Initial commit: Linux kernel 4.19 for Unplugged OS

This commit is contained in:
ziv committed 2025-10-06 13:52:42 +00:00
commit d30235b04c
80519 files changed
+35800681

No files matched your search

+88
View File
@@ -0,0 +1,88 @@
# SPDX-License-Identifier: GPL-2.0
#
# Makefile for the linux kernel.
#
ifdef CONFIG_FUNCTION_TRACER
# Do not trace tracer code
CFLAGS_REMOVE_ftrace.o = $(CC_FLAGS_FTRACE)
# Do not trace early setup code
CFLAGS_REMOVE_early.o = $(CC_FLAGS_FTRACE)
CFLAGS_REMOVE_early_nobss.o = $(CC_FLAGS_FTRACE)
endif
GCOV_PROFILE_early.o := n
GCOV_PROFILE_early_nobss.o := n
KCOV_INSTRUMENT_early.o := n
KCOV_INSTRUMENT_early_nobss.o := n
UBSAN_SANITIZE_early.o := n
UBSAN_SANITIZE_early_nobss.o := n
KASAN_SANITIZE_early_nobss.o := n
#
# Passing null pointers is ok for smp code, since we access the lowcore here.
#
CFLAGS_smp.o := -Wno-nonnull
#
# Disable tailcall optimizations for stack / callchain walking functions
# since this might generate broken code when accessing register 15 and
# passing its content to other functions.
#
CFLAGS_stacktrace.o += -fno-optimize-sibling-calls
CFLAGS_dumpstack.o += -fno-optimize-sibling-calls
#
# Pass UTS_MACHINE for user_regset definition
#
CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"'
obj-y := traps.o time.o process.o base.o early.o setup.o idle.o vtime.o
obj-y += processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o nmi.o
obj-y += debug.o irq.o ipl.o dis.o diag.o vdso.o early_nobss.o
obj-y += sysinfo.o lgr.o os_info.o machine_kexec.o pgm_check.o
obj-y += runtime_instr.o cache.o fpu.o dumpstack.o guarded_storage.o sthyi.o
obj-y += entry.o reipl.o relocate_kernel.o kdebugfs.o alternative.o
obj-y += nospec-branch.o
extra-y += head64.o vmlinux.lds
obj-$(CONFIG_SYSFS) += nospec-sysfs.o
CFLAGS_REMOVE_nospec-branch.o += $(CC_FLAGS_EXPOLINE)
obj-$(CONFIG_MODULES) += module.o
obj-$(CONFIG_SMP) += smp.o
obj-$(CONFIG_SCHED_TOPOLOGY) += topology.o
obj-$(CONFIG_HIBERNATION) += suspend.o swsusp.o
obj-$(CONFIG_AUDIT) += audit.o
compat-obj-$(CONFIG_AUDIT) += compat_audit.o
obj-$(CONFIG_COMPAT) += compat_linux.o compat_signal.o
obj-$(CONFIG_COMPAT) += compat_wrapper.o $(compat-obj-y)
obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
obj-$(CONFIG_STACKTRACE) += stacktrace.o
obj-$(CONFIG_KPROBES) += kprobes.o
obj-$(CONFIG_FUNCTION_TRACER) += mcount.o ftrace.o
obj-$(CONFIG_CRASH_DUMP) += crash_dump.o
obj-$(CONFIG_UPROBES) += uprobes.o
obj-$(CONFIG_JUMP_LABEL) += jump_label.o
obj-$(CONFIG_KEXEC_FILE) += machine_kexec_file.o kexec_image.o
obj-$(CONFIG_KEXEC_FILE) += kexec_elf.o
obj-$(CONFIG_PERF_EVENTS) += perf_event.o perf_cpum_cf.o perf_cpum_sf.o
obj-$(CONFIG_PERF_EVENTS) += perf_cpum_cf_events.o perf_regs.o
obj-$(CONFIG_TRACEPOINTS) += trace.o
# vdso
obj-y += vdso64/
obj-$(CONFIG_COMPAT) += vdso32/
chkbss := head64.o early_nobss.o
include $(srctree)/arch/s390/scripts/Makefile.chkbss
+113
View File
@@ -0,0 +1,113 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/module.h>
#include <asm/alternative.h>
#include <asm/facility.h>
#include <asm/nospec-branch.h>
#define MAX_PATCH_LEN (255 - 1)
static int __initdata_or_module alt_instr_disabled;
static int __init disable_alternative_instructions(char *str)
{
alt_instr_disabled = 1;
return 0;
}
early_param("noaltinstr", disable_alternative_instructions);
struct brcl_insn {
u16 opc;
s32 disp;
} __packed;
static u16 __initdata_or_module nop16 = 0x0700;
static u32 __initdata_or_module nop32 = 0x47000000;
static struct brcl_insn __initdata_or_module nop48 = {
0xc004, 0
};
static const void *nops[] __initdata_or_module = {
&nop16,
&nop32,
&nop48
};
static void __init_or_module add_jump_padding(void *insns, unsigned int len)
{
struct brcl_insn brcl = {
0xc0f4,
len / 2
};
memcpy(insns, &brcl, sizeof(brcl));
insns += sizeof(brcl);
len -= sizeof(brcl);
while (len > 0) {
memcpy(insns, &nop16, 2);
insns += 2;
len -= 2;
}
}
static void __init_or_module add_padding(void *insns, unsigned int len)
{
if (len > 6)
add_jump_padding(insns, len);
else if (len >= 2)
memcpy(insns, nops[len / 2 - 1], len);
}
static void __init_or_module __apply_alternatives(struct alt_instr *start,
struct alt_instr *end)
{
struct alt_instr *a;
u8 *instr, *replacement;
u8 insnbuf[MAX_PATCH_LEN];
/*
* The scan order should be from start to end. A later scanned
* alternative code can overwrite previously scanned alternative code.
*/
for (a = start; a < end; a++) {
int insnbuf_sz = 0;
instr = (u8 *)&a->instr_offset + a->instr_offset;
replacement = (u8 *)&a->repl_offset + a->repl_offset;
if (!__test_facility(a->facility,
S390_lowcore.alt_stfle_fac_list))
continue;
if (unlikely(a->instrlen % 2 || a->replacementlen % 2)) {
WARN_ONCE(1, "cpu alternatives instructions length is "
"odd, skipping patching\n");
continue;
}
memcpy(insnbuf, replacement, a->replacementlen);
insnbuf_sz = a->replacementlen;
if (a->instrlen > a->replacementlen) {
add_padding(insnbuf + a->replacementlen,
a->instrlen - a->replacementlen);
insnbuf_sz += a->instrlen - a->replacementlen;
}
s390_kernel_write(instr, insnbuf, insnbuf_sz);
}
}
void __init_or_module apply_alternatives(struct alt_instr *start,
struct alt_instr *end)
{
if (!alt_instr_disabled)
__apply_alternatives(start, end);
}
extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
void __init apply_alternative_instructions(void)
{
apply_alternatives(__alt_instructions, __alt_instructions_end);
}
+206
View File
@@ -0,0 +1,206 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Generate definitions needed by assembly language modules.
* This code generates raw asm output which is post-processed to extract
* and format the required data.
*/
#define ASM_OFFSETS_C
#include <linux/kbuild.h>
#include <linux/kvm_host.h>
#include <linux/sched.h>
#include <linux/purgatory.h>
#include <asm/idle.h>
#include <asm/vdso.h>
#include <asm/pgtable.h>
#include <asm/gmap.h>
#include <asm/nmi.h>
int main(void)
{
/* task struct offsets */
OFFSET(__TASK_stack, task_struct, stack);
OFFSET(__TASK_thread, task_struct, thread);
OFFSET(__TASK_pid, task_struct, pid);
BLANK();
/* thread struct offsets */
OFFSET(__THREAD_ksp, thread_struct, ksp);
OFFSET(__THREAD_sysc_table, thread_struct, sys_call_table);
OFFSET(__THREAD_last_break, thread_struct, last_break);
OFFSET(__THREAD_FPU_fpc, thread_struct, fpu.fpc);
OFFSET(__THREAD_FPU_regs, thread_struct, fpu.regs);
OFFSET(__THREAD_per_cause, thread_struct, per_event.cause);
OFFSET(__THREAD_per_address, thread_struct, per_event.address);
OFFSET(__THREAD_per_paid, thread_struct, per_event.paid);
OFFSET(__THREAD_trap_tdb, thread_struct, trap_tdb);
BLANK();
/* thread info offsets */
OFFSET(__TI_flags, task_struct, thread_info.flags);
BLANK();
/* pt_regs offsets */
OFFSET(__PT_ARGS, pt_regs, args);
OFFSET(__PT_PSW, pt_regs, psw);
OFFSET(__PT_GPRS, pt_regs, gprs);
OFFSET(__PT_ORIG_GPR2, pt_regs, orig_gpr2);
OFFSET(__PT_INT_CODE, pt_regs, int_code);
OFFSET(__PT_INT_PARM, pt_regs, int_parm);
OFFSET(__PT_INT_PARM_LONG, pt_regs, int_parm_long);
OFFSET(__PT_FLAGS, pt_regs, flags);
DEFINE(__PT_SIZE, sizeof(struct pt_regs));
BLANK();
/* stack_frame offsets */
OFFSET(__SF_BACKCHAIN, stack_frame, back_chain);
OFFSET(__SF_GPRS, stack_frame, gprs);
OFFSET(__SF_EMPTY, stack_frame, empty1);
OFFSET(__SF_SIE_CONTROL, stack_frame, empty1[0]);
OFFSET(__SF_SIE_SAVEAREA, stack_frame, empty1[1]);
OFFSET(__SF_SIE_REASON, stack_frame, empty1[2]);
OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[3]);
BLANK();
/* timeval/timezone offsets for use by vdso */
OFFSET(__VDSO_UPD_COUNT, vdso_data, tb_update_count);
OFFSET(__VDSO_XTIME_STAMP, vdso_data, xtime_tod_stamp);
OFFSET(__VDSO_XTIME_SEC, vdso_data, xtime_clock_sec);
OFFSET(__VDSO_XTIME_NSEC, vdso_data, xtime_clock_nsec);
OFFSET(__VDSO_XTIME_CRS_SEC, vdso_data, xtime_coarse_sec);
OFFSET(__VDSO_XTIME_CRS_NSEC, vdso_data, xtime_coarse_nsec);
OFFSET(__VDSO_WTOM_SEC, vdso_data, wtom_clock_sec);
OFFSET(__VDSO_WTOM_NSEC, vdso_data, wtom_clock_nsec);
OFFSET(__VDSO_WTOM_CRS_SEC, vdso_data, wtom_coarse_sec);
OFFSET(__VDSO_WTOM_CRS_NSEC, vdso_data, wtom_coarse_nsec);
OFFSET(__VDSO_TIMEZONE, vdso_data, tz_minuteswest);
OFFSET(__VDSO_ECTG_OK, vdso_data, ectg_available);
OFFSET(__VDSO_TK_MULT, vdso_data, tk_mult);
OFFSET(__VDSO_TK_SHIFT, vdso_data, tk_shift);
OFFSET(__VDSO_TS_DIR, vdso_data, ts_dir);
OFFSET(__VDSO_TS_END, vdso_data, ts_end);
OFFSET(__VDSO_CLOCK_REALTIME_RES, vdso_data, hrtimer_res);
OFFSET(__VDSO_ECTG_BASE, vdso_per_cpu_data, ectg_timer_base);
OFFSET(__VDSO_ECTG_USER, vdso_per_cpu_data, ectg_user_time);
OFFSET(__VDSO_CPU_NR, vdso_per_cpu_data, cpu_nr);
OFFSET(__VDSO_NODE_ID, vdso_per_cpu_data, node_id);
BLANK();
/* constants used by the vdso */
DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME);
DEFINE(__CLOCK_MONOTONIC, CLOCK_MONOTONIC);
DEFINE(__CLOCK_REALTIME_COARSE, CLOCK_REALTIME_COARSE);
DEFINE(__CLOCK_MONOTONIC_COARSE, CLOCK_MONOTONIC_COARSE);
DEFINE(__CLOCK_THREAD_CPUTIME_ID, CLOCK_THREAD_CPUTIME_ID);
DEFINE(__CLOCK_COARSE_RES, LOW_RES_NSEC);
BLANK();
/* idle data offsets */
OFFSET(__CLOCK_IDLE_ENTER, s390_idle_data, clock_idle_enter);
OFFSET(__CLOCK_IDLE_EXIT, s390_idle_data, clock_idle_exit);
OFFSET(__TIMER_IDLE_ENTER, s390_idle_data, timer_idle_enter);
OFFSET(__TIMER_IDLE_EXIT, s390_idle_data, timer_idle_exit);
BLANK();
/* hardware defined lowcore locations 0x000 - 0x1ff */
OFFSET(__LC_EXT_PARAMS, lowcore, ext_params);
OFFSET(__LC_EXT_CPU_ADDR, lowcore, ext_cpu_addr);
OFFSET(__LC_EXT_INT_CODE, lowcore, ext_int_code);
OFFSET(__LC_SVC_ILC, lowcore, svc_ilc);
OFFSET(__LC_SVC_INT_CODE, lowcore, svc_code);
OFFSET(__LC_PGM_ILC, lowcore, pgm_ilc);
OFFSET(__LC_PGM_INT_CODE, lowcore, pgm_code);
OFFSET(__LC_DATA_EXC_CODE, lowcore, data_exc_code);
OFFSET(__LC_MON_CLASS_NR, lowcore, mon_class_num);
OFFSET(__LC_PER_CODE, lowcore, per_code);
OFFSET(__LC_PER_ATMID, lowcore, per_atmid);
OFFSET(__LC_PER_ADDRESS, lowcore, per_address);
OFFSET(__LC_EXC_ACCESS_ID, lowcore, exc_access_id);
OFFSET(__LC_PER_ACCESS_ID, lowcore, per_access_id);
OFFSET(__LC_OP_ACCESS_ID, lowcore, op_access_id);
OFFSET(__LC_AR_MODE_ID, lowcore, ar_mode_id);
OFFSET(__LC_TRANS_EXC_CODE, lowcore, trans_exc_code);
OFFSET(__LC_MON_CODE, lowcore, monitor_code);
OFFSET(__LC_SUBCHANNEL_ID, lowcore, subchannel_id);
OFFSET(__LC_SUBCHANNEL_NR, lowcore, subchannel_nr);
OFFSET(__LC_IO_INT_PARM, lowcore, io_int_parm);
OFFSET(__LC_IO_INT_WORD, lowcore, io_int_word);
OFFSET(__LC_STFL_FAC_LIST, lowcore, stfl_fac_list);
OFFSET(__LC_STFLE_FAC_LIST, lowcore, stfle_fac_list);
OFFSET(__LC_MCCK_CODE, lowcore, mcck_interruption_code);
OFFSET(__LC_EXT_DAMAGE_CODE, lowcore, external_damage_code);
OFFSET(__LC_MCCK_FAIL_STOR_ADDR, lowcore, failing_storage_address);
OFFSET(__LC_LAST_BREAK, lowcore, breaking_event_addr);
OFFSET(__LC_RST_OLD_PSW, lowcore, restart_old_psw);
OFFSET(__LC_EXT_OLD_PSW, lowcore, external_old_psw);
OFFSET(__LC_SVC_OLD_PSW, lowcore, svc_old_psw);
OFFSET(__LC_PGM_OLD_PSW, lowcore, program_old_psw);
OFFSET(__LC_MCK_OLD_PSW, lowcore, mcck_old_psw);
OFFSET(__LC_IO_OLD_PSW, lowcore, io_old_psw);
OFFSET(__LC_RST_NEW_PSW, lowcore, restart_psw);
OFFSET(__LC_EXT_NEW_PSW, lowcore, external_new_psw);
OFFSET(__LC_SVC_NEW_PSW, lowcore, svc_new_psw);
OFFSET(__LC_PGM_NEW_PSW, lowcore, program_new_psw);
OFFSET(__LC_MCK_NEW_PSW, lowcore, mcck_new_psw);
OFFSET(__LC_IO_NEW_PSW, lowcore, io_new_psw);
/* software defined lowcore locations 0x200 - 0xdff*/
OFFSET(__LC_SAVE_AREA_SYNC, lowcore, save_area_sync);
OFFSET(__LC_SAVE_AREA_ASYNC, lowcore, save_area_async);
OFFSET(__LC_SAVE_AREA_RESTART, lowcore, save_area_restart);
OFFSET(__LC_CPU_FLAGS, lowcore, cpu_flags);
OFFSET(__LC_RETURN_PSW, lowcore, return_psw);
OFFSET(__LC_RETURN_MCCK_PSW, lowcore, return_mcck_psw);
OFFSET(__LC_SYNC_ENTER_TIMER, lowcore, sync_enter_timer);
OFFSET(__LC_ASYNC_ENTER_TIMER, lowcore, async_enter_timer);
OFFSET(__LC_MCCK_ENTER_TIMER, lowcore, mcck_enter_timer);
OFFSET(__LC_EXIT_TIMER, lowcore, exit_timer);
OFFSET(__LC_USER_TIMER, lowcore, user_timer);
OFFSET(__LC_SYSTEM_TIMER, lowcore, system_timer);
OFFSET(__LC_STEAL_TIMER, lowcore, steal_timer);
OFFSET(__LC_LAST_UPDATE_TIMER, lowcore, last_update_timer);
OFFSET(__LC_LAST_UPDATE_CLOCK, lowcore, last_update_clock);
OFFSET(__LC_INT_CLOCK, lowcore, int_clock);
OFFSET(__LC_MCCK_CLOCK, lowcore, mcck_clock);
OFFSET(__LC_CLOCK_COMPARATOR, lowcore, clock_comparator);
OFFSET(__LC_BOOT_CLOCK, lowcore, boot_clock);
OFFSET(__LC_CURRENT, lowcore, current_task);
OFFSET(__LC_KERNEL_STACK, lowcore, kernel_stack);
OFFSET(__LC_ASYNC_STACK, lowcore, async_stack);
OFFSET(__LC_PANIC_STACK, lowcore, panic_stack);
OFFSET(__LC_RESTART_STACK, lowcore, restart_stack);
OFFSET(__LC_RESTART_FN, lowcore, restart_fn);
OFFSET(__LC_RESTART_DATA, lowcore, restart_data);
OFFSET(__LC_RESTART_SOURCE, lowcore, restart_source);
OFFSET(__LC_USER_ASCE, lowcore, user_asce);
OFFSET(__LC_VDSO_ASCE, lowcore, vdso_asce);
OFFSET(__LC_LPP, lowcore, lpp);
OFFSET(__LC_CURRENT_PID, lowcore, current_pid);
OFFSET(__LC_PERCPU_OFFSET, lowcore, percpu_offset);
OFFSET(__LC_VDSO_PER_CPU, lowcore, vdso_per_cpu_data);
OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags);
OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count);
OFFSET(__LC_GMAP, lowcore, gmap);
OFFSET(__LC_BR_R1, lowcore, br_r1_trampoline);
/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
/* hardware defined lowcore locations 0x1000 - 0x18ff */
OFFSET(__LC_MCESAD, lowcore, mcesad);
OFFSET(__LC_EXT_PARAMS2, lowcore, ext_params2);
OFFSET(__LC_FPREGS_SAVE_AREA, lowcore, floating_pt_save_area);
OFFSET(__LC_GPREGS_SAVE_AREA, lowcore, gpregs_save_area);
OFFSET(__LC_PSW_SAVE_AREA, lowcore, psw_save_area);
OFFSET(__LC_PREFIX_SAVE_AREA, lowcore, prefixreg_save_area);
OFFSET(__LC_FP_CREG_SAVE_AREA, lowcore, fpt_creg_save_area);
OFFSET(__LC_TOD_PROGREG_SAVE_AREA, lowcore, tod_progreg_save_area);
OFFSET(__LC_CPU_TIMER_SAVE_AREA, lowcore, cpu_timer_save_area);
OFFSET(__LC_CLOCK_COMP_SAVE_AREA, lowcore, clock_comp_save_area);
OFFSET(__LC_AREGS_SAVE_AREA, lowcore, access_regs_save_area);
OFFSET(__LC_CREGS_SAVE_AREA, lowcore, cregs_save_area);
OFFSET(__LC_PGM_TDB, lowcore, pgm_tdb);
BLANK();
/* extended machine check save area */
OFFSET(__MCESA_GS_SAVE_AREA, mcesa, guarded_storage_save_area);
BLANK();
/* gmap/sie offsets */
OFFSET(__GMAP_ASCE, gmap, asce);
OFFSET(__SIE_PROG0C, kvm_s390_sie_block, prog0c);
OFFSET(__SIE_PROG20, kvm_s390_sie_block, prog20);
/* kexec_sha_region */
OFFSET(__KEXEC_SHA_REGION_START, kexec_sha_region, start);
OFFSET(__KEXEC_SHA_REGION_LEN, kexec_sha_region, len);
DEFINE(__KEXEC_SHA_REGION_SIZE, sizeof(struct kexec_sha_region));
return 0;
}
+79
View File
@@ -0,0 +1,79 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/init.h>
#include <linux/types.h>
#include <linux/audit.h>
#include <asm/unistd.h>
#include "audit.h"
static unsigned dir_class[] = {
#include <asm-generic/audit_dir_write.h>
~0U
};
static unsigned read_class[] = {
#include <asm-generic/audit_read.h>
~0U
};
static unsigned write_class[] = {
#include <asm-generic/audit_write.h>
~0U
};
static unsigned chattr_class[] = {
#include <asm-generic/audit_change_attr.h>
~0U
};
static unsigned signal_class[] = {
#include <asm-generic/audit_signal.h>
~0U
};
int audit_classify_arch(int arch)
{
#ifdef CONFIG_COMPAT
if (arch == AUDIT_ARCH_S390)
return 1;
#endif
return 0;
}
int audit_classify_syscall(int abi, unsigned syscall)
{
#ifdef CONFIG_COMPAT
if (abi == AUDIT_ARCH_S390)
return s390_classify_syscall(syscall);
#endif
switch(syscall) {
case __NR_open:
return 2;
case __NR_openat:
return 3;
case __NR_socketcall:
return 4;
case __NR_execve:
return 5;
default:
return 0;
}
}
static int __init audit_classes_init(void)
{
#ifdef CONFIG_COMPAT
audit_register_class(AUDIT_CLASS_WRITE_32, s390_write_class);
audit_register_class(AUDIT_CLASS_READ_32, s390_read_class);
audit_register_class(AUDIT_CLASS_DIR_WRITE_32, s390_dir_class);
audit_register_class(AUDIT_CLASS_CHATTR_32, s390_chattr_class);
audit_register_class(AUDIT_CLASS_SIGNAL_32, s390_signal_class);
#endif
audit_register_class(AUDIT_CLASS_WRITE, write_class);
audit_register_class(AUDIT_CLASS_READ, read_class);
audit_register_class(AUDIT_CLASS_DIR_WRITE, dir_class);
audit_register_class(AUDIT_CLASS_CHATTR, chattr_class);
audit_register_class(AUDIT_CLASS_SIGNAL, signal_class);
return 0;
}
__initcall(audit_classes_init);
+16
View File
@@ -0,0 +1,16 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __ARCH_S390_KERNEL_AUDIT_H
#define __ARCH_S390_KERNEL_AUDIT_H
#include <linux/types.h>
#ifdef CONFIG_COMPAT
extern int s390_classify_syscall(unsigned);
extern __u32 s390_dir_class[];
extern __u32 s390_write_class[];
extern __u32 s390_read_class[];
extern __u32 s390_chattr_class[];
extern __u32 s390_signal_class[];
#endif /* CONFIG_COMPAT */
#endif /* __ARCH_S390_KERNEL_AUDIT_H */
+149
View File
@@ -0,0 +1,149 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* arch/s390/kernel/base.S
*
* Copyright IBM Corp. 2006, 2007
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
* Michael Holzheu <holzheu@de.ibm.com>
*/
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/sigp.h>
GEN_BR_THUNK %r9
GEN_BR_THUNK %r14
ENTRY(s390_base_mcck_handler)
basr %r13,0
0: lg %r15,__LC_PANIC_STACK # load panic stack
aghi %r15,-STACK_FRAME_OVERHEAD
larl %r1,s390_base_mcck_handler_fn
lg %r9,0(%r1)
ltgr %r9,%r9
jz 1f
BASR_EX %r14,%r9
1: la %r1,4095
lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)
lpswe __LC_MCK_OLD_PSW
.section .bss
.align 8
.globl s390_base_mcck_handler_fn
s390_base_mcck_handler_fn:
.quad 0
.previous
ENTRY(s390_base_ext_handler)
stmg %r0,%r15,__LC_SAVE_AREA_ASYNC
basr %r13,0
0: aghi %r15,-STACK_FRAME_OVERHEAD
larl %r1,s390_base_ext_handler_fn
lg %r9,0(%r1)
ltgr %r9,%r9
jz 1f
BASR_EX %r14,%r9
1: lmg %r0,%r15,__LC_SAVE_AREA_ASYNC
ni __LC_EXT_OLD_PSW+1,0xfd # clear wait state bit
lpswe __LC_EXT_OLD_PSW
.section .bss
.align 8
.globl s390_base_ext_handler_fn
s390_base_ext_handler_fn:
.quad 0
.previous
ENTRY(s390_base_pgm_handler)
stmg %r0,%r15,__LC_SAVE_AREA_SYNC
basr %r13,0
0: aghi %r15,-STACK_FRAME_OVERHEAD
larl %r1,s390_base_pgm_handler_fn
lg %r9,0(%r1)
ltgr %r9,%r9
jz 1f
BASR_EX %r14,%r9
lmg %r0,%r15,__LC_SAVE_AREA_SYNC
lpswe __LC_PGM_OLD_PSW
1: lpswe disabled_wait_psw-0b(%r13)
.align 8
disabled_wait_psw:
.quad 0x0002000180000000,0x0000000000000000 + s390_base_pgm_handler
.section .bss
.align 8
.globl s390_base_pgm_handler_fn
s390_base_pgm_handler_fn:
.quad 0
.previous
#
# Calls diag 308 subcode 1 and continues execution
#
ENTRY(diag308_reset)
larl %r4,.Lctlregs # Save control registers
stctg %c0,%c15,0(%r4)
lg %r2,0(%r4) # Disable lowcore protection
nilh %r2,0xefff
larl %r4,.Lctlreg0
stg %r2,0(%r4)
lctlg %c0,%c0,0(%r4)
larl %r4,.Lfpctl # Floating point control register
stfpc 0(%r4)
larl %r4,.Lprefix # Save prefix register
stpx 0(%r4)
larl %r4,.Lprefix_zero # Set prefix register to 0
spx 0(%r4)
larl %r4,.Lcontinue_psw # Save PSW flags
epsw %r2,%r3
stm %r2,%r3,0(%r4)
larl %r4,.Lrestart_psw # Setup restart PSW at absolute 0
lghi %r3,0
lg %r4,0(%r4) # Save PSW
sturg %r4,%r3 # Use sturg, because of large pages
lghi %r1,1
lghi %r0,0
diag %r0,%r1,0x308
.Lrestart_part2:
lhi %r0,0 # Load r0 with zero
lhi %r1,2 # Use mode 2 = ESAME (dump)
sigp %r1,%r0,SIGP_SET_ARCHITECTURE # Switch to ESAME mode
sam64 # Switch to 64 bit addressing mode
larl %r4,.Lctlregs # Restore control registers
lctlg %c0,%c15,0(%r4)
larl %r4,.Lfpctl # Restore floating point ctl register
lfpc 0(%r4)
larl %r4,.Lprefix # Restore prefix register
spx 0(%r4)
larl %r4,.Lcontinue_psw # Restore PSW flags
lpswe 0(%r4)
.Lcontinue:
BR_EX %r14
.align 16
.Lrestart_psw:
.long 0x00080000,0x80000000 + .Lrestart_part2
.section .data..nosave,"aw",@progbits
.align 8
.Lcontinue_psw:
.quad 0,.Lcontinue
.previous
.section .bss
.align 8
.Lctlreg0:
.quad 0
.Lctlregs:
.rept 16
.quad 0
.endr
.Lfpctl:
.long 0
.Lprefix:
.long 0
.Lprefix_zero:
.long 0
.previous
+177
View File
@@ -0,0 +1,177 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Extract CPU cache information and expose them via sysfs.
*
* Copyright IBM Corp. 2012
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
*/
#include <linux/seq_file.h>
#include <linux/cpu.h>
#include <linux/cacheinfo.h>
#include <asm/facility.h>
enum {
CACHE_SCOPE_NOTEXISTS,
CACHE_SCOPE_PRIVATE,
CACHE_SCOPE_SHARED,
CACHE_SCOPE_RESERVED,
};
enum {
CTYPE_SEPARATE,
CTYPE_DATA,
CTYPE_INSTRUCTION,
CTYPE_UNIFIED,
};
enum {
EXTRACT_TOPOLOGY,
EXTRACT_LINE_SIZE,
EXTRACT_SIZE,
EXTRACT_ASSOCIATIVITY,
};
enum {
CACHE_TI_UNIFIED = 0,
CACHE_TI_DATA = 0,
CACHE_TI_INSTRUCTION,
};
struct cache_info {
unsigned char : 4;
unsigned char scope : 2;
unsigned char type : 2;
};
#define CACHE_MAX_LEVEL 8
union cache_topology {
struct cache_info ci[CACHE_MAX_LEVEL];
unsigned long long raw;
};
static const char * const cache_type_string[] = {
"",
"Instruction",
"Data",
"",
"Unified",
};
static const enum cache_type cache_type_map[] = {
[CTYPE_SEPARATE] = CACHE_TYPE_SEPARATE,
[CTYPE_DATA] = CACHE_TYPE_DATA,
[CTYPE_INSTRUCTION] = CACHE_TYPE_INST,
[CTYPE_UNIFIED] = CACHE_TYPE_UNIFIED,
};
void show_cacheinfo(struct seq_file *m)
{
struct cpu_cacheinfo *this_cpu_ci;
struct cacheinfo *cache;
int idx;
if (!test_facility(34))
return;
this_cpu_ci = get_cpu_cacheinfo(cpumask_any(cpu_online_mask));
for (idx = 0; idx < this_cpu_ci->num_leaves; idx++) {
cache = this_cpu_ci->info_list + idx;
seq_printf(m, "cache%-11d: ", idx);
seq_printf(m, "level=%d ", cache->level);
seq_printf(m, "type=%s ", cache_type_string[cache->type]);
seq_printf(m, "scope=%s ",
cache->disable_sysfs ? "Shared" : "Private");
seq_printf(m, "size=%dK ", cache->size >> 10);
seq_printf(m, "line_size=%u ", cache->coherency_line_size);
seq_printf(m, "associativity=%d", cache->ways_of_associativity);
seq_puts(m, "\n");
}
}
static inline enum cache_type get_cache_type(struct cache_info *ci, int level)
{
if (level >= CACHE_MAX_LEVEL)
return CACHE_TYPE_NOCACHE;
ci += level;
if (ci->scope != CACHE_SCOPE_SHARED && ci->scope != CACHE_SCOPE_PRIVATE)
return CACHE_TYPE_NOCACHE;
return cache_type_map[ci->type];
}
static inline unsigned long ecag(int ai, int li, int ti)
{
return __ecag(ECAG_CACHE_ATTRIBUTE, ai << 4 | li << 1 | ti);
}
static void ci_leaf_init(struct cacheinfo *this_leaf, int private,
enum cache_type type, unsigned int level, int cpu)
{
int ti, num_sets;
if (type == CACHE_TYPE_INST)
ti = CACHE_TI_INSTRUCTION;
else
ti = CACHE_TI_UNIFIED;
this_leaf->level = level + 1;
this_leaf->type = type;
this_leaf->coherency_line_size = ecag(EXTRACT_LINE_SIZE, level, ti);
this_leaf->ways_of_associativity = ecag(EXTRACT_ASSOCIATIVITY, level, ti);
this_leaf->size = ecag(EXTRACT_SIZE, level, ti);
num_sets = this_leaf->size / this_leaf->coherency_line_size;
num_sets /= this_leaf->ways_of_associativity;
this_leaf->number_of_sets = num_sets;
cpumask_set_cpu(cpu, &this_leaf->shared_cpu_map);
if (!private)
this_leaf->disable_sysfs = true;
}
int init_cache_level(unsigned int cpu)
{
struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
unsigned int level = 0, leaves = 0;
union cache_topology ct;
enum cache_type ctype;
if (!test_facility(34))
return -EOPNOTSUPP;
if (!this_cpu_ci)
return -EINVAL;
ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0);
do {
ctype = get_cache_type(&ct.ci[0], level);
if (ctype == CACHE_TYPE_NOCACHE)
break;
/* Separate instruction and data caches */
leaves += (ctype == CACHE_TYPE_SEPARATE) ? 2 : 1;
} while (++level < CACHE_MAX_LEVEL);
this_cpu_ci->num_levels = level;
this_cpu_ci->num_leaves = leaves;
return 0;
}
int populate_cache_leaves(unsigned int cpu)
{
struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
struct cacheinfo *this_leaf = this_cpu_ci->info_list;
unsigned int level, idx, pvt;
union cache_topology ct;
enum cache_type ctype;
if (!test_facility(34))
return -EOPNOTSUPP;
ct.raw = ecag(EXTRACT_TOPOLOGY, 0, 0);
for (idx = 0, level = 0; level < this_cpu_ci->num_levels &&
idx < this_cpu_ci->num_leaves; idx++, level++) {
if (!this_leaf)
return -EINVAL;
pvt = (ct.ci[level].scope == CACHE_SCOPE_PRIVATE) ? 1 : 0;
ctype = get_cache_type(&ct.ci[0], level);
if (ctype == CACHE_TYPE_SEPARATE) {
ci_leaf_init(this_leaf++, pvt, CACHE_TYPE_DATA, level, cpu);
ci_leaf_init(this_leaf++, pvt, CACHE_TYPE_INST, level, cpu);
} else {
ci_leaf_init(this_leaf++, pvt, ctype, level, cpu);
}
}
return 0;
}
+45
View File
@@ -0,0 +1,45 @@
// SPDX-License-Identifier: GPL-2.0
#undef __s390x__
#include <asm/unistd.h>
#include "audit.h"
unsigned s390_dir_class[] = {
#include <asm-generic/audit_dir_write.h>
~0U
};
unsigned s390_chattr_class[] = {
#include <asm-generic/audit_change_attr.h>
~0U
};
unsigned s390_write_class[] = {
#include <asm-generic/audit_write.h>
~0U
};
unsigned s390_read_class[] = {
#include <asm-generic/audit_read.h>
~0U
};
unsigned s390_signal_class[] = {
#include <asm-generic/audit_signal.h>
~0U
};
int s390_classify_syscall(unsigned syscall)
{
switch(syscall) {
case __NR_open:
return 2;
case __NR_openat:
return 3;
case __NR_socketcall:
return 4;
case __NR_execve:
return 5;
default:
return 1;
}
}
+522
View File
@@ -0,0 +1,522 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 version
* Copyright IBM Corp. 2000
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
* Gerhard Tonn (ton@de.ibm.com)
* Thomas Spatzier (tspat@de.ibm.com)
*
* Conversion between 31bit and 64bit native syscalls.
*
* Heavily inspired by the 32-bit Sparc compat code which is
* Copyright (C) 1997,1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
* Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu)
*
*/
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/file.h>
#include <linux/signal.h>
#include <linux/resource.h>
#include <linux/times.h>
#include <linux/smp.h>
#include <linux/sem.h>
#include <linux/msg.h>
#include <linux/shm.h>
#include <linux/uio.h>
#include <linux/quota.h>
#include <linux/poll.h>
#include <linux/personality.h>
#include <linux/stat.h>
#include <linux/filter.h>
#include <linux/highmem.h>
#include <linux/highuid.h>
#include <linux/mman.h>
#include <linux/ipv6.h>
#include <linux/in.h>
#include <linux/icmpv6.h>
#include <linux/syscalls.h>
#include <linux/sysctl.h>
#include <linux/binfmts.h>
#include <linux/capability.h>
#include <linux/compat.h>
#include <linux/vfs.h>
#include <linux/ptrace.h>
#include <linux/fadvise.h>
#include <linux/ipc.h>
#include <linux/slab.h>
#include <asm/types.h>
#include <linux/uaccess.h>
#include <net/scm.h>
#include <net/sock.h>
#include "compat_linux.h"
/* For this source file, we want overflow handling. */
#undef high2lowuid
#undef high2lowgid
#undef low2highuid
#undef low2highgid
#undef SET_UID16
#undef SET_GID16
#undef NEW_TO_OLD_UID
#undef NEW_TO_OLD_GID
#undef SET_OLDSTAT_UID
#undef SET_OLDSTAT_GID
#undef SET_STAT_UID
#undef SET_STAT_GID
#define high2lowuid(uid) ((uid) > 65535) ? (u16)overflowuid : (u16)(uid)
#define high2lowgid(gid) ((gid) > 65535) ? (u16)overflowgid : (u16)(gid)
#define low2highuid(uid) ((uid) == (u16)-1) ? (uid_t)-1 : (uid_t)(uid)
#define low2highgid(gid) ((gid) == (u16)-1) ? (gid_t)-1 : (gid_t)(gid)
#define SET_UID16(var, uid) var = high2lowuid(uid)
#define SET_GID16(var, gid) var = high2lowgid(gid)
#define NEW_TO_OLD_UID(uid) high2lowuid(uid)
#define NEW_TO_OLD_GID(gid) high2lowgid(gid)
#define SET_OLDSTAT_UID(stat, uid) (stat).st_uid = high2lowuid(uid)
#define SET_OLDSTAT_GID(stat, gid) (stat).st_gid = high2lowgid(gid)
#define SET_STAT_UID(stat, uid) (stat).st_uid = high2lowuid(uid)
#define SET_STAT_GID(stat, gid) (stat).st_gid = high2lowgid(gid)
COMPAT_SYSCALL_DEFINE3(s390_chown16, const char __user *, filename,
u16, user, u16, group)
{
return ksys_chown(filename, low2highuid(user), low2highgid(group));
}
COMPAT_SYSCALL_DEFINE3(s390_lchown16, const char __user *,
filename, u16, user, u16, group)
{
return ksys_lchown(filename, low2highuid(user), low2highgid(group));
}
COMPAT_SYSCALL_DEFINE3(s390_fchown16, unsigned int, fd, u16, user, u16, group)
{
return ksys_fchown(fd, low2highuid(user), low2highgid(group));
}
COMPAT_SYSCALL_DEFINE2(s390_setregid16, u16, rgid, u16, egid)
{
return sys_setregid(low2highgid(rgid), low2highgid(egid));
}
COMPAT_SYSCALL_DEFINE1(s390_setgid16, u16, gid)
{
return sys_setgid(low2highgid(gid));
}
COMPAT_SYSCALL_DEFINE2(s390_setreuid16, u16, ruid, u16, euid)
{
return sys_setreuid(low2highuid(ruid), low2highuid(euid));
}
COMPAT_SYSCALL_DEFINE1(s390_setuid16, u16, uid)
{
return sys_setuid(low2highuid(uid));
}
COMPAT_SYSCALL_DEFINE3(s390_setresuid16, u16, ruid, u16, euid, u16, suid)
{
return sys_setresuid(low2highuid(ruid), low2highuid(euid),
low2highuid(suid));
}
COMPAT_SYSCALL_DEFINE3(s390_getresuid16, u16 __user *, ruidp,
u16 __user *, euidp, u16 __user *, suidp)
{
const struct cred *cred = current_cred();
int retval;
u16 ruid, euid, suid;
ruid = high2lowuid(from_kuid_munged(cred->user_ns, cred->uid));
euid = high2lowuid(from_kuid_munged(cred->user_ns, cred->euid));
suid = high2lowuid(from_kuid_munged(cred->user_ns, cred->suid));
if (!(retval = put_user(ruid, ruidp)) &&
!(retval = put_user(euid, euidp)))
retval = put_user(suid, suidp);
return retval;
}
COMPAT_SYSCALL_DEFINE3(s390_setresgid16, u16, rgid, u16, egid, u16, sgid)
{
return sys_setresgid(low2highgid(rgid), low2highgid(egid),
low2highgid(sgid));
}
COMPAT_SYSCALL_DEFINE3(s390_getresgid16, u16 __user *, rgidp,
u16 __user *, egidp, u16 __user *, sgidp)
{
const struct cred *cred = current_cred();
int retval;
u16 rgid, egid, sgid;
rgid = high2lowgid(from_kgid_munged(cred->user_ns, cred->gid));
egid = high2lowgid(from_kgid_munged(cred->user_ns, cred->egid));
sgid = high2lowgid(from_kgid_munged(cred->user_ns, cred->sgid));
if (!(retval = put_user(rgid, rgidp)) &&
!(retval = put_user(egid, egidp)))
retval = put_user(sgid, sgidp);
return retval;
}
COMPAT_SYSCALL_DEFINE1(s390_setfsuid16, u16, uid)
{
return sys_setfsuid(low2highuid(uid));
}
COMPAT_SYSCALL_DEFINE1(s390_setfsgid16, u16, gid)
{
return sys_setfsgid(low2highgid(gid));
}
static int groups16_to_user(u16 __user *grouplist, struct group_info *group_info)
{
struct user_namespace *user_ns = current_user_ns();
int i;
u16 group;
kgid_t kgid;
for (i = 0; i < group_info->ngroups; i++) {
kgid = group_info->gid[i];
group = (u16)from_kgid_munged(user_ns, kgid);
if (put_user(group, grouplist+i))
return -EFAULT;
}
return 0;
}
static int groups16_from_user(struct group_info *group_info, u16 __user *grouplist)
{
struct user_namespace *user_ns = current_user_ns();
int i;
u16 group;
kgid_t kgid;
for (i = 0; i < group_info->ngroups; i++) {
if (get_user(group, grouplist+i))
return -EFAULT;
kgid = make_kgid(user_ns, (gid_t)group);
if (!gid_valid(kgid))
return -EINVAL;
group_info->gid[i] = kgid;
}
return 0;
}
COMPAT_SYSCALL_DEFINE2(s390_getgroups16, int, gidsetsize, u16 __user *, grouplist)
{
const struct cred *cred = current_cred();
int i;
if (gidsetsize < 0)
return -EINVAL;
get_group_info(cred->group_info);
i = cred->group_info->ngroups;
if (gidsetsize) {
if (i > gidsetsize) {
i = -EINVAL;
goto out;
}
if (groups16_to_user(grouplist, cred->group_info)) {
i = -EFAULT;
goto out;
}
}
out:
put_group_info(cred->group_info);
return i;
}
COMPAT_SYSCALL_DEFINE2(s390_setgroups16, int, gidsetsize, u16 __user *, grouplist)
{
struct group_info *group_info;
int retval;
if (!may_setgroups())
return -EPERM;
if ((unsigned)gidsetsize > NGROUPS_MAX)
return -EINVAL;
group_info = groups_alloc(gidsetsize);
if (!group_info)
return -ENOMEM;
retval = groups16_from_user(group_info, grouplist);
if (retval) {
put_group_info(group_info);
return retval;
}
groups_sort(group_info);
retval = set_current_groups(group_info);
put_group_info(group_info);
return retval;
}
COMPAT_SYSCALL_DEFINE0(s390_getuid16)
{
return high2lowuid(from_kuid_munged(current_user_ns(), current_uid()));
}
COMPAT_SYSCALL_DEFINE0(s390_geteuid16)
{
return high2lowuid(from_kuid_munged(current_user_ns(), current_euid()));
}
COMPAT_SYSCALL_DEFINE0(s390_getgid16)
{
return high2lowgid(from_kgid_munged(current_user_ns(), current_gid()));
}
COMPAT_SYSCALL_DEFINE0(s390_getegid16)
{
return high2lowgid(from_kgid_munged(current_user_ns(), current_egid()));
}
#ifdef CONFIG_SYSVIPC
COMPAT_SYSCALL_DEFINE5(s390_ipc, uint, call, int, first, compat_ulong_t, second,
compat_ulong_t, third, compat_uptr_t, ptr)
{
if (call >> 16) /* hack for backward compatibility */
return -EINVAL;
return compat_sys_ipc(call, first, second, third, ptr, third);
}
#endif
COMPAT_SYSCALL_DEFINE3(s390_truncate64, const char __user *, path, u32, high, u32, low)
{
return ksys_truncate(path, (unsigned long)high << 32 | low);
}
COMPAT_SYSCALL_DEFINE3(s390_ftruncate64, unsigned int, fd, u32, high, u32, low)
{
return ksys_ftruncate(fd, (unsigned long)high << 32 | low);
}
COMPAT_SYSCALL_DEFINE5(s390_pread64, unsigned int, fd, char __user *, ubuf,
compat_size_t, count, u32, high, u32, low)
{
if ((compat_ssize_t) count < 0)
return -EINVAL;
return ksys_pread64(fd, ubuf, count, (unsigned long)high << 32 | low);
}
COMPAT_SYSCALL_DEFINE5(s390_pwrite64, unsigned int, fd, const char __user *, ubuf,
compat_size_t, count, u32, high, u32, low)
{
if ((compat_ssize_t) count < 0)
return -EINVAL;
return ksys_pwrite64(fd, ubuf, count, (unsigned long)high << 32 | low);
}
COMPAT_SYSCALL_DEFINE4(s390_readahead, int, fd, u32, high, u32, low, s32, count)
{
return ksys_readahead(fd, (unsigned long)high << 32 | low, count);
}
struct stat64_emu31 {
unsigned long long st_dev;
unsigned int __pad1;
#define STAT64_HAS_BROKEN_ST_INO 1
u32 __st_ino;
unsigned int st_mode;
unsigned int st_nlink;
u32 st_uid;
u32 st_gid;
unsigned long long st_rdev;
unsigned int __pad3;
long st_size;
u32 st_blksize;
unsigned char __pad4[4];
u32 __pad5; /* future possible st_blocks high bits */
u32 st_blocks; /* Number 512-byte blocks allocated. */
u32 st_atime;
u32 __pad6;
u32 st_mtime;
u32 __pad7;
u32 st_ctime;
u32 __pad8; /* will be high 32 bits of ctime someday */
unsigned long st_ino;
};
static int cp_stat64(struct stat64_emu31 __user *ubuf, struct kstat *stat)
{
struct stat64_emu31 tmp;
memset(&tmp, 0, sizeof(tmp));
tmp.st_dev = huge_encode_dev(stat->dev);
tmp.st_ino = stat->ino;
tmp.__st_ino = (u32)stat->ino;
tmp.st_mode = stat->mode;
tmp.st_nlink = (unsigned int)stat->nlink;
tmp.st_uid = from_kuid_munged(current_user_ns(), stat->uid);
tmp.st_gid = from_kgid_munged(current_user_ns(), stat->gid);
tmp.st_rdev = huge_encode_dev(stat->rdev);
tmp.st_size = stat->size;
tmp.st_blksize = (u32)stat->blksize;
tmp.st_blocks = (u32)stat->blocks;
tmp.st_atime = (u32)stat->atime.tv_sec;
tmp.st_mtime = (u32)stat->mtime.tv_sec;
tmp.st_ctime = (u32)stat->ctime.tv_sec;
return copy_to_user(ubuf,&tmp,sizeof(tmp)) ? -EFAULT : 0;
}
COMPAT_SYSCALL_DEFINE2(s390_stat64, const char __user *, filename, struct stat64_emu31 __user *, statbuf)
{
struct kstat stat;
int ret = vfs_stat(filename, &stat);
if (!ret)
ret = cp_stat64(statbuf, &stat);
return ret;
}
COMPAT_SYSCALL_DEFINE2(s390_lstat64, const char __user *, filename, struct stat64_emu31 __user *, statbuf)
{
struct kstat stat;
int ret = vfs_lstat(filename, &stat);
if (!ret)
ret = cp_stat64(statbuf, &stat);
return ret;
}
COMPAT_SYSCALL_DEFINE2(s390_fstat64, unsigned int, fd, struct stat64_emu31 __user *, statbuf)
{
struct kstat stat;
int ret = vfs_fstat(fd, &stat);
if (!ret)
ret = cp_stat64(statbuf, &stat);
return ret;
}
COMPAT_SYSCALL_DEFINE4(s390_fstatat64, unsigned int, dfd, const char __user *, filename,
struct stat64_emu31 __user *, statbuf, int, flag)
{
struct kstat stat;
int error;
error = vfs_fstatat(dfd, filename, &stat, flag);
if (error)
return error;
return cp_stat64(statbuf, &stat);
}
/*
* Linux/i386 didn't use to be able to handle more than
* 4 system call parameters, so these system calls used a memory
* block for parameter passing..
*/
struct mmap_arg_struct_emu31 {
compat_ulong_t addr;
compat_ulong_t len;
compat_ulong_t prot;
compat_ulong_t flags;
compat_ulong_t fd;
compat_ulong_t offset;
};
COMPAT_SYSCALL_DEFINE1(s390_old_mmap, struct mmap_arg_struct_emu31 __user *, arg)
{
struct mmap_arg_struct_emu31 a;
if (copy_from_user(&a, arg, sizeof(a)))
return -EFAULT;
if (a.offset & ~PAGE_MASK)
return -EINVAL;
return ksys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd,
a.offset >> PAGE_SHIFT);
}
COMPAT_SYSCALL_DEFINE1(s390_mmap2, struct mmap_arg_struct_emu31 __user *, arg)
{
struct mmap_arg_struct_emu31 a;
if (copy_from_user(&a, arg, sizeof(a)))
return -EFAULT;
return ksys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset);
}
COMPAT_SYSCALL_DEFINE3(s390_read, unsigned int, fd, char __user *, buf, compat_size_t, count)
{
if ((compat_ssize_t) count < 0)
return -EINVAL;
return ksys_read(fd, buf, count);
}
COMPAT_SYSCALL_DEFINE3(s390_write, unsigned int, fd, const char __user *, buf, compat_size_t, count)
{
if ((compat_ssize_t) count < 0)
return -EINVAL;
return ksys_write(fd, buf, count);
}
/*
* 31 bit emulation wrapper functions for sys_fadvise64/fadvise64_64.
* These need to rewrite the advise values for POSIX_FADV_{DONTNEED,NOREUSE}
* because the 31 bit values differ from the 64 bit values.
*/
COMPAT_SYSCALL_DEFINE5(s390_fadvise64, int, fd, u32, high, u32, low, compat_size_t, len, int, advise)
{
if (advise == 4)
advise = POSIX_FADV_DONTNEED;
else if (advise == 5)
advise = POSIX_FADV_NOREUSE;
return ksys_fadvise64_64(fd, (unsigned long)high << 32 | low, len,
advise);
}
struct fadvise64_64_args {
int fd;
long long offset;
long long len;
int advice;
};
COMPAT_SYSCALL_DEFINE1(s390_fadvise64_64, struct fadvise64_64_args __user *, args)
{
struct fadvise64_64_args a;
if ( copy_from_user(&a, args, sizeof(a)) )
return -EFAULT;
if (a.advice == 4)
a.advice = POSIX_FADV_DONTNEED;
else if (a.advice == 5)
a.advice = POSIX_FADV_NOREUSE;
return ksys_fadvise64_64(a.fd, a.offset, a.len, a.advice);
}
COMPAT_SYSCALL_DEFINE6(s390_sync_file_range, int, fd, u32, offhigh, u32, offlow,
u32, nhigh, u32, nlow, unsigned int, flags)
{
return ksys_sync_file_range(fd, ((loff_t)offhigh << 32) + offlow,
((u64)nhigh << 32) + nlow, flags);
}
COMPAT_SYSCALL_DEFINE6(s390_fallocate, int, fd, int, mode, u32, offhigh, u32, offlow,
u32, lenhigh, u32, lenlow)
{
return ksys_fallocate(fd, mode, ((loff_t)offhigh << 32) + offlow,
((u64)lenhigh << 32) + lenlow);
}
+130
View File
@@ -0,0 +1,130 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _ASM_S390X_S390_H
#define _ASM_S390X_S390_H
#include <linux/compat.h>
#include <linux/socket.h>
#include <linux/syscalls.h>
/* Macro that masks the high order bit of an 32 bit pointer and converts it*/
/* to a 64 bit pointer */
#define A(__x) ((unsigned long)((__x) & 0x7FFFFFFFUL))
#define AA(__x) \
((unsigned long)(__x))
/* Now 32bit compatibility types */
struct ipc_kludge_32 {
__u32 msgp; /* pointer */
__s32 msgtyp;
};
/* asm/sigcontext.h */
typedef union
{
__u64 d;
__u32 f;
} freg_t32;
typedef struct
{
unsigned int fpc;
unsigned int pad;
freg_t32 fprs[__NUM_FPRS];
} _s390_fp_regs32;
typedef struct
{
__u32 mask;
__u32 addr;
} _psw_t32 __attribute__ ((aligned(8)));
typedef struct
{
_psw_t32 psw;
__u32 gprs[__NUM_GPRS];
__u32 acrs[__NUM_ACRS];
} _s390_regs_common32;
typedef struct
{
_s390_regs_common32 regs;
_s390_fp_regs32 fpregs;
} _sigregs32;
typedef struct
{
__u32 gprs_high[__NUM_GPRS];
__u64 vxrs_low[__NUM_VXRS_LOW];
__vector128 vxrs_high[__NUM_VXRS_HIGH];
__u8 __reserved[128];
} _sigregs_ext32;
#define _SIGCONTEXT_NSIG32 64
#define _SIGCONTEXT_NSIG_BPW32 32
#define __SIGNAL_FRAMESIZE32 96
#define _SIGMASK_COPY_SIZE32 (sizeof(u32)*2)
struct sigcontext32
{
__u32 oldmask[_COMPAT_NSIG_WORDS];
__u32 sregs; /* pointer */
};
/* asm/signal.h */
/* asm/ucontext.h */
struct ucontext32 {
__u32 uc_flags;
__u32 uc_link; /* pointer */
compat_stack_t uc_stack;
_sigregs32 uc_mcontext;
compat_sigset_t uc_sigmask;
/* Allow for uc_sigmask growth. Glibc uses a 1024-bit sigset_t. */
unsigned char __unused[128 - sizeof(compat_sigset_t)];
_sigregs_ext32 uc_mcontext_ext;
};
struct stat64_emu31;
struct mmap_arg_struct_emu31;
struct fadvise64_64_args;
long compat_sys_s390_chown16(const char __user *filename, u16 user, u16 group);
long compat_sys_s390_lchown16(const char __user *filename, u16 user, u16 group);
long compat_sys_s390_fchown16(unsigned int fd, u16 user, u16 group);
long compat_sys_s390_setregid16(u16 rgid, u16 egid);
long compat_sys_s390_setgid16(u16 gid);
long compat_sys_s390_setreuid16(u16 ruid, u16 euid);
long compat_sys_s390_setuid16(u16 uid);
long compat_sys_s390_setresuid16(u16 ruid, u16 euid, u16 suid);
long compat_sys_s390_getresuid16(u16 __user *ruid, u16 __user *euid, u16 __user *suid);
long compat_sys_s390_setresgid16(u16 rgid, u16 egid, u16 sgid);
long compat_sys_s390_getresgid16(u16 __user *rgid, u16 __user *egid, u16 __user *sgid);
long compat_sys_s390_setfsuid16(u16 uid);
long compat_sys_s390_setfsgid16(u16 gid);
long compat_sys_s390_getgroups16(int gidsetsize, u16 __user *grouplist);
long compat_sys_s390_setgroups16(int gidsetsize, u16 __user *grouplist);
long compat_sys_s390_getuid16(void);
long compat_sys_s390_geteuid16(void);
long compat_sys_s390_getgid16(void);
long compat_sys_s390_getegid16(void);
long compat_sys_s390_truncate64(const char __user *path, u32 high, u32 low);
long compat_sys_s390_ftruncate64(unsigned int fd, u32 high, u32 low);
long compat_sys_s390_pread64(unsigned int fd, char __user *ubuf, compat_size_t count, u32 high, u32 low);
long compat_sys_s390_pwrite64(unsigned int fd, const char __user *ubuf, compat_size_t count, u32 high, u32 low);
long compat_sys_s390_readahead(int fd, u32 high, u32 low, s32 count);
long compat_sys_s390_stat64(const char __user *filename, struct stat64_emu31 __user *statbuf);
long compat_sys_s390_lstat64(const char __user *filename, struct stat64_emu31 __user *statbuf);
long compat_sys_s390_fstat64(unsigned int fd, struct stat64_emu31 __user *statbuf);
long compat_sys_s390_fstatat64(unsigned int dfd, const char __user *filename, struct stat64_emu31 __user *statbuf, int flag);
long compat_sys_s390_old_mmap(struct mmap_arg_struct_emu31 __user *arg);
long compat_sys_s390_mmap2(struct mmap_arg_struct_emu31 __user *arg);
long compat_sys_s390_read(unsigned int fd, char __user * buf, compat_size_t count);
long compat_sys_s390_write(unsigned int fd, const char __user * buf, compat_size_t count);
long compat_sys_s390_fadvise64(int fd, u32 high, u32 low, compat_size_t len, int advise);
long compat_sys_s390_fadvise64_64(struct fadvise64_64_args __user *args);
long compat_sys_s390_sync_file_range(int fd, u32 offhigh, u32 offlow, u32 nhigh, u32 nlow, unsigned int flags);
long compat_sys_s390_fallocate(int fd, int mode, u32 offhigh, u32 offlow, u32 lenhigh, u32 lenlow);
long compat_sys_sigreturn(void);
long compat_sys_rt_sigreturn(void);
#endif /* _ASM_S390X_S390_H */
+64
View File
@@ -0,0 +1,64 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _PTRACE32_H
#define _PTRACE32_H
#include <asm/ptrace.h> /* needed for NUM_CR_WORDS */
#include "compat_linux.h" /* needed for psw_compat_t */
struct compat_per_struct_kernel {
__u32 cr9; /* PER control bits */
__u32 cr10; /* PER starting address */
__u32 cr11; /* PER ending address */
__u32 bits; /* Obsolete software bits */
__u32 starting_addr; /* User specified start address */
__u32 ending_addr; /* User specified end address */
__u16 perc_atmid; /* PER trap ATMID */
__u32 address; /* PER trap instruction address */
__u8 access_id; /* PER trap access identification */
};
struct compat_user_regs_struct
{
psw_compat_t psw;
u32 gprs[NUM_GPRS];
u32 acrs[NUM_ACRS];
u32 orig_gpr2;
/* nb: there's a 4-byte hole here */
s390_fp_regs fp_regs;
/*
* These per registers are in here so that gdb can modify them
* itself as there is no "official" ptrace interface for hardware
* watchpoints. This is the way intel does it.
*/
struct compat_per_struct_kernel per_info;
u32 ieee_instruction_pointer; /* obsolete, always 0 */
};
struct compat_user {
/* We start with the registers, to mimic the way that "memory"
is returned from the ptrace(3,...) function. */
struct compat_user_regs_struct regs;
/* The rest of this junk is to help gdb figure out what goes where */
u32 u_tsize; /* Text segment size (pages). */
u32 u_dsize; /* Data segment size (pages). */
u32 u_ssize; /* Stack segment size (pages). */
u32 start_code; /* Starting virtual address of text. */
u32 start_stack; /* Starting virtual address of stack area.
This is actually the bottom of the stack,
the top of the stack is always found in the
esp register. */
s32 signal; /* Signal that caused the core dump. */
u32 u_ar0; /* Used by gdb to help find the values for */
/* the registers. */
u32 magic; /* To uniquely identify a core file */
char u_comm[32]; /* User command that was responsible */
};
typedef struct
{
__u32 len;
__u32 kernel_addr;
__u32 process_addr;
} compat_ptrace_area;
#endif /* _PTRACE32_H */
+428
View File
@@ -0,0 +1,428 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2000, 2006
* Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
* Gerhard Tonn (ton@de.ibm.com)
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* 1997-11-28 Modified for POSIX.1b signals by Richard Henderson
*/
#include <linux/compat.h>
#include <linux/sched.h>
#include <linux/sched/task_stack.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/kernel.h>
#include <linux/signal.h>
#include <linux/errno.h>
#include <linux/wait.h>
#include <linux/ptrace.h>
#include <linux/unistd.h>
#include <linux/stddef.h>
#include <linux/tty.h>
#include <linux/personality.h>
#include <linux/binfmts.h>
#include <asm/ucontext.h>
#include <linux/uaccess.h>
#include <asm/lowcore.h>
#include <asm/switch_to.h>
#include "compat_linux.h"
#include "compat_ptrace.h"
#include "entry.h"
typedef struct
{
__u8 callee_used_stack[__SIGNAL_FRAMESIZE32];
struct sigcontext32 sc;
_sigregs32 sregs;
int signo;
_sigregs_ext32 sregs_ext;
__u16 svc_insn; /* Offset of svc_insn is NOT fixed! */
} sigframe32;
typedef struct
{
__u8 callee_used_stack[__SIGNAL_FRAMESIZE32];
__u16 svc_insn;
compat_siginfo_t info;
struct ucontext32 uc;
} rt_sigframe32;
/* Store registers needed to create the signal frame */
static void store_sigregs(void)
{
save_access_regs(current->thread.acrs);
save_fpu_regs();
}
/* Load registers after signal return */
static void load_sigregs(void)
{
restore_access_regs(current->thread.acrs);
}
static int save_sigregs32(struct pt_regs *regs, _sigregs32 __user *sregs)
{
_sigregs32 user_sregs;
int i;
user_sregs.regs.psw.mask = (__u32)(regs->psw.mask >> 32);
user_sregs.regs.psw.mask &= PSW32_MASK_USER | PSW32_MASK_RI;
user_sregs.regs.psw.mask |= PSW32_USER_BITS;
user_sregs.regs.psw.addr = (__u32) regs->psw.addr |
(__u32)(regs->psw.mask & PSW_MASK_BA);
for (i = 0; i < NUM_GPRS; i++)
user_sregs.regs.gprs[i] = (__u32) regs->gprs[i];
memcpy(&user_sregs.regs.acrs, current->thread.acrs,
sizeof(user_sregs.regs.acrs));
fpregs_store((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs32)))
return -EFAULT;
return 0;
}
static int restore_sigregs32(struct pt_regs *regs,_sigregs32 __user *sregs)
{
_sigregs32 user_sregs;
int i;
/* Alwys make any pending restarted system call return -EINTR */
current->restart_block.fn = do_no_restart_syscall;
if (__copy_from_user(&user_sregs, &sregs->regs, sizeof(user_sregs)))
return -EFAULT;
if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW32_MASK_RI))
return -EINVAL;
/* Test the floating-point-control word. */
if (test_fp_ctl(user_sregs.fpregs.fpc))
return -EINVAL;
/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
regs->psw.mask = (regs->psw.mask & ~(PSW_MASK_USER | PSW_MASK_RI)) |
(__u64)(user_sregs.regs.psw.mask & PSW32_MASK_USER) << 32 |
(__u64)(user_sregs.regs.psw.mask & PSW32_MASK_RI) << 32 |
(__u64)(user_sregs.regs.psw.addr & PSW32_ADDR_AMODE);
/* Check for invalid user address space control. */
if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_HOME)
regs->psw.mask = PSW_ASC_PRIMARY |
(regs->psw.mask & ~PSW_MASK_ASC);
regs->psw.addr = (__u64)(user_sregs.regs.psw.addr & PSW32_ADDR_INSN);
for (i = 0; i < NUM_GPRS; i++)
regs->gprs[i] = (__u64) user_sregs.regs.gprs[i];
memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
sizeof(current->thread.acrs));
fpregs_load((_s390_fp_regs *) &user_sregs.fpregs, &current->thread.fpu);
clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
return 0;
}
static int save_sigregs_ext32(struct pt_regs *regs,
_sigregs_ext32 __user *sregs_ext)
{
__u32 gprs_high[NUM_GPRS];
__u64 vxrs[__NUM_VXRS_LOW];
int i;
/* Save high gprs to signal stack */
for (i = 0; i < NUM_GPRS; i++)
gprs_high[i] = regs->gprs[i] >> 32;
if (__copy_to_user(&sregs_ext->gprs_high, &gprs_high,
sizeof(sregs_ext->gprs_high)))
return -EFAULT;
/* Save vector registers to signal stack */
if (MACHINE_HAS_VX) {
for (i = 0; i < __NUM_VXRS_LOW; i++)
vxrs[i] = *((__u64 *)(current->thread.fpu.vxrs + i) + 1);
if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
sizeof(sregs_ext->vxrs_low)) ||
__copy_to_user(&sregs_ext->vxrs_high,
current->thread.fpu.vxrs + __NUM_VXRS_LOW,
sizeof(sregs_ext->vxrs_high)))
return -EFAULT;
}
return 0;
}
static int restore_sigregs_ext32(struct pt_regs *regs,
_sigregs_ext32 __user *sregs_ext)
{
__u32 gprs_high[NUM_GPRS];
__u64 vxrs[__NUM_VXRS_LOW];
int i;
/* Restore high gprs from signal stack */
if (__copy_from_user(&gprs_high, &sregs_ext->gprs_high,
sizeof(sregs_ext->gprs_high)))
return -EFAULT;
for (i = 0; i < NUM_GPRS; i++)
*(__u32 *)&regs->gprs[i] = gprs_high[i];
/* Restore vector registers from signal stack */
if (MACHINE_HAS_VX) {
if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
sizeof(sregs_ext->vxrs_low)) ||
__copy_from_user(current->thread.fpu.vxrs + __NUM_VXRS_LOW,
&sregs_ext->vxrs_high,
sizeof(sregs_ext->vxrs_high)))
return -EFAULT;
for (i = 0; i < __NUM_VXRS_LOW; i++)
*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
}
return 0;
}
COMPAT_SYSCALL_DEFINE0(sigreturn)
{
struct pt_regs *regs = task_pt_regs(current);
sigframe32 __user *frame = (sigframe32 __user *)regs->gprs[15];
sigset_t set;
if (get_compat_sigset(&set, (compat_sigset_t __user *)frame->sc.oldmask))
goto badframe;
set_current_blocked(&set);
save_fpu_regs();
if (restore_sigregs32(regs, &frame->sregs))
goto badframe;
if (restore_sigregs_ext32(regs, &frame->sregs_ext))
goto badframe;
load_sigregs();
return regs->gprs[2];
badframe:
force_sig(SIGSEGV, current);
return 0;
}
COMPAT_SYSCALL_DEFINE0(rt_sigreturn)
{
struct pt_regs *regs = task_pt_regs(current);
rt_sigframe32 __user *frame = (rt_sigframe32 __user *)regs->gprs[15];
sigset_t set;
if (get_compat_sigset(&set, &frame->uc.uc_sigmask))
goto badframe;
set_current_blocked(&set);
if (compat_restore_altstack(&frame->uc.uc_stack))
goto badframe;
save_fpu_regs();
if (restore_sigregs32(regs, &frame->uc.uc_mcontext))
goto badframe;
if (restore_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
goto badframe;
load_sigregs();
return regs->gprs[2];
badframe:
force_sig(SIGSEGV, current);
return 0;
}
/*
* Set up a signal frame.
*/
/*
* Determine which stack to use..
*/
static inline void __user *
get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size)
{
unsigned long sp;
/* Default to using normal stack */
sp = (unsigned long) A(regs->gprs[15]);
/* Overflow on alternate signal stack gives SIGSEGV. */
if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL))
return (void __user *) -1UL;
/* This is the X/Open sanctioned signal stack switching. */
if (ka->sa.sa_flags & SA_ONSTACK) {
if (! sas_ss_flags(sp))
sp = current->sas_ss_sp + current->sas_ss_size;
}
return (void __user *)((sp - frame_size) & -8ul);
}
static int setup_frame32(struct ksignal *ksig, sigset_t *set,
struct pt_regs *regs)
{
int sig = ksig->sig;
sigframe32 __user *frame;
unsigned long restorer;
size_t frame_size;
/*
* gprs_high are always present for 31-bit compat tasks.
* The space for vector registers is only allocated if
* the machine supports it
*/
frame_size = sizeof(*frame) - sizeof(frame->sregs_ext.__reserved);
if (!MACHINE_HAS_VX)
frame_size -= sizeof(frame->sregs_ext.vxrs_low) +
sizeof(frame->sregs_ext.vxrs_high);
frame = get_sigframe(&ksig->ka, regs, frame_size);
if (frame == (void __user *) -1UL)
return -EFAULT;
/* Set up backchain. */
if (__put_user(regs->gprs[15], (unsigned int __user *) frame))
return -EFAULT;
/* Create struct sigcontext32 on the signal stack */
if (put_compat_sigset((compat_sigset_t __user *)frame->sc.oldmask,
set, sizeof(compat_sigset_t)))
return -EFAULT;
if (__put_user(ptr_to_compat(&frame->sregs), &frame->sc.sregs))
return -EFAULT;
/* Store registers needed to create the signal frame */
store_sigregs();
/* Create _sigregs32 on the signal stack */
if (save_sigregs32(regs, &frame->sregs))
return -EFAULT;
/* Place signal number on stack to allow backtrace from handler. */
if (__put_user(regs->gprs[2], (int __force __user *) &frame->signo))
return -EFAULT;
/* Create _sigregs_ext32 on the signal stack */
if (save_sigregs_ext32(regs, &frame->sregs_ext))
return -EFAULT;
/* Set up to return from userspace. If provided, use a stub
already in userspace. */
if (ksig->ka.sa.sa_flags & SA_RESTORER) {
restorer = (unsigned long __force)
ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
} else {
/* Signal frames without vectors registers are short ! */
__u16 __user *svc = (void __user *) frame + frame_size - 2;
if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
return -EFAULT;
restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE;
}
/* Set up registers for signal handler */
regs->gprs[14] = restorer;
regs->gprs[15] = (__force __u64) frame;
/* Force 31 bit amode and default user address space control. */
regs->psw.mask = PSW_MASK_BA |
(PSW_USER_BITS & PSW_MASK_ASC) |
(regs->psw.mask & ~PSW_MASK_ASC);
regs->psw.addr = (__force __u64) ksig->ka.sa.sa_handler;
regs->gprs[2] = sig;
regs->gprs[3] = (__force __u64) &frame->sc;
/* We forgot to include these in the sigcontext.
To avoid breaking binary compatibility, they are passed as args. */
if (sig == SIGSEGV || sig == SIGBUS || sig == SIGILL ||
sig == SIGTRAP || sig == SIGFPE) {
/* set extra registers only for synchronous signals */
regs->gprs[4] = regs->int_code & 127;
regs->gprs[5] = regs->int_parm_long;
regs->gprs[6] = current->thread.last_break;
}
return 0;
}
static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set,
struct pt_regs *regs)
{
rt_sigframe32 __user *frame;
unsigned long restorer;
size_t frame_size;
u32 uc_flags;
frame_size = sizeof(*frame) -
sizeof(frame->uc.uc_mcontext_ext.__reserved);
/*
* gprs_high are always present for 31-bit compat tasks.
* The space for vector registers is only allocated if
* the machine supports it
*/
uc_flags = UC_GPRS_HIGH;
if (MACHINE_HAS_VX) {
uc_flags |= UC_VXRS;
} else
frame_size -= sizeof(frame->uc.uc_mcontext_ext.vxrs_low) +
sizeof(frame->uc.uc_mcontext_ext.vxrs_high);
frame = get_sigframe(&ksig->ka, regs, frame_size);
if (frame == (void __user *) -1UL)
return -EFAULT;
/* Set up backchain. */
if (__put_user(regs->gprs[15], (unsigned int __force __user *) frame))
return -EFAULT;
/* Set up to return from userspace. If provided, use a stub
already in userspace. */
if (ksig->ka.sa.sa_flags & SA_RESTORER) {
restorer = (unsigned long __force)
ksig->ka.sa.sa_restorer | PSW32_ADDR_AMODE;
} else {
__u16 __user *svc = &frame->svc_insn;
if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc))
return -EFAULT;
restorer = (unsigned long __force) svc | PSW32_ADDR_AMODE;
}
/* Create siginfo on the signal stack */
if (copy_siginfo_to_user32(&frame->info, &ksig->info))
return -EFAULT;
/* Store registers needed to create the signal frame */
store_sigregs();
/* Create ucontext on the signal stack. */
if (__put_user(uc_flags, &frame->uc.uc_flags) ||
__put_user(0, &frame->uc.uc_link) ||
__compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]) ||
save_sigregs32(regs, &frame->uc.uc_mcontext) ||
put_compat_sigset(&frame->uc.uc_sigmask, set, sizeof(compat_sigset_t)) ||
save_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext))
return -EFAULT;
/* Set up registers for signal handler */
regs->gprs[14] = restorer;
regs->gprs[15] = (__force __u64) frame;
/* Force 31 bit amode and default user address space control. */
regs->psw.mask = PSW_MASK_BA |
(PSW_USER_BITS & PSW_MASK_ASC) |
(regs->psw.mask & ~PSW_MASK_ASC);
regs->psw.addr = (__u64 __force) ksig->ka.sa.sa_handler;
regs->gprs[2] = ksig->sig;
regs->gprs[3] = (__force __u64) &frame->info;
regs->gprs[4] = (__force __u64) &frame->uc;
regs->gprs[5] = current->thread.last_break;
return 0;
}
/*
* OK, we're invoking a handler
*/
void handle_signal32(struct ksignal *ksig, sigset_t *oldset,
struct pt_regs *regs)
{
int ret;
/* Set up the stack frame */
if (ksig->ka.sa.sa_flags & SA_SIGINFO)
ret = setup_rt_frame32(ksig, oldset, regs);
else
ret = setup_frame32(ksig, oldset, regs);
signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLE_STEP));
}
+186
View File
@@ -0,0 +1,186 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Compat system call wrappers.
*
* Copyright IBM Corp. 2014
*/
#include <linux/syscalls.h>
#include <linux/compat.h>
#include "entry.h"
#define COMPAT_SYSCALL_WRAP1(name, ...) \
COMPAT_SYSCALL_WRAPx(1, _##name, __VA_ARGS__)
#define COMPAT_SYSCALL_WRAP2(name, ...) \
COMPAT_SYSCALL_WRAPx(2, _##name, __VA_ARGS__)
#define COMPAT_SYSCALL_WRAP3(name, ...) \
COMPAT_SYSCALL_WRAPx(3, _##name, __VA_ARGS__)
#define COMPAT_SYSCALL_WRAP4(name, ...) \
COMPAT_SYSCALL_WRAPx(4, _##name, __VA_ARGS__)
#define COMPAT_SYSCALL_WRAP5(name, ...) \
COMPAT_SYSCALL_WRAPx(5, _##name, __VA_ARGS__)
#define COMPAT_SYSCALL_WRAP6(name, ...) \
COMPAT_SYSCALL_WRAPx(6, _##name, __VA_ARGS__)
#define __SC_COMPAT_TYPE(t, a) \
__typeof(__builtin_choose_expr(sizeof(t) > 4, 0L, (t)0)) a
#define __SC_COMPAT_CAST(t, a) \
({ \
long __ReS = a; \
\
BUILD_BUG_ON((sizeof(t) > 4) && !__TYPE_IS_L(t) && \
!__TYPE_IS_UL(t) && !__TYPE_IS_PTR(t)); \
if (__TYPE_IS_L(t)) \
__ReS = (s32)a; \
if (__TYPE_IS_UL(t)) \
__ReS = (u32)a; \
if (__TYPE_IS_PTR(t)) \
__ReS = a & 0x7fffffff; \
(t)__ReS; \
})
/*
* The COMPAT_SYSCALL_WRAP macro generates system call wrappers to be used by
* compat tasks. These wrappers will only be used for system calls where only
* the system call arguments need sign or zero extension or zeroing of the upper
* 33 bits of pointers.
* Note: since the wrapper function will afterwards call a system call which
* again performs zero and sign extension for all system call arguments with
* a size of less than eight bytes, these compat wrappers only touch those
* system call arguments with a size of eight bytes ((unsigned) long and
* pointers). Zero and sign extension for e.g. int parameters will be done by
* the regular system call wrappers.
*/
#define COMPAT_SYSCALL_WRAPx(x, name, ...) \
asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)); \
asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\
asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__)) \
{ \
return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__)); \
}
COMPAT_SYSCALL_WRAP2(creat, const char __user *, pathname, umode_t, mode);
COMPAT_SYSCALL_WRAP2(link, const char __user *, oldname, const char __user *, newname);
COMPAT_SYSCALL_WRAP1(unlink, const char __user *, pathname);
COMPAT_SYSCALL_WRAP1(chdir, const char __user *, filename);
COMPAT_SYSCALL_WRAP3(mknod, const char __user *, filename, umode_t, mode, unsigned, dev);
COMPAT_SYSCALL_WRAP2(chmod, const char __user *, filename, umode_t, mode);
COMPAT_SYSCALL_WRAP1(oldumount, char __user *, name);
COMPAT_SYSCALL_WRAP2(access, const char __user *, filename, int, mode);
COMPAT_SYSCALL_WRAP2(rename, const char __user *, oldname, const char __user *, newname);
COMPAT_SYSCALL_WRAP2(mkdir, const char __user *, pathname, umode_t, mode);
COMPAT_SYSCALL_WRAP1(rmdir, const char __user *, pathname);
COMPAT_SYSCALL_WRAP1(pipe, int __user *, fildes);
COMPAT_SYSCALL_WRAP1(brk, unsigned long, brk);
COMPAT_SYSCALL_WRAP2(signal, int, sig, __sighandler_t, handler);
COMPAT_SYSCALL_WRAP1(acct, const char __user *, name);
COMPAT_SYSCALL_WRAP2(umount, char __user *, name, int, flags);
COMPAT_SYSCALL_WRAP1(chroot, const char __user *, filename);
COMPAT_SYSCALL_WRAP3(sigsuspend, int, unused1, int, unused2, old_sigset_t, mask);
COMPAT_SYSCALL_WRAP2(sethostname, char __user *, name, int, len);
COMPAT_SYSCALL_WRAP2(symlink, const char __user *, old, const char __user *, new);
COMPAT_SYSCALL_WRAP3(readlink, const char __user *, path, char __user *, buf, int, bufsiz);
COMPAT_SYSCALL_WRAP1(uselib, const char __user *, library);
COMPAT_SYSCALL_WRAP2(swapon, const char __user *, specialfile, int, swap_flags);
COMPAT_SYSCALL_WRAP4(reboot, int, magic1, int, magic2, unsigned int, cmd, void __user *, arg);
COMPAT_SYSCALL_WRAP2(munmap, unsigned long, addr, size_t, len);
COMPAT_SYSCALL_WRAP3(syslog, int, type, char __user *, buf, int, len);
COMPAT_SYSCALL_WRAP1(swapoff, const char __user *, specialfile);
COMPAT_SYSCALL_WRAP2(setdomainname, char __user *, name, int, len);
COMPAT_SYSCALL_WRAP1(newuname, struct new_utsname __user *, name);
COMPAT_SYSCALL_WRAP3(mprotect, unsigned long, start, size_t, len, unsigned long, prot);
COMPAT_SYSCALL_WRAP3(init_module, void __user *, umod, unsigned long, len, const char __user *, uargs);
COMPAT_SYSCALL_WRAP2(delete_module, const char __user *, name_user, unsigned int, flags);
COMPAT_SYSCALL_WRAP4(quotactl, unsigned int, cmd, const char __user *, special, qid_t, id, void __user *, addr);
COMPAT_SYSCALL_WRAP2(bdflush, int, func, long, data);
COMPAT_SYSCALL_WRAP3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2);
COMPAT_SYSCALL_WRAP5(llseek, unsigned int, fd, unsigned long, high, unsigned long, low, loff_t __user *, result, unsigned int, whence);
COMPAT_SYSCALL_WRAP3(msync, unsigned long, start, size_t, len, int, flags);
COMPAT_SYSCALL_WRAP2(mlock, unsigned long, start, size_t, len);
COMPAT_SYSCALL_WRAP2(munlock, unsigned long, start, size_t, len);
COMPAT_SYSCALL_WRAP2(sched_setparam, pid_t, pid, struct sched_param __user *, param);
COMPAT_SYSCALL_WRAP2(sched_getparam, pid_t, pid, struct sched_param __user *, param);
COMPAT_SYSCALL_WRAP3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param);
COMPAT_SYSCALL_WRAP5(mremap, unsigned long, addr, unsigned long, old_len, unsigned long, new_len, unsigned long, flags, unsigned long, new_addr);
COMPAT_SYSCALL_WRAP3(poll, struct pollfd __user *, ufds, unsigned int, nfds, int, timeout);
COMPAT_SYSCALL_WRAP5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5);
COMPAT_SYSCALL_WRAP2(getcwd, char __user *, buf, unsigned long, size);
COMPAT_SYSCALL_WRAP2(capget, cap_user_header_t, header, cap_user_data_t, dataptr);
COMPAT_SYSCALL_WRAP2(capset, cap_user_header_t, header, const cap_user_data_t, data);
COMPAT_SYSCALL_WRAP3(lchown, const char __user *, filename, uid_t, user, gid_t, group);
COMPAT_SYSCALL_WRAP2(getgroups, int, gidsetsize, gid_t __user *, grouplist);
COMPAT_SYSCALL_WRAP2(setgroups, int, gidsetsize, gid_t __user *, grouplist);
COMPAT_SYSCALL_WRAP3(getresuid, uid_t __user *, ruid, uid_t __user *, euid, uid_t __user *, suid);
COMPAT_SYSCALL_WRAP3(getresgid, gid_t __user *, rgid, gid_t __user *, egid, gid_t __user *, sgid);
COMPAT_SYSCALL_WRAP3(chown, const char __user *, filename, uid_t, user, gid_t, group);
COMPAT_SYSCALL_WRAP2(pivot_root, const char __user *, new_root, const char __user *, put_old);
COMPAT_SYSCALL_WRAP3(mincore, unsigned long, start, size_t, len, unsigned char __user *, vec);
COMPAT_SYSCALL_WRAP3(madvise, unsigned long, start, size_t, len, int, behavior);
COMPAT_SYSCALL_WRAP5(setxattr, const char __user *, path, const char __user *, name, const void __user *, value, size_t, size, int, flags);
COMPAT_SYSCALL_WRAP5(lsetxattr, const char __user *, path, const char __user *, name, const void __user *, value, size_t, size, int, flags);
COMPAT_SYSCALL_WRAP5(fsetxattr, int, fd, const char __user *, name, const void __user *, value, size_t, size, int, flags);
COMPAT_SYSCALL_WRAP3(getdents64, unsigned int, fd, struct linux_dirent64 __user *, dirent, unsigned int, count);
COMPAT_SYSCALL_WRAP4(getxattr, const char __user *, path, const char __user *, name, void __user *, value, size_t, size);
COMPAT_SYSCALL_WRAP4(lgetxattr, const char __user *, path, const char __user *, name, void __user *, value, size_t, size);
COMPAT_SYSCALL_WRAP4(fgetxattr, int, fd, const char __user *, name, void __user *, value, size_t, size);
COMPAT_SYSCALL_WRAP3(listxattr, const char __user *, path, char __user *, list, size_t, size);
COMPAT_SYSCALL_WRAP3(llistxattr, const char __user *, path, char __user *, list, size_t, size);
COMPAT_SYSCALL_WRAP3(flistxattr, int, fd, char __user *, list, size_t, size);
COMPAT_SYSCALL_WRAP2(removexattr, const char __user *, path, const char __user *, name);
COMPAT_SYSCALL_WRAP2(lremovexattr, const char __user *, path, const char __user *, name);
COMPAT_SYSCALL_WRAP2(fremovexattr, int, fd, const char __user *, name);
COMPAT_SYSCALL_WRAP1(set_tid_address, int __user *, tidptr);
COMPAT_SYSCALL_WRAP4(epoll_ctl, int, epfd, int, op, int, fd, struct epoll_event __user *, event);
COMPAT_SYSCALL_WRAP4(epoll_wait, int, epfd, struct epoll_event __user *, events, int, maxevents, int, timeout);
COMPAT_SYSCALL_WRAP1(io_destroy, aio_context_t, ctx);
COMPAT_SYSCALL_WRAP3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, struct io_event __user *, result);
COMPAT_SYSCALL_WRAP1(mq_unlink, const char __user *, name);
COMPAT_SYSCALL_WRAP5(add_key, const char __user *, tp, const char __user *, dsc, const void __user *, pld, size_t, len, key_serial_t, id);
COMPAT_SYSCALL_WRAP4(request_key, const char __user *, tp, const char __user *, dsc, const char __user *, info, key_serial_t, id);
COMPAT_SYSCALL_WRAP5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags);
COMPAT_SYSCALL_WRAP3(inotify_add_watch, int, fd, const char __user *, path, u32, mask);
COMPAT_SYSCALL_WRAP3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode);
COMPAT_SYSCALL_WRAP4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, unsigned, dev);
COMPAT_SYSCALL_WRAP5(fchownat, int, dfd, const char __user *, filename, uid_t, user, gid_t, group, int, flag);
COMPAT_SYSCALL_WRAP3(unlinkat, int, dfd, const char __user *, pathname, int, flag);
COMPAT_SYSCALL_WRAP4(renameat, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname);
COMPAT_SYSCALL_WRAP5(linkat, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname, int, flags);
COMPAT_SYSCALL_WRAP3(symlinkat, const char __user *, oldname, int, newdfd, const char __user *, newname);
COMPAT_SYSCALL_WRAP4(readlinkat, int, dfd, const char __user *, path, char __user *, buf, int, bufsiz);
COMPAT_SYSCALL_WRAP3(fchmodat, int, dfd, const char __user *, filename, umode_t, mode);
COMPAT_SYSCALL_WRAP3(faccessat, int, dfd, const char __user *, filename, int, mode);
COMPAT_SYSCALL_WRAP1(unshare, unsigned long, unshare_flags);
COMPAT_SYSCALL_WRAP6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags);
COMPAT_SYSCALL_WRAP4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags);
COMPAT_SYSCALL_WRAP3(getcpu, unsigned __user *, cpu, unsigned __user *, node, struct getcpu_cache __user *, cache);
COMPAT_SYSCALL_WRAP2(pipe2, int __user *, fildes, int, flags);
COMPAT_SYSCALL_WRAP5(perf_event_open, struct perf_event_attr __user *, attr_uptr, pid_t, pid, int, cpu, int, group_fd, unsigned long, flags);
COMPAT_SYSCALL_WRAP5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, unsigned long, tls);
COMPAT_SYSCALL_WRAP4(prlimit64, pid_t, pid, unsigned int, resource, const struct rlimit64 __user *, new_rlim, struct rlimit64 __user *, old_rlim);
COMPAT_SYSCALL_WRAP5(name_to_handle_at, int, dfd, const char __user *, name, struct file_handle __user *, handle, int __user *, mnt_id, int, flag);
COMPAT_SYSCALL_WRAP5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2);
COMPAT_SYSCALL_WRAP3(finit_module, int, fd, const char __user *, uargs, int, flags);
COMPAT_SYSCALL_WRAP3(sched_setattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, flags);
COMPAT_SYSCALL_WRAP4(sched_getattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, size, unsigned int, flags);
COMPAT_SYSCALL_WRAP5(renameat2, int, olddfd, const char __user *, oldname, int, newdfd, const char __user *, newname, unsigned int, flags);
COMPAT_SYSCALL_WRAP3(seccomp, unsigned int, op, unsigned int, flags, const char __user *, uargs)
COMPAT_SYSCALL_WRAP3(getrandom, char __user *, buf, size_t, count, unsigned int, flags)
COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, flags)
COMPAT_SYSCALL_WRAP3(bpf, int, cmd, union bpf_attr *, attr, unsigned int, size);
COMPAT_SYSCALL_WRAP3(s390_pci_mmio_write, const unsigned long, mmio_addr, const void __user *, user_buffer, const size_t, length);
COMPAT_SYSCALL_WRAP3(s390_pci_mmio_read, const unsigned long, mmio_addr, void __user *, user_buffer, const size_t, length);
COMPAT_SYSCALL_WRAP4(socketpair, int, family, int, type, int, protocol, int __user *, usockvec);
COMPAT_SYSCALL_WRAP3(bind, int, fd, struct sockaddr __user *, umyaddr, int, addrlen);
COMPAT_SYSCALL_WRAP3(connect, int, fd, struct sockaddr __user *, uservaddr, int, addrlen);
COMPAT_SYSCALL_WRAP4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr, int __user *, upeer_addrlen, int, flags);
COMPAT_SYSCALL_WRAP3(getsockname, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
COMPAT_SYSCALL_WRAP3(getpeername, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len);
COMPAT_SYSCALL_WRAP6(sendto, int, fd, void __user *, buff, size_t, len, unsigned int, flags, struct sockaddr __user *, addr, int, addr_len);
COMPAT_SYSCALL_WRAP3(mlock2, unsigned long, start, size_t, len, int, flags);
COMPAT_SYSCALL_WRAP6(copy_file_range, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags);
COMPAT_SYSCALL_WRAP2(s390_guarded_storage, int, command, struct gs_cb *, gs_cb);
COMPAT_SYSCALL_WRAP5(statx, int, dfd, const char __user *, path, unsigned, flags, unsigned, mask, struct statx __user *, buffer);
COMPAT_SYSCALL_WRAP4(s390_sthyi, unsigned long, code, void __user *, info, u64 __user *, rc, unsigned long, flags);
COMPAT_SYSCALL_WRAP5(kexec_file_load, int, kernel_fd, int, initrd_fd, unsigned long, cmdline_len, const char __user *, cmdline_ptr, unsigned long, flags)
COMPAT_SYSCALL_WRAP4(rseq, struct rseq __user *, rseq, u32, rseq_len, int, flags, u32, sig)
+112
View File
@@ -0,0 +1,112 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 version
* Copyright IBM Corp. 1999, 2007
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
* Christian Borntraeger (cborntra@de.ibm.com),
*/
#define KMSG_COMPONENT "cpcmd"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/stddef.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <asm/diag.h>
#include <asm/ebcdic.h>
#include <asm/cpcmd.h>
#include <asm/io.h>
static DEFINE_SPINLOCK(cpcmd_lock);
static char cpcmd_buf[241];
static int diag8_noresponse(int cmdlen)
{
register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf;
register unsigned long reg3 asm ("3") = cmdlen;
asm volatile(
" diag %1,%0,0x8\n"
: "+d" (reg3) : "d" (reg2) : "cc");
return reg3;
}
static int diag8_response(int cmdlen, char *response, int *rlen)
{
unsigned long _cmdlen = cmdlen | 0x40000000L;
unsigned long _rlen = *rlen;
register unsigned long reg2 asm ("2") = (addr_t) cpcmd_buf;
register unsigned long reg3 asm ("3") = (addr_t) response;
register unsigned long reg4 asm ("4") = _cmdlen;
register unsigned long reg5 asm ("5") = _rlen;
asm volatile(
" diag %2,%0,0x8\n"
" brc 8,1f\n"
" agr %1,%4\n"
"1:\n"
: "+d" (reg4), "+d" (reg5)
: "d" (reg2), "d" (reg3), "d" (*rlen) : "cc");
*rlen = reg5;
return reg4;
}
/*
* __cpcmd has some restrictions over cpcmd
* - __cpcmd is unlocked and therefore not SMP-safe
*/
int __cpcmd(const char *cmd, char *response, int rlen, int *response_code)
{
int cmdlen;
int rc;
int response_len;
cmdlen = strlen(cmd);
BUG_ON(cmdlen > 240);
memcpy(cpcmd_buf, cmd, cmdlen);
ASCEBC(cpcmd_buf, cmdlen);
diag_stat_inc(DIAG_STAT_X008);
if (response) {
memset(response, 0, rlen);
response_len = rlen;
rc = diag8_response(cmdlen, response, &rlen);
EBCASC(response, response_len);
} else {
rc = diag8_noresponse(cmdlen);
}
if (response_code)
*response_code = rc;
return rlen;
}
EXPORT_SYMBOL(__cpcmd);
int cpcmd(const char *cmd, char *response, int rlen, int *response_code)
{
unsigned long flags;
char *lowbuf;
int len;
if (is_vmalloc_or_module_addr(response)) {
lowbuf = kmalloc(rlen, GFP_KERNEL);
if (!lowbuf) {
pr_warn("The cpcmd kernel function failed to allocate a response buffer\n");
return -ENOMEM;
}
spin_lock_irqsave(&cpcmd_lock, flags);
len = __cpcmd(cmd, lowbuf, rlen, response_code);
spin_unlock_irqrestore(&cpcmd_lock, flags);
memcpy(response, lowbuf, rlen);
kfree(lowbuf);
} else {
spin_lock_irqsave(&cpcmd_lock, flags);
len = __cpcmd(cmd, response, rlen, response_code);
spin_unlock_irqrestore(&cpcmd_lock, flags);
}
return len;
}
EXPORT_SYMBOL(cpcmd);
+711
View File
@@ -0,0 +1,711 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 kdump implementation
*
* Copyright IBM Corp. 2011
* Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com>
*/
#include <linux/crash_dump.h>
#include <asm/lowcore.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/gfp.h>
#include <linux/slab.h>
#include <linux/bootmem.h>
#include <linux/elf.h>
#include <asm/asm-offsets.h>
#include <linux/memblock.h>
#include <asm/os_info.h>
#include <asm/elf.h>
#include <asm/ipl.h>
#include <asm/sclp.h>
#define PTR_ADD(x, y) (((char *) (x)) + ((unsigned long) (y)))
#define PTR_SUB(x, y) (((char *) (x)) - ((unsigned long) (y)))
#define PTR_DIFF(x, y) ((unsigned long)(((char *) (x)) - ((unsigned long) (y))))
static struct memblock_region oldmem_region;
static struct memblock_type oldmem_type = {
.cnt = 1,
.max = 1,
.total_size = 0,
.regions = &oldmem_region,
.name = "oldmem",
};
struct save_area {
struct list_head list;
u64 psw[2];
u64 ctrs[16];
u64 gprs[16];
u32 acrs[16];
u64 fprs[16];
u32 fpc;
u32 prefix;
u64 todpreg;
u64 timer;
u64 todcmp;
u64 vxrs_low[16];
__vector128 vxrs_high[16];
};
static LIST_HEAD(dump_save_areas);
/*
* Allocate a save area
*/
struct save_area * __init save_area_alloc(bool is_boot_cpu)
{
struct save_area *sa;
sa = (void *) memblock_alloc(sizeof(*sa), 8);
if (is_boot_cpu)
list_add(&sa->list, &dump_save_areas);
else
list_add_tail(&sa->list, &dump_save_areas);
return sa;
}
/*
* Return the address of the save area for the boot CPU
*/
struct save_area * __init save_area_boot_cpu(void)
{
return list_first_entry_or_null(&dump_save_areas, struct save_area, list);
}
/*
* Copy CPU registers into the save area
*/
void __init save_area_add_regs(struct save_area *sa, void *regs)
{
struct lowcore *lc;
lc = (struct lowcore *)(regs - __LC_FPREGS_SAVE_AREA);
memcpy(&sa->psw, &lc->psw_save_area, sizeof(sa->psw));
memcpy(&sa->ctrs, &lc->cregs_save_area, sizeof(sa->ctrs));
memcpy(&sa->gprs, &lc->gpregs_save_area, sizeof(sa->gprs));
memcpy(&sa->acrs, &lc->access_regs_save_area, sizeof(sa->acrs));
memcpy(&sa->fprs, &lc->floating_pt_save_area, sizeof(sa->fprs));
memcpy(&sa->fpc, &lc->fpt_creg_save_area, sizeof(sa->fpc));
memcpy(&sa->prefix, &lc->prefixreg_save_area, sizeof(sa->prefix));
memcpy(&sa->todpreg, &lc->tod_progreg_save_area, sizeof(sa->todpreg));
memcpy(&sa->timer, &lc->cpu_timer_save_area, sizeof(sa->timer));
memcpy(&sa->todcmp, &lc->clock_comp_save_area, sizeof(sa->todcmp));
}
/*
* Copy vector registers into the save area
*/
void __init save_area_add_vxrs(struct save_area *sa, __vector128 *vxrs)
{
int i;
/* Copy lower halves of vector registers 0-15 */
for (i = 0; i < 16; i++)
memcpy(&sa->vxrs_low[i], &vxrs[i].u[2], 8);
/* Copy vector registers 16-31 */
memcpy(sa->vxrs_high, vxrs + 16, 16 * sizeof(__vector128));
}
/*
* Return physical address for virtual address
*/
static inline void *load_real_addr(void *addr)
{
unsigned long real_addr;
asm volatile(
" lra %0,0(%1)\n"
" jz 0f\n"
" la %0,0\n"
"0:"
: "=a" (real_addr) : "a" (addr) : "cc");
return (void *)real_addr;
}
/*
* Copy memory of the old, dumped system to a kernel space virtual address
*/
int copy_oldmem_kernel(void *dst, void *src, size_t count)
{
unsigned long from, len;
void *ra;
int rc;
while (count) {
from = __pa(src);
if (!OLDMEM_BASE && from < sclp.hsa_size) {
/* Copy from zfcpdump HSA area */
len = min(count, sclp.hsa_size - from);
rc = memcpy_hsa_kernel(dst, from, len);
if (rc)
return rc;
} else {
/* Check for swapped kdump oldmem areas */
if (OLDMEM_BASE && from - OLDMEM_BASE < OLDMEM_SIZE) {
from -= OLDMEM_BASE;
len = min(count, OLDMEM_SIZE - from);
} else if (OLDMEM_BASE && from < OLDMEM_SIZE) {
len = min(count, OLDMEM_SIZE - from);
from += OLDMEM_BASE;
} else {
len = count;
}
if (is_vmalloc_or_module_addr(dst)) {
ra = load_real_addr(dst);
len = min(PAGE_SIZE - offset_in_page(ra), len);
} else {
ra = dst;
}
if (memcpy_real(ra, (void *) from, len))
return -EFAULT;
}
dst += len;
src += len;
count -= len;
}
return 0;
}
/*
* Copy memory of the old, dumped system to a user space virtual address
*/
static int copy_oldmem_user(void __user *dst, void *src, size_t count)
{
unsigned long from, len;
int rc;
while (count) {
from = __pa(src);
if (!OLDMEM_BASE && from < sclp.hsa_size) {
/* Copy from zfcpdump HSA area */
len = min(count, sclp.hsa_size - from);
rc = memcpy_hsa_user(dst, from, len);
if (rc)
return rc;
} else {
/* Check for swapped kdump oldmem areas */
if (OLDMEM_BASE && from - OLDMEM_BASE < OLDMEM_SIZE) {
from -= OLDMEM_BASE;
len = min(count, OLDMEM_SIZE - from);
} else if (OLDMEM_BASE && from < OLDMEM_SIZE) {
len = min(count, OLDMEM_SIZE - from);
from += OLDMEM_BASE;
} else {
len = count;
}
rc = copy_to_user_real(dst, (void *) from, count);
if (rc)
return rc;
}
dst += len;
src += len;
count -= len;
}
return 0;
}
/*
* Copy one page from "oldmem"
*/
ssize_t copy_oldmem_page(unsigned long pfn, char *buf, size_t csize,
unsigned long offset, int userbuf)
{
void *src;
int rc;
if (!csize)
return 0;
src = (void *) (pfn << PAGE_SHIFT) + offset;
if (userbuf)
rc = copy_oldmem_user((void __force __user *) buf, src, csize);
else
rc = copy_oldmem_kernel((void *) buf, src, csize);
return rc;
}
/*
* Remap "oldmem" for kdump
*
* For the kdump reserved memory this functions performs a swap operation:
* [0 - OLDMEM_SIZE] is mapped to [OLDMEM_BASE - OLDMEM_BASE + OLDMEM_SIZE]
*/
static int remap_oldmem_pfn_range_kdump(struct vm_area_struct *vma,
unsigned long from, unsigned long pfn,
unsigned long size, pgprot_t prot)
{
unsigned long size_old;
int rc;
if (pfn < OLDMEM_SIZE >> PAGE_SHIFT) {
size_old = min(size, OLDMEM_SIZE - (pfn << PAGE_SHIFT));
rc = remap_pfn_range(vma, from,
pfn + (OLDMEM_BASE >> PAGE_SHIFT),
size_old, prot);
if (rc || size == size_old)
return rc;
size -= size_old;
from += size_old;
pfn += size_old >> PAGE_SHIFT;
}
return remap_pfn_range(vma, from, pfn, size, prot);
}
/*
* Remap "oldmem" for zfcpdump
*
* We only map available memory above HSA size. Memory below HSA size
* is read on demand using the copy_oldmem_page() function.
*/
static int remap_oldmem_pfn_range_zfcpdump(struct vm_area_struct *vma,
unsigned long from,
unsigned long pfn,
unsigned long size, pgprot_t prot)
{
unsigned long hsa_end = sclp.hsa_size;
unsigned long size_hsa;
if (pfn < hsa_end >> PAGE_SHIFT) {
size_hsa = min(size, hsa_end - (pfn << PAGE_SHIFT));
if (size == size_hsa)
return 0;
size -= size_hsa;
from += size_hsa;
pfn += size_hsa >> PAGE_SHIFT;
}
return remap_pfn_range(vma, from, pfn, size, prot);
}
/*
* Remap "oldmem" for kdump or zfcpdump
*/
int remap_oldmem_pfn_range(struct vm_area_struct *vma, unsigned long from,
unsigned long pfn, unsigned long size, pgprot_t prot)
{
if (OLDMEM_BASE)
return remap_oldmem_pfn_range_kdump(vma, from, pfn, size, prot);
else
return remap_oldmem_pfn_range_zfcpdump(vma, from, pfn, size,
prot);
}
static const char *nt_name(Elf64_Word type)
{
const char *name = "LINUX";
if (type == NT_PRPSINFO || type == NT_PRSTATUS || type == NT_PRFPREG)
name = KEXEC_CORE_NOTE_NAME;
return name;
}
/*
* Initialize ELF note
*/
static void *nt_init_name(void *buf, Elf64_Word type, void *desc, int d_len,
const char *name)
{
Elf64_Nhdr *note;
u64 len;
note = (Elf64_Nhdr *)buf;
note->n_namesz = strlen(name) + 1;
note->n_descsz = d_len;
note->n_type = type;
len = sizeof(Elf64_Nhdr);
memcpy(buf + len, name, note->n_namesz);
len = roundup(len + note->n_namesz, 4);
memcpy(buf + len, desc, note->n_descsz);
len = roundup(len + note->n_descsz, 4);
return PTR_ADD(buf, len);
}
static inline void *nt_init(void *buf, Elf64_Word type, void *desc, int d_len)
{
return nt_init_name(buf, type, desc, d_len, nt_name(type));
}
/*
* Calculate the size of ELF note
*/
static size_t nt_size_name(int d_len, const char *name)
{
size_t size;
size = sizeof(Elf64_Nhdr);
size += roundup(strlen(name) + 1, 4);
size += roundup(d_len, 4);
return size;
}
static inline size_t nt_size(Elf64_Word type, int d_len)
{
return nt_size_name(d_len, nt_name(type));
}
/*
* Fill ELF notes for one CPU with save area registers
*/
static void *fill_cpu_elf_notes(void *ptr, int cpu, struct save_area *sa)
{
struct elf_prstatus nt_prstatus;
elf_fpregset_t nt_fpregset;
/* Prepare prstatus note */
memset(&nt_prstatus, 0, sizeof(nt_prstatus));
memcpy(&nt_prstatus.pr_reg.gprs, sa->gprs, sizeof(sa->gprs));
memcpy(&nt_prstatus.pr_reg.psw, sa->psw, sizeof(sa->psw));
memcpy(&nt_prstatus.pr_reg.acrs, sa->acrs, sizeof(sa->acrs));
nt_prstatus.pr_pid = cpu;
/* Prepare fpregset (floating point) note */
memset(&nt_fpregset, 0, sizeof(nt_fpregset));
memcpy(&nt_fpregset.fpc, &sa->fpc, sizeof(sa->fpc));
memcpy(&nt_fpregset.fprs, &sa->fprs, sizeof(sa->fprs));
/* Create ELF notes for the CPU */
ptr = nt_init(ptr, NT_PRSTATUS, &nt_prstatus, sizeof(nt_prstatus));
ptr = nt_init(ptr, NT_PRFPREG, &nt_fpregset, sizeof(nt_fpregset));
ptr = nt_init(ptr, NT_S390_TIMER, &sa->timer, sizeof(sa->timer));
ptr = nt_init(ptr, NT_S390_TODCMP, &sa->todcmp, sizeof(sa->todcmp));
ptr = nt_init(ptr, NT_S390_TODPREG, &sa->todpreg, sizeof(sa->todpreg));
ptr = nt_init(ptr, NT_S390_CTRS, &sa->ctrs, sizeof(sa->ctrs));
ptr = nt_init(ptr, NT_S390_PREFIX, &sa->prefix, sizeof(sa->prefix));
if (MACHINE_HAS_VX) {
ptr = nt_init(ptr, NT_S390_VXRS_HIGH,
&sa->vxrs_high, sizeof(sa->vxrs_high));
ptr = nt_init(ptr, NT_S390_VXRS_LOW,
&sa->vxrs_low, sizeof(sa->vxrs_low));
}
return ptr;
}
/*
* Calculate size of ELF notes per cpu
*/
static size_t get_cpu_elf_notes_size(void)
{
struct save_area *sa = NULL;
size_t size;
size = nt_size(NT_PRSTATUS, sizeof(struct elf_prstatus));
size += nt_size(NT_PRFPREG, sizeof(elf_fpregset_t));
size += nt_size(NT_S390_TIMER, sizeof(sa->timer));
size += nt_size(NT_S390_TODCMP, sizeof(sa->todcmp));
size += nt_size(NT_S390_TODPREG, sizeof(sa->todpreg));
size += nt_size(NT_S390_CTRS, sizeof(sa->ctrs));
size += nt_size(NT_S390_PREFIX, sizeof(sa->prefix));
if (MACHINE_HAS_VX) {
size += nt_size(NT_S390_VXRS_HIGH, sizeof(sa->vxrs_high));
size += nt_size(NT_S390_VXRS_LOW, sizeof(sa->vxrs_low));
}
return size;
}
/*
* Initialize prpsinfo note (new kernel)
*/
static void *nt_prpsinfo(void *ptr)
{
struct elf_prpsinfo prpsinfo;
memset(&prpsinfo, 0, sizeof(prpsinfo));
prpsinfo.pr_sname = 'R';
strcpy(prpsinfo.pr_fname, "vmlinux");
return nt_init(ptr, NT_PRPSINFO, &prpsinfo, sizeof(prpsinfo));
}
/*
* Get vmcoreinfo using lowcore->vmcore_info (new kernel)
*/
static void *get_vmcoreinfo_old(unsigned long *size)
{
char nt_name[11], *vmcoreinfo;
Elf64_Nhdr note;
void *addr;
if (copy_oldmem_kernel(&addr, &S390_lowcore.vmcore_info, sizeof(addr)))
return NULL;
memset(nt_name, 0, sizeof(nt_name));
if (copy_oldmem_kernel(&note, addr, sizeof(note)))
return NULL;
if (copy_oldmem_kernel(nt_name, addr + sizeof(note),
sizeof(nt_name) - 1))
return NULL;
if (strcmp(nt_name, VMCOREINFO_NOTE_NAME) != 0)
return NULL;
vmcoreinfo = kzalloc(note.n_descsz, GFP_KERNEL);
if (!vmcoreinfo)
return NULL;
if (copy_oldmem_kernel(vmcoreinfo, addr + 24, note.n_descsz)) {
kfree(vmcoreinfo);
return NULL;
}
*size = note.n_descsz;
return vmcoreinfo;
}
/*
* Initialize vmcoreinfo note (new kernel)
*/
static void *nt_vmcoreinfo(void *ptr)
{
const char *name = VMCOREINFO_NOTE_NAME;
unsigned long size;
void *vmcoreinfo;
vmcoreinfo = os_info_old_entry(OS_INFO_VMCOREINFO, &size);
if (vmcoreinfo)
return nt_init_name(ptr, 0, vmcoreinfo, size, name);
vmcoreinfo = get_vmcoreinfo_old(&size);
if (!vmcoreinfo)
return ptr;
ptr = nt_init_name(ptr, 0, vmcoreinfo, size, name);
kfree(vmcoreinfo);
return ptr;
}
static size_t nt_vmcoreinfo_size(void)
{
const char *name = VMCOREINFO_NOTE_NAME;
unsigned long size;
void *vmcoreinfo;
vmcoreinfo = os_info_old_entry(OS_INFO_VMCOREINFO, &size);
if (vmcoreinfo)
return nt_size_name(size, name);
vmcoreinfo = get_vmcoreinfo_old(&size);
if (!vmcoreinfo)
return 0;
kfree(vmcoreinfo);
return nt_size_name(size, name);
}
/*
* Initialize final note (needed for /proc/vmcore code)
*/
static void *nt_final(void *ptr)
{
Elf64_Nhdr *note;
note = (Elf64_Nhdr *) ptr;
note->n_namesz = 0;
note->n_descsz = 0;
note->n_type = 0;
return PTR_ADD(ptr, sizeof(Elf64_Nhdr));
}
/*
* Initialize ELF header (new kernel)
*/
static void *ehdr_init(Elf64_Ehdr *ehdr, int mem_chunk_cnt)
{
memset(ehdr, 0, sizeof(*ehdr));
memcpy(ehdr->e_ident, ELFMAG, SELFMAG);
ehdr->e_ident[EI_CLASS] = ELFCLASS64;
ehdr->e_ident[EI_DATA] = ELFDATA2MSB;
ehdr->e_ident[EI_VERSION] = EV_CURRENT;
memset(ehdr->e_ident + EI_PAD, 0, EI_NIDENT - EI_PAD);
ehdr->e_type = ET_CORE;
ehdr->e_machine = EM_S390;
ehdr->e_version = EV_CURRENT;
ehdr->e_phoff = sizeof(Elf64_Ehdr);
ehdr->e_ehsize = sizeof(Elf64_Ehdr);
ehdr->e_phentsize = sizeof(Elf64_Phdr);
ehdr->e_phnum = mem_chunk_cnt + 1;
return ehdr + 1;
}
/*
* Return CPU count for ELF header (new kernel)
*/
static int get_cpu_cnt(void)
{
struct save_area *sa;
int cpus = 0;
list_for_each_entry(sa, &dump_save_areas, list)
if (sa->prefix != 0)
cpus++;
return cpus;
}
/*
* Return memory chunk count for ELF header (new kernel)
*/
static int get_mem_chunk_cnt(void)
{
int cnt = 0;
u64 idx;
for_each_mem_range(idx, &memblock.physmem, &oldmem_type, NUMA_NO_NODE,
MEMBLOCK_NONE, NULL, NULL, NULL)
cnt++;
return cnt;
}
/*
* Initialize ELF loads (new kernel)
*/
static void loads_init(Elf64_Phdr *phdr, u64 loads_offset)
{
phys_addr_t start, end;
u64 idx;
for_each_mem_range(idx, &memblock.physmem, &oldmem_type, NUMA_NO_NODE,
MEMBLOCK_NONE, &start, &end, NULL) {
phdr->p_filesz = end - start;
phdr->p_type = PT_LOAD;
phdr->p_offset = start;
phdr->p_vaddr = start;
phdr->p_paddr = start;
phdr->p_memsz = end - start;
phdr->p_flags = PF_R | PF_W | PF_X;
phdr->p_align = PAGE_SIZE;
phdr++;
}
}
/*
* Initialize notes (new kernel)
*/
static void *notes_init(Elf64_Phdr *phdr, void *ptr, u64 notes_offset)
{
struct save_area *sa;
void *ptr_start = ptr;
int cpu;
ptr = nt_prpsinfo(ptr);
cpu = 1;
list_for_each_entry(sa, &dump_save_areas, list)
if (sa->prefix != 0)
ptr = fill_cpu_elf_notes(ptr, cpu++, sa);
ptr = nt_vmcoreinfo(ptr);
ptr = nt_final(ptr);
memset(phdr, 0, sizeof(*phdr));
phdr->p_type = PT_NOTE;
phdr->p_offset = notes_offset;
phdr->p_filesz = (unsigned long) PTR_SUB(ptr, ptr_start);
phdr->p_memsz = phdr->p_filesz;
return ptr;
}
static size_t get_elfcorehdr_size(int mem_chunk_cnt)
{
size_t size;
size = sizeof(Elf64_Ehdr);
/* PT_NOTES */
size += sizeof(Elf64_Phdr);
/* nt_prpsinfo */
size += nt_size(NT_PRPSINFO, sizeof(struct elf_prpsinfo));
/* regsets */
size += get_cpu_cnt() * get_cpu_elf_notes_size();
/* nt_vmcoreinfo */
size += nt_vmcoreinfo_size();
/* nt_final */
size += sizeof(Elf64_Nhdr);
/* PT_LOADS */
size += mem_chunk_cnt * sizeof(Elf64_Phdr);
return size;
}
/*
* Create ELF core header (new kernel)
*/
int elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size)
{
Elf64_Phdr *phdr_notes, *phdr_loads;
int mem_chunk_cnt;
void *ptr, *hdr;
u32 alloc_size;
u64 hdr_off;
/* If we are not in kdump or zfcpdump mode return */
if (!OLDMEM_BASE && ipl_info.type != IPL_TYPE_FCP_DUMP)
return 0;
/* If we cannot get HSA size for zfcpdump return error */
if (ipl_info.type == IPL_TYPE_FCP_DUMP && !sclp.hsa_size)
return -ENODEV;
/* For kdump, exclude previous crashkernel memory */
if (OLDMEM_BASE) {
oldmem_region.base = OLDMEM_BASE;
oldmem_region.size = OLDMEM_SIZE;
oldmem_type.total_size = OLDMEM_SIZE;
}
mem_chunk_cnt = get_mem_chunk_cnt();
alloc_size = get_elfcorehdr_size(mem_chunk_cnt);
hdr = kzalloc(alloc_size, GFP_KERNEL);
/* Without elfcorehdr /proc/vmcore cannot be created. Thus creating
* a dump with this crash kernel will fail. Panic now to allow other
* dump mechanisms to take over.
*/
if (!hdr)
panic("s390 kdump allocating elfcorehdr failed");
/* Init elf header */
ptr = ehdr_init(hdr, mem_chunk_cnt);
/* Init program headers */
phdr_notes = ptr;
ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr));
phdr_loads = ptr;
ptr = PTR_ADD(ptr, sizeof(Elf64_Phdr) * mem_chunk_cnt);
/* Init notes */
hdr_off = PTR_DIFF(ptr, hdr);
ptr = notes_init(phdr_notes, ptr, ((unsigned long) hdr) + hdr_off);
/* Init loads */
hdr_off = PTR_DIFF(ptr, hdr);
loads_init(phdr_loads, hdr_off);
*addr = (unsigned long long) hdr;
*size = (unsigned long long) hdr_off;
BUG_ON(elfcorehdr_size > alloc_size);
return 0;
}
/*
* Free ELF core header (new kernel)
*/
void elfcorehdr_free(unsigned long long addr)
{
kfree((void *)(unsigned long)addr);
}
/*
* Read from ELF header
*/
ssize_t elfcorehdr_read(char *buf, size_t count, u64 *ppos)
{
void *src = (void *)(unsigned long)*ppos;
memcpy(buf, src, count);
*ppos += count;
return count;
}
/*
* Read from ELF notes data
*/
ssize_t elfcorehdr_read_notes(char *buf, size_t count, u64 *ppos)
{
void *src = (void *)(unsigned long)*ppos;
memcpy(buf, src, count);
*ppos += count;
return count;
}
File diff suppressed because it is too large. Load diff
+268
View File
@@ -0,0 +1,268 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Implementation of s390 diagnose codes
*
* Copyright IBM Corp. 2007
* Author(s): Michael Holzheu <holzheu@de.ibm.com>
*/
#include <linux/export.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/seq_file.h>
#include <linux/debugfs.h>
#include <asm/diag.h>
#include <asm/trace/diag.h>
struct diag_stat {
unsigned int counter[NR_DIAG_STAT];
};
static DEFINE_PER_CPU(struct diag_stat, diag_stat);
struct diag_desc {
int code;
char *name;
};
static const struct diag_desc diag_map[NR_DIAG_STAT] = {
[DIAG_STAT_X008] = { .code = 0x008, .name = "Console Function" },
[DIAG_STAT_X00C] = { .code = 0x00c, .name = "Pseudo Timer" },
[DIAG_STAT_X010] = { .code = 0x010, .name = "Release Pages" },
[DIAG_STAT_X014] = { .code = 0x014, .name = "Spool File Services" },
[DIAG_STAT_X044] = { .code = 0x044, .name = "Voluntary Timeslice End" },
[DIAG_STAT_X064] = { .code = 0x064, .name = "NSS Manipulation" },
[DIAG_STAT_X09C] = { .code = 0x09c, .name = "Relinquish Timeslice" },
[DIAG_STAT_X0DC] = { .code = 0x0dc, .name = "Appldata Control" },
[DIAG_STAT_X204] = { .code = 0x204, .name = "Logical-CPU Utilization" },
[DIAG_STAT_X210] = { .code = 0x210, .name = "Device Information" },
[DIAG_STAT_X224] = { .code = 0x224, .name = "EBCDIC-Name Table" },
[DIAG_STAT_X250] = { .code = 0x250, .name = "Block I/O" },
[DIAG_STAT_X258] = { .code = 0x258, .name = "Page-Reference Services" },
[DIAG_STAT_X26C] = { .code = 0x26c, .name = "Certain System Information" },
[DIAG_STAT_X288] = { .code = 0x288, .name = "Time Bomb" },
[DIAG_STAT_X2C4] = { .code = 0x2c4, .name = "FTP Services" },
[DIAG_STAT_X2FC] = { .code = 0x2fc, .name = "Guest Performance Data" },
[DIAG_STAT_X304] = { .code = 0x304, .name = "Partition-Resource Service" },
[DIAG_STAT_X308] = { .code = 0x308, .name = "List-Directed IPL" },
[DIAG_STAT_X500] = { .code = 0x500, .name = "Virtio Service" },
};
static int show_diag_stat(struct seq_file *m, void *v)
{
struct diag_stat *stat;
unsigned long n = (unsigned long) v - 1;
int cpu, prec, tmp;
get_online_cpus();
if (n == 0) {
seq_puts(m, " ");
for_each_online_cpu(cpu) {
prec = 10;
for (tmp = 10; cpu >= tmp; tmp *= 10)
prec--;
seq_printf(m, "%*s%d", prec, "CPU", cpu);
}
seq_putc(m, '\n');
} else if (n <= NR_DIAG_STAT) {
seq_printf(m, "diag %03x:", diag_map[n-1].code);
for_each_online_cpu(cpu) {
stat = &per_cpu(diag_stat, cpu);
seq_printf(m, " %10u", stat->counter[n-1]);
}
seq_printf(m, " %s\n", diag_map[n-1].name);
}
put_online_cpus();
return 0;
}
static void *show_diag_stat_start(struct seq_file *m, loff_t *pos)
{
return *pos <= NR_DIAG_STAT ? (void *)((unsigned long) *pos + 1) : NULL;
}
static void *show_diag_stat_next(struct seq_file *m, void *v, loff_t *pos)
{
++*pos;
return show_diag_stat_start(m, pos);
}
static void show_diag_stat_stop(struct seq_file *m, void *v)
{
}
static const struct seq_operations show_diag_stat_sops = {
.start = show_diag_stat_start,
.next = show_diag_stat_next,
.stop = show_diag_stat_stop,
.show = show_diag_stat,
};
static int show_diag_stat_open(struct inode *inode, struct file *file)
{
return seq_open(file, &show_diag_stat_sops);
}
static const struct file_operations show_diag_stat_fops = {
.open = show_diag_stat_open,
.read = seq_read,
.llseek = seq_lseek,
.release = seq_release,
};
static int __init show_diag_stat_init(void)
{
debugfs_create_file("diag_stat", 0400, NULL, NULL,
&show_diag_stat_fops);
return 0;
}
device_initcall(show_diag_stat_init);
void diag_stat_inc(enum diag_stat_enum nr)
{
this_cpu_inc(diag_stat.counter[nr]);
trace_s390_diagnose(diag_map[nr].code);
}
EXPORT_SYMBOL(diag_stat_inc);
void notrace diag_stat_inc_norecursion(enum diag_stat_enum nr)
{
this_cpu_inc(diag_stat.counter[nr]);
trace_s390_diagnose_norecursion(diag_map[nr].code);
}
EXPORT_SYMBOL(diag_stat_inc_norecursion);
/*
* Diagnose 14: Input spool file manipulation
*/
static inline int __diag14(unsigned long rx, unsigned long ry1,
unsigned long subcode)
{
register unsigned long _ry1 asm("2") = ry1;
register unsigned long _ry2 asm("3") = subcode;
int rc = 0;
asm volatile(
" sam31\n"
" diag %2,2,0x14\n"
" sam64\n"
" ipm %0\n"
" srl %0,28\n"
: "=d" (rc), "+d" (_ry2)
: "d" (rx), "d" (_ry1)
: "cc");
return rc;
}
int diag14(unsigned long rx, unsigned long ry1, unsigned long subcode)
{
diag_stat_inc(DIAG_STAT_X014);
return __diag14(rx, ry1, subcode);
}
EXPORT_SYMBOL(diag14);
static inline int __diag204(unsigned long *subcode, unsigned long size, void *addr)
{
register unsigned long _subcode asm("0") = *subcode;
register unsigned long _size asm("1") = size;
asm volatile(
" diag %2,%0,0x204\n"
"0: nopr %%r7\n"
EX_TABLE(0b,0b)
: "+d" (_subcode), "+d" (_size) : "d" (addr) : "memory");
*subcode = _subcode;
return _size;
}
int diag204(unsigned long subcode, unsigned long size, void *addr)
{
diag_stat_inc(DIAG_STAT_X204);
size = __diag204(&subcode, size, addr);
if (subcode)
return -1;
return size;
}
EXPORT_SYMBOL(diag204);
/*
* Diagnose 210: Get information about a virtual device
*/
int diag210(struct diag210 *addr)
{
/*
* diag 210 needs its data below the 2GB border, so we
* use a static data area to be sure
*/
static struct diag210 diag210_tmp;
static DEFINE_SPINLOCK(diag210_lock);
unsigned long flags;
int ccode;
spin_lock_irqsave(&diag210_lock, flags);
diag210_tmp = *addr;
diag_stat_inc(DIAG_STAT_X210);
asm volatile(
" lhi %0,-1\n"
" sam31\n"
" diag %1,0,0x210\n"
"0: ipm %0\n"
" srl %0,28\n"
"1: sam64\n"
EX_TABLE(0b, 1b)
: "=&d" (ccode) : "a" (&diag210_tmp) : "cc", "memory");
*addr = diag210_tmp;
spin_unlock_irqrestore(&diag210_lock, flags);
return ccode;
}
EXPORT_SYMBOL(diag210);
int diag224(void *ptr)
{
int rc = -EOPNOTSUPP;
diag_stat_inc(DIAG_STAT_X224);
asm volatile(
" diag %1,%2,0x224\n"
"0: lhi %0,0x0\n"
"1:\n"
EX_TABLE(0b,1b)
: "+d" (rc) :"d" (0), "d" (ptr) : "memory");
return rc;
}
EXPORT_SYMBOL(diag224);
/*
* Diagnose 26C: Access Certain System Information
*/
static inline int __diag26c(void *req, void *resp, enum diag26c_sc subcode)
{
register unsigned long _req asm("2") = (addr_t) req;
register unsigned long _resp asm("3") = (addr_t) resp;
register unsigned long _subcode asm("4") = subcode;
register unsigned long _rc asm("5") = -EOPNOTSUPP;
asm volatile(
" sam31\n"
" diag %[rx],%[ry],0x26c\n"
"0: sam64\n"
EX_TABLE(0b,0b)
: "+d" (_rc)
: [rx] "d" (_req), "d" (_resp), [ry] "d" (_subcode)
: "cc", "memory");
return _rc;
}
int diag26c(void *req, void *resp, enum diag26c_sc subcode)
{
diag_stat_inc(DIAG_STAT_X26C);
return __diag26c(req, resp, subcode);
}
EXPORT_SYMBOL(diag26c);
+582
View File
@@ -0,0 +1,582 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Disassemble s390 instructions.
*
* Copyright IBM Corp. 2007
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
*/
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/string.h>
#include <linux/errno.h>
#include <linux/ptrace.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/delay.h>
#include <linux/export.h>
#include <linux/kallsyms.h>
#include <linux/reboot.h>
#include <linux/kprobes.h>
#include <linux/kdebug.h>
#include <linux/uaccess.h>
#include <linux/atomic.h>
#include <asm/dis.h>
#include <asm/io.h>
#include <asm/cpcmd.h>
#include <asm/lowcore.h>
#include <asm/debug.h>
#include <asm/irq.h>
/* Type of operand */
#define OPERAND_GPR 0x1 /* Operand printed as %rx */
#define OPERAND_FPR 0x2 /* Operand printed as %fx */
#define OPERAND_AR 0x4 /* Operand printed as %ax */
#define OPERAND_CR 0x8 /* Operand printed as %cx */
#define OPERAND_VR 0x10 /* Operand printed as %vx */
#define OPERAND_DISP 0x20 /* Operand printed as displacement */
#define OPERAND_BASE 0x40 /* Operand printed as base register */
#define OPERAND_INDEX 0x80 /* Operand printed as index register */
#define OPERAND_PCREL 0x100 /* Operand printed as pc-relative symbol */
#define OPERAND_SIGNED 0x200 /* Operand printed as signed value */
#define OPERAND_LENGTH 0x400 /* Operand printed as length (+1) */
struct s390_operand {
unsigned char bits; /* The number of bits in the operand. */
unsigned char shift; /* The number of bits to shift. */
unsigned short flags; /* One bit syntax flags. */
};
struct s390_insn {
union {
const char name[5];
struct {
unsigned char zero;
unsigned int offset;
} __packed;
};
unsigned char opfrag;
unsigned char format;
};
struct s390_opcode_offset {
unsigned char opcode;
unsigned char mask;
unsigned char byte;
unsigned short offset;
unsigned short count;
} __packed;
enum {
UNUSED,
A_8, /* Access reg. starting at position 8 */
A_12, /* Access reg. starting at position 12 */
A_24, /* Access reg. starting at position 24 */
A_28, /* Access reg. starting at position 28 */
B_16, /* Base register starting at position 16 */
B_32, /* Base register starting at position 32 */
C_8, /* Control reg. starting at position 8 */
C_12, /* Control reg. starting at position 12 */
D20_20, /* 20 bit displacement starting at 20 */
D_20, /* Displacement starting at position 20 */
D_36, /* Displacement starting at position 36 */
F_8, /* FPR starting at position 8 */
F_12, /* FPR starting at position 12 */
F_16, /* FPR starting at position 16 */
F_24, /* FPR starting at position 24 */
F_28, /* FPR starting at position 28 */
F_32, /* FPR starting at position 32 */
I8_8, /* 8 bit signed value starting at 8 */
I8_32, /* 8 bit signed value starting at 32 */
I16_16, /* 16 bit signed value starting at 16 */
I16_32, /* 16 bit signed value starting at 32 */
I32_16, /* 32 bit signed value starting at 16 */
J12_12, /* 12 bit PC relative offset at 12 */
J16_16, /* 16 bit PC relative offset at 16 */
J16_32, /* 16 bit PC relative offset at 32 */
J24_24, /* 24 bit PC relative offset at 24 */
J32_16, /* 32 bit PC relative offset at 16 */
L4_8, /* 4 bit length starting at position 8 */
L4_12, /* 4 bit length starting at position 12 */
L8_8, /* 8 bit length starting at position 8 */
R_8, /* GPR starting at position 8 */
R_12, /* GPR starting at position 12 */
R_16, /* GPR starting at position 16 */
R_24, /* GPR starting at position 24 */
R_28, /* GPR starting at position 28 */
U4_8, /* 4 bit unsigned value starting at 8 */
U4_12, /* 4 bit unsigned value starting at 12 */
U4_16, /* 4 bit unsigned value starting at 16 */
U4_20, /* 4 bit unsigned value starting at 20 */
U4_24, /* 4 bit unsigned value starting at 24 */
U4_28, /* 4 bit unsigned value starting at 28 */
U4_32, /* 4 bit unsigned value starting at 32 */
U4_36, /* 4 bit unsigned value starting at 36 */
U8_8, /* 8 bit unsigned value starting at 8 */
U8_16, /* 8 bit unsigned value starting at 16 */
U8_24, /* 8 bit unsigned value starting at 24 */
U8_28, /* 8 bit unsigned value starting at 28 */
U8_32, /* 8 bit unsigned value starting at 32 */
U12_16, /* 12 bit unsigned value starting at 16 */
U16_16, /* 16 bit unsigned value starting at 16 */
U16_32, /* 16 bit unsigned value starting at 32 */
U32_16, /* 32 bit unsigned value starting at 16 */
VX_12, /* Vector index register starting at position 12 */
V_8, /* Vector reg. starting at position 8 */
V_12, /* Vector reg. starting at position 12 */
V_16, /* Vector reg. starting at position 16 */
V_32, /* Vector reg. starting at position 32 */
X_12, /* Index register starting at position 12 */
};
static const struct s390_operand operands[] = {
[UNUSED] = { 0, 0, 0 },
[A_8] = { 4, 8, OPERAND_AR },
[A_12] = { 4, 12, OPERAND_AR },
[A_24] = { 4, 24, OPERAND_AR },
[A_28] = { 4, 28, OPERAND_AR },
[B_16] = { 4, 16, OPERAND_BASE | OPERAND_GPR },
[B_32] = { 4, 32, OPERAND_BASE | OPERAND_GPR },
[C_8] = { 4, 8, OPERAND_CR },
[C_12] = { 4, 12, OPERAND_CR },
[D20_20] = { 20, 20, OPERAND_DISP | OPERAND_SIGNED },
[D_20] = { 12, 20, OPERAND_DISP },
[D_36] = { 12, 36, OPERAND_DISP },
[F_8] = { 4, 8, OPERAND_FPR },
[F_12] = { 4, 12, OPERAND_FPR },
[F_16] = { 4, 16, OPERAND_FPR },
[F_24] = { 4, 24, OPERAND_FPR },
[F_28] = { 4, 28, OPERAND_FPR },
[F_32] = { 4, 32, OPERAND_FPR },
[I8_8] = { 8, 8, OPERAND_SIGNED },
[I8_32] = { 8, 32, OPERAND_SIGNED },
[I16_16] = { 16, 16, OPERAND_SIGNED },
[I16_32] = { 16, 32, OPERAND_SIGNED },
[I32_16] = { 32, 16, OPERAND_SIGNED },
[J12_12] = { 12, 12, OPERAND_PCREL },
[J16_16] = { 16, 16, OPERAND_PCREL },
[J16_32] = { 16, 32, OPERAND_PCREL },
[J24_24] = { 24, 24, OPERAND_PCREL },
[J32_16] = { 32, 16, OPERAND_PCREL },
[L4_8] = { 4, 8, OPERAND_LENGTH },
[L4_12] = { 4, 12, OPERAND_LENGTH },
[L8_8] = { 8, 8, OPERAND_LENGTH },
[R_8] = { 4, 8, OPERAND_GPR },
[R_12] = { 4, 12, OPERAND_GPR },
[R_16] = { 4, 16, OPERAND_GPR },
[R_24] = { 4, 24, OPERAND_GPR },
[R_28] = { 4, 28, OPERAND_GPR },
[U4_8] = { 4, 8, 0 },
[U4_12] = { 4, 12, 0 },
[U4_16] = { 4, 16, 0 },
[U4_20] = { 4, 20, 0 },
[U4_24] = { 4, 24, 0 },
[U4_28] = { 4, 28, 0 },
[U4_32] = { 4, 32, 0 },
[U4_36] = { 4, 36, 0 },
[U8_8] = { 8, 8, 0 },
[U8_16] = { 8, 16, 0 },
[U8_24] = { 8, 24, 0 },
[U8_28] = { 8, 28, 0 },
[U8_32] = { 8, 32, 0 },
[U12_16] = { 12, 16, 0 },
[U16_16] = { 16, 16, 0 },
[U16_32] = { 16, 32, 0 },
[U32_16] = { 32, 16, 0 },
[VX_12] = { 4, 12, OPERAND_INDEX | OPERAND_VR },
[V_8] = { 4, 8, OPERAND_VR },
[V_12] = { 4, 12, OPERAND_VR },
[V_16] = { 4, 16, OPERAND_VR },
[V_32] = { 4, 32, OPERAND_VR },
[X_12] = { 4, 12, OPERAND_INDEX | OPERAND_GPR },
};
static const unsigned char formats[][6] = {
[INSTR_E] = { 0, 0, 0, 0, 0, 0 },
[INSTR_IE_UU] = { U4_24, U4_28, 0, 0, 0, 0 },
[INSTR_MII_UPP] = { U4_8, J12_12, J24_24 },
[INSTR_RIE_R0IU] = { R_8, I16_16, U4_32, 0, 0, 0 },
[INSTR_RIE_R0UU] = { R_8, U16_16, U4_32, 0, 0, 0 },
[INSTR_RIE_RRI0] = { R_8, R_12, I16_16, 0, 0, 0 },
[INSTR_RIE_RRP] = { R_8, R_12, J16_16, 0, 0, 0 },
[INSTR_RIE_RRPU] = { R_8, R_12, U4_32, J16_16, 0, 0 },
[INSTR_RIE_RRUUU] = { R_8, R_12, U8_16, U8_24, U8_32, 0 },
[INSTR_RIE_RUI0] = { R_8, I16_16, U4_12, 0, 0, 0 },
[INSTR_RIE_RUPI] = { R_8, I8_32, U4_12, J16_16, 0, 0 },
[INSTR_RIE_RUPU] = { R_8, U8_32, U4_12, J16_16, 0, 0 },
[INSTR_RIL_RI] = { R_8, I32_16, 0, 0, 0, 0 },
[INSTR_RIL_RP] = { R_8, J32_16, 0, 0, 0, 0 },
[INSTR_RIL_RU] = { R_8, U32_16, 0, 0, 0, 0 },
[INSTR_RIL_UP] = { U4_8, J32_16, 0, 0, 0, 0 },
[INSTR_RIS_RURDI] = { R_8, I8_32, U4_12, D_20, B_16, 0 },
[INSTR_RIS_RURDU] = { R_8, U8_32, U4_12, D_20, B_16, 0 },
[INSTR_RI_RI] = { R_8, I16_16, 0, 0, 0, 0 },
[INSTR_RI_RP] = { R_8, J16_16, 0, 0, 0, 0 },
[INSTR_RI_RU] = { R_8, U16_16, 0, 0, 0, 0 },
[INSTR_RI_UP] = { U4_8, J16_16, 0, 0, 0, 0 },
[INSTR_RRE_00] = { 0, 0, 0, 0, 0, 0 },
[INSTR_RRE_AA] = { A_24, A_28, 0, 0, 0, 0 },
[INSTR_RRE_AR] = { A_24, R_28, 0, 0, 0, 0 },
[INSTR_RRE_F0] = { F_24, 0, 0, 0, 0, 0 },
[INSTR_RRE_FF] = { F_24, F_28, 0, 0, 0, 0 },
[INSTR_RRE_FR] = { F_24, R_28, 0, 0, 0, 0 },
[INSTR_RRE_R0] = { R_24, 0, 0, 0, 0, 0 },
[INSTR_RRE_RA] = { R_24, A_28, 0, 0, 0, 0 },
[INSTR_RRE_RF] = { R_24, F_28, 0, 0, 0, 0 },
[INSTR_RRE_RR] = { R_24, R_28, 0, 0, 0, 0 },
[INSTR_RRF_0UFF] = { F_24, F_28, U4_20, 0, 0, 0 },
[INSTR_RRF_0URF] = { R_24, F_28, U4_20, 0, 0, 0 },
[INSTR_RRF_F0FF] = { F_16, F_24, F_28, 0, 0, 0 },
[INSTR_RRF_F0FF2] = { F_24, F_16, F_28, 0, 0, 0 },
[INSTR_RRF_F0FR] = { F_24, F_16, R_28, 0, 0, 0 },
[INSTR_RRF_FFRU] = { F_24, F_16, R_28, U4_20, 0, 0 },
[INSTR_RRF_FUFF] = { F_24, F_16, F_28, U4_20, 0, 0 },
[INSTR_RRF_FUFF2] = { F_24, F_28, F_16, U4_20, 0, 0 },
[INSTR_RRF_R0RR] = { R_24, R_16, R_28, 0, 0, 0 },
[INSTR_RRF_R0RR2] = { R_24, R_28, R_16, 0, 0, 0 },
[INSTR_RRF_RURR] = { R_24, R_28, R_16, U4_20, 0, 0 },
[INSTR_RRF_RURR2] = { R_24, R_16, R_28, U4_20, 0, 0 },
[INSTR_RRF_U0FF] = { F_24, U4_16, F_28, 0, 0, 0 },
[INSTR_RRF_U0RF] = { R_24, U4_16, F_28, 0, 0, 0 },
[INSTR_RRF_U0RR] = { R_24, R_28, U4_16, 0, 0, 0 },
[INSTR_RRF_UUFF] = { F_24, U4_16, F_28, U4_20, 0, 0 },
[INSTR_RRF_UUFR] = { F_24, U4_16, R_28, U4_20, 0, 0 },
[INSTR_RRF_UURF] = { R_24, U4_16, F_28, U4_20, 0, 0 },
[INSTR_RRS_RRRDU] = { R_8, R_12, U4_32, D_20, B_16 },
[INSTR_RR_FF] = { F_8, F_12, 0, 0, 0, 0 },
[INSTR_RR_R0] = { R_8, 0, 0, 0, 0, 0 },
[INSTR_RR_RR] = { R_8, R_12, 0, 0, 0, 0 },
[INSTR_RR_U0] = { U8_8, 0, 0, 0, 0, 0 },
[INSTR_RR_UR] = { U4_8, R_12, 0, 0, 0, 0 },
[INSTR_RSI_RRP] = { R_8, R_12, J16_16, 0, 0, 0 },
[INSTR_RSL_LRDFU] = { F_32, D_20, L8_8, B_16, U4_36, 0 },
[INSTR_RSL_R0RD] = { D_20, L4_8, B_16, 0, 0, 0 },
[INSTR_RSY_AARD] = { A_8, A_12, D20_20, B_16, 0, 0 },
[INSTR_RSY_CCRD] = { C_8, C_12, D20_20, B_16, 0, 0 },
[INSTR_RSY_RDRU] = { R_8, D20_20, B_16, U4_12, 0, 0 },
[INSTR_RSY_RRRD] = { R_8, R_12, D20_20, B_16, 0, 0 },
[INSTR_RSY_RURD] = { R_8, U4_12, D20_20, B_16, 0, 0 },
[INSTR_RSY_RURD2] = { R_8, D20_20, B_16, U4_12, 0, 0 },
[INSTR_RS_AARD] = { A_8, A_12, D_20, B_16, 0, 0 },
[INSTR_RS_CCRD] = { C_8, C_12, D_20, B_16, 0, 0 },
[INSTR_RS_R0RD] = { R_8, D_20, B_16, 0, 0, 0 },
[INSTR_RS_RRRD] = { R_8, R_12, D_20, B_16, 0, 0 },
[INSTR_RS_RURD] = { R_8, U4_12, D_20, B_16, 0, 0 },
[INSTR_RXE_FRRD] = { F_8, D_20, X_12, B_16, 0, 0 },
[INSTR_RXE_RRRDU] = { R_8, D_20, X_12, B_16, U4_32, 0 },
[INSTR_RXF_FRRDF] = { F_32, F_8, D_20, X_12, B_16, 0 },
[INSTR_RXY_FRRD] = { F_8, D20_20, X_12, B_16, 0, 0 },
[INSTR_RXY_RRRD] = { R_8, D20_20, X_12, B_16, 0, 0 },
[INSTR_RXY_URRD] = { U4_8, D20_20, X_12, B_16, 0, 0 },
[INSTR_RX_FRRD] = { F_8, D_20, X_12, B_16, 0, 0 },
[INSTR_RX_RRRD] = { R_8, D_20, X_12, B_16, 0, 0 },
[INSTR_RX_URRD] = { U4_8, D_20, X_12, B_16, 0, 0 },
[INSTR_SIL_RDI] = { D_20, B_16, I16_32, 0, 0, 0 },
[INSTR_SIL_RDU] = { D_20, B_16, U16_32, 0, 0, 0 },
[INSTR_SIY_IRD] = { D20_20, B_16, I8_8, 0, 0, 0 },
[INSTR_SIY_URD] = { D20_20, B_16, U8_8, 0, 0, 0 },
[INSTR_SI_RD] = { D_20, B_16, 0, 0, 0, 0 },
[INSTR_SI_URD] = { D_20, B_16, U8_8, 0, 0, 0 },
[INSTR_SMI_U0RDP] = { U4_8, J16_32, D_20, B_16, 0, 0 },
[INSTR_SSE_RDRD] = { D_20, B_16, D_36, B_32, 0, 0 },
[INSTR_SSF_RRDRD] = { D_20, B_16, D_36, B_32, R_8, 0 },
[INSTR_SSF_RRDRD2] = { R_8, D_20, B_16, D_36, B_32, 0 },
[INSTR_SS_L0RDRD] = { D_20, L8_8, B_16, D_36, B_32, 0 },
[INSTR_SS_L2RDRD] = { D_20, B_16, D_36, L8_8, B_32, 0 },
[INSTR_SS_LIRDRD] = { D_20, L4_8, B_16, D_36, B_32, U4_12 },
[INSTR_SS_LLRDRD] = { D_20, L4_8, B_16, D_36, L4_12, B_32 },
[INSTR_SS_RRRDRD] = { D_20, R_8, B_16, D_36, B_32, R_12 },
[INSTR_SS_RRRDRD2] = { R_8, D_20, B_16, R_12, D_36, B_32 },
[INSTR_SS_RRRDRD3] = { R_8, R_12, D_20, B_16, D_36, B_32 },
[INSTR_S_00] = { 0, 0, 0, 0, 0, 0 },
[INSTR_S_RD] = { D_20, B_16, 0, 0, 0, 0 },
[INSTR_VRI_V0IU] = { V_8, I16_16, U4_32, 0, 0, 0 },
[INSTR_VRI_V0U] = { V_8, U16_16, 0, 0, 0, 0 },
[INSTR_VRI_V0UU2] = { V_8, U16_16, U4_32, 0, 0, 0 },
[INSTR_VRI_V0UUU] = { V_8, U8_16, U8_24, U4_32, 0, 0 },
[INSTR_VRI_VR0UU] = { V_8, R_12, U8_28, U4_24, 0, 0 },
[INSTR_VRI_VVUU] = { V_8, V_12, U16_16, U4_32, 0, 0 },
[INSTR_VRI_VVUUU] = { V_8, V_12, U12_16, U4_32, U4_28, 0 },
[INSTR_VRI_VVUUU2] = { V_8, V_12, U8_28, U8_16, U4_24, 0 },
[INSTR_VRI_VVV0U] = { V_8, V_12, V_16, U8_24, 0, 0 },
[INSTR_VRI_VVV0UU] = { V_8, V_12, V_16, U8_24, U4_32, 0 },
[INSTR_VRI_VVV0UU2] = { V_8, V_12, V_16, U8_28, U4_24, 0 },
[INSTR_VRR_0V] = { V_12, 0, 0, 0, 0, 0 },
[INSTR_VRR_0VV0U] = { V_12, V_16, U4_24, 0, 0, 0 },
[INSTR_VRR_RV0U] = { R_8, V_12, U4_24, 0, 0, 0 },
[INSTR_VRR_VRR] = { V_8, R_12, R_16, 0, 0, 0 },
[INSTR_VRR_VV] = { V_8, V_12, 0, 0, 0, 0 },
[INSTR_VRR_VV0U] = { V_8, V_12, U4_32, 0, 0, 0 },
[INSTR_VRR_VV0U0U] = { V_8, V_12, U4_32, U4_24, 0, 0 },
[INSTR_VRR_VV0UU2] = { V_8, V_12, U4_32, U4_28, 0, 0 },
[INSTR_VRR_VV0UUU] = { V_8, V_12, U4_32, U4_28, U4_24, 0 },
[INSTR_VRR_VVV] = { V_8, V_12, V_16, 0, 0, 0 },
[INSTR_VRR_VVV0U] = { V_8, V_12, V_16, U4_32, 0, 0 },
[INSTR_VRR_VVV0U0U] = { V_8, V_12, V_16, U4_32, U4_24, 0 },
[INSTR_VRR_VVV0UU] = { V_8, V_12, V_16, U4_32, U4_28, 0 },
[INSTR_VRR_VVV0UUU] = { V_8, V_12, V_16, U4_32, U4_28, U4_24 },
[INSTR_VRR_VVV0V] = { V_8, V_12, V_16, V_32, 0, 0 },
[INSTR_VRR_VVVU0UV] = { V_8, V_12, V_16, V_32, U4_28, U4_20 },
[INSTR_VRR_VVVU0V] = { V_8, V_12, V_16, V_32, U4_20, 0 },
[INSTR_VRR_VVVUU0V] = { V_8, V_12, V_16, V_32, U4_20, U4_24 },
[INSTR_VRS_RRDV] = { V_32, R_12, D_20, B_16, 0, 0 },
[INSTR_VRS_RVRDU] = { R_8, V_12, D_20, B_16, U4_32, 0 },
[INSTR_VRS_VRRD] = { V_8, R_12, D_20, B_16, 0, 0 },
[INSTR_VRS_VRRDU] = { V_8, R_12, D_20, B_16, U4_32, 0 },
[INSTR_VRS_VVRD] = { V_8, V_12, D_20, B_16, 0, 0 },
[INSTR_VRS_VVRDU] = { V_8, V_12, D_20, B_16, U4_32, 0 },
[INSTR_VRV_VVXRDU] = { V_8, D_20, VX_12, B_16, U4_32, 0 },
[INSTR_VRX_VRRD] = { V_8, D_20, X_12, B_16, 0, 0 },
[INSTR_VRX_VRRDU] = { V_8, D_20, X_12, B_16, U4_32, 0 },
[INSTR_VRX_VV] = { V_8, V_12, 0, 0, 0, 0 },
[INSTR_VSI_URDV] = { V_32, D_20, B_16, U8_8, 0, 0 },
};
static char long_insn_name[][7] = LONG_INSN_INITIALIZER;
static struct s390_insn opcode[] = OPCODE_TABLE_INITIALIZER;
static struct s390_opcode_offset opcode_offset[] = OPCODE_OFFSET_INITIALIZER;
/* Extracts an operand value from an instruction. */
static unsigned int extract_operand(unsigned char *code,
const struct s390_operand *operand)
{
unsigned char *cp;
unsigned int val;
int bits;
/* Extract fragments of the operand byte for byte. */
cp = code + operand->shift / 8;
bits = (operand->shift & 7) + operand->bits;
val = 0;
do {
val <<= 8;
val |= (unsigned int) *cp++;
bits -= 8;
} while (bits > 0);
val >>= -bits;
val &= ((1U << (operand->bits - 1)) << 1) - 1;
/* Check for special long displacement case. */
if (operand->bits == 20 && operand->shift == 20)
val = (val & 0xff) << 12 | (val & 0xfff00) >> 8;
/* Check for register extensions bits for vector registers. */
if (operand->flags & OPERAND_VR) {
if (operand->shift == 8)
val |= (code[4] & 8) << 1;
else if (operand->shift == 12)
val |= (code[4] & 4) << 2;
else if (operand->shift == 16)
val |= (code[4] & 2) << 3;
else if (operand->shift == 32)
val |= (code[4] & 1) << 4;
}
/* Sign extend value if the operand is signed or pc relative. */
if ((operand->flags & (OPERAND_SIGNED | OPERAND_PCREL)) &&
(val & (1U << (operand->bits - 1))))
val |= (-1U << (operand->bits - 1)) << 1;
/* Double value if the operand is pc relative. */
if (operand->flags & OPERAND_PCREL)
val <<= 1;
/* Length x in an instructions has real length x + 1. */
if (operand->flags & OPERAND_LENGTH)
val++;
return val;
}
struct s390_insn *find_insn(unsigned char *code)
{
struct s390_opcode_offset *entry;
struct s390_insn *insn;
unsigned char opfrag;
int i;
/* Search the opcode offset table to find an entry which
* matches the beginning of the opcode. If there is no match
* the last entry will be used, which is the default entry for
* unknown instructions as well as 1-byte opcode instructions.
*/
for (i = 0; i < ARRAY_SIZE(opcode_offset); i++) {
entry = &opcode_offset[i];
if (entry->opcode == code[0])
break;
}
opfrag = *(code + entry->byte) & entry->mask;
insn = &opcode[entry->offset];
for (i = 0; i < entry->count; i++) {
if (insn->opfrag == opfrag)
return insn;
insn++;
}
return NULL;
}
static int print_insn(char *buffer, unsigned char *code, unsigned long addr)
{
struct s390_insn *insn;
const unsigned char *ops;
const struct s390_operand *operand;
unsigned int value;
char separator;
char *ptr;
int i;
ptr = buffer;
insn = find_insn(code);
if (insn) {
if (insn->zero == 0)
ptr += sprintf(ptr, "%.7s\t",
long_insn_name[insn->offset]);
else
ptr += sprintf(ptr, "%.5s\t", insn->name);
/* Extract the operands. */
separator = 0;
for (ops = formats[insn->format], i = 0;
*ops != 0 && i < 6; ops++, i++) {
operand = operands + *ops;
value = extract_operand(code, operand);
if ((operand->flags & OPERAND_INDEX) && value == 0)
continue;
if ((operand->flags & OPERAND_BASE) &&
value == 0 && separator == '(') {
separator = ',';
continue;
}
if (separator)
ptr += sprintf(ptr, "%c", separator);
if (operand->flags & OPERAND_GPR)
ptr += sprintf(ptr, "%%r%i", value);
else if (operand->flags & OPERAND_FPR)
ptr += sprintf(ptr, "%%f%i", value);
else if (operand->flags & OPERAND_AR)
ptr += sprintf(ptr, "%%a%i", value);
else if (operand->flags & OPERAND_CR)
ptr += sprintf(ptr, "%%c%i", value);
else if (operand->flags & OPERAND_VR)
ptr += sprintf(ptr, "%%v%i", value);
else if (operand->flags & OPERAND_PCREL) {
void *pcrel = (void *)((int)value + addr);
ptr += sprintf(ptr, "%px", pcrel);
} else if (operand->flags & OPERAND_SIGNED)
ptr += sprintf(ptr, "%i", value);
else
ptr += sprintf(ptr, "%u", value);
if (operand->flags & OPERAND_DISP)
separator = '(';
else if (operand->flags & OPERAND_BASE) {
ptr += sprintf(ptr, ")");
separator = ',';
} else
separator = ',';
}
} else
ptr += sprintf(ptr, "unknown");
return (int) (ptr - buffer);
}
void show_code(struct pt_regs *regs)
{
char *mode = user_mode(regs) ? "User" : "Krnl";
unsigned char code[64];
char buffer[128], *ptr;
mm_segment_t old_fs;
unsigned long addr;
int start, end, opsize, hops, i;
/* Get a snapshot of the 64 bytes surrounding the fault address. */
old_fs = get_fs();
set_fs(user_mode(regs) ? USER_DS : KERNEL_DS);
for (start = 32; start && regs->psw.addr >= 34 - start; start -= 2) {
addr = regs->psw.addr - 34 + start;
if (__copy_from_user(code + start - 2,
(char __user *) addr, 2))
break;
}
for (end = 32; end < 64; end += 2) {
addr = regs->psw.addr + end - 32;
if (__copy_from_user(code + end,
(char __user *) addr, 2))
break;
}
set_fs(old_fs);
/* Code snapshot useable ? */
if ((regs->psw.addr & 1) || start >= end) {
printk("%s Code: Bad PSW.\n", mode);
return;
}
/* Find a starting point for the disassembly. */
while (start < 32) {
for (i = 0, hops = 0; start + i < 32 && hops < 3; hops++) {
if (!find_insn(code + start + i))
break;
i += insn_length(code[start + i]);
}
if (start + i == 32)
/* Looks good, sequence ends at PSW. */
break;
start += 2;
}
/* Decode the instructions. */
ptr = buffer;
ptr += sprintf(ptr, "%s Code:", mode);
hops = 0;
while (start < end && hops < 8) {
opsize = insn_length(code[start]);
if (start + opsize == 32)
*ptr++ = '#';
else if (start == 32)
*ptr++ = '>';
else
*ptr++ = ' ';
addr = regs->psw.addr + start - 32;
ptr += sprintf(ptr, "%px: ", (void *)addr);
if (start + opsize >= end)
break;
for (i = 0; i < opsize; i++)
ptr += sprintf(ptr, "%02x", code[start + i]);
*ptr++ = '\t';
if (i < 6)
*ptr++ = '\t';
ptr += print_insn(ptr, code + start, addr);
start += opsize;
pr_cont("%s", buffer);
ptr = buffer;
ptr += sprintf(ptr, "\n ");
hops++;
}
pr_cont("\n");
}
void print_fn_code(unsigned char *code, unsigned long len)
{
char buffer[128], *ptr;
int opsize, i;
while (len) {
ptr = buffer;
opsize = insn_length(*code);
if (opsize > len)
break;
ptr += sprintf(ptr, "%px: ", code);
for (i = 0; i < opsize; i++)
ptr += sprintf(ptr, "%02x", code[i]);
*ptr++ = '\t';
if (i < 4)
*ptr++ = '\t';
ptr += print_insn(ptr, code, (unsigned long) code);
*ptr++ = '\n';
*ptr++ = 0;
printk("%s", buffer);
code += opsize;
len -= opsize;
}
}
+191
View File
@@ -0,0 +1,191 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Stack dumping functions
*
* Copyright IBM Corp. 1999, 2013
*/
#include <linux/kallsyms.h>
#include <linux/hardirq.h>
#include <linux/kprobes.h>
#include <linux/utsname.h>
#include <linux/export.h>
#include <linux/kdebug.h>
#include <linux/ptrace.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/sched/task_stack.h>
#include <asm/processor.h>
#include <asm/debug.h>
#include <asm/dis.h>
#include <asm/ipl.h>
/*
* For dump_trace we have tree different stack to consider:
* - the panic stack which is used if the kernel stack has overflown
* - the asynchronous interrupt stack (cpu related)
* - the synchronous kernel stack (process related)
* The stack trace can start at any of the three stacks and can potentially
* touch all of them. The order is: panic stack, async stack, sync stack.
*/
static unsigned long
__dump_trace(dump_trace_func_t func, void *data, unsigned long sp,
unsigned long low, unsigned long high)
{
struct stack_frame *sf;
struct pt_regs *regs;
while (1) {
if (sp < low || sp > high - sizeof(*sf))
return sp;
sf = (struct stack_frame *) sp;
if (func(data, sf->gprs[8], 0))
return sp;
/* Follow the backchain. */
while (1) {
low = sp;
sp = sf->back_chain;
if (!sp)
break;
if (sp <= low || sp > high - sizeof(*sf))
return sp;
sf = (struct stack_frame *) sp;
if (func(data, sf->gprs[8], 1))
return sp;
}
/* Zero backchain detected, check for interrupt frame. */
sp = (unsigned long) (sf + 1);
if (sp <= low || sp > high - sizeof(*regs))
return sp;
regs = (struct pt_regs *) sp;
if (!user_mode(regs)) {
if (func(data, regs->psw.addr, 1))
return sp;
}
low = sp;
sp = regs->gprs[15];
}
}
void dump_trace(dump_trace_func_t func, void *data, struct task_struct *task,
unsigned long sp)
{
unsigned long frame_size;
frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
#ifdef CONFIG_CHECK_STACK
sp = __dump_trace(func, data, sp,
S390_lowcore.panic_stack + frame_size - PAGE_SIZE,
S390_lowcore.panic_stack + frame_size);
#endif
sp = __dump_trace(func, data, sp,
S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
S390_lowcore.async_stack + frame_size);
task = task ?: current;
__dump_trace(func, data, sp,
(unsigned long)task_stack_page(task),
(unsigned long)task_stack_page(task) + THREAD_SIZE);
}
EXPORT_SYMBOL_GPL(dump_trace);
static int show_address(void *data, unsigned long address, int reliable)
{
if (reliable)
printk(" [<%016lx>] %pSR \n", address, (void *)address);
else
printk("([<%016lx>] %pSR)\n", address, (void *)address);
return 0;
}
void show_stack(struct task_struct *task, unsigned long *stack)
{
unsigned long sp = (unsigned long) stack;
if (!sp)
sp = task ? task->thread.ksp : current_stack_pointer();
printk("Call Trace:\n");
dump_trace(show_address, NULL, task, sp);
if (!task)
task = current;
debug_show_held_locks(task);
}
static void show_last_breaking_event(struct pt_regs *regs)
{
printk("Last Breaking-Event-Address:\n");
printk(" [<%016lx>] %pSR\n", regs->args[0], (void *)regs->args[0]);
}
void show_registers(struct pt_regs *regs)
{
struct psw_bits *psw = &psw_bits(regs->psw);
char *mode;
mode = user_mode(regs) ? "User" : "Krnl";
printk("%s PSW : %p %p", mode, (void *)regs->psw.mask, (void *)regs->psw.addr);
if (!user_mode(regs))
pr_cont(" (%pSR)", (void *)regs->psw.addr);
pr_cont("\n");
printk(" R:%x T:%x IO:%x EX:%x Key:%x M:%x W:%x "
"P:%x AS:%x CC:%x PM:%x", psw->per, psw->dat, psw->io, psw->ext,
psw->key, psw->mcheck, psw->wait, psw->pstate, psw->as, psw->cc, psw->pm);
pr_cont(" RI:%x EA:%x\n", psw->ri, psw->eaba);
printk("%s GPRS: %016lx %016lx %016lx %016lx\n", mode,
regs->gprs[0], regs->gprs[1], regs->gprs[2], regs->gprs[3]);
printk(" %016lx %016lx %016lx %016lx\n",
regs->gprs[4], regs->gprs[5], regs->gprs[6], regs->gprs[7]);
printk(" %016lx %016lx %016lx %016lx\n",
regs->gprs[8], regs->gprs[9], regs->gprs[10], regs->gprs[11]);
printk(" %016lx %016lx %016lx %016lx\n",
regs->gprs[12], regs->gprs[13], regs->gprs[14], regs->gprs[15]);
show_code(regs);
}
void show_regs(struct pt_regs *regs)
{
show_regs_print_info(KERN_DEFAULT);
show_registers(regs);
/* Show stack backtrace if pt_regs is from kernel mode */
if (!user_mode(regs))
show_stack(NULL, (unsigned long *) regs->gprs[15]);
show_last_breaking_event(regs);
}
static DEFINE_SPINLOCK(die_lock);
void die(struct pt_regs *regs, const char *str)
{
static int die_counter;
oops_enter();
lgr_info_log();
debug_stop_all();
console_verbose();
spin_lock_irq(&die_lock);
bust_spinlocks(1);
printk("%s: %04x ilc:%d [#%d] ", str, regs->int_code & 0xffff,
regs->int_code >> 17, ++die_counter);
#ifdef CONFIG_PREEMPT
pr_cont("PREEMPT ");
#endif
#ifdef CONFIG_SMP
pr_cont("SMP ");
#endif
if (debug_pagealloc_enabled())
pr_cont("DEBUG_PAGEALLOC");
pr_cont("\n");
notify_die(DIE_OOPS, str, regs, 0, regs->int_code & 0xffff, SIGSEGV);
print_modules();
show_regs(regs);
bust_spinlocks(0);
add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE);
spin_unlock_irq(&die_lock);
if (in_interrupt())
panic("Fatal exception in interrupt");
if (panic_on_oops)
panic("Fatal exception: panic_on_oops");
oops_exit();
do_exit(SIGSEGV);
}
+366
View File
@@ -0,0 +1,366 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2007, 2009
* Author(s): Hongjie Yang <hongjie@us.ibm.com>,
* Heiko Carstens <heiko.carstens@de.ibm.com>
*/
#define KMSG_COMPONENT "setup"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/compiler.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/ctype.h>
#include <linux/lockdep.h>
#include <linux/extable.h>
#include <linux/pfn.h>
#include <linux/uaccess.h>
#include <linux/kernel.h>
#include <asm/diag.h>
#include <asm/ebcdic.h>
#include <asm/ipl.h>
#include <asm/lowcore.h>
#include <asm/processor.h>
#include <asm/sections.h>
#include <asm/setup.h>
#include <asm/sysinfo.h>
#include <asm/cpcmd.h>
#include <asm/sclp.h>
#include <asm/facility.h>
#include "entry.h"
static void __init setup_boot_command_line(void);
/*
* Initialize storage key for kernel pages
*/
static noinline __init void init_kernel_storage_key(void)
{
#if PAGE_DEFAULT_KEY
unsigned long end_pfn, init_pfn;
end_pfn = PFN_UP(__pa(_end));
for (init_pfn = 0 ; init_pfn < end_pfn; init_pfn++)
page_set_storage_key(init_pfn << PAGE_SHIFT,
PAGE_DEFAULT_KEY, 0);
#endif
}
static __initdata char sysinfo_page[PAGE_SIZE] __aligned(PAGE_SIZE);
static noinline __init void detect_machine_type(void)
{
struct sysinfo_3_2_2 *vmms = (struct sysinfo_3_2_2 *)&sysinfo_page;
/* Check current-configuration-level */
if (stsi(NULL, 0, 0, 0) <= 2) {
S390_lowcore.machine_flags |= MACHINE_FLAG_LPAR;
return;
}
/* Get virtual-machine cpu information. */
if (stsi(vmms, 3, 2, 2) || !vmms->count)
return;
/* Detect known hypervisors */
if (!memcmp(vmms->vm[0].cpi, "\xd2\xe5\xd4", 3))
S390_lowcore.machine_flags |= MACHINE_FLAG_KVM;
else if (!memcmp(vmms->vm[0].cpi, "\xa9\x61\xe5\xd4", 4))
S390_lowcore.machine_flags |= MACHINE_FLAG_VM;
}
/* Remove leading, trailing and double whitespace. */
static inline void strim_all(char *str)
{
char *s;
s = strim(str);
if (s != str)
memmove(str, s, strlen(s));
while (*str) {
if (!isspace(*str++))
continue;
if (isspace(*str)) {
s = skip_spaces(str);
memmove(str, s, strlen(s) + 1);
}
}
}
static noinline __init void setup_arch_string(void)
{
struct sysinfo_1_1_1 *mach = (struct sysinfo_1_1_1 *)&sysinfo_page;
struct sysinfo_3_2_2 *vm = (struct sysinfo_3_2_2 *)&sysinfo_page;
char mstr[80], hvstr[17];
if (stsi(mach, 1, 1, 1))
return;
EBCASC(mach->manufacturer, sizeof(mach->manufacturer));
EBCASC(mach->type, sizeof(mach->type));
EBCASC(mach->model, sizeof(mach->model));
EBCASC(mach->model_capacity, sizeof(mach->model_capacity));
sprintf(mstr, "%-16.16s %-4.4s %-16.16s %-16.16s",
mach->manufacturer, mach->type,
mach->model, mach->model_capacity);
strim_all(mstr);
if (stsi(vm, 3, 2, 2) == 0 && vm->count) {
EBCASC(vm->vm[0].cpi, sizeof(vm->vm[0].cpi));
sprintf(hvstr, "%-16.16s", vm->vm[0].cpi);
strim_all(hvstr);
} else {
sprintf(hvstr, "%s",
MACHINE_IS_LPAR ? "LPAR" :
MACHINE_IS_VM ? "z/VM" :
MACHINE_IS_KVM ? "KVM" : "unknown");
}
dump_stack_set_arch_desc("%s (%s)", mstr, hvstr);
}
static __init void setup_topology(void)
{
int max_mnest;
if (!test_facility(11))
return;
S390_lowcore.machine_flags |= MACHINE_FLAG_TOPOLOGY;
for (max_mnest = 6; max_mnest > 1; max_mnest--) {
if (stsi(&sysinfo_page, 15, 1, max_mnest) == 0)
break;
}
topology_max_mnest = max_mnest;
}
static void early_pgm_check_handler(void)
{
const struct exception_table_entry *fixup;
unsigned long cr0, cr0_new;
unsigned long addr;
addr = S390_lowcore.program_old_psw.addr;
fixup = search_exception_tables(addr);
if (!fixup)
disabled_wait(0);
/* Disable low address protection before storing into lowcore. */
__ctl_store(cr0, 0, 0);
cr0_new = cr0 & ~(1UL << 28);
__ctl_load(cr0_new, 0, 0);
S390_lowcore.program_old_psw.addr = extable_fixup(fixup);
__ctl_load(cr0, 0, 0);
}
static noinline __init void setup_lowcore_early(void)
{
psw_t psw;
psw.mask = PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA;
if (IS_ENABLED(CONFIG_KASAN))
psw.mask |= PSW_MASK_DAT;
psw.addr = (unsigned long) s390_base_ext_handler;
S390_lowcore.external_new_psw = psw;
psw.addr = (unsigned long) s390_base_pgm_handler;
S390_lowcore.program_new_psw = psw;
s390_base_pgm_handler_fn = early_pgm_check_handler;
S390_lowcore.preempt_count = INIT_PREEMPT_COUNT;
}
static noinline __init void setup_facility_list(void)
{
stfle(S390_lowcore.stfle_fac_list,
ARRAY_SIZE(S390_lowcore.stfle_fac_list));
memcpy(S390_lowcore.alt_stfle_fac_list,
S390_lowcore.stfle_fac_list,
sizeof(S390_lowcore.alt_stfle_fac_list));
if (!IS_ENABLED(CONFIG_KERNEL_NOBP))
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
}
static __init void detect_diag9c(void)
{
unsigned int cpu_address;
int rc;
cpu_address = stap();
diag_stat_inc(DIAG_STAT_X09C);
asm volatile(
" diag %2,0,0x9c\n"
"0: la %0,0\n"
"1:\n"
EX_TABLE(0b,1b)
: "=d" (rc) : "0" (-EOPNOTSUPP), "d" (cpu_address) : "cc");
if (!rc)
S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG9C;
}
static __init void detect_diag44(void)
{
int rc;
diag_stat_inc(DIAG_STAT_X044);
asm volatile(
" diag 0,0,0x44\n"
"0: la %0,0\n"
"1:\n"
EX_TABLE(0b,1b)
: "=d" (rc) : "0" (-EOPNOTSUPP) : "cc");
if (!rc)
S390_lowcore.machine_flags |= MACHINE_FLAG_DIAG44;
}
static __init void detect_machine_facilities(void)
{
if (test_facility(8)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT1;
__ctl_set_bit(0, 23);
}
if (test_facility(78))
S390_lowcore.machine_flags |= MACHINE_FLAG_EDAT2;
if (test_facility(3))
S390_lowcore.machine_flags |= MACHINE_FLAG_IDTE;
if (test_facility(50) && test_facility(73)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_TE;
__ctl_set_bit(0, 55);
}
if (test_facility(51))
S390_lowcore.machine_flags |= MACHINE_FLAG_TLB_LC;
if (test_facility(129)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_VX;
__ctl_set_bit(0, 17);
}
if (test_facility(130)) {
S390_lowcore.machine_flags |= MACHINE_FLAG_NX;
__ctl_set_bit(0, 20);
}
if (test_facility(133))
S390_lowcore.machine_flags |= MACHINE_FLAG_GS;
if (test_facility(139) && (tod_clock_base[1] & 0x80)) {
/* Enabled signed clock comparator comparisons */
S390_lowcore.machine_flags |= MACHINE_FLAG_SCC;
clock_comparator_max = -1ULL >> 1;
__ctl_set_bit(0, 53);
}
}
static inline void save_vector_registers(void)
{
#ifdef CONFIG_CRASH_DUMP
if (test_facility(129))
save_vx_regs(boot_cpu_vector_save_area);
#endif
}
static int __init disable_vector_extension(char *str)
{
S390_lowcore.machine_flags &= ~MACHINE_FLAG_VX;
__ctl_clear_bit(0, 17);
return 0;
}
early_param("novx", disable_vector_extension);
static int __init noexec_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (!rc && !enabled) {
/* Disable no-execute support */
S390_lowcore.machine_flags &= ~MACHINE_FLAG_NX;
__ctl_clear_bit(0, 20);
}
return rc;
}
early_param("noexec", noexec_setup);
static int __init cad_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (!rc && enabled && test_facility(128))
/* Enable problem state CAD. */
__ctl_set_bit(2, 3);
return rc;
}
early_param("cad", cad_setup);
/* Set up boot command line */
static void __init append_to_cmdline(size_t (*ipl_data)(char *, size_t))
{
char *parm, *delim;
size_t rc, len;
len = strlen(boot_command_line);
delim = boot_command_line + len; /* '\0' character position */
parm = boot_command_line + len + 1; /* append right after '\0' */
rc = ipl_data(parm, COMMAND_LINE_SIZE - len - 1);
if (rc) {
if (*parm == '=')
memmove(boot_command_line, parm + 1, rc);
else
*delim = ' '; /* replace '\0' with space */
}
}
static inline int has_ebcdic_char(const char *str)
{
int i;
for (i = 0; str[i]; i++)
if (str[i] & 0x80)
return 1;
return 0;
}
static void __init setup_boot_command_line(void)
{
COMMAND_LINE[ARCH_COMMAND_LINE_SIZE - 1] = 0;
/* convert arch command line to ascii if necessary */
if (has_ebcdic_char(COMMAND_LINE))
EBCASC(COMMAND_LINE, ARCH_COMMAND_LINE_SIZE);
/* copy arch command line */
strlcpy(boot_command_line, strstrip(COMMAND_LINE),
ARCH_COMMAND_LINE_SIZE);
/* append IPL PARM data to the boot command line */
if (MACHINE_IS_VM)
append_to_cmdline(append_ipl_vmparm);
append_to_cmdline(append_ipl_scpdata);
}
static void __init check_image_bootable(void)
{
if (!memcmp(EP_STRING, (void *)EP_OFFSET, strlen(EP_STRING)))
return;
sclp_early_printk("Linux kernel boot failure: An attempt to boot a vmlinux ELF image failed.\n");
sclp_early_printk("This image does not contain all parts necessary for starting up. Use\n");
sclp_early_printk("bzImage or arch/s390/boot/compressed/vmlinux instead.\n");
disabled_wait(0xbadb007);
}
void __init startup_init(void)
{
check_image_bootable();
time_early_init();
init_kernel_storage_key();
lockdep_off();
setup_lowcore_early();
setup_facility_list();
detect_machine_type();
setup_arch_string();
ipl_store_parameters();
setup_boot_command_line();
detect_diag9c();
detect_diag44();
detect_machine_facilities();
save_vector_registers();
setup_topology();
sclp_early_detect();
lockdep_on();
}
+65
View File
@@ -0,0 +1,65 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2007, 2018
*/
/*
* Early setup functions which may not rely on an initialized bss
* section. The last thing that is supposed to happen here is
* initialization of the bss section.
*/
#include <linux/processor.h>
#include <linux/string.h>
#include <asm/sections.h>
#include <asm/lowcore.h>
#include <asm/setup.h>
#include <asm/timex.h>
#include "entry.h"
static void __init reset_tod_clock(void)
{
u64 time;
if (store_tod_clock(&time) == 0)
return;
/* TOD clock not running. Set the clock to Unix Epoch. */
if (set_tod_clock(TOD_UNIX_EPOCH) != 0 || store_tod_clock(&time) != 0)
disabled_wait(0);
memset(tod_clock_base, 0, 16);
*(__u64 *) &tod_clock_base[1] = TOD_UNIX_EPOCH;
S390_lowcore.last_update_clock = TOD_UNIX_EPOCH;
}
static void __init rescue_initrd(void)
{
unsigned long min_initrd_addr = (unsigned long) _end + (4UL << 20);
/*
* Just like in case of IPL from VM reader we make sure there is a
* gap of 4MB between end of kernel and start of initrd.
* That way we can also be sure that saving an NSS will succeed,
* which however only requires different segments.
*/
if (!IS_ENABLED(CONFIG_BLK_DEV_INITRD))
return;
if (!INITRD_START || !INITRD_SIZE)
return;
if (INITRD_START >= min_initrd_addr)
return;
memmove((void *) min_initrd_addr, (void *) INITRD_START, INITRD_SIZE);
INITRD_START = min_initrd_addr;
}
static void __init clear_bss_section(void)
{
memset(__bss_start, 0, __bss_stop - __bss_start);
}
void __init startup_init_nobss(void)
{
reset_tod_clock();
rescue_initrd();
clear_bss_section();
}
+36
View File
@@ -0,0 +1,36 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2017
*/
#include <linux/console.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <asm/sclp.h>
static void sclp_early_write(struct console *con, const char *s, unsigned int len)
{
__sclp_early_printk(s, len, 0);
}
static struct console sclp_early_console = {
.name = "earlysclp",
.write = sclp_early_write,
.flags = CON_PRINTBUFFER | CON_BOOT,
.index = -1,
};
static int __init setup_early_printk(char *buf)
{
if (early_console)
return 0;
/* Accept only "earlyprintk" and "earlyprintk=sclp" */
if (buf && strncmp(buf, "sclp", 4))
return 0;
if (!sclp.has_linemode && !sclp.has_vt220)
return 0;
early_console = &sclp_early_console;
register_console(early_console);
return 0;
}
early_param("earlyprintk", setup_early_printk);
+401
View File
@@ -0,0 +1,401 @@
// SPDX-License-Identifier: GPL-2.0
/*
* ECBDIC -> ASCII, ASCII -> ECBDIC,
* upper to lower case (EBCDIC) conversion tables.
*
* S390 version
* Copyright IBM Corp. 1999
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
* Martin Peschke <peschke@fh-brandenburg.de>
*/
#include <linux/types.h>
#include <linux/export.h>
#include <asm/ebcdic.h>
/*
* ASCII (IBM PC 437) -> EBCDIC 037
*/
__u8 _ascebc[256] =
{
/*00 NUL SOH STX ETX EOT ENQ ACK BEL */
0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F,
/*08 BS HT LF VT FF CR SO SI */
/* ->NL */
0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
/*10 DLE DC1 DC2 DC3 DC4 NAK SYN ETB */
0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26,
/*18 CAN EM SUB ESC FS GS RS US */
/* ->IGS ->IRS ->IUS */
0x18, 0x19, 0x3F, 0x27, 0x22, 0x1D, 0x1E, 0x1F,
/*20 SP ! " # $ % & ' */
0x40, 0x5A, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D,
/*28 ( ) * + , - . / */
0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
/*30 0 1 2 3 4 5 6 7 */
0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
/*38 8 9 : ; < = > ? */
0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
/*40 @ A B C D E F G */
0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
/*48 H I J K L M N O */
0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
/*50 P Q R S T U V W */
0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6,
/*58 X Y Z [ \ ] ^ _ */
0xE7, 0xE8, 0xE9, 0xBA, 0xE0, 0xBB, 0xB0, 0x6D,
/*60 ` a b c d e f g */
0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
/*68 h i j k l m n o */
0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
/*70 p q r s t u v w */
0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6,
/*78 x y z { | } ~ DL */
0xA7, 0xA8, 0xA9, 0xC0, 0x4F, 0xD0, 0xA1, 0x07,
/*80*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*88*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*90*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*98*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*A0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*A8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*B0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*B8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*C0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*C8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*D0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*D8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*E0 sz */
0x3F, 0x59, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*E8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*F0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*F8*/
0x90, 0x3F, 0x3F, 0x3F, 0x3F, 0xEA, 0x3F, 0xFF
};
/*
* EBCDIC 037 -> ASCII (IBM PC 437)
*/
__u8 _ebcasc[256] =
{
/* 0x00 NUL SOH STX ETX *SEL HT *RNL DEL */
0x00, 0x01, 0x02, 0x03, 0x07, 0x09, 0x07, 0x7F,
/* 0x08 -GE -SPS -RPT VT FF CR SO SI */
0x07, 0x07, 0x07, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
/* 0x10 DLE DC1 DC2 DC3 -RES -NL BS -POC
-ENP ->LF */
0x10, 0x11, 0x12, 0x13, 0x07, 0x0A, 0x08, 0x07,
/* 0x18 CAN EM -UBS -CU1 -IFS -IGS -IRS -ITB
-IUS */
0x18, 0x19, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07,
/* 0x20 -DS -SOS FS -WUS -BYP LF ETB ESC
-INP */
0x07, 0x07, 0x1C, 0x07, 0x07, 0x0A, 0x17, 0x1B,
/* 0x28 -SA -SFE -SM -CSP -MFA ENQ ACK BEL
-SW */
0x07, 0x07, 0x07, 0x07, 0x07, 0x05, 0x06, 0x07,
/* 0x30 ---- ---- SYN -IR -PP -TRN -NBS EOT */
0x07, 0x07, 0x16, 0x07, 0x07, 0x07, 0x07, 0x04,
/* 0x38 -SBS -IT -RFF -CU3 DC4 NAK ---- SUB */
0x07, 0x07, 0x07, 0x07, 0x14, 0x15, 0x07, 0x1A,
/* 0x40 SP RSP ä ---- */
0x20, 0xFF, 0x83, 0x84, 0x85, 0xA0, 0x07, 0x86,
/* 0x48 . < ( + | */
0x87, 0xA4, 0x9B, 0x2E, 0x3C, 0x28, 0x2B, 0x7C,
/* 0x50 & ---- */
0x26, 0x82, 0x88, 0x89, 0x8A, 0xA1, 0x8C, 0x07,
/* 0x58 ß ! $ * ) ; */
0x8D, 0xE1, 0x21, 0x24, 0x2A, 0x29, 0x3B, 0xAA,
/* 0x60 - / ---- Ä ---- ---- ---- */
0x2D, 0x2F, 0x07, 0x8E, 0x07, 0x07, 0x07, 0x8F,
/* 0x68 ---- , % _ > ? */
0x80, 0xA5, 0x07, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
/* 0x70 ---- ---- ---- ---- ---- ---- ---- */
0x07, 0x90, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07,
/* 0x78 * ` : # @ ' = " */
0x70, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
/* 0x80 * a b c d e f g */
0x07, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
/* 0x88 h i ---- ---- ---- */
0x68, 0x69, 0xAE, 0xAF, 0x07, 0x07, 0x07, 0xF1,
/* 0x90 ° j k l m n o p */
0xF8, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70,
/* 0x98 q r ---- ---- */
0x71, 0x72, 0xA6, 0xA7, 0x91, 0x07, 0x92, 0x07,
/* 0xA0 ~ s t u v w x */
0xE6, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78,
/* 0xA8 y z ---- ---- ---- ---- */
0x79, 0x7A, 0xAD, 0xAB, 0x07, 0x07, 0x07, 0x07,
/* 0xB0 ^ ---- § ---- */
0x5E, 0x9C, 0x9D, 0xFA, 0x07, 0x07, 0x07, 0xAC,
/* 0xB8 ---- [ ] ---- ---- ---- ---- */
0xAB, 0x07, 0x5B, 0x5D, 0x07, 0x07, 0x07, 0x07,
/* 0xC0 { A B C D E F G */
0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
/* 0xC8 H I ---- ö ---- */
0x48, 0x49, 0x07, 0x93, 0x94, 0x95, 0xA2, 0x07,
/* 0xD0 } J K L M N O P */
0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50,
/* 0xD8 Q R ---- ü */
0x51, 0x52, 0x07, 0x96, 0x81, 0x97, 0xA3, 0x98,
/* 0xE0 \ S T U V W X */
0x5C, 0xF6, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58,
/* 0xE8 Y Z ---- Ö ---- ---- ---- */
0x59, 0x5A, 0xFD, 0x07, 0x99, 0x07, 0x07, 0x07,
/* 0xF0 0 1 2 3 4 5 6 7 */
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
/* 0xF8 8 9 ---- ---- Ü ---- ---- ---- */
0x38, 0x39, 0x07, 0x07, 0x9A, 0x07, 0x07, 0x07
};
/*
* ASCII (IBM PC 437) -> EBCDIC 500
*/
__u8 _ascebc_500[256] =
{
/*00 NUL SOH STX ETX EOT ENQ ACK BEL */
0x00, 0x01, 0x02, 0x03, 0x37, 0x2D, 0x2E, 0x2F,
/*08 BS HT LF VT FF CR SO SI */
/* ->NL */
0x16, 0x05, 0x15, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
/*10 DLE DC1 DC2 DC3 DC4 NAK SYN ETB */
0x10, 0x11, 0x12, 0x13, 0x3C, 0x3D, 0x32, 0x26,
/*18 CAN EM SUB ESC FS GS RS US */
/* ->IGS ->IRS ->IUS */
0x18, 0x19, 0x3F, 0x27, 0x22, 0x1D, 0x1E, 0x1F,
/*20 SP ! " # $ % & ' */
0x40, 0x4F, 0x7F, 0x7B, 0x5B, 0x6C, 0x50, 0x7D,
/*28 ( ) * + , - . / */
0x4D, 0x5D, 0x5C, 0x4E, 0x6B, 0x60, 0x4B, 0x61,
/*30 0 1 2 3 4 5 6 7 */
0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
/*38 8 9 : ; < = > ? */
0xF8, 0xF9, 0x7A, 0x5E, 0x4C, 0x7E, 0x6E, 0x6F,
/*40 @ A B C D E F G */
0x7C, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
/*48 H I J K L M N O */
0xC8, 0xC9, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6,
/*50 P Q R S T U V W */
0xD7, 0xD8, 0xD9, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6,
/*58 X Y Z [ \ ] ^ _ */
0xE7, 0xE8, 0xE9, 0x4A, 0xE0, 0x5A, 0x5F, 0x6D,
/*60 ` a b c d e f g */
0x79, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
/*68 h i j k l m n o */
0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96,
/*70 p q r s t u v w */
0x97, 0x98, 0x99, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6,
/*78 x y z { | } ~ DL */
0xA7, 0xA8, 0xA9, 0xC0, 0xBB, 0xD0, 0xA1, 0x07,
/*80*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*88*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*90*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*98*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*A0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*A8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*B0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*B8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*C0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*C8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*D0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*D8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*E0 sz */
0x3F, 0x59, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*E8*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*F0*/
0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F, 0x3F,
/*F8*/
0x90, 0x3F, 0x3F, 0x3F, 0x3F, 0xEA, 0x3F, 0xFF
};
/*
* EBCDIC 500 -> ASCII (IBM PC 437)
*/
__u8 _ebcasc_500[256] =
{
/* 0x00 NUL SOH STX ETX *SEL HT *RNL DEL */
0x00, 0x01, 0x02, 0x03, 0x07, 0x09, 0x07, 0x7F,
/* 0x08 -GE -SPS -RPT VT FF CR SO SI */
0x07, 0x07, 0x07, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
/* 0x10 DLE DC1 DC2 DC3 -RES -NL BS -POC
-ENP ->LF */
0x10, 0x11, 0x12, 0x13, 0x07, 0x0A, 0x08, 0x07,
/* 0x18 CAN EM -UBS -CU1 -IFS -IGS -IRS -ITB
-IUS */
0x18, 0x19, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07,
/* 0x20 -DS -SOS FS -WUS -BYP LF ETB ESC
-INP */
0x07, 0x07, 0x1C, 0x07, 0x07, 0x0A, 0x17, 0x1B,
/* 0x28 -SA -SFE -SM -CSP -MFA ENQ ACK BEL
-SW */
0x07, 0x07, 0x07, 0x07, 0x07, 0x05, 0x06, 0x07,
/* 0x30 ---- ---- SYN -IR -PP -TRN -NBS EOT */
0x07, 0x07, 0x16, 0x07, 0x07, 0x07, 0x07, 0x04,
/* 0x38 -SBS -IT -RFF -CU3 DC4 NAK ---- SUB */
0x07, 0x07, 0x07, 0x07, 0x14, 0x15, 0x07, 0x1A,
/* 0x40 SP RSP ä ---- */
0x20, 0xFF, 0x83, 0x84, 0x85, 0xA0, 0x07, 0x86,
/* 0x48 [ . < ( + ! */
0x87, 0xA4, 0x5B, 0x2E, 0x3C, 0x28, 0x2B, 0x21,
/* 0x50 & ---- */
0x26, 0x82, 0x88, 0x89, 0x8A, 0xA1, 0x8C, 0x07,
/* 0x58 ß ] $ * ) ; ^ */
0x8D, 0xE1, 0x5D, 0x24, 0x2A, 0x29, 0x3B, 0x5E,
/* 0x60 - / ---- Ä ---- ---- ---- */
0x2D, 0x2F, 0x07, 0x8E, 0x07, 0x07, 0x07, 0x8F,
/* 0x68 ---- , % _ > ? */
0x80, 0xA5, 0x07, 0x2C, 0x25, 0x5F, 0x3E, 0x3F,
/* 0x70 ---- ---- ---- ---- ---- ---- ---- */
0x07, 0x90, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07,
/* 0x78 * ` : # @ ' = " */
0x70, 0x60, 0x3A, 0x23, 0x40, 0x27, 0x3D, 0x22,
/* 0x80 * a b c d e f g */
0x07, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
/* 0x88 h i ---- ---- ---- */
0x68, 0x69, 0xAE, 0xAF, 0x07, 0x07, 0x07, 0xF1,
/* 0x90 ° j k l m n o p */
0xF8, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70,
/* 0x98 q r ---- ---- */
0x71, 0x72, 0xA6, 0xA7, 0x91, 0x07, 0x92, 0x07,
/* 0xA0 ~ s t u v w x */
0xE6, 0x7E, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78,
/* 0xA8 y z ---- ---- ---- ---- */
0x79, 0x7A, 0xAD, 0xAB, 0x07, 0x07, 0x07, 0x07,
/* 0xB0 ---- § ---- */
0x9B, 0x9C, 0x9D, 0xFA, 0x07, 0x07, 0x07, 0xAC,
/* 0xB8 ---- | ---- ---- ---- ---- */
0xAB, 0x07, 0xAA, 0x7C, 0x07, 0x07, 0x07, 0x07,
/* 0xC0 { A B C D E F G */
0x7B, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
/* 0xC8 H I ---- ö ---- */
0x48, 0x49, 0x07, 0x93, 0x94, 0x95, 0xA2, 0x07,
/* 0xD0 } J K L M N O P */
0x7D, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50,
/* 0xD8 Q R ---- ü */
0x51, 0x52, 0x07, 0x96, 0x81, 0x97, 0xA3, 0x98,
/* 0xE0 \ S T U V W X */
0x5C, 0xF6, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58,
/* 0xE8 Y Z ---- Ö ---- ---- ---- */
0x59, 0x5A, 0xFD, 0x07, 0x99, 0x07, 0x07, 0x07,
/* 0xF0 0 1 2 3 4 5 6 7 */
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
/* 0xF8 8 9 ---- ---- Ü ---- ---- ---- */
0x38, 0x39, 0x07, 0x07, 0x9A, 0x07, 0x07, 0x07
};
/*
* EBCDIC 037/500 conversion table:
* from upper to lower case
*/
__u8 _ebc_tolower[256] =
{
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F,
0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F,
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F,
0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F,
0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F,
0x60, 0x61, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
0x48, 0x49, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
0x70, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
0x58, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F,
0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F,
0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
0x98, 0x99, 0x9A, 0x9B, 0x9C, 0x9D, 0x9C, 0x9F,
0xA0, 0xA1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7,
0xA8, 0xA9, 0xAA, 0xAB, 0x8C, 0x8D, 0x8E, 0xAF,
0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7,
0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF,
0xC0, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
0x88, 0x89, 0xCA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF,
0xD0, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97,
0x98, 0x99, 0xDA, 0xDB, 0xDC, 0xDD, 0xDE, 0xDF,
0xE0, 0xE1, 0xA2, 0xA3, 0xA4, 0xA5, 0xA6, 0xA7,
0xA8, 0xA9, 0xEA, 0xCB, 0xCC, 0xCD, 0xCE, 0xCF,
0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
0xF8, 0xF9, 0xFA, 0xDB, 0xDC, 0xDD, 0xDE, 0xFF
};
/*
* EBCDIC 037/500 conversion table:
* from lower to upper case
*/
__u8 _ebc_toupper[256] =
{
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F,
0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F,
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F,
0x40, 0x41, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
0x68, 0x69, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F,
0x50, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
0x78, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x5F,
0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F,
0x80, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
0xC8, 0xC9, 0x8A, 0x8B, 0xAC, 0xAD, 0xAE, 0x8F,
0x90, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7,
0xD8, 0xD9, 0x9A, 0x9B, 0x9E, 0x9D, 0x9E, 0x9F,
0xA0, 0xA1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7,
0xE8, 0xE9, 0xAA, 0xAB, 0xAC, 0xAD, 0xAE, 0xAF,
0xB0, 0xB1, 0xB2, 0xB3, 0xB4, 0xB5, 0xB6, 0xB7,
0xB8, 0xB9, 0xBA, 0xBB, 0xBC, 0xBD, 0xBE, 0xBF,
0xC0, 0xC1, 0xC2, 0xC3, 0xC4, 0xC5, 0xC6, 0xC7,
0xC8, 0xC9, 0xCA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF,
0xD0, 0xD1, 0xD2, 0xD3, 0xD4, 0xD5, 0xD6, 0xD7,
0xD8, 0xD9, 0xDA, 0xFB, 0xFC, 0xFD, 0xFE, 0xDF,
0xE0, 0xE1, 0xE2, 0xE3, 0xE4, 0xE5, 0xE6, 0xE7,
0xE8, 0xE9, 0xEA, 0xEB, 0xEC, 0xED, 0xEE, 0xEF,
0xF0, 0xF1, 0xF2, 0xF3, 0xF4, 0xF5, 0xF6, 0xF7,
0xF8, 0xF9, 0xFA, 0xFB, 0xFC, 0xFD, 0xFE, 0xFF
};
EXPORT_SYMBOL(_ascebc_500);
EXPORT_SYMBOL(_ebcasc_500);
EXPORT_SYMBOL(_ascebc);
EXPORT_SYMBOL(_ebcasc);
EXPORT_SYMBOL(_ebc_tolower);
EXPORT_SYMBOL(_ebc_toupper);
File diff suppressed because it is too large. Load diff
+89
View File
@@ -0,0 +1,89 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _ENTRY_H
#define _ENTRY_H
#include <linux/percpu.h>
#include <linux/types.h>
#include <linux/signal.h>
#include <asm/ptrace.h>
#include <asm/idle.h>
extern void *restart_stack;
extern unsigned long suspend_zero_pages;
void system_call(void);
void pgm_check_handler(void);
void ext_int_handler(void);
void io_int_handler(void);
void mcck_int_handler(void);
void restart_int_handler(void);
void restart_call_handler(void);
asmlinkage long do_syscall_trace_enter(struct pt_regs *regs);
asmlinkage void do_syscall_trace_exit(struct pt_regs *regs);
void do_protection_exception(struct pt_regs *regs);
void do_dat_exception(struct pt_regs *regs);
void addressing_exception(struct pt_regs *regs);
void data_exception(struct pt_regs *regs);
void default_trap_handler(struct pt_regs *regs);
void divide_exception(struct pt_regs *regs);
void execute_exception(struct pt_regs *regs);
void hfp_divide_exception(struct pt_regs *regs);
void hfp_overflow_exception(struct pt_regs *regs);
void hfp_significance_exception(struct pt_regs *regs);
void hfp_sqrt_exception(struct pt_regs *regs);
void hfp_underflow_exception(struct pt_regs *regs);
void illegal_op(struct pt_regs *regs);
void operand_exception(struct pt_regs *regs);
void overflow_exception(struct pt_regs *regs);
void privileged_op(struct pt_regs *regs);
void space_switch_exception(struct pt_regs *regs);
void special_op_exception(struct pt_regs *regs);
void specification_exception(struct pt_regs *regs);
void transaction_exception(struct pt_regs *regs);
void translation_exception(struct pt_regs *regs);
void vector_exception(struct pt_regs *regs);
void do_per_trap(struct pt_regs *regs);
void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str);
void syscall_trace(struct pt_regs *regs, int entryexit);
void kernel_stack_overflow(struct pt_regs * regs);
void do_signal(struct pt_regs *regs);
void handle_signal32(struct ksignal *ksig, sigset_t *oldset,
struct pt_regs *regs);
void do_notify_resume(struct pt_regs *regs);
void __init init_IRQ(void);
void do_IRQ(struct pt_regs *regs, int irq);
void do_restart(void);
void __init startup_init_nobss(void);
void __init startup_init(void);
void die(struct pt_regs *regs, const char *str);
int setup_profiling_timer(unsigned int multiplier);
void __init time_init(void);
int pfn_is_nosave(unsigned long);
void s390_early_resume(void);
unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip);
struct s390_mmap_arg_struct;
struct fadvise64_64_args;
struct old_sigaction;
long sys_rt_sigreturn(void);
long sys_sigreturn(void);
long sys_s390_personality(unsigned int personality);
long sys_s390_runtime_instr(int command, int signum);
long sys_s390_guarded_storage(int command, struct gs_cb __user *);
long sys_s390_pci_mmio_write(unsigned long, const void __user *, size_t);
long sys_s390_pci_mmio_read(unsigned long, void __user *, size_t);
long sys_s390_sthyi(unsigned long function_code, void __user *buffer, u64 __user *return_code, unsigned long flags);
DECLARE_PER_CPU(u64, mt_cycles[8]);
void gs_load_bc_cb(struct pt_regs *regs);
void set_fs_fixup(void);
#endif /* _ENTRY_H */
+177
View File
@@ -0,0 +1,177 @@
// SPDX-License-Identifier: GPL-2.0
/*
* In-kernel vector facility support functions
*
* Copyright IBM Corp. 2015
* Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
*/
#include <linux/kernel.h>
#include <linux/cpu.h>
#include <linux/sched.h>
#include <asm/fpu/types.h>
#include <asm/fpu/api.h>
asm(".include \"asm/vx-insn.h\"\n");
void __kernel_fpu_begin(struct kernel_fpu *state, u32 flags)
{
/*
* Limit the save to the FPU/vector registers already
* in use by the previous context
*/
flags &= state->mask;
if (flags & KERNEL_FPC)
/* Save floating point control */
asm volatile("stfpc %0" : "=m" (state->fpc));
if (!MACHINE_HAS_VX) {
if (flags & KERNEL_VXR_V0V7) {
/* Save floating-point registers */
asm volatile("std 0,%0" : "=Q" (state->fprs[0]));
asm volatile("std 1,%0" : "=Q" (state->fprs[1]));
asm volatile("std 2,%0" : "=Q" (state->fprs[2]));
asm volatile("std 3,%0" : "=Q" (state->fprs[3]));
asm volatile("std 4,%0" : "=Q" (state->fprs[4]));
asm volatile("std 5,%0" : "=Q" (state->fprs[5]));
asm volatile("std 6,%0" : "=Q" (state->fprs[6]));
asm volatile("std 7,%0" : "=Q" (state->fprs[7]));
asm volatile("std 8,%0" : "=Q" (state->fprs[8]));
asm volatile("std 9,%0" : "=Q" (state->fprs[9]));
asm volatile("std 10,%0" : "=Q" (state->fprs[10]));
asm volatile("std 11,%0" : "=Q" (state->fprs[11]));
asm volatile("std 12,%0" : "=Q" (state->fprs[12]));
asm volatile("std 13,%0" : "=Q" (state->fprs[13]));
asm volatile("std 14,%0" : "=Q" (state->fprs[14]));
asm volatile("std 15,%0" : "=Q" (state->fprs[15]));
}
return;
}
/* Test and save vector registers */
asm volatile (
/*
* Test if any vector register must be saved and, if so,
* test if all register can be saved.
*/
" la 1,%[vxrs]\n" /* load save area */
" tmll %[m],30\n" /* KERNEL_VXR */
" jz 7f\n" /* no work -> done */
" jo 5f\n" /* -> save V0..V31 */
/*
* Test for special case KERNEL_FPU_MID only. In this
* case a vstm V8..V23 is the best instruction
*/
" chi %[m],12\n" /* KERNEL_VXR_MID */
" jne 0f\n" /* -> save V8..V23 */
" VSTM 8,23,128,1\n" /* vstm %v8,%v23,128(%r1) */
" j 7f\n"
/* Test and save the first half of 16 vector registers */
"0: tmll %[m],6\n" /* KERNEL_VXR_LOW */
" jz 3f\n" /* -> KERNEL_VXR_HIGH */
" jo 2f\n" /* 11 -> save V0..V15 */
" brc 2,1f\n" /* 10 -> save V8..V15 */
" VSTM 0,7,0,1\n" /* vstm %v0,%v7,0(%r1) */
" j 3f\n"
"1: VSTM 8,15,128,1\n" /* vstm %v8,%v15,128(%r1) */
" j 3f\n"
"2: VSTM 0,15,0,1\n" /* vstm %v0,%v15,0(%r1) */
/* Test and save the second half of 16 vector registers */
"3: tmll %[m],24\n" /* KERNEL_VXR_HIGH */
" jz 7f\n"
" jo 6f\n" /* 11 -> save V16..V31 */
" brc 2,4f\n" /* 10 -> save V24..V31 */
" VSTM 16,23,256,1\n" /* vstm %v16,%v23,256(%r1) */
" j 7f\n"
"4: VSTM 24,31,384,1\n" /* vstm %v24,%v31,384(%r1) */
" j 7f\n"
"5: VSTM 0,15,0,1\n" /* vstm %v0,%v15,0(%r1) */
"6: VSTM 16,31,256,1\n" /* vstm %v16,%v31,256(%r1) */
"7:"
: [vxrs] "=Q" (*(struct vx_array *) &state->vxrs)
: [m] "d" (flags)
: "1", "cc");
}
EXPORT_SYMBOL(__kernel_fpu_begin);
void __kernel_fpu_end(struct kernel_fpu *state, u32 flags)
{
/*
* Limit the restore to the FPU/vector registers of the
* previous context that have been overwritte by the
* current context
*/
flags &= state->mask;
if (flags & KERNEL_FPC)
/* Restore floating-point controls */
asm volatile("lfpc %0" : : "Q" (state->fpc));
if (!MACHINE_HAS_VX) {
if (flags & KERNEL_VXR_V0V7) {
/* Restore floating-point registers */
asm volatile("ld 0,%0" : : "Q" (state->fprs[0]));
asm volatile("ld 1,%0" : : "Q" (state->fprs[1]));
asm volatile("ld 2,%0" : : "Q" (state->fprs[2]));
asm volatile("ld 3,%0" : : "Q" (state->fprs[3]));
asm volatile("ld 4,%0" : : "Q" (state->fprs[4]));
asm volatile("ld 5,%0" : : "Q" (state->fprs[5]));
asm volatile("ld 6,%0" : : "Q" (state->fprs[6]));
asm volatile("ld 7,%0" : : "Q" (state->fprs[7]));
asm volatile("ld 8,%0" : : "Q" (state->fprs[8]));
asm volatile("ld 9,%0" : : "Q" (state->fprs[9]));
asm volatile("ld 10,%0" : : "Q" (state->fprs[10]));
asm volatile("ld 11,%0" : : "Q" (state->fprs[11]));
asm volatile("ld 12,%0" : : "Q" (state->fprs[12]));
asm volatile("ld 13,%0" : : "Q" (state->fprs[13]));
asm volatile("ld 14,%0" : : "Q" (state->fprs[14]));
asm volatile("ld 15,%0" : : "Q" (state->fprs[15]));
}
return;
}
/* Test and restore (load) vector registers */
asm volatile (
/*
* Test if any vector register must be loaded and, if so,
* test if all registers can be loaded at once.
*/
" la 1,%[vxrs]\n" /* load restore area */
" tmll %[m],30\n" /* KERNEL_VXR */
" jz 7f\n" /* no work -> done */
" jo 5f\n" /* -> restore V0..V31 */
/*
* Test for special case KERNEL_FPU_MID only. In this
* case a vlm V8..V23 is the best instruction
*/
" chi %[m],12\n" /* KERNEL_VXR_MID */
" jne 0f\n" /* -> restore V8..V23 */
" VLM 8,23,128,1\n" /* vlm %v8,%v23,128(%r1) */
" j 7f\n"
/* Test and restore the first half of 16 vector registers */
"0: tmll %[m],6\n" /* KERNEL_VXR_LOW */
" jz 3f\n" /* -> KERNEL_VXR_HIGH */
" jo 2f\n" /* 11 -> restore V0..V15 */
" brc 2,1f\n" /* 10 -> restore V8..V15 */
" VLM 0,7,0,1\n" /* vlm %v0,%v7,0(%r1) */
" j 3f\n"
"1: VLM 8,15,128,1\n" /* vlm %v8,%v15,128(%r1) */
" j 3f\n"
"2: VLM 0,15,0,1\n" /* vlm %v0,%v15,0(%r1) */
/* Test and restore the second half of 16 vector registers */
"3: tmll %[m],24\n" /* KERNEL_VXR_HIGH */
" jz 7f\n"
" jo 6f\n" /* 11 -> restore V16..V31 */
" brc 2,4f\n" /* 10 -> restore V24..V31 */
" VLM 16,23,256,1\n" /* vlm %v16,%v23,256(%r1) */
" j 7f\n"
"4: VLM 24,31,384,1\n" /* vlm %v24,%v31,384(%r1) */
" j 7f\n"
"5: VLM 0,15,0,1\n" /* vlm %v0,%v15,0(%r1) */
"6: VLM 16,31,256,1\n" /* vlm %v16,%v31,256(%r1) */
"7:"
: [vxrs] "=Q" (*(struct vx_array *) &state->vxrs)
: [m] "d" (flags)
: "1", "cc");
}
EXPORT_SYMBOL(__kernel_fpu_end);
+242
View File
@@ -0,0 +1,242 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Dynamic function tracer architecture backend.
*
* Copyright IBM Corp. 2009,2014
*
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
* Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <linux/moduleloader.h>
#include <linux/hardirq.h>
#include <linux/uaccess.h>
#include <linux/ftrace.h>
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/kprobes.h>
#include <trace/syscall.h>
#include <asm/asm-offsets.h>
#include <asm/cacheflush.h>
#include <asm/set_memory.h>
#include "entry.h"
/*
* The mcount code looks like this:
* stg %r14,8(%r15) # offset 0
* larl %r1,<&counter> # offset 6
* brasl %r14,_mcount # offset 12
* lg %r14,8(%r15) # offset 18
* Total length is 24 bytes. Only the first instruction will be patched
* by ftrace_make_call / ftrace_make_nop.
* The enabled ftrace code block looks like this:
* > brasl %r0,ftrace_caller # offset 0
* larl %r1,<&counter> # offset 6
* brasl %r14,_mcount # offset 12
* lg %r14,8(%r15) # offset 18
* The ftrace function gets called with a non-standard C function call ABI
* where r0 contains the return address. It is also expected that the called
* function only clobbers r0 and r1, but restores r2-r15.
* For module code we can't directly jump to ftrace caller, but need a
* trampoline (ftrace_plt), which clobbers also r1.
* The return point of the ftrace function has offset 24, so execution
* continues behind the mcount block.
* The disabled ftrace code block looks like this:
* > jg .+24 # offset 0
* larl %r1,<&counter> # offset 6
* brasl %r14,_mcount # offset 12
* lg %r14,8(%r15) # offset 18
* The jg instruction branches to offset 24 to skip as many instructions
* as possible.
* In case we use gcc's hotpatch feature the original and also the disabled
* function prologue contains only a single six byte instruction and looks
* like this:
* > brcl 0,0 # offset 0
* To enable ftrace the code gets patched like above and afterwards looks
* like this:
* > brasl %r0,ftrace_caller # offset 0
*/
unsigned long ftrace_plt;
static inline void ftrace_generate_orig_insn(struct ftrace_insn *insn)
{
#if defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT)
/* brcl 0,0 */
insn->opc = 0xc004;
insn->disp = 0;
#else
/* stg r14,8(r15) */
insn->opc = 0xe3e0;
insn->disp = 0xf0080024;
#endif
}
static inline int is_kprobe_on_ftrace(struct ftrace_insn *insn)
{
#ifdef CONFIG_KPROBES
if (insn->opc == BREAKPOINT_INSTRUCTION)
return 1;
#endif
return 0;
}
static inline void ftrace_generate_kprobe_nop_insn(struct ftrace_insn *insn)
{
#ifdef CONFIG_KPROBES
insn->opc = BREAKPOINT_INSTRUCTION;
insn->disp = KPROBE_ON_FTRACE_NOP;
#endif
}
static inline void ftrace_generate_kprobe_call_insn(struct ftrace_insn *insn)
{
#ifdef CONFIG_KPROBES
insn->opc = BREAKPOINT_INSTRUCTION;
insn->disp = KPROBE_ON_FTRACE_CALL;
#endif
}
int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
unsigned long addr)
{
return 0;
}
int ftrace_make_nop(struct module *mod, struct dyn_ftrace *rec,
unsigned long addr)
{
struct ftrace_insn orig, new, old;
if (probe_kernel_read(&old, (void *) rec->ip, sizeof(old)))
return -EFAULT;
if (addr == MCOUNT_ADDR) {
/* Initial code replacement */
ftrace_generate_orig_insn(&orig);
ftrace_generate_nop_insn(&new);
} else if (is_kprobe_on_ftrace(&old)) {
/*
* If we find a breakpoint instruction, a kprobe has been
* placed at the beginning of the function. We write the
* constant KPROBE_ON_FTRACE_NOP into the remaining four
* bytes of the original instruction so that the kprobes
* handler can execute a nop, if it reaches this breakpoint.
*/
ftrace_generate_kprobe_call_insn(&orig);
ftrace_generate_kprobe_nop_insn(&new);
} else {
/* Replace ftrace call with a nop. */
ftrace_generate_call_insn(&orig, rec->ip);
ftrace_generate_nop_insn(&new);
}
/* Verify that the to be replaced code matches what we expect. */
if (memcmp(&orig, &old, sizeof(old)))
return -EINVAL;
s390_kernel_write((void *) rec->ip, &new, sizeof(new));
return 0;
}
int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
{
struct ftrace_insn orig, new, old;
if (probe_kernel_read(&old, (void *) rec->ip, sizeof(old)))
return -EFAULT;
if (is_kprobe_on_ftrace(&old)) {
/*
* If we find a breakpoint instruction, a kprobe has been
* placed at the beginning of the function. We write the
* constant KPROBE_ON_FTRACE_CALL into the remaining four
* bytes of the original instruction so that the kprobes
* handler can execute a brasl if it reaches this breakpoint.
*/
ftrace_generate_kprobe_nop_insn(&orig);
ftrace_generate_kprobe_call_insn(&new);
} else {
/* Replace nop with an ftrace call. */
ftrace_generate_nop_insn(&orig);
ftrace_generate_call_insn(&new, rec->ip);
}
/* Verify that the to be replaced code matches what we expect. */
if (memcmp(&orig, &old, sizeof(old)))
return -EINVAL;
s390_kernel_write((void *) rec->ip, &new, sizeof(new));
return 0;
}
int ftrace_update_ftrace_func(ftrace_func_t func)
{
return 0;
}
int __init ftrace_dyn_arch_init(void)
{
return 0;
}
#ifdef CONFIG_MODULES
static int __init ftrace_plt_init(void)
{
unsigned int *ip;
ftrace_plt = (unsigned long) module_alloc(PAGE_SIZE);
if (!ftrace_plt)
panic("cannot allocate ftrace plt\n");
ip = (unsigned int *) ftrace_plt;
ip[0] = 0x0d10e310; /* basr 1,0; lg 1,10(1); br 1 */
ip[1] = 0x100a0004;
ip[2] = 0x07f10000;
ip[3] = FTRACE_ADDR >> 32;
ip[4] = FTRACE_ADDR & 0xffffffff;
set_memory_ro(ftrace_plt, 1);
return 0;
}
device_initcall(ftrace_plt_init);
#endif /* CONFIG_MODULES */
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
/*
* Hook the return address and push it in the stack of return addresses
* in current thread info.
*/
unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip)
{
if (unlikely(ftrace_graph_is_dead()))
goto out;
if (unlikely(atomic_read(&current->tracing_graph_pause)))
goto out;
ip -= MCOUNT_INSN_SIZE;
if (!function_graph_enter(parent, ip, 0, NULL))
parent = (unsigned long) return_to_handler;
out:
return parent;
}
NOKPROBE_SYMBOL(prepare_ftrace_return);
/*
* Patch the kernel code at ftrace_graph_caller location. The instruction
* there is branch relative on condition. To enable the ftrace graph code
* block, we simply patch the mask field of the instruction to zero and
* turn the instruction into a nop.
* To disable the ftrace graph code the mask field will be patched to
* all ones, which turns the instruction into an unconditional branch.
*/
int ftrace_enable_ftrace_graph_caller(void)
{
u8 op = 0x04; /* set mask field to zero */
s390_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op));
return 0;
}
int ftrace_disable_ftrace_graph_caller(void)
{
u8 op = 0xf4; /* set mask field to all ones */
s390_kernel_write(__va(ftrace_graph_caller)+1, &op, sizeof(op));
return 0;
}
#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
+128
View File
@@ -0,0 +1,128 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2016
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <linux/kernel.h>
#include <linux/syscalls.h>
#include <linux/signal.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <asm/guarded_storage.h>
#include "entry.h"
void guarded_storage_release(struct task_struct *tsk)
{
kfree(tsk->thread.gs_cb);
kfree(tsk->thread.gs_bc_cb);
}
static int gs_enable(void)
{
struct gs_cb *gs_cb;
if (!current->thread.gs_cb) {
gs_cb = kzalloc(sizeof(*gs_cb), GFP_KERNEL);
if (!gs_cb)
return -ENOMEM;
gs_cb->gsd = 25;
preempt_disable();
__ctl_set_bit(2, 4);
load_gs_cb(gs_cb);
current->thread.gs_cb = gs_cb;
preempt_enable();
}
return 0;
}
static int gs_disable(void)
{
if (current->thread.gs_cb) {
preempt_disable();
kfree(current->thread.gs_cb);
current->thread.gs_cb = NULL;
__ctl_clear_bit(2, 4);
preempt_enable();
}
return 0;
}
static int gs_set_bc_cb(struct gs_cb __user *u_gs_cb)
{
struct gs_cb *gs_cb;
gs_cb = current->thread.gs_bc_cb;
if (!gs_cb) {
gs_cb = kzalloc(sizeof(*gs_cb), GFP_KERNEL);
if (!gs_cb)
return -ENOMEM;
current->thread.gs_bc_cb = gs_cb;
}
if (copy_from_user(gs_cb, u_gs_cb, sizeof(*gs_cb)))
return -EFAULT;
return 0;
}
static int gs_clear_bc_cb(void)
{
struct gs_cb *gs_cb;
gs_cb = current->thread.gs_bc_cb;
current->thread.gs_bc_cb = NULL;
kfree(gs_cb);
return 0;
}
void gs_load_bc_cb(struct pt_regs *regs)
{
struct gs_cb *gs_cb;
preempt_disable();
clear_thread_flag(TIF_GUARDED_STORAGE);
gs_cb = current->thread.gs_bc_cb;
if (gs_cb) {
kfree(current->thread.gs_cb);
current->thread.gs_bc_cb = NULL;
__ctl_set_bit(2, 4);
load_gs_cb(gs_cb);
current->thread.gs_cb = gs_cb;
}
preempt_enable();
}
static int gs_broadcast(void)
{
struct task_struct *sibling;
read_lock(&tasklist_lock);
for_each_thread(current, sibling) {
if (!sibling->thread.gs_bc_cb)
continue;
if (test_and_set_tsk_thread_flag(sibling, TIF_GUARDED_STORAGE))
kick_process(sibling);
}
read_unlock(&tasklist_lock);
return 0;
}
SYSCALL_DEFINE2(s390_guarded_storage, int, command,
struct gs_cb __user *, gs_cb)
{
if (!MACHINE_HAS_GS)
return -EOPNOTSUPP;
switch (command) {
case GS_ENABLE:
return gs_enable();
case GS_DISABLE:
return gs_disable();
case GS_SET_BC_CB:
return gs_set_bc_cb(gs_cb);
case GS_CLEAR_BC_CB:
return gs_clear_bc_cb();
case GS_BROADCAST:
return gs_broadcast();
default:
return -EINVAL;
}
}
+101
View File
@@ -0,0 +1,101 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright IBM Corp. 1999, 2010
*
* Author(s): Hartmut Penner <hp@de.ibm.com>
* Martin Schwidefsky <schwidefsky@de.ibm.com>
* Rob van der Heij <rvdhei@iae.nl>
* Heiko Carstens <heiko.carstens@de.ibm.com>
*
*/
#include <linux/init.h>
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/thread_info.h>
#include <asm/page.h>
__HEAD
ENTRY(startup_continue)
tm __LC_STFLE_FAC_LIST+5,0x80 # LPP available ?
jz 0f
xc __LC_LPP+1(7,0),__LC_LPP+1 # clear lpp and current_pid
mvi __LC_LPP,0x80 # and set LPP_MAGIC
.insn s,0xb2800000,__LC_LPP # load program parameter
0: larl %r1,tod_clock_base
mvc 0(16,%r1),__LC_BOOT_CLOCK
larl %r13,.LPG1 # get base
lctlg %c0,%c15,.Lctl-.LPG1(%r13) # load control registers
lg %r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area
# move IPL device to lowcore
larl %r0,boot_vdso_data
stg %r0,__LC_VDSO_PER_CPU
#
# Setup stack
#
larl %r14,init_task
stg %r14,__LC_CURRENT
larl %r15,init_thread_union
aghi %r15,1<<(PAGE_SHIFT+THREAD_SIZE_ORDER) # init_task_union + THREAD_SIZE
stg %r15,__LC_KERNEL_STACK # set end of kernel stack
aghi %r15,-160
#
# Early setup functions that may not rely on an initialized bss section,
# like moving the initrd. Returns with an initialized bss section.
#
brasl %r14,startup_init_nobss
#
# Early machine initialization and detection functions.
#
brasl %r14,startup_init
# check control registers
stctg %c0,%c15,0(%r15)
oi 6(%r15),0x60 # enable sigp emergency & external call
oi 4(%r15),0x10 # switch on low address proctection
lctlg %c0,%c15,0(%r15)
lam 0,15,.Laregs-.LPG1(%r13) # load acrs needed by uaccess
brasl %r14,start_kernel # go to C code
#
# We returned from start_kernel ?!? PANIK
#
basr %r13,0
lpswe .Ldw-.(%r13) # load disabled wait psw
.align 16
.LPG1:
.Lctl: .quad 0x04040000 # cr0: AFP registers & secondary space
.quad 0 # cr1: primary space segment table
.quad .Lduct # cr2: dispatchable unit control table
.quad 0 # cr3: instruction authorization
.quad 0xffff # cr4: instruction authorization
.quad .Lduct # cr5: primary-aste origin
.quad 0 # cr6: I/O interrupts
.quad 0 # cr7: secondary space segment table
.quad 0 # cr8: access registers translation
.quad 0 # cr9: tracing off
.quad 0 # cr10: tracing off
.quad 0 # cr11: tracing off
.quad 0 # cr12: tracing off
.quad 0 # cr13: home space segment table
.quad 0xc0000000 # cr14: machine check handling off
.quad .Llinkage_stack # cr15: linkage stack operations
.Lpcmsk:.quad 0x0000000180000000
.L4malign:.quad 0xffffffffffc00000
.Lscan2g:.quad 0x80000000 + 0x20000 - 8 # 2GB + 128K - 8
.Lnop: .long 0x07000700
.Lparmaddr:
.quad PARMAREA
.align 64
.Lduct: .long 0,.Laste,.Laste,0,.Lduald,0,0,0
.long 0,0,0,0,0,0,0,0
.Laste: .quad 0,0xffffffffffffffff,0,0,0,0,0,0
.align 128
.Lduald:.rept 8
.long 0x80000000,0,0,0 # invalid access-list entries
.endr
.Llinkage_stack:
.long 0,0,0x89000000,0,0,0,0x8a000000,0
.Ldw: .quad 0x0002000180000000,0x0000000000000000
.Laregs:.long 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+142
View File
@@ -0,0 +1,142 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Idle functions for s390.
*
* Copyright IBM Corp. 2014
*
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <linux/kernel.h>
#include <linux/kernel_stat.h>
#include <linux/kprobes.h>
#include <linux/notifier.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/sched/cputime.h>
#include <asm/nmi.h>
#include <asm/smp.h>
#include "entry.h"
static DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
void enabled_wait(void)
{
struct s390_idle_data *idle = this_cpu_ptr(&s390_idle);
unsigned long long idle_time;
unsigned long psw_mask;
trace_hardirqs_on();
/* Wait for external, I/O or machine check interrupt. */
psw_mask = PSW_KERNEL_BITS | PSW_MASK_WAIT | PSW_MASK_DAT |
PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
clear_cpu_flag(CIF_NOHZ_DELAY);
/* Call the assembler magic in entry.S */
psw_idle(idle, psw_mask);
trace_hardirqs_off();
/* Account time spent with enabled wait psw loaded as idle time. */
write_seqcount_begin(&idle->seqcount);
idle_time = idle->clock_idle_exit - idle->clock_idle_enter;
idle->clock_idle_enter = idle->clock_idle_exit = 0ULL;
idle->idle_time += idle_time;
idle->idle_count++;
account_idle_time(cputime_to_nsecs(idle_time));
write_seqcount_end(&idle->seqcount);
}
NOKPROBE_SYMBOL(enabled_wait);
static ssize_t show_idle_count(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
unsigned long long idle_count;
unsigned int seq;
do {
seq = read_seqcount_begin(&idle->seqcount);
idle_count = READ_ONCE(idle->idle_count);
if (READ_ONCE(idle->clock_idle_enter))
idle_count++;
} while (read_seqcount_retry(&idle->seqcount, seq));
return sprintf(buf, "%llu\n", idle_count);
}
DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
static ssize_t show_idle_time(struct device *dev,
struct device_attribute *attr, char *buf)
{
unsigned long long now, idle_time, idle_enter, idle_exit, in_idle;
struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
unsigned int seq;
do {
seq = read_seqcount_begin(&idle->seqcount);
idle_time = READ_ONCE(idle->idle_time);
idle_enter = READ_ONCE(idle->clock_idle_enter);
idle_exit = READ_ONCE(idle->clock_idle_exit);
} while (read_seqcount_retry(&idle->seqcount, seq));
in_idle = 0;
now = get_tod_clock();
if (idle_enter) {
if (idle_exit) {
in_idle = idle_exit - idle_enter;
} else if (now > idle_enter) {
in_idle = now - idle_enter;
}
}
idle_time += in_idle;
return sprintf(buf, "%llu\n", idle_time >> 12);
}
DEVICE_ATTR(idle_time_us, 0444, show_idle_time, NULL);
u64 arch_cpu_idle_time(int cpu)
{
struct s390_idle_data *idle = &per_cpu(s390_idle, cpu);
unsigned long long now, idle_enter, idle_exit, in_idle;
unsigned int seq;
do {
seq = read_seqcount_begin(&idle->seqcount);
idle_enter = READ_ONCE(idle->clock_idle_enter);
idle_exit = READ_ONCE(idle->clock_idle_exit);
} while (read_seqcount_retry(&idle->seqcount, seq));
in_idle = 0;
now = get_tod_clock();
if (idle_enter) {
if (idle_exit) {
in_idle = idle_exit - idle_enter;
} else if (now > idle_enter) {
in_idle = now - idle_enter;
}
}
return cputime_to_nsecs(in_idle);
}
void arch_cpu_idle_enter(void)
{
local_mcck_disable();
}
void arch_cpu_idle(void)
{
if (!test_cpu_flag(CIF_MCCK_PENDING))
/* Halt the cpu and keep track of cpu time accounting. */
enabled_wait();
local_irq_enable();
}
void arch_cpu_idle_exit(void)
{
local_mcck_enable();
if (test_cpu_flag(CIF_MCCK_PENDING))
s390_handle_mcck();
}
void arch_cpu_idle_dead(void)
{
cpu_die();
}
File diff suppressed because it is too large. Load diff
+311
View File
@@ -0,0 +1,311 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2004, 2011
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Holger Smolinski <Holger.Smolinski@de.ibm.com>,
* Thomas Spatzier <tspat@de.ibm.com>,
*
* This file contains interrupt related functions.
*/
#include <linux/kernel_stat.h>
#include <linux/interrupt.h>
#include <linux/seq_file.h>
#include <linux/proc_fs.h>
#include <linux/profile.h>
#include <linux/export.h>
#include <linux/kernel.h>
#include <linux/ftrace.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/irq.h>
#include <asm/irq_regs.h>
#include <asm/cputime.h>
#include <asm/lowcore.h>
#include <asm/irq.h>
#include <asm/hw_irq.h>
#include "entry.h"
DEFINE_PER_CPU_SHARED_ALIGNED(struct irq_stat, irq_stat);
EXPORT_PER_CPU_SYMBOL_GPL(irq_stat);
struct irq_class {
int irq;
char *name;
char *desc;
};
/*
* The list of "main" irq classes on s390. This is the list of interrupts
* that appear both in /proc/stat ("intr" line) and /proc/interrupts.
* Historically only external and I/O interrupts have been part of /proc/stat.
* We can't add the split external and I/O sub classes since the first field
* in the "intr" line in /proc/stat is supposed to be the sum of all other
* fields.
* Since the external and I/O interrupt fields are already sums we would end
* up with having a sum which accounts each interrupt twice.
*/
static const struct irq_class irqclass_main_desc[NR_IRQS_BASE] = {
{.irq = EXT_INTERRUPT, .name = "EXT"},
{.irq = IO_INTERRUPT, .name = "I/O"},
{.irq = THIN_INTERRUPT, .name = "AIO"},
};
/*
* The list of split external and I/O interrupts that appear only in
* /proc/interrupts.
* In addition this list contains non external / I/O events like NMIs.
*/
static const struct irq_class irqclass_sub_desc[] = {
{.irq = IRQEXT_CLK, .name = "CLK", .desc = "[EXT] Clock Comparator"},
{.irq = IRQEXT_EXC, .name = "EXC", .desc = "[EXT] External Call"},
{.irq = IRQEXT_EMS, .name = "EMS", .desc = "[EXT] Emergency Signal"},
{.irq = IRQEXT_TMR, .name = "TMR", .desc = "[EXT] CPU Timer"},
{.irq = IRQEXT_TLA, .name = "TAL", .desc = "[EXT] Timing Alert"},
{.irq = IRQEXT_PFL, .name = "PFL", .desc = "[EXT] Pseudo Page Fault"},
{.irq = IRQEXT_DSD, .name = "DSD", .desc = "[EXT] DASD Diag"},
{.irq = IRQEXT_VRT, .name = "VRT", .desc = "[EXT] Virtio"},
{.irq = IRQEXT_SCP, .name = "SCP", .desc = "[EXT] Service Call"},
{.irq = IRQEXT_IUC, .name = "IUC", .desc = "[EXT] IUCV"},
{.irq = IRQEXT_CMS, .name = "CMS", .desc = "[EXT] CPU-Measurement: Sampling"},
{.irq = IRQEXT_CMC, .name = "CMC", .desc = "[EXT] CPU-Measurement: Counter"},
{.irq = IRQEXT_FTP, .name = "FTP", .desc = "[EXT] HMC FTP Service"},
{.irq = IRQIO_CIO, .name = "CIO", .desc = "[I/O] Common I/O Layer Interrupt"},
{.irq = IRQIO_QAI, .name = "QAI", .desc = "[I/O] QDIO Adapter Interrupt"},
{.irq = IRQIO_DAS, .name = "DAS", .desc = "[I/O] DASD"},
{.irq = IRQIO_C15, .name = "C15", .desc = "[I/O] 3215"},
{.irq = IRQIO_C70, .name = "C70", .desc = "[I/O] 3270"},
{.irq = IRQIO_TAP, .name = "TAP", .desc = "[I/O] Tape"},
{.irq = IRQIO_VMR, .name = "VMR", .desc = "[I/O] Unit Record Devices"},
{.irq = IRQIO_LCS, .name = "LCS", .desc = "[I/O] LCS"},
{.irq = IRQIO_CTC, .name = "CTC", .desc = "[I/O] CTC"},
{.irq = IRQIO_APB, .name = "APB", .desc = "[I/O] AP Bus"},
{.irq = IRQIO_ADM, .name = "ADM", .desc = "[I/O] EADM Subchannel"},
{.irq = IRQIO_CSC, .name = "CSC", .desc = "[I/O] CHSC Subchannel"},
{.irq = IRQIO_PCI, .name = "PCI", .desc = "[I/O] PCI Interrupt" },
{.irq = IRQIO_MSI, .name = "MSI", .desc = "[I/O] MSI Interrupt" },
{.irq = IRQIO_VIR, .name = "VIR", .desc = "[I/O] Virtual I/O Devices"},
{.irq = IRQIO_VAI, .name = "VAI", .desc = "[I/O] Virtual I/O Devices AI"},
{.irq = NMI_NMI, .name = "NMI", .desc = "[NMI] Machine Check"},
{.irq = CPU_RST, .name = "RST", .desc = "[CPU] CPU Restart"},
};
void __init init_IRQ(void)
{
BUILD_BUG_ON(ARRAY_SIZE(irqclass_sub_desc) != NR_ARCH_IRQS);
init_cio_interrupts();
init_airq_interrupts();
init_ext_interrupts();
}
void do_IRQ(struct pt_regs *regs, int irq)
{
struct pt_regs *old_regs;
old_regs = set_irq_regs(regs);
irq_enter();
if (tod_after_eq(S390_lowcore.int_clock,
S390_lowcore.clock_comparator))
/* Serve timer interrupts first. */
clock_comparator_work();
generic_handle_irq(irq);
irq_exit();
set_irq_regs(old_regs);
}
/*
* show_interrupts is needed by /proc/interrupts.
*/
int show_interrupts(struct seq_file *p, void *v)
{
int index = *(loff_t *) v;
int cpu, irq;
get_online_cpus();
if (index == 0) {
seq_puts(p, " ");
for_each_online_cpu(cpu)
seq_printf(p, "CPU%d ", cpu);
seq_putc(p, '\n');
}
if (index < NR_IRQS_BASE) {
seq_printf(p, "%s: ", irqclass_main_desc[index].name);
irq = irqclass_main_desc[index].irq;
for_each_online_cpu(cpu)
seq_printf(p, "%10u ", kstat_irqs_cpu(irq, cpu));
seq_putc(p, '\n');
goto out;
}
if (index > NR_IRQS_BASE)
goto out;
for (index = 0; index < NR_ARCH_IRQS; index++) {
seq_printf(p, "%s: ", irqclass_sub_desc[index].name);
irq = irqclass_sub_desc[index].irq;
for_each_online_cpu(cpu)
seq_printf(p, "%10u ",
per_cpu(irq_stat, cpu).irqs[irq]);
if (irqclass_sub_desc[index].desc)
seq_printf(p, " %s", irqclass_sub_desc[index].desc);
seq_putc(p, '\n');
}
out:
put_online_cpus();
return 0;
}
unsigned int arch_dynirq_lower_bound(unsigned int from)
{
return from < NR_IRQS_BASE ? NR_IRQS_BASE : from;
}
/*
* Switch to the asynchronous interrupt stack for softirq execution.
*/
void do_softirq_own_stack(void)
{
unsigned long old, new;
old = current_stack_pointer();
/* Check against async. stack address range. */
new = S390_lowcore.async_stack;
if (((new - old) >> (PAGE_SHIFT + THREAD_SIZE_ORDER)) != 0) {
/* Need to switch to the async. stack. */
new -= STACK_FRAME_OVERHEAD;
((struct stack_frame *) new)->back_chain = old;
asm volatile(" la 15,0(%0)\n"
" brasl 14,__do_softirq\n"
" la 15,0(%1)\n"
: : "a" (new), "a" (old)
: "0", "1", "2", "3", "4", "5", "14",
"cc", "memory" );
} else {
/* We are already on the async stack. */
__do_softirq();
}
}
/*
* ext_int_hash[index] is the list head for all external interrupts that hash
* to this index.
*/
static struct hlist_head ext_int_hash[32] ____cacheline_aligned;
struct ext_int_info {
ext_int_handler_t handler;
struct hlist_node entry;
struct rcu_head rcu;
u16 code;
};
/* ext_int_hash_lock protects the handler lists for external interrupts */
static DEFINE_SPINLOCK(ext_int_hash_lock);
static inline int ext_hash(u16 code)
{
BUILD_BUG_ON(!is_power_of_2(ARRAY_SIZE(ext_int_hash)));
return (code + (code >> 9)) & (ARRAY_SIZE(ext_int_hash) - 1);
}
int register_external_irq(u16 code, ext_int_handler_t handler)
{
struct ext_int_info *p;
unsigned long flags;
int index;
p = kmalloc(sizeof(*p), GFP_ATOMIC);
if (!p)
return -ENOMEM;
p->code = code;
p->handler = handler;
index = ext_hash(code);
spin_lock_irqsave(&ext_int_hash_lock, flags);
hlist_add_head_rcu(&p->entry, &ext_int_hash[index]);
spin_unlock_irqrestore(&ext_int_hash_lock, flags);
return 0;
}
EXPORT_SYMBOL(register_external_irq);
int unregister_external_irq(u16 code, ext_int_handler_t handler)
{
struct ext_int_info *p;
unsigned long flags;
int index = ext_hash(code);
spin_lock_irqsave(&ext_int_hash_lock, flags);
hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) {
if (p->code == code && p->handler == handler) {
hlist_del_rcu(&p->entry);
kfree_rcu(p, rcu);
}
}
spin_unlock_irqrestore(&ext_int_hash_lock, flags);
return 0;
}
EXPORT_SYMBOL(unregister_external_irq);
static irqreturn_t do_ext_interrupt(int irq, void *dummy)
{
struct pt_regs *regs = get_irq_regs();
struct ext_code ext_code;
struct ext_int_info *p;
int index;
ext_code = *(struct ext_code *) &regs->int_code;
if (ext_code.code != EXT_IRQ_CLK_COMP)
set_cpu_flag(CIF_NOHZ_DELAY);
index = ext_hash(ext_code.code);
rcu_read_lock();
hlist_for_each_entry_rcu(p, &ext_int_hash[index], entry) {
if (unlikely(p->code != ext_code.code))
continue;
p->handler(ext_code, regs->int_parm, regs->int_parm_long);
}
rcu_read_unlock();
return IRQ_HANDLED;
}
static struct irqaction external_interrupt = {
.name = "EXT",
.handler = do_ext_interrupt,
};
void __init init_ext_interrupts(void)
{
int idx;
for (idx = 0; idx < ARRAY_SIZE(ext_int_hash); idx++)
INIT_HLIST_HEAD(&ext_int_hash[idx]);
irq_set_chip_and_handler(EXT_INTERRUPT,
&dummy_irq_chip, handle_percpu_irq);
setup_irq(EXT_INTERRUPT, &external_interrupt);
}
static DEFINE_SPINLOCK(irq_subclass_lock);
static unsigned char irq_subclass_refcount[64];
void irq_subclass_register(enum irq_subclass subclass)
{
spin_lock(&irq_subclass_lock);
if (!irq_subclass_refcount[subclass])
ctl_set_bit(0, subclass);
irq_subclass_refcount[subclass]++;
spin_unlock(&irq_subclass_lock);
}
EXPORT_SYMBOL(irq_subclass_register);
void irq_subclass_unregister(enum irq_subclass subclass)
{
spin_lock(&irq_subclass_lock);
irq_subclass_refcount[subclass]--;
if (!irq_subclass_refcount[subclass])
ctl_clear_bit(0, subclass);
spin_unlock(&irq_subclass_lock);
}
EXPORT_SYMBOL(irq_subclass_unregister);
+102
View File
@@ -0,0 +1,102 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Jump label s390 support
*
* Copyright IBM Corp. 2011
* Author(s): Jan Glauber <jang@linux.vnet.ibm.com>
*/
#include <linux/uaccess.h>
#include <linux/stop_machine.h>
#include <linux/jump_label.h>
#include <asm/ipl.h>
struct insn {
u16 opcode;
s32 offset;
} __packed;
struct insn_args {
struct jump_entry *entry;
enum jump_label_type type;
};
static void jump_label_make_nop(struct jump_entry *entry, struct insn *insn)
{
/* brcl 0,0 */
insn->opcode = 0xc004;
insn->offset = 0;
}
static void jump_label_make_branch(struct jump_entry *entry, struct insn *insn)
{
/* brcl 15,offset */
insn->opcode = 0xc0f4;
insn->offset = (entry->target - entry->code) >> 1;
}
static void jump_label_bug(struct jump_entry *entry, struct insn *expected,
struct insn *new)
{
unsigned char *ipc = (unsigned char *)entry->code;
unsigned char *ipe = (unsigned char *)expected;
unsigned char *ipn = (unsigned char *)new;
pr_emerg("Jump label code mismatch at %pS [%p]\n", ipc, ipc);
pr_emerg("Found: %6ph\n", ipc);
pr_emerg("Expected: %6ph\n", ipe);
pr_emerg("New: %6ph\n", ipn);
panic("Corrupted kernel text");
}
static struct insn orignop = {
.opcode = 0xc004,
.offset = JUMP_LABEL_NOP_OFFSET >> 1,
};
static void __jump_label_transform(struct jump_entry *entry,
enum jump_label_type type,
int init)
{
struct insn old, new;
if (type == JUMP_LABEL_JMP) {
jump_label_make_nop(entry, &old);
jump_label_make_branch(entry, &new);
} else {
jump_label_make_branch(entry, &old);
jump_label_make_nop(entry, &new);
}
if (init) {
if (memcmp((void *)entry->code, &orignop, sizeof(orignop)))
jump_label_bug(entry, &orignop, &new);
} else {
if (memcmp((void *)entry->code, &old, sizeof(old)))
jump_label_bug(entry, &old, &new);
}
s390_kernel_write((void *)entry->code, &new, sizeof(new));
}
static int __sm_arch_jump_label_transform(void *data)
{
struct insn_args *args = data;
__jump_label_transform(args->entry, args->type, 0);
return 0;
}
void arch_jump_label_transform(struct jump_entry *entry,
enum jump_label_type type)
{
struct insn_args args;
args.entry = entry;
args.type = type;
stop_machine_cpuslocked(__sm_arch_jump_label_transform, &args, NULL);
}
void arch_jump_label_transform_static(struct jump_entry *entry,
enum jump_label_type type)
{
__jump_label_transform(entry, type, 1);
}
+16
View File
@@ -0,0 +1,16 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/debugfs.h>
#include <linux/export.h>
#include <linux/init.h>
struct dentry *arch_debugfs_dir;
EXPORT_SYMBOL(arch_debugfs_dir);
static int __init arch_kdebugfs_init(void)
{
arch_debugfs_dir = debugfs_create_dir("s390", NULL);
if (IS_ERR(arch_debugfs_dir))
arch_debugfs_dir = NULL;
return 0;
}
postcore_initcall(arch_kdebugfs_init);
+152
View File
@@ -0,0 +1,152 @@
// SPDX-License-Identifier: GPL-2.0
/*
* ELF loader for kexec_file_load system call.
*
* Copyright IBM Corp. 2018
*
* Author(s): Philipp Rudo <prudo@linux.vnet.ibm.com>
*/
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/kexec.h>
#include <asm/setup.h>
static int kexec_file_add_elf_kernel(struct kimage *image,
struct s390_load_data *data,
char *kernel, unsigned long kernel_len)
{
struct kexec_buf buf;
const Elf_Ehdr *ehdr;
const Elf_Phdr *phdr;
Elf_Addr entry;
int i, ret;
ehdr = (Elf_Ehdr *)kernel;
buf.image = image;
if (image->type == KEXEC_TYPE_CRASH)
entry = STARTUP_KDUMP_OFFSET;
else
entry = ehdr->e_entry;
phdr = (void *)ehdr + ehdr->e_phoff;
for (i = 0; i < ehdr->e_phnum; i++, phdr++) {
if (phdr->p_type != PT_LOAD)
continue;
buf.buffer = kernel + phdr->p_offset;
buf.bufsz = phdr->p_filesz;
buf.mem = ALIGN(phdr->p_paddr, phdr->p_align);
buf.memsz = phdr->p_memsz;
if (entry - phdr->p_paddr < phdr->p_memsz) {
data->kernel_buf = buf.buffer;
data->memsz += STARTUP_NORMAL_OFFSET;
buf.buffer += STARTUP_NORMAL_OFFSET;
buf.bufsz -= STARTUP_NORMAL_OFFSET;
buf.mem += STARTUP_NORMAL_OFFSET;
buf.memsz -= STARTUP_NORMAL_OFFSET;
}
if (image->type == KEXEC_TYPE_CRASH)
buf.mem += crashk_res.start;
ret = kexec_add_buffer(&buf);
if (ret)
return ret;
data->memsz = ALIGN(data->memsz, phdr->p_align) + buf.memsz;
}
return 0;
}
static void *s390_elf_load(struct kimage *image,
char *kernel, unsigned long kernel_len,
char *initrd, unsigned long initrd_len,
char *cmdline, unsigned long cmdline_len)
{
struct s390_load_data data = {0};
const Elf_Ehdr *ehdr;
const Elf_Phdr *phdr;
size_t size;
int i, ret;
/* image->fobs->probe already checked for valid ELF magic number. */
ehdr = (Elf_Ehdr *)kernel;
if (ehdr->e_type != ET_EXEC ||
ehdr->e_ident[EI_CLASS] != ELFCLASS64 ||
!elf_check_arch(ehdr))
return ERR_PTR(-EINVAL);
if (!ehdr->e_phnum || ehdr->e_phentsize != sizeof(Elf_Phdr))
return ERR_PTR(-EINVAL);
size = ehdr->e_ehsize + ehdr->e_phoff;
size += ehdr->e_phentsize * ehdr->e_phnum;
if (size > kernel_len)
return ERR_PTR(-EINVAL);
phdr = (void *)ehdr + ehdr->e_phoff;
size = ALIGN(size, phdr->p_align);
for (i = 0; i < ehdr->e_phnum; i++, phdr++) {
if (phdr->p_type == PT_INTERP)
return ERR_PTR(-EINVAL);
if (phdr->p_offset > kernel_len)
return ERR_PTR(-EINVAL);
size += ALIGN(phdr->p_filesz, phdr->p_align);
}
if (size > kernel_len)
return ERR_PTR(-EINVAL);
ret = kexec_file_add_elf_kernel(image, &data, kernel, kernel_len);
if (ret)
return ERR_PTR(ret);
if (!data.memsz)
return ERR_PTR(-EINVAL);
if (initrd) {
ret = kexec_file_add_initrd(image, &data, initrd, initrd_len);
if (ret)
return ERR_PTR(ret);
}
ret = kexec_file_add_purgatory(image, &data);
if (ret)
return ERR_PTR(ret);
return kexec_file_update_kernel(image, &data);
}
static int s390_elf_probe(const char *buf, unsigned long len)
{
const Elf_Ehdr *ehdr;
if (len < sizeof(Elf_Ehdr))
return -ENOEXEC;
ehdr = (Elf_Ehdr *)buf;
/* Only check the ELF magic number here and do proper validity check
* in the loader. Any check here that fails would send the erroneous
* ELF file to the image loader that does not care what it gets.
* (Most likely) causing behavior not intended by the user.
*/
if (memcmp(ehdr->e_ident, ELFMAG, SELFMAG) != 0)
return -ENOEXEC;
return 0;
}
const struct kexec_file_ops s390_kexec_elf_ops = {
.probe = s390_elf_probe,
.load = s390_elf_load,
};
+76
View File
@@ -0,0 +1,76 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Image loader for kexec_file_load system call.
*
* Copyright IBM Corp. 2018
*
* Author(s): Philipp Rudo <prudo@linux.vnet.ibm.com>
*/
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/kexec.h>
#include <asm/setup.h>
static int kexec_file_add_image_kernel(struct kimage *image,
struct s390_load_data *data,
char *kernel, unsigned long kernel_len)
{
struct kexec_buf buf;
int ret;
buf.image = image;
buf.buffer = kernel + STARTUP_NORMAL_OFFSET;
buf.bufsz = kernel_len - STARTUP_NORMAL_OFFSET;
buf.mem = STARTUP_NORMAL_OFFSET;
if (image->type == KEXEC_TYPE_CRASH)
buf.mem += crashk_res.start;
buf.memsz = buf.bufsz;
ret = kexec_add_buffer(&buf);
data->kernel_buf = kernel;
data->memsz += buf.memsz + STARTUP_NORMAL_OFFSET;
return ret;
}
static void *s390_image_load(struct kimage *image,
char *kernel, unsigned long kernel_len,
char *initrd, unsigned long initrd_len,
char *cmdline, unsigned long cmdline_len)
{
struct s390_load_data data = {0};
int ret;
ret = kexec_file_add_image_kernel(image, &data, kernel, kernel_len);
if (ret)
return ERR_PTR(ret);
if (initrd) {
ret = kexec_file_add_initrd(image, &data, initrd, initrd_len);
if (ret)
return ERR_PTR(ret);
}
ret = kexec_file_add_purgatory(image, &data);
if (ret)
return ERR_PTR(ret);
return kexec_file_update_kernel(image, &data);
}
static int s390_image_probe(const char *buf, unsigned long len)
{
/* Can't reliably tell if an image is valid. Therefore give the
* user whatever he wants.
*/
return 0;
}
const struct kexec_file_ops s390_kexec_image_ops = {
.probe = s390_image_probe,
.load = s390_image_load,
};
+658
View File
@@ -0,0 +1,658 @@
// SPDX-License-Identifier: GPL-2.0+
/*
* Kernel Probes (KProbes)
*
* Copyright IBM Corp. 2002, 2006
*
* s390 port, used ppc64 as template. Mike Grundy <grundym@us.ibm.com>
*/
#include <linux/kprobes.h>
#include <linux/ptrace.h>
#include <linux/preempt.h>
#include <linux/stop_machine.h>
#include <linux/kdebug.h>
#include <linux/uaccess.h>
#include <linux/extable.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/hardirq.h>
#include <linux/ftrace.h>
#include <asm/set_memory.h>
#include <asm/sections.h>
#include <asm/dis.h>
DEFINE_PER_CPU(struct kprobe *, current_kprobe);
DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
struct kretprobe_blackpoint kretprobe_blacklist[] = { };
DEFINE_INSN_CACHE_OPS(dmainsn);
static void *alloc_dmainsn_page(void)
{
void *page;
page = (void *) __get_free_page(GFP_KERNEL | GFP_DMA);
if (page)
set_memory_x((unsigned long) page, 1);
return page;
}
static void free_dmainsn_page(void *page)
{
set_memory_nx((unsigned long) page, 1);
free_page((unsigned long)page);
}
struct kprobe_insn_cache kprobe_dmainsn_slots = {
.mutex = __MUTEX_INITIALIZER(kprobe_dmainsn_slots.mutex),
.alloc = alloc_dmainsn_page,
.free = free_dmainsn_page,
.pages = LIST_HEAD_INIT(kprobe_dmainsn_slots.pages),
.insn_size = MAX_INSN_SIZE,
};
static void copy_instruction(struct kprobe *p)
{
unsigned long ip = (unsigned long) p->addr;
s64 disp, new_disp;
u64 addr, new_addr;
if (ftrace_location(ip) == ip) {
/*
* If kprobes patches the instruction that is morphed by
* ftrace make sure that kprobes always sees the branch
* "jg .+24" that skips the mcount block or the "brcl 0,0"
* in case of hotpatch.
*/
ftrace_generate_nop_insn((struct ftrace_insn *)p->ainsn.insn);
p->ainsn.is_ftrace_insn = 1;
} else
memcpy(p->ainsn.insn, p->addr, insn_length(*p->addr >> 8));
p->opcode = p->ainsn.insn[0];
if (!probe_is_insn_relative_long(p->ainsn.insn))
return;
/*
* For pc-relative instructions in RIL-b or RIL-c format patch the
* RI2 displacement field. We have already made sure that the insn
* slot for the patched instruction is within the same 2GB area
* as the original instruction (either kernel image or module area).
* Therefore the new displacement will always fit.
*/
disp = *(s32 *)&p->ainsn.insn[1];
addr = (u64)(unsigned long)p->addr;
new_addr = (u64)(unsigned long)p->ainsn.insn;
new_disp = ((addr + (disp * 2)) - new_addr) / 2;
*(s32 *)&p->ainsn.insn[1] = new_disp;
}
NOKPROBE_SYMBOL(copy_instruction);
static inline int is_kernel_addr(void *addr)
{
return addr < (void *)_end;
}
static int s390_get_insn_slot(struct kprobe *p)
{
/*
* Get an insn slot that is within the same 2GB area like the original
* instruction. That way instructions with a 32bit signed displacement
* field can be patched and executed within the insn slot.
*/
p->ainsn.insn = NULL;
if (is_kernel_addr(p->addr))
p->ainsn.insn = get_dmainsn_slot();
else if (is_module_addr(p->addr))
p->ainsn.insn = get_insn_slot();
return p->ainsn.insn ? 0 : -ENOMEM;
}
NOKPROBE_SYMBOL(s390_get_insn_slot);
static void s390_free_insn_slot(struct kprobe *p)
{
if (!p->ainsn.insn)
return;
if (is_kernel_addr(p->addr))
free_dmainsn_slot(p->ainsn.insn, 0);
else
free_insn_slot(p->ainsn.insn, 0);
p->ainsn.insn = NULL;
}
NOKPROBE_SYMBOL(s390_free_insn_slot);
int arch_prepare_kprobe(struct kprobe *p)
{
if ((unsigned long) p->addr & 0x01)
return -EINVAL;
/* Make sure the probe isn't going on a difficult instruction */
if (probe_is_prohibited_opcode(p->addr))
return -EINVAL;
if (s390_get_insn_slot(p))
return -ENOMEM;
copy_instruction(p);
return 0;
}
NOKPROBE_SYMBOL(arch_prepare_kprobe);
int arch_check_ftrace_location(struct kprobe *p)
{
return 0;
}
struct swap_insn_args {
struct kprobe *p;
unsigned int arm_kprobe : 1;
};
static int swap_instruction(void *data)
{
struct swap_insn_args *args = data;
struct ftrace_insn new_insn, *insn;
struct kprobe *p = args->p;
size_t len;
new_insn.opc = args->arm_kprobe ? BREAKPOINT_INSTRUCTION : p->opcode;
len = sizeof(new_insn.opc);
if (!p->ainsn.is_ftrace_insn)
goto skip_ftrace;
len = sizeof(new_insn);
insn = (struct ftrace_insn *) p->addr;
if (args->arm_kprobe) {
if (is_ftrace_nop(insn))
new_insn.disp = KPROBE_ON_FTRACE_NOP;
else
new_insn.disp = KPROBE_ON_FTRACE_CALL;
} else {
ftrace_generate_call_insn(&new_insn, (unsigned long)p->addr);
if (insn->disp == KPROBE_ON_FTRACE_NOP)
ftrace_generate_nop_insn(&new_insn);
}
skip_ftrace:
s390_kernel_write(p->addr, &new_insn, len);
return 0;
}
NOKPROBE_SYMBOL(swap_instruction);
void arch_arm_kprobe(struct kprobe *p)
{
struct swap_insn_args args = {.p = p, .arm_kprobe = 1};
stop_machine_cpuslocked(swap_instruction, &args, NULL);
}
NOKPROBE_SYMBOL(arch_arm_kprobe);
void arch_disarm_kprobe(struct kprobe *p)
{
struct swap_insn_args args = {.p = p, .arm_kprobe = 0};
stop_machine_cpuslocked(swap_instruction, &args, NULL);
}
NOKPROBE_SYMBOL(arch_disarm_kprobe);
void arch_remove_kprobe(struct kprobe *p)
{
s390_free_insn_slot(p);
}
NOKPROBE_SYMBOL(arch_remove_kprobe);
static void enable_singlestep(struct kprobe_ctlblk *kcb,
struct pt_regs *regs,
unsigned long ip)
{
struct per_regs per_kprobe;
/* Set up the PER control registers %cr9-%cr11 */
per_kprobe.control = PER_EVENT_IFETCH;
per_kprobe.start = ip;
per_kprobe.end = ip;
/* Save control regs and psw mask */
__ctl_store(kcb->kprobe_saved_ctl, 9, 11);
kcb->kprobe_saved_imask = regs->psw.mask &
(PSW_MASK_PER | PSW_MASK_IO | PSW_MASK_EXT);
/* Set PER control regs, turns on single step for the given address */
__ctl_load(per_kprobe, 9, 11);
regs->psw.mask |= PSW_MASK_PER;
regs->psw.mask &= ~(PSW_MASK_IO | PSW_MASK_EXT);
regs->psw.addr = ip;
}
NOKPROBE_SYMBOL(enable_singlestep);
static void disable_singlestep(struct kprobe_ctlblk *kcb,
struct pt_regs *regs,
unsigned long ip)
{
/* Restore control regs and psw mask, set new psw address */
__ctl_load(kcb->kprobe_saved_ctl, 9, 11);
regs->psw.mask &= ~PSW_MASK_PER;
regs->psw.mask |= kcb->kprobe_saved_imask;
regs->psw.addr = ip;
}
NOKPROBE_SYMBOL(disable_singlestep);
/*
* Activate a kprobe by storing its pointer to current_kprobe. The
* previous kprobe is stored in kcb->prev_kprobe. A stack of up to
* two kprobes can be active, see KPROBE_REENTER.
*/
static void push_kprobe(struct kprobe_ctlblk *kcb, struct kprobe *p)
{
kcb->prev_kprobe.kp = __this_cpu_read(current_kprobe);
kcb->prev_kprobe.status = kcb->kprobe_status;
__this_cpu_write(current_kprobe, p);
}
NOKPROBE_SYMBOL(push_kprobe);
/*
* Deactivate a kprobe by backing up to the previous state. If the
* current state is KPROBE_REENTER prev_kprobe.kp will be non-NULL,
* for any other state prev_kprobe.kp will be NULL.
*/
static void pop_kprobe(struct kprobe_ctlblk *kcb)
{
__this_cpu_write(current_kprobe, kcb->prev_kprobe.kp);
kcb->kprobe_status = kcb->prev_kprobe.status;
}
NOKPROBE_SYMBOL(pop_kprobe);
void arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs)
{
ri->ret_addr = (kprobe_opcode_t *) regs->gprs[14];
/* Replace the return addr with trampoline addr */
regs->gprs[14] = (unsigned long) &kretprobe_trampoline;
}
NOKPROBE_SYMBOL(arch_prepare_kretprobe);
static void kprobe_reenter_check(struct kprobe_ctlblk *kcb, struct kprobe *p)
{
switch (kcb->kprobe_status) {
case KPROBE_HIT_SSDONE:
case KPROBE_HIT_ACTIVE:
kprobes_inc_nmissed_count(p);
break;
case KPROBE_HIT_SS:
case KPROBE_REENTER:
default:
/*
* A kprobe on the code path to single step an instruction
* is a BUG. The code path resides in the .kprobes.text
* section and is executed with interrupts disabled.
*/
pr_err("Invalid kprobe detected.\n");
dump_kprobe(p);
BUG();
}
}
NOKPROBE_SYMBOL(kprobe_reenter_check);
static int kprobe_handler(struct pt_regs *regs)
{
struct kprobe_ctlblk *kcb;
struct kprobe *p;
/*
* We want to disable preemption for the entire duration of kprobe
* processing. That includes the calls to the pre/post handlers
* and single stepping the kprobe instruction.
*/
preempt_disable();
kcb = get_kprobe_ctlblk();
p = get_kprobe((void *)(regs->psw.addr - 2));
if (p) {
if (kprobe_running()) {
/*
* We have hit a kprobe while another is still
* active. This can happen in the pre and post
* handler. Single step the instruction of the
* new probe but do not call any handler function
* of this secondary kprobe.
* push_kprobe and pop_kprobe saves and restores
* the currently active kprobe.
*/
kprobe_reenter_check(kcb, p);
push_kprobe(kcb, p);
kcb->kprobe_status = KPROBE_REENTER;
} else {
/*
* If we have no pre-handler or it returned 0, we
* continue with single stepping. If we have a
* pre-handler and it returned non-zero, it prepped
* for changing execution path, so get out doing
* nothing more here.
*/
push_kprobe(kcb, p);
kcb->kprobe_status = KPROBE_HIT_ACTIVE;
if (p->pre_handler && p->pre_handler(p, regs)) {
pop_kprobe(kcb);
preempt_enable_no_resched();
return 1;
}
kcb->kprobe_status = KPROBE_HIT_SS;
}
enable_singlestep(kcb, regs, (unsigned long) p->ainsn.insn);
return 1;
} /* else:
* No kprobe at this address and no active kprobe. The trap has
* not been caused by a kprobe breakpoint. The race of breakpoint
* vs. kprobe remove does not exist because on s390 as we use
* stop_machine to arm/disarm the breakpoints.
*/
preempt_enable_no_resched();
return 0;
}
NOKPROBE_SYMBOL(kprobe_handler);
/*
* Function return probe trampoline:
* - init_kprobes() establishes a probepoint here
* - When the probed function returns, this probe
* causes the handlers to fire
*/
static void __used kretprobe_trampoline_holder(void)
{
asm volatile(".global kretprobe_trampoline\n"
"kretprobe_trampoline: bcr 0,0\n");
}
/*
* Called when the probe at kretprobe trampoline is hit
*/
static int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
{
struct kretprobe_instance *ri;
struct hlist_head *head, empty_rp;
struct hlist_node *tmp;
unsigned long flags, orig_ret_address;
unsigned long trampoline_address;
kprobe_opcode_t *correct_ret_addr;
INIT_HLIST_HEAD(&empty_rp);
kretprobe_hash_lock(current, &head, &flags);
/*
* It is possible to have multiple instances associated with a given
* task either because an multiple functions in the call path
* have a return probe installed on them, and/or more than one return
* return probe was registered for a target function.
*
* We can handle this because:
* - instances are always inserted at the head of the list
* - when multiple return probes are registered for the same
* function, the first instance's ret_addr will point to the
* real return address, and all the rest will point to
* kretprobe_trampoline
*/
ri = NULL;
orig_ret_address = 0;
correct_ret_addr = NULL;
trampoline_address = (unsigned long) &kretprobe_trampoline;
hlist_for_each_entry_safe(ri, tmp, head, hlist) {
if (ri->task != current)
/* another task is sharing our hash bucket */
continue;
orig_ret_address = (unsigned long) ri->ret_addr;
if (orig_ret_address != trampoline_address)
/*
* This is the real return address. Any other
* instances associated with this task are for
* other calls deeper on the call stack
*/
break;
}
kretprobe_assert(ri, orig_ret_address, trampoline_address);
correct_ret_addr = ri->ret_addr;
hlist_for_each_entry_safe(ri, tmp, head, hlist) {
if (ri->task != current)
/* another task is sharing our hash bucket */
continue;
orig_ret_address = (unsigned long) ri->ret_addr;
if (ri->rp && ri->rp->handler) {
ri->ret_addr = correct_ret_addr;
ri->rp->handler(ri, regs);
}
recycle_rp_inst(ri, &empty_rp);
if (orig_ret_address != trampoline_address)
/*
* This is the real return address. Any other
* instances associated with this task are for
* other calls deeper on the call stack
*/
break;
}
regs->psw.addr = orig_ret_address;
kretprobe_hash_unlock(current, &flags);
hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) {
hlist_del(&ri->hlist);
kfree(ri);
}
/*
* By returning a non-zero value, we are telling
* kprobe_handler() that we don't want the post_handler
* to run (and have re-enabled preemption)
*/
return 1;
}
NOKPROBE_SYMBOL(trampoline_probe_handler);
/*
* Called after single-stepping. p->addr is the address of the
* instruction whose first byte has been replaced by the "breakpoint"
* instruction. To avoid the SMP problems that can occur when we
* temporarily put back the original opcode to single-step, we
* single-stepped a copy of the instruction. The address of this
* copy is p->ainsn.insn.
*/
static void resume_execution(struct kprobe *p, struct pt_regs *regs)
{
struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
unsigned long ip = regs->psw.addr;
int fixup = probe_get_fixup_type(p->ainsn.insn);
/* Check if the kprobes location is an enabled ftrace caller */
if (p->ainsn.is_ftrace_insn) {
struct ftrace_insn *insn = (struct ftrace_insn *) p->addr;
struct ftrace_insn call_insn;
ftrace_generate_call_insn(&call_insn, (unsigned long) p->addr);
/*
* A kprobe on an enabled ftrace call site actually single
* stepped an unconditional branch (ftrace nop equivalent).
* Now we need to fixup things and pretend that a brasl r0,...
* was executed instead.
*/
if (insn->disp == KPROBE_ON_FTRACE_CALL) {
ip += call_insn.disp * 2 - MCOUNT_INSN_SIZE;
regs->gprs[0] = (unsigned long)p->addr + sizeof(*insn);
}
}
if (fixup & FIXUP_PSW_NORMAL)
ip += (unsigned long) p->addr - (unsigned long) p->ainsn.insn;
if (fixup & FIXUP_BRANCH_NOT_TAKEN) {
int ilen = insn_length(p->ainsn.insn[0] >> 8);
if (ip - (unsigned long) p->ainsn.insn == ilen)
ip = (unsigned long) p->addr + ilen;
}
if (fixup & FIXUP_RETURN_REGISTER) {
int reg = (p->ainsn.insn[0] & 0xf0) >> 4;
regs->gprs[reg] += (unsigned long) p->addr -
(unsigned long) p->ainsn.insn;
}
disable_singlestep(kcb, regs, ip);
}
NOKPROBE_SYMBOL(resume_execution);
static int post_kprobe_handler(struct pt_regs *regs)
{
struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
struct kprobe *p = kprobe_running();
if (!p)
return 0;
if (kcb->kprobe_status != KPROBE_REENTER && p->post_handler) {
kcb->kprobe_status = KPROBE_HIT_SSDONE;
p->post_handler(p, regs, 0);
}
resume_execution(p, regs);
pop_kprobe(kcb);
preempt_enable_no_resched();
/*
* if somebody else is singlestepping across a probe point, psw mask
* will have PER set, in which case, continue the remaining processing
* of do_single_step, as if this is not a probe hit.
*/
if (regs->psw.mask & PSW_MASK_PER)
return 0;
return 1;
}
NOKPROBE_SYMBOL(post_kprobe_handler);
static int kprobe_trap_handler(struct pt_regs *regs, int trapnr)
{
struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
struct kprobe *p = kprobe_running();
const struct exception_table_entry *entry;
switch(kcb->kprobe_status) {
case KPROBE_HIT_SS:
case KPROBE_REENTER:
/*
* We are here because the instruction being single
* stepped caused a page fault. We reset the current
* kprobe and the nip points back to the probe address
* and allow the page fault handler to continue as a
* normal page fault.
*/
disable_singlestep(kcb, regs, (unsigned long) p->addr);
pop_kprobe(kcb);
preempt_enable_no_resched();
break;
case KPROBE_HIT_ACTIVE:
case KPROBE_HIT_SSDONE:
/*
* We increment the nmissed count for accounting,
* we can also use npre/npostfault count for accounting
* these specific fault cases.
*/
kprobes_inc_nmissed_count(p);
/*
* We come here because instructions in the pre/post
* handler caused the page_fault, this could happen
* if handler tries to access user space by
* copy_from_user(), get_user() etc. Let the
* user-specified handler try to fix it first.
*/
if (p->fault_handler && p->fault_handler(p, regs, trapnr))
return 1;
/*
* In case the user-specified fault handler returned
* zero, try to fix up.
*/
entry = search_exception_tables(regs->psw.addr);
if (entry) {
regs->psw.addr = extable_fixup(entry);
return 1;
}
/*
* fixup_exception() could not handle it,
* Let do_page_fault() fix it.
*/
break;
default:
break;
}
return 0;
}
NOKPROBE_SYMBOL(kprobe_trap_handler);
int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
{
int ret;
if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT))
local_irq_disable();
ret = kprobe_trap_handler(regs, trapnr);
if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT))
local_irq_restore(regs->psw.mask & ~PSW_MASK_PER);
return ret;
}
NOKPROBE_SYMBOL(kprobe_fault_handler);
/*
* Wrapper routine to for handling exceptions.
*/
int kprobe_exceptions_notify(struct notifier_block *self,
unsigned long val, void *data)
{
struct die_args *args = (struct die_args *) data;
struct pt_regs *regs = args->regs;
int ret = NOTIFY_DONE;
if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT))
local_irq_disable();
switch (val) {
case DIE_BPT:
if (kprobe_handler(regs))
ret = NOTIFY_STOP;
break;
case DIE_SSTEP:
if (post_kprobe_handler(regs))
ret = NOTIFY_STOP;
break;
case DIE_TRAP:
if (!preemptible() && kprobe_running() &&
kprobe_trap_handler(regs, args->trapnr))
ret = NOTIFY_STOP;
break;
default:
break;
}
if (regs->psw.mask & (PSW_MASK_IO | PSW_MASK_EXT))
local_irq_restore(regs->psw.mask & ~PSW_MASK_PER);
return ret;
}
NOKPROBE_SYMBOL(kprobe_exceptions_notify);
static struct kprobe trampoline = {
.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
.pre_handler = trampoline_probe_handler
};
int __init arch_init_kprobes(void)
{
return register_kprobe(&trampoline);
}
int arch_trampoline_kprobe(struct kprobe *p)
{
return p->addr == (kprobe_opcode_t *) &kretprobe_trampoline;
}
NOKPROBE_SYMBOL(arch_trampoline_kprobe);
+188
View File
@@ -0,0 +1,188 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Linux Guest Relocation (LGR) detection
*
* Copyright IBM Corp. 2012
* Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com>
*/
#include <linux/init.h>
#include <linux/export.h>
#include <linux/timer.h>
#include <linux/slab.h>
#include <asm/facility.h>
#include <asm/sysinfo.h>
#include <asm/ebcdic.h>
#include <asm/debug.h>
#include <asm/ipl.h>
#define LGR_TIMER_INTERVAL_SECS (30 * 60)
#define VM_LEVEL_MAX 2 /* Maximum is 8, but we only record two levels */
/*
* LGR info: Contains stfle and stsi data
*/
struct lgr_info {
/* Bit field with facility information: 4 DWORDs are stored */
u64 stfle_fac_list[4];
/* Level of system (1 = CEC, 2 = LPAR, 3 = z/VM */
u32 level;
/* Level 1: CEC info (stsi 1.1.1) */
char manufacturer[16];
char type[4];
char sequence[16];
char plant[4];
char model[16];
/* Level 2: LPAR info (stsi 2.2.2) */
u16 lpar_number;
char name[8];
/* Level 3: VM info (stsi 3.2.2) */
u8 vm_count;
struct {
char name[8];
char cpi[16];
} vm[VM_LEVEL_MAX];
} __packed __aligned(8);
/*
* LGR globals
*/
static char lgr_page[PAGE_SIZE] __aligned(PAGE_SIZE);
static struct lgr_info lgr_info_last;
static struct lgr_info lgr_info_cur;
static struct debug_info *lgr_dbf;
/*
* Copy buffer and then convert it to ASCII
*/
static void cpascii(char *dst, char *src, int size)
{
memcpy(dst, src, size);
EBCASC(dst, size);
}
/*
* Fill LGR info with 1.1.1 stsi data
*/
static void lgr_stsi_1_1_1(struct lgr_info *lgr_info)
{
struct sysinfo_1_1_1 *si = (void *) lgr_page;
if (stsi(si, 1, 1, 1))
return;
cpascii(lgr_info->manufacturer, si->manufacturer,
sizeof(si->manufacturer));
cpascii(lgr_info->type, si->type, sizeof(si->type));
cpascii(lgr_info->model, si->model, sizeof(si->model));
cpascii(lgr_info->sequence, si->sequence, sizeof(si->sequence));
cpascii(lgr_info->plant, si->plant, sizeof(si->plant));
}
/*
* Fill LGR info with 2.2.2 stsi data
*/
static void lgr_stsi_2_2_2(struct lgr_info *lgr_info)
{
struct sysinfo_2_2_2 *si = (void *) lgr_page;
if (stsi(si, 2, 2, 2))
return;
cpascii(lgr_info->name, si->name, sizeof(si->name));
memcpy(&lgr_info->lpar_number, &si->lpar_number,
sizeof(lgr_info->lpar_number));
}
/*
* Fill LGR info with 3.2.2 stsi data
*/
static void lgr_stsi_3_2_2(struct lgr_info *lgr_info)
{
struct sysinfo_3_2_2 *si = (void *) lgr_page;
int i;
if (stsi(si, 3, 2, 2))
return;
for (i = 0; i < min_t(u8, si->count, VM_LEVEL_MAX); i++) {
cpascii(lgr_info->vm[i].name, si->vm[i].name,
sizeof(si->vm[i].name));
cpascii(lgr_info->vm[i].cpi, si->vm[i].cpi,
sizeof(si->vm[i].cpi));
}
lgr_info->vm_count = si->count;
}
/*
* Fill LGR info with current data
*/
static void lgr_info_get(struct lgr_info *lgr_info)
{
int level;
memset(lgr_info, 0, sizeof(*lgr_info));
stfle(lgr_info->stfle_fac_list, ARRAY_SIZE(lgr_info->stfle_fac_list));
level = stsi(NULL, 0, 0, 0);
lgr_info->level = level;
if (level >= 1)
lgr_stsi_1_1_1(lgr_info);
if (level >= 2)
lgr_stsi_2_2_2(lgr_info);
if (level >= 3)
lgr_stsi_3_2_2(lgr_info);
}
/*
* Check if LGR info has changed and if yes log new LGR info to s390dbf
*/
void lgr_info_log(void)
{
static DEFINE_SPINLOCK(lgr_info_lock);
unsigned long flags;
if (!spin_trylock_irqsave(&lgr_info_lock, flags))
return;
lgr_info_get(&lgr_info_cur);
if (memcmp(&lgr_info_last, &lgr_info_cur, sizeof(lgr_info_cur)) != 0) {
debug_event(lgr_dbf, 1, &lgr_info_cur, sizeof(lgr_info_cur));
lgr_info_last = lgr_info_cur;
}
spin_unlock_irqrestore(&lgr_info_lock, flags);
}
EXPORT_SYMBOL_GPL(lgr_info_log);
static void lgr_timer_set(void);
/*
* LGR timer callback
*/
static void lgr_timer_fn(struct timer_list *unused)
{
lgr_info_log();
lgr_timer_set();
}
static struct timer_list lgr_timer;
/*
* Setup next LGR timer
*/
static void lgr_timer_set(void)
{
mod_timer(&lgr_timer, jiffies + LGR_TIMER_INTERVAL_SECS * HZ);
}
/*
* Initialize LGR: Add s390dbf, write initial lgr_info and setup timer
*/
static int __init lgr_init(void)
{
lgr_dbf = debug_register("lgr", 1, 1, sizeof(struct lgr_info));
if (!lgr_dbf)
return -ENOMEM;
debug_register_view(lgr_dbf, &debug_hex_ascii_view);
lgr_info_get(&lgr_info_last);
debug_event(lgr_dbf, 1, &lgr_info_last, sizeof(lgr_info_last));
timer_setup(&lgr_timer, lgr_timer_fn, TIMER_DEFERRABLE);
lgr_timer_set();
return 0;
}
device_initcall(lgr_init);
+304
View File
@@ -0,0 +1,304 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2005, 2011
*
* Author(s): Rolf Adelsberger,
* Heiko Carstens <heiko.carstens@de.ibm.com>
* Michael Holzheu <holzheu@linux.vnet.ibm.com>
*/
#include <linux/device.h>
#include <linux/mm.h>
#include <linux/kexec.h>
#include <linux/delay.h>
#include <linux/reboot.h>
#include <linux/ftrace.h>
#include <linux/debug_locks.h>
#include <linux/suspend.h>
#include <asm/cio.h>
#include <asm/setup.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/smp.h>
#include <asm/ipl.h>
#include <asm/diag.h>
#include <asm/elf.h>
#include <asm/asm-offsets.h>
#include <asm/cacheflush.h>
#include <asm/os_info.h>
#include <asm/set_memory.h>
#include <asm/switch_to.h>
#include <asm/nmi.h>
typedef void (*relocate_kernel_t)(kimage_entry_t *, unsigned long);
extern const unsigned char relocate_kernel[];
extern const unsigned long long relocate_kernel_len;
#ifdef CONFIG_CRASH_DUMP
/*
* PM notifier callback for kdump
*/
static int machine_kdump_pm_cb(struct notifier_block *nb, unsigned long action,
void *ptr)
{
switch (action) {
case PM_SUSPEND_PREPARE:
case PM_HIBERNATION_PREPARE:
if (kexec_crash_image)
arch_kexec_unprotect_crashkres();
break;
case PM_POST_SUSPEND:
case PM_POST_HIBERNATION:
if (kexec_crash_image)
arch_kexec_protect_crashkres();
break;
default:
return NOTIFY_DONE;
}
return NOTIFY_OK;
}
static int __init machine_kdump_pm_init(void)
{
pm_notifier(machine_kdump_pm_cb, 0);
return 0;
}
arch_initcall(machine_kdump_pm_init);
/*
* Reset the system, copy boot CPU registers to absolute zero,
* and jump to the kdump image
*/
static void __do_machine_kdump(void *image)
{
int (*start_kdump)(int);
unsigned long prefix;
/* store_status() saved the prefix register to lowcore */
prefix = (unsigned long) S390_lowcore.prefixreg_save_area;
/* Now do the reset */
s390_reset_system();
/*
* Copy dump CPU store status info to absolute zero.
* This need to be done *after* s390_reset_system set the
* prefix register of this CPU to zero
*/
memcpy((void *) __LC_FPREGS_SAVE_AREA,
(void *)(prefix + __LC_FPREGS_SAVE_AREA), 512);
__load_psw_mask(PSW_MASK_BASE | PSW_DEFAULT_KEY | PSW_MASK_EA | PSW_MASK_BA);
start_kdump = (void *)((struct kimage *) image)->start;
start_kdump(1);
/* Die if start_kdump returns */
disabled_wait((unsigned long) __builtin_return_address(0));
}
/*
* Start kdump: create a LGR log entry, store status of all CPUs and
* branch to __do_machine_kdump.
*/
static noinline void __machine_kdump(void *image)
{
struct mcesa *mcesa;
union ctlreg2 cr2_old, cr2_new;
int this_cpu, cpu;
lgr_info_log();
/* Get status of the other CPUs */
this_cpu = smp_find_processor_id(stap());
for_each_online_cpu(cpu) {
if (cpu == this_cpu)
continue;
if (smp_store_status(cpu))
continue;
}
/* Store status of the boot CPU */
mcesa = (struct mcesa *)(S390_lowcore.mcesad & MCESA_ORIGIN_MASK);
if (MACHINE_HAS_VX)
save_vx_regs((__vector128 *) mcesa->vector_save_area);
if (MACHINE_HAS_GS) {
__ctl_store(cr2_old.val, 2, 2);
cr2_new = cr2_old;
cr2_new.gse = 1;
__ctl_load(cr2_new.val, 2, 2);
save_gs_cb((struct gs_cb *) mcesa->guarded_storage_save_area);
__ctl_load(cr2_old.val, 2, 2);
}
/*
* To create a good backchain for this CPU in the dump store_status
* is passed the address of a function. The address is saved into
* the PSW save area of the boot CPU and the function is invoked as
* a tail call of store_status. The backchain in the dump will look
* like this:
* restart_int_handler -> __machine_kexec -> __do_machine_kdump
* The call to store_status() will not return.
*/
store_status(__do_machine_kdump, image);
}
#endif
/*
* Check if kdump checksums are valid: We call purgatory with parameter "0"
*/
static bool kdump_csum_valid(struct kimage *image)
{
#ifdef CONFIG_CRASH_DUMP
int (*start_kdump)(int) = (void *)image->start;
int rc;
__arch_local_irq_stnsm(0xfb); /* disable DAT */
rc = start_kdump(0);
__arch_local_irq_stosm(0x04); /* enable DAT */
return rc == 0;
#else
return false;
#endif
}
#ifdef CONFIG_CRASH_DUMP
void crash_free_reserved_phys_range(unsigned long begin, unsigned long end)
{
unsigned long addr, size;
for (addr = begin; addr < end; addr += PAGE_SIZE)
free_reserved_page(pfn_to_page(addr >> PAGE_SHIFT));
size = begin - crashk_res.start;
if (size)
os_info_crashkernel_add(crashk_res.start, size);
else
os_info_crashkernel_add(0, 0);
}
static void crash_protect_pages(int protect)
{
unsigned long size;
if (!crashk_res.end)
return;
size = resource_size(&crashk_res);
if (protect)
set_memory_ro(crashk_res.start, size >> PAGE_SHIFT);
else
set_memory_rw(crashk_res.start, size >> PAGE_SHIFT);
}
void arch_kexec_protect_crashkres(void)
{
crash_protect_pages(1);
}
void arch_kexec_unprotect_crashkres(void)
{
crash_protect_pages(0);
}
#endif
/*
* Give back memory to hypervisor before new kdump is loaded
*/
static int machine_kexec_prepare_kdump(void)
{
#ifdef CONFIG_CRASH_DUMP
if (MACHINE_IS_VM)
diag10_range(PFN_DOWN(crashk_res.start),
PFN_DOWN(crashk_res.end - crashk_res.start + 1));
return 0;
#else
return -EINVAL;
#endif
}
int machine_kexec_prepare(struct kimage *image)
{
void *reboot_code_buffer;
if (image->type == KEXEC_TYPE_CRASH)
return machine_kexec_prepare_kdump();
/* We don't support anything but the default image type for now. */
if (image->type != KEXEC_TYPE_DEFAULT)
return -EINVAL;
/* Get the destination where the assembler code should be copied to.*/
reboot_code_buffer = (void *) page_to_phys(image->control_code_page);
/* Then copy it */
memcpy(reboot_code_buffer, relocate_kernel, relocate_kernel_len);
return 0;
}
void machine_kexec_cleanup(struct kimage *image)
{
}
void arch_crash_save_vmcoreinfo(void)
{
VMCOREINFO_SYMBOL(lowcore_ptr);
VMCOREINFO_SYMBOL(high_memory);
VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS);
mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note());
}
void machine_shutdown(void)
{
}
void machine_crash_shutdown(struct pt_regs *regs)
{
set_os_info_reipl_block();
}
/*
* Do normal kexec
*/
static void __do_machine_kexec(void *data)
{
relocate_kernel_t data_mover;
struct kimage *image = data;
s390_reset_system();
data_mover = (relocate_kernel_t) page_to_phys(image->control_code_page);
__arch_local_irq_stnsm(0xfb); /* disable DAT - avoid no-execute */
/* Call the moving routine */
(*data_mover)(&image->head, image->start);
/* Die if kexec returns */
disabled_wait((unsigned long) __builtin_return_address(0));
}
/*
* Reset system and call either kdump or normal kexec
*/
static void __machine_kexec(void *data)
{
__arch_local_irq_stosm(0x04); /* enable DAT */
pfault_fini();
tracing_off();
debug_locks_off();
#ifdef CONFIG_CRASH_DUMP
if (((struct kimage *) data)->type == KEXEC_TYPE_CRASH)
__machine_kdump(data);
#endif
__do_machine_kexec(data);
}
/*
* Do either kdump or normal kexec. In case of kdump we first ask
* purgatory, if kdump checksums are valid.
*/
void machine_kexec(struct kimage *image)
{
if (image->type == KEXEC_TYPE_CRASH && !kdump_csum_valid(image))
return;
tracer_disable();
smp_send_stop();
smp_call_ipl_cpu(__machine_kexec, image);
}
+245
View File
@@ -0,0 +1,245 @@
// SPDX-License-Identifier: GPL-2.0
/*
* s390 code for kexec_file_load system call
*
* Copyright IBM Corp. 2018
*
* Author(s): Philipp Rudo <prudo@linux.vnet.ibm.com>
*/
#include <linux/elf.h>
#include <linux/kexec.h>
#include <asm/setup.h>
const struct kexec_file_ops * const kexec_file_loaders[] = {
&s390_kexec_elf_ops,
&s390_kexec_image_ops,
NULL,
};
int *kexec_file_update_kernel(struct kimage *image,
struct s390_load_data *data)
{
unsigned long *loc;
if (image->cmdline_buf_len >= ARCH_COMMAND_LINE_SIZE)
return ERR_PTR(-EINVAL);
if (image->cmdline_buf_len)
memcpy(data->kernel_buf + COMMAND_LINE_OFFSET,
image->cmdline_buf, image->cmdline_buf_len);
if (image->type == KEXEC_TYPE_CRASH) {
loc = (unsigned long *)(data->kernel_buf + OLDMEM_BASE_OFFSET);
*loc = crashk_res.start;
loc = (unsigned long *)(data->kernel_buf + OLDMEM_SIZE_OFFSET);
*loc = crashk_res.end - crashk_res.start + 1;
}
if (image->initrd_buf) {
loc = (unsigned long *)(data->kernel_buf + INITRD_START_OFFSET);
*loc = data->initrd_load_addr;
loc = (unsigned long *)(data->kernel_buf + INITRD_SIZE_OFFSET);
*loc = image->initrd_buf_len;
}
return NULL;
}
static int kexec_file_update_purgatory(struct kimage *image)
{
u64 entry, type;
int ret;
if (image->type == KEXEC_TYPE_CRASH) {
entry = STARTUP_KDUMP_OFFSET;
type = KEXEC_TYPE_CRASH;
} else {
entry = STARTUP_NORMAL_OFFSET;
type = KEXEC_TYPE_DEFAULT;
}
ret = kexec_purgatory_get_set_symbol(image, "kernel_entry", &entry,
sizeof(entry), false);
if (ret)
return ret;
ret = kexec_purgatory_get_set_symbol(image, "kernel_type", &type,
sizeof(type), false);
if (ret)
return ret;
if (image->type == KEXEC_TYPE_CRASH) {
u64 crash_size;
ret = kexec_purgatory_get_set_symbol(image, "crash_start",
&crashk_res.start,
sizeof(crashk_res.start),
false);
if (ret)
return ret;
crash_size = crashk_res.end - crashk_res.start + 1;
ret = kexec_purgatory_get_set_symbol(image, "crash_size",
&crash_size,
sizeof(crash_size),
false);
}
return ret;
}
int kexec_file_add_purgatory(struct kimage *image, struct s390_load_data *data)
{
struct kexec_buf buf;
int ret;
buf.image = image;
data->memsz = ALIGN(data->memsz, PAGE_SIZE);
buf.mem = data->memsz;
if (image->type == KEXEC_TYPE_CRASH)
buf.mem += crashk_res.start;
ret = kexec_load_purgatory(image, &buf);
if (ret)
return ret;
ret = kexec_file_update_purgatory(image);
return ret;
}
int kexec_file_add_initrd(struct kimage *image, struct s390_load_data *data,
char *initrd, unsigned long initrd_len)
{
struct kexec_buf buf;
int ret;
buf.image = image;
buf.buffer = initrd;
buf.bufsz = initrd_len;
data->memsz = ALIGN(data->memsz, PAGE_SIZE);
buf.mem = data->memsz;
if (image->type == KEXEC_TYPE_CRASH)
buf.mem += crashk_res.start;
buf.memsz = buf.bufsz;
data->initrd_load_addr = buf.mem;
data->memsz += buf.memsz;
ret = kexec_add_buffer(&buf);
return ret;
}
/*
* The kernel is loaded to a fixed location. Turn off kexec_locate_mem_hole
* and provide kbuf->mem by hand.
*/
int arch_kexec_walk_mem(struct kexec_buf *kbuf,
int (*func)(struct resource *, void *))
{
return 1;
}
int arch_kexec_apply_relocations_add(struct purgatory_info *pi,
Elf_Shdr *section,
const Elf_Shdr *relsec,
const Elf_Shdr *symtab)
{
Elf_Rela *relas;
int i;
relas = (void *)pi->ehdr + relsec->sh_offset;
for (i = 0; i < relsec->sh_size / sizeof(*relas); i++) {
const Elf_Sym *sym; /* symbol to relocate */
unsigned long addr; /* final location after relocation */
unsigned long val; /* relocated symbol value */
void *loc; /* tmp location to modify */
sym = (void *)pi->ehdr + symtab->sh_offset;
sym += ELF64_R_SYM(relas[i].r_info);
if (sym->st_shndx == SHN_UNDEF)
return -ENOEXEC;
if (sym->st_shndx == SHN_COMMON)
return -ENOEXEC;
if (sym->st_shndx >= pi->ehdr->e_shnum &&
sym->st_shndx != SHN_ABS)
return -ENOEXEC;
loc = pi->purgatory_buf;
loc += section->sh_offset;
loc += relas[i].r_offset;
val = sym->st_value;
if (sym->st_shndx != SHN_ABS)
val += pi->sechdrs[sym->st_shndx].sh_addr;
val += relas[i].r_addend;
addr = section->sh_addr + relas[i].r_offset;
switch (ELF64_R_TYPE(relas[i].r_info)) {
case R_390_8: /* Direct 8 bit. */
*(u8 *)loc = val;
break;
case R_390_12: /* Direct 12 bit. */
*(u16 *)loc &= 0xf000;
*(u16 *)loc |= val & 0xfff;
break;
case R_390_16: /* Direct 16 bit. */
*(u16 *)loc = val;
break;
case R_390_20: /* Direct 20 bit. */
*(u32 *)loc &= 0xf00000ff;
*(u32 *)loc |= (val & 0xfff) << 16; /* DL */
*(u32 *)loc |= (val & 0xff000) >> 4; /* DH */
break;
case R_390_32: /* Direct 32 bit. */
*(u32 *)loc = val;
break;
case R_390_64: /* Direct 64 bit. */
*(u64 *)loc = val;
break;
case R_390_PC16: /* PC relative 16 bit. */
*(u16 *)loc = (val - addr);
break;
case R_390_PC16DBL: /* PC relative 16 bit shifted by 1. */
*(u16 *)loc = (val - addr) >> 1;
break;
case R_390_PC32DBL: /* PC relative 32 bit shifted by 1. */
*(u32 *)loc = (val - addr) >> 1;
break;
case R_390_PC32: /* PC relative 32 bit. */
*(u32 *)loc = (val - addr);
break;
case R_390_PC64: /* PC relative 64 bit. */
*(u64 *)loc = (val - addr);
break;
default:
break;
}
}
return 0;
}
int arch_kexec_kernel_image_probe(struct kimage *image, void *buf,
unsigned long buf_len)
{
/* A kernel must be at least large enough to contain head.S. During
* load memory in head.S will be accessed, e.g. to register the next
* command line. If the next kernel were smaller the current kernel
* will panic at load.
*
* 0x11000 = sizeof(head.S)
*/
if (buf_len < 0x11000)
return -ENOEXEC;
return kexec_image_probe_default(image, buf, buf_len);
}
+104
View File
@@ -0,0 +1,104 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright IBM Corp. 2008, 2009
*
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
*
*/
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/ftrace.h>
#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/export.h>
GEN_BR_THUNK %r1
GEN_BR_THUNK %r14
.section .kprobes.text, "ax"
ENTRY(ftrace_stub)
BR_EX %r14
#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE)
#define STACK_PTREGS (STACK_FRAME_OVERHEAD)
#define STACK_PTREGS_GPRS (STACK_PTREGS + __PT_GPRS)
#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW)
#ifdef __PACK_STACK
/* allocate just enough for r14, r15 and backchain */
#define TRACED_FUNC_FRAME_SIZE 24
#else
#define TRACED_FUNC_FRAME_SIZE STACK_FRAME_OVERHEAD
#endif
ENTRY(_mcount)
BR_EX %r14
EXPORT_SYMBOL(_mcount)
ENTRY(ftrace_caller)
.globl ftrace_regs_caller
.set ftrace_regs_caller,ftrace_caller
stg %r14,(__SF_GPRS+8*8)(%r15) # save traced function caller
lgr %r1,%r15
#if !(defined(CC_USING_HOTPATCH) || defined(CC_USING_NOP_MCOUNT))
aghi %r0,MCOUNT_RETURN_FIXUP
#endif
# allocate stack frame for ftrace_caller to contain traced function
aghi %r15,-TRACED_FUNC_FRAME_SIZE
stg %r1,__SF_BACKCHAIN(%r15)
stg %r0,(__SF_GPRS+8*8)(%r15)
stg %r15,(__SF_GPRS+9*8)(%r15)
# allocate pt_regs and stack frame for ftrace_trace_function
aghi %r15,-STACK_FRAME_SIZE
stg %r1,(STACK_PTREGS_GPRS+15*8)(%r15)
aghi %r1,-TRACED_FUNC_FRAME_SIZE
stg %r1,__SF_BACKCHAIN(%r15)
stg %r0,(STACK_PTREGS_PSW+8)(%r15)
stmg %r2,%r14,(STACK_PTREGS_GPRS+2*8)(%r15)
#ifdef CONFIG_HAVE_MARCH_Z196_FEATURES
aghik %r2,%r0,-MCOUNT_INSN_SIZE
lgrl %r4,function_trace_op
lgrl %r1,ftrace_trace_function
#else
lgr %r2,%r0
aghi %r2,-MCOUNT_INSN_SIZE
larl %r4,function_trace_op
lg %r4,0(%r4)
larl %r1,ftrace_trace_function
lg %r1,0(%r1)
#endif
lgr %r3,%r14
la %r5,STACK_PTREGS(%r15)
BASR_EX %r14,%r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
# The j instruction gets runtime patched to a nop instruction.
# See ftrace_enable_ftrace_graph_caller.
ENTRY(ftrace_graph_caller)
j ftrace_graph_caller_end
lg %r2,(STACK_PTREGS_GPRS+14*8)(%r15)
lg %r3,(STACK_PTREGS_PSW+8)(%r15)
brasl %r14,prepare_ftrace_return
stg %r2,(STACK_PTREGS_GPRS+14*8)(%r15)
ftrace_graph_caller_end:
.globl ftrace_graph_caller_end
#endif
lg %r1,(STACK_PTREGS_PSW+8)(%r15)
lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
BR_EX %r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
ENTRY(return_to_handler)
stmg %r2,%r5,32(%r15)
lgr %r1,%r15
aghi %r15,-STACK_FRAME_OVERHEAD
stg %r1,__SF_BACKCHAIN(%r15)
brasl %r14,ftrace_return_to_handler
aghi %r15,STACK_FRAME_OVERHEAD
lgr %r14,%r2
lmg %r2,%r5,32(%r15)
BR_EX %r14
#endif
+478
View File
@@ -0,0 +1,478 @@
// SPDX-License-Identifier: GPL-2.0+
/*
* Kernel module help for s390.
*
* S390 version
* Copyright IBM Corp. 2002, 2003
* Author(s): Arnd Bergmann (arndb@de.ibm.com)
* Martin Schwidefsky (schwidefsky@de.ibm.com)
*
* based on i386 version
* Copyright (C) 2001 Rusty Russell.
*/
#include <linux/module.h>
#include <linux/elf.h>
#include <linux/vmalloc.h>
#include <linux/fs.h>
#include <linux/string.h>
#include <linux/kernel.h>
#include <linux/moduleloader.h>
#include <linux/bug.h>
#include <asm/alternative.h>
#include <asm/nospec-branch.h>
#include <asm/facility.h>
#if 0
#define DEBUGP printk
#else
#define DEBUGP(fmt , ...)
#endif
#define PLT_ENTRY_SIZE 20
void *module_alloc(unsigned long size)
{
if (PAGE_ALIGN(size) > MODULES_LEN)
return NULL;
return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END,
GFP_KERNEL, PAGE_KERNEL_EXEC,
0, NUMA_NO_NODE,
__builtin_return_address(0));
}
void module_arch_freeing_init(struct module *mod)
{
if (is_livepatch_module(mod) &&
mod->state == MODULE_STATE_LIVE)
return;
vfree(mod->arch.syminfo);
mod->arch.syminfo = NULL;
}
static void check_rela(Elf_Rela *rela, struct module *me)
{
struct mod_arch_syminfo *info;
info = me->arch.syminfo + ELF_R_SYM (rela->r_info);
switch (ELF_R_TYPE (rela->r_info)) {
case R_390_GOT12: /* 12 bit GOT offset. */
case R_390_GOT16: /* 16 bit GOT offset. */
case R_390_GOT20: /* 20 bit GOT offset. */
case R_390_GOT32: /* 32 bit GOT offset. */
case R_390_GOT64: /* 64 bit GOT offset. */
case R_390_GOTENT: /* 32 bit PC rel. to GOT entry shifted by 1. */
case R_390_GOTPLT12: /* 12 bit offset to jump slot. */
case R_390_GOTPLT16: /* 16 bit offset to jump slot. */
case R_390_GOTPLT20: /* 20 bit offset to jump slot. */
case R_390_GOTPLT32: /* 32 bit offset to jump slot. */
case R_390_GOTPLT64: /* 64 bit offset to jump slot. */
case R_390_GOTPLTENT: /* 32 bit rel. offset to jump slot >> 1. */
if (info->got_offset == -1UL) {
info->got_offset = me->arch.got_size;
me->arch.got_size += sizeof(void*);
}
break;
case R_390_PLT16DBL: /* 16 bit PC rel. PLT shifted by 1. */
case R_390_PLT32DBL: /* 32 bit PC rel. PLT shifted by 1. */
case R_390_PLT32: /* 32 bit PC relative PLT address. */
case R_390_PLT64: /* 64 bit PC relative PLT address. */
case R_390_PLTOFF16: /* 16 bit offset from GOT to PLT. */
case R_390_PLTOFF32: /* 32 bit offset from GOT to PLT. */
case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */
if (info->plt_offset == -1UL) {
info->plt_offset = me->arch.plt_size;
me->arch.plt_size += PLT_ENTRY_SIZE;
}
break;
case R_390_COPY:
case R_390_GLOB_DAT:
case R_390_JMP_SLOT:
case R_390_RELATIVE:
/* Only needed if we want to support loading of
modules linked with -shared. */
break;
}
}
/*
* Account for GOT and PLT relocations. We can't add sections for
* got and plt but we can increase the core module size.
*/
int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
char *secstrings, struct module *me)
{
Elf_Shdr *symtab;
Elf_Sym *symbols;
Elf_Rela *rela;
char *strings;
int nrela, i, j;
/* Find symbol table and string table. */
symtab = NULL;
for (i = 0; i < hdr->e_shnum; i++)
switch (sechdrs[i].sh_type) {
case SHT_SYMTAB:
symtab = sechdrs + i;
break;
}
if (!symtab) {
printk(KERN_ERR "module %s: no symbol table\n", me->name);
return -ENOEXEC;
}
/* Allocate one syminfo structure per symbol. */
me->arch.nsyms = symtab->sh_size / sizeof(Elf_Sym);
me->arch.syminfo = vmalloc(array_size(sizeof(struct mod_arch_syminfo),
me->arch.nsyms));
if (!me->arch.syminfo)
return -ENOMEM;
symbols = (void *) hdr + symtab->sh_offset;
strings = (void *) hdr + sechdrs[symtab->sh_link].sh_offset;
for (i = 0; i < me->arch.nsyms; i++) {
if (symbols[i].st_shndx == SHN_UNDEF &&
strcmp(strings + symbols[i].st_name,
"_GLOBAL_OFFSET_TABLE_") == 0)
/* "Define" it as absolute. */
symbols[i].st_shndx = SHN_ABS;
me->arch.syminfo[i].got_offset = -1UL;
me->arch.syminfo[i].plt_offset = -1UL;
me->arch.syminfo[i].got_initialized = 0;
me->arch.syminfo[i].plt_initialized = 0;
}
/* Search for got/plt relocations. */
me->arch.got_size = me->arch.plt_size = 0;
for (i = 0; i < hdr->e_shnum; i++) {
if (sechdrs[i].sh_type != SHT_RELA)
continue;
nrela = sechdrs[i].sh_size / sizeof(Elf_Rela);
rela = (void *) hdr + sechdrs[i].sh_offset;
for (j = 0; j < nrela; j++)
check_rela(rela + j, me);
}
/* Increase core size by size of got & plt and set start
offsets for got and plt. */
me->core_layout.size = ALIGN(me->core_layout.size, 4);
me->arch.got_offset = me->core_layout.size;
me->core_layout.size += me->arch.got_size;
me->arch.plt_offset = me->core_layout.size;
if (me->arch.plt_size) {
if (IS_ENABLED(CONFIG_EXPOLINE) && !nospec_disable)
me->arch.plt_size += PLT_ENTRY_SIZE;
me->core_layout.size += me->arch.plt_size;
}
return 0;
}
static int apply_rela_bits(Elf_Addr loc, Elf_Addr val,
int sign, int bits, int shift)
{
unsigned long umax;
long min, max;
if (val & ((1UL << shift) - 1))
return -ENOEXEC;
if (sign) {
val = (Elf_Addr)(((long) val) >> shift);
min = -(1L << (bits - 1));
max = (1L << (bits - 1)) - 1;
if ((long) val < min || (long) val > max)
return -ENOEXEC;
} else {
val >>= shift;
umax = ((1UL << (bits - 1)) << 1) - 1;
if ((unsigned long) val > umax)
return -ENOEXEC;
}
if (bits == 8)
*(unsigned char *) loc = val;
else if (bits == 12)
*(unsigned short *) loc = (val & 0xfff) |
(*(unsigned short *) loc & 0xf000);
else if (bits == 16)
*(unsigned short *) loc = val;
else if (bits == 20)
*(unsigned int *) loc = (val & 0xfff) << 16 |
(val & 0xff000) >> 4 |
(*(unsigned int *) loc & 0xf00000ff);
else if (bits == 32)
*(unsigned int *) loc = val;
else if (bits == 64)
*(unsigned long *) loc = val;
return 0;
}
static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
const char *strtab, struct module *me)
{
struct mod_arch_syminfo *info;
Elf_Addr loc, val;
int r_type, r_sym;
int rc = -ENOEXEC;
/* This is where to make the change */
loc = base + rela->r_offset;
/* This is the symbol it is referring to. Note that all
undefined symbols have been resolved. */
r_sym = ELF_R_SYM(rela->r_info);
r_type = ELF_R_TYPE(rela->r_info);
info = me->arch.syminfo + r_sym;
val = symtab[r_sym].st_value;
switch (r_type) {
case R_390_NONE: /* No relocation. */
rc = 0;
break;
case R_390_8: /* Direct 8 bit. */
case R_390_12: /* Direct 12 bit. */
case R_390_16: /* Direct 16 bit. */
case R_390_20: /* Direct 20 bit. */
case R_390_32: /* Direct 32 bit. */
case R_390_64: /* Direct 64 bit. */
val += rela->r_addend;
if (r_type == R_390_8)
rc = apply_rela_bits(loc, val, 0, 8, 0);
else if (r_type == R_390_12)
rc = apply_rela_bits(loc, val, 0, 12, 0);
else if (r_type == R_390_16)
rc = apply_rela_bits(loc, val, 0, 16, 0);
else if (r_type == R_390_20)
rc = apply_rela_bits(loc, val, 1, 20, 0);
else if (r_type == R_390_32)
rc = apply_rela_bits(loc, val, 0, 32, 0);
else if (r_type == R_390_64)
rc = apply_rela_bits(loc, val, 0, 64, 0);
break;
case R_390_PC16: /* PC relative 16 bit. */
case R_390_PC16DBL: /* PC relative 16 bit shifted by 1. */
case R_390_PC32DBL: /* PC relative 32 bit shifted by 1. */
case R_390_PC32: /* PC relative 32 bit. */
case R_390_PC64: /* PC relative 64 bit. */
val += rela->r_addend - loc;
if (r_type == R_390_PC16)
rc = apply_rela_bits(loc, val, 1, 16, 0);
else if (r_type == R_390_PC16DBL)
rc = apply_rela_bits(loc, val, 1, 16, 1);
else if (r_type == R_390_PC32DBL)
rc = apply_rela_bits(loc, val, 1, 32, 1);
else if (r_type == R_390_PC32)
rc = apply_rela_bits(loc, val, 1, 32, 0);
else if (r_type == R_390_PC64)
rc = apply_rela_bits(loc, val, 1, 64, 0);
break;
case R_390_GOT12: /* 12 bit GOT offset. */
case R_390_GOT16: /* 16 bit GOT offset. */
case R_390_GOT20: /* 20 bit GOT offset. */
case R_390_GOT32: /* 32 bit GOT offset. */
case R_390_GOT64: /* 64 bit GOT offset. */
case R_390_GOTENT: /* 32 bit PC rel. to GOT entry shifted by 1. */
case R_390_GOTPLT12: /* 12 bit offset to jump slot. */
case R_390_GOTPLT20: /* 20 bit offset to jump slot. */
case R_390_GOTPLT16: /* 16 bit offset to jump slot. */
case R_390_GOTPLT32: /* 32 bit offset to jump slot. */
case R_390_GOTPLT64: /* 64 bit offset to jump slot. */
case R_390_GOTPLTENT: /* 32 bit rel. offset to jump slot >> 1. */
if (info->got_initialized == 0) {
Elf_Addr *gotent;
gotent = me->core_layout.base + me->arch.got_offset +
info->got_offset;
*gotent = val;
info->got_initialized = 1;
}
val = info->got_offset + rela->r_addend;
if (r_type == R_390_GOT12 ||
r_type == R_390_GOTPLT12)
rc = apply_rela_bits(loc, val, 0, 12, 0);
else if (r_type == R_390_GOT16 ||
r_type == R_390_GOTPLT16)
rc = apply_rela_bits(loc, val, 0, 16, 0);
else if (r_type == R_390_GOT20 ||
r_type == R_390_GOTPLT20)
rc = apply_rela_bits(loc, val, 1, 20, 0);
else if (r_type == R_390_GOT32 ||
r_type == R_390_GOTPLT32)
rc = apply_rela_bits(loc, val, 0, 32, 0);
else if (r_type == R_390_GOT64 ||
r_type == R_390_GOTPLT64)
rc = apply_rela_bits(loc, val, 0, 64, 0);
else if (r_type == R_390_GOTENT ||
r_type == R_390_GOTPLTENT) {
val += (Elf_Addr) me->core_layout.base - loc;
rc = apply_rela_bits(loc, val, 1, 32, 1);
}
break;
case R_390_PLT16DBL: /* 16 bit PC rel. PLT shifted by 1. */
case R_390_PLT32DBL: /* 32 bit PC rel. PLT shifted by 1. */
case R_390_PLT32: /* 32 bit PC relative PLT address. */
case R_390_PLT64: /* 64 bit PC relative PLT address. */
case R_390_PLTOFF16: /* 16 bit offset from GOT to PLT. */
case R_390_PLTOFF32: /* 32 bit offset from GOT to PLT. */
case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */
if (info->plt_initialized == 0) {
unsigned int *ip;
ip = me->core_layout.base + me->arch.plt_offset +
info->plt_offset;
ip[0] = 0x0d10e310; /* basr 1,0 */
ip[1] = 0x100a0004; /* lg 1,10(1) */
if (IS_ENABLED(CONFIG_EXPOLINE) && !nospec_disable) {
unsigned int *ij;
ij = me->core_layout.base +
me->arch.plt_offset +
me->arch.plt_size - PLT_ENTRY_SIZE;
ip[2] = 0xa7f40000 + /* j __jump_r1 */
(unsigned int)(u16)
(((unsigned long) ij - 8 -
(unsigned long) ip) / 2);
} else {
ip[2] = 0x07f10000; /* br %r1 */
}
ip[3] = (unsigned int) (val >> 32);
ip[4] = (unsigned int) val;
info->plt_initialized = 1;
}
if (r_type == R_390_PLTOFF16 ||
r_type == R_390_PLTOFF32 ||
r_type == R_390_PLTOFF64)
val = me->arch.plt_offset - me->arch.got_offset +
info->plt_offset + rela->r_addend;
else {
if (!((r_type == R_390_PLT16DBL &&
val - loc + 0xffffUL < 0x1ffffeUL) ||
(r_type == R_390_PLT32DBL &&
val - loc + 0xffffffffULL < 0x1fffffffeULL)))
val = (Elf_Addr) me->core_layout.base +
me->arch.plt_offset +
info->plt_offset;
val += rela->r_addend - loc;
}
if (r_type == R_390_PLT16DBL)
rc = apply_rela_bits(loc, val, 1, 16, 1);
else if (r_type == R_390_PLTOFF16)
rc = apply_rela_bits(loc, val, 0, 16, 0);
else if (r_type == R_390_PLT32DBL)
rc = apply_rela_bits(loc, val, 1, 32, 1);
else if (r_type == R_390_PLT32 ||
r_type == R_390_PLTOFF32)
rc = apply_rela_bits(loc, val, 0, 32, 0);
else if (r_type == R_390_PLT64 ||
r_type == R_390_PLTOFF64)
rc = apply_rela_bits(loc, val, 0, 64, 0);
break;
case R_390_GOTOFF16: /* 16 bit offset to GOT. */
case R_390_GOTOFF32: /* 32 bit offset to GOT. */
case R_390_GOTOFF64: /* 64 bit offset to GOT. */
val = val + rela->r_addend -
((Elf_Addr) me->core_layout.base + me->arch.got_offset);
if (r_type == R_390_GOTOFF16)
rc = apply_rela_bits(loc, val, 0, 16, 0);
else if (r_type == R_390_GOTOFF32)
rc = apply_rela_bits(loc, val, 0, 32, 0);
else if (r_type == R_390_GOTOFF64)
rc = apply_rela_bits(loc, val, 0, 64, 0);
break;
case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */
case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */
val = (Elf_Addr) me->core_layout.base + me->arch.got_offset +
rela->r_addend - loc;
if (r_type == R_390_GOTPC)
rc = apply_rela_bits(loc, val, 1, 32, 0);
else if (r_type == R_390_GOTPCDBL)
rc = apply_rela_bits(loc, val, 1, 32, 1);
break;
case R_390_COPY:
case R_390_GLOB_DAT: /* Create GOT entry. */
case R_390_JMP_SLOT: /* Create PLT entry. */
case R_390_RELATIVE: /* Adjust by program base. */
/* Only needed if we want to support loading of
modules linked with -shared. */
return -ENOEXEC;
default:
printk(KERN_ERR "module %s: unknown relocation: %u\n",
me->name, r_type);
return -ENOEXEC;
}
if (rc) {
printk(KERN_ERR "module %s: relocation error for symbol %s "
"(r_type %i, value 0x%lx)\n",
me->name, strtab + symtab[r_sym].st_name,
r_type, (unsigned long) val);
return rc;
}
return 0;
}
int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab,
unsigned int symindex, unsigned int relsec,
struct module *me)
{
Elf_Addr base;
Elf_Sym *symtab;
Elf_Rela *rela;
unsigned long i, n;
int rc;
DEBUGP("Applying relocate section %u to %u\n",
relsec, sechdrs[relsec].sh_info);
base = sechdrs[sechdrs[relsec].sh_info].sh_addr;
symtab = (Elf_Sym *) sechdrs[symindex].sh_addr;
rela = (Elf_Rela *) sechdrs[relsec].sh_addr;
n = sechdrs[relsec].sh_size / sizeof(Elf_Rela);
for (i = 0; i < n; i++, rela++) {
rc = apply_rela(rela, base, symtab, strtab, me);
if (rc)
return rc;
}
return 0;
}
int module_finalize(const Elf_Ehdr *hdr,
const Elf_Shdr *sechdrs,
struct module *me)
{
const Elf_Shdr *s;
char *secstrings, *secname;
void *aseg;
if (IS_ENABLED(CONFIG_EXPOLINE) &&
!nospec_disable && me->arch.plt_size) {
unsigned int *ij;
ij = me->core_layout.base + me->arch.plt_offset +
me->arch.plt_size - PLT_ENTRY_SIZE;
if (test_facility(35)) {
ij[0] = 0xc6000000; /* exrl %r0,.+10 */
ij[1] = 0x0005a7f4; /* j . */
ij[2] = 0x000007f1; /* br %r1 */
} else {
ij[0] = 0x44000000 | (unsigned int)
offsetof(struct lowcore, br_r1_trampoline);
ij[1] = 0xa7f40000; /* j . */
}
}
secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) {
aseg = (void *) s->sh_addr;
secname = secstrings + s->sh_name;
if (!strcmp(".altinstructions", secname))
/* patch .altinstructions */
apply_alternatives(aseg, aseg + s->sh_size);
if (IS_ENABLED(CONFIG_EXPOLINE) &&
(!strncmp(".s390_indirect", secname, 14)))
nospec_revert(aseg, aseg + s->sh_size);
if (IS_ENABLED(CONFIG_EXPOLINE) &&
(!strncmp(".s390_return", secname, 12)))
nospec_revert(aseg, aseg + s->sh_size);
}
jump_label_apply_nops(me);
return 0;
}
+476
View File
@@ -0,0 +1,476 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Machine check handler
*
* Copyright IBM Corp. 2000, 2009
* Author(s): Ingo Adlung <adlung@de.ibm.com>,
* Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Cornelia Huck <cornelia.huck@de.ibm.com>,
* Heiko Carstens <heiko.carstens@de.ibm.com>,
*/
#include <linux/kernel_stat.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/hardirq.h>
#include <linux/log2.h>
#include <linux/kprobes.h>
#include <linux/kmemleak.h>
#include <linux/time.h>
#include <linux/module.h>
#include <linux/sched/signal.h>
#include <linux/export.h>
#include <asm/lowcore.h>
#include <asm/smp.h>
#include <asm/stp.h>
#include <asm/cputime.h>
#include <asm/nmi.h>
#include <asm/crw.h>
#include <asm/switch_to.h>
#include <asm/ctl_reg.h>
#include <asm/asm-offsets.h>
#include <linux/kvm_host.h>
struct mcck_struct {
unsigned int kill_task : 1;
unsigned int channel_report : 1;
unsigned int warning : 1;
unsigned int stp_queue : 1;
unsigned long mcck_code;
};
static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
static struct kmem_cache *mcesa_cache;
static unsigned long mcesa_origin_lc;
static inline int nmi_needs_mcesa(void)
{
return MACHINE_HAS_VX || MACHINE_HAS_GS;
}
static inline unsigned long nmi_get_mcesa_size(void)
{
if (MACHINE_HAS_GS)
return MCESA_MAX_SIZE;
return MCESA_MIN_SIZE;
}
/*
* The initial machine check extended save area for the boot CPU.
* It will be replaced by nmi_init() with an allocated structure.
* The structure is required for machine check happening early in
* the boot process.
*/
static struct mcesa boot_mcesa __initdata __aligned(MCESA_MAX_SIZE);
void __init nmi_alloc_boot_cpu(struct lowcore *lc)
{
if (!nmi_needs_mcesa())
return;
lc->mcesad = (unsigned long) &boot_mcesa;
if (MACHINE_HAS_GS)
lc->mcesad |= ilog2(MCESA_MAX_SIZE);
}
static int __init nmi_init(void)
{
unsigned long origin, cr0, size;
if (!nmi_needs_mcesa())
return 0;
size = nmi_get_mcesa_size();
if (size > MCESA_MIN_SIZE)
mcesa_origin_lc = ilog2(size);
/* create slab cache for the machine-check-extended-save-areas */
mcesa_cache = kmem_cache_create("nmi_save_areas", size, size, 0, NULL);
if (!mcesa_cache)
panic("Couldn't create nmi save area cache");
origin = (unsigned long) kmem_cache_alloc(mcesa_cache, GFP_KERNEL);
if (!origin)
panic("Couldn't allocate nmi save area");
/* The pointer is stored with mcesa_bits ORed in */
kmemleak_not_leak((void *) origin);
__ctl_store(cr0, 0, 0);
__ctl_clear_bit(0, 28); /* disable lowcore protection */
/* Replace boot_mcesa on the boot CPU */
S390_lowcore.mcesad = origin | mcesa_origin_lc;
__ctl_load(cr0, 0, 0);
return 0;
}
early_initcall(nmi_init);
int nmi_alloc_per_cpu(struct lowcore *lc)
{
unsigned long origin;
if (!nmi_needs_mcesa())
return 0;
origin = (unsigned long) kmem_cache_alloc(mcesa_cache, GFP_KERNEL);
if (!origin)
return -ENOMEM;
/* The pointer is stored with mcesa_bits ORed in */
kmemleak_not_leak((void *) origin);
lc->mcesad = origin | mcesa_origin_lc;
return 0;
}
void nmi_free_per_cpu(struct lowcore *lc)
{
if (!nmi_needs_mcesa())
return;
kmem_cache_free(mcesa_cache, (void *)(lc->mcesad & MCESA_ORIGIN_MASK));
}
static notrace void s390_handle_damage(void)
{
smp_emergency_stop();
disabled_wait((unsigned long) __builtin_return_address(0));
while (1);
}
NOKPROBE_SYMBOL(s390_handle_damage);
/*
* Main machine check handler function. Will be called with interrupts enabled
* or disabled and machine checks enabled or disabled.
*/
void s390_handle_mcck(void)
{
unsigned long flags;
struct mcck_struct mcck;
/*
* Disable machine checks and get the current state of accumulated
* machine checks. Afterwards delete the old state and enable machine
* checks again.
*/
local_irq_save(flags);
local_mcck_disable();
mcck = *this_cpu_ptr(&cpu_mcck);
memset(this_cpu_ptr(&cpu_mcck), 0, sizeof(mcck));
clear_cpu_flag(CIF_MCCK_PENDING);
local_mcck_enable();
local_irq_restore(flags);
if (mcck.channel_report)
crw_handle_channel_report();
/*
* A warning may remain for a prolonged period on the bare iron.
* (actually until the machine is powered off, or the problem is gone)
* So we just stop listening for the WARNING MCH and avoid continuously
* being interrupted. One caveat is however, that we must do this per
* processor and cannot use the smp version of ctl_clear_bit().
* On VM we only get one interrupt per virtally presented machinecheck.
* Though one suffices, we may get one interrupt per (virtual) cpu.
*/
if (mcck.warning) { /* WARNING pending ? */
static int mchchk_wng_posted = 0;
/* Use single cpu clear, as we cannot handle smp here. */
__ctl_clear_bit(14, 24); /* Disable WARNING MCH */
if (xchg(&mchchk_wng_posted, 1) == 0)
kill_cad_pid(SIGPWR, 1);
}
if (mcck.stp_queue)
stp_queue_work();
if (mcck.kill_task) {
local_irq_enable();
printk(KERN_EMERG "mcck: Terminating task because of machine "
"malfunction (code 0x%016lx).\n", mcck.mcck_code);
printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
current->comm, current->pid);
do_exit(SIGSEGV);
}
}
EXPORT_SYMBOL_GPL(s390_handle_mcck);
/*
* returns 0 if all required registers are available
* returns 1 otherwise
*/
static int notrace s390_check_registers(union mci mci, int umode)
{
union ctlreg2 cr2;
int kill_task;
kill_task = 0;
if (!mci.gr) {
/*
* General purpose registers couldn't be restored and have
* unknown contents. Stop system or terminate process.
*/
if (!umode)
s390_handle_damage();
kill_task = 1;
}
/* Check control registers */
if (!mci.cr) {
/*
* Control registers have unknown contents.
* Can't recover and therefore stopping machine.
*/
s390_handle_damage();
}
if (!mci.fp) {
/*
* Floating point registers can't be restored. If the
* kernel currently uses floating point registers the
* system is stopped. If the process has its floating
* pointer registers loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_VXR_V0V7)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
if (!mci.fc) {
/*
* Floating point control register can't be restored.
* If the kernel currently uses the floating pointer
* registers and needs the FPC register the system is
* stopped. If the process has its floating pointer
* registers loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_FPC)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
if (MACHINE_HAS_VX) {
if (!mci.vr) {
/*
* Vector registers can't be restored. If the kernel
* currently uses vector registers the system is
* stopped. If the process has its vector registers
* loaded it is terminated.
*/
if (S390_lowcore.fpu_flags & KERNEL_VXR)
s390_handle_damage();
if (!test_cpu_flag(CIF_FPU))
kill_task = 1;
}
}
/* Check if access registers are valid */
if (!mci.ar) {
/*
* Access registers have unknown contents.
* Terminating task.
*/
kill_task = 1;
}
/* Check guarded storage registers */
cr2.val = S390_lowcore.cregs_save_area[2];
if (cr2.gse) {
if (!mci.gs) {
/*
* Guarded storage register can't be restored and
* the current processes uses guarded storage.
* It has to be terminated.
*/
kill_task = 1;
}
}
/* Check if old PSW is valid */
if (!mci.wp) {
/*
* Can't tell if we come from user or kernel mode
* -> stopping machine.
*/
s390_handle_damage();
}
/* Check for invalid kernel instruction address */
if (!mci.ia && !umode) {
/*
* The instruction address got lost while running
* in the kernel -> stopping machine.
*/
s390_handle_damage();
}
if (!mci.ms || !mci.pm || !mci.ia)
kill_task = 1;
return kill_task;
}
NOKPROBE_SYMBOL(s390_check_registers);
/*
* Backup the guest's machine check info to its description block
*/
static void notrace s390_backup_mcck_info(struct pt_regs *regs)
{
struct mcck_volatile_info *mcck_backup;
struct sie_page *sie_page;
/* r14 contains the sie block, which was set in sie64a */
struct kvm_s390_sie_block *sie_block =
(struct kvm_s390_sie_block *) regs->gprs[14];
if (sie_block == NULL)
/* Something's seriously wrong, stop system. */
s390_handle_damage();
sie_page = container_of(sie_block, struct sie_page, sie_block);
mcck_backup = &sie_page->mcck_info;
mcck_backup->mcic = S390_lowcore.mcck_interruption_code &
~(MCCK_CODE_CP | MCCK_CODE_EXT_DAMAGE);
mcck_backup->ext_damage_code = S390_lowcore.external_damage_code;
mcck_backup->failing_storage_address
= S390_lowcore.failing_storage_address;
}
NOKPROBE_SYMBOL(s390_backup_mcck_info);
#define MAX_IPD_COUNT 29
#define MAX_IPD_TIME (5 * 60 * USEC_PER_SEC) /* 5 minutes */
#define ED_STP_ISLAND 6 /* External damage STP island check */
#define ED_STP_SYNC 7 /* External damage STP sync check */
#define MCCK_CODE_NO_GUEST (MCCK_CODE_CP | MCCK_CODE_EXT_DAMAGE)
/*
* machine check handler.
*/
void notrace s390_do_machine_check(struct pt_regs *regs)
{
static int ipd_count;
static DEFINE_SPINLOCK(ipd_lock);
static unsigned long long last_ipd;
struct mcck_struct *mcck;
unsigned long long tmp;
union mci mci;
unsigned long mcck_dam_code;
nmi_enter();
inc_irq_stat(NMI_NMI);
mci.val = S390_lowcore.mcck_interruption_code;
mcck = this_cpu_ptr(&cpu_mcck);
if (mci.sd) {
/* System damage -> stopping machine */
s390_handle_damage();
}
/*
* Reinject the instruction processing damages' machine checks
* including Delayed Access Exception into the guest
* instead of damaging the host if they happen in the guest.
*/
if (mci.pd && !test_cpu_flag(CIF_MCCK_GUEST)) {
if (mci.b) {
/* Processing backup -> verify if we can survive this */
u64 z_mcic, o_mcic, t_mcic;
z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29);
o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 |
1ULL<<16);
t_mcic = mci.val;
if (((t_mcic & z_mcic) != 0) ||
((t_mcic & o_mcic) != o_mcic)) {
s390_handle_damage();
}
/*
* Nullifying exigent condition, therefore we might
* retry this instruction.
*/
spin_lock(&ipd_lock);
tmp = get_tod_clock();
if (((tmp - last_ipd) >> 12) < MAX_IPD_TIME)
ipd_count++;
else
ipd_count = 1;
last_ipd = tmp;
if (ipd_count == MAX_IPD_COUNT)
s390_handle_damage();
spin_unlock(&ipd_lock);
} else {
/* Processing damage -> stopping machine */
s390_handle_damage();
}
}
if (s390_check_registers(mci, user_mode(regs))) {
/*
* Couldn't restore all register contents for the
* user space process -> mark task for termination.
*/
mcck->kill_task = 1;
mcck->mcck_code = mci.val;
set_cpu_flag(CIF_MCCK_PENDING);
}
/*
* Backup the machine check's info if it happens when the guest
* is running.
*/
if (test_cpu_flag(CIF_MCCK_GUEST))
s390_backup_mcck_info(regs);
if (mci.cd) {
/* Timing facility damage */
s390_handle_damage();
}
if (mci.ed && mci.ec) {
/* External damage */
if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC))
mcck->stp_queue |= stp_sync_check();
if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND))
mcck->stp_queue |= stp_island_check();
if (mcck->stp_queue)
set_cpu_flag(CIF_MCCK_PENDING);
}
/*
* Reinject storage related machine checks into the guest if they
* happen when the guest is running.
*/
if (!test_cpu_flag(CIF_MCCK_GUEST)) {
if (mci.se)
/* Storage error uncorrected */
s390_handle_damage();
if (mci.ke)
/* Storage key-error uncorrected */
s390_handle_damage();
if (mci.ds && mci.fa)
/* Storage degradation */
s390_handle_damage();
}
if (mci.cp) {
/* Channel report word pending */
mcck->channel_report = 1;
set_cpu_flag(CIF_MCCK_PENDING);
}
if (mci.w) {
/* Warning pending */
mcck->warning = 1;
set_cpu_flag(CIF_MCCK_PENDING);
}
/*
* If there are only Channel Report Pending and External Damage
* machine checks, they will not be reinjected into the guest
* because they refer to host conditions only.
*/
mcck_dam_code = (mci.val & MCIC_SUBCLASS_MASK);
if (test_cpu_flag(CIF_MCCK_GUEST) &&
(mcck_dam_code & MCCK_CODE_NO_GUEST) != mcck_dam_code) {
/* Set exit reason code for host's later handling */
*((long *)(regs->gprs[15] + __SF_SIE_REASON)) = -EINTR;
}
clear_cpu_flag(CIF_MCCK_GUEST);
nmi_exit();
}
NOKPROBE_SYMBOL(s390_do_machine_check);
static int __init machine_check_init(void)
{
ctl_set_bit(14, 25); /* enable external damage MCH */
ctl_set_bit(14, 27); /* enable system recovery MCH */
ctl_set_bit(14, 24); /* enable warning MCH */
return 0;
}
early_initcall(machine_check_init);
+176
View File
@@ -0,0 +1,176 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/module.h>
#include <linux/device.h>
#include <linux/cpu.h>
#include <asm/nospec-branch.h>
static int __init nobp_setup_early(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (rc)
return rc;
if (enabled && test_facility(82)) {
/*
* The user explicitely requested nobp=1, enable it and
* disable the expoline support.
*/
__set_facility(82, S390_lowcore.alt_stfle_fac_list);
if (IS_ENABLED(CONFIG_EXPOLINE))
nospec_disable = 1;
} else {
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
}
return 0;
}
early_param("nobp", nobp_setup_early);
static int __init nospec_setup_early(char *str)
{
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
return 0;
}
early_param("nospec", nospec_setup_early);
static int __init nospec_report(void)
{
if (test_facility(156))
pr_info("Spectre V2 mitigation: etokens\n");
if (IS_ENABLED(CC_USING_EXPOLINE) && !nospec_disable)
pr_info("Spectre V2 mitigation: execute trampolines\n");
if (__test_facility(82, S390_lowcore.alt_stfle_fac_list))
pr_info("Spectre V2 mitigation: limited branch prediction\n");
return 0;
}
arch_initcall(nospec_report);
#ifdef CONFIG_EXPOLINE
int nospec_disable = IS_ENABLED(CONFIG_EXPOLINE_OFF);
static int __init nospectre_v2_setup_early(char *str)
{
nospec_disable = 1;
return 0;
}
early_param("nospectre_v2", nospectre_v2_setup_early);
void __init nospec_auto_detect(void)
{
if (test_facility(156) || cpu_mitigations_off()) {
/*
* The machine supports etokens.
* Disable expolines and disable nobp.
*/
if (IS_ENABLED(CC_USING_EXPOLINE))
nospec_disable = 1;
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
} else if (IS_ENABLED(CC_USING_EXPOLINE)) {
/*
* The kernel has been compiled with expolines.
* Keep expolines enabled and disable nobp.
*/
nospec_disable = 0;
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
}
/*
* If the kernel has not been compiled with expolines the
* nobp setting decides what is done, this depends on the
* CONFIG_KERNEL_NP option and the nobp/nospec parameters.
*/
}
static int __init spectre_v2_setup_early(char *str)
{
if (str && !strncmp(str, "on", 2)) {
nospec_disable = 0;
__clear_facility(82, S390_lowcore.alt_stfle_fac_list);
}
if (str && !strncmp(str, "off", 3))
nospec_disable = 1;
if (str && !strncmp(str, "auto", 4))
nospec_auto_detect();
return 0;
}
early_param("spectre_v2", spectre_v2_setup_early);
static void __init_or_module __nospec_revert(s32 *start, s32 *end)
{
enum { BRCL_EXPOLINE, BRASL_EXPOLINE } type;
u8 *instr, *thunk, *br;
u8 insnbuf[6];
s32 *epo;
/* Second part of the instruction replace is always a nop */
for (epo = start; epo < end; epo++) {
instr = (u8 *) epo + *epo;
if (instr[0] == 0xc0 && (instr[1] & 0x0f) == 0x04)
type = BRCL_EXPOLINE; /* brcl instruction */
else if (instr[0] == 0xc0 && (instr[1] & 0x0f) == 0x05)
type = BRASL_EXPOLINE; /* brasl instruction */
else
continue;
thunk = instr + (*(int *)(instr + 2)) * 2;
if (thunk[0] == 0xc6 && thunk[1] == 0x00)
/* exrl %r0,<target-br> */
br = thunk + (*(int *)(thunk + 2)) * 2;
else if (thunk[0] == 0xc0 && (thunk[1] & 0x0f) == 0x00 &&
thunk[6] == 0x44 && thunk[7] == 0x00 &&
(thunk[8] & 0x0f) == 0x00 && thunk[9] == 0x00 &&
(thunk[1] & 0xf0) == (thunk[8] & 0xf0))
/* larl %rx,<target br> + ex %r0,0(%rx) */
br = thunk + (*(int *)(thunk + 2)) * 2;
else
continue;
/* Check for unconditional branch 0x07f? or 0x47f???? */
if ((br[0] & 0xbf) != 0x07 || (br[1] & 0xf0) != 0xf0)
continue;
memcpy(insnbuf + 2, (char[]) { 0x47, 0x00, 0x07, 0x00 }, 4);
switch (type) {
case BRCL_EXPOLINE:
insnbuf[0] = br[0];
insnbuf[1] = (instr[1] & 0xf0) | (br[1] & 0x0f);
if (br[0] == 0x47) {
/* brcl to b, replace with bc + nopr */
insnbuf[2] = br[2];
insnbuf[3] = br[3];
} else {
/* brcl to br, replace with bcr + nop */
}
break;
case BRASL_EXPOLINE:
insnbuf[1] = (instr[1] & 0xf0) | (br[1] & 0x0f);
if (br[0] == 0x47) {
/* brasl to b, replace with bas + nopr */
insnbuf[0] = 0x4d;
insnbuf[2] = br[2];
insnbuf[3] = br[3];
} else {
/* brasl to br, replace with basr + nop */
insnbuf[0] = 0x0d;
}
break;
}
s390_kernel_write(instr, insnbuf, 6);
}
}
void __init_or_module nospec_revert(s32 *start, s32 *end)
{
if (nospec_disable)
__nospec_revert(start, end);
}
extern s32 __nospec_call_start[], __nospec_call_end[];
extern s32 __nospec_return_start[], __nospec_return_end[];
void __init nospec_init_branches(void)
{
nospec_revert(__nospec_call_start, __nospec_call_end);
nospec_revert(__nospec_return_start, __nospec_return_end);
}
#endif /* CONFIG_EXPOLINE */
+23
View File
@@ -0,0 +1,23 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/device.h>
#include <linux/cpu.h>
#include <asm/facility.h>
#include <asm/nospec-branch.h>
ssize_t cpu_show_spectre_v1(struct device *dev,
struct device_attribute *attr, char *buf)
{
return sprintf(buf, "Mitigation: __user pointer sanitization\n");
}
ssize_t cpu_show_spectre_v2(struct device *dev,
struct device_attribute *attr, char *buf)
{
if (test_facility(156))
return sprintf(buf, "Mitigation: etokens\n");
if (IS_ENABLED(CC_USING_EXPOLINE) && !nospec_disable)
return sprintf(buf, "Mitigation: execute trampolines\n");
if (__test_facility(82, S390_lowcore.alt_stfle_fac_list))
return sprintf(buf, "Mitigation: limited branch prediction\n");
return sprintf(buf, "Vulnerable\n");
}
+170
View File
@@ -0,0 +1,170 @@
// SPDX-License-Identifier: GPL-2.0
/*
* OS info memory interface
*
* Copyright IBM Corp. 2012
* Author(s): Michael Holzheu <holzheu@linux.vnet.ibm.com>
*/
#define KMSG_COMPONENT "os_info"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/crash_dump.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <asm/checksum.h>
#include <asm/lowcore.h>
#include <asm/os_info.h>
/*
* OS info structure has to be page aligned
*/
static struct os_info os_info __page_aligned_data;
/*
* Compute checksum over OS info structure
*/
u32 os_info_csum(struct os_info *os_info)
{
int size = sizeof(*os_info) - offsetof(struct os_info, version_major);
return (__force u32)csum_partial(&os_info->version_major, size, 0);
}
/*
* Add crashkernel info to OS info and update checksum
*/
void os_info_crashkernel_add(unsigned long base, unsigned long size)
{
os_info.crashkernel_addr = (u64)(unsigned long)base;
os_info.crashkernel_size = (u64)(unsigned long)size;
os_info.csum = os_info_csum(&os_info);
}
/*
* Add OS info entry and update checksum
*/
void os_info_entry_add(int nr, void *ptr, u64 size)
{
os_info.entry[nr].addr = (u64)(unsigned long)ptr;
os_info.entry[nr].size = size;
os_info.entry[nr].csum = (__force u32)csum_partial(ptr, size, 0);
os_info.csum = os_info_csum(&os_info);
}
/*
* Initialize OS info struture and set lowcore pointer
*/
void __init os_info_init(void)
{
void *ptr = &os_info;
os_info.version_major = OS_INFO_VERSION_MAJOR;
os_info.version_minor = OS_INFO_VERSION_MINOR;
os_info.magic = OS_INFO_MAGIC;
os_info.csum = os_info_csum(&os_info);
mem_assign_absolute(S390_lowcore.os_info, (unsigned long) ptr);
}
#ifdef CONFIG_CRASH_DUMP
static struct os_info *os_info_old;
/*
* Allocate and copy OS info entry from oldmem
*/
static void os_info_old_alloc(int nr, int align)
{
unsigned long addr, size = 0;
char *buf, *buf_align, *msg;
u32 csum;
addr = os_info_old->entry[nr].addr;
if (!addr) {
msg = "not available";
goto fail;
}
size = os_info_old->entry[nr].size;
buf = kmalloc(size + align - 1, GFP_KERNEL);
if (!buf) {
msg = "alloc failed";
goto fail;
}
buf_align = PTR_ALIGN(buf, align);
if (copy_oldmem_kernel(buf_align, (void *) addr, size)) {
msg = "copy failed";
goto fail_free;
}
csum = (__force u32)csum_partial(buf_align, size, 0);
if (csum != os_info_old->entry[nr].csum) {
msg = "checksum failed";
goto fail_free;
}
os_info_old->entry[nr].addr = (u64)(unsigned long)buf_align;
msg = "copied";
goto out;
fail_free:
kfree(buf);
fail:
os_info_old->entry[nr].addr = 0;
out:
pr_info("entry %i: %s (addr=0x%lx size=%lu)\n",
nr, msg, addr, size);
}
/*
* Initialize os info and os info entries from oldmem
*/
static void os_info_old_init(void)
{
static int os_info_init;
unsigned long addr;
if (os_info_init)
return;
if (!OLDMEM_BASE)
goto fail;
if (copy_oldmem_kernel(&addr, &S390_lowcore.os_info, sizeof(addr)))
goto fail;
if (addr == 0 || addr % PAGE_SIZE)
goto fail;
os_info_old = kzalloc(sizeof(*os_info_old), GFP_KERNEL);
if (!os_info_old)
goto fail;
if (copy_oldmem_kernel(os_info_old, (void *) addr,
sizeof(*os_info_old)))
goto fail_free;
if (os_info_old->magic != OS_INFO_MAGIC)
goto fail_free;
if (os_info_old->csum != os_info_csum(os_info_old))
goto fail_free;
if (os_info_old->version_major > OS_INFO_VERSION_MAJOR)
goto fail_free;
os_info_old_alloc(OS_INFO_VMCOREINFO, 1);
os_info_old_alloc(OS_INFO_REIPL_BLOCK, 1);
pr_info("crashkernel: addr=0x%lx size=%lu\n",
(unsigned long) os_info_old->crashkernel_addr,
(unsigned long) os_info_old->crashkernel_size);
os_info_init = 1;
return;
fail_free:
kfree(os_info_old);
fail:
os_info_init = 1;
os_info_old = NULL;
}
/*
* Return pointer to os infor entry and its size
*/
void *os_info_old_entry(int nr, unsigned long *size)
{
os_info_old_init();
if (!os_info_old)
return NULL;
if (!os_info_old->entry[nr].addr)
return NULL;
*size = (unsigned long) os_info_old->entry[nr].size;
return (void *)(unsigned long)os_info_old->entry[nr].addr;
}
#endif
+724
View File
@@ -0,0 +1,724 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Performance event support for s390x - CPU-measurement Counter Facility
*
* Copyright IBM Corp. 2012, 2017
* Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
*/
#define KMSG_COMPONENT "cpum_cf"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/kernel.h>
#include <linux/kernel_stat.h>
#include <linux/perf_event.h>
#include <linux/percpu.h>
#include <linux/notifier.h>
#include <linux/init.h>
#include <linux/export.h>
#include <asm/ctl_reg.h>
#include <asm/irq.h>
#include <asm/cpu_mf.h>
enum cpumf_ctr_set {
CPUMF_CTR_SET_BASIC = 0, /* Basic Counter Set */
CPUMF_CTR_SET_USER = 1, /* Problem-State Counter Set */
CPUMF_CTR_SET_CRYPTO = 2, /* Crypto-Activity Counter Set */
CPUMF_CTR_SET_EXT = 3, /* Extended Counter Set */
CPUMF_CTR_SET_MT_DIAG = 4, /* MT-diagnostic Counter Set */
/* Maximum number of counter sets */
CPUMF_CTR_SET_MAX,
};
#define CPUMF_LCCTL_ENABLE_SHIFT 16
#define CPUMF_LCCTL_ACTCTL_SHIFT 0
static const u64 cpumf_state_ctl[CPUMF_CTR_SET_MAX] = {
[CPUMF_CTR_SET_BASIC] = 0x02,
[CPUMF_CTR_SET_USER] = 0x04,
[CPUMF_CTR_SET_CRYPTO] = 0x08,
[CPUMF_CTR_SET_EXT] = 0x01,
[CPUMF_CTR_SET_MT_DIAG] = 0x20,
};
static void ctr_set_enable(u64 *state, int ctr_set)
{
*state |= cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ENABLE_SHIFT;
}
static void ctr_set_disable(u64 *state, int ctr_set)
{
*state &= ~(cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ENABLE_SHIFT);
}
static void ctr_set_start(u64 *state, int ctr_set)
{
*state |= cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ACTCTL_SHIFT;
}
static void ctr_set_stop(u64 *state, int ctr_set)
{
*state &= ~(cpumf_state_ctl[ctr_set] << CPUMF_LCCTL_ACTCTL_SHIFT);
}
/* Local CPUMF event structure */
struct cpu_hw_events {
struct cpumf_ctr_info info;
atomic_t ctr_set[CPUMF_CTR_SET_MAX];
u64 state, tx_state;
unsigned int flags;
unsigned int txn_flags;
};
static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
.ctr_set = {
[CPUMF_CTR_SET_BASIC] = ATOMIC_INIT(0),
[CPUMF_CTR_SET_USER] = ATOMIC_INIT(0),
[CPUMF_CTR_SET_CRYPTO] = ATOMIC_INIT(0),
[CPUMF_CTR_SET_EXT] = ATOMIC_INIT(0),
[CPUMF_CTR_SET_MT_DIAG] = ATOMIC_INIT(0),
},
.state = 0,
.flags = 0,
.txn_flags = 0,
};
static enum cpumf_ctr_set get_counter_set(u64 event)
{
int set = CPUMF_CTR_SET_MAX;
if (event < 32)
set = CPUMF_CTR_SET_BASIC;
else if (event < 64)
set = CPUMF_CTR_SET_USER;
else if (event < 128)
set = CPUMF_CTR_SET_CRYPTO;
else if (event < 256)
set = CPUMF_CTR_SET_EXT;
else if (event >= 448 && event < 496)
set = CPUMF_CTR_SET_MT_DIAG;
return set;
}
static int validate_ctr_version(const struct hw_perf_event *hwc)
{
struct cpu_hw_events *cpuhw;
int err = 0;
u16 mtdiag_ctl;
cpuhw = &get_cpu_var(cpu_hw_events);
/* check required version for counter sets */
switch (hwc->config_base) {
case CPUMF_CTR_SET_BASIC:
case CPUMF_CTR_SET_USER:
if (cpuhw->info.cfvn < 1)
err = -EOPNOTSUPP;
break;
case CPUMF_CTR_SET_CRYPTO:
case CPUMF_CTR_SET_EXT:
if (cpuhw->info.csvn < 1)
err = -EOPNOTSUPP;
if ((cpuhw->info.csvn == 1 && hwc->config > 159) ||
(cpuhw->info.csvn == 2 && hwc->config > 175) ||
(cpuhw->info.csvn > 2 && hwc->config > 255))
err = -EOPNOTSUPP;
break;
case CPUMF_CTR_SET_MT_DIAG:
if (cpuhw->info.csvn <= 3)
err = -EOPNOTSUPP;
/*
* MT-diagnostic counters are read-only. The counter set
* is automatically enabled and activated on all CPUs with
* multithreading (SMT). Deactivation of multithreading
* also disables the counter set. State changes are ignored
* by lcctl(). Because Linux controls SMT enablement through
* a kernel parameter only, the counter set is either disabled
* or enabled and active.
*
* Thus, the counters can only be used if SMT is on and the
* counter set is enabled and active.
*/
mtdiag_ctl = cpumf_state_ctl[CPUMF_CTR_SET_MT_DIAG];
if (!((cpuhw->info.auth_ctl & mtdiag_ctl) &&
(cpuhw->info.enable_ctl & mtdiag_ctl) &&
(cpuhw->info.act_ctl & mtdiag_ctl)))
err = -EOPNOTSUPP;
break;
}
put_cpu_var(cpu_hw_events);
return err;
}
static int validate_ctr_auth(const struct hw_perf_event *hwc)
{
struct cpu_hw_events *cpuhw;
u64 ctrs_state;
int err = 0;
cpuhw = &get_cpu_var(cpu_hw_events);
/* Check authorization for cpu counter sets.
* If the particular CPU counter set is not authorized,
* return with -ENOENT in order to fall back to other
* PMUs that might suffice the event request.
*/
ctrs_state = cpumf_state_ctl[hwc->config_base];
if (!(ctrs_state & cpuhw->info.auth_ctl))
err = -ENOENT;
put_cpu_var(cpu_hw_events);
return err;
}
/*
* Change the CPUMF state to active.
* Enable and activate the CPU-counter sets according
* to the per-cpu control state.
*/
static void cpumf_pmu_enable(struct pmu *pmu)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
int err;
if (cpuhw->flags & PMU_F_ENABLED)
return;
err = lcctl(cpuhw->state);
if (err) {
pr_err("Enabling the performance measuring unit "
"failed with rc=%x\n", err);
return;
}
cpuhw->flags |= PMU_F_ENABLED;
}
/*
* Change the CPUMF state to inactive.
* Disable and enable (inactive) the CPU-counter sets according
* to the per-cpu control state.
*/
static void cpumf_pmu_disable(struct pmu *pmu)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
int err;
u64 inactive;
if (!(cpuhw->flags & PMU_F_ENABLED))
return;
inactive = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1);
err = lcctl(inactive);
if (err) {
pr_err("Disabling the performance measuring unit "
"failed with rc=%x\n", err);
return;
}
cpuhw->flags &= ~PMU_F_ENABLED;
}
/* Number of perf events counting hardware events */
static atomic_t num_events = ATOMIC_INIT(0);
/* Used to avoid races in calling reserve/release_cpumf_hardware */
static DEFINE_MUTEX(pmc_reserve_mutex);
/* CPU-measurement alerts for the counter facility */
static void cpumf_measurement_alert(struct ext_code ext_code,
unsigned int alert, unsigned long unused)
{
struct cpu_hw_events *cpuhw;
if (!(alert & CPU_MF_INT_CF_MASK))
return;
inc_irq_stat(IRQEXT_CMC);
cpuhw = this_cpu_ptr(&cpu_hw_events);
/* Measurement alerts are shared and might happen when the PMU
* is not reserved. Ignore these alerts in this case. */
if (!(cpuhw->flags & PMU_F_RESERVED))
return;
/* counter authorization change alert */
if (alert & CPU_MF_INT_CF_CACA)
qctri(&cpuhw->info);
/* loss of counter data alert */
if (alert & CPU_MF_INT_CF_LCDA)
pr_err("CPU[%i] Counter data was lost\n", smp_processor_id());
/* loss of MT counter data alert */
if (alert & CPU_MF_INT_CF_MTDA)
pr_warn("CPU[%i] MT counter data was lost\n",
smp_processor_id());
}
#define PMC_INIT 0
#define PMC_RELEASE 1
static void setup_pmc_cpu(void *flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
switch (*((int *) flags)) {
case PMC_INIT:
memset(&cpuhw->info, 0, sizeof(cpuhw->info));
qctri(&cpuhw->info);
cpuhw->flags |= PMU_F_RESERVED;
break;
case PMC_RELEASE:
cpuhw->flags &= ~PMU_F_RESERVED;
break;
}
/* Disable CPU counter sets */
lcctl(0);
}
/* Initialize the CPU-measurement facility */
static int reserve_pmc_hardware(void)
{
int flags = PMC_INIT;
on_each_cpu(setup_pmc_cpu, &flags, 1);
irq_subclass_register(IRQ_SUBCLASS_MEASUREMENT_ALERT);
return 0;
}
/* Release the CPU-measurement facility */
static void release_pmc_hardware(void)
{
int flags = PMC_RELEASE;
on_each_cpu(setup_pmc_cpu, &flags, 1);
irq_subclass_unregister(IRQ_SUBCLASS_MEASUREMENT_ALERT);
}
/* Release the PMU if event is the last perf event */
static void hw_perf_event_destroy(struct perf_event *event)
{
if (!atomic_add_unless(&num_events, -1, 1)) {
mutex_lock(&pmc_reserve_mutex);
if (atomic_dec_return(&num_events) == 0)
release_pmc_hardware();
mutex_unlock(&pmc_reserve_mutex);
}
}
/* CPUMF <-> perf event mappings for kernel+userspace (basic set) */
static const int cpumf_generic_events_basic[] = {
[PERF_COUNT_HW_CPU_CYCLES] = 0,
[PERF_COUNT_HW_INSTRUCTIONS] = 1,
[PERF_COUNT_HW_CACHE_REFERENCES] = -1,
[PERF_COUNT_HW_CACHE_MISSES] = -1,
[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = -1,
[PERF_COUNT_HW_BRANCH_MISSES] = -1,
[PERF_COUNT_HW_BUS_CYCLES] = -1,
};
/* CPUMF <-> perf event mappings for userspace (problem-state set) */
static const int cpumf_generic_events_user[] = {
[PERF_COUNT_HW_CPU_CYCLES] = 32,
[PERF_COUNT_HW_INSTRUCTIONS] = 33,
[PERF_COUNT_HW_CACHE_REFERENCES] = -1,
[PERF_COUNT_HW_CACHE_MISSES] = -1,
[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = -1,
[PERF_COUNT_HW_BRANCH_MISSES] = -1,
[PERF_COUNT_HW_BUS_CYCLES] = -1,
};
static int __hw_perf_event_init(struct perf_event *event)
{
struct perf_event_attr *attr = &event->attr;
struct hw_perf_event *hwc = &event->hw;
enum cpumf_ctr_set set;
int err;
u64 ev;
switch (attr->type) {
case PERF_TYPE_RAW:
/* Raw events are used to access counters directly,
* hence do not permit excludes */
if (attr->exclude_kernel || attr->exclude_user ||
attr->exclude_hv)
return -EOPNOTSUPP;
ev = attr->config;
break;
case PERF_TYPE_HARDWARE:
if (is_sampling_event(event)) /* No sampling support */
return -ENOENT;
ev = attr->config;
/* Count user space (problem-state) only */
if (!attr->exclude_user && attr->exclude_kernel) {
if (ev >= ARRAY_SIZE(cpumf_generic_events_user))
return -EOPNOTSUPP;
ev = cpumf_generic_events_user[ev];
/* No support for kernel space counters only */
} else if (!attr->exclude_kernel && attr->exclude_user) {
return -EOPNOTSUPP;
/* Count user and kernel space */
} else {
if (ev >= ARRAY_SIZE(cpumf_generic_events_basic))
return -EOPNOTSUPP;
ev = cpumf_generic_events_basic[ev];
}
break;
default:
return -ENOENT;
}
if (ev == -1)
return -ENOENT;
if (ev > PERF_CPUM_CF_MAX_CTR)
return -ENOENT;
/* Obtain the counter set to which the specified counter belongs */
set = get_counter_set(ev);
switch (set) {
case CPUMF_CTR_SET_BASIC:
case CPUMF_CTR_SET_USER:
case CPUMF_CTR_SET_CRYPTO:
case CPUMF_CTR_SET_EXT:
case CPUMF_CTR_SET_MT_DIAG:
/*
* Use the hardware perf event structure to store the
* counter number in the 'config' member and the counter
* set number in the 'config_base'. The counter set number
* is then later used to enable/disable the counter(s).
*/
hwc->config = ev;
hwc->config_base = set;
break;
case CPUMF_CTR_SET_MAX:
/* The counter could not be associated to a counter set */
return -EINVAL;
};
/* Initialize for using the CPU-measurement counter facility */
if (!atomic_inc_not_zero(&num_events)) {
mutex_lock(&pmc_reserve_mutex);
if (atomic_read(&num_events) == 0 && reserve_pmc_hardware())
err = -EBUSY;
else
atomic_inc(&num_events);
mutex_unlock(&pmc_reserve_mutex);
}
event->destroy = hw_perf_event_destroy;
/* Finally, validate version and authorization of the counter set */
err = validate_ctr_auth(hwc);
if (!err)
err = validate_ctr_version(hwc);
return err;
}
static int cpumf_pmu_event_init(struct perf_event *event)
{
int err;
switch (event->attr.type) {
case PERF_TYPE_HARDWARE:
case PERF_TYPE_HW_CACHE:
case PERF_TYPE_RAW:
err = __hw_perf_event_init(event);
break;
default:
return -ENOENT;
}
if (unlikely(err) && event->destroy)
event->destroy(event);
return err;
}
static int hw_perf_event_reset(struct perf_event *event)
{
u64 prev, new;
int err;
do {
prev = local64_read(&event->hw.prev_count);
err = ecctr(event->hw.config, &new);
if (err) {
if (err != 3)
break;
/* The counter is not (yet) available. This
* might happen if the counter set to which
* this counter belongs is in the disabled
* state.
*/
new = 0;
}
} while (local64_cmpxchg(&event->hw.prev_count, prev, new) != prev);
return err;
}
static void hw_perf_event_update(struct perf_event *event)
{
u64 prev, new, delta;
int err;
do {
prev = local64_read(&event->hw.prev_count);
err = ecctr(event->hw.config, &new);
if (err)
return;
} while (local64_cmpxchg(&event->hw.prev_count, prev, new) != prev);
delta = (prev <= new) ? new - prev
: (-1ULL - prev) + new + 1; /* overflow */
local64_add(delta, &event->count);
}
static void cpumf_pmu_read(struct perf_event *event)
{
if (event->hw.state & PERF_HES_STOPPED)
return;
hw_perf_event_update(event);
}
static void cpumf_pmu_start(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
return;
if (WARN_ON_ONCE(hwc->config == -1))
return;
if (flags & PERF_EF_RELOAD)
WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
hwc->state = 0;
/* (Re-)enable and activate the counter set */
ctr_set_enable(&cpuhw->state, hwc->config_base);
ctr_set_start(&cpuhw->state, hwc->config_base);
/* The counter set to which this counter belongs can be already active.
* Because all counters in a set are active, the event->hw.prev_count
* needs to be synchronized. At this point, the counter set can be in
* the inactive or disabled state.
*/
hw_perf_event_reset(event);
/* increment refcount for this counter set */
atomic_inc(&cpuhw->ctr_set[hwc->config_base]);
}
static void cpumf_pmu_stop(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
if (!(hwc->state & PERF_HES_STOPPED)) {
/* Decrement reference count for this counter set and if this
* is the last used counter in the set, clear activation
* control and set the counter set state to inactive.
*/
if (!atomic_dec_return(&cpuhw->ctr_set[hwc->config_base]))
ctr_set_stop(&cpuhw->state, hwc->config_base);
event->hw.state |= PERF_HES_STOPPED;
}
if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
hw_perf_event_update(event);
event->hw.state |= PERF_HES_UPTODATE;
}
}
static int cpumf_pmu_add(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
/* Check authorization for the counter set to which this
* counter belongs.
* For group events transaction, the authorization check is
* done in cpumf_pmu_commit_txn().
*/
if (!(cpuhw->txn_flags & PERF_PMU_TXN_ADD))
if (validate_ctr_auth(&event->hw))
return -ENOENT;
ctr_set_enable(&cpuhw->state, event->hw.config_base);
event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
if (flags & PERF_EF_START)
cpumf_pmu_start(event, PERF_EF_RELOAD);
perf_event_update_userpage(event);
return 0;
}
static void cpumf_pmu_del(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
cpumf_pmu_stop(event, PERF_EF_UPDATE);
/* Check if any counter in the counter set is still used. If not used,
* change the counter set to the disabled state. This also clears the
* content of all counters in the set.
*
* When a new perf event has been added but not yet started, this can
* clear enable control and resets all counters in a set. Therefore,
* cpumf_pmu_start() always has to reenable a counter set.
*/
if (!atomic_read(&cpuhw->ctr_set[event->hw.config_base]))
ctr_set_disable(&cpuhw->state, event->hw.config_base);
perf_event_update_userpage(event);
}
/*
* Start group events scheduling transaction.
* Set flags to perform a single test at commit time.
*
* We only support PERF_PMU_TXN_ADD transactions. Save the
* transaction flags but otherwise ignore non-PERF_PMU_TXN_ADD
* transactions.
*/
static void cpumf_pmu_start_txn(struct pmu *pmu, unsigned int txn_flags)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
WARN_ON_ONCE(cpuhw->txn_flags); /* txn already in flight */
cpuhw->txn_flags = txn_flags;
if (txn_flags & ~PERF_PMU_TXN_ADD)
return;
perf_pmu_disable(pmu);
cpuhw->tx_state = cpuhw->state;
}
/*
* Stop and cancel a group events scheduling tranctions.
* Assumes cpumf_pmu_del() is called for each successful added
* cpumf_pmu_add() during the transaction.
*/
static void cpumf_pmu_cancel_txn(struct pmu *pmu)
{
unsigned int txn_flags;
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
WARN_ON_ONCE(!cpuhw->txn_flags); /* no txn in flight */
txn_flags = cpuhw->txn_flags;
cpuhw->txn_flags = 0;
if (txn_flags & ~PERF_PMU_TXN_ADD)
return;
WARN_ON(cpuhw->tx_state != cpuhw->state);
perf_pmu_enable(pmu);
}
/*
* Commit the group events scheduling transaction. On success, the
* transaction is closed. On error, the transaction is kept open
* until cpumf_pmu_cancel_txn() is called.
*/
static int cpumf_pmu_commit_txn(struct pmu *pmu)
{
struct cpu_hw_events *cpuhw = this_cpu_ptr(&cpu_hw_events);
u64 state;
WARN_ON_ONCE(!cpuhw->txn_flags); /* no txn in flight */
if (cpuhw->txn_flags & ~PERF_PMU_TXN_ADD) {
cpuhw->txn_flags = 0;
return 0;
}
/* check if the updated state can be scheduled */
state = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1);
state >>= CPUMF_LCCTL_ENABLE_SHIFT;
if ((state & cpuhw->info.auth_ctl) != state)
return -ENOENT;
cpuhw->txn_flags = 0;
perf_pmu_enable(pmu);
return 0;
}
/* Performance monitoring unit for s390x */
static struct pmu cpumf_pmu = {
.task_ctx_nr = perf_sw_context,
.capabilities = PERF_PMU_CAP_NO_INTERRUPT,
.pmu_enable = cpumf_pmu_enable,
.pmu_disable = cpumf_pmu_disable,
.event_init = cpumf_pmu_event_init,
.add = cpumf_pmu_add,
.del = cpumf_pmu_del,
.start = cpumf_pmu_start,
.stop = cpumf_pmu_stop,
.read = cpumf_pmu_read,
.start_txn = cpumf_pmu_start_txn,
.commit_txn = cpumf_pmu_commit_txn,
.cancel_txn = cpumf_pmu_cancel_txn,
};
static int cpumf_pmf_setup(unsigned int cpu, int flags)
{
local_irq_disable();
setup_pmc_cpu(&flags);
local_irq_enable();
return 0;
}
static int s390_pmu_online_cpu(unsigned int cpu)
{
return cpumf_pmf_setup(cpu, PMC_INIT);
}
static int s390_pmu_offline_cpu(unsigned int cpu)
{
return cpumf_pmf_setup(cpu, PMC_RELEASE);
}
static int __init cpumf_pmu_init(void)
{
int rc;
if (!cpum_cf_avail())
return -ENODEV;
/* clear bit 15 of cr0 to unauthorize problem-state to
* extract measurement counters */
ctl_clear_bit(0, 48);
/* register handler for measurement-alert interruptions */
rc = register_external_irq(EXT_IRQ_MEASURE_ALERT,
cpumf_measurement_alert);
if (rc) {
pr_err("Registering for CPU-measurement alerts "
"failed with rc=%i\n", rc);
return rc;
}
cpumf_pmu.attr_groups = cpumf_cf_event_group();
rc = perf_pmu_register(&cpumf_pmu, "cpum_cf", PERF_TYPE_RAW);
if (rc) {
pr_err("Registering the cpum_cf PMU failed with rc=%i\n", rc);
unregister_external_irq(EXT_IRQ_MEASURE_ALERT,
cpumf_measurement_alert);
return rc;
}
return cpuhp_setup_state(CPUHP_AP_PERF_S390_CF_ONLINE,
"perf/s390/cf:online",
s390_pmu_online_cpu, s390_pmu_offline_cpu);
}
early_initcall(cpumf_pmu_init);
+598
View File
@@ -0,0 +1,598 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Perf PMU sysfs events attributes for available CPU-measurement counters
*
*/
#include <linux/slab.h>
#include <linux/perf_event.h>
/* BEGIN: CPUM_CF COUNTER DEFINITIONS =================================== */
CPUMF_EVENT_ATTR(cf_fvn1, CPU_CYCLES, 0x0000);
CPUMF_EVENT_ATTR(cf_fvn1, INSTRUCTIONS, 0x0001);
CPUMF_EVENT_ATTR(cf_fvn1, L1I_DIR_WRITES, 0x0002);
CPUMF_EVENT_ATTR(cf_fvn1, L1I_PENALTY_CYCLES, 0x0003);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_CPU_CYCLES, 0x0020);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_INSTRUCTIONS, 0x0021);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_L1I_DIR_WRITES, 0x0022);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_L1I_PENALTY_CYCLES, 0x0023);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_L1D_DIR_WRITES, 0x0024);
CPUMF_EVENT_ATTR(cf_fvn1, PROBLEM_STATE_L1D_PENALTY_CYCLES, 0x0025);
CPUMF_EVENT_ATTR(cf_fvn1, L1D_DIR_WRITES, 0x0004);
CPUMF_EVENT_ATTR(cf_fvn1, L1D_PENALTY_CYCLES, 0x0005);
CPUMF_EVENT_ATTR(cf_fvn3, CPU_CYCLES, 0x0000);
CPUMF_EVENT_ATTR(cf_fvn3, INSTRUCTIONS, 0x0001);
CPUMF_EVENT_ATTR(cf_fvn3, L1I_DIR_WRITES, 0x0002);
CPUMF_EVENT_ATTR(cf_fvn3, L1I_PENALTY_CYCLES, 0x0003);
CPUMF_EVENT_ATTR(cf_fvn3, PROBLEM_STATE_CPU_CYCLES, 0x0020);
CPUMF_EVENT_ATTR(cf_fvn3, PROBLEM_STATE_INSTRUCTIONS, 0x0021);
CPUMF_EVENT_ATTR(cf_fvn3, L1D_DIR_WRITES, 0x0004);
CPUMF_EVENT_ATTR(cf_fvn3, L1D_PENALTY_CYCLES, 0x0005);
CPUMF_EVENT_ATTR(cf_svn_generic, PRNG_FUNCTIONS, 0x0040);
CPUMF_EVENT_ATTR(cf_svn_generic, PRNG_CYCLES, 0x0041);
CPUMF_EVENT_ATTR(cf_svn_generic, PRNG_BLOCKED_FUNCTIONS, 0x0042);
CPUMF_EVENT_ATTR(cf_svn_generic, PRNG_BLOCKED_CYCLES, 0x0043);
CPUMF_EVENT_ATTR(cf_svn_generic, SHA_FUNCTIONS, 0x0044);
CPUMF_EVENT_ATTR(cf_svn_generic, SHA_CYCLES, 0x0045);
CPUMF_EVENT_ATTR(cf_svn_generic, SHA_BLOCKED_FUNCTIONS, 0x0046);
CPUMF_EVENT_ATTR(cf_svn_generic, SHA_BLOCKED_CYCLES, 0x0047);
CPUMF_EVENT_ATTR(cf_svn_generic, DEA_FUNCTIONS, 0x0048);
CPUMF_EVENT_ATTR(cf_svn_generic, DEA_CYCLES, 0x0049);
CPUMF_EVENT_ATTR(cf_svn_generic, DEA_BLOCKED_FUNCTIONS, 0x004a);
CPUMF_EVENT_ATTR(cf_svn_generic, DEA_BLOCKED_CYCLES, 0x004b);
CPUMF_EVENT_ATTR(cf_svn_generic, AES_FUNCTIONS, 0x004c);
CPUMF_EVENT_ATTR(cf_svn_generic, AES_CYCLES, 0x004d);
CPUMF_EVENT_ATTR(cf_svn_generic, AES_BLOCKED_FUNCTIONS, 0x004e);
CPUMF_EVENT_ATTR(cf_svn_generic, AES_BLOCKED_CYCLES, 0x004f);
CPUMF_EVENT_ATTR(cf_z10, L1I_L2_SOURCED_WRITES, 0x0080);
CPUMF_EVENT_ATTR(cf_z10, L1D_L2_SOURCED_WRITES, 0x0081);
CPUMF_EVENT_ATTR(cf_z10, L1I_L3_LOCAL_WRITES, 0x0082);
CPUMF_EVENT_ATTR(cf_z10, L1D_L3_LOCAL_WRITES, 0x0083);
CPUMF_EVENT_ATTR(cf_z10, L1I_L3_REMOTE_WRITES, 0x0084);
CPUMF_EVENT_ATTR(cf_z10, L1D_L3_REMOTE_WRITES, 0x0085);
CPUMF_EVENT_ATTR(cf_z10, L1D_LMEM_SOURCED_WRITES, 0x0086);
CPUMF_EVENT_ATTR(cf_z10, L1I_LMEM_SOURCED_WRITES, 0x0087);
CPUMF_EVENT_ATTR(cf_z10, L1D_RO_EXCL_WRITES, 0x0088);
CPUMF_EVENT_ATTR(cf_z10, L1I_CACHELINE_INVALIDATES, 0x0089);
CPUMF_EVENT_ATTR(cf_z10, ITLB1_WRITES, 0x008a);
CPUMF_EVENT_ATTR(cf_z10, DTLB1_WRITES, 0x008b);
CPUMF_EVENT_ATTR(cf_z10, TLB2_PTE_WRITES, 0x008c);
CPUMF_EVENT_ATTR(cf_z10, TLB2_CRSTE_WRITES, 0x008d);
CPUMF_EVENT_ATTR(cf_z10, TLB2_CRSTE_HPAGE_WRITES, 0x008e);
CPUMF_EVENT_ATTR(cf_z10, ITLB1_MISSES, 0x0091);
CPUMF_EVENT_ATTR(cf_z10, DTLB1_MISSES, 0x0092);
CPUMF_EVENT_ATTR(cf_z10, L2C_STORES_SENT, 0x0093);
CPUMF_EVENT_ATTR(cf_z196, L1D_L2_SOURCED_WRITES, 0x0080);
CPUMF_EVENT_ATTR(cf_z196, L1I_L2_SOURCED_WRITES, 0x0081);
CPUMF_EVENT_ATTR(cf_z196, DTLB1_MISSES, 0x0082);
CPUMF_EVENT_ATTR(cf_z196, ITLB1_MISSES, 0x0083);
CPUMF_EVENT_ATTR(cf_z196, L2C_STORES_SENT, 0x0085);
CPUMF_EVENT_ATTR(cf_z196, L1D_OFFBOOK_L3_SOURCED_WRITES, 0x0086);
CPUMF_EVENT_ATTR(cf_z196, L1D_ONBOOK_L4_SOURCED_WRITES, 0x0087);
CPUMF_EVENT_ATTR(cf_z196, L1I_ONBOOK_L4_SOURCED_WRITES, 0x0088);
CPUMF_EVENT_ATTR(cf_z196, L1D_RO_EXCL_WRITES, 0x0089);
CPUMF_EVENT_ATTR(cf_z196, L1D_OFFBOOK_L4_SOURCED_WRITES, 0x008a);
CPUMF_EVENT_ATTR(cf_z196, L1I_OFFBOOK_L4_SOURCED_WRITES, 0x008b);
CPUMF_EVENT_ATTR(cf_z196, DTLB1_HPAGE_WRITES, 0x008c);
CPUMF_EVENT_ATTR(cf_z196, L1D_LMEM_SOURCED_WRITES, 0x008d);
CPUMF_EVENT_ATTR(cf_z196, L1I_LMEM_SOURCED_WRITES, 0x008e);
CPUMF_EVENT_ATTR(cf_z196, L1I_OFFBOOK_L3_SOURCED_WRITES, 0x008f);
CPUMF_EVENT_ATTR(cf_z196, DTLB1_WRITES, 0x0090);
CPUMF_EVENT_ATTR(cf_z196, ITLB1_WRITES, 0x0091);
CPUMF_EVENT_ATTR(cf_z196, TLB2_PTE_WRITES, 0x0092);
CPUMF_EVENT_ATTR(cf_z196, TLB2_CRSTE_HPAGE_WRITES, 0x0093);
CPUMF_EVENT_ATTR(cf_z196, TLB2_CRSTE_WRITES, 0x0094);
CPUMF_EVENT_ATTR(cf_z196, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0096);
CPUMF_EVENT_ATTR(cf_z196, L1D_OFFCHIP_L3_SOURCED_WRITES, 0x0098);
CPUMF_EVENT_ATTR(cf_z196, L1I_ONCHIP_L3_SOURCED_WRITES, 0x0099);
CPUMF_EVENT_ATTR(cf_z196, L1I_OFFCHIP_L3_SOURCED_WRITES, 0x009b);
CPUMF_EVENT_ATTR(cf_zec12, DTLB1_MISSES, 0x0080);
CPUMF_EVENT_ATTR(cf_zec12, ITLB1_MISSES, 0x0081);
CPUMF_EVENT_ATTR(cf_zec12, L1D_L2I_SOURCED_WRITES, 0x0082);
CPUMF_EVENT_ATTR(cf_zec12, L1I_L2I_SOURCED_WRITES, 0x0083);
CPUMF_EVENT_ATTR(cf_zec12, L1D_L2D_SOURCED_WRITES, 0x0084);
CPUMF_EVENT_ATTR(cf_zec12, DTLB1_WRITES, 0x0085);
CPUMF_EVENT_ATTR(cf_zec12, L1D_LMEM_SOURCED_WRITES, 0x0087);
CPUMF_EVENT_ATTR(cf_zec12, L1I_LMEM_SOURCED_WRITES, 0x0089);
CPUMF_EVENT_ATTR(cf_zec12, L1D_RO_EXCL_WRITES, 0x008a);
CPUMF_EVENT_ATTR(cf_zec12, DTLB1_HPAGE_WRITES, 0x008b);
CPUMF_EVENT_ATTR(cf_zec12, ITLB1_WRITES, 0x008c);
CPUMF_EVENT_ATTR(cf_zec12, TLB2_PTE_WRITES, 0x008d);
CPUMF_EVENT_ATTR(cf_zec12, TLB2_CRSTE_HPAGE_WRITES, 0x008e);
CPUMF_EVENT_ATTR(cf_zec12, TLB2_CRSTE_WRITES, 0x008f);
CPUMF_EVENT_ATTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0090);
CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES, 0x0091);
CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES, 0x0092);
CPUMF_EVENT_ATTR(cf_zec12, L1D_ONBOOK_L4_SOURCED_WRITES, 0x0093);
CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L4_SOURCED_WRITES, 0x0094);
CPUMF_EVENT_ATTR(cf_zec12, TX_NC_TEND, 0x0095);
CPUMF_EVENT_ATTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES_IV, 0x0096);
CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES_IV, 0x0097);
CPUMF_EVENT_ATTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES_IV, 0x0098);
CPUMF_EVENT_ATTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES, 0x0099);
CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES, 0x009a);
CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES, 0x009b);
CPUMF_EVENT_ATTR(cf_zec12, L1I_ONBOOK_L4_SOURCED_WRITES, 0x009c);
CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L4_SOURCED_WRITES, 0x009d);
CPUMF_EVENT_ATTR(cf_zec12, TX_C_TEND, 0x009e);
CPUMF_EVENT_ATTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES_IV, 0x009f);
CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES_IV, 0x00a0);
CPUMF_EVENT_ATTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES_IV, 0x00a1);
CPUMF_EVENT_ATTR(cf_zec12, TX_NC_TABORT, 0x00b1);
CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_NO_SPECIAL, 0x00b2);
CPUMF_EVENT_ATTR(cf_zec12, TX_C_TABORT_SPECIAL, 0x00b3);
CPUMF_EVENT_ATTR(cf_z13, L1D_RO_EXCL_WRITES, 0x0080);
CPUMF_EVENT_ATTR(cf_z13, DTLB1_WRITES, 0x0081);
CPUMF_EVENT_ATTR(cf_z13, DTLB1_MISSES, 0x0082);
CPUMF_EVENT_ATTR(cf_z13, DTLB1_HPAGE_WRITES, 0x0083);
CPUMF_EVENT_ATTR(cf_z13, DTLB1_GPAGE_WRITES, 0x0084);
CPUMF_EVENT_ATTR(cf_z13, L1D_L2D_SOURCED_WRITES, 0x0085);
CPUMF_EVENT_ATTR(cf_z13, ITLB1_WRITES, 0x0086);
CPUMF_EVENT_ATTR(cf_z13, ITLB1_MISSES, 0x0087);
CPUMF_EVENT_ATTR(cf_z13, L1I_L2I_SOURCED_WRITES, 0x0088);
CPUMF_EVENT_ATTR(cf_z13, TLB2_PTE_WRITES, 0x0089);
CPUMF_EVENT_ATTR(cf_z13, TLB2_CRSTE_HPAGE_WRITES, 0x008a);
CPUMF_EVENT_ATTR(cf_z13, TLB2_CRSTE_WRITES, 0x008b);
CPUMF_EVENT_ATTR(cf_z13, TX_C_TEND, 0x008c);
CPUMF_EVENT_ATTR(cf_z13, TX_NC_TEND, 0x008d);
CPUMF_EVENT_ATTR(cf_z13, L1C_TLB1_MISSES, 0x008f);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0090);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES_IV, 0x0091);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L4_SOURCED_WRITES, 0x0092);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES_IV, 0x0093);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES, 0x0094);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L4_SOURCED_WRITES, 0x0095);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES_IV, 0x0096);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES, 0x0097);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L4_SOURCED_WRITES, 0x0098);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV, 0x0099);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES, 0x009a);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L4_SOURCED_WRITES, 0x009b);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV, 0x009c);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES, 0x009d);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONNODE_MEM_SOURCED_WRITES, 0x009e);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONDRAWER_MEM_SOURCED_WRITES, 0x009f);
CPUMF_EVENT_ATTR(cf_z13, L1D_OFFDRAWER_MEM_SOURCED_WRITES, 0x00a0);
CPUMF_EVENT_ATTR(cf_z13, L1D_ONCHIP_MEM_SOURCED_WRITES, 0x00a1);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES, 0x00a2);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES_IV, 0x00a3);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L4_SOURCED_WRITES, 0x00a4);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES_IV, 0x00a5);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES, 0x00a6);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L4_SOURCED_WRITES, 0x00a7);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES_IV, 0x00a8);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES, 0x00a9);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L4_SOURCED_WRITES, 0x00aa);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV, 0x00ab);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES, 0x00ac);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L4_SOURCED_WRITES, 0x00ad);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV, 0x00ae);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES, 0x00af);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONNODE_MEM_SOURCED_WRITES, 0x00b0);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONDRAWER_MEM_SOURCED_WRITES, 0x00b1);
CPUMF_EVENT_ATTR(cf_z13, L1I_OFFDRAWER_MEM_SOURCED_WRITES, 0x00b2);
CPUMF_EVENT_ATTR(cf_z13, L1I_ONCHIP_MEM_SOURCED_WRITES, 0x00b3);
CPUMF_EVENT_ATTR(cf_z13, TX_NC_TABORT, 0x00da);
CPUMF_EVENT_ATTR(cf_z13, TX_C_TABORT_NO_SPECIAL, 0x00db);
CPUMF_EVENT_ATTR(cf_z13, TX_C_TABORT_SPECIAL, 0x00dc);
CPUMF_EVENT_ATTR(cf_z13, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
CPUMF_EVENT_ATTR(cf_z13, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
CPUMF_EVENT_ATTR(cf_z14, L1D_RO_EXCL_WRITES, 0x0080);
CPUMF_EVENT_ATTR(cf_z14, DTLB2_WRITES, 0x0081);
CPUMF_EVENT_ATTR(cf_z14, DTLB2_MISSES, 0x0082);
CPUMF_EVENT_ATTR(cf_z14, DTLB2_HPAGE_WRITES, 0x0083);
CPUMF_EVENT_ATTR(cf_z14, DTLB2_GPAGE_WRITES, 0x0084);
CPUMF_EVENT_ATTR(cf_z14, L1D_L2D_SOURCED_WRITES, 0x0085);
CPUMF_EVENT_ATTR(cf_z14, ITLB2_WRITES, 0x0086);
CPUMF_EVENT_ATTR(cf_z14, ITLB2_MISSES, 0x0087);
CPUMF_EVENT_ATTR(cf_z14, L1I_L2I_SOURCED_WRITES, 0x0088);
CPUMF_EVENT_ATTR(cf_z14, TLB2_PTE_WRITES, 0x0089);
CPUMF_EVENT_ATTR(cf_z14, TLB2_CRSTE_WRITES, 0x008a);
CPUMF_EVENT_ATTR(cf_z14, TLB2_ENGINES_BUSY, 0x008b);
CPUMF_EVENT_ATTR(cf_z14, TX_C_TEND, 0x008c);
CPUMF_EVENT_ATTR(cf_z14, TX_NC_TEND, 0x008d);
CPUMF_EVENT_ATTR(cf_z14, L1C_TLB2_MISSES, 0x008f);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES, 0x0090);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCHIP_MEMORY_SOURCED_WRITES, 0x0091);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES_IV, 0x0092);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCLUSTER_L3_SOURCED_WRITES, 0x0093);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCLUSTER_MEMORY_SOURCED_WRITES, 0x0094);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCLUSTER_L3_SOURCED_WRITES_IV, 0x0095);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFCLUSTER_L3_SOURCED_WRITES, 0x0096);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFCLUSTER_MEMORY_SOURCED_WRITES, 0x0097);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFCLUSTER_L3_SOURCED_WRITES_IV, 0x0098);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFDRAWER_L3_SOURCED_WRITES, 0x0099);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFDRAWER_MEMORY_SOURCED_WRITES, 0x009a);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFDRAWER_L3_SOURCED_WRITES_IV, 0x009b);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONDRAWER_L4_SOURCED_WRITES, 0x009c);
CPUMF_EVENT_ATTR(cf_z14, L1D_OFFDRAWER_L4_SOURCED_WRITES, 0x009d);
CPUMF_EVENT_ATTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES_RO, 0x009e);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCHIP_L3_SOURCED_WRITES, 0x00a2);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCHIP_MEMORY_SOURCED_WRITES, 0x00a3);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCHIP_L3_SOURCED_WRITES_IV, 0x00a4);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCLUSTER_L3_SOURCED_WRITES, 0x00a5);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCLUSTER_MEMORY_SOURCED_WRITES, 0x00a6);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONCLUSTER_L3_SOURCED_WRITES_IV, 0x00a7);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFCLUSTER_L3_SOURCED_WRITES, 0x00a8);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFCLUSTER_MEMORY_SOURCED_WRITES, 0x00a9);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFCLUSTER_L3_SOURCED_WRITES_IV, 0x00aa);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFDRAWER_L3_SOURCED_WRITES, 0x00ab);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFDRAWER_MEMORY_SOURCED_WRITES, 0x00ac);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFDRAWER_L3_SOURCED_WRITES_IV, 0x00ad);
CPUMF_EVENT_ATTR(cf_z14, L1I_ONDRAWER_L4_SOURCED_WRITES, 0x00ae);
CPUMF_EVENT_ATTR(cf_z14, L1I_OFFDRAWER_L4_SOURCED_WRITES, 0x00af);
CPUMF_EVENT_ATTR(cf_z14, BCD_DFP_EXECUTION_SLOTS, 0x00e0);
CPUMF_EVENT_ATTR(cf_z14, VX_BCD_EXECUTION_SLOTS, 0x00e1);
CPUMF_EVENT_ATTR(cf_z14, DECIMAL_INSTRUCTIONS, 0x00e2);
CPUMF_EVENT_ATTR(cf_z14, LAST_HOST_TRANSLATIONS, 0x00e8);
CPUMF_EVENT_ATTR(cf_z14, TX_NC_TABORT, 0x00f3);
CPUMF_EVENT_ATTR(cf_z14, TX_C_TABORT_NO_SPECIAL, 0x00f4);
CPUMF_EVENT_ATTR(cf_z14, TX_C_TABORT_SPECIAL, 0x00f5);
CPUMF_EVENT_ATTR(cf_z14, MT_DIAG_CYCLES_ONE_THR_ACTIVE, 0x01c0);
CPUMF_EVENT_ATTR(cf_z14, MT_DIAG_CYCLES_TWO_THR_ACTIVE, 0x01c1);
static struct attribute *cpumcf_fvn1_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_fvn1, CPU_CYCLES),
CPUMF_EVENT_PTR(cf_fvn1, INSTRUCTIONS),
CPUMF_EVENT_PTR(cf_fvn1, L1I_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn1, L1I_PENALTY_CYCLES),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_CPU_CYCLES),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_INSTRUCTIONS),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_L1I_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_L1I_PENALTY_CYCLES),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_L1D_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn1, PROBLEM_STATE_L1D_PENALTY_CYCLES),
CPUMF_EVENT_PTR(cf_fvn1, L1D_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn1, L1D_PENALTY_CYCLES),
NULL,
};
static struct attribute *cpumcf_fvn3_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_fvn3, CPU_CYCLES),
CPUMF_EVENT_PTR(cf_fvn3, INSTRUCTIONS),
CPUMF_EVENT_PTR(cf_fvn3, L1I_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn3, L1I_PENALTY_CYCLES),
CPUMF_EVENT_PTR(cf_fvn3, PROBLEM_STATE_CPU_CYCLES),
CPUMF_EVENT_PTR(cf_fvn3, PROBLEM_STATE_INSTRUCTIONS),
CPUMF_EVENT_PTR(cf_fvn3, L1D_DIR_WRITES),
CPUMF_EVENT_PTR(cf_fvn3, L1D_PENALTY_CYCLES),
NULL,
};
static struct attribute *cpumcf_svn_generic_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_svn_generic, PRNG_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, PRNG_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, PRNG_BLOCKED_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, PRNG_BLOCKED_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, SHA_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, SHA_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, SHA_BLOCKED_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, SHA_BLOCKED_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, DEA_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, DEA_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, DEA_BLOCKED_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, DEA_BLOCKED_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, AES_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, AES_CYCLES),
CPUMF_EVENT_PTR(cf_svn_generic, AES_BLOCKED_FUNCTIONS),
CPUMF_EVENT_PTR(cf_svn_generic, AES_BLOCKED_CYCLES),
NULL,
};
static struct attribute *cpumcf_z10_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_z10, L1I_L2_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1D_L2_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1I_L3_LOCAL_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1D_L3_LOCAL_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1I_L3_REMOTE_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1D_L3_REMOTE_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1D_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1I_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1D_RO_EXCL_WRITES),
CPUMF_EVENT_PTR(cf_z10, L1I_CACHELINE_INVALIDATES),
CPUMF_EVENT_PTR(cf_z10, ITLB1_WRITES),
CPUMF_EVENT_PTR(cf_z10, DTLB1_WRITES),
CPUMF_EVENT_PTR(cf_z10, TLB2_PTE_WRITES),
CPUMF_EVENT_PTR(cf_z10, TLB2_CRSTE_WRITES),
CPUMF_EVENT_PTR(cf_z10, TLB2_CRSTE_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z10, ITLB1_MISSES),
CPUMF_EVENT_PTR(cf_z10, DTLB1_MISSES),
CPUMF_EVENT_PTR(cf_z10, L2C_STORES_SENT),
NULL,
};
static struct attribute *cpumcf_z196_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_z196, L1D_L2_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_L2_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, DTLB1_MISSES),
CPUMF_EVENT_PTR(cf_z196, ITLB1_MISSES),
CPUMF_EVENT_PTR(cf_z196, L2C_STORES_SENT),
CPUMF_EVENT_PTR(cf_z196, L1D_OFFBOOK_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_ONBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_ONBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_RO_EXCL_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_OFFBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_OFFBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, DTLB1_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_OFFBOOK_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, DTLB1_WRITES),
CPUMF_EVENT_PTR(cf_z196, ITLB1_WRITES),
CPUMF_EVENT_PTR(cf_z196, TLB2_PTE_WRITES),
CPUMF_EVENT_PTR(cf_z196, TLB2_CRSTE_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z196, TLB2_CRSTE_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1D_OFFCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z196, L1I_OFFCHIP_L3_SOURCED_WRITES),
NULL,
};
static struct attribute *cpumcf_zec12_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_zec12, DTLB1_MISSES),
CPUMF_EVENT_PTR(cf_zec12, ITLB1_MISSES),
CPUMF_EVENT_PTR(cf_zec12, L1D_L2I_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_L2I_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_L2D_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, DTLB1_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_LMEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_RO_EXCL_WRITES),
CPUMF_EVENT_PTR(cf_zec12, DTLB1_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_zec12, ITLB1_WRITES),
CPUMF_EVENT_PTR(cf_zec12, TLB2_PTE_WRITES),
CPUMF_EVENT_PTR(cf_zec12, TLB2_CRSTE_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_zec12, TLB2_CRSTE_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_ONBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, TX_NC_TEND),
CPUMF_EVENT_PTR(cf_zec12, L1D_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, L1D_OFFCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, L1D_OFFBOOK_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_ONBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_zec12, TX_C_TEND),
CPUMF_EVENT_PTR(cf_zec12, L1I_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, L1I_OFFCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, L1I_OFFBOOK_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_zec12, TX_NC_TABORT),
CPUMF_EVENT_PTR(cf_zec12, TX_C_TABORT_NO_SPECIAL),
CPUMF_EVENT_PTR(cf_zec12, TX_C_TABORT_SPECIAL),
NULL,
};
static struct attribute *cpumcf_z13_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_z13, L1D_RO_EXCL_WRITES),
CPUMF_EVENT_PTR(cf_z13, DTLB1_WRITES),
CPUMF_EVENT_PTR(cf_z13, DTLB1_MISSES),
CPUMF_EVENT_PTR(cf_z13, DTLB1_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z13, DTLB1_GPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_L2D_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, ITLB1_WRITES),
CPUMF_EVENT_PTR(cf_z13, ITLB1_MISSES),
CPUMF_EVENT_PTR(cf_z13, L1I_L2I_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, TLB2_PTE_WRITES),
CPUMF_EVENT_PTR(cf_z13, TLB2_CRSTE_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z13, TLB2_CRSTE_WRITES),
CPUMF_EVENT_PTR(cf_z13, TX_C_TEND),
CPUMF_EVENT_PTR(cf_z13, TX_NC_TEND),
CPUMF_EVENT_PTR(cf_z13, L1C_TLB1_MISSES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_SCOL_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_FCOL_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONNODE_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONDRAWER_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_OFFDRAWER_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1D_ONCHIP_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_SCOL_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_FCOL_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONNODE_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONDRAWER_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_OFFDRAWER_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, L1I_ONCHIP_MEM_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z13, TX_NC_TABORT),
CPUMF_EVENT_PTR(cf_z13, TX_C_TABORT_NO_SPECIAL),
CPUMF_EVENT_PTR(cf_z13, TX_C_TABORT_SPECIAL),
CPUMF_EVENT_PTR(cf_z13, MT_DIAG_CYCLES_ONE_THR_ACTIVE),
CPUMF_EVENT_PTR(cf_z13, MT_DIAG_CYCLES_TWO_THR_ACTIVE),
NULL,
};
static struct attribute *cpumcf_z14_pmu_event_attr[] __initdata = {
CPUMF_EVENT_PTR(cf_z14, L1D_RO_EXCL_WRITES),
CPUMF_EVENT_PTR(cf_z14, DTLB2_WRITES),
CPUMF_EVENT_PTR(cf_z14, DTLB2_MISSES),
CPUMF_EVENT_PTR(cf_z14, DTLB2_HPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z14, DTLB2_GPAGE_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_L2D_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, ITLB2_WRITES),
CPUMF_EVENT_PTR(cf_z14, ITLB2_MISSES),
CPUMF_EVENT_PTR(cf_z14, L1I_L2I_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, TLB2_PTE_WRITES),
CPUMF_EVENT_PTR(cf_z14, TLB2_CRSTE_WRITES),
CPUMF_EVENT_PTR(cf_z14, TLB2_ENGINES_BUSY),
CPUMF_EVENT_PTR(cf_z14, TX_C_TEND),
CPUMF_EVENT_PTR(cf_z14, TX_NC_TEND),
CPUMF_EVENT_PTR(cf_z14, L1C_TLB2_MISSES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCHIP_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCLUSTER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCLUSTER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCLUSTER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFCLUSTER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFCLUSTER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFCLUSTER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFDRAWER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFDRAWER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFDRAWER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1D_ONDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_OFFDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1D_ONCHIP_L3_SOURCED_WRITES_RO),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCHIP_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCHIP_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCHIP_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCLUSTER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCLUSTER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_ONCLUSTER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFCLUSTER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFCLUSTER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFCLUSTER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFDRAWER_L3_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFDRAWER_MEMORY_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFDRAWER_L3_SOURCED_WRITES_IV),
CPUMF_EVENT_PTR(cf_z14, L1I_ONDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, L1I_OFFDRAWER_L4_SOURCED_WRITES),
CPUMF_EVENT_PTR(cf_z14, BCD_DFP_EXECUTION_SLOTS),
CPUMF_EVENT_PTR(cf_z14, VX_BCD_EXECUTION_SLOTS),
CPUMF_EVENT_PTR(cf_z14, DECIMAL_INSTRUCTIONS),
CPUMF_EVENT_PTR(cf_z14, LAST_HOST_TRANSLATIONS),
CPUMF_EVENT_PTR(cf_z14, TX_NC_TABORT),
CPUMF_EVENT_PTR(cf_z14, TX_C_TABORT_NO_SPECIAL),
CPUMF_EVENT_PTR(cf_z14, TX_C_TABORT_SPECIAL),
CPUMF_EVENT_PTR(cf_z14, MT_DIAG_CYCLES_ONE_THR_ACTIVE),
CPUMF_EVENT_PTR(cf_z14, MT_DIAG_CYCLES_TWO_THR_ACTIVE),
NULL,
};
/* END: CPUM_CF COUNTER DEFINITIONS ===================================== */
static struct attribute_group cpumcf_pmu_events_group = {
.name = "events",
};
PMU_FORMAT_ATTR(event, "config:0-63");
static struct attribute *cpumcf_pmu_format_attr[] = {
&format_attr_event.attr,
NULL,
};
static struct attribute_group cpumcf_pmu_format_group = {
.name = "format",
.attrs = cpumcf_pmu_format_attr,
};
static const struct attribute_group *cpumcf_pmu_attr_groups[] = {
&cpumcf_pmu_events_group,
&cpumcf_pmu_format_group,
NULL,
};
static __init struct attribute **merge_attr(struct attribute **a,
struct attribute **b,
struct attribute **c)
{
struct attribute **new;
int j, i;
for (j = 0; a[j]; j++)
;
for (i = 0; b[i]; i++)
j++;
for (i = 0; c[i]; i++)
j++;
j++;
new = kmalloc_array(j, sizeof(struct attribute *), GFP_KERNEL);
if (!new)
return NULL;
j = 0;
for (i = 0; a[i]; i++)
new[j++] = a[i];
for (i = 0; b[i]; i++)
new[j++] = b[i];
for (i = 0; c[i]; i++)
new[j++] = c[i];
new[j] = NULL;
return new;
}
__init const struct attribute_group **cpumf_cf_event_group(void)
{
struct attribute **combined, **model, **cfvn, **csvn;
struct attribute *none[] = { NULL };
struct cpumf_ctr_info ci;
struct cpuid cpu_id;
/* Determine generic counters set(s) */
qctri(&ci);
switch (ci.cfvn) {
case 1:
cfvn = cpumcf_fvn1_pmu_event_attr;
break;
case 3:
cfvn = cpumcf_fvn3_pmu_event_attr;
break;
default:
cfvn = none;
}
csvn = cpumcf_svn_generic_pmu_event_attr;
/* Determine model-specific counter set(s) */
get_cpu_id(&cpu_id);
switch (cpu_id.machine) {
case 0x2097:
case 0x2098:
model = cpumcf_z10_pmu_event_attr;
break;
case 0x2817:
case 0x2818:
model = cpumcf_z196_pmu_event_attr;
break;
case 0x2827:
case 0x2828:
model = cpumcf_zec12_pmu_event_attr;
break;
case 0x2964:
case 0x2965:
model = cpumcf_z13_pmu_event_attr;
break;
case 0x3906:
case 0x3907:
model = cpumcf_z14_pmu_event_attr;
break;
default:
model = none;
break;
}
combined = merge_attr(cfvn, csvn, model);
if (combined)
cpumcf_pmu_events_group.attrs = combined;
return cpumcf_pmu_attr_groups;
}
File diff suppressed because it is too large. Load diff
+246
View File
@@ -0,0 +1,246 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Performance event support for s390x
*
* Copyright IBM Corp. 2012, 2013
* Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
*/
#define KMSG_COMPONENT "perf"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/kernel.h>
#include <linux/perf_event.h>
#include <linux/kvm_host.h>
#include <linux/percpu.h>
#include <linux/export.h>
#include <linux/seq_file.h>
#include <linux/spinlock.h>
#include <linux/sysfs.h>
#include <asm/irq.h>
#include <asm/cpu_mf.h>
#include <asm/lowcore.h>
#include <asm/processor.h>
#include <asm/sysinfo.h>
const char *perf_pmu_name(void)
{
if (cpum_cf_avail() || cpum_sf_avail())
return "CPU-Measurement Facilities (CPU-MF)";
return "pmu";
}
EXPORT_SYMBOL(perf_pmu_name);
int perf_num_counters(void)
{
int num = 0;
if (cpum_cf_avail())
num += PERF_CPUM_CF_MAX_CTR;
if (cpum_sf_avail())
num += PERF_CPUM_SF_MAX_CTR;
return num;
}
EXPORT_SYMBOL(perf_num_counters);
static struct kvm_s390_sie_block *sie_block(struct pt_regs *regs)
{
struct stack_frame *stack = (struct stack_frame *) regs->gprs[15];
if (!stack)
return NULL;
return (struct kvm_s390_sie_block *) stack->empty1[0];
}
static bool is_in_guest(struct pt_regs *regs)
{
if (user_mode(regs))
return false;
#if IS_ENABLED(CONFIG_KVM)
return instruction_pointer(regs) == (unsigned long) &sie_exit;
#else
return false;
#endif
}
static unsigned long guest_is_user_mode(struct pt_regs *regs)
{
return sie_block(regs)->gpsw.mask & PSW_MASK_PSTATE;
}
static unsigned long instruction_pointer_guest(struct pt_regs *regs)
{
return sie_block(regs)->gpsw.addr;
}
unsigned long perf_instruction_pointer(struct pt_regs *regs)
{
return is_in_guest(regs) ? instruction_pointer_guest(regs)
: instruction_pointer(regs);
}
static unsigned long perf_misc_guest_flags(struct pt_regs *regs)
{
return guest_is_user_mode(regs) ? PERF_RECORD_MISC_GUEST_USER
: PERF_RECORD_MISC_GUEST_KERNEL;
}
static unsigned long perf_misc_flags_sf(struct pt_regs *regs)
{
struct perf_sf_sde_regs *sde_regs;
unsigned long flags;
sde_regs = (struct perf_sf_sde_regs *) &regs->int_parm_long;
if (sde_regs->in_guest)
flags = user_mode(regs) ? PERF_RECORD_MISC_GUEST_USER
: PERF_RECORD_MISC_GUEST_KERNEL;
else
flags = user_mode(regs) ? PERF_RECORD_MISC_USER
: PERF_RECORD_MISC_KERNEL;
return flags;
}
unsigned long perf_misc_flags(struct pt_regs *regs)
{
/* Check if the cpum_sf PMU has created the pt_regs structure.
* In this case, perf misc flags can be easily extracted. Otherwise,
* do regular checks on the pt_regs content.
*/
if (regs->int_code == 0x1407 && regs->int_parm == CPU_MF_INT_SF_PRA)
if (!regs->gprs[15])
return perf_misc_flags_sf(regs);
if (is_in_guest(regs))
return perf_misc_guest_flags(regs);
return user_mode(regs) ? PERF_RECORD_MISC_USER
: PERF_RECORD_MISC_KERNEL;
}
static void print_debug_cf(void)
{
struct cpumf_ctr_info cf_info;
int cpu = smp_processor_id();
memset(&cf_info, 0, sizeof(cf_info));
if (!qctri(&cf_info))
pr_info("CPU[%i] CPUM_CF: ver=%u.%u A=%04x E=%04x C=%04x\n",
cpu, cf_info.cfvn, cf_info.csvn,
cf_info.auth_ctl, cf_info.enable_ctl, cf_info.act_ctl);
}
static void print_debug_sf(void)
{
struct hws_qsi_info_block si;
int cpu = smp_processor_id();
memset(&si, 0, sizeof(si));
if (qsi(&si))
return;
pr_info("CPU[%i] CPUM_SF: basic=%i diag=%i min=%lu max=%lu cpu_speed=%u\n",
cpu, si.as, si.ad, si.min_sampl_rate, si.max_sampl_rate,
si.cpu_speed);
if (si.as)
pr_info("CPU[%i] CPUM_SF: Basic-sampling: a=%i e=%i c=%i"
" bsdes=%i tear=%016lx dear=%016lx\n", cpu,
si.as, si.es, si.cs, si.bsdes, si.tear, si.dear);
if (si.ad)
pr_info("CPU[%i] CPUM_SF: Diagnostic-sampling: a=%i e=%i c=%i"
" dsdes=%i tear=%016lx dear=%016lx\n", cpu,
si.ad, si.ed, si.cd, si.dsdes, si.tear, si.dear);
}
void perf_event_print_debug(void)
{
unsigned long flags;
local_irq_save(flags);
if (cpum_cf_avail())
print_debug_cf();
if (cpum_sf_avail())
print_debug_sf();
local_irq_restore(flags);
}
/* Service level infrastructure */
static void sl_print_counter(struct seq_file *m)
{
struct cpumf_ctr_info ci;
memset(&ci, 0, sizeof(ci));
if (qctri(&ci))
return;
seq_printf(m, "CPU-MF: Counter facility: version=%u.%u "
"authorization=%04x\n", ci.cfvn, ci.csvn, ci.auth_ctl);
}
static void sl_print_sampling(struct seq_file *m)
{
struct hws_qsi_info_block si;
memset(&si, 0, sizeof(si));
if (qsi(&si))
return;
if (!si.as && !si.ad)
return;
seq_printf(m, "CPU-MF: Sampling facility: min_rate=%lu max_rate=%lu"
" cpu_speed=%u\n", si.min_sampl_rate, si.max_sampl_rate,
si.cpu_speed);
if (si.as)
seq_printf(m, "CPU-MF: Sampling facility: mode=basic"
" sample_size=%u\n", si.bsdes);
if (si.ad)
seq_printf(m, "CPU-MF: Sampling facility: mode=diagnostic"
" sample_size=%u\n", si.dsdes);
}
static void service_level_perf_print(struct seq_file *m,
struct service_level *sl)
{
if (cpum_cf_avail())
sl_print_counter(m);
if (cpum_sf_avail())
sl_print_sampling(m);
}
static struct service_level service_level_perf = {
.seq_print = service_level_perf_print,
};
static int __init service_level_perf_register(void)
{
return register_service_level(&service_level_perf);
}
arch_initcall(service_level_perf_register);
static int __perf_callchain_kernel(void *data, unsigned long address, int reliable)
{
struct perf_callchain_entry_ctx *entry = data;
perf_callchain_store(entry, address);
return 0;
}
void perf_callchain_kernel(struct perf_callchain_entry_ctx *entry,
struct pt_regs *regs)
{
if (user_mode(regs))
return;
dump_trace(__perf_callchain_kernel, entry, NULL, regs->gprs[15]);
}
/* Perf definitions for PMU event attributes in sysfs */
ssize_t cpumf_events_sysfs_show(struct device *dev,
struct device_attribute *attr, char *page)
{
struct perf_pmu_events_attr *pmu_attr;
pmu_attr = container_of(attr, struct perf_pmu_events_attr, attr);
return sprintf(page, "event=0x%04llx\n", pmu_attr->id);
}
+69
View File
@@ -0,0 +1,69 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/perf_event.h>
#include <linux/perf_regs.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/bug.h>
#include <asm/ptrace.h>
#include <asm/fpu/api.h>
#include <asm/fpu/types.h>
u64 perf_reg_value(struct pt_regs *regs, int idx)
{
freg_t fp;
if (idx >= PERF_REG_S390_R0 && idx <= PERF_REG_S390_R15)
return regs->gprs[idx];
if (idx >= PERF_REG_S390_FP0 && idx <= PERF_REG_S390_FP15) {
if (!user_mode(regs))
return 0;
idx -= PERF_REG_S390_FP0;
fp = MACHINE_HAS_VX ? *(freg_t *)(current->thread.fpu.vxrs + idx)
: current->thread.fpu.fprs[idx];
return fp.ui;
}
if (idx == PERF_REG_S390_MASK)
return regs->psw.mask;
if (idx == PERF_REG_S390_PC)
return regs->psw.addr;
WARN_ON_ONCE((u32)idx >= PERF_REG_S390_MAX);
return 0;
}
#define REG_RESERVED (~((1UL << PERF_REG_S390_MAX) - 1))
int perf_reg_validate(u64 mask)
{
if (!mask || mask & REG_RESERVED)
return -EINVAL;
return 0;
}
u64 perf_reg_abi(struct task_struct *task)
{
if (test_tsk_thread_flag(task, TIF_31BIT))
return PERF_SAMPLE_REGS_ABI_32;
return PERF_SAMPLE_REGS_ABI_64;
}
void perf_get_regs_user(struct perf_regs *regs_user,
struct pt_regs *regs,
struct pt_regs *regs_user_copy)
{
/*
* Use the regs from the first interruption and let
* perf_sample_regs_intr() handle interrupts (regs == get_irq_regs()).
*
* Also save FPU registers for user-space tasks only.
*/
regs_user->regs = task_pt_regs(current);
if (user_mode(regs_user->regs))
save_fpu_regs();
regs_user->abi = perf_reg_abi(current);
}
+147
View File
@@ -0,0 +1,147 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Program check table.
*
* Copyright IBM Corp. 2012
*/
#include <linux/linkage.h>
#define PGM_CHECK(handler) .long handler
#define PGM_CHECK_DEFAULT PGM_CHECK(default_trap_handler)
/*
* The program check table contains exactly 128 (0x00-0x7f) entries. Each
* line defines the function to be called corresponding to the program check
* interruption code.
*/
.section .rodata, "a"
ENTRY(pgm_check_table)
PGM_CHECK_DEFAULT /* 00 */
PGM_CHECK(illegal_op) /* 01 */
PGM_CHECK(privileged_op) /* 02 */
PGM_CHECK(execute_exception) /* 03 */
PGM_CHECK(do_protection_exception) /* 04 */
PGM_CHECK(addressing_exception) /* 05 */
PGM_CHECK(specification_exception) /* 06 */
PGM_CHECK(data_exception) /* 07 */
PGM_CHECK(overflow_exception) /* 08 */
PGM_CHECK(divide_exception) /* 09 */
PGM_CHECK(overflow_exception) /* 0a */
PGM_CHECK(divide_exception) /* 0b */
PGM_CHECK(hfp_overflow_exception) /* 0c */
PGM_CHECK(hfp_underflow_exception) /* 0d */
PGM_CHECK(hfp_significance_exception) /* 0e */
PGM_CHECK(hfp_divide_exception) /* 0f */
PGM_CHECK(do_dat_exception) /* 10 */
PGM_CHECK(do_dat_exception) /* 11 */
PGM_CHECK(translation_exception) /* 12 */
PGM_CHECK(special_op_exception) /* 13 */
PGM_CHECK_DEFAULT /* 14 */
PGM_CHECK(operand_exception) /* 15 */
PGM_CHECK_DEFAULT /* 16 */
PGM_CHECK_DEFAULT /* 17 */
PGM_CHECK(transaction_exception) /* 18 */
PGM_CHECK_DEFAULT /* 19 */
PGM_CHECK_DEFAULT /* 1a */
PGM_CHECK(vector_exception) /* 1b */
PGM_CHECK(space_switch_exception) /* 1c */
PGM_CHECK(hfp_sqrt_exception) /* 1d */
PGM_CHECK_DEFAULT /* 1e */
PGM_CHECK_DEFAULT /* 1f */
PGM_CHECK_DEFAULT /* 20 */
PGM_CHECK_DEFAULT /* 21 */
PGM_CHECK_DEFAULT /* 22 */
PGM_CHECK_DEFAULT /* 23 */
PGM_CHECK_DEFAULT /* 24 */
PGM_CHECK_DEFAULT /* 25 */
PGM_CHECK_DEFAULT /* 26 */
PGM_CHECK_DEFAULT /* 27 */
PGM_CHECK_DEFAULT /* 28 */
PGM_CHECK_DEFAULT /* 29 */
PGM_CHECK_DEFAULT /* 2a */
PGM_CHECK_DEFAULT /* 2b */
PGM_CHECK_DEFAULT /* 2c */
PGM_CHECK_DEFAULT /* 2d */
PGM_CHECK_DEFAULT /* 2e */
PGM_CHECK_DEFAULT /* 2f */
PGM_CHECK_DEFAULT /* 30 */
PGM_CHECK_DEFAULT /* 31 */
PGM_CHECK_DEFAULT /* 32 */
PGM_CHECK_DEFAULT /* 33 */
PGM_CHECK_DEFAULT /* 34 */
PGM_CHECK_DEFAULT /* 35 */
PGM_CHECK_DEFAULT /* 36 */
PGM_CHECK_DEFAULT /* 37 */
PGM_CHECK(do_dat_exception) /* 38 */
PGM_CHECK(do_dat_exception) /* 39 */
PGM_CHECK(do_dat_exception) /* 3a */
PGM_CHECK(do_dat_exception) /* 3b */
PGM_CHECK_DEFAULT /* 3c */
PGM_CHECK_DEFAULT /* 3d */
PGM_CHECK_DEFAULT /* 3e */
PGM_CHECK_DEFAULT /* 3f */
PGM_CHECK_DEFAULT /* 40 */
PGM_CHECK_DEFAULT /* 41 */
PGM_CHECK_DEFAULT /* 42 */
PGM_CHECK_DEFAULT /* 43 */
PGM_CHECK_DEFAULT /* 44 */
PGM_CHECK_DEFAULT /* 45 */
PGM_CHECK_DEFAULT /* 46 */
PGM_CHECK_DEFAULT /* 47 */
PGM_CHECK_DEFAULT /* 48 */
PGM_CHECK_DEFAULT /* 49 */
PGM_CHECK_DEFAULT /* 4a */
PGM_CHECK_DEFAULT /* 4b */
PGM_CHECK_DEFAULT /* 4c */
PGM_CHECK_DEFAULT /* 4d */
PGM_CHECK_DEFAULT /* 4e */
PGM_CHECK_DEFAULT /* 4f */
PGM_CHECK_DEFAULT /* 50 */
PGM_CHECK_DEFAULT /* 51 */
PGM_CHECK_DEFAULT /* 52 */
PGM_CHECK_DEFAULT /* 53 */
PGM_CHECK_DEFAULT /* 54 */
PGM_CHECK_DEFAULT /* 55 */
PGM_CHECK_DEFAULT /* 56 */
PGM_CHECK_DEFAULT /* 57 */
PGM_CHECK_DEFAULT /* 58 */
PGM_CHECK_DEFAULT /* 59 */
PGM_CHECK_DEFAULT /* 5a */
PGM_CHECK_DEFAULT /* 5b */
PGM_CHECK_DEFAULT /* 5c */
PGM_CHECK_DEFAULT /* 5d */
PGM_CHECK_DEFAULT /* 5e */
PGM_CHECK_DEFAULT /* 5f */
PGM_CHECK_DEFAULT /* 60 */
PGM_CHECK_DEFAULT /* 61 */
PGM_CHECK_DEFAULT /* 62 */
PGM_CHECK_DEFAULT /* 63 */
PGM_CHECK_DEFAULT /* 64 */
PGM_CHECK_DEFAULT /* 65 */
PGM_CHECK_DEFAULT /* 66 */
PGM_CHECK_DEFAULT /* 67 */
PGM_CHECK_DEFAULT /* 68 */
PGM_CHECK_DEFAULT /* 69 */
PGM_CHECK_DEFAULT /* 6a */
PGM_CHECK_DEFAULT /* 6b */
PGM_CHECK_DEFAULT /* 6c */
PGM_CHECK_DEFAULT /* 6d */
PGM_CHECK_DEFAULT /* 6e */
PGM_CHECK_DEFAULT /* 6f */
PGM_CHECK_DEFAULT /* 70 */
PGM_CHECK_DEFAULT /* 71 */
PGM_CHECK_DEFAULT /* 72 */
PGM_CHECK_DEFAULT /* 73 */
PGM_CHECK_DEFAULT /* 74 */
PGM_CHECK_DEFAULT /* 75 */
PGM_CHECK_DEFAULT /* 76 */
PGM_CHECK_DEFAULT /* 77 */
PGM_CHECK_DEFAULT /* 78 */
PGM_CHECK_DEFAULT /* 79 */
PGM_CHECK_DEFAULT /* 7a */
PGM_CHECK_DEFAULT /* 7b */
PGM_CHECK_DEFAULT /* 7c */
PGM_CHECK_DEFAULT /* 7d */
PGM_CHECK_DEFAULT /* 7e */
PGM_CHECK_DEFAULT /* 7f */
+243
View File
@@ -0,0 +1,243 @@
// SPDX-License-Identifier: GPL-2.0
/*
* This file handles the architecture dependent parts of process handling.
*
* Copyright IBM Corp. 1999, 2009
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
* Hartmut Penner <hp@de.ibm.com>,
* Denis Joseph Barrow,
*/
#include <linux/elf-randomize.h>
#include <linux/compiler.h>
#include <linux/cpu.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/sched/task.h>
#include <linux/sched/task_stack.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/elfcore.h>
#include <linux/smp.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/tick.h>
#include <linux/personality.h>
#include <linux/syscalls.h>
#include <linux/compat.h>
#include <linux/kprobes.h>
#include <linux/random.h>
#include <linux/export.h>
#include <linux/init_task.h>
#include <asm/cpu_mf.h>
#include <asm/io.h>
#include <asm/processor.h>
#include <asm/vtimer.h>
#include <asm/exec.h>
#include <asm/irq.h>
#include <asm/nmi.h>
#include <asm/smp.h>
#include <asm/switch_to.h>
#include <asm/runtime_instr.h>
#include "entry.h"
asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
extern void kernel_thread_starter(void);
void flush_thread(void)
{
}
void arch_setup_new_exec(void)
{
if (S390_lowcore.current_pid != current->pid) {
S390_lowcore.current_pid = current->pid;
if (test_facility(40))
lpp(&S390_lowcore.lpp);
}
}
void arch_release_task_struct(struct task_struct *tsk)
{
runtime_instr_release(tsk);
guarded_storage_release(tsk);
}
int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
{
/*
* Save the floating-point or vector register state of the current
* task and set the CIF_FPU flag to lazy restore the FPU register
* state when returning to user space.
*/
save_fpu_regs();
memcpy(dst, src, arch_task_struct_size);
dst->thread.fpu.regs = dst->thread.fpu.fprs;
return 0;
}
int copy_thread_tls(unsigned long clone_flags, unsigned long new_stackp,
unsigned long arg, struct task_struct *p, unsigned long tls)
{
struct fake_frame
{
struct stack_frame sf;
struct pt_regs childregs;
} *frame;
frame = container_of(task_pt_regs(p), struct fake_frame, childregs);
p->thread.ksp = (unsigned long) frame;
/* Save access registers to new thread structure. */
save_access_regs(&p->thread.acrs[0]);
/* start new process with ar4 pointing to the correct address space */
p->thread.mm_segment = get_fs();
/* Don't copy debug registers */
memset(&p->thread.per_user, 0, sizeof(p->thread.per_user));
memset(&p->thread.per_event, 0, sizeof(p->thread.per_event));
clear_tsk_thread_flag(p, TIF_SINGLE_STEP);
p->thread.per_flags = 0;
/* Initialize per thread user and system timer values */
p->thread.user_timer = 0;
p->thread.guest_timer = 0;
p->thread.system_timer = 0;
p->thread.hardirq_timer = 0;
p->thread.softirq_timer = 0;
frame->sf.back_chain = 0;
/* new return point is ret_from_fork */
frame->sf.gprs[8] = (unsigned long) ret_from_fork;
/* fake return stack for resume(), don't go back to schedule */
frame->sf.gprs[9] = (unsigned long) frame;
/* Store access registers to kernel stack of new process. */
if (unlikely(p->flags & PF_KTHREAD)) {
/* kernel thread */
memset(&frame->childregs, 0, sizeof(struct pt_regs));
frame->childregs.psw.mask = PSW_KERNEL_BITS | PSW_MASK_DAT |
PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
frame->childregs.psw.addr =
(unsigned long) kernel_thread_starter;
frame->childregs.gprs[9] = new_stackp; /* function */
frame->childregs.gprs[10] = arg;
frame->childregs.gprs[11] = (unsigned long) do_exit;
frame->childregs.orig_gpr2 = -1;
return 0;
}
frame->childregs = *current_pt_regs();
frame->childregs.gprs[2] = 0; /* child returns 0 on fork. */
frame->childregs.flags = 0;
if (new_stackp)
frame->childregs.gprs[15] = new_stackp;
/* Don't copy runtime instrumentation info */
p->thread.ri_cb = NULL;
frame->childregs.psw.mask &= ~PSW_MASK_RI;
/* Don't copy guarded storage control block */
p->thread.gs_cb = NULL;
p->thread.gs_bc_cb = NULL;
/* Set a new TLS ? */
if (clone_flags & CLONE_SETTLS) {
if (is_compat_task()) {
p->thread.acrs[0] = (unsigned int)tls;
} else {
p->thread.acrs[0] = (unsigned int)(tls >> 32);
p->thread.acrs[1] = (unsigned int)tls;
}
}
return 0;
}
asmlinkage void execve_tail(void)
{
current->thread.fpu.fpc = 0;
asm volatile("sfpc %0" : : "d" (0));
}
/*
* fill in the FPU structure for a core dump.
*/
int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
{
save_fpu_regs();
fpregs->fpc = current->thread.fpu.fpc;
fpregs->pad = 0;
if (MACHINE_HAS_VX)
convert_vx_to_fp((freg_t *)&fpregs->fprs,
current->thread.fpu.vxrs);
else
memcpy(&fpregs->fprs, current->thread.fpu.fprs,
sizeof(fpregs->fprs));
return 1;
}
EXPORT_SYMBOL(dump_fpu);
unsigned long get_wchan(struct task_struct *p)
{
struct stack_frame *sf, *low, *high;
unsigned long return_address;
int count;
if (!p || p == current || p->state == TASK_RUNNING || !task_stack_page(p))
return 0;
if (!try_get_task_stack(p))
return 0;
low = task_stack_page(p);
high = (struct stack_frame *) task_pt_regs(p);
sf = (struct stack_frame *) p->thread.ksp;
if (sf <= low || sf > high) {
return_address = 0;
goto out;
}
for (count = 0; count < 16; count++) {
sf = (struct stack_frame *) sf->back_chain;
if (sf <= low || sf > high) {
return_address = 0;
goto out;
}
return_address = sf->gprs[8];
if (!in_sched_functions(return_address))
goto out;
}
out:
put_task_stack(p);
return return_address;
}
unsigned long arch_align_stack(unsigned long sp)
{
if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
sp -= get_random_int() & ~PAGE_MASK;
return sp & ~0xf;
}
static inline unsigned long brk_rnd(void)
{
return (get_random_int() & BRK_RND_MASK) << PAGE_SHIFT;
}
unsigned long arch_randomize_brk(struct mm_struct *mm)
{
unsigned long ret;
ret = PAGE_ALIGN(mm->brk + brk_rnd());
return (ret > mm->brk) ? ret : mm->brk;
}
void set_fs_fixup(void)
{
struct pt_regs *regs = current_pt_regs();
static bool warned;
set_fs(USER_DS);
if (warned)
return;
WARN(1, "Unbalanced set_fs - int code: 0x%x\n", regs->int_code);
show_registers(regs);
warned = true;
}
+220
View File
@@ -0,0 +1,220 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#define KMSG_COMPONENT "cpu"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/cpufeature.h>
#include <linux/bitops.h>
#include <linux/kernel.h>
#include <linux/sched/mm.h>
#include <linux/init.h>
#include <linux/seq_file.h>
#include <linux/mm_types.h>
#include <linux/delay.h>
#include <linux/cpu.h>
#include <asm/diag.h>
#include <asm/facility.h>
#include <asm/elf.h>
#include <asm/lowcore.h>
#include <asm/param.h>
#include <asm/smp.h>
struct cpu_info {
unsigned int cpu_mhz_dynamic;
unsigned int cpu_mhz_static;
struct cpuid cpu_id;
};
static DEFINE_PER_CPU(struct cpu_info, cpu_info);
static bool machine_has_cpu_mhz;
void __init cpu_detect_mhz_feature(void)
{
if (test_facility(34) && __ecag(ECAG_CPU_ATTRIBUTE, 0) != -1UL)
machine_has_cpu_mhz = true;
}
static void update_cpu_mhz(void *arg)
{
unsigned long mhz;
struct cpu_info *c;
mhz = __ecag(ECAG_CPU_ATTRIBUTE, 0);
c = this_cpu_ptr(&cpu_info);
c->cpu_mhz_dynamic = mhz >> 32;
c->cpu_mhz_static = mhz & 0xffffffff;
}
void s390_update_cpu_mhz(void)
{
s390_adjust_jiffies();
if (machine_has_cpu_mhz)
on_each_cpu(update_cpu_mhz, NULL, 0);
}
void notrace cpu_relax_yield(void)
{
if (!smp_cpu_mtid && MACHINE_HAS_DIAG44) {
diag_stat_inc(DIAG_STAT_X044);
asm volatile("diag 0,0,0x44");
}
barrier();
}
EXPORT_SYMBOL(cpu_relax_yield);
/*
* cpu_init - initializes state that is per-CPU.
*/
void cpu_init(void)
{
struct cpuid *id = this_cpu_ptr(&cpu_info.cpu_id);
get_cpu_id(id);
if (machine_has_cpu_mhz)
update_cpu_mhz(NULL);
mmgrab(&init_mm);
current->active_mm = &init_mm;
BUG_ON(current->mm);
enter_lazy_tlb(&init_mm, current);
}
/*
* cpu_have_feature - Test CPU features on module initialization
*/
int cpu_have_feature(unsigned int num)
{
return elf_hwcap & (1UL << num);
}
EXPORT_SYMBOL(cpu_have_feature);
static void show_facilities(struct seq_file *m)
{
unsigned int bit;
long *facilities;
facilities = (long *)&S390_lowcore.stfle_fac_list;
seq_puts(m, "facilities :");
for_each_set_bit_inv(bit, facilities, MAX_FACILITY_BIT)
seq_printf(m, " %d", bit);
seq_putc(m, '\n');
}
static void show_cpu_summary(struct seq_file *m, void *v)
{
static const char *hwcap_str[] = {
"esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp",
"edat", "etf3eh", "highgprs", "te", "vx", "vxd", "vxe", "gs"
};
static const char * const int_hwcap_str[] = {
"sie"
};
int i, cpu;
seq_printf(m, "vendor_id : IBM/S390\n"
"# processors : %i\n"
"bogomips per cpu: %lu.%02lu\n",
num_online_cpus(), loops_per_jiffy/(500000/HZ),
(loops_per_jiffy/(5000/HZ))%100);
seq_printf(m, "max thread id : %d\n", smp_cpu_mtid);
seq_puts(m, "features\t: ");
for (i = 0; i < ARRAY_SIZE(hwcap_str); i++)
if (hwcap_str[i] && (elf_hwcap & (1UL << i)))
seq_printf(m, "%s ", hwcap_str[i]);
for (i = 0; i < ARRAY_SIZE(int_hwcap_str); i++)
if (int_hwcap_str[i] && (int_hwcap & (1UL << i)))
seq_printf(m, "%s ", int_hwcap_str[i]);
seq_puts(m, "\n");
show_facilities(m);
show_cacheinfo(m);
for_each_online_cpu(cpu) {
struct cpuid *id = &per_cpu(cpu_info.cpu_id, cpu);
seq_printf(m, "processor %d: "
"version = %02X, "
"identification = %06X, "
"machine = %04X\n",
cpu, id->version, id->ident, id->machine);
}
}
static void show_cpu_mhz(struct seq_file *m, unsigned long n)
{
struct cpu_info *c = per_cpu_ptr(&cpu_info, n);
seq_printf(m, "cpu MHz dynamic : %d\n", c->cpu_mhz_dynamic);
seq_printf(m, "cpu MHz static : %d\n", c->cpu_mhz_static);
}
/*
* show_cpuinfo - Get information on one CPU for use by procfs.
*/
static int show_cpuinfo(struct seq_file *m, void *v)
{
unsigned long n = (unsigned long) v - 1;
unsigned long first = cpumask_first(cpu_online_mask);
if (n == first)
show_cpu_summary(m, v);
if (!machine_has_cpu_mhz)
return 0;
seq_printf(m, "\ncpu number : %ld\n", n);
show_cpu_mhz(m, n);
return 0;
}
static inline void *c_update(loff_t *pos)
{
if (*pos)
*pos = cpumask_next(*pos - 1, cpu_online_mask);
else
*pos = cpumask_first(cpu_online_mask);
return *pos < nr_cpu_ids ? (void *)*pos + 1 : NULL;
}
static void *c_start(struct seq_file *m, loff_t *pos)
{
get_online_cpus();
return c_update(pos);
}
static void *c_next(struct seq_file *m, void *v, loff_t *pos)
{
++*pos;
return c_update(pos);
}
static void c_stop(struct seq_file *m, void *v)
{
put_online_cpus();
}
const struct seq_operations cpuinfo_op = {
.start = c_start,
.next = c_next,
.stop = c_stop,
.show = show_cpuinfo,
};
int s390_isolate_bp(void)
{
if (!test_facility(82))
return -EOPNOTSUPP;
set_thread_flag(TIF_ISOLATE_BP);
return 0;
}
EXPORT_SYMBOL(s390_isolate_bp);
int s390_isolate_bp_guest(void)
{
if (!test_facility(82))
return -EOPNOTSUPP;
set_thread_flag(TIF_ISOLATE_BP_GUEST);
return 0;
}
EXPORT_SYMBOL(s390_isolate_bp_guest);
File diff suppressed because it is too large. Load diff
+80
View File
@@ -0,0 +1,80 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright IBM Corp 2000, 2011
* Author(s): Holger Smolinski <Holger.Smolinski@de.ibm.com>,
* Denis Joseph Barrow,
*/
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/nospec-insn.h>
#include <asm/sigp.h>
GEN_BR_THUNK %r9
#
# Issue "store status" for the current CPU to its prefix page
# and call passed function afterwards
#
# r2 = Function to be called after store status
# r3 = Parameter for function
#
ENTRY(store_status)
/* Save register one and load save area base */
stg %r1,__LC_SAVE_AREA_RESTART
/* General purpose registers */
lghi %r1,__LC_GPREGS_SAVE_AREA
stmg %r0,%r15,0(%r1)
mvc 8(8,%r1),__LC_SAVE_AREA_RESTART
/* Control registers */
lghi %r1,__LC_CREGS_SAVE_AREA
stctg %c0,%c15,0(%r1)
/* Access registers */
lghi %r1,__LC_AREGS_SAVE_AREA
stam %a0,%a15,0(%r1)
/* Floating point registers */
lghi %r1,__LC_FPREGS_SAVE_AREA
std %f0, 0x00(%r1)
std %f1, 0x08(%r1)
std %f2, 0x10(%r1)
std %f3, 0x18(%r1)
std %f4, 0x20(%r1)
std %f5, 0x28(%r1)
std %f6, 0x30(%r1)
std %f7, 0x38(%r1)
std %f8, 0x40(%r1)
std %f9, 0x48(%r1)
std %f10,0x50(%r1)
std %f11,0x58(%r1)
std %f12,0x60(%r1)
std %f13,0x68(%r1)
std %f14,0x70(%r1)
std %f15,0x78(%r1)
/* Floating point control register */
lghi %r1,__LC_FP_CREG_SAVE_AREA
stfpc 0(%r1)
/* CPU timer */
lghi %r1,__LC_CPU_TIMER_SAVE_AREA
stpt 0(%r1)
/* Store prefix register */
lghi %r1,__LC_PREFIX_SAVE_AREA
stpx 0(%r1)
/* Clock comparator - seven bytes */
lghi %r1,__LC_CLOCK_COMP_SAVE_AREA
larl %r4,.Lclkcmp
stckc 0(%r4)
mvc 1(7,%r1),1(%r4)
/* Program status word */
lghi %r1,__LC_PSW_SAVE_AREA
epsw %r4,%r5
st %r4,0(%r1)
st %r5,4(%r1)
stg %r2,8(%r1)
lgr %r9,%r2
lgr %r2,%r3
BR_EX %r9
.section .bss
.align 8
.Lclkcmp: .quad 0x0000000000000000
.previous
+74
View File
@@ -0,0 +1,74 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright IBM Corp. 2005
*
* Author(s): Rolf Adelsberger,
* Heiko Carstens <heiko.carstens@de.ibm.com>
*
*/
#include <linux/linkage.h>
#include <asm/page.h>
#include <asm/sigp.h>
/*
* moves the new kernel to its destination...
* %r2 = pointer to first kimage_entry_t
* %r3 = start address - where to jump to after the job is done...
*
* %r5 will be used as temp. storage
* %r6 holds the destination address
* %r7 = PAGE_SIZE
* %r8 holds the source address
* %r9 = PAGE_SIZE
*
* 0xf000 is a page_mask
*/
.text
ENTRY(relocate_kernel)
basr %r13,0 # base address
.base:
lghi %r7,PAGE_SIZE # load PAGE_SIZE in r7
lghi %r9,PAGE_SIZE # load PAGE_SIZE in r9
lg %r5,0(%r2) # read another word for indirection page
aghi %r2,8 # increment pointer
tml %r5,0x1 # is it a destination page?
je .indir_check # NO, goto "indir_check"
lgr %r6,%r5 # r6 = r5
nill %r6,0xf000 # mask it out and...
j .base # ...next iteration
.indir_check:
tml %r5,0x2 # is it a indirection page?
je .done_test # NO, goto "done_test"
nill %r5,0xf000 # YES, mask out,
lgr %r2,%r5 # move it into the right register,
j .base # and read next...
.done_test:
tml %r5,0x4 # is it the done indicator?
je .source_test # NO! Well, then it should be the source indicator...
j .done # ok, lets finish it here...
.source_test:
tml %r5,0x8 # it should be a source indicator...
je .base # NO, ignore it...
lgr %r8,%r5 # r8 = r5
nill %r8,0xf000 # masking
0: mvcle %r6,%r8,0x0 # copy PAGE_SIZE bytes from r8 to r6 - pad with 0
jo 0b
j .base
.done:
sgr %r0,%r0 # clear register r0
la %r4,load_psw-.base(%r13) # load psw-address into the register
o %r3,4(%r4) # or load address into psw
st %r3,4(%r4)
mvc 0(8,%r0),0(%r4) # copy psw to absolute address 0
diag %r0,%r0,0x308
.align 8
load_psw:
.long 0x00080000,0x80000000
relocate_kernel_end:
.align 8
.globl relocate_kernel_len
relocate_kernel_len:
.quad relocate_kernel_end - relocate_kernel
+102
View File
@@ -0,0 +1,102 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2012
* Author(s): Jan Glauber <jang@linux.vnet.ibm.com>
*/
#include <linux/kernel.h>
#include <linux/syscalls.h>
#include <linux/signal.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/kernel_stat.h>
#include <linux/sched/task_stack.h>
#include <asm/runtime_instr.h>
#include <asm/cpu_mf.h>
#include <asm/irq.h>
#include "entry.h"
/* empty control block to disable RI by loading it */
struct runtime_instr_cb runtime_instr_empty_cb;
void runtime_instr_release(struct task_struct *tsk)
{
kfree(tsk->thread.ri_cb);
}
static void disable_runtime_instr(void)
{
struct task_struct *task = current;
struct pt_regs *regs;
if (!task->thread.ri_cb)
return;
regs = task_pt_regs(task);
preempt_disable();
load_runtime_instr_cb(&runtime_instr_empty_cb);
kfree(task->thread.ri_cb);
task->thread.ri_cb = NULL;
preempt_enable();
/*
* Make sure the RI bit is deleted from the PSW. If the user did not
* switch off RI before the system call the process will get a
* specification exception otherwise.
*/
regs->psw.mask &= ~PSW_MASK_RI;
}
static void init_runtime_instr_cb(struct runtime_instr_cb *cb)
{
cb->rla = 0xfff;
cb->s = 1;
cb->k = 1;
cb->ps = 1;
cb->pc = 1;
cb->key = PAGE_DEFAULT_KEY >> 4;
cb->v = 1;
}
/*
* The signum argument is unused. In older kernels it was used to
* specify a real-time signal. For backwards compatibility user space
* should pass a valid real-time signal number (the signum argument
* was checked in older kernels).
*/
SYSCALL_DEFINE2(s390_runtime_instr, int, command, int, signum)
{
struct runtime_instr_cb *cb;
if (!test_facility(64))
return -EOPNOTSUPP;
if (command == S390_RUNTIME_INSTR_STOP) {
disable_runtime_instr();
return 0;
}
if (command != S390_RUNTIME_INSTR_START)
return -EINVAL;
if (!current->thread.ri_cb) {
cb = kzalloc(sizeof(*cb), GFP_KERNEL);
if (!cb)
return -ENOMEM;
} else {
cb = current->thread.ri_cb;
memset(cb, 0, sizeof(*cb));
}
init_runtime_instr_cb(cb);
/* now load the control block to make it available */
preempt_disable();
current->thread.ri_cb = cb;
load_runtime_instr_cb(cb);
preempt_enable();
return 0;
}
+989
View File
@@ -0,0 +1,989 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 version
* Copyright IBM Corp. 1999, 2012
* Author(s): Hartmut Penner (hp@de.ibm.com),
* Martin Schwidefsky (schwidefsky@de.ibm.com)
*
* Derived from "arch/i386/kernel/setup.c"
* Copyright (C) 1995, Linus Torvalds
*/
/*
* This file handles the architecture-dependent parts of initialization
*/
#define KMSG_COMPONENT "setup"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/errno.h>
#include <linux/export.h>
#include <linux/sched.h>
#include <linux/sched/task.h>
#include <linux/cpu.h>
#include <linux/kernel.h>
#include <linux/memblock.h>
#include <linux/mm.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/ptrace.h>
#include <linux/random.h>
#include <linux/user.h>
#include <linux/tty.h>
#include <linux/ioport.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/initrd.h>
#include <linux/bootmem.h>
#include <linux/root_dev.h>
#include <linux/console.h>
#include <linux/kernel_stat.h>
#include <linux/dma-contiguous.h>
#include <linux/device.h>
#include <linux/notifier.h>
#include <linux/pfn.h>
#include <linux/ctype.h>
#include <linux/reboot.h>
#include <linux/topology.h>
#include <linux/kexec.h>
#include <linux/crash_dump.h>
#include <linux/memory.h>
#include <linux/compat.h>
#include <asm/ipl.h>
#include <asm/facility.h>
#include <asm/smp.h>
#include <asm/mmu_context.h>
#include <asm/cpcmd.h>
#include <asm/lowcore.h>
#include <asm/nmi.h>
#include <asm/irq.h>
#include <asm/page.h>
#include <asm/ptrace.h>
#include <asm/sections.h>
#include <asm/ebcdic.h>
#include <asm/diag.h>
#include <asm/os_info.h>
#include <asm/sclp.h>
#include <asm/sysinfo.h>
#include <asm/numa.h>
#include <asm/alternative.h>
#include <asm/nospec-branch.h>
#include "entry.h"
/*
* Machine setup..
*/
unsigned int console_mode = 0;
EXPORT_SYMBOL(console_mode);
unsigned int console_devno = -1;
EXPORT_SYMBOL(console_devno);
unsigned int console_irq = -1;
EXPORT_SYMBOL(console_irq);
unsigned long elf_hwcap __read_mostly = 0;
char elf_platform[ELF_PLATFORM_SIZE];
unsigned long int_hwcap = 0;
int __initdata memory_end_set;
unsigned long __initdata memory_end;
unsigned long __initdata max_physmem_end;
unsigned long VMALLOC_START;
EXPORT_SYMBOL(VMALLOC_START);
unsigned long VMALLOC_END;
EXPORT_SYMBOL(VMALLOC_END);
struct page *vmemmap;
EXPORT_SYMBOL(vmemmap);
unsigned long MODULES_VADDR;
unsigned long MODULES_END;
/* An array with a pointer to the lowcore of every CPU. */
struct lowcore *lowcore_ptr[NR_CPUS];
EXPORT_SYMBOL(lowcore_ptr);
/*
* This is set up by the setup-routine at boot-time
* for S390 need to find out, what we have to setup
* using address 0x10400 ...
*/
#include <asm/setup.h>
/*
* condev= and conmode= setup parameter.
*/
static int __init condev_setup(char *str)
{
int vdev;
vdev = simple_strtoul(str, &str, 0);
if (vdev >= 0 && vdev < 65536) {
console_devno = vdev;
console_irq = -1;
}
return 1;
}
__setup("condev=", condev_setup);
static void __init set_preferred_console(void)
{
if (CONSOLE_IS_3215 || CONSOLE_IS_SCLP)
add_preferred_console("ttyS", 0, NULL);
else if (CONSOLE_IS_3270)
add_preferred_console("tty3270", 0, NULL);
else if (CONSOLE_IS_VT220)
add_preferred_console("ttyS", 1, NULL);
else if (CONSOLE_IS_HVC)
add_preferred_console("hvc", 0, NULL);
}
static int __init conmode_setup(char *str)
{
#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
if (strncmp(str, "hwc", 4) == 0 || strncmp(str, "sclp", 5) == 0)
SET_CONSOLE_SCLP;
#endif
#if defined(CONFIG_TN3215_CONSOLE)
if (strncmp(str, "3215", 5) == 0)
SET_CONSOLE_3215;
#endif
#if defined(CONFIG_TN3270_CONSOLE)
if (strncmp(str, "3270", 5) == 0)
SET_CONSOLE_3270;
#endif
set_preferred_console();
return 1;
}
__setup("conmode=", conmode_setup);
static void __init conmode_default(void)
{
char query_buffer[1024];
char *ptr;
if (MACHINE_IS_VM) {
cpcmd("QUERY CONSOLE", query_buffer, 1024, NULL);
console_devno = simple_strtoul(query_buffer + 5, NULL, 16);
ptr = strstr(query_buffer, "SUBCHANNEL =");
console_irq = simple_strtoul(ptr + 13, NULL, 16);
cpcmd("QUERY TERM", query_buffer, 1024, NULL);
ptr = strstr(query_buffer, "CONMODE");
/*
* Set the conmode to 3215 so that the device recognition
* will set the cu_type of the console to 3215. If the
* conmode is 3270 and we don't set it back then both
* 3215 and the 3270 driver will try to access the console
* device (3215 as console and 3270 as normal tty).
*/
cpcmd("TERM CONMODE 3215", NULL, 0, NULL);
if (ptr == NULL) {
#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
SET_CONSOLE_SCLP;
#endif
return;
}
if (strncmp(ptr + 8, "3270", 4) == 0) {
#if defined(CONFIG_TN3270_CONSOLE)
SET_CONSOLE_3270;
#elif defined(CONFIG_TN3215_CONSOLE)
SET_CONSOLE_3215;
#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
SET_CONSOLE_SCLP;
#endif
} else if (strncmp(ptr + 8, "3215", 4) == 0) {
#if defined(CONFIG_TN3215_CONSOLE)
SET_CONSOLE_3215;
#elif defined(CONFIG_TN3270_CONSOLE)
SET_CONSOLE_3270;
#elif defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
SET_CONSOLE_SCLP;
#endif
}
} else if (MACHINE_IS_KVM) {
if (sclp.has_vt220 && IS_ENABLED(CONFIG_SCLP_VT220_CONSOLE))
SET_CONSOLE_VT220;
else if (sclp.has_linemode && IS_ENABLED(CONFIG_SCLP_CONSOLE))
SET_CONSOLE_SCLP;
else
SET_CONSOLE_HVC;
} else {
#if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE)
SET_CONSOLE_SCLP;
#endif
}
if (IS_ENABLED(CONFIG_VT) && IS_ENABLED(CONFIG_DUMMY_CONSOLE))
conswitchp = &dummy_con;
}
#ifdef CONFIG_CRASH_DUMP
static void __init setup_zfcpdump(void)
{
if (ipl_info.type != IPL_TYPE_FCP_DUMP)
return;
if (OLDMEM_BASE)
return;
strcat(boot_command_line, " cio_ignore=all,!ipldev,!condev");
console_loglevel = 2;
}
#else
static inline void setup_zfcpdump(void) {}
#endif /* CONFIG_CRASH_DUMP */
/*
* Reboot, halt and power_off stubs. They just call _machine_restart,
* _machine_halt or _machine_power_off.
*/
void machine_restart(char *command)
{
if ((!in_interrupt() && !in_atomic()) || oops_in_progress)
/*
* Only unblank the console if we are called in enabled
* context or a bust_spinlocks cleared the way for us.
*/
console_unblank();
_machine_restart(command);
}
void machine_halt(void)
{
if (!in_interrupt() || oops_in_progress)
/*
* Only unblank the console if we are called in enabled
* context or a bust_spinlocks cleared the way for us.
*/
console_unblank();
_machine_halt();
}
void machine_power_off(void)
{
if (!in_interrupt() || oops_in_progress)
/*
* Only unblank the console if we are called in enabled
* context or a bust_spinlocks cleared the way for us.
*/
console_unblank();
_machine_power_off();
}
/*
* Dummy power off function.
*/
void (*pm_power_off)(void) = machine_power_off;
EXPORT_SYMBOL_GPL(pm_power_off);
static int __init early_parse_mem(char *p)
{
memory_end = memparse(p, &p);
memory_end &= PAGE_MASK;
memory_end_set = 1;
return 0;
}
early_param("mem", early_parse_mem);
static int __init parse_vmalloc(char *arg)
{
if (!arg)
return -EINVAL;
VMALLOC_END = (memparse(arg, &arg) + PAGE_SIZE - 1) & PAGE_MASK;
return 0;
}
early_param("vmalloc", parse_vmalloc);
void *restart_stack __section(.data);
static void __init setup_lowcore_dat_off(void)
{
struct lowcore *lc;
/*
* Setup lowcore for boot cpu
*/
BUILD_BUG_ON(sizeof(struct lowcore) != LC_PAGES * PAGE_SIZE);
lc = memblock_virt_alloc_low(sizeof(*lc), sizeof(*lc));
lc->restart_psw.mask = PSW_KERNEL_BITS;
lc->restart_psw.addr = (unsigned long) restart_int_handler;
lc->external_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK;
lc->external_new_psw.addr = (unsigned long) ext_int_handler;
lc->svc_new_psw.mask = PSW_KERNEL_BITS |
PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK;
lc->svc_new_psw.addr = (unsigned long) system_call;
lc->program_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK;
lc->program_new_psw.addr = (unsigned long) pgm_check_handler;
lc->mcck_new_psw.mask = PSW_KERNEL_BITS;
lc->mcck_new_psw.addr = (unsigned long) mcck_int_handler;
lc->io_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_MCHECK;
lc->io_new_psw.addr = (unsigned long) io_int_handler;
lc->clock_comparator = clock_comparator_max;
lc->kernel_stack = ((unsigned long) &init_thread_union)
+ THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
lc->async_stack = (unsigned long)
memblock_virt_alloc(ASYNC_SIZE, ASYNC_SIZE)
+ ASYNC_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
lc->panic_stack = (unsigned long)
memblock_virt_alloc(PAGE_SIZE, PAGE_SIZE)
+ PAGE_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
lc->current_task = (unsigned long)&init_task;
lc->lpp = LPP_MAGIC;
lc->machine_flags = S390_lowcore.machine_flags;
lc->preempt_count = S390_lowcore.preempt_count;
lc->stfl_fac_list = S390_lowcore.stfl_fac_list;
memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,
sizeof(lc->stfle_fac_list));
memcpy(lc->alt_stfle_fac_list, S390_lowcore.alt_stfle_fac_list,
sizeof(lc->alt_stfle_fac_list));
nmi_alloc_boot_cpu(lc);
vdso_alloc_boot_cpu(lc);
lc->sync_enter_timer = S390_lowcore.sync_enter_timer;
lc->async_enter_timer = S390_lowcore.async_enter_timer;
lc->exit_timer = S390_lowcore.exit_timer;
lc->user_timer = S390_lowcore.user_timer;
lc->system_timer = S390_lowcore.system_timer;
lc->steal_timer = S390_lowcore.steal_timer;
lc->last_update_timer = S390_lowcore.last_update_timer;
lc->last_update_clock = S390_lowcore.last_update_clock;
restart_stack = memblock_virt_alloc(ASYNC_SIZE, ASYNC_SIZE);
restart_stack += ASYNC_SIZE;
/*
* Set up PSW restart to call ipl.c:do_restart(). Copy the relevant
* restart data to the absolute zero lowcore. This is necessary if
* PSW restart is done on an offline CPU that has lowcore zero.
*/
lc->restart_stack = (unsigned long) restart_stack;
lc->restart_fn = (unsigned long) do_restart;
lc->restart_data = 0;
lc->restart_source = -1UL;
/* Setup absolute zero lowcore */
mem_assign_absolute(S390_lowcore.restart_stack, lc->restart_stack);
mem_assign_absolute(S390_lowcore.restart_fn, lc->restart_fn);
mem_assign_absolute(S390_lowcore.restart_data, lc->restart_data);
mem_assign_absolute(S390_lowcore.restart_source, lc->restart_source);
mem_assign_absolute(S390_lowcore.restart_psw, lc->restart_psw);
#ifdef CONFIG_SMP
lc->spinlock_lockval = arch_spin_lockval(0);
lc->spinlock_index = 0;
arch_spin_lock_setup(0);
#endif
lc->br_r1_trampoline = 0x07f1; /* br %r1 */
set_prefix((u32)(unsigned long) lc);
lowcore_ptr[0] = lc;
}
static void __init setup_lowcore_dat_on(void)
{
__ctl_clear_bit(0, 28);
S390_lowcore.external_new_psw.mask |= PSW_MASK_DAT;
S390_lowcore.svc_new_psw.mask |= PSW_MASK_DAT;
S390_lowcore.program_new_psw.mask |= PSW_MASK_DAT;
S390_lowcore.io_new_psw.mask |= PSW_MASK_DAT;
__ctl_set_bit(0, 28);
}
static struct resource code_resource = {
.name = "Kernel code",
.flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM,
};
static struct resource data_resource = {
.name = "Kernel data",
.flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM,
};
static struct resource bss_resource = {
.name = "Kernel bss",
.flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM,
};
static struct resource __initdata *standard_resources[] = {
&code_resource,
&data_resource,
&bss_resource,
};
static void __init setup_resources(void)
{
struct resource *res, *std_res, *sub_res;
struct memblock_region *reg;
int j;
code_resource.start = (unsigned long) _text;
code_resource.end = (unsigned long) _etext - 1;
data_resource.start = (unsigned long) _etext;
data_resource.end = (unsigned long) _edata - 1;
bss_resource.start = (unsigned long) __bss_start;
bss_resource.end = (unsigned long) __bss_stop - 1;
for_each_memblock(memory, reg) {
res = memblock_virt_alloc(sizeof(*res), 8);
res->flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM;
res->name = "System RAM";
res->start = reg->base;
res->end = reg->base + reg->size - 1;
request_resource(&iomem_resource, res);
for (j = 0; j < ARRAY_SIZE(standard_resources); j++) {
std_res = standard_resources[j];
if (std_res->start < res->start ||
std_res->start > res->end)
continue;
if (std_res->end > res->end) {
sub_res = memblock_virt_alloc(sizeof(*sub_res), 8);
*sub_res = *std_res;
sub_res->end = res->end;
std_res->start = res->end + 1;
request_resource(res, sub_res);
} else {
request_resource(res, std_res);
}
}
}
#ifdef CONFIG_CRASH_DUMP
/*
* Re-add removed crash kernel memory as reserved memory. This makes
* sure it will be mapped with the identity mapping and struct pages
* will be created, so it can be resized later on.
* However add it later since the crash kernel resource should not be
* part of the System RAM resource.
*/
if (crashk_res.end) {
memblock_add_node(crashk_res.start, resource_size(&crashk_res), 0);
memblock_reserve(crashk_res.start, resource_size(&crashk_res));
insert_resource(&iomem_resource, &crashk_res);
}
#endif
}
static void __init setup_memory_end(void)
{
unsigned long vmax, vmalloc_size, tmp;
/* Choose kernel address space layout: 2, 3, or 4 levels. */
vmalloc_size = VMALLOC_END ?: (128UL << 30) - MODULES_LEN;
tmp = (memory_end ?: max_physmem_end) / PAGE_SIZE;
tmp = tmp * (sizeof(struct page) + PAGE_SIZE);
if (tmp + vmalloc_size + MODULES_LEN <= _REGION2_SIZE)
vmax = _REGION2_SIZE; /* 3-level kernel page table */
else
vmax = _REGION1_SIZE; /* 4-level kernel page table */
/* module area is at the end of the kernel address space. */
MODULES_END = vmax;
MODULES_VADDR = MODULES_END - MODULES_LEN;
VMALLOC_END = MODULES_VADDR;
VMALLOC_START = vmax - vmalloc_size;
/* Split remaining virtual space between 1:1 mapping & vmemmap array */
tmp = VMALLOC_START / (PAGE_SIZE + sizeof(struct page));
/* vmemmap contains a multiple of PAGES_PER_SECTION struct pages */
tmp = SECTION_ALIGN_UP(tmp);
tmp = VMALLOC_START - tmp * sizeof(struct page);
tmp &= ~((vmax >> 11) - 1); /* align to page table level */
tmp = min(tmp, 1UL << MAX_PHYSMEM_BITS);
vmemmap = (struct page *) tmp;
/* Take care that memory_end is set and <= vmemmap */
memory_end = min(memory_end ?: max_physmem_end, tmp);
max_pfn = max_low_pfn = PFN_DOWN(memory_end);
memblock_remove(memory_end, ULONG_MAX);
pr_notice("The maximum memory size is %luMB\n", memory_end >> 20);
}
#ifdef CONFIG_CRASH_DUMP
/*
* When kdump is enabled, we have to ensure that no memory from
* the area [0 - crashkernel memory size] and
* [crashk_res.start - crashk_res.end] is set offline.
*/
static int kdump_mem_notifier(struct notifier_block *nb,
unsigned long action, void *data)
{
struct memory_notify *arg = data;
if (action != MEM_GOING_OFFLINE)
return NOTIFY_OK;
if (arg->start_pfn < PFN_DOWN(resource_size(&crashk_res)))
return NOTIFY_BAD;
if (arg->start_pfn > PFN_DOWN(crashk_res.end))
return NOTIFY_OK;
if (arg->start_pfn + arg->nr_pages - 1 < PFN_DOWN(crashk_res.start))
return NOTIFY_OK;
return NOTIFY_BAD;
}
static struct notifier_block kdump_mem_nb = {
.notifier_call = kdump_mem_notifier,
};
#endif
/*
* Make sure that the area behind memory_end is protected
*/
static void __init reserve_memory_end(void)
{
#ifdef CONFIG_CRASH_DUMP
if (ipl_info.type == IPL_TYPE_FCP_DUMP &&
!OLDMEM_BASE && sclp.hsa_size) {
memory_end = sclp.hsa_size;
memory_end &= PAGE_MASK;
memory_end_set = 1;
}
#endif
if (!memory_end_set)
return;
memblock_reserve(memory_end, ULONG_MAX);
}
/*
* Make sure that oldmem, where the dump is stored, is protected
*/
static void __init reserve_oldmem(void)
{
#ifdef CONFIG_CRASH_DUMP
if (OLDMEM_BASE)
/* Forget all memory above the running kdump system */
memblock_reserve(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX);
#endif
}
/*
* Make sure that oldmem, where the dump is stored, is protected
*/
static void __init remove_oldmem(void)
{
#ifdef CONFIG_CRASH_DUMP
if (OLDMEM_BASE)
/* Forget all memory above the running kdump system */
memblock_remove(OLDMEM_SIZE, (phys_addr_t)ULONG_MAX);
#endif
}
/*
* Reserve memory for kdump kernel to be loaded with kexec
*/
static void __init reserve_crashkernel(void)
{
#ifdef CONFIG_CRASH_DUMP
unsigned long long crash_base, crash_size;
phys_addr_t low, high;
int rc;
rc = parse_crashkernel(boot_command_line, memory_end, &crash_size,
&crash_base);
crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN);
crash_size = ALIGN(crash_size, KEXEC_CRASH_MEM_ALIGN);
if (rc || crash_size == 0)
return;
if (memblock.memory.regions[0].size < crash_size) {
pr_info("crashkernel reservation failed: %s\n",
"first memory chunk must be at least crashkernel size");
return;
}
low = crash_base ?: OLDMEM_BASE;
high = low + crash_size;
if (low >= OLDMEM_BASE && high <= OLDMEM_BASE + OLDMEM_SIZE) {
/* The crashkernel fits into OLDMEM, reuse OLDMEM */
crash_base = low;
} else {
/* Find suitable area in free memory */
low = max_t(unsigned long, crash_size, sclp.hsa_size);
high = crash_base ? crash_base + crash_size : ULONG_MAX;
if (crash_base && crash_base < low) {
pr_info("crashkernel reservation failed: %s\n",
"crash_base too low");
return;
}
low = crash_base ?: low;
crash_base = memblock_find_in_range(low, high, crash_size,
KEXEC_CRASH_MEM_ALIGN);
}
if (!crash_base) {
pr_info("crashkernel reservation failed: %s\n",
"no suitable area found");
return;
}
if (register_memory_notifier(&kdump_mem_nb))
return;
if (!OLDMEM_BASE && MACHINE_IS_VM)
diag10_range(PFN_DOWN(crash_base), PFN_DOWN(crash_size));
crashk_res.start = crash_base;
crashk_res.end = crash_base + crash_size - 1;
memblock_remove(crash_base, crash_size);
pr_info("Reserving %lluMB of memory at %lluMB "
"for crashkernel (System RAM: %luMB)\n",
crash_size >> 20, crash_base >> 20,
(unsigned long)memblock.memory.total_size >> 20);
os_info_crashkernel_add(crash_base, crash_size);
#endif
}
/*
* Reserve the initrd from being used by memblock
*/
static void __init reserve_initrd(void)
{
#ifdef CONFIG_BLK_DEV_INITRD
if (!INITRD_START || !INITRD_SIZE)
return;
initrd_start = INITRD_START;
initrd_end = initrd_start + INITRD_SIZE;
memblock_reserve(INITRD_START, INITRD_SIZE);
#endif
}
/*
* Check for initrd being in usable memory
*/
static void __init check_initrd(void)
{
#ifdef CONFIG_BLK_DEV_INITRD
if (INITRD_START && INITRD_SIZE &&
!memblock_is_region_memory(INITRD_START, INITRD_SIZE)) {
pr_err("The initial RAM disk does not fit into the memory\n");
memblock_free(INITRD_START, INITRD_SIZE);
initrd_start = initrd_end = 0;
}
#endif
}
/*
* Reserve memory used for lowcore/command line/kernel image.
*/
static void __init reserve_kernel(void)
{
unsigned long start_pfn = PFN_UP(__pa(_end));
#ifdef CONFIG_DMA_API_DEBUG
/*
* DMA_API_DEBUG code stumbles over addresses from the
* range [PARMAREA_END, _stext]. Mark the memory as reserved
* so it is not used for CONFIG_DMA_API_DEBUG=y.
*/
memblock_reserve(0, PFN_PHYS(start_pfn));
#else
memblock_reserve(0, PARMAREA_END);
memblock_reserve((unsigned long)_stext, PFN_PHYS(start_pfn)
- (unsigned long)_stext);
#endif
}
static void __init setup_memory(void)
{
struct memblock_region *reg;
/*
* Init storage key for present memory
*/
for_each_memblock(memory, reg) {
storage_key_init_range(reg->base, reg->base + reg->size);
}
psw_set_key(PAGE_DEFAULT_KEY);
/* Only cosmetics */
memblock_enforce_memory_limit(memblock_end_of_DRAM());
}
/*
* Setup hardware capabilities.
*/
static int __init setup_hwcaps(void)
{
static const int stfl_bits[6] = { 0, 2, 7, 17, 19, 21 };
struct cpuid cpu_id;
int i;
/*
* The store facility list bits numbers as found in the principles
* of operation are numbered with bit 1UL<<31 as number 0 to
* bit 1UL<<0 as number 31.
* Bit 0: instructions named N3, "backported" to esa-mode
* Bit 2: z/Architecture mode is active
* Bit 7: the store-facility-list-extended facility is installed
* Bit 17: the message-security assist is installed
* Bit 19: the long-displacement facility is installed
* Bit 21: the extended-immediate facility is installed
* Bit 22: extended-translation facility 3 is installed
* Bit 30: extended-translation facility 3 enhancement facility
* These get translated to:
* HWCAP_S390_ESAN3 bit 0, HWCAP_S390_ZARCH bit 1,
* HWCAP_S390_STFLE bit 2, HWCAP_S390_MSA bit 3,
* HWCAP_S390_LDISP bit 4, HWCAP_S390_EIMM bit 5 and
* HWCAP_S390_ETF3EH bit 8 (22 && 30).
*/
for (i = 0; i < 6; i++)
if (test_facility(stfl_bits[i]))
elf_hwcap |= 1UL << i;
if (test_facility(22) && test_facility(30))
elf_hwcap |= HWCAP_S390_ETF3EH;
/*
* Check for additional facilities with store-facility-list-extended.
* stfle stores doublewords (8 byte) with bit 1ULL<<63 as bit 0
* and 1ULL<<0 as bit 63. Bits 0-31 contain the same information
* as stored by stfl, bits 32-xxx contain additional facilities.
* How many facility words are stored depends on the number of
* doublewords passed to the instruction. The additional facilities
* are:
* Bit 42: decimal floating point facility is installed
* Bit 44: perform floating point operation facility is installed
* translated to:
* HWCAP_S390_DFP bit 6 (42 && 44).
*/
if ((elf_hwcap & (1UL << 2)) && test_facility(42) && test_facility(44))
elf_hwcap |= HWCAP_S390_DFP;
/*
* Huge page support HWCAP_S390_HPAGE is bit 7.
*/
if (MACHINE_HAS_EDAT1)
elf_hwcap |= HWCAP_S390_HPAGE;
/*
* 64-bit register support for 31-bit processes
* HWCAP_S390_HIGH_GPRS is bit 9.
*/
elf_hwcap |= HWCAP_S390_HIGH_GPRS;
/*
* Transactional execution support HWCAP_S390_TE is bit 10.
*/
if (MACHINE_HAS_TE)
elf_hwcap |= HWCAP_S390_TE;
/*
* Vector extension HWCAP_S390_VXRS is bit 11. The Vector extension
* can be disabled with the "novx" parameter. Use MACHINE_HAS_VX
* instead of facility bit 129.
*/
if (MACHINE_HAS_VX) {
elf_hwcap |= HWCAP_S390_VXRS;
if (test_facility(134))
elf_hwcap |= HWCAP_S390_VXRS_EXT;
if (test_facility(135))
elf_hwcap |= HWCAP_S390_VXRS_BCD;
}
/*
* Guarded storage support HWCAP_S390_GS is bit 12.
*/
if (MACHINE_HAS_GS)
elf_hwcap |= HWCAP_S390_GS;
get_cpu_id(&cpu_id);
add_device_randomness(&cpu_id, sizeof(cpu_id));
switch (cpu_id.machine) {
case 0x2064:
case 0x2066:
default: /* Use "z900" as default for 64 bit kernels. */
strcpy(elf_platform, "z900");
break;
case 0x2084:
case 0x2086:
strcpy(elf_platform, "z990");
break;
case 0x2094:
case 0x2096:
strcpy(elf_platform, "z9-109");
break;
case 0x2097:
case 0x2098:
strcpy(elf_platform, "z10");
break;
case 0x2817:
case 0x2818:
strcpy(elf_platform, "z196");
break;
case 0x2827:
case 0x2828:
strcpy(elf_platform, "zEC12");
break;
case 0x2964:
case 0x2965:
strcpy(elf_platform, "z13");
break;
case 0x3906:
case 0x3907:
strcpy(elf_platform, "z14");
break;
}
/*
* Virtualization support HWCAP_INT_SIE is bit 0.
*/
if (sclp.has_sief2)
int_hwcap |= HWCAP_INT_SIE;
return 0;
}
arch_initcall(setup_hwcaps);
/*
* Add system information as device randomness
*/
static void __init setup_randomness(void)
{
struct sysinfo_3_2_2 *vmms;
vmms = (struct sysinfo_3_2_2 *) memblock_alloc(PAGE_SIZE, PAGE_SIZE);
if (stsi(vmms, 3, 2, 2) == 0 && vmms->count)
add_device_randomness(&vmms->vm, sizeof(vmms->vm[0]) * vmms->count);
memblock_free((unsigned long) vmms, PAGE_SIZE);
}
/*
* Find the correct size for the task_struct. This depends on
* the size of the struct fpu at the end of the thread_struct
* which is embedded in the task_struct.
*/
static void __init setup_task_size(void)
{
int task_size = sizeof(struct task_struct);
if (!MACHINE_HAS_VX) {
task_size -= sizeof(__vector128) * __NUM_VXRS;
task_size += sizeof(freg_t) * __NUM_FPRS;
}
arch_task_struct_size = task_size;
}
/*
* Setup function called from init/main.c just after the banner
* was printed.
*/
void __init setup_arch(char **cmdline_p)
{
/*
* print what head.S has found out about the machine
*/
if (MACHINE_IS_VM)
pr_info("Linux is running as a z/VM "
"guest operating system in 64-bit mode\n");
else if (MACHINE_IS_KVM)
pr_info("Linux is running under KVM in 64-bit mode\n");
else if (MACHINE_IS_LPAR)
pr_info("Linux is running natively in 64-bit mode\n");
else
pr_info("Linux is running as a guest in 64-bit mode\n");
/* Have one command line that is parsed and saved in /proc/cmdline */
/* boot_command_line has been already set up in early.c */
*cmdline_p = boot_command_line;
ROOT_DEV = Root_RAM0;
/* Is init_mm really needed? */
init_mm.start_code = PAGE_OFFSET;
init_mm.end_code = (unsigned long) _etext;
init_mm.end_data = (unsigned long) _edata;
init_mm.brk = (unsigned long) _end;
if (IS_ENABLED(CONFIG_EXPOLINE_AUTO))
nospec_auto_detect();
jump_label_init();
parse_early_param();
#ifdef CONFIG_CRASH_DUMP
/* Deactivate elfcorehdr= kernel parameter */
elfcorehdr_addr = ELFCORE_ADDR_MAX;
#endif
os_info_init();
setup_ipl();
setup_task_size();
/* Do some memory reservations *before* memory is added to memblock */
reserve_memory_end();
reserve_oldmem();
reserve_kernel();
reserve_initrd();
memblock_allow_resize();
/* Get information about *all* installed memory */
detect_memory_memblock();
remove_oldmem();
/*
* Make sure all chunks are MAX_ORDER aligned so we don't need the
* extra checks that HOLES_IN_ZONE would require.
*
* Is this still required?
*/
memblock_trim_memory(1UL << (MAX_ORDER - 1 + PAGE_SHIFT));
setup_memory_end();
setup_memory();
dma_contiguous_reserve(memory_end);
vmcp_cma_reserve();
check_initrd();
reserve_crashkernel();
#ifdef CONFIG_CRASH_DUMP
/*
* Be aware that smp_save_dump_cpus() triggers a system reset.
* Therefore CPU and device initialization should be done afterwards.
*/
smp_save_dump_cpus();
#endif
setup_resources();
setup_lowcore_dat_off();
smp_fill_possible_mask();
cpu_detect_mhz_feature();
cpu_init();
numa_setup();
smp_detect_cpus();
topology_init_early();
/*
* Create kernel page tables and switch to virtual addressing.
*/
paging_init();
/*
* After paging_init created the kernel page table, the new PSWs
* in lowcore can now run with DAT enabled.
*/
setup_lowcore_dat_on();
/* Setup default console */
conmode_default();
set_preferred_console();
apply_alternative_instructions();
if (IS_ENABLED(CONFIG_EXPOLINE))
nospec_init_branches();
/* Setup zfcpdump support */
setup_zfcpdump();
/* Add system specific data to the random pool */
setup_randomness();
}
+541
View File
@@ -0,0 +1,541 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 1999, 2006
* Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
*
* Based on Intel version
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* 1997-11-28 Modified for POSIX.1b signals by Richard Henderson
*/
#include <linux/sched.h>
#include <linux/sched/task_stack.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/kernel.h>
#include <linux/signal.h>
#include <linux/errno.h>
#include <linux/wait.h>
#include <linux/ptrace.h>
#include <linux/unistd.h>
#include <linux/stddef.h>
#include <linux/tty.h>
#include <linux/personality.h>
#include <linux/binfmts.h>
#include <linux/tracehook.h>
#include <linux/syscalls.h>
#include <linux/compat.h>
#include <asm/ucontext.h>
#include <linux/uaccess.h>
#include <asm/lowcore.h>
#include <asm/switch_to.h>
#include "entry.h"
/*
* Layout of an old-style signal-frame:
* -----------------------------------------
* | save area (_SIGNAL_FRAMESIZE) |
* -----------------------------------------
* | struct sigcontext |
* | oldmask |
* | _sigregs * |
* -----------------------------------------
* | _sigregs with |
* | _s390_regs_common |
* | _s390_fp_regs |
* -----------------------------------------
* | int signo |
* -----------------------------------------
* | _sigregs_ext with |
* | gprs_high 64 byte (opt) |
* | vxrs_low 128 byte (opt) |
* | vxrs_high 256 byte (opt) |
* | reserved 128 byte (opt) |
* -----------------------------------------
* | __u16 svc_insn |
* -----------------------------------------
* The svc_insn entry with the sigreturn system call opcode does not
* have a fixed position and moves if gprs_high or vxrs exist.
* Future extensions will be added to _sigregs_ext.
*/
struct sigframe
{
__u8 callee_used_stack[__SIGNAL_FRAMESIZE];
struct sigcontext sc;
_sigregs sregs;
int signo;
_sigregs_ext sregs_ext;
__u16 svc_insn; /* Offset of svc_insn is NOT fixed! */
};
/*
* Layout of an rt signal-frame:
* -----------------------------------------
* | save area (_SIGNAL_FRAMESIZE) |
* -----------------------------------------
* | svc __NR_rt_sigreturn 2 byte |
* -----------------------------------------
* | struct siginfo |
* -----------------------------------------
* | struct ucontext_extended with |
* | unsigned long uc_flags |
* | struct ucontext *uc_link |
* | stack_t uc_stack |
* | _sigregs uc_mcontext with |
* | _s390_regs_common |
* | _s390_fp_regs |
* | sigset_t uc_sigmask |
* | _sigregs_ext uc_mcontext_ext |
* | gprs_high 64 byte (opt) |
* | vxrs_low 128 byte (opt) |
* | vxrs_high 256 byte (opt)|
* | reserved 128 byte (opt) |
* -----------------------------------------
* Future extensions will be added to _sigregs_ext.
*/
struct rt_sigframe
{
__u8 callee_used_stack[__SIGNAL_FRAMESIZE];
__u16 svc_insn;
struct siginfo info;
struct ucontext_extended uc;
};
/* Store registers needed to create the signal frame */
static void store_sigregs(void)
{
save_access_regs(current->thread.acrs);
save_fpu_regs();
}
/* Load registers after signal return */
static void load_sigregs(void)
{
restore_access_regs(current->thread.acrs);
}
/* Returns non-zero on fault. */
static int save_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
{
_sigregs user_sregs;
/* Copy a 'clean' PSW mask to the user to avoid leaking
information about whether PER is currently on. */
user_sregs.regs.psw.mask = PSW_USER_BITS |
(regs->psw.mask & (PSW_MASK_USER | PSW_MASK_RI));
user_sregs.regs.psw.addr = regs->psw.addr;
memcpy(&user_sregs.regs.gprs, &regs->gprs, sizeof(sregs->regs.gprs));
memcpy(&user_sregs.regs.acrs, current->thread.acrs,
sizeof(user_sregs.regs.acrs));
fpregs_store(&user_sregs.fpregs, &current->thread.fpu);
if (__copy_to_user(sregs, &user_sregs, sizeof(_sigregs)))
return -EFAULT;
return 0;
}
static int restore_sigregs(struct pt_regs *regs, _sigregs __user *sregs)
{
_sigregs user_sregs;
/* Alwys make any pending restarted system call return -EINTR */
current->restart_block.fn = do_no_restart_syscall;
if (__copy_from_user(&user_sregs, sregs, sizeof(user_sregs)))
return -EFAULT;
if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW_MASK_RI))
return -EINVAL;
/* Test the floating-point-control word. */
if (test_fp_ctl(user_sregs.fpregs.fpc))
return -EINVAL;
/* Use regs->psw.mask instead of PSW_USER_BITS to preserve PER bit. */
regs->psw.mask = (regs->psw.mask & ~(PSW_MASK_USER | PSW_MASK_RI)) |
(user_sregs.regs.psw.mask & (PSW_MASK_USER | PSW_MASK_RI));
/* Check for invalid user address space control. */
if ((regs->psw.mask & PSW_MASK_ASC) == PSW_ASC_HOME)
regs->psw.mask = PSW_ASC_PRIMARY |
(regs->psw.mask & ~PSW_MASK_ASC);
/* Check for invalid amode */
if (regs->psw.mask & PSW_MASK_EA)
regs->psw.mask |= PSW_MASK_BA;
regs->psw.addr = user_sregs.regs.psw.addr;
memcpy(&regs->gprs, &user_sregs.regs.gprs, sizeof(sregs->regs.gprs));
memcpy(&current->thread.acrs, &user_sregs.regs.acrs,
sizeof(current->thread.acrs));
fpregs_load(&user_sregs.fpregs, &current->thread.fpu);
clear_pt_regs_flag(regs, PIF_SYSCALL); /* No longer in a system call */
return 0;
}
/* Returns non-zero on fault. */
static int save_sigregs_ext(struct pt_regs *regs,
_sigregs_ext __user *sregs_ext)
{
__u64 vxrs[__NUM_VXRS_LOW];
int i;
/* Save vector registers to signal stack */
if (MACHINE_HAS_VX) {
for (i = 0; i < __NUM_VXRS_LOW; i++)
vxrs[i] = *((__u64 *)(current->thread.fpu.vxrs + i) + 1);
if (__copy_to_user(&sregs_ext->vxrs_low, vxrs,
sizeof(sregs_ext->vxrs_low)) ||
__copy_to_user(&sregs_ext->vxrs_high,
current->thread.fpu.vxrs + __NUM_VXRS_LOW,
sizeof(sregs_ext->vxrs_high)))
return -EFAULT;
}
return 0;
}
static int restore_sigregs_ext(struct pt_regs *regs,
_sigregs_ext __user *sregs_ext)
{
__u64 vxrs[__NUM_VXRS_LOW];
int i;
/* Restore vector registers from signal stack */
if (MACHINE_HAS_VX) {
if (__copy_from_user(vxrs, &sregs_ext->vxrs_low,
sizeof(sregs_ext->vxrs_low)) ||
__copy_from_user(current->thread.fpu.vxrs + __NUM_VXRS_LOW,
&sregs_ext->vxrs_high,
sizeof(sregs_ext->vxrs_high)))
return -EFAULT;
for (i = 0; i < __NUM_VXRS_LOW; i++)
*((__u64 *)(current->thread.fpu.vxrs + i) + 1) = vxrs[i];
}
return 0;
}
SYSCALL_DEFINE0(sigreturn)
{
struct pt_regs *regs = task_pt_regs(current);
struct sigframe __user *frame =
(struct sigframe __user *) regs->gprs[15];
sigset_t set;
if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE))
goto badframe;
set_current_blocked(&set);
save_fpu_regs();
if (restore_sigregs(regs, &frame->sregs))
goto badframe;
if (restore_sigregs_ext(regs, &frame->sregs_ext))
goto badframe;
load_sigregs();
return regs->gprs[2];
badframe:
force_sig(SIGSEGV, current);
return 0;
}
SYSCALL_DEFINE0(rt_sigreturn)
{
struct pt_regs *regs = task_pt_regs(current);
struct rt_sigframe __user *frame =
(struct rt_sigframe __user *)regs->gprs[15];
sigset_t set;
if (__copy_from_user(&set.sig, &frame->uc.uc_sigmask, sizeof(set)))
goto badframe;
set_current_blocked(&set);
if (restore_altstack(&frame->uc.uc_stack))
goto badframe;
save_fpu_regs();
if (restore_sigregs(regs, &frame->uc.uc_mcontext))
goto badframe;
if (restore_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
goto badframe;
load_sigregs();
return regs->gprs[2];
badframe:
force_sig(SIGSEGV, current);
return 0;
}
/*
* Determine which stack to use..
*/
static inline void __user *
get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size)
{
unsigned long sp;
/* Default to using normal stack */
sp = regs->gprs[15];
/* Overflow on alternate signal stack gives SIGSEGV. */
if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL))
return (void __user *) -1UL;
/* This is the X/Open sanctioned signal stack switching. */
if (ka->sa.sa_flags & SA_ONSTACK) {
if (! sas_ss_flags(sp))
sp = current->sas_ss_sp + current->sas_ss_size;
}
return (void __user *)((sp - frame_size) & -8ul);
}
static int setup_frame(int sig, struct k_sigaction *ka,
sigset_t *set, struct pt_regs * regs)
{
struct sigframe __user *frame;
struct sigcontext sc;
unsigned long restorer;
size_t frame_size;
/*
* gprs_high are only present for a 31-bit task running on
* a 64-bit kernel (see compat_signal.c) but the space for
* gprs_high need to be allocated if vector registers are
* included in the signal frame on a 31-bit system.
*/
frame_size = sizeof(*frame) - sizeof(frame->sregs_ext);
if (MACHINE_HAS_VX)
frame_size += sizeof(frame->sregs_ext);
frame = get_sigframe(ka, regs, frame_size);
if (frame == (void __user *) -1UL)
return -EFAULT;
/* Set up backchain. */
if (__put_user(regs->gprs[15], (addr_t __user *) frame))
return -EFAULT;
/* Create struct sigcontext on the signal stack */
memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE);
sc.sregs = (_sigregs __user __force *) &frame->sregs;
if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc)))
return -EFAULT;
/* Store registers needed to create the signal frame */
store_sigregs();
/* Create _sigregs on the signal stack */
if (save_sigregs(regs, &frame->sregs))
return -EFAULT;
/* Place signal number on stack to allow backtrace from handler. */
if (__put_user(regs->gprs[2], (int __user *) &frame->signo))
return -EFAULT;
/* Create _sigregs_ext on the signal stack */
if (save_sigregs_ext(regs, &frame->sregs_ext))
return -EFAULT;
/* Set up to return from userspace. If provided, use a stub
already in userspace. */
if (ka->sa.sa_flags & SA_RESTORER) {
restorer = (unsigned long) ka->sa.sa_restorer;
} else {
/* Signal frame without vector registers are short ! */
__u16 __user *svc = (void __user *) frame + frame_size - 2;
if (__put_user(S390_SYSCALL_OPCODE | __NR_sigreturn, svc))
return -EFAULT;
restorer = (unsigned long) svc;
}
/* Set up registers for signal handler */
regs->gprs[14] = restorer;
regs->gprs[15] = (unsigned long) frame;
/* Force default amode and default user address space control. */
regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA |
(PSW_USER_BITS & PSW_MASK_ASC) |
(regs->psw.mask & ~PSW_MASK_ASC);
regs->psw.addr = (unsigned long) ka->sa.sa_handler;
regs->gprs[2] = sig;
regs->gprs[3] = (unsigned long) &frame->sc;
/* We forgot to include these in the sigcontext.
To avoid breaking binary compatibility, they are passed as args. */
if (sig == SIGSEGV || sig == SIGBUS || sig == SIGILL ||
sig == SIGTRAP || sig == SIGFPE) {
/* set extra registers only for synchronous signals */
regs->gprs[4] = regs->int_code & 127;
regs->gprs[5] = regs->int_parm_long;
regs->gprs[6] = current->thread.last_break;
}
return 0;
}
static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
struct pt_regs *regs)
{
struct rt_sigframe __user *frame;
unsigned long uc_flags, restorer;
size_t frame_size;
frame_size = sizeof(struct rt_sigframe) - sizeof(_sigregs_ext);
/*
* gprs_high are only present for a 31-bit task running on
* a 64-bit kernel (see compat_signal.c) but the space for
* gprs_high need to be allocated if vector registers are
* included in the signal frame on a 31-bit system.
*/
uc_flags = 0;
if (MACHINE_HAS_VX) {
frame_size += sizeof(_sigregs_ext);
uc_flags |= UC_VXRS;
}
frame = get_sigframe(&ksig->ka, regs, frame_size);
if (frame == (void __user *) -1UL)
return -EFAULT;
/* Set up backchain. */
if (__put_user(regs->gprs[15], (addr_t __user *) frame))
return -EFAULT;
/* Set up to return from userspace. If provided, use a stub
already in userspace. */
if (ksig->ka.sa.sa_flags & SA_RESTORER) {
restorer = (unsigned long) ksig->ka.sa.sa_restorer;
} else {
__u16 __user *svc = &frame->svc_insn;
if (__put_user(S390_SYSCALL_OPCODE | __NR_rt_sigreturn, svc))
return -EFAULT;
restorer = (unsigned long) svc;
}
/* Create siginfo on the signal stack */
if (copy_siginfo_to_user(&frame->info, &ksig->info))
return -EFAULT;
/* Store registers needed to create the signal frame */
store_sigregs();
/* Create ucontext on the signal stack. */
if (__put_user(uc_flags, &frame->uc.uc_flags) ||
__put_user(NULL, &frame->uc.uc_link) ||
__save_altstack(&frame->uc.uc_stack, regs->gprs[15]) ||
save_sigregs(regs, &frame->uc.uc_mcontext) ||
__copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) ||
save_sigregs_ext(regs, &frame->uc.uc_mcontext_ext))
return -EFAULT;
/* Set up registers for signal handler */
regs->gprs[14] = restorer;
regs->gprs[15] = (unsigned long) frame;
/* Force default amode and default user address space control. */
regs->psw.mask = PSW_MASK_EA | PSW_MASK_BA |
(PSW_USER_BITS & PSW_MASK_ASC) |
(regs->psw.mask & ~PSW_MASK_ASC);
regs->psw.addr = (unsigned long) ksig->ka.sa.sa_handler;
regs->gprs[2] = ksig->sig;
regs->gprs[3] = (unsigned long) &frame->info;
regs->gprs[4] = (unsigned long) &frame->uc;
regs->gprs[5] = current->thread.last_break;
return 0;
}
static void handle_signal(struct ksignal *ksig, sigset_t *oldset,
struct pt_regs *regs)
{
int ret;
/* Set up the stack frame */
if (ksig->ka.sa.sa_flags & SA_SIGINFO)
ret = setup_rt_frame(ksig, oldset, regs);
else
ret = setup_frame(ksig->sig, &ksig->ka, oldset, regs);
signal_setup_done(ret, ksig, test_thread_flag(TIF_SINGLE_STEP));
}
/*
* Note that 'init' is a special process: it doesn't get signals it doesn't
* want to handle. Thus you cannot kill init even with a SIGKILL even by
* mistake.
*
* Note that we go through the signals twice: once to check the signals that
* the kernel can handle, and then we build all the user-level signal handling
* stack-frames in one go after that.
*/
void do_signal(struct pt_regs *regs)
{
struct ksignal ksig;
sigset_t *oldset = sigmask_to_save();
/*
* Get signal to deliver. When running under ptrace, at this point
* the debugger may change all our registers, including the system
* call information.
*/
current->thread.system_call =
test_pt_regs_flag(regs, PIF_SYSCALL) ? regs->int_code : 0;
if (get_signal(&ksig)) {
/* Whee! Actually deliver the signal. */
if (current->thread.system_call) {
regs->int_code = current->thread.system_call;
/* Check for system call restarting. */
switch (regs->gprs[2]) {
case -ERESTART_RESTARTBLOCK:
case -ERESTARTNOHAND:
regs->gprs[2] = -EINTR;
break;
case -ERESTARTSYS:
if (!(ksig.ka.sa.sa_flags & SA_RESTART)) {
regs->gprs[2] = -EINTR;
break;
}
/* fallthrough */
case -ERESTARTNOINTR:
regs->gprs[2] = regs->orig_gpr2;
regs->psw.addr =
__rewind_psw(regs->psw,
regs->int_code >> 16);
break;
}
}
/* No longer in a system call */
clear_pt_regs_flag(regs, PIF_SYSCALL);
rseq_signal_deliver(&ksig, regs);
if (is_compat_task())
handle_signal32(&ksig, oldset, regs);
else
handle_signal(&ksig, oldset, regs);
return;
}
/* No handlers present - check for system call restart */
clear_pt_regs_flag(regs, PIF_SYSCALL);
if (current->thread.system_call) {
regs->int_code = current->thread.system_call;
switch (regs->gprs[2]) {
case -ERESTART_RESTARTBLOCK:
/* Restart with sys_restart_syscall */
regs->int_code = __NR_restart_syscall;
/* fallthrough */
case -ERESTARTNOHAND:
case -ERESTARTSYS:
case -ERESTARTNOINTR:
/* Restart system call with magic TIF bit. */
regs->gprs[2] = regs->orig_gpr2;
set_pt_regs_flag(regs, PIF_SYSCALL);
if (test_thread_flag(TIF_SINGLE_STEP))
clear_pt_regs_flag(regs, PIF_PER_TRAP);
break;
}
}
/*
* If there's no signal to deliver, we just put the saved sigmask back.
*/
restore_saved_sigmask();
}
void do_notify_resume(struct pt_regs *regs)
{
clear_thread_flag(TIF_NOTIFY_RESUME);
tracehook_notify_resume(regs);
rseq_handle_notify_resume(NULL, regs);
}
File diff suppressed because it is too large. Load diff
+75
View File
@@ -0,0 +1,75 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Stack trace management functions
*
* Copyright IBM Corp. 2006
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
*/
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/stacktrace.h>
#include <linux/kallsyms.h>
#include <linux/export.h>
static int __save_address(void *data, unsigned long address, int nosched)
{
struct stack_trace *trace = data;
if (nosched && in_sched_functions(address))
return 0;
if (trace->skip > 0) {
trace->skip--;
return 0;
}
if (trace->nr_entries < trace->max_entries) {
trace->entries[trace->nr_entries++] = address;
return 0;
}
return 1;
}
static int save_address(void *data, unsigned long address, int reliable)
{
return __save_address(data, address, 0);
}
static int save_address_nosched(void *data, unsigned long address, int reliable)
{
return __save_address(data, address, 1);
}
void save_stack_trace(struct stack_trace *trace)
{
unsigned long sp;
sp = current_stack_pointer();
dump_trace(save_address, trace, NULL, sp);
if (trace->nr_entries < trace->max_entries)
trace->entries[trace->nr_entries++] = ULONG_MAX;
}
EXPORT_SYMBOL_GPL(save_stack_trace);
void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
{
unsigned long sp;
sp = tsk->thread.ksp;
if (tsk == current)
sp = current_stack_pointer();
dump_trace(save_address_nosched, trace, tsk, sp);
if (trace->nr_entries < trace->max_entries)
trace->entries[trace->nr_entries++] = ULONG_MAX;
}
EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
void save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace)
{
unsigned long sp;
sp = kernel_stack_pointer(regs);
dump_trace(save_address, trace, NULL, sp);
if (trace->nr_entries < trace->max_entries)
trace->entries[trace->nr_entries++] = ULONG_MAX;
}
EXPORT_SYMBOL_GPL(save_stack_trace_regs);
+516
View File
@@ -0,0 +1,516 @@
// SPDX-License-Identifier: GPL-2.0
/*
* store hypervisor information instruction emulation functions.
*
* Copyright IBM Corp. 2016
* Author(s): Janosch Frank <frankja@linux.vnet.ibm.com>
*/
#include <linux/errno.h>
#include <linux/pagemap.h>
#include <linux/vmalloc.h>
#include <linux/syscalls.h>
#include <linux/mutex.h>
#include <asm/asm-offsets.h>
#include <asm/sclp.h>
#include <asm/diag.h>
#include <asm/sysinfo.h>
#include <asm/ebcdic.h>
#include <asm/facility.h>
#include <asm/sthyi.h>
#include "entry.h"
#define DED_WEIGHT 0xffff
/*
* CP and IFL as EBCDIC strings, SP/0x40 determines the end of string
* as they are justified with spaces.
*/
#define CP 0xc3d7404040404040UL
#define IFL 0xc9c6d34040404040UL
enum hdr_flags {
HDR_NOT_LPAR = 0x10,
HDR_STACK_INCM = 0x20,
HDR_STSI_UNAV = 0x40,
HDR_PERF_UNAV = 0x80,
};
enum mac_validity {
MAC_NAME_VLD = 0x20,
MAC_ID_VLD = 0x40,
MAC_CNT_VLD = 0x80,
};
enum par_flag {
PAR_MT_EN = 0x80,
};
enum par_validity {
PAR_GRP_VLD = 0x08,
PAR_ID_VLD = 0x10,
PAR_ABS_VLD = 0x20,
PAR_WGHT_VLD = 0x40,
PAR_PCNT_VLD = 0x80,
};
struct hdr_sctn {
u8 infhflg1;
u8 infhflg2; /* reserved */
u8 infhval1; /* reserved */
u8 infhval2; /* reserved */
u8 reserved[3];
u8 infhygct;
u16 infhtotl;
u16 infhdln;
u16 infmoff;
u16 infmlen;
u16 infpoff;
u16 infplen;
u16 infhoff1;
u16 infhlen1;
u16 infgoff1;
u16 infglen1;
u16 infhoff2;
u16 infhlen2;
u16 infgoff2;
u16 infglen2;
u16 infhoff3;
u16 infhlen3;
u16 infgoff3;
u16 infglen3;
u8 reserved2[4];
} __packed;
struct mac_sctn {
u8 infmflg1; /* reserved */
u8 infmflg2; /* reserved */
u8 infmval1;
u8 infmval2; /* reserved */
u16 infmscps;
u16 infmdcps;
u16 infmsifl;
u16 infmdifl;
char infmname[8];
char infmtype[4];
char infmmanu[16];
char infmseq[16];
char infmpman[4];
u8 reserved[4];
} __packed;
struct par_sctn {
u8 infpflg1;
u8 infpflg2; /* reserved */
u8 infpval1;
u8 infpval2; /* reserved */
u16 infppnum;
u16 infpscps;
u16 infpdcps;
u16 infpsifl;
u16 infpdifl;
u16 reserved;
char infppnam[8];
u32 infpwbcp;
u32 infpabcp;
u32 infpwbif;
u32 infpabif;
char infplgnm[8];
u32 infplgcp;
u32 infplgif;
} __packed;
struct sthyi_sctns {
struct hdr_sctn hdr;
struct mac_sctn mac;
struct par_sctn par;
} __packed;
struct cpu_inf {
u64 lpar_cap;
u64 lpar_grp_cap;
u64 lpar_weight;
u64 all_weight;
int cpu_num_ded;
int cpu_num_shd;
};
struct lpar_cpu_inf {
struct cpu_inf cp;
struct cpu_inf ifl;
};
/*
* STHYI requires extensive locking in the higher hypervisors
* and is very computational/memory expensive. Therefore we
* cache the retrieved data whose valid period is 1s.
*/
#define CACHE_VALID_JIFFIES HZ
struct sthyi_info {
void *info;
unsigned long end;
};
static DEFINE_MUTEX(sthyi_mutex);
static struct sthyi_info sthyi_cache;
static inline u64 cpu_id(u8 ctidx, void *diag224_buf)
{
return *((u64 *)(diag224_buf + (ctidx + 1) * DIAG204_CPU_NAME_LEN));
}
/*
* Scales the cpu capping from the lpar range to the one expected in
* sthyi data.
*
* diag204 reports a cap in hundredths of processor units.
* z/VM's range for one core is 0 - 0x10000.
*/
static u32 scale_cap(u32 in)
{
return (0x10000 * in) / 100;
}
static void fill_hdr(struct sthyi_sctns *sctns)
{
sctns->hdr.infhdln = sizeof(sctns->hdr);
sctns->hdr.infmoff = sizeof(sctns->hdr);
sctns->hdr.infmlen = sizeof(sctns->mac);
sctns->hdr.infplen = sizeof(sctns->par);
sctns->hdr.infpoff = sctns->hdr.infhdln + sctns->hdr.infmlen;
sctns->hdr.infhtotl = sctns->hdr.infpoff + sctns->hdr.infplen;
}
static void fill_stsi_mac(struct sthyi_sctns *sctns,
struct sysinfo_1_1_1 *sysinfo)
{
sclp_ocf_cpc_name_copy(sctns->mac.infmname);
if (*(u64 *)sctns->mac.infmname != 0)
sctns->mac.infmval1 |= MAC_NAME_VLD;
if (stsi(sysinfo, 1, 1, 1))
return;
memcpy(sctns->mac.infmtype, sysinfo->type, sizeof(sctns->mac.infmtype));
memcpy(sctns->mac.infmmanu, sysinfo->manufacturer, sizeof(sctns->mac.infmmanu));
memcpy(sctns->mac.infmpman, sysinfo->plant, sizeof(sctns->mac.infmpman));
memcpy(sctns->mac.infmseq, sysinfo->sequence, sizeof(sctns->mac.infmseq));
sctns->mac.infmval1 |= MAC_ID_VLD;
}
static void fill_stsi_par(struct sthyi_sctns *sctns,
struct sysinfo_2_2_2 *sysinfo)
{
if (stsi(sysinfo, 2, 2, 2))
return;
sctns->par.infppnum = sysinfo->lpar_number;
memcpy(sctns->par.infppnam, sysinfo->name, sizeof(sctns->par.infppnam));
sctns->par.infpval1 |= PAR_ID_VLD;
}
static void fill_stsi(struct sthyi_sctns *sctns)
{
void *sysinfo;
/* Errors are handled through the validity bits in the response. */
sysinfo = (void *)__get_free_page(GFP_KERNEL);
if (!sysinfo)
return;
fill_stsi_mac(sctns, sysinfo);
fill_stsi_par(sctns, sysinfo);
free_pages((unsigned long)sysinfo, 0);
}
static void fill_diag_mac(struct sthyi_sctns *sctns,
struct diag204_x_phys_block *block,
void *diag224_buf)
{
int i;
for (i = 0; i < block->hdr.cpus; i++) {
switch (cpu_id(block->cpus[i].ctidx, diag224_buf)) {
case CP:
if (block->cpus[i].weight == DED_WEIGHT)
sctns->mac.infmdcps++;
else
sctns->mac.infmscps++;
break;
case IFL:
if (block->cpus[i].weight == DED_WEIGHT)
sctns->mac.infmdifl++;
else
sctns->mac.infmsifl++;
break;
}
}
sctns->mac.infmval1 |= MAC_CNT_VLD;
}
/* Returns a pointer to the the next partition block. */
static struct diag204_x_part_block *lpar_cpu_inf(struct lpar_cpu_inf *part_inf,
bool this_lpar,
void *diag224_buf,
struct diag204_x_part_block *block)
{
int i, capped = 0, weight_cp = 0, weight_ifl = 0;
struct cpu_inf *cpu_inf;
for (i = 0; i < block->hdr.rcpus; i++) {
if (!(block->cpus[i].cflag & DIAG204_CPU_ONLINE))
continue;
switch (cpu_id(block->cpus[i].ctidx, diag224_buf)) {
case CP:
cpu_inf = &part_inf->cp;
if (block->cpus[i].cur_weight < DED_WEIGHT)
weight_cp |= block->cpus[i].cur_weight;
break;
case IFL:
cpu_inf = &part_inf->ifl;
if (block->cpus[i].cur_weight < DED_WEIGHT)
weight_ifl |= block->cpus[i].cur_weight;
break;
default:
continue;
}
if (!this_lpar)
continue;
capped |= block->cpus[i].cflag & DIAG204_CPU_CAPPED;
cpu_inf->lpar_cap |= block->cpus[i].cpu_type_cap;
cpu_inf->lpar_grp_cap |= block->cpus[i].group_cpu_type_cap;
if (block->cpus[i].weight == DED_WEIGHT)
cpu_inf->cpu_num_ded += 1;
else
cpu_inf->cpu_num_shd += 1;
}
if (this_lpar && capped) {
part_inf->cp.lpar_weight = weight_cp;
part_inf->ifl.lpar_weight = weight_ifl;
}
part_inf->cp.all_weight += weight_cp;
part_inf->ifl.all_weight += weight_ifl;
return (struct diag204_x_part_block *)&block->cpus[i];
}
static void fill_diag(struct sthyi_sctns *sctns)
{
int i, r, pages;
bool this_lpar;
void *diag204_buf;
void *diag224_buf = NULL;
struct diag204_x_info_blk_hdr *ti_hdr;
struct diag204_x_part_block *part_block;
struct diag204_x_phys_block *phys_block;
struct lpar_cpu_inf lpar_inf = {};
/* Errors are handled through the validity bits in the response. */
pages = diag204((unsigned long)DIAG204_SUBC_RSI |
(unsigned long)DIAG204_INFO_EXT, 0, NULL);
if (pages <= 0)
return;
diag204_buf = vmalloc(array_size(pages, PAGE_SIZE));
if (!diag204_buf)
return;
r = diag204((unsigned long)DIAG204_SUBC_STIB7 |
(unsigned long)DIAG204_INFO_EXT, pages, diag204_buf);
if (r < 0)
goto out;
diag224_buf = (void *)__get_free_page(GFP_KERNEL | GFP_DMA);
if (!diag224_buf || diag224(diag224_buf))
goto out;
ti_hdr = diag204_buf;
part_block = diag204_buf + sizeof(*ti_hdr);
for (i = 0; i < ti_hdr->npar; i++) {
/*
* For the calling lpar we also need to get the cpu
* caps and weights. The time information block header
* specifies the offset to the partition block of the
* caller lpar, so we know when we process its data.
*/
this_lpar = (void *)part_block - diag204_buf == ti_hdr->this_part;
part_block = lpar_cpu_inf(&lpar_inf, this_lpar, diag224_buf,
part_block);
}
phys_block = (struct diag204_x_phys_block *)part_block;
part_block = diag204_buf + ti_hdr->this_part;
if (part_block->hdr.mtid)
sctns->par.infpflg1 = PAR_MT_EN;
sctns->par.infpval1 |= PAR_GRP_VLD;
sctns->par.infplgcp = scale_cap(lpar_inf.cp.lpar_grp_cap);
sctns->par.infplgif = scale_cap(lpar_inf.ifl.lpar_grp_cap);
memcpy(sctns->par.infplgnm, part_block->hdr.hardware_group_name,
sizeof(sctns->par.infplgnm));
sctns->par.infpscps = lpar_inf.cp.cpu_num_shd;
sctns->par.infpdcps = lpar_inf.cp.cpu_num_ded;
sctns->par.infpsifl = lpar_inf.ifl.cpu_num_shd;
sctns->par.infpdifl = lpar_inf.ifl.cpu_num_ded;
sctns->par.infpval1 |= PAR_PCNT_VLD;
sctns->par.infpabcp = scale_cap(lpar_inf.cp.lpar_cap);
sctns->par.infpabif = scale_cap(lpar_inf.ifl.lpar_cap);
sctns->par.infpval1 |= PAR_ABS_VLD;
/*
* Everything below needs global performance data to be
* meaningful.
*/
if (!(ti_hdr->flags & DIAG204_LPAR_PHYS_FLG)) {
sctns->hdr.infhflg1 |= HDR_PERF_UNAV;
goto out;
}
fill_diag_mac(sctns, phys_block, diag224_buf);
if (lpar_inf.cp.lpar_weight) {
sctns->par.infpwbcp = sctns->mac.infmscps * 0x10000 *
lpar_inf.cp.lpar_weight / lpar_inf.cp.all_weight;
}
if (lpar_inf.ifl.lpar_weight) {
sctns->par.infpwbif = sctns->mac.infmsifl * 0x10000 *
lpar_inf.ifl.lpar_weight / lpar_inf.ifl.all_weight;
}
sctns->par.infpval1 |= PAR_WGHT_VLD;
out:
free_page((unsigned long)diag224_buf);
vfree(diag204_buf);
}
static int sthyi(u64 vaddr, u64 *rc)
{
register u64 code asm("0") = 0;
register u64 addr asm("2") = vaddr;
register u64 rcode asm("3");
int cc;
asm volatile(
".insn rre,0xB2560000,%[code],%[addr]\n"
"ipm %[cc]\n"
"srl %[cc],28\n"
: [cc] "=d" (cc), "=d" (rcode)
: [code] "d" (code), [addr] "a" (addr)
: "memory", "cc");
*rc = rcode;
return cc;
}
static int fill_dst(void *dst, u64 *rc)
{
struct sthyi_sctns *sctns = (struct sthyi_sctns *)dst;
/*
* If the facility is on, we don't want to emulate the instruction.
* We ask the hypervisor to provide the data.
*/
if (test_facility(74))
return sthyi((u64)dst, rc);
fill_hdr(sctns);
fill_stsi(sctns);
fill_diag(sctns);
*rc = 0;
return 0;
}
static int sthyi_init_cache(void)
{
if (sthyi_cache.info)
return 0;
sthyi_cache.info = (void *)get_zeroed_page(GFP_KERNEL);
if (!sthyi_cache.info)
return -ENOMEM;
sthyi_cache.end = jiffies - 1; /* expired */
return 0;
}
static int sthyi_update_cache(u64 *rc)
{
int r;
memset(sthyi_cache.info, 0, PAGE_SIZE);
r = fill_dst(sthyi_cache.info, rc);
if (r)
return r;
sthyi_cache.end = jiffies + CACHE_VALID_JIFFIES;
return r;
}
/*
* sthyi_fill - Fill page with data returned by the STHYI instruction
*
* @dst: Pointer to zeroed page
* @rc: Pointer for storing the return code of the instruction
*
* Fills the destination with system information returned by the STHYI
* instruction. The data is generated by emulation or execution of STHYI,
* if available. The return value is the condition code that would be
* returned, the rc parameter is the return code which is passed in
* register R2 + 1.
*/
int sthyi_fill(void *dst, u64 *rc)
{
int r;
mutex_lock(&sthyi_mutex);
r = sthyi_init_cache();
if (r)
goto out;
if (time_is_before_jiffies(sthyi_cache.end)) {
/* cache expired */
r = sthyi_update_cache(rc);
if (r)
goto out;
}
*rc = 0;
memcpy(dst, sthyi_cache.info, PAGE_SIZE);
out:
mutex_unlock(&sthyi_mutex);
return r;
}
EXPORT_SYMBOL_GPL(sthyi_fill);
SYSCALL_DEFINE4(s390_sthyi, unsigned long, function_code, void __user *, buffer,
u64 __user *, return_code, unsigned long, flags)
{
u64 sthyi_rc;
void *info;
int r;
if (flags)
return -EINVAL;
if (function_code != STHYI_FC_CP_IFL_CAP)
return -EOPNOTSUPP;
info = (void *)get_zeroed_page(GFP_KERNEL);
if (!info)
return -ENOMEM;
r = sthyi_fill(info, &sthyi_rc);
if (r < 0)
goto out;
if (return_code && put_user(sthyi_rc, return_code)) {
r = -EFAULT;
goto out;
}
if (copy_to_user(buffer, info, PAGE_SIZE))
r = -EFAULT;
out:
free_page((unsigned long)info);
return r;
}
+240
View File
@@ -0,0 +1,240 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Suspend support specific for s390.
*
* Copyright IBM Corp. 2009
*
* Author(s): Hans-Joachim Picht <hans@linux.vnet.ibm.com>
*/
#include <linux/pfn.h>
#include <linux/suspend.h>
#include <linux/mm.h>
#include <linux/pci.h>
#include <asm/ctl_reg.h>
#include <asm/ipl.h>
#include <asm/cio.h>
#include <asm/sections.h>
#include "entry.h"
/*
* The restore of the saved pages in an hibernation image will set
* the change and referenced bits in the storage key for each page.
* Overindication of the referenced bits after an hibernation cycle
* does not cause any harm but the overindication of the change bits
* would cause trouble.
* Use the ARCH_SAVE_PAGE_KEYS hooks to save the storage key of each
* page to the most significant byte of the associated page frame
* number in the hibernation image.
*/
/*
* Key storage is allocated as a linked list of pages.
* The size of the keys array is (PAGE_SIZE - sizeof(long))
*/
struct page_key_data {
struct page_key_data *next;
unsigned char data[];
};
#define PAGE_KEY_DATA_SIZE (PAGE_SIZE - sizeof(struct page_key_data *))
static struct page_key_data *page_key_data;
static struct page_key_data *page_key_rp, *page_key_wp;
static unsigned long page_key_rx, page_key_wx;
unsigned long suspend_zero_pages;
/*
* For each page in the hibernation image one additional byte is
* stored in the most significant byte of the page frame number.
* On suspend no additional memory is required but on resume the
* keys need to be memorized until the page data has been restored.
* Only then can the storage keys be set to their old state.
*/
unsigned long page_key_additional_pages(unsigned long pages)
{
return DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE);
}
/*
* Free page_key_data list of arrays.
*/
void page_key_free(void)
{
struct page_key_data *pkd;
while (page_key_data) {
pkd = page_key_data;
page_key_data = pkd->next;
free_page((unsigned long) pkd);
}
}
/*
* Allocate page_key_data list of arrays with enough room to store
* one byte for each page in the hibernation image.
*/
int page_key_alloc(unsigned long pages)
{
struct page_key_data *pk;
unsigned long size;
size = DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE);
while (size--) {
pk = (struct page_key_data *) get_zeroed_page(GFP_KERNEL);
if (!pk) {
page_key_free();
return -ENOMEM;
}
pk->next = page_key_data;
page_key_data = pk;
}
page_key_rp = page_key_wp = page_key_data;
page_key_rx = page_key_wx = 0;
return 0;
}
/*
* Save the storage key into the upper 8 bits of the page frame number.
*/
void page_key_read(unsigned long *pfn)
{
struct page *page;
unsigned long addr;
unsigned char key;
page = pfn_to_page(*pfn);
addr = (unsigned long) page_address(page);
key = (unsigned char) page_get_storage_key(addr) & 0x7f;
if (arch_test_page_nodat(page))
key |= 0x80;
*(unsigned char *) pfn = key;
}
/*
* Extract the storage key from the upper 8 bits of the page frame number
* and store it in the page_key_data list of arrays.
*/
void page_key_memorize(unsigned long *pfn)
{
page_key_wp->data[page_key_wx] = *(unsigned char *) pfn;
*(unsigned char *) pfn = 0;
if (++page_key_wx < PAGE_KEY_DATA_SIZE)
return;
page_key_wp = page_key_wp->next;
page_key_wx = 0;
}
/*
* Get the next key from the page_key_data list of arrays and set the
* storage key of the page referred by @address. If @address refers to
* a "safe" page the swsusp_arch_resume code will transfer the storage
* key from the buffer page to the original page.
*/
void page_key_write(void *address)
{
struct page *page;
unsigned char key;
key = page_key_rp->data[page_key_rx];
page_set_storage_key((unsigned long) address, key & 0x7f, 0);
page = virt_to_page(address);
if (key & 0x80)
arch_set_page_nodat(page, 0);
else
arch_set_page_dat(page, 0);
if (++page_key_rx >= PAGE_KEY_DATA_SIZE)
return;
page_key_rp = page_key_rp->next;
page_key_rx = 0;
}
int pfn_is_nosave(unsigned long pfn)
{
unsigned long nosave_begin_pfn = PFN_DOWN(__pa(&__nosave_begin));
unsigned long nosave_end_pfn = PFN_DOWN(__pa(&__nosave_end));
unsigned long end_rodata_pfn = PFN_DOWN(__pa(__end_rodata)) - 1;
unsigned long stext_pfn = PFN_DOWN(__pa(_stext));
/* Always save lowcore pages (LC protection might be enabled). */
if (pfn <= LC_PAGES)
return 0;
if (pfn >= nosave_begin_pfn && pfn < nosave_end_pfn)
return 1;
/* Skip memory holes and read-only pages (DCSS, ...). */
if (pfn >= stext_pfn && pfn <= end_rodata_pfn)
return 0;
if (tprot(PFN_PHYS(pfn)))
return 1;
return 0;
}
/*
* PM notifier callback for suspend
*/
static int suspend_pm_cb(struct notifier_block *nb, unsigned long action,
void *ptr)
{
switch (action) {
case PM_SUSPEND_PREPARE:
case PM_HIBERNATION_PREPARE:
suspend_zero_pages = __get_free_pages(GFP_KERNEL, LC_ORDER);
if (!suspend_zero_pages)
return NOTIFY_BAD;
break;
case PM_POST_SUSPEND:
case PM_POST_HIBERNATION:
free_pages(suspend_zero_pages, LC_ORDER);
break;
default:
return NOTIFY_DONE;
}
return NOTIFY_OK;
}
static int __init suspend_pm_init(void)
{
pm_notifier(suspend_pm_cb, 0);
return 0;
}
arch_initcall(suspend_pm_init);
void save_processor_state(void)
{
/* swsusp_arch_suspend() actually saves all cpu register contents.
* Machine checks must be disabled since swsusp_arch_suspend() stores
* register contents to their lowcore save areas. That's the same
* place where register contents on machine checks would be saved.
* To avoid register corruption disable machine checks.
* We must also disable machine checks in the new psw mask for
* program checks, since swsusp_arch_suspend() may generate program
* checks. Disabling machine checks for all other new psw masks is
* just paranoia.
*/
local_mcck_disable();
/* Disable lowcore protection */
__ctl_clear_bit(0,28);
S390_lowcore.external_new_psw.mask &= ~PSW_MASK_MCHECK;
S390_lowcore.svc_new_psw.mask &= ~PSW_MASK_MCHECK;
S390_lowcore.io_new_psw.mask &= ~PSW_MASK_MCHECK;
S390_lowcore.program_new_psw.mask &= ~PSW_MASK_MCHECK;
}
void restore_processor_state(void)
{
S390_lowcore.external_new_psw.mask |= PSW_MASK_MCHECK;
S390_lowcore.svc_new_psw.mask |= PSW_MASK_MCHECK;
S390_lowcore.io_new_psw.mask |= PSW_MASK_MCHECK;
S390_lowcore.program_new_psw.mask |= PSW_MASK_MCHECK;
/* Enable lowcore protection */
__ctl_set_bit(0,28);
local_mcck_enable();
}
/* Called at the end of swsusp_arch_resume */
void s390_early_resume(void)
{
lgr_info_log();
channel_subsystem_reinit();
zpci_rescan();
}
+286
View File
@@ -0,0 +1,286 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* S390 64-bit swsusp implementation
*
* Copyright IBM Corp. 2009
*
* Author(s): Hans-Joachim Picht <hans@linux.vnet.ibm.com>
* Michael Holzheu <holzheu@linux.vnet.ibm.com>
*/
#include <linux/linkage.h>
#include <asm/page.h>
#include <asm/ptrace.h>
#include <asm/thread_info.h>
#include <asm/asm-offsets.h>
#include <asm/nospec-insn.h>
#include <asm/sigp.h>
/*
* Save register context in absolute 0 lowcore and call swsusp_save() to
* create in-memory kernel image. The context is saved in the designated
* "store status" memory locations (see POP).
* We return from this function twice. The first time during the suspend to
* disk process. The second time via the swsusp_arch_resume() function
* (see below) in the resume process.
* This function runs with disabled interrupts.
*/
GEN_BR_THUNK %r14
.section .text
ENTRY(swsusp_arch_suspend)
stmg %r6,%r15,__SF_GPRS(%r15)
lgr %r1,%r15
aghi %r15,-STACK_FRAME_OVERHEAD
stg %r1,__SF_BACKCHAIN(%r15)
/* Store FPU registers */
brasl %r14,save_fpu_regs
/* Deactivate DAT */
stnsm __SF_EMPTY(%r15),0xfb
/* Store prefix register on stack */
stpx __SF_EMPTY(%r15)
/* Save prefix register contents for lowcore copy */
llgf %r10,__SF_EMPTY(%r15)
/* Get pointer to save area */
lghi %r1,0x1000
/* Save CPU address */
stap __LC_EXT_CPU_ADDR(%r0)
/* Store registers */
mvc 0x318(4,%r1),__SF_EMPTY(%r15) /* move prefix to lowcore */
stam %a0,%a15,0x340(%r1) /* store access registers */
stctg %c0,%c15,0x380(%r1) /* store control registers */
stmg %r0,%r15,0x280(%r1) /* store general registers */
stpt 0x328(%r1) /* store timer */
stck __SF_EMPTY(%r15) /* store clock */
stckc 0x330(%r1) /* store clock comparator */
/* Update cputime accounting before going to sleep */
lg %r0,__LC_LAST_UPDATE_TIMER
slg %r0,0x328(%r1)
alg %r0,__LC_SYSTEM_TIMER
stg %r0,__LC_SYSTEM_TIMER
mvc __LC_LAST_UPDATE_TIMER(8),0x328(%r1)
lg %r0,__LC_LAST_UPDATE_CLOCK
slg %r0,__SF_EMPTY(%r15)
alg %r0,__LC_STEAL_TIMER
stg %r0,__LC_STEAL_TIMER
mvc __LC_LAST_UPDATE_CLOCK(8),__SF_EMPTY(%r15)
/* Activate DAT */
stosm __SF_EMPTY(%r15),0x04
/* Set prefix page to zero */
xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15)
spx __SF_EMPTY(%r15)
/* Save absolute zero pages */
larl %r2,suspend_zero_pages
lg %r2,0(%r2)
lghi %r4,0
lghi %r3,2*PAGE_SIZE
lghi %r5,2*PAGE_SIZE
1: mvcle %r2,%r4,0
jo 1b
/* Copy lowcore to absolute zero lowcore */
lghi %r2,0
lgr %r4,%r10
lghi %r3,2*PAGE_SIZE
lghi %r5,2*PAGE_SIZE
1: mvcle %r2,%r4,0
jo 1b
/* Save image */
brasl %r14,swsusp_save
/* Restore prefix register and return */
lghi %r1,0x1000
spx 0x318(%r1)
lmg %r6,%r15,STACK_FRAME_OVERHEAD + __SF_GPRS(%r15)
lghi %r2,0
BR_EX %r14
/*
* Restore saved memory image to correct place and restore register context.
* Then we return to the function that called swsusp_arch_suspend().
* swsusp_arch_resume() runs with disabled interrupts.
*/
ENTRY(swsusp_arch_resume)
stmg %r6,%r15,__SF_GPRS(%r15)
lgr %r1,%r15
aghi %r15,-STACK_FRAME_OVERHEAD
stg %r1,__SF_BACKCHAIN(%r15)
/* Make all free pages stable */
lghi %r2,1
brasl %r14,arch_set_page_states
/* Deactivate DAT */
stnsm __SF_EMPTY(%r15),0xfb
/* Set prefix page to zero */
xc __SF_EMPTY(4,%r15),__SF_EMPTY(%r15)
spx __SF_EMPTY(%r15)
/* Restore saved image */
larl %r1,restore_pblist
lg %r1,0(%r1)
ltgr %r1,%r1
jz 2f
0:
lg %r2,8(%r1)
lg %r4,0(%r1)
iske %r0,%r4
lghi %r3,PAGE_SIZE
lghi %r5,PAGE_SIZE
1:
mvcle %r2,%r4,0
jo 1b
lg %r2,8(%r1)
sske %r0,%r2
lg %r1,16(%r1)
ltgr %r1,%r1
jnz 0b
2:
ptlb /* flush tlb */
/* Reset System */
larl %r1,restart_entry
larl %r2,.Lrestart_diag308_psw
og %r1,0(%r2)
stg %r1,0(%r0)
larl %r1,.Lnew_pgm_check_psw
epsw %r2,%r3
stm %r2,%r3,0(%r1)
mvc __LC_PGM_NEW_PSW(16,%r0),0(%r1)
lghi %r0,0
diag %r0,%r0,0x308
restart_entry:
lhi %r1,1
sigp %r1,%r0,SIGP_SET_ARCHITECTURE
sam64
#ifdef CONFIG_SMP
larl %r1,smp_cpu_mt_shift
icm %r1,15,0(%r1)
jz smt_done
llgfr %r1,%r1
smt_loop:
sigp %r1,%r0,SIGP_SET_MULTI_THREADING
brc 8,smt_done /* accepted */
brc 2,smt_loop /* busy, try again */
smt_done:
#endif
larl %r1,.Lnew_pgm_check_psw
lpswe 0(%r1)
pgm_check_entry:
/* Switch to original suspend CPU */
larl %r1,.Lresume_cpu /* Resume CPU address: r2 */
stap 0(%r1)
llgh %r2,0(%r1)
llgh %r1,__LC_EXT_CPU_ADDR(%r0) /* Suspend CPU address: r1 */
cgr %r1,%r2
je restore_registers /* r1 = r2 -> nothing to do */
larl %r4,.Lrestart_suspend_psw /* Set new restart PSW */
mvc __LC_RST_NEW_PSW(16,%r0),0(%r4)
3:
sigp %r9,%r1,SIGP_INITIAL_CPU_RESET /* sigp initial cpu reset */
brc 8,4f /* accepted */
brc 2,3b /* busy, try again */
/* Suspend CPU not available -> panic */
larl %r15,init_thread_union
aghi %r15,1<<(PAGE_SHIFT+THREAD_SIZE_ORDER)
aghi %r15,-STACK_FRAME_OVERHEAD
larl %r2,.Lpanic_string
brasl %r14,sclp_early_printk_force
larl %r3,.Ldisabled_wait_31
lpsw 0(%r3)
4:
/* Switch to suspend CPU */
sigp %r9,%r1,SIGP_RESTART /* sigp restart to suspend CPU */
brc 2,4b /* busy, try again */
5:
sigp %r9,%r2,SIGP_STOP /* sigp stop to current resume CPU */
brc 2,5b /* busy, try again */
6: j 6b
restart_suspend:
larl %r1,.Lresume_cpu
llgh %r2,0(%r1)
7:
sigp %r9,%r2,SIGP_SENSE /* sigp sense, wait for resume CPU */
brc 8,7b /* accepted, status 0, still running */
brc 2,7b /* busy, try again */
tmll %r9,0x40 /* Test if resume CPU is stopped */
jz 7b
restore_registers:
/* Restore registers */
lghi %r13,0x1000 /* %r1 = pointer to save area */
/* Ignore time spent in suspended state. */
llgf %r1,0x318(%r13)
stck __LC_LAST_UPDATE_CLOCK(%r1)
spt 0x328(%r13) /* reprogram timer */
//sckc 0x330(%r13) /* set clock comparator */
lctlg %c0,%c15,0x380(%r13) /* load control registers */
lam %a0,%a15,0x340(%r13) /* load access registers */
/* Load old stack */
lg %r15,0x2f8(%r13)
/* Save prefix register */
mvc __SF_EMPTY(4,%r15),0x318(%r13)
/* Restore absolute zero pages */
lghi %r2,0
larl %r4,suspend_zero_pages
lg %r4,0(%r4)
lghi %r3,2*PAGE_SIZE
lghi %r5,2*PAGE_SIZE
1: mvcle %r2,%r4,0
jo 1b
/* Restore prefix register */
spx __SF_EMPTY(%r15)
/* Activate DAT */
stosm __SF_EMPTY(%r15),0x04
/* Make all free pages unstable */
lghi %r2,0
brasl %r14,arch_set_page_states
/* Call arch specific early resume code */
brasl %r14,s390_early_resume
/* Return 0 */
lmg %r6,%r15,STACK_FRAME_OVERHEAD + __SF_GPRS(%r15)
lghi %r2,0
BR_EX %r14
.section .data..nosave,"aw",@progbits
.align 8
.Ldisabled_wait_31:
.long 0x000a0000,0x00000000
.Lpanic_string:
.asciz "Resume not possible because suspend CPU is no longer available\n"
.align 8
.Lrestart_diag308_psw:
.long 0x00080000,0x80000000
.Lrestart_suspend_psw:
.quad 0x0000000180000000,restart_suspend
.Lnew_pgm_check_psw:
.quad 0,pgm_check_entry
.Lresume_cpu:
.byte 0,0
+92
View File
@@ -0,0 +1,92 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 version
* Copyright IBM Corp. 1999, 2000
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
* Thomas Spatzier (tspat@de.ibm.com)
*
* Derived from "arch/i386/kernel/sys_i386.c"
*
* This file contains various random system calls that
* have a non-standard calling sequence on the Linux/s390
* platform.
*/
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/mm.h>
#include <linux/fs.h>
#include <linux/smp.h>
#include <linux/sem.h>
#include <linux/msg.h>
#include <linux/shm.h>
#include <linux/stat.h>
#include <linux/syscalls.h>
#include <linux/mman.h>
#include <linux/file.h>
#include <linux/utsname.h>
#include <linux/personality.h>
#include <linux/unistd.h>
#include <linux/ipc.h>
#include <linux/uaccess.h>
#include "entry.h"
/*
* Perform the mmap() system call. Linux for S/390 isn't able to handle more
* than 5 system call parameters, so this system call uses a memory block
* for parameter passing.
*/
struct s390_mmap_arg_struct {
unsigned long addr;
unsigned long len;
unsigned long prot;
unsigned long flags;
unsigned long fd;
unsigned long offset;
};
SYSCALL_DEFINE1(mmap2, struct s390_mmap_arg_struct __user *, arg)
{
struct s390_mmap_arg_struct a;
int error = -EFAULT;
if (copy_from_user(&a, arg, sizeof(a)))
goto out;
error = ksys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd, a.offset);
out:
return error;
}
/*
* sys_ipc() is the de-multiplexer for the SysV IPC calls.
*/
SYSCALL_DEFINE5(s390_ipc, uint, call, int, first, unsigned long, second,
unsigned long, third, void __user *, ptr)
{
if (call >> 16)
return -EINVAL;
/* The s390 sys_ipc variant has only five parameters instead of six
* like the generic variant. The only difference is the handling of
* the SEMTIMEDOP subcall where on s390 the third parameter is used
* as a pointer to a struct timespec where the generic variant uses
* the fifth parameter.
* Therefore we can call the generic variant by simply passing the
* third parameter also as fifth parameter.
*/
return sys_ipc(call, first, second, third, ptr, third);
}
SYSCALL_DEFINE1(s390_personality, unsigned int, personality)
{
unsigned int ret;
if (personality(current->personality) == PER_LINUX32 &&
personality(personality) == PER_LINUX)
personality |= PER_LINUX32;
ret = sys_personality(personality);
if (personality(ret) == PER_LINUX32)
ret &= ~PER_LINUX32;
return ret;
}
+52
View File
@@ -0,0 +1,52 @@
# SPDX-License-Identifier: GPL-2.0
gen := arch/$(ARCH)/include/generated
kapi := $(gen)/asm
uapi := $(gen)/uapi/asm
syscall := $(srctree)/$(src)/syscall.tbl
systbl := $(srctree)/$(src)/syscalltbl
gen-y := $(kapi)/syscall_table.h
kapi-hdrs-y := $(kapi)/unistd_nr.h
uapi-hdrs-y := $(uapi)/unistd_32.h
uapi-hdrs-y += $(uapi)/unistd_64.h
targets += $(addprefix ../../../,$(gen-y) $(kapi-hdrs-y) $(uapi-hdrs-y))
PHONY += kapi uapi
kapi: $(gen-y) $(kapi-hdrs-y)
uapi: $(uapi-hdrs-y)
# Create output directory if not already present
_dummy := $(shell [ -d '$(uapi)' ] || mkdir -p '$(uapi)') \
$(shell [ -d '$(kapi)' ] || mkdir -p '$(kapi)')
define filechk_syshdr
$(CONFIG_SHELL) '$(systbl)' -H -a $(syshdr_abi_$(basetarget)) -f "$2" < $<
endef
define filechk_sysnr
$(CONFIG_SHELL) '$(systbl)' -N -a $(sysnr_abi_$(basetarget)) < $<
endef
define filechk_syscalls
$(CONFIG_SHELL) '$(systbl)' -S < $<
endef
syshdr_abi_unistd_32 := common,32
$(uapi)/unistd_32.h: $(syscall) FORCE
$(call filechk,syshdr,$@)
syshdr_abi_unistd_64 := common,64
$(uapi)/unistd_64.h: $(syscall) FORCE
$(call filechk,syshdr,$@)
$(kapi)/syscall_table.h: $(syscall) FORCE
$(call filechk,syscalls)
sysnr_abi_unistd_nr := common,32,64
$(kapi)/unistd_nr.h: $(syscall) FORCE
$(call filechk,sysnr)
+395
View File
@@ -0,0 +1,395 @@
# SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note
#
# System call table for s390
#
# Format:
#
# <nr> <abi> <syscall> <entry-64bit> <compat-entry>
#
# where <abi> can be common, 64, or 32
1 common exit sys_exit sys_exit
2 common fork sys_fork sys_fork
3 common read sys_read compat_sys_s390_read
4 common write sys_write compat_sys_s390_write
5 common open sys_open compat_sys_open
6 common close sys_close sys_close
7 common restart_syscall sys_restart_syscall sys_restart_syscall
8 common creat sys_creat compat_sys_creat
9 common link sys_link compat_sys_link
10 common unlink sys_unlink compat_sys_unlink
11 common execve sys_execve compat_sys_execve
12 common chdir sys_chdir compat_sys_chdir
13 32 time - compat_sys_time
14 common mknod sys_mknod compat_sys_mknod
15 common chmod sys_chmod compat_sys_chmod
16 32 lchown - compat_sys_s390_lchown16
19 common lseek sys_lseek compat_sys_lseek
20 common getpid sys_getpid sys_getpid
21 common mount sys_mount compat_sys_mount
22 common umount sys_oldumount compat_sys_oldumount
23 32 setuid - compat_sys_s390_setuid16
24 32 getuid - compat_sys_s390_getuid16
25 32 stime - compat_sys_stime
26 common ptrace sys_ptrace compat_sys_ptrace
27 common alarm sys_alarm sys_alarm
29 common pause sys_pause sys_pause
30 common utime sys_utime compat_sys_utime
33 common access sys_access compat_sys_access
34 common nice sys_nice sys_nice
36 common sync sys_sync sys_sync
37 common kill sys_kill sys_kill
38 common rename sys_rename compat_sys_rename
39 common mkdir sys_mkdir compat_sys_mkdir
40 common rmdir sys_rmdir compat_sys_rmdir
41 common dup sys_dup sys_dup
42 common pipe sys_pipe compat_sys_pipe
43 common times sys_times compat_sys_times
45 common brk sys_brk compat_sys_brk
46 32 setgid - compat_sys_s390_setgid16
47 32 getgid - compat_sys_s390_getgid16
48 common signal sys_signal compat_sys_signal
49 32 geteuid - compat_sys_s390_geteuid16
50 32 getegid - compat_sys_s390_getegid16
51 common acct sys_acct compat_sys_acct
52 common umount2 sys_umount compat_sys_umount
54 common ioctl sys_ioctl compat_sys_ioctl
55 common fcntl sys_fcntl compat_sys_fcntl
57 common setpgid sys_setpgid sys_setpgid
60 common umask sys_umask sys_umask
61 common chroot sys_chroot compat_sys_chroot
62 common ustat sys_ustat compat_sys_ustat
63 common dup2 sys_dup2 sys_dup2
64 common getppid sys_getppid sys_getppid
65 common getpgrp sys_getpgrp sys_getpgrp
66 common setsid sys_setsid sys_setsid
67 common sigaction sys_sigaction compat_sys_sigaction
70 32 setreuid - compat_sys_s390_setreuid16
71 32 setregid - compat_sys_s390_setregid16
72 common sigsuspend sys_sigsuspend compat_sys_sigsuspend
73 common sigpending sys_sigpending compat_sys_sigpending
74 common sethostname sys_sethostname compat_sys_sethostname
75 common setrlimit sys_setrlimit compat_sys_setrlimit
76 32 getrlimit - compat_sys_old_getrlimit
77 common getrusage sys_getrusage compat_sys_getrusage
78 common gettimeofday sys_gettimeofday compat_sys_gettimeofday
79 common settimeofday sys_settimeofday compat_sys_settimeofday
80 32 getgroups - compat_sys_s390_getgroups16
81 32 setgroups - compat_sys_s390_setgroups16
83 common symlink sys_symlink compat_sys_symlink
85 common readlink sys_readlink compat_sys_readlink
86 common uselib sys_uselib compat_sys_uselib
87 common swapon sys_swapon compat_sys_swapon
88 common reboot sys_reboot compat_sys_reboot
89 common readdir - compat_sys_old_readdir
90 common mmap sys_old_mmap compat_sys_s390_old_mmap
91 common munmap sys_munmap compat_sys_munmap
92 common truncate sys_truncate compat_sys_truncate
93 common ftruncate sys_ftruncate compat_sys_ftruncate
94 common fchmod sys_fchmod sys_fchmod
95 32 fchown - compat_sys_s390_fchown16
96 common getpriority sys_getpriority sys_getpriority
97 common setpriority sys_setpriority sys_setpriority
99 common statfs sys_statfs compat_sys_statfs
100 common fstatfs sys_fstatfs compat_sys_fstatfs
101 32 ioperm - -
102 common socketcall sys_socketcall compat_sys_socketcall
103 common syslog sys_syslog compat_sys_syslog
104 common setitimer sys_setitimer compat_sys_setitimer
105 common getitimer sys_getitimer compat_sys_getitimer
106 common stat sys_newstat compat_sys_newstat
107 common lstat sys_newlstat compat_sys_newlstat
108 common fstat sys_newfstat compat_sys_newfstat
110 common lookup_dcookie sys_lookup_dcookie compat_sys_lookup_dcookie
111 common vhangup sys_vhangup sys_vhangup
112 common idle - -
114 common wait4 sys_wait4 compat_sys_wait4
115 common swapoff sys_swapoff compat_sys_swapoff
116 common sysinfo sys_sysinfo compat_sys_sysinfo
117 common ipc sys_s390_ipc compat_sys_s390_ipc
118 common fsync sys_fsync sys_fsync
119 common sigreturn sys_sigreturn compat_sys_sigreturn
120 common clone sys_clone compat_sys_clone
121 common setdomainname sys_setdomainname compat_sys_setdomainname
122 common uname sys_newuname compat_sys_newuname
124 common adjtimex sys_adjtimex compat_sys_adjtimex
125 common mprotect sys_mprotect compat_sys_mprotect
126 common sigprocmask sys_sigprocmask compat_sys_sigprocmask
127 common create_module - -
128 common init_module sys_init_module compat_sys_init_module
129 common delete_module sys_delete_module compat_sys_delete_module
130 common get_kernel_syms - -
131 common quotactl sys_quotactl compat_sys_quotactl
132 common getpgid sys_getpgid sys_getpgid
133 common fchdir sys_fchdir sys_fchdir
134 common bdflush sys_bdflush compat_sys_bdflush
135 common sysfs sys_sysfs compat_sys_sysfs
136 common personality sys_s390_personality sys_s390_personality
137 common afs_syscall - -
138 32 setfsuid - compat_sys_s390_setfsuid16
139 32 setfsgid - compat_sys_s390_setfsgid16
140 32 _llseek - compat_sys_llseek
141 common getdents sys_getdents compat_sys_getdents
142 32 _newselect - compat_sys_select
142 64 select sys_select -
143 common flock sys_flock sys_flock
144 common msync sys_msync compat_sys_msync
145 common readv sys_readv compat_sys_readv
146 common writev sys_writev compat_sys_writev
147 common getsid sys_getsid sys_getsid
148 common fdatasync sys_fdatasync sys_fdatasync
149 common _sysctl sys_sysctl compat_sys_sysctl
150 common mlock sys_mlock compat_sys_mlock
151 common munlock sys_munlock compat_sys_munlock
152 common mlockall sys_mlockall sys_mlockall
153 common munlockall sys_munlockall sys_munlockall
154 common sched_setparam sys_sched_setparam compat_sys_sched_setparam
155 common sched_getparam sys_sched_getparam compat_sys_sched_getparam
156 common sched_setscheduler sys_sched_setscheduler compat_sys_sched_setscheduler
157 common sched_getscheduler sys_sched_getscheduler sys_sched_getscheduler
158 common sched_yield sys_sched_yield sys_sched_yield
159 common sched_get_priority_max sys_sched_get_priority_max sys_sched_get_priority_max
160 common sched_get_priority_min sys_sched_get_priority_min sys_sched_get_priority_min
161 common sched_rr_get_interval sys_sched_rr_get_interval compat_sys_sched_rr_get_interval
162 common nanosleep sys_nanosleep compat_sys_nanosleep
163 common mremap sys_mremap compat_sys_mremap
164 32 setresuid - compat_sys_s390_setresuid16
165 32 getresuid - compat_sys_s390_getresuid16
167 common query_module - -
168 common poll sys_poll compat_sys_poll
169 common nfsservctl - -
170 32 setresgid - compat_sys_s390_setresgid16
171 32 getresgid - compat_sys_s390_getresgid16
172 common prctl sys_prctl compat_sys_prctl
173 common rt_sigreturn sys_rt_sigreturn compat_sys_rt_sigreturn
174 common rt_sigaction sys_rt_sigaction compat_sys_rt_sigaction
175 common rt_sigprocmask sys_rt_sigprocmask compat_sys_rt_sigprocmask
176 common rt_sigpending sys_rt_sigpending compat_sys_rt_sigpending
177 common rt_sigtimedwait sys_rt_sigtimedwait compat_sys_rt_sigtimedwait
178 common rt_sigqueueinfo sys_rt_sigqueueinfo compat_sys_rt_sigqueueinfo
179 common rt_sigsuspend sys_rt_sigsuspend compat_sys_rt_sigsuspend
180 common pread64 sys_pread64 compat_sys_s390_pread64
181 common pwrite64 sys_pwrite64 compat_sys_s390_pwrite64
182 32 chown - compat_sys_s390_chown16
183 common getcwd sys_getcwd compat_sys_getcwd
184 common capget sys_capget compat_sys_capget
185 common capset sys_capset compat_sys_capset
186 common sigaltstack sys_sigaltstack compat_sys_sigaltstack
187 common sendfile sys_sendfile64 compat_sys_sendfile
188 common getpmsg - -
189 common putpmsg - -
190 common vfork sys_vfork sys_vfork
191 32 ugetrlimit - compat_sys_getrlimit
191 64 getrlimit sys_getrlimit -
192 32 mmap2 - compat_sys_s390_mmap2
193 32 truncate64 - compat_sys_s390_truncate64
194 32 ftruncate64 - compat_sys_s390_ftruncate64
195 32 stat64 - compat_sys_s390_stat64
196 32 lstat64 - compat_sys_s390_lstat64
197 32 fstat64 - compat_sys_s390_fstat64
198 32 lchown32 - compat_sys_lchown
198 64 lchown sys_lchown -
199 32 getuid32 - sys_getuid
199 64 getuid sys_getuid -
200 32 getgid32 - sys_getgid
200 64 getgid sys_getgid -
201 32 geteuid32 - sys_geteuid
201 64 geteuid sys_geteuid -
202 32 getegid32 - sys_getegid
202 64 getegid sys_getegid -
203 32 setreuid32 - sys_setreuid
203 64 setreuid sys_setreuid -
204 32 setregid32 - sys_setregid
204 64 setregid sys_setregid -
205 32 getgroups32 - compat_sys_getgroups
205 64 getgroups sys_getgroups -
206 32 setgroups32 - compat_sys_setgroups
206 64 setgroups sys_setgroups -
207 32 fchown32 - sys_fchown
207 64 fchown sys_fchown -
208 32 setresuid32 - sys_setresuid
208 64 setresuid sys_setresuid -
209 32 getresuid32 - compat_sys_getresuid
209 64 getresuid sys_getresuid -
210 32 setresgid32 - sys_setresgid
210 64 setresgid sys_setresgid -
211 32 getresgid32 - compat_sys_getresgid
211 64 getresgid sys_getresgid -
212 32 chown32 - compat_sys_chown
212 64 chown sys_chown -
213 32 setuid32 - sys_setuid
213 64 setuid sys_setuid -
214 32 setgid32 - sys_setgid
214 64 setgid sys_setgid -
215 32 setfsuid32 - sys_setfsuid
215 64 setfsuid sys_setfsuid -
216 32 setfsgid32 - sys_setfsgid
216 64 setfsgid sys_setfsgid -
217 common pivot_root sys_pivot_root compat_sys_pivot_root
218 common mincore sys_mincore compat_sys_mincore
219 common madvise sys_madvise compat_sys_madvise
220 common getdents64 sys_getdents64 compat_sys_getdents64
221 32 fcntl64 - compat_sys_fcntl64
222 common readahead sys_readahead compat_sys_s390_readahead
223 32 sendfile64 - compat_sys_sendfile64
224 common setxattr sys_setxattr compat_sys_setxattr
225 common lsetxattr sys_lsetxattr compat_sys_lsetxattr
226 common fsetxattr sys_fsetxattr compat_sys_fsetxattr
227 common getxattr sys_getxattr compat_sys_getxattr
228 common lgetxattr sys_lgetxattr compat_sys_lgetxattr
229 common fgetxattr sys_fgetxattr compat_sys_fgetxattr
230 common listxattr sys_listxattr compat_sys_listxattr
231 common llistxattr sys_llistxattr compat_sys_llistxattr
232 common flistxattr sys_flistxattr compat_sys_flistxattr
233 common removexattr sys_removexattr compat_sys_removexattr
234 common lremovexattr sys_lremovexattr compat_sys_lremovexattr
235 common fremovexattr sys_fremovexattr compat_sys_fremovexattr
236 common gettid sys_gettid sys_gettid
237 common tkill sys_tkill sys_tkill
238 common futex sys_futex compat_sys_futex
239 common sched_setaffinity sys_sched_setaffinity compat_sys_sched_setaffinity
240 common sched_getaffinity sys_sched_getaffinity compat_sys_sched_getaffinity
241 common tgkill sys_tgkill sys_tgkill
243 common io_setup sys_io_setup compat_sys_io_setup
244 common io_destroy sys_io_destroy compat_sys_io_destroy
245 common io_getevents sys_io_getevents compat_sys_io_getevents
246 common io_submit sys_io_submit compat_sys_io_submit
247 common io_cancel sys_io_cancel compat_sys_io_cancel
248 common exit_group sys_exit_group sys_exit_group
249 common epoll_create sys_epoll_create sys_epoll_create
250 common epoll_ctl sys_epoll_ctl compat_sys_epoll_ctl
251 common epoll_wait sys_epoll_wait compat_sys_epoll_wait
252 common set_tid_address sys_set_tid_address compat_sys_set_tid_address
253 common fadvise64 sys_fadvise64_64 compat_sys_s390_fadvise64
254 common timer_create sys_timer_create compat_sys_timer_create
255 common timer_settime sys_timer_settime compat_sys_timer_settime
256 common timer_gettime sys_timer_gettime compat_sys_timer_gettime
257 common timer_getoverrun sys_timer_getoverrun sys_timer_getoverrun
258 common timer_delete sys_timer_delete sys_timer_delete
259 common clock_settime sys_clock_settime compat_sys_clock_settime
260 common clock_gettime sys_clock_gettime compat_sys_clock_gettime
261 common clock_getres sys_clock_getres compat_sys_clock_getres
262 common clock_nanosleep sys_clock_nanosleep compat_sys_clock_nanosleep
264 32 fadvise64_64 - compat_sys_s390_fadvise64_64
265 common statfs64 sys_statfs64 compat_sys_statfs64
266 common fstatfs64 sys_fstatfs64 compat_sys_fstatfs64
267 common remap_file_pages sys_remap_file_pages compat_sys_remap_file_pages
268 common mbind sys_mbind compat_sys_mbind
269 common get_mempolicy sys_get_mempolicy compat_sys_get_mempolicy
270 common set_mempolicy sys_set_mempolicy compat_sys_set_mempolicy
271 common mq_open sys_mq_open compat_sys_mq_open
272 common mq_unlink sys_mq_unlink compat_sys_mq_unlink
273 common mq_timedsend sys_mq_timedsend compat_sys_mq_timedsend
274 common mq_timedreceive sys_mq_timedreceive compat_sys_mq_timedreceive
275 common mq_notify sys_mq_notify compat_sys_mq_notify
276 common mq_getsetattr sys_mq_getsetattr compat_sys_mq_getsetattr
277 common kexec_load sys_kexec_load compat_sys_kexec_load
278 common add_key sys_add_key compat_sys_add_key
279 common request_key sys_request_key compat_sys_request_key
280 common keyctl sys_keyctl compat_sys_keyctl
281 common waitid sys_waitid compat_sys_waitid
282 common ioprio_set sys_ioprio_set sys_ioprio_set
283 common ioprio_get sys_ioprio_get sys_ioprio_get
284 common inotify_init sys_inotify_init sys_inotify_init
285 common inotify_add_watch sys_inotify_add_watch compat_sys_inotify_add_watch
286 common inotify_rm_watch sys_inotify_rm_watch sys_inotify_rm_watch
287 common migrate_pages sys_migrate_pages compat_sys_migrate_pages
288 common openat sys_openat compat_sys_openat
289 common mkdirat sys_mkdirat compat_sys_mkdirat
290 common mknodat sys_mknodat compat_sys_mknodat
291 common fchownat sys_fchownat compat_sys_fchownat
292 common futimesat sys_futimesat compat_sys_futimesat
293 32 fstatat64 - compat_sys_s390_fstatat64
293 64 newfstatat sys_newfstatat -
294 common unlinkat sys_unlinkat compat_sys_unlinkat
295 common renameat sys_renameat compat_sys_renameat
296 common linkat sys_linkat compat_sys_linkat
297 common symlinkat sys_symlinkat compat_sys_symlinkat
298 common readlinkat sys_readlinkat compat_sys_readlinkat
299 common fchmodat sys_fchmodat compat_sys_fchmodat
300 common faccessat sys_faccessat compat_sys_faccessat
301 common pselect6 sys_pselect6 compat_sys_pselect6
302 common ppoll sys_ppoll compat_sys_ppoll
303 common unshare sys_unshare compat_sys_unshare
304 common set_robust_list sys_set_robust_list compat_sys_set_robust_list
305 common get_robust_list sys_get_robust_list compat_sys_get_robust_list
306 common splice sys_splice compat_sys_splice
307 common sync_file_range sys_sync_file_range compat_sys_s390_sync_file_range
308 common tee sys_tee compat_sys_tee
309 common vmsplice sys_vmsplice compat_sys_vmsplice
310 common move_pages sys_move_pages compat_sys_move_pages
311 common getcpu sys_getcpu compat_sys_getcpu
312 common epoll_pwait sys_epoll_pwait compat_sys_epoll_pwait
313 common utimes sys_utimes compat_sys_utimes
314 common fallocate sys_fallocate compat_sys_s390_fallocate
315 common utimensat sys_utimensat compat_sys_utimensat
316 common signalfd sys_signalfd compat_sys_signalfd
317 common timerfd - -
318 common eventfd sys_eventfd sys_eventfd
319 common timerfd_create sys_timerfd_create sys_timerfd_create
320 common timerfd_settime sys_timerfd_settime compat_sys_timerfd_settime
321 common timerfd_gettime sys_timerfd_gettime compat_sys_timerfd_gettime
322 common signalfd4 sys_signalfd4 compat_sys_signalfd4
323 common eventfd2 sys_eventfd2 sys_eventfd2
324 common inotify_init1 sys_inotify_init1 sys_inotify_init1
325 common pipe2 sys_pipe2 compat_sys_pipe2
326 common dup3 sys_dup3 sys_dup3
327 common epoll_create1 sys_epoll_create1 sys_epoll_create1
328 common preadv sys_preadv compat_sys_preadv
329 common pwritev sys_pwritev compat_sys_pwritev
330 common rt_tgsigqueueinfo sys_rt_tgsigqueueinfo compat_sys_rt_tgsigqueueinfo
331 common perf_event_open sys_perf_event_open compat_sys_perf_event_open
332 common fanotify_init sys_fanotify_init sys_fanotify_init
333 common fanotify_mark sys_fanotify_mark compat_sys_fanotify_mark
334 common prlimit64 sys_prlimit64 compat_sys_prlimit64
335 common name_to_handle_at sys_name_to_handle_at compat_sys_name_to_handle_at
336 common open_by_handle_at sys_open_by_handle_at compat_sys_open_by_handle_at
337 common clock_adjtime sys_clock_adjtime compat_sys_clock_adjtime
338 common syncfs sys_syncfs sys_syncfs
339 common setns sys_setns sys_setns
340 common process_vm_readv sys_process_vm_readv compat_sys_process_vm_readv
341 common process_vm_writev sys_process_vm_writev compat_sys_process_vm_writev
342 common s390_runtime_instr sys_s390_runtime_instr sys_s390_runtime_instr
343 common kcmp sys_kcmp compat_sys_kcmp
344 common finit_module sys_finit_module compat_sys_finit_module
345 common sched_setattr sys_sched_setattr compat_sys_sched_setattr
346 common sched_getattr sys_sched_getattr compat_sys_sched_getattr
347 common renameat2 sys_renameat2 compat_sys_renameat2
348 common seccomp sys_seccomp compat_sys_seccomp
349 common getrandom sys_getrandom compat_sys_getrandom
350 common memfd_create sys_memfd_create compat_sys_memfd_create
351 common bpf sys_bpf compat_sys_bpf
352 common s390_pci_mmio_write sys_s390_pci_mmio_write compat_sys_s390_pci_mmio_write
353 common s390_pci_mmio_read sys_s390_pci_mmio_read compat_sys_s390_pci_mmio_read
354 common execveat sys_execveat compat_sys_execveat
355 common userfaultfd sys_userfaultfd sys_userfaultfd
356 common membarrier sys_membarrier sys_membarrier
357 common recvmmsg sys_recvmmsg compat_sys_recvmmsg
358 common sendmmsg sys_sendmmsg compat_sys_sendmmsg
359 common socket sys_socket sys_socket
360 common socketpair sys_socketpair compat_sys_socketpair
361 common bind sys_bind compat_sys_bind
362 common connect sys_connect compat_sys_connect
363 common listen sys_listen sys_listen
364 common accept4 sys_accept4 compat_sys_accept4
365 common getsockopt sys_getsockopt compat_sys_getsockopt
366 common setsockopt sys_setsockopt compat_sys_setsockopt
367 common getsockname sys_getsockname compat_sys_getsockname
368 common getpeername sys_getpeername compat_sys_getpeername
369 common sendto sys_sendto compat_sys_sendto
370 common sendmsg sys_sendmsg compat_sys_sendmsg
371 common recvfrom sys_recvfrom compat_sys_recvfrom
372 common recvmsg sys_recvmsg compat_sys_recvmsg
373 common shutdown sys_shutdown sys_shutdown
374 common mlock2 sys_mlock2 compat_sys_mlock2
375 common copy_file_range sys_copy_file_range compat_sys_copy_file_range
376 common preadv2 sys_preadv2 compat_sys_preadv2
377 common pwritev2 sys_pwritev2 compat_sys_pwritev2
378 common s390_guarded_storage sys_s390_guarded_storage compat_sys_s390_guarded_storage
379 common statx sys_statx compat_sys_statx
380 common s390_sthyi sys_s390_sthyi compat_sys_s390_sthyi
381 common kexec_file_load sys_kexec_file_load compat_sys_kexec_file_load
382 common io_pgetevents sys_io_pgetevents compat_sys_io_pgetevents
383 common rseq sys_rseq compat_sys_rseq
424 common pidfd_send_signal sys_pidfd_send_signal sys_pidfd_send_signal
434 common pidfd_open sys_pidfd_open sys_pidfd_open
+232
View File
@@ -0,0 +1,232 @@
#!/bin/sh
# SPDX-License-Identifier: GPL-2.0
#
# Generate system call table and header files
#
# Copyright IBM Corp. 2018
# Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
#
# File path to the system call table definition.
# You can set the path with the -i option. If omitted,
# system call table definitions are read from standard input.
#
SYSCALL_TBL=""
create_syscall_table_entries()
{
local nr abi name entry64 entry32 _ignore
local temp=$(mktemp ${TMPDIR:-/tmp}/syscalltbl-common.XXXXXXXXX)
(
#
# Initialize with 0 to create an NI_SYSCALL for 0
#
local prev_nr=0 prev_32=sys_ni_syscall prev_64=sys_ni_syscall
while read nr abi name entry64 entry32 _ignore; do
test x$entry32 = x- && entry32=sys_ni_syscall
test x$entry64 = x- && entry64=sys_ni_syscall
if test $prev_nr -eq $nr; then
#
# Same syscall but different ABI, just update
# the respective entry point
#
case $abi in
32)
prev_32=$entry32
;;
64)
prev_64=$entry64
;;
esac
continue;
else
printf "%d\t%s\t%s\n" $prev_nr $prev_64 $prev_32
fi
prev_nr=$nr
prev_64=$entry64
prev_32=$entry32
done
printf "%d\t%s\t%s\n" $prev_nr $prev_64 $prev_32
) >> $temp
#
# Check for duplicate syscall numbers
#
if ! cat $temp |cut -f1 |uniq -d 2>&1; then
echo "Error: generated system call table contains duplicate entries: $temp" >&2
exit 1
fi
#
# Generate syscall table
#
prev_nr=0
while read nr entry64 entry32; do
while test $prev_nr -lt $((nr - 1)); do
printf "NI_SYSCALL\n"
prev_nr=$((prev_nr + 1))
done
if test x$entry64 = xsys_ni_syscall &&
test x$entry32 = xsys_ni_syscall; then
printf "NI_SYSCALL\n"
else
printf "SYSCALL(%s,%s)\n" $entry64 $entry32
fi
prev_nr=$nr
done < $temp
rm $temp
}
generate_syscall_table()
{
cat <<-EoHEADER
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Definitions for sys_call_table, each line represents an
* entry in the table in the form
* SYSCALL(64 bit syscall, 31 bit emulated syscall)
*
* This file is meant to be included from entry.S.
*/
#define NI_SYSCALL SYSCALL(sys_ni_syscall,sys_ni_syscall)
EoHEADER
grep -Ev '^(#|[[:blank:]]*$)' $SYSCALL_TBL \
|sort -k1 -n \
|create_syscall_table_entries
}
create_header_defines()
{
local nr abi name _ignore
while read nr abi name _ignore; do
printf "#define __NR_%s %d\n" $name $nr
done
}
normalize_fileguard()
{
local fileguard="$1"
echo "$1" |tr '[[:lower:]]' '[[:upper:]]' \
|sed -e 's/[^A-Z0-9_]/_/g' -e 's/__/_/g'
}
generate_syscall_header()
{
local abis=$(echo "($1)" | tr ',' '|')
local filename="$2"
local fileguard suffix
if test "$filename"; then
fileguard=$(normalize_fileguard "__UAPI_ASM_S390_$2")
else
case "$abis" in
*64*) suffix=64 ;;
*32*) suffix=32 ;;
esac
fileguard=$(normalize_fileguard "__UAPI_ASM_S390_SYSCALLS_$suffix")
fi
cat <<-EoHEADER
/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
#ifndef ${fileguard}
#define ${fileguard}
EoHEADER
grep -E "^[[:digit:]]+[[:space:]]+${abis}" $SYSCALL_TBL \
|sort -k1 -n \
|create_header_defines
cat <<-EoFOOTER
#endif /* ${fileguard} */
EoFOOTER
}
__max_syscall_nr()
{
local abis=$(echo "($1)" | tr ',' '|')
grep -E "^[[:digit:]]+[[:space:]]+${abis}" $SYSCALL_TBL \
|sed -ne 's/^\([[:digit:]]*\)[[:space:]].*/\1/p' \
|sort -n \
|tail -1
}
generate_syscall_nr()
{
local abis="$1"
local max_syscall_nr num_syscalls
max_syscall_nr=$(__max_syscall_nr "$abis")
num_syscalls=$((max_syscall_nr + 1))
cat <<-EoHEADER
/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
#ifndef __ASM_S390_SYSCALLS_NR
#define __ASM_S390_SYSCALLS_NR
#define NR_syscalls ${num_syscalls}
#endif /* __ASM_S390_SYSCALLS_NR */
EoHEADER
}
#
# Parse command line arguments
#
do_syscall_header=""
do_syscall_table=""
do_syscall_nr=""
output_file=""
abi_list="common,64"
filename=""
while getopts ":HNSXi:a:f:" arg; do
case $arg in
a)
abi_list="$OPTARG"
;;
i)
SYSCALL_TBL="$OPTARG"
;;
f)
filename=${OPTARG##*/}
;;
H)
do_syscall_header=1
;;
N)
do_syscall_nr=1
;;
S)
do_syscall_table=1
;;
X)
set -x
;;
:)
echo "Missing argument for -$OPTARG" >&2
exit 1
;;
\?)
echo "Invalid option specified" >&2
exit 1
;;
esac
done
test "$do_syscall_header" && generate_syscall_header "$abi_list" "$filename"
test "$do_syscall_table" && generate_syscall_table
test "$do_syscall_nr" && generate_syscall_nr "$abi_list"
exit 0
+568
View File
@@ -0,0 +1,568 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2001, 2009
* Author(s): Ulrich Weigand <Ulrich.Weigand@de.ibm.com>,
* Martin Schwidefsky <schwidefsky@de.ibm.com>,
*/
#include <linux/debugfs.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/init.h>
#include <linux/delay.h>
#include <linux/export.h>
#include <linux/slab.h>
#include <asm/ebcdic.h>
#include <asm/debug.h>
#include <asm/sysinfo.h>
#include <asm/cpcmd.h>
#include <asm/topology.h>
#include <asm/fpu/api.h>
int topology_max_mnest;
static inline int __stsi(void *sysinfo, int fc, int sel1, int sel2, int *lvl)
{
register int r0 asm("0") = (fc << 28) | sel1;
register int r1 asm("1") = sel2;
int rc = 0;
asm volatile(
" stsi 0(%3)\n"
"0: jz 2f\n"
"1: lhi %1,%4\n"
"2:\n"
EX_TABLE(0b, 1b)
: "+d" (r0), "+d" (rc)
: "d" (r1), "a" (sysinfo), "K" (-EOPNOTSUPP)
: "cc", "memory");
*lvl = ((unsigned int) r0) >> 28;
return rc;
}
/*
* stsi - store system information
*
* Returns the current configuration level if function code 0 was specified.
* Otherwise returns 0 on success or a negative value on error.
*/
int stsi(void *sysinfo, int fc, int sel1, int sel2)
{
int lvl, rc;
rc = __stsi(sysinfo, fc, sel1, sel2, &lvl);
if (rc)
return rc;
return fc ? 0 : lvl;
}
EXPORT_SYMBOL(stsi);
#ifdef CONFIG_PROC_FS
static bool convert_ext_name(unsigned char encoding, char *name, size_t len)
{
switch (encoding) {
case 1: /* EBCDIC */
EBCASC(name, len);
break;
case 2: /* UTF-8 */
break;
default:
return false;
}
return true;
}
static void stsi_1_1_1(struct seq_file *m, struct sysinfo_1_1_1 *info)
{
int i;
if (stsi(info, 1, 1, 1))
return;
EBCASC(info->manufacturer, sizeof(info->manufacturer));
EBCASC(info->type, sizeof(info->type));
EBCASC(info->model, sizeof(info->model));
EBCASC(info->sequence, sizeof(info->sequence));
EBCASC(info->plant, sizeof(info->plant));
EBCASC(info->model_capacity, sizeof(info->model_capacity));
EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap));
EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap));
seq_printf(m, "Manufacturer: %-16.16s\n", info->manufacturer);
seq_printf(m, "Type: %-4.4s\n", info->type);
if (info->lic)
seq_printf(m, "LIC Identifier: %016lx\n", info->lic);
/*
* Sigh: the model field has been renamed with System z9
* to model_capacity and a new model field has been added
* after the plant field. To avoid confusing older programs
* the "Model:" prints "model_capacity model" or just
* "model_capacity" if the model string is empty .
*/
seq_printf(m, "Model: %-16.16s", info->model_capacity);
if (info->model[0] != '\0')
seq_printf(m, " %-16.16s", info->model);
seq_putc(m, '\n');
seq_printf(m, "Sequence Code: %-16.16s\n", info->sequence);
seq_printf(m, "Plant: %-4.4s\n", info->plant);
seq_printf(m, "Model Capacity: %-16.16s %08u\n",
info->model_capacity, info->model_cap_rating);
if (info->model_perm_cap_rating)
seq_printf(m, "Model Perm. Capacity: %-16.16s %08u\n",
info->model_perm_cap,
info->model_perm_cap_rating);
if (info->model_temp_cap_rating)
seq_printf(m, "Model Temp. Capacity: %-16.16s %08u\n",
info->model_temp_cap,
info->model_temp_cap_rating);
if (info->ncr)
seq_printf(m, "Nominal Cap. Rating: %08u\n", info->ncr);
if (info->npr)
seq_printf(m, "Nominal Perm. Rating: %08u\n", info->npr);
if (info->ntr)
seq_printf(m, "Nominal Temp. Rating: %08u\n", info->ntr);
if (info->cai) {
seq_printf(m, "Capacity Adj. Ind.: %d\n", info->cai);
seq_printf(m, "Capacity Ch. Reason: %d\n", info->ccr);
seq_printf(m, "Capacity Transient: %d\n", info->t);
}
if (info->p) {
for (i = 1; i <= ARRAY_SIZE(info->typepct); i++) {
seq_printf(m, "Type %d Percentage: %d\n",
i, info->typepct[i - 1]);
}
}
}
static void stsi_15_1_x(struct seq_file *m, struct sysinfo_15_1_x *info)
{
int i;
seq_putc(m, '\n');
if (!MACHINE_HAS_TOPOLOGY)
return;
if (stsi(info, 15, 1, topology_max_mnest))
return;
seq_printf(m, "CPU Topology HW: ");
for (i = 0; i < TOPOLOGY_NR_MAG; i++)
seq_printf(m, " %d", info->mag[i]);
seq_putc(m, '\n');
#ifdef CONFIG_SCHED_TOPOLOGY
store_topology(info);
seq_printf(m, "CPU Topology SW: ");
for (i = 0; i < TOPOLOGY_NR_MAG; i++)
seq_printf(m, " %d", info->mag[i]);
seq_putc(m, '\n');
#endif
}
static void stsi_1_2_2(struct seq_file *m, struct sysinfo_1_2_2 *info)
{
struct sysinfo_1_2_2_extension *ext;
int i;
if (stsi(info, 1, 2, 2))
return;
ext = (struct sysinfo_1_2_2_extension *)
((unsigned long) info + info->acc_offset);
seq_printf(m, "CPUs Total: %d\n", info->cpus_total);
seq_printf(m, "CPUs Configured: %d\n", info->cpus_configured);
seq_printf(m, "CPUs Standby: %d\n", info->cpus_standby);
seq_printf(m, "CPUs Reserved: %d\n", info->cpus_reserved);
if (info->mt_installed) {
seq_printf(m, "CPUs G-MTID: %d\n", info->mt_gtid);
seq_printf(m, "CPUs S-MTID: %d\n", info->mt_stid);
}
/*
* Sigh 2. According to the specification the alternate
* capability field is a 32 bit floating point number
* if the higher order 8 bits are not zero. Printing
* a floating point number in the kernel is a no-no,
* always print the number as 32 bit unsigned integer.
* The user-space needs to know about the strange
* encoding of the alternate cpu capability.
*/
seq_printf(m, "Capability: %u", info->capability);
if (info->format == 1)
seq_printf(m, " %u", ext->alt_capability);
seq_putc(m, '\n');
if (info->nominal_cap)
seq_printf(m, "Nominal Capability: %d\n", info->nominal_cap);
if (info->secondary_cap)
seq_printf(m, "Secondary Capability: %d\n", info->secondary_cap);
for (i = 2; i <= info->cpus_total; i++) {
seq_printf(m, "Adjustment %02d-way: %u",
i, info->adjustment[i-2]);
if (info->format == 1)
seq_printf(m, " %u", ext->alt_adjustment[i-2]);
seq_putc(m, '\n');
}
}
static void stsi_2_2_2(struct seq_file *m, struct sysinfo_2_2_2 *info)
{
if (stsi(info, 2, 2, 2))
return;
EBCASC(info->name, sizeof(info->name));
seq_putc(m, '\n');
seq_printf(m, "LPAR Number: %d\n", info->lpar_number);
seq_printf(m, "LPAR Characteristics: ");
if (info->characteristics & LPAR_CHAR_DEDICATED)
seq_printf(m, "Dedicated ");
if (info->characteristics & LPAR_CHAR_SHARED)
seq_printf(m, "Shared ");
if (info->characteristics & LPAR_CHAR_LIMITED)
seq_printf(m, "Limited ");
seq_putc(m, '\n');
seq_printf(m, "LPAR Name: %-8.8s\n", info->name);
seq_printf(m, "LPAR Adjustment: %d\n", info->caf);
seq_printf(m, "LPAR CPUs Total: %d\n", info->cpus_total);
seq_printf(m, "LPAR CPUs Configured: %d\n", info->cpus_configured);
seq_printf(m, "LPAR CPUs Standby: %d\n", info->cpus_standby);
seq_printf(m, "LPAR CPUs Reserved: %d\n", info->cpus_reserved);
seq_printf(m, "LPAR CPUs Dedicated: %d\n", info->cpus_dedicated);
seq_printf(m, "LPAR CPUs Shared: %d\n", info->cpus_shared);
if (info->mt_installed) {
seq_printf(m, "LPAR CPUs G-MTID: %d\n", info->mt_gtid);
seq_printf(m, "LPAR CPUs S-MTID: %d\n", info->mt_stid);
seq_printf(m, "LPAR CPUs PS-MTID: %d\n", info->mt_psmtid);
}
if (convert_ext_name(info->vsne, info->ext_name, sizeof(info->ext_name))) {
seq_printf(m, "LPAR Extended Name: %-.256s\n", info->ext_name);
seq_printf(m, "LPAR UUID: %pUb\n", &info->uuid);
}
}
static void print_ext_name(struct seq_file *m, int lvl,
struct sysinfo_3_2_2 *info)
{
size_t len = sizeof(info->ext_names[lvl]);
if (!convert_ext_name(info->vm[lvl].evmne, info->ext_names[lvl], len))
return;
seq_printf(m, "VM%02d Extended Name: %-.256s\n", lvl,
info->ext_names[lvl]);
}
static void print_uuid(struct seq_file *m, int i, struct sysinfo_3_2_2 *info)
{
if (uuid_is_null(&info->vm[i].uuid))
return;
seq_printf(m, "VM%02d UUID: %pUb\n", i, &info->vm[i].uuid);
}
static void stsi_3_2_2(struct seq_file *m, struct sysinfo_3_2_2 *info)
{
int i;
if (stsi(info, 3, 2, 2))
return;
for (i = 0; i < info->count; i++) {
EBCASC(info->vm[i].name, sizeof(info->vm[i].name));
EBCASC(info->vm[i].cpi, sizeof(info->vm[i].cpi));
seq_putc(m, '\n');
seq_printf(m, "VM%02d Name: %-8.8s\n", i, info->vm[i].name);
seq_printf(m, "VM%02d Control Program: %-16.16s\n", i, info->vm[i].cpi);
seq_printf(m, "VM%02d Adjustment: %d\n", i, info->vm[i].caf);
seq_printf(m, "VM%02d CPUs Total: %d\n", i, info->vm[i].cpus_total);
seq_printf(m, "VM%02d CPUs Configured: %d\n", i, info->vm[i].cpus_configured);
seq_printf(m, "VM%02d CPUs Standby: %d\n", i, info->vm[i].cpus_standby);
seq_printf(m, "VM%02d CPUs Reserved: %d\n", i, info->vm[i].cpus_reserved);
print_ext_name(m, i, info);
print_uuid(m, i, info);
}
}
static int sysinfo_show(struct seq_file *m, void *v)
{
void *info = (void *)get_zeroed_page(GFP_KERNEL);
int level;
if (!info)
return 0;
level = stsi(NULL, 0, 0, 0);
if (level >= 1)
stsi_1_1_1(m, info);
if (level >= 1)
stsi_15_1_x(m, info);
if (level >= 1)
stsi_1_2_2(m, info);
if (level >= 2)
stsi_2_2_2(m, info);
if (level >= 3)
stsi_3_2_2(m, info);
free_page((unsigned long)info);
return 0;
}
static int __init sysinfo_create_proc(void)
{
proc_create_single("sysinfo", 0444, NULL, sysinfo_show);
return 0;
}
device_initcall(sysinfo_create_proc);
#endif /* CONFIG_PROC_FS */
/*
* Service levels interface.
*/
static DECLARE_RWSEM(service_level_sem);
static LIST_HEAD(service_level_list);
int register_service_level(struct service_level *slr)
{
struct service_level *ptr;
down_write(&service_level_sem);
list_for_each_entry(ptr, &service_level_list, list)
if (ptr == slr) {
up_write(&service_level_sem);
return -EEXIST;
}
list_add_tail(&slr->list, &service_level_list);
up_write(&service_level_sem);
return 0;
}
EXPORT_SYMBOL(register_service_level);
int unregister_service_level(struct service_level *slr)
{
struct service_level *ptr, *next;
int rc = -ENOENT;
down_write(&service_level_sem);
list_for_each_entry_safe(ptr, next, &service_level_list, list) {
if (ptr != slr)
continue;
list_del(&ptr->list);
rc = 0;
break;
}
up_write(&service_level_sem);
return rc;
}
EXPORT_SYMBOL(unregister_service_level);
static void *service_level_start(struct seq_file *m, loff_t *pos)
{
down_read(&service_level_sem);
return seq_list_start(&service_level_list, *pos);
}
static void *service_level_next(struct seq_file *m, void *p, loff_t *pos)
{
return seq_list_next(p, &service_level_list, pos);
}
static void service_level_stop(struct seq_file *m, void *p)
{
up_read(&service_level_sem);
}
static int service_level_show(struct seq_file *m, void *p)
{
struct service_level *slr;
slr = list_entry(p, struct service_level, list);
slr->seq_print(m, slr);
return 0;
}
static const struct seq_operations service_level_seq_ops = {
.start = service_level_start,
.next = service_level_next,
.stop = service_level_stop,
.show = service_level_show
};
static void service_level_vm_print(struct seq_file *m,
struct service_level *slr)
{
char *query_buffer, *str;
query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA);
if (!query_buffer)
return;
cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL);
str = strchr(query_buffer, '\n');
if (str)
*str = 0;
seq_printf(m, "VM: %s\n", query_buffer);
kfree(query_buffer);
}
static struct service_level service_level_vm = {
.seq_print = service_level_vm_print
};
static __init int create_proc_service_level(void)
{
proc_create_seq("service_levels", 0, NULL, &service_level_seq_ops);
if (MACHINE_IS_VM)
register_service_level(&service_level_vm);
return 0;
}
subsys_initcall(create_proc_service_level);
/*
* CPU capability might have changed. Therefore recalculate loops_per_jiffy.
*/
void s390_adjust_jiffies(void)
{
struct sysinfo_1_2_2 *info;
unsigned long capability;
struct kernel_fpu fpu;
info = (void *) get_zeroed_page(GFP_KERNEL);
if (!info)
return;
if (stsi(info, 1, 2, 2) == 0) {
/*
* Major sigh. The cpu capability encoding is "special".
* If the first 9 bits of info->capability are 0 then it
* is a 32 bit unsigned integer in the range 0 .. 2^23.
* If the first 9 bits are != 0 then it is a 32 bit float.
* In addition a lower value indicates a proportionally
* higher cpu capacity. Bogomips are the other way round.
* To get to a halfway suitable number we divide 1e7
* by the cpu capability number. Yes, that means a floating
* point division ..
*/
kernel_fpu_begin(&fpu, KERNEL_FPR);
asm volatile(
" sfpc %3\n"
" l %0,%1\n"
" tmlh %0,0xff80\n"
" jnz 0f\n"
" cefbr %%f2,%0\n"
" j 1f\n"
"0: le %%f2,%1\n"
"1: cefbr %%f0,%2\n"
" debr %%f0,%%f2\n"
" cgebr %0,5,%%f0\n"
: "=&d" (capability)
: "Q" (info->capability), "d" (10000000), "d" (0)
: "cc"
);
kernel_fpu_end(&fpu, KERNEL_FPR);
} else
/*
* Really old machine without stsi block for basic
* cpu information. Report 42.0 bogomips.
*/
capability = 42;
loops_per_jiffy = capability * (500000/HZ);
free_page((unsigned long) info);
}
/*
* calibrate the delay loop
*/
void calibrate_delay(void)
{
s390_adjust_jiffies();
/* Print the good old Bogomips line .. */
printk(KERN_DEBUG "Calibrating delay loop (skipped)... "
"%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ),
(loops_per_jiffy/(5000/HZ)) % 100);
}
#ifdef CONFIG_DEBUG_FS
#define STSI_FILE(fc, s1, s2) \
static int stsi_open_##fc##_##s1##_##s2(struct inode *inode, struct file *file)\
{ \
file->private_data = (void *) get_zeroed_page(GFP_KERNEL); \
if (!file->private_data) \
return -ENOMEM; \
if (stsi(file->private_data, fc, s1, s2)) { \
free_page((unsigned long)file->private_data); \
file->private_data = NULL; \
return -EACCES; \
} \
return nonseekable_open(inode, file); \
} \
\
static const struct file_operations stsi_##fc##_##s1##_##s2##_fs_ops = { \
.open = stsi_open_##fc##_##s1##_##s2, \
.release = stsi_release, \
.read = stsi_read, \
.llseek = no_llseek, \
};
static int stsi_release(struct inode *inode, struct file *file)
{
free_page((unsigned long)file->private_data);
return 0;
}
static ssize_t stsi_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
{
return simple_read_from_buffer(buf, size, ppos, file->private_data, PAGE_SIZE);
}
STSI_FILE( 1, 1, 1);
STSI_FILE( 1, 2, 1);
STSI_FILE( 1, 2, 2);
STSI_FILE( 2, 2, 1);
STSI_FILE( 2, 2, 2);
STSI_FILE( 3, 2, 2);
STSI_FILE(15, 1, 2);
STSI_FILE(15, 1, 3);
STSI_FILE(15, 1, 4);
STSI_FILE(15, 1, 5);
STSI_FILE(15, 1, 6);
struct stsi_file {
const struct file_operations *fops;
char *name;
};
static struct stsi_file stsi_file[] __initdata = {
{.fops = &stsi_1_1_1_fs_ops, .name = "1_1_1"},
{.fops = &stsi_1_2_1_fs_ops, .name = "1_2_1"},
{.fops = &stsi_1_2_2_fs_ops, .name = "1_2_2"},
{.fops = &stsi_2_2_1_fs_ops, .name = "2_2_1"},
{.fops = &stsi_2_2_2_fs_ops, .name = "2_2_2"},
{.fops = &stsi_3_2_2_fs_ops, .name = "3_2_2"},
{.fops = &stsi_15_1_2_fs_ops, .name = "15_1_2"},
{.fops = &stsi_15_1_3_fs_ops, .name = "15_1_3"},
{.fops = &stsi_15_1_4_fs_ops, .name = "15_1_4"},
{.fops = &stsi_15_1_5_fs_ops, .name = "15_1_5"},
{.fops = &stsi_15_1_6_fs_ops, .name = "15_1_6"},
};
static u8 stsi_0_0_0;
static __init int stsi_init_debugfs(void)
{
struct dentry *stsi_root;
struct stsi_file *sf;
int lvl, i;
stsi_root = debugfs_create_dir("stsi", arch_debugfs_dir);
if (IS_ERR_OR_NULL(stsi_root))
return 0;
lvl = stsi(NULL, 0, 0, 0);
if (lvl > 0)
stsi_0_0_0 = lvl;
debugfs_create_u8("0_0_0", 0400, stsi_root, &stsi_0_0_0);
for (i = 0; i < ARRAY_SIZE(stsi_file); i++) {
sf = &stsi_file[i];
debugfs_create_file(sf->name, 0400, stsi_root, NULL, sf->fops);
}
if (IS_ENABLED(CONFIG_SCHED_TOPOLOGY) && MACHINE_HAS_TOPOLOGY) {
char link_to[10];
sprintf(link_to, "15_1_%d", topology_mnest_limit());
debugfs_create_symlink("topology", stsi_root, link_to);
}
return 0;
}
device_initcall(stsi_init_debugfs);
#endif /* CONFIG_DEBUG_FS */
+923
View File
@@ -0,0 +1,923 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Time of day based timer functions.
*
* S390 version
* Copyright IBM Corp. 1999, 2008
* Author(s): Hartmut Penner (hp@de.ibm.com),
* Martin Schwidefsky (schwidefsky@de.ibm.com),
* Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
*
* Derived from "arch/i386/kernel/time.c"
* Copyright (C) 1991, 1992, 1995 Linus Torvalds
*/
#define KMSG_COMPONENT "time"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/kernel_stat.h>
#include <linux/errno.h>
#include <linux/export.h>
#include <linux/sched.h>
#include <linux/sched/clock.h>
#include <linux/kernel.h>
#include <linux/param.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/interrupt.h>
#include <linux/cpu.h>
#include <linux/stop_machine.h>
#include <linux/time.h>
#include <linux/device.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/smp.h>
#include <linux/types.h>
#include <linux/profile.h>
#include <linux/timex.h>
#include <linux/notifier.h>
#include <linux/timekeeper_internal.h>
#include <linux/clockchips.h>
#include <linux/gfp.h>
#include <linux/kprobes.h>
#include <linux/uaccess.h>
#include <asm/facility.h>
#include <asm/delay.h>
#include <asm/div64.h>
#include <asm/vdso.h>
#include <asm/irq.h>
#include <asm/irq_regs.h>
#include <asm/vtimer.h>
#include <asm/stp.h>
#include <asm/cio.h>
#include "entry.h"
unsigned char tod_clock_base[16] __aligned(8) = {
/* Force to data section. */
0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff
};
EXPORT_SYMBOL_GPL(tod_clock_base);
u64 clock_comparator_max = -1ULL;
EXPORT_SYMBOL_GPL(clock_comparator_max);
static DEFINE_PER_CPU(struct clock_event_device, comparators);
ATOMIC_NOTIFIER_HEAD(s390_epoch_delta_notifier);
EXPORT_SYMBOL(s390_epoch_delta_notifier);
unsigned char ptff_function_mask[16];
static unsigned long long lpar_offset;
static unsigned long long initial_leap_seconds;
static unsigned long long tod_steering_end;
static long long tod_steering_delta;
/*
* Get time offsets with PTFF
*/
void __init time_early_init(void)
{
struct ptff_qto qto;
struct ptff_qui qui;
/* Initialize TOD steering parameters */
tod_steering_end = *(unsigned long long *) &tod_clock_base[1];
vdso_data->ts_end = tod_steering_end;
if (!test_facility(28))
return;
ptff(&ptff_function_mask, sizeof(ptff_function_mask), PTFF_QAF);
/* get LPAR offset */
if (ptff_query(PTFF_QTO) && ptff(&qto, sizeof(qto), PTFF_QTO) == 0)
lpar_offset = qto.tod_epoch_difference;
/* get initial leap seconds */
if (ptff_query(PTFF_QUI) && ptff(&qui, sizeof(qui), PTFF_QUI) == 0)
initial_leap_seconds = (unsigned long long)
((long) qui.old_leap * 4096000000L);
}
/*
* Scheduler clock - returns current time in nanosec units.
*/
unsigned long long notrace sched_clock(void)
{
return tod_to_ns(get_tod_clock_monotonic());
}
NOKPROBE_SYMBOL(sched_clock);
/*
* Monotonic_clock - returns # of nanoseconds passed since time_init()
*/
unsigned long long monotonic_clock(void)
{
return sched_clock();
}
EXPORT_SYMBOL(monotonic_clock);
static void ext_to_timespec64(unsigned char *clk, struct timespec64 *xt)
{
unsigned long long high, low, rem, sec, nsec;
/* Split extendnd TOD clock to micro-seconds and sub-micro-seconds */
high = (*(unsigned long long *) clk) >> 4;
low = (*(unsigned long long *)&clk[7]) << 4;
/* Calculate seconds and nano-seconds */
sec = high;
rem = do_div(sec, 1000000);
nsec = (((low >> 32) + (rem << 32)) * 1000) >> 32;
xt->tv_sec = sec;
xt->tv_nsec = nsec;
}
void clock_comparator_work(void)
{
struct clock_event_device *cd;
S390_lowcore.clock_comparator = clock_comparator_max;
cd = this_cpu_ptr(&comparators);
cd->event_handler(cd);
}
static int s390_next_event(unsigned long delta,
struct clock_event_device *evt)
{
S390_lowcore.clock_comparator = get_tod_clock() + delta;
set_clock_comparator(S390_lowcore.clock_comparator);
return 0;
}
/*
* Set up lowcore and control register of the current cpu to
* enable TOD clock and clock comparator interrupts.
*/
void init_cpu_timer(void)
{
struct clock_event_device *cd;
int cpu;
S390_lowcore.clock_comparator = clock_comparator_max;
set_clock_comparator(S390_lowcore.clock_comparator);
cpu = smp_processor_id();
cd = &per_cpu(comparators, cpu);
cd->name = "comparator";
cd->features = CLOCK_EVT_FEAT_ONESHOT;
cd->mult = 16777;
cd->shift = 12;
cd->min_delta_ns = 1;
cd->min_delta_ticks = 1;
cd->max_delta_ns = LONG_MAX;
cd->max_delta_ticks = ULONG_MAX;
cd->rating = 400;
cd->cpumask = cpumask_of(cpu);
cd->set_next_event = s390_next_event;
clockevents_register_device(cd);
/* Enable clock comparator timer interrupt. */
__ctl_set_bit(0,11);
/* Always allow the timing alert external interrupt. */
__ctl_set_bit(0, 4);
}
static void clock_comparator_interrupt(struct ext_code ext_code,
unsigned int param32,
unsigned long param64)
{
inc_irq_stat(IRQEXT_CLK);
if (S390_lowcore.clock_comparator == clock_comparator_max)
set_clock_comparator(S390_lowcore.clock_comparator);
}
static void stp_timing_alert(struct stp_irq_parm *);
static void timing_alert_interrupt(struct ext_code ext_code,
unsigned int param32, unsigned long param64)
{
inc_irq_stat(IRQEXT_TLA);
if (param32 & 0x00038000)
stp_timing_alert((struct stp_irq_parm *) &param32);
}
static void stp_reset(void);
void read_persistent_clock64(struct timespec64 *ts)
{
unsigned char clk[STORE_CLOCK_EXT_SIZE];
__u64 delta;
delta = initial_leap_seconds + TOD_UNIX_EPOCH;
get_tod_clock_ext(clk);
*(__u64 *) &clk[1] -= delta;
if (*(__u64 *) &clk[1] > delta)
clk[0]--;
ext_to_timespec64(clk, ts);
}
void __init read_persistent_wall_and_boot_offset(struct timespec64 *wall_time,
struct timespec64 *boot_offset)
{
unsigned char clk[STORE_CLOCK_EXT_SIZE];
struct timespec64 boot_time;
__u64 delta;
delta = initial_leap_seconds + TOD_UNIX_EPOCH;
memcpy(clk, tod_clock_base, STORE_CLOCK_EXT_SIZE);
*(__u64 *)&clk[1] -= delta;
if (*(__u64 *)&clk[1] > delta)
clk[0]--;
ext_to_timespec64(clk, &boot_time);
read_persistent_clock64(wall_time);
*boot_offset = timespec64_sub(*wall_time, boot_time);
}
static u64 read_tod_clock(struct clocksource *cs)
{
unsigned long long now, adj;
preempt_disable(); /* protect from changes to steering parameters */
now = get_tod_clock();
adj = tod_steering_end - now;
if (unlikely((s64) adj >= 0))
/*
* manually steer by 1 cycle every 2^16 cycles. This
* corresponds to shifting the tod delta by 15. 1s is
* therefore steered in ~9h. The adjust will decrease
* over time, until it finally reaches 0.
*/
now += (tod_steering_delta < 0) ? (adj >> 15) : -(adj >> 15);
preempt_enable();
return now;
}
static struct clocksource clocksource_tod = {
.name = "tod",
.rating = 400,
.read = read_tod_clock,
.mask = -1ULL,
.mult = 1000,
.shift = 12,
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
};
struct clocksource * __init clocksource_default_clock(void)
{
return &clocksource_tod;
}
void update_vsyscall(struct timekeeper *tk)
{
u64 nsecps;
if (tk->tkr_mono.clock != &clocksource_tod)
return;
/* Make userspace gettimeofday spin until we're done. */
++vdso_data->tb_update_count;
smp_wmb();
vdso_data->xtime_tod_stamp = tk->tkr_mono.cycle_last;
vdso_data->xtime_clock_sec = tk->xtime_sec;
vdso_data->xtime_clock_nsec = tk->tkr_mono.xtime_nsec;
vdso_data->wtom_clock_sec =
tk->xtime_sec + tk->wall_to_monotonic.tv_sec;
vdso_data->wtom_clock_nsec = tk->tkr_mono.xtime_nsec +
+ ((u64) tk->wall_to_monotonic.tv_nsec << tk->tkr_mono.shift);
nsecps = (u64) NSEC_PER_SEC << tk->tkr_mono.shift;
while (vdso_data->wtom_clock_nsec >= nsecps) {
vdso_data->wtom_clock_nsec -= nsecps;
vdso_data->wtom_clock_sec++;
}
vdso_data->xtime_coarse_sec = tk->xtime_sec;
vdso_data->xtime_coarse_nsec =
(long)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift);
vdso_data->wtom_coarse_sec =
vdso_data->xtime_coarse_sec + tk->wall_to_monotonic.tv_sec;
vdso_data->wtom_coarse_nsec =
vdso_data->xtime_coarse_nsec + tk->wall_to_monotonic.tv_nsec;
while (vdso_data->wtom_coarse_nsec >= NSEC_PER_SEC) {
vdso_data->wtom_coarse_nsec -= NSEC_PER_SEC;
vdso_data->wtom_coarse_sec++;
}
vdso_data->tk_mult = tk->tkr_mono.mult;
vdso_data->tk_shift = tk->tkr_mono.shift;
vdso_data->hrtimer_res = hrtimer_resolution;
smp_wmb();
++vdso_data->tb_update_count;
}
extern struct timezone sys_tz;
void update_vsyscall_tz(void)
{
vdso_data->tz_minuteswest = sys_tz.tz_minuteswest;
vdso_data->tz_dsttime = sys_tz.tz_dsttime;
}
/*
* Initialize the TOD clock and the CPU timer of
* the boot cpu.
*/
void __init time_init(void)
{
/* Reset time synchronization interfaces. */
stp_reset();
/* request the clock comparator external interrupt */
if (register_external_irq(EXT_IRQ_CLK_COMP, clock_comparator_interrupt))
panic("Couldn't request external interrupt 0x1004");
/* request the timing alert external interrupt */
if (register_external_irq(EXT_IRQ_TIMING_ALERT, timing_alert_interrupt))
panic("Couldn't request external interrupt 0x1406");
if (__clocksource_register(&clocksource_tod) != 0)
panic("Could not register TOD clock source");
/* Enable TOD clock interrupts on the boot cpu. */
init_cpu_timer();
/* Enable cpu timer interrupts on the boot cpu. */
vtime_init();
}
static DEFINE_PER_CPU(atomic_t, clock_sync_word);
static DEFINE_MUTEX(clock_sync_mutex);
static unsigned long clock_sync_flags;
#define CLOCK_SYNC_HAS_STP 0
#define CLOCK_SYNC_STP 1
#define CLOCK_SYNC_STPINFO_VALID 2
/*
* The get_clock function for the physical clock. It will get the current
* TOD clock, subtract the LPAR offset and write the result to *clock.
* The function returns 0 if the clock is in sync with the external time
* source. If the clock mode is local it will return -EOPNOTSUPP and
* -EAGAIN if the clock is not in sync with the external reference.
*/
int get_phys_clock(unsigned long *clock)
{
atomic_t *sw_ptr;
unsigned int sw0, sw1;
sw_ptr = &get_cpu_var(clock_sync_word);
sw0 = atomic_read(sw_ptr);
*clock = get_tod_clock() - lpar_offset;
sw1 = atomic_read(sw_ptr);
put_cpu_var(clock_sync_word);
if (sw0 == sw1 && (sw0 & 0x80000000U))
/* Success: time is in sync. */
return 0;
if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags))
return -EOPNOTSUPP;
if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags))
return -EACCES;
return -EAGAIN;
}
EXPORT_SYMBOL(get_phys_clock);
/*
* Make get_phys_clock() return -EAGAIN.
*/
static void disable_sync_clock(void *dummy)
{
atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word);
/*
* Clear the in-sync bit 2^31. All get_phys_clock calls will
* fail until the sync bit is turned back on. In addition
* increase the "sequence" counter to avoid the race of an
* stp event and the complete recovery against get_phys_clock.
*/
atomic_andnot(0x80000000, sw_ptr);
atomic_inc(sw_ptr);
}
/*
* Make get_phys_clock() return 0 again.
* Needs to be called from a context disabled for preemption.
*/
static void enable_sync_clock(void)
{
atomic_t *sw_ptr = this_cpu_ptr(&clock_sync_word);
atomic_or(0x80000000, sw_ptr);
}
/*
* Function to check if the clock is in sync.
*/
static inline int check_sync_clock(void)
{
atomic_t *sw_ptr;
int rc;
sw_ptr = &get_cpu_var(clock_sync_word);
rc = (atomic_read(sw_ptr) & 0x80000000U) != 0;
put_cpu_var(clock_sync_word);
return rc;
}
/*
* Apply clock delta to the global data structures.
* This is called once on the CPU that performed the clock sync.
*/
static void clock_sync_global(unsigned long long delta)
{
unsigned long now, adj;
struct ptff_qto qto;
/* Fixup the monotonic sched clock. */
*(unsigned long long *) &tod_clock_base[1] += delta;
if (*(unsigned long long *) &tod_clock_base[1] < delta)
/* Epoch overflow */
tod_clock_base[0]++;
/* Adjust TOD steering parameters. */
vdso_data->tb_update_count++;
now = get_tod_clock();
adj = tod_steering_end - now;
if (unlikely((s64) adj >= 0))
/* Calculate how much of the old adjustment is left. */
tod_steering_delta = (tod_steering_delta < 0) ?
-(adj >> 15) : (adj >> 15);
tod_steering_delta += delta;
if ((abs(tod_steering_delta) >> 48) != 0)
panic("TOD clock sync offset %lli is too large to drift\n",
tod_steering_delta);
tod_steering_end = now + (abs(tod_steering_delta) << 15);
vdso_data->ts_dir = (tod_steering_delta < 0) ? 0 : 1;
vdso_data->ts_end = tod_steering_end;
vdso_data->tb_update_count++;
/* Update LPAR offset. */
if (ptff_query(PTFF_QTO) && ptff(&qto, sizeof(qto), PTFF_QTO) == 0)
lpar_offset = qto.tod_epoch_difference;
/* Call the TOD clock change notifier. */
atomic_notifier_call_chain(&s390_epoch_delta_notifier, 0, &delta);
}
/*
* Apply clock delta to the per-CPU data structures of this CPU.
* This is called for each online CPU after the call to clock_sync_global.
*/
static void clock_sync_local(unsigned long long delta)
{
/* Add the delta to the clock comparator. */
if (S390_lowcore.clock_comparator != clock_comparator_max) {
S390_lowcore.clock_comparator += delta;
set_clock_comparator(S390_lowcore.clock_comparator);
}
/* Adjust the last_update_clock time-stamp. */
S390_lowcore.last_update_clock += delta;
}
/* Single threaded workqueue used for stp sync events */
static struct workqueue_struct *time_sync_wq;
static void __init time_init_wq(void)
{
if (time_sync_wq)
return;
time_sync_wq = create_singlethread_workqueue("timesync");
}
struct clock_sync_data {
atomic_t cpus;
int in_sync;
unsigned long long clock_delta;
};
/*
* Server Time Protocol (STP) code.
*/
static bool stp_online;
static struct stp_sstpi stp_info;
static void *stp_page;
static void stp_work_fn(struct work_struct *work);
static DEFINE_MUTEX(stp_work_mutex);
static DECLARE_WORK(stp_work, stp_work_fn);
static struct timer_list stp_timer;
static int __init early_parse_stp(char *p)
{
return kstrtobool(p, &stp_online);
}
early_param("stp", early_parse_stp);
/*
* Reset STP attachment.
*/
static void __init stp_reset(void)
{
int rc;
stp_page = (void *) get_zeroed_page(GFP_ATOMIC);
rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000, NULL);
if (rc == 0)
set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags);
else if (stp_online) {
pr_warn("The real or virtual hardware system does not provide an STP interface\n");
free_page((unsigned long) stp_page);
stp_page = NULL;
stp_online = false;
}
}
static void stp_timeout(struct timer_list *unused)
{
queue_work(time_sync_wq, &stp_work);
}
static int __init stp_init(void)
{
if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags))
return 0;
timer_setup(&stp_timer, stp_timeout, 0);
time_init_wq();
if (!stp_online)
return 0;
queue_work(time_sync_wq, &stp_work);
return 0;
}
arch_initcall(stp_init);
/*
* STP timing alert. There are three causes:
* 1) timing status change
* 2) link availability change
* 3) time control parameter change
* In all three cases we are only interested in the clock source state.
* If a STP clock source is now available use it.
*/
static void stp_timing_alert(struct stp_irq_parm *intparm)
{
if (intparm->tsc || intparm->lac || intparm->tcpc)
queue_work(time_sync_wq, &stp_work);
}
/*
* STP sync check machine check. This is called when the timing state
* changes from the synchronized state to the unsynchronized state.
* After a STP sync check the clock is not in sync. The machine check
* is broadcasted to all cpus at the same time.
*/
int stp_sync_check(void)
{
disable_sync_clock(NULL);
return 1;
}
/*
* STP island condition machine check. This is called when an attached
* server attempts to communicate over an STP link and the servers
* have matching CTN ids and have a valid stratum-1 configuration
* but the configurations do not match.
*/
int stp_island_check(void)
{
disable_sync_clock(NULL);
return 1;
}
void stp_queue_work(void)
{
queue_work(time_sync_wq, &stp_work);
}
static int __store_stpinfo(void)
{
int rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi));
if (rc)
clear_bit(CLOCK_SYNC_STPINFO_VALID, &clock_sync_flags);
else
set_bit(CLOCK_SYNC_STPINFO_VALID, &clock_sync_flags);
return rc;
}
static int stpinfo_valid(void)
{
return stp_online && test_bit(CLOCK_SYNC_STPINFO_VALID, &clock_sync_flags);
}
static int stp_sync_clock(void *data)
{
struct clock_sync_data *sync = data;
unsigned long long clock_delta;
static int first;
int rc;
enable_sync_clock();
if (xchg(&first, 1) == 0) {
/* Wait until all other cpus entered the sync function. */
while (atomic_read(&sync->cpus) != 0)
cpu_relax();
rc = 0;
if (stp_info.todoff[0] || stp_info.todoff[1] ||
stp_info.todoff[2] || stp_info.todoff[3] ||
stp_info.tmd != 2) {
rc = chsc_sstpc(stp_page, STP_OP_SYNC, 0,
&clock_delta);
if (rc == 0) {
sync->clock_delta = clock_delta;
clock_sync_global(clock_delta);
rc = __store_stpinfo();
if (rc == 0 && stp_info.tmd != 2)
rc = -EAGAIN;
}
}
sync->in_sync = rc ? -EAGAIN : 1;
xchg(&first, 0);
} else {
/* Slave */
atomic_dec(&sync->cpus);
/* Wait for in_sync to be set. */
while (READ_ONCE(sync->in_sync) == 0)
__udelay(1);
}
if (sync->in_sync != 1)
/* Didn't work. Clear per-cpu in sync bit again. */
disable_sync_clock(NULL);
/* Apply clock delta to per-CPU fields of this CPU. */
clock_sync_local(sync->clock_delta);
return 0;
}
/*
* STP work. Check for the STP state and take over the clock
* synchronization if the STP clock source is usable.
*/
static void stp_work_fn(struct work_struct *work)
{
struct clock_sync_data stp_sync;
int rc;
/* prevent multiple execution. */
mutex_lock(&stp_work_mutex);
if (!stp_online) {
chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000, NULL);
del_timer_sync(&stp_timer);
goto out_unlock;
}
rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0, NULL);
if (rc)
goto out_unlock;
rc = __store_stpinfo();
if (rc || stp_info.c == 0)
goto out_unlock;
/* Skip synchronization if the clock is already in sync. */
if (check_sync_clock())
goto out_unlock;
memset(&stp_sync, 0, sizeof(stp_sync));
cpus_read_lock();
atomic_set(&stp_sync.cpus, num_online_cpus() - 1);
stop_machine_cpuslocked(stp_sync_clock, &stp_sync, cpu_online_mask);
cpus_read_unlock();
if (!check_sync_clock())
/*
* There is a usable clock but the synchonization failed.
* Retry after a second.
*/
mod_timer(&stp_timer, jiffies + HZ);
out_unlock:
mutex_unlock(&stp_work_mutex);
}
/*
* STP subsys sysfs interface functions
*/
static struct bus_type stp_subsys = {
.name = "stp",
.dev_name = "stp",
};
static ssize_t stp_ctn_id_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid())
ret = sprintf(buf, "%016llx\n",
*(unsigned long long *) stp_info.ctnid);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(ctn_id, 0400, stp_ctn_id_show, NULL);
static ssize_t stp_ctn_type_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid())
ret = sprintf(buf, "%i\n", stp_info.ctn);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(ctn_type, 0400, stp_ctn_type_show, NULL);
static ssize_t stp_dst_offset_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid() && (stp_info.vbits & 0x2000))
ret = sprintf(buf, "%i\n", (int)(s16) stp_info.dsto);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(dst_offset, 0400, stp_dst_offset_show, NULL);
static ssize_t stp_leap_seconds_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid() && (stp_info.vbits & 0x8000))
ret = sprintf(buf, "%i\n", (int)(s16) stp_info.leaps);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(leap_seconds, 0400, stp_leap_seconds_show, NULL);
static ssize_t stp_stratum_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid())
ret = sprintf(buf, "%i\n", (int)(s16) stp_info.stratum);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(stratum, 0400, stp_stratum_show, NULL);
static ssize_t stp_time_offset_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid() && (stp_info.vbits & 0x0800))
ret = sprintf(buf, "%i\n", (int) stp_info.tto);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(time_offset, 0400, stp_time_offset_show, NULL);
static ssize_t stp_time_zone_offset_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid() && (stp_info.vbits & 0x4000))
ret = sprintf(buf, "%i\n", (int)(s16) stp_info.tzo);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(time_zone_offset, 0400,
stp_time_zone_offset_show, NULL);
static ssize_t stp_timing_mode_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid())
ret = sprintf(buf, "%i\n", stp_info.tmd);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(timing_mode, 0400, stp_timing_mode_show, NULL);
static ssize_t stp_timing_state_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t ret = -ENODATA;
mutex_lock(&stp_work_mutex);
if (stpinfo_valid())
ret = sprintf(buf, "%i\n", stp_info.tst);
mutex_unlock(&stp_work_mutex);
return ret;
}
static DEVICE_ATTR(timing_state, 0400, stp_timing_state_show, NULL);
static ssize_t stp_online_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
return sprintf(buf, "%i\n", stp_online);
}
static ssize_t stp_online_store(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
unsigned int value;
value = simple_strtoul(buf, NULL, 0);
if (value != 0 && value != 1)
return -EINVAL;
if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags))
return -EOPNOTSUPP;
mutex_lock(&clock_sync_mutex);
stp_online = value;
if (stp_online)
set_bit(CLOCK_SYNC_STP, &clock_sync_flags);
else
clear_bit(CLOCK_SYNC_STP, &clock_sync_flags);
queue_work(time_sync_wq, &stp_work);
mutex_unlock(&clock_sync_mutex);
return count;
}
/*
* Can't use DEVICE_ATTR because the attribute should be named
* stp/online but dev_attr_online already exists in this file ..
*/
static struct device_attribute dev_attr_stp_online = {
.attr = { .name = "online", .mode = 0600 },
.show = stp_online_show,
.store = stp_online_store,
};
static struct device_attribute *stp_attributes[] = {
&dev_attr_ctn_id,
&dev_attr_ctn_type,
&dev_attr_dst_offset,
&dev_attr_leap_seconds,
&dev_attr_stp_online,
&dev_attr_stratum,
&dev_attr_time_offset,
&dev_attr_time_zone_offset,
&dev_attr_timing_mode,
&dev_attr_timing_state,
NULL
};
static int __init stp_init_sysfs(void)
{
struct device_attribute **attr;
int rc;
rc = subsys_system_register(&stp_subsys, NULL);
if (rc)
goto out;
for (attr = stp_attributes; *attr; attr++) {
rc = device_create_file(stp_subsys.dev_root, *attr);
if (rc)
goto out_unreg;
}
return 0;
out_unreg:
for (; attr >= stp_attributes; attr--)
device_remove_file(stp_subsys.dev_root, *attr);
bus_unregister(&stp_subsys);
out:
return rc;
}
device_initcall(stp_init_sysfs);
+641
View File
@@ -0,0 +1,641 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright IBM Corp. 2007, 2011
* Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
*/
#define KMSG_COMPONENT "cpu"
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#include <linux/workqueue.h>
#include <linux/bootmem.h>
#include <linux/uaccess.h>
#include <linux/sysctl.h>
#include <linux/cpuset.h>
#include <linux/device.h>
#include <linux/export.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/sched/topology.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/cpu.h>
#include <linux/smp.h>
#include <linux/mm.h>
#include <linux/nodemask.h>
#include <linux/node.h>
#include <asm/sysinfo.h>
#include <asm/numa.h>
#define PTF_HORIZONTAL (0UL)
#define PTF_VERTICAL (1UL)
#define PTF_CHECK (2UL)
enum {
TOPOLOGY_MODE_HW,
TOPOLOGY_MODE_SINGLE,
TOPOLOGY_MODE_PACKAGE,
TOPOLOGY_MODE_UNINITIALIZED
};
struct mask_info {
struct mask_info *next;
unsigned char id;
cpumask_t mask;
};
static int topology_mode = TOPOLOGY_MODE_UNINITIALIZED;
static void set_topology_timer(void);
static void topology_work_fn(struct work_struct *work);
static struct sysinfo_15_1_x *tl_info;
static DECLARE_WORK(topology_work, topology_work_fn);
/*
* Socket/Book linked lists and cpu_topology updates are
* protected by "sched_domains_mutex".
*/
static struct mask_info socket_info;
static struct mask_info book_info;
static struct mask_info drawer_info;
struct cpu_topology_s390 cpu_topology[NR_CPUS];
EXPORT_SYMBOL_GPL(cpu_topology);
cpumask_t cpus_with_topology;
static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu)
{
cpumask_t mask;
cpumask_copy(&mask, cpumask_of(cpu));
switch (topology_mode) {
case TOPOLOGY_MODE_HW:
while (info) {
if (cpumask_test_cpu(cpu, &info->mask)) {
mask = info->mask;
break;
}
info = info->next;
}
if (cpumask_empty(&mask))
cpumask_copy(&mask, cpumask_of(cpu));
break;
case TOPOLOGY_MODE_PACKAGE:
cpumask_copy(&mask, cpu_present_mask);
break;
default:
/* fallthrough */
case TOPOLOGY_MODE_SINGLE:
cpumask_copy(&mask, cpumask_of(cpu));
break;
}
return mask;
}
static cpumask_t cpu_thread_map(unsigned int cpu)
{
cpumask_t mask;
int i;
cpumask_copy(&mask, cpumask_of(cpu));
if (topology_mode != TOPOLOGY_MODE_HW)
return mask;
cpu -= cpu % (smp_cpu_mtid + 1);
for (i = 0; i <= smp_cpu_mtid; i++)
if (cpu_present(cpu + i))
cpumask_set_cpu(cpu + i, &mask);
return mask;
}
#define TOPOLOGY_CORE_BITS 64
static void add_cpus_to_mask(struct topology_core *tl_core,
struct mask_info *drawer,
struct mask_info *book,
struct mask_info *socket)
{
struct cpu_topology_s390 *topo;
unsigned int core;
for_each_set_bit(core, &tl_core->mask, TOPOLOGY_CORE_BITS) {
unsigned int rcore;
int lcpu, i;
rcore = TOPOLOGY_CORE_BITS - 1 - core + tl_core->origin;
lcpu = smp_find_processor_id(rcore << smp_cpu_mt_shift);
if (lcpu < 0)
continue;
for (i = 0; i <= smp_cpu_mtid; i++) {
topo = &cpu_topology[lcpu + i];
topo->drawer_id = drawer->id;
topo->book_id = book->id;
topo->socket_id = socket->id;
topo->core_id = rcore;
topo->thread_id = lcpu + i;
topo->dedicated = tl_core->d;
cpumask_set_cpu(lcpu + i, &drawer->mask);
cpumask_set_cpu(lcpu + i, &book->mask);
cpumask_set_cpu(lcpu + i, &socket->mask);
cpumask_set_cpu(lcpu + i, &cpus_with_topology);
smp_cpu_set_polarization(lcpu + i, tl_core->pp);
}
}
}
static void clear_masks(void)
{
struct mask_info *info;
info = &socket_info;
while (info) {
cpumask_clear(&info->mask);
info = info->next;
}
info = &book_info;
while (info) {
cpumask_clear(&info->mask);
info = info->next;
}
info = &drawer_info;
while (info) {
cpumask_clear(&info->mask);
info = info->next;
}
}
static union topology_entry *next_tle(union topology_entry *tle)
{
if (!tle->nl)
return (union topology_entry *)((struct topology_core *)tle + 1);
return (union topology_entry *)((struct topology_container *)tle + 1);
}
static void tl_to_masks(struct sysinfo_15_1_x *info)
{
struct mask_info *socket = &socket_info;
struct mask_info *book = &book_info;
struct mask_info *drawer = &drawer_info;
union topology_entry *tle, *end;
clear_masks();
tle = info->tle;
end = (union topology_entry *)((unsigned long)info + info->length);
while (tle < end) {
switch (tle->nl) {
case 3:
drawer = drawer->next;
drawer->id = tle->container.id;
break;
case 2:
book = book->next;
book->id = tle->container.id;
break;
case 1:
socket = socket->next;
socket->id = tle->container.id;
break;
case 0:
add_cpus_to_mask(&tle->cpu, drawer, book, socket);
break;
default:
clear_masks();
return;
}
tle = next_tle(tle);
}
}
static void topology_update_polarization_simple(void)
{
int cpu;
for_each_possible_cpu(cpu)
smp_cpu_set_polarization(cpu, POLARIZATION_HRZ);
}
static int ptf(unsigned long fc)
{
int rc;
asm volatile(
" .insn rre,0xb9a20000,%1,%1\n"
" ipm %0\n"
" srl %0,28\n"
: "=d" (rc)
: "d" (fc) : "cc");
return rc;
}
int topology_set_cpu_management(int fc)
{
int cpu, rc;
if (!MACHINE_HAS_TOPOLOGY)
return -EOPNOTSUPP;
if (fc)
rc = ptf(PTF_VERTICAL);
else
rc = ptf(PTF_HORIZONTAL);
if (rc)
return -EBUSY;
for_each_possible_cpu(cpu)
smp_cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
return rc;
}
static void update_cpu_masks(void)
{
struct cpu_topology_s390 *topo;
int cpu, id;
for_each_possible_cpu(cpu) {
topo = &cpu_topology[cpu];
topo->thread_mask = cpu_thread_map(cpu);
topo->core_mask = cpu_group_map(&socket_info, cpu);
topo->book_mask = cpu_group_map(&book_info, cpu);
topo->drawer_mask = cpu_group_map(&drawer_info, cpu);
if (topology_mode != TOPOLOGY_MODE_HW) {
id = topology_mode == TOPOLOGY_MODE_PACKAGE ? 0 : cpu;
topo->thread_id = cpu;
topo->core_id = cpu;
topo->socket_id = id;
topo->book_id = id;
topo->drawer_id = id;
if (cpu_present(cpu))
cpumask_set_cpu(cpu, &cpus_with_topology);
}
}
numa_update_cpu_topology();
}
void store_topology(struct sysinfo_15_1_x *info)
{
stsi(info, 15, 1, topology_mnest_limit());
}
static void __arch_update_dedicated_flag(void *arg)
{
if (topology_cpu_dedicated(smp_processor_id()))
set_cpu_flag(CIF_DEDICATED_CPU);
else
clear_cpu_flag(CIF_DEDICATED_CPU);
}
static int __arch_update_cpu_topology(void)
{
struct sysinfo_15_1_x *info = tl_info;
int rc = 0;
mutex_lock(&smp_cpu_state_mutex);
cpumask_clear(&cpus_with_topology);
if (MACHINE_HAS_TOPOLOGY) {
rc = 1;
store_topology(info);
tl_to_masks(info);
}
update_cpu_masks();
if (!MACHINE_HAS_TOPOLOGY)
topology_update_polarization_simple();
mutex_unlock(&smp_cpu_state_mutex);
return rc;
}
int arch_update_cpu_topology(void)
{
struct device *dev;
int cpu, rc;
rc = __arch_update_cpu_topology();
on_each_cpu(__arch_update_dedicated_flag, NULL, 0);
for_each_online_cpu(cpu) {
dev = get_cpu_device(cpu);
if (dev)
kobject_uevent(&dev->kobj, KOBJ_CHANGE);
}
return rc;
}
static void topology_work_fn(struct work_struct *work)
{
rebuild_sched_domains();
}
void topology_schedule_update(void)
{
schedule_work(&topology_work);
}
static void topology_flush_work(void)
{
flush_work(&topology_work);
}
static void topology_timer_fn(struct timer_list *unused)
{
if (ptf(PTF_CHECK))
topology_schedule_update();
set_topology_timer();
}
static struct timer_list topology_timer;
static atomic_t topology_poll = ATOMIC_INIT(0);
static void set_topology_timer(void)
{
if (atomic_add_unless(&topology_poll, -1, 0))
mod_timer(&topology_timer, jiffies + HZ / 10);
else
mod_timer(&topology_timer, jiffies + HZ * 60);
}
void topology_expect_change(void)
{
if (!MACHINE_HAS_TOPOLOGY)
return;
/* This is racy, but it doesn't matter since it is just a heuristic.
* Worst case is that we poll in a higher frequency for a bit longer.
*/
if (atomic_read(&topology_poll) > 60)
return;
atomic_add(60, &topology_poll);
set_topology_timer();
}
static int cpu_management;
static ssize_t dispatching_show(struct device *dev,
struct device_attribute *attr,
char *buf)
{
ssize_t count;
mutex_lock(&smp_cpu_state_mutex);
count = sprintf(buf, "%d\n", cpu_management);
mutex_unlock(&smp_cpu_state_mutex);
return count;
}
static ssize_t dispatching_store(struct device *dev,
struct device_attribute *attr,
const char *buf,
size_t count)
{
int val, rc;
char delim;
if (sscanf(buf, "%d %c", &val, &delim) != 1)
return -EINVAL;
if (val != 0 && val != 1)
return -EINVAL;
rc = 0;
get_online_cpus();
mutex_lock(&smp_cpu_state_mutex);
if (cpu_management == val)
goto out;
rc = topology_set_cpu_management(val);
if (rc)
goto out;
cpu_management = val;
topology_expect_change();
out:
mutex_unlock(&smp_cpu_state_mutex);
put_online_cpus();
return rc ? rc : count;
}
static DEVICE_ATTR_RW(dispatching);
static ssize_t cpu_polarization_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
int cpu = dev->id;
ssize_t count;
mutex_lock(&smp_cpu_state_mutex);
switch (smp_cpu_get_polarization(cpu)) {
case POLARIZATION_HRZ:
count = sprintf(buf, "horizontal\n");
break;
case POLARIZATION_VL:
count = sprintf(buf, "vertical:low\n");
break;
case POLARIZATION_VM:
count = sprintf(buf, "vertical:medium\n");
break;
case POLARIZATION_VH:
count = sprintf(buf, "vertical:high\n");
break;
default:
count = sprintf(buf, "unknown\n");
break;
}
mutex_unlock(&smp_cpu_state_mutex);
return count;
}
static DEVICE_ATTR(polarization, 0444, cpu_polarization_show, NULL);
static struct attribute *topology_cpu_attrs[] = {
&dev_attr_polarization.attr,
NULL,
};
static struct attribute_group topology_cpu_attr_group = {
.attrs = topology_cpu_attrs,
};
static ssize_t cpu_dedicated_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
int cpu = dev->id;
ssize_t count;
mutex_lock(&smp_cpu_state_mutex);
count = sprintf(buf, "%d\n", topology_cpu_dedicated(cpu));
mutex_unlock(&smp_cpu_state_mutex);
return count;
}
static DEVICE_ATTR(dedicated, 0444, cpu_dedicated_show, NULL);
static struct attribute *topology_extra_cpu_attrs[] = {
&dev_attr_dedicated.attr,
NULL,
};
static struct attribute_group topology_extra_cpu_attr_group = {
.attrs = topology_extra_cpu_attrs,
};
int topology_cpu_init(struct cpu *cpu)
{
int rc;
rc = sysfs_create_group(&cpu->dev.kobj, &topology_cpu_attr_group);
if (rc || !MACHINE_HAS_TOPOLOGY)
return rc;
rc = sysfs_create_group(&cpu->dev.kobj, &topology_extra_cpu_attr_group);
if (rc)
sysfs_remove_group(&cpu->dev.kobj, &topology_cpu_attr_group);
return rc;
}
static const struct cpumask *cpu_thread_mask(int cpu)
{
return &cpu_topology[cpu].thread_mask;
}
const struct cpumask *cpu_coregroup_mask(int cpu)
{
return &cpu_topology[cpu].core_mask;
}
static const struct cpumask *cpu_book_mask(int cpu)
{
return &cpu_topology[cpu].book_mask;
}
static const struct cpumask *cpu_drawer_mask(int cpu)
{
return &cpu_topology[cpu].drawer_mask;
}
static struct sched_domain_topology_level s390_topology[] = {
{ cpu_thread_mask, cpu_smt_flags, SD_INIT_NAME(SMT) },
{ cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) },
{ cpu_book_mask, SD_INIT_NAME(BOOK) },
{ cpu_drawer_mask, SD_INIT_NAME(DRAWER) },
{ cpu_cpu_mask, SD_INIT_NAME(DIE) },
{ NULL, },
};
static void __init alloc_masks(struct sysinfo_15_1_x *info,
struct mask_info *mask, int offset)
{
int i, nr_masks;
nr_masks = info->mag[TOPOLOGY_NR_MAG - offset];
for (i = 0; i < info->mnest - offset; i++)
nr_masks *= info->mag[TOPOLOGY_NR_MAG - offset - 1 - i];
nr_masks = max(nr_masks, 1);
for (i = 0; i < nr_masks; i++) {
mask->next = memblock_virt_alloc(sizeof(*mask->next), 8);
mask = mask->next;
}
}
void __init topology_init_early(void)
{
struct sysinfo_15_1_x *info;
set_sched_topology(s390_topology);
if (topology_mode == TOPOLOGY_MODE_UNINITIALIZED) {
if (MACHINE_HAS_TOPOLOGY)
topology_mode = TOPOLOGY_MODE_HW;
else
topology_mode = TOPOLOGY_MODE_SINGLE;
}
if (!MACHINE_HAS_TOPOLOGY)
goto out;
tl_info = memblock_virt_alloc(PAGE_SIZE, PAGE_SIZE);
info = tl_info;
store_topology(info);
pr_info("The CPU configuration topology of the machine is: %d %d %d %d %d %d / %d\n",
info->mag[0], info->mag[1], info->mag[2], info->mag[3],
info->mag[4], info->mag[5], info->mnest);
alloc_masks(info, &socket_info, 1);
alloc_masks(info, &book_info, 2);
alloc_masks(info, &drawer_info, 3);
out:
__arch_update_cpu_topology();
__arch_update_dedicated_flag(NULL);
}
static inline int topology_get_mode(int enabled)
{
if (!enabled)
return TOPOLOGY_MODE_SINGLE;
return MACHINE_HAS_TOPOLOGY ? TOPOLOGY_MODE_HW : TOPOLOGY_MODE_PACKAGE;
}
static inline int topology_is_enabled(void)
{
return topology_mode != TOPOLOGY_MODE_SINGLE;
}
static int __init topology_setup(char *str)
{
bool enabled;
int rc;
rc = kstrtobool(str, &enabled);
if (rc)
return rc;
topology_mode = topology_get_mode(enabled);
return 0;
}
early_param("topology", topology_setup);
static int topology_ctl_handler(struct ctl_table *ctl, int write,
void __user *buffer, size_t *lenp, loff_t *ppos)
{
int enabled = topology_is_enabled();
int new_mode;
int zero = 0;
int one = 1;
int rc;
struct ctl_table ctl_entry = {
.procname = ctl->procname,
.data = &enabled,
.maxlen = sizeof(int),
.extra1 = &zero,
.extra2 = &one,
};
rc = proc_douintvec_minmax(&ctl_entry, write, buffer, lenp, ppos);
if (rc < 0 || !write)
return rc;
mutex_lock(&smp_cpu_state_mutex);
new_mode = topology_get_mode(enabled);
if (topology_mode != new_mode) {
topology_mode = new_mode;
topology_schedule_update();
}
mutex_unlock(&smp_cpu_state_mutex);
topology_flush_work();
return rc;
}
static struct ctl_table topology_ctl_table[] = {
{
.procname = "topology",
.mode = 0644,
.proc_handler = topology_ctl_handler,
},
{ },
};
static struct ctl_table topology_dir_table[] = {
{
.procname = "s390",
.maxlen = 0,
.mode = 0555,
.child = topology_ctl_table,
},
{ },
};
static int __init topology_init(void)
{
timer_setup(&topology_timer, topology_timer_fn, TIMER_DEFERRABLE);
if (MACHINE_HAS_TOPOLOGY)
set_topology_timer();
else
topology_update_polarization_simple();
register_sysctl_table(topology_dir_table);
return device_create_file(cpu_subsys.dev_root, &dev_attr_dispatching);
}
device_initcall(topology_init);
+33
View File
@@ -0,0 +1,33 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Tracepoint definitions for s390
*
* Copyright IBM Corp. 2015
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <linux/percpu.h>
#define CREATE_TRACE_POINTS
#include <asm/trace/diag.h>
EXPORT_TRACEPOINT_SYMBOL(s390_diagnose);
static DEFINE_PER_CPU(unsigned int, diagnose_trace_depth);
void notrace trace_s390_diagnose_norecursion(int diag_nr)
{
unsigned long flags;
unsigned int *depth;
/* Avoid lockdep recursion. */
if (IS_ENABLED(CONFIG_LOCKDEP))
return;
local_irq_save(flags);
depth = this_cpu_ptr(&diagnose_trace_depth);
if (*depth == 0) {
(*depth)++;
trace_s390_diagnose(diag_nr);
(*depth)--;
}
local_irq_restore(flags);
}
+267
View File
@@ -0,0 +1,267 @@
// SPDX-License-Identifier: GPL-2.0
/*
* S390 version
* Copyright IBM Corp. 1999, 2000
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
* Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
*
* Derived from "arch/i386/kernel/traps.c"
* Copyright (C) 1991, 1992 Linus Torvalds
*/
/*
* 'Traps.c' handles hardware traps and faults after we have saved some
* state in 'asm.s'.
*/
#include <linux/kprobes.h>
#include <linux/kdebug.h>
#include <linux/extable.h>
#include <linux/ptrace.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/uaccess.h>
#include <linux/cpu.h>
#include <asm/fpu/api.h>
#include "entry.h"
static inline void __user *get_trap_ip(struct pt_regs *regs)
{
unsigned long address;
if (regs->int_code & 0x200)
address = *(unsigned long *)(current->thread.trap_tdb + 24);
else
address = regs->psw.addr;
return (void __user *) (address - (regs->int_code >> 16));
}
int is_valid_bugaddr(unsigned long addr)
{
return 1;
}
void do_report_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
{
if (user_mode(regs)) {
force_sig_fault(si_signo, si_code, get_trap_ip(regs), current);
report_user_fault(regs, si_signo, 0);
} else {
const struct exception_table_entry *fixup;
fixup = search_exception_tables(regs->psw.addr);
if (fixup)
regs->psw.addr = extable_fixup(fixup);
else {
enum bug_trap_type btt;
btt = report_bug(regs->psw.addr, regs);
if (btt == BUG_TRAP_TYPE_WARN)
return;
die(regs, str);
}
}
}
static void do_trap(struct pt_regs *regs, int si_signo, int si_code, char *str)
{
if (notify_die(DIE_TRAP, str, regs, 0,
regs->int_code, si_signo) == NOTIFY_STOP)
return;
do_report_trap(regs, si_signo, si_code, str);
}
NOKPROBE_SYMBOL(do_trap);
void do_per_trap(struct pt_regs *regs)
{
if (notify_die(DIE_SSTEP, "sstep", regs, 0, 0, SIGTRAP) == NOTIFY_STOP)
return;
if (!current->ptrace)
return;
force_sig_fault(SIGTRAP, TRAP_HWBKPT,
(void __force __user *) current->thread.per_event.address, current);
}
NOKPROBE_SYMBOL(do_per_trap);
void default_trap_handler(struct pt_regs *regs)
{
if (user_mode(regs)) {
report_user_fault(regs, SIGSEGV, 0);
do_exit(SIGSEGV);
} else
die(regs, "Unknown program exception");
}
#define DO_ERROR_INFO(name, signr, sicode, str) \
void name(struct pt_regs *regs) \
{ \
do_trap(regs, signr, sicode, str); \
}
DO_ERROR_INFO(addressing_exception, SIGILL, ILL_ILLADR,
"addressing exception")
DO_ERROR_INFO(execute_exception, SIGILL, ILL_ILLOPN,
"execute exception")
DO_ERROR_INFO(divide_exception, SIGFPE, FPE_INTDIV,
"fixpoint divide exception")
DO_ERROR_INFO(overflow_exception, SIGFPE, FPE_INTOVF,
"fixpoint overflow exception")
DO_ERROR_INFO(hfp_overflow_exception, SIGFPE, FPE_FLTOVF,
"HFP overflow exception")
DO_ERROR_INFO(hfp_underflow_exception, SIGFPE, FPE_FLTUND,
"HFP underflow exception")
DO_ERROR_INFO(hfp_significance_exception, SIGFPE, FPE_FLTRES,
"HFP significance exception")
DO_ERROR_INFO(hfp_divide_exception, SIGFPE, FPE_FLTDIV,
"HFP divide exception")
DO_ERROR_INFO(hfp_sqrt_exception, SIGFPE, FPE_FLTINV,
"HFP square root exception")
DO_ERROR_INFO(operand_exception, SIGILL, ILL_ILLOPN,
"operand exception")
DO_ERROR_INFO(privileged_op, SIGILL, ILL_PRVOPC,
"privileged operation")
DO_ERROR_INFO(special_op_exception, SIGILL, ILL_ILLOPN,
"special operation exception")
DO_ERROR_INFO(transaction_exception, SIGILL, ILL_ILLOPN,
"transaction constraint exception")
static inline void do_fp_trap(struct pt_regs *regs, __u32 fpc)
{
int si_code = 0;
/* FPC[2] is Data Exception Code */
if ((fpc & 0x00000300) == 0) {
/* bits 6 and 7 of DXC are 0 iff IEEE exception */
if (fpc & 0x8000) /* invalid fp operation */
si_code = FPE_FLTINV;
else if (fpc & 0x4000) /* div by 0 */
si_code = FPE_FLTDIV;
else if (fpc & 0x2000) /* overflow */
si_code = FPE_FLTOVF;
else if (fpc & 0x1000) /* underflow */
si_code = FPE_FLTUND;
else if (fpc & 0x0800) /* inexact */
si_code = FPE_FLTRES;
}
do_trap(regs, SIGFPE, si_code, "floating point exception");
}
void translation_exception(struct pt_regs *regs)
{
/* May never happen. */
panic("Translation exception");
}
void illegal_op(struct pt_regs *regs)
{
__u8 opcode[6];
__u16 __user *location;
int is_uprobe_insn = 0;
int signal = 0;
location = get_trap_ip(regs);
if (user_mode(regs)) {
if (get_user(*((__u16 *) opcode), (__u16 __user *) location))
return;
if (*((__u16 *) opcode) == S390_BREAKPOINT_U16) {
if (current->ptrace)
force_sig_fault(SIGTRAP, TRAP_BRKPT, location, current);
else
signal = SIGILL;
#ifdef CONFIG_UPROBES
} else if (*((__u16 *) opcode) == UPROBE_SWBP_INSN) {
is_uprobe_insn = 1;
#endif
} else
signal = SIGILL;
}
/*
* We got either an illegal op in kernel mode, or user space trapped
* on a uprobes illegal instruction. See if kprobes or uprobes picks
* it up. If not, SIGILL.
*/
if (is_uprobe_insn || !user_mode(regs)) {
if (notify_die(DIE_BPT, "bpt", regs, 0,
3, SIGTRAP) != NOTIFY_STOP)
signal = SIGILL;
}
if (signal)
do_trap(regs, signal, ILL_ILLOPC, "illegal operation");
}
NOKPROBE_SYMBOL(illegal_op);
DO_ERROR_INFO(specification_exception, SIGILL, ILL_ILLOPN,
"specification exception");
void vector_exception(struct pt_regs *regs)
{
int si_code, vic;
if (!MACHINE_HAS_VX) {
do_trap(regs, SIGILL, ILL_ILLOPN, "illegal operation");
return;
}
/* get vector interrupt code from fpc */
save_fpu_regs();
vic = (current->thread.fpu.fpc & 0xf00) >> 8;
switch (vic) {
case 1: /* invalid vector operation */
si_code = FPE_FLTINV;
break;
case 2: /* division by zero */
si_code = FPE_FLTDIV;
break;
case 3: /* overflow */
si_code = FPE_FLTOVF;
break;
case 4: /* underflow */
si_code = FPE_FLTUND;
break;
case 5: /* inexact */
si_code = FPE_FLTRES;
break;
default: /* unknown cause */
si_code = 0;
}
do_trap(regs, SIGFPE, si_code, "vector exception");
}
void data_exception(struct pt_regs *regs)
{
int signal = 0;
save_fpu_regs();
if (current->thread.fpu.fpc & FPC_DXC_MASK)
signal = SIGFPE;
else
signal = SIGILL;
if (signal == SIGFPE)
do_fp_trap(regs, current->thread.fpu.fpc);
else if (signal)
do_trap(regs, signal, ILL_ILLOPN, "data exception");
}
void space_switch_exception(struct pt_regs *regs)
{
/* Set user psw back to home space mode. */
if (user_mode(regs))
regs->psw.mask |= PSW_ASC_HOME;
/* Send SIGILL. */
do_trap(regs, SIGILL, ILL_PRVOPC, "space switch event");
}
void kernel_stack_overflow(struct pt_regs *regs)
{
bust_spinlocks(1);
printk("Kernel stack overflow.\n");
show_regs(regs);
bust_spinlocks(0);
panic("Corrupt kernel stack, can't continue.");
}
NOKPROBE_SYMBOL(kernel_stack_overflow);
void __init trap_init(void)
{
local_mcck_enable();
}
+397
View File
@@ -0,0 +1,397 @@
// SPDX-License-Identifier: GPL-2.0
/*
* User-space Probes (UProbes) for s390
*
* Copyright IBM Corp. 2014
* Author(s): Jan Willeke,
*/
#include <linux/uaccess.h>
#include <linux/uprobes.h>
#include <linux/compat.h>
#include <linux/kdebug.h>
#include <linux/sched/task_stack.h>
#include <asm/switch_to.h>
#include <asm/facility.h>
#include <asm/kprobes.h>
#include <asm/dis.h>
#include "entry.h"
#define UPROBE_TRAP_NR UINT_MAX
int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm,
unsigned long addr)
{
return probe_is_prohibited_opcode(auprobe->insn);
}
int arch_uprobe_pre_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
{
if (psw_bits(regs->psw).eaba == PSW_BITS_AMODE_24BIT)
return -EINVAL;
if (!is_compat_task() && psw_bits(regs->psw).eaba == PSW_BITS_AMODE_31BIT)
return -EINVAL;
clear_pt_regs_flag(regs, PIF_PER_TRAP);
auprobe->saved_per = psw_bits(regs->psw).per;
auprobe->saved_int_code = regs->int_code;
regs->int_code = UPROBE_TRAP_NR;
regs->psw.addr = current->utask->xol_vaddr;
set_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP);
update_cr_regs(current);
return 0;
}
bool arch_uprobe_xol_was_trapped(struct task_struct *tsk)
{
struct pt_regs *regs = task_pt_regs(tsk);
if (regs->int_code != UPROBE_TRAP_NR)
return true;
return false;
}
static int check_per_event(unsigned short cause, unsigned long control,
struct pt_regs *regs)
{
if (!(regs->psw.mask & PSW_MASK_PER))
return 0;
/* user space single step */
if (control == 0)
return 1;
/* over indication for storage alteration */
if ((control & 0x20200000) && (cause & 0x2000))
return 1;
if (cause & 0x8000) {
/* all branches */
if ((control & 0x80800000) == 0x80000000)
return 1;
/* branch into selected range */
if (((control & 0x80800000) == 0x80800000) &&
regs->psw.addr >= current->thread.per_user.start &&
regs->psw.addr <= current->thread.per_user.end)
return 1;
}
return 0;
}
int arch_uprobe_post_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
{
int fixup = probe_get_fixup_type(auprobe->insn);
struct uprobe_task *utask = current->utask;
clear_tsk_thread_flag(current, TIF_UPROBE_SINGLESTEP);
update_cr_regs(current);
psw_bits(regs->psw).per = auprobe->saved_per;
regs->int_code = auprobe->saved_int_code;
if (fixup & FIXUP_PSW_NORMAL)
regs->psw.addr += utask->vaddr - utask->xol_vaddr;
if (fixup & FIXUP_RETURN_REGISTER) {
int reg = (auprobe->insn[0] & 0xf0) >> 4;
regs->gprs[reg] += utask->vaddr - utask->xol_vaddr;
}
if (fixup & FIXUP_BRANCH_NOT_TAKEN) {
int ilen = insn_length(auprobe->insn[0] >> 8);
if (regs->psw.addr - utask->xol_vaddr == ilen)
regs->psw.addr = utask->vaddr + ilen;
}
if (check_per_event(current->thread.per_event.cause,
current->thread.per_user.control, regs)) {
/* fix per address */
current->thread.per_event.address = utask->vaddr;
/* trigger per event */
set_pt_regs_flag(regs, PIF_PER_TRAP);
}
return 0;
}
int arch_uprobe_exception_notify(struct notifier_block *self, unsigned long val,
void *data)
{
struct die_args *args = data;
struct pt_regs *regs = args->regs;
if (!user_mode(regs))
return NOTIFY_DONE;
if (regs->int_code & 0x200) /* Trap during transaction */
return NOTIFY_DONE;
switch (val) {
case DIE_BPT:
if (uprobe_pre_sstep_notifier(regs))
return NOTIFY_STOP;
break;
case DIE_SSTEP:
if (uprobe_post_sstep_notifier(regs))
return NOTIFY_STOP;
default:
break;
}
return NOTIFY_DONE;
}
void arch_uprobe_abort_xol(struct arch_uprobe *auprobe, struct pt_regs *regs)
{
clear_thread_flag(TIF_UPROBE_SINGLESTEP);
regs->int_code = auprobe->saved_int_code;
regs->psw.addr = current->utask->vaddr;
current->thread.per_event.address = current->utask->vaddr;
}
unsigned long arch_uretprobe_hijack_return_addr(unsigned long trampoline,
struct pt_regs *regs)
{
unsigned long orig;
orig = regs->gprs[14];
regs->gprs[14] = trampoline;
return orig;
}
bool arch_uretprobe_is_alive(struct return_instance *ret, enum rp_check ctx,
struct pt_regs *regs)
{
if (ctx == RP_CHECK_CHAIN_CALL)
return user_stack_pointer(regs) <= ret->stack;
else
return user_stack_pointer(regs) < ret->stack;
}
/* Instruction Emulation */
static void adjust_psw_addr(psw_t *psw, unsigned long len)
{
psw->addr = __rewind_psw(*psw, -len);
}
#define EMU_ILLEGAL_OP 1
#define EMU_SPECIFICATION 2
#define EMU_ADDRESSING 3
#define emu_load_ril(ptr, output) \
({ \
unsigned int mask = sizeof(*(ptr)) - 1; \
__typeof__(*(ptr)) input; \
int __rc = 0; \
\
if (!test_facility(34)) \
__rc = EMU_ILLEGAL_OP; \
else if ((u64 __force)ptr & mask) \
__rc = EMU_SPECIFICATION; \
else if (get_user(input, ptr)) \
__rc = EMU_ADDRESSING; \
else \
*(output) = input; \
__rc; \
})
#define emu_store_ril(regs, ptr, input) \
({ \
unsigned int mask = sizeof(*(ptr)) - 1; \
__typeof__(ptr) __ptr = (ptr); \
int __rc = 0; \
\
if (!test_facility(34)) \
__rc = EMU_ILLEGAL_OP; \
else if ((u64 __force)__ptr & mask) \
__rc = EMU_SPECIFICATION; \
else if (put_user(*(input), __ptr)) \
__rc = EMU_ADDRESSING; \
if (__rc == 0) \
sim_stor_event(regs, \
(void __force *)__ptr, \
mask + 1); \
__rc; \
})
#define emu_cmp_ril(regs, ptr, cmp) \
({ \
unsigned int mask = sizeof(*(ptr)) - 1; \
__typeof__(*(ptr)) input; \
int __rc = 0; \
\
if (!test_facility(34)) \
__rc = EMU_ILLEGAL_OP; \
else if ((u64 __force)ptr & mask) \
__rc = EMU_SPECIFICATION; \
else if (get_user(input, ptr)) \
__rc = EMU_ADDRESSING; \
else if (input > *(cmp)) \
psw_bits((regs)->psw).cc = 1; \
else if (input < *(cmp)) \
psw_bits((regs)->psw).cc = 2; \
else \
psw_bits((regs)->psw).cc = 0; \
__rc; \
})
struct insn_ril {
u8 opc0;
u8 reg : 4;
u8 opc1 : 4;
s32 disp;
} __packed;
union split_register {
u64 u64;
u32 u32[2];
u16 u16[4];
s64 s64;
s32 s32[2];
s16 s16[4];
};
/*
* If user per registers are setup to trace storage alterations and an
* emulated store took place on a fitting address a user trap is generated.
*/
static void sim_stor_event(struct pt_regs *regs, void *addr, int len)
{
if (!(regs->psw.mask & PSW_MASK_PER))
return;
if (!(current->thread.per_user.control & PER_EVENT_STORE))
return;
if ((void *)current->thread.per_user.start > (addr + len))
return;
if ((void *)current->thread.per_user.end < addr)
return;
current->thread.per_event.address = regs->psw.addr;
current->thread.per_event.cause = PER_EVENT_STORE >> 16;
set_pt_regs_flag(regs, PIF_PER_TRAP);
}
/*
* pc relative instructions are emulated, since parameters may not be
* accessible from the xol area due to range limitations.
*/
static void handle_insn_ril(struct arch_uprobe *auprobe, struct pt_regs *regs)
{
union split_register *rx;
struct insn_ril *insn;
unsigned int ilen;
void *uptr;
int rc = 0;
insn = (struct insn_ril *) &auprobe->insn;
rx = (union split_register *) &regs->gprs[insn->reg];
uptr = (void *)(regs->psw.addr + (insn->disp * 2));
ilen = insn_length(insn->opc0);
switch (insn->opc0) {
case 0xc0:
switch (insn->opc1) {
case 0x00: /* larl */
rx->u64 = (unsigned long)uptr;
break;
}
break;
case 0xc4:
switch (insn->opc1) {
case 0x02: /* llhrl */
rc = emu_load_ril((u16 __user *)uptr, &rx->u32[1]);
break;
case 0x04: /* lghrl */
rc = emu_load_ril((s16 __user *)uptr, &rx->u64);
break;
case 0x05: /* lhrl */
rc = emu_load_ril((s16 __user *)uptr, &rx->u32[1]);
break;
case 0x06: /* llghrl */
rc = emu_load_ril((u16 __user *)uptr, &rx->u64);
break;
case 0x08: /* lgrl */
rc = emu_load_ril((u64 __user *)uptr, &rx->u64);
break;
case 0x0c: /* lgfrl */
rc = emu_load_ril((s32 __user *)uptr, &rx->u64);
break;
case 0x0d: /* lrl */
rc = emu_load_ril((u32 __user *)uptr, &rx->u32[1]);
break;
case 0x0e: /* llgfrl */
rc = emu_load_ril((u32 __user *)uptr, &rx->u64);
break;
case 0x07: /* sthrl */
rc = emu_store_ril(regs, (u16 __user *)uptr, &rx->u16[3]);
break;
case 0x0b: /* stgrl */
rc = emu_store_ril(regs, (u64 __user *)uptr, &rx->u64);
break;
case 0x0f: /* strl */
rc = emu_store_ril(regs, (u32 __user *)uptr, &rx->u32[1]);
break;
}
break;
case 0xc6:
switch (insn->opc1) {
case 0x02: /* pfdrl */
if (!test_facility(34))
rc = EMU_ILLEGAL_OP;
break;
case 0x04: /* cghrl */
rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s64);
break;
case 0x05: /* chrl */
rc = emu_cmp_ril(regs, (s16 __user *)uptr, &rx->s32[1]);
break;
case 0x06: /* clghrl */
rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u64);
break;
case 0x07: /* clhrl */
rc = emu_cmp_ril(regs, (u16 __user *)uptr, &rx->u32[1]);
break;
case 0x08: /* cgrl */
rc = emu_cmp_ril(regs, (s64 __user *)uptr, &rx->s64);
break;
case 0x0a: /* clgrl */
rc = emu_cmp_ril(regs, (u64 __user *)uptr, &rx->u64);
break;
case 0x0c: /* cgfrl */
rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s64);
break;
case 0x0d: /* crl */
rc = emu_cmp_ril(regs, (s32 __user *)uptr, &rx->s32[1]);
break;
case 0x0e: /* clgfrl */
rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u64);
break;
case 0x0f: /* clrl */
rc = emu_cmp_ril(regs, (u32 __user *)uptr, &rx->u32[1]);
break;
}
break;
}
adjust_psw_addr(&regs->psw, ilen);
switch (rc) {
case EMU_ILLEGAL_OP:
regs->int_code = ilen << 16 | 0x0001;
do_report_trap(regs, SIGILL, ILL_ILLOPC, NULL);
break;
case EMU_SPECIFICATION:
regs->int_code = ilen << 16 | 0x0006;
do_report_trap(regs, SIGILL, ILL_ILLOPC , NULL);
break;
case EMU_ADDRESSING:
regs->int_code = ilen << 16 | 0x0005;
do_report_trap(regs, SIGSEGV, SEGV_MAPERR, NULL);
break;
}
}
bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs)
{
if ((psw_bits(regs->psw).eaba == PSW_BITS_AMODE_24BIT) ||
((psw_bits(regs->psw).eaba == PSW_BITS_AMODE_31BIT) &&
!is_compat_task())) {
regs->psw.addr = __rewind_psw(regs->psw, UPROBE_SWBP_INSN_SIZE);
do_report_trap(regs, SIGILL, ILL_ILLADR, NULL);
return true;
}
if (probe_is_insn_relative_long(auprobe->insn)) {
handle_insn_ril(auprobe, regs);
return true;
}
return false;
}
+325
View File
@@ -0,0 +1,325 @@
// SPDX-License-Identifier: GPL-2.0
/*
* vdso setup for s390
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/stddef.h>
#include <linux/unistd.h>
#include <linux/slab.h>
#include <linux/user.h>
#include <linux/elf.h>
#include <linux/security.h>
#include <linux/bootmem.h>
#include <linux/compat.h>
#include <asm/asm-offsets.h>
#include <asm/pgtable.h>
#include <asm/processor.h>
#include <asm/mmu.h>
#include <asm/mmu_context.h>
#include <asm/sections.h>
#include <asm/vdso.h>
#include <asm/facility.h>
#ifdef CONFIG_COMPAT
extern char vdso32_start, vdso32_end;
static void *vdso32_kbase = &vdso32_start;
static unsigned int vdso32_pages;
static struct page **vdso32_pagelist;
#endif
extern char vdso64_start, vdso64_end;
static void *vdso64_kbase = &vdso64_start;
static unsigned int vdso64_pages;
static struct page **vdso64_pagelist;
/*
* Should the kernel map a VDSO page into processes and pass its
* address down to glibc upon exec()?
*/
unsigned int __read_mostly vdso_enabled = 1;
static vm_fault_t vdso_fault(const struct vm_special_mapping *sm,
struct vm_area_struct *vma, struct vm_fault *vmf)
{
struct page **vdso_pagelist;
unsigned long vdso_pages;
vdso_pagelist = vdso64_pagelist;
vdso_pages = vdso64_pages;
#ifdef CONFIG_COMPAT
if (vma->vm_mm->context.compat_mm) {
vdso_pagelist = vdso32_pagelist;
vdso_pages = vdso32_pages;
}
#endif
if (vmf->pgoff >= vdso_pages)
return VM_FAULT_SIGBUS;
vmf->page = vdso_pagelist[vmf->pgoff];
get_page(vmf->page);
return 0;
}
static int vdso_mremap(const struct vm_special_mapping *sm,
struct vm_area_struct *vma)
{
unsigned long vdso_pages;
vdso_pages = vdso64_pages;
#ifdef CONFIG_COMPAT
if (vma->vm_mm->context.compat_mm)
vdso_pages = vdso32_pages;
#endif
if ((vdso_pages << PAGE_SHIFT) != vma->vm_end - vma->vm_start)
return -EINVAL;
if (WARN_ON_ONCE(current->mm != vma->vm_mm))
return -EFAULT;
current->mm->context.vdso_base = vma->vm_start;
return 0;
}
static const struct vm_special_mapping vdso_mapping = {
.name = "[vdso]",
.fault = vdso_fault,
.mremap = vdso_mremap,
};
static int __init vdso_setup(char *s)
{
unsigned long val;
int rc;
rc = 0;
if (strncmp(s, "on", 3) == 0)
vdso_enabled = 1;
else if (strncmp(s, "off", 4) == 0)
vdso_enabled = 0;
else {
rc = kstrtoul(s, 0, &val);
vdso_enabled = rc ? 0 : !!val;
}
return !rc;
}
__setup("vdso=", vdso_setup);
/*
* The vdso data page
*/
static union {
struct vdso_data data;
u8 page[PAGE_SIZE];
} vdso_data_store __page_aligned_data;
struct vdso_data *vdso_data = &vdso_data_store.data;
/*
* Setup vdso data page.
*/
static void __init vdso_init_data(struct vdso_data *vd)
{
vd->ectg_available = test_facility(31);
}
/*
* Allocate/free per cpu vdso data.
*/
#define SEGMENT_ORDER 2
/*
* The initial vdso_data structure for the boot CPU. Eventually
* it is replaced with a properly allocated structure in vdso_init.
* This is necessary because a valid S390_lowcore.vdso_per_cpu_data
* pointer is required to be able to return from an interrupt or
* program check. See the exit paths in entry.S.
*/
struct vdso_data boot_vdso_data __initdata;
void __init vdso_alloc_boot_cpu(struct lowcore *lowcore)
{
lowcore->vdso_per_cpu_data = (unsigned long) &boot_vdso_data;
}
int vdso_alloc_per_cpu(struct lowcore *lowcore)
{
unsigned long segment_table, page_table, page_frame;
struct vdso_per_cpu_data *vd;
segment_table = __get_free_pages(GFP_KERNEL, SEGMENT_ORDER);
page_table = get_zeroed_page(GFP_KERNEL);
page_frame = get_zeroed_page(GFP_KERNEL);
if (!segment_table || !page_table || !page_frame)
goto out;
arch_set_page_dat(virt_to_page(segment_table), SEGMENT_ORDER);
arch_set_page_dat(virt_to_page(page_table), 0);
/* Initialize per-cpu vdso data page */
vd = (struct vdso_per_cpu_data *) page_frame;
vd->cpu_nr = lowcore->cpu_nr;
vd->node_id = cpu_to_node(vd->cpu_nr);
/* Set up page table for the vdso address space */
memset64((u64 *)segment_table, _SEGMENT_ENTRY_EMPTY, _CRST_ENTRIES);
memset64((u64 *)page_table, _PAGE_INVALID, PTRS_PER_PTE);
*(unsigned long *) segment_table = _SEGMENT_ENTRY + page_table;
*(unsigned long *) page_table = _PAGE_PROTECT + page_frame;
lowcore->vdso_asce = segment_table +
_ASCE_TABLE_LENGTH + _ASCE_USER_BITS + _ASCE_TYPE_SEGMENT;
lowcore->vdso_per_cpu_data = page_frame;
return 0;
out:
free_page(page_frame);
free_page(page_table);
free_pages(segment_table, SEGMENT_ORDER);
return -ENOMEM;
}
void vdso_free_per_cpu(struct lowcore *lowcore)
{
unsigned long segment_table, page_table, page_frame;
segment_table = lowcore->vdso_asce & PAGE_MASK;
page_table = *(unsigned long *) segment_table;
page_frame = *(unsigned long *) page_table;
free_page(page_frame);
free_page(page_table);
free_pages(segment_table, SEGMENT_ORDER);
}
/*
* This is called from binfmt_elf, we create the special vma for the
* vDSO and insert it into the mm struct tree
*/
int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
{
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma;
unsigned long vdso_pages;
unsigned long vdso_base;
int rc;
if (!vdso_enabled)
return 0;
/*
* Only map the vdso for dynamically linked elf binaries.
*/
if (!uses_interp)
return 0;
vdso_pages = vdso64_pages;
#ifdef CONFIG_COMPAT
mm->context.compat_mm = is_compat_task();
if (mm->context.compat_mm)
vdso_pages = vdso32_pages;
#endif
/*
* vDSO has a problem and was disabled, just don't "enable" it for
* the process
*/
if (vdso_pages == 0)
return 0;
/*
* pick a base address for the vDSO in process space. We try to put
* it at vdso_base which is the "natural" base for it, but we might
* fail and end up putting it elsewhere.
*/
if (down_write_killable(&mm->mmap_sem))
return -EINTR;
vdso_base = get_unmapped_area(NULL, 0, vdso_pages << PAGE_SHIFT, 0, 0);
if (IS_ERR_VALUE(vdso_base)) {
rc = vdso_base;
goto out_up;
}
/*
* our vma flags don't have VM_WRITE so by default, the process
* isn't allowed to write those pages.
* gdb can break that with ptrace interface, and thus trigger COW
* on those pages but it's then your responsibility to never do that
* on the "data" page of the vDSO or you'll stop getting kernel
* updates and your nice userland gettimeofday will be totally dead.
* It's fine to use that for setting breakpoints in the vDSO code
* pages though.
*/
vma = _install_special_mapping(mm, vdso_base, vdso_pages << PAGE_SHIFT,
VM_READ|VM_EXEC|
VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC,
&vdso_mapping);
if (IS_ERR(vma)) {
rc = PTR_ERR(vma);
goto out_up;
}
current->mm->context.vdso_base = vdso_base;
rc = 0;
out_up:
up_write(&mm->mmap_sem);
return rc;
}
static int __init vdso_init(void)
{
int i;
vdso_init_data(vdso_data);
#ifdef CONFIG_COMPAT
/* Calculate the size of the 32 bit vDSO */
vdso32_pages = ((&vdso32_end - &vdso32_start
+ PAGE_SIZE - 1) >> PAGE_SHIFT) + 1;
/* Make sure pages are in the correct state */
vdso32_pagelist = kcalloc(vdso32_pages + 1, sizeof(struct page *),
GFP_KERNEL);
BUG_ON(vdso32_pagelist == NULL);
for (i = 0; i < vdso32_pages - 1; i++) {
struct page *pg = virt_to_page(vdso32_kbase + i*PAGE_SIZE);
ClearPageReserved(pg);
get_page(pg);
vdso32_pagelist[i] = pg;
}
vdso32_pagelist[vdso32_pages - 1] = virt_to_page(vdso_data);
vdso32_pagelist[vdso32_pages] = NULL;
#endif
/* Calculate the size of the 64 bit vDSO */
vdso64_pages = ((&vdso64_end - &vdso64_start
+ PAGE_SIZE - 1) >> PAGE_SHIFT) + 1;
/* Make sure pages are in the correct state */
vdso64_pagelist = kcalloc(vdso64_pages + 1, sizeof(struct page *),
GFP_KERNEL);
BUG_ON(vdso64_pagelist == NULL);
for (i = 0; i < vdso64_pages - 1; i++) {
struct page *pg = virt_to_page(vdso64_kbase + i*PAGE_SIZE);
ClearPageReserved(pg);
get_page(pg);
vdso64_pagelist[i] = pg;
}
vdso64_pagelist[vdso64_pages - 1] = virt_to_page(vdso_data);
vdso64_pagelist[vdso64_pages] = NULL;
if (vdso_alloc_per_cpu(&S390_lowcore))
BUG();
get_page(virt_to_page(vdso_data));
return 0;
}
early_initcall(vdso_init);
+66
View File
@@ -0,0 +1,66 @@
# SPDX-License-Identifier: GPL-2.0
# List of files in the vdso, has to be asm only for now
KCOV_INSTRUMENT := n
obj-vdso32 = gettimeofday.o clock_getres.o clock_gettime.o note.o getcpu.o
# Build rules
targets := $(obj-vdso32) vdso32.so vdso32.so.dbg
obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32))
KBUILD_AFLAGS += -DBUILD_VDSO
KBUILD_CFLAGS += -DBUILD_VDSO
KBUILD_AFLAGS_31 := $(filter-out -m64,$(KBUILD_AFLAGS))
KBUILD_AFLAGS_31 += -m31 -s
KBUILD_CFLAGS_31 := $(filter-out -m64,$(KBUILD_CFLAGS))
KBUILD_CFLAGS_31 += -m31 -fPIC -shared -fno-common -fno-builtin
KBUILD_CFLAGS_31 += -nostdlib -Wl,-soname=linux-vdso32.so.1 \
$(call cc-ldoption, -Wl$(comma)--hash-style=both)
$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31)
$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31)
obj-y += vdso32_wrapper.o
extra-y += vdso32.lds
CPPFLAGS_vdso32.lds += -P -C -U$(ARCH)
# Disable gcov profiling, ubsan and kasan for VDSO code
GCOV_PROFILE := n
UBSAN_SANITIZE := n
KASAN_SANITIZE := n
# Force dependency (incbin is bad)
$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so
# link rule for the .so file, .lds has to be first
$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) FORCE
$(call if_changed,vdso32ld)
# strip rule for the .so file
$(obj)/%.so: OBJCOPYFLAGS := -S
$(obj)/%.so: $(obj)/%.so.dbg FORCE
$(call if_changed,objcopy)
# assembly rules for the .S files
$(obj-vdso32): %.o: %.S FORCE
$(call if_changed_dep,vdso32as)
# actual build commands
quiet_cmd_vdso32ld = VDSO32L $@
cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $(filter %.lds %.o,$^) -o $@
quiet_cmd_vdso32as = VDSO32A $@
cmd_vdso32as = $(CC) $(a_flags) -c -o $@ $<
# install commands for the unstripped file
quiet_cmd_vdso_install = INSTALL $@
cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@
vdso32.so: $(obj)/vdso32.so.dbg
@mkdir -p $(MODLIB)/vdso
$(call cmd,vdso_install)
vdso_install: vdso32.so
+44
View File
@@ -0,0 +1,44 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of clock_getres() for 32 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
.text
.align 4
.globl __kernel_clock_getres
.type __kernel_clock_getres,@function
__kernel_clock_getres:
CFI_STARTPROC
basr %r1,0
la %r1,4f-.(%r1)
chi %r2,__CLOCK_REALTIME
je 0f
chi %r2,__CLOCK_MONOTONIC
je 0f
la %r1,5f-4f(%r1)
chi %r2,__CLOCK_REALTIME_COARSE
je 0f
chi %r2,__CLOCK_MONOTONIC_COARSE
jne 3f
0: ltr %r3,%r3
jz 2f /* res == NULL */
1: l %r0,0(%r1)
xc 0(4,%r3),0(%r3) /* set tp->tv_sec to zero */
st %r0,4(%r3) /* store tp->tv_usec */
2: lhi %r2,0
br %r14
3: lhi %r1,__NR_clock_getres /* fallback to svc */
svc 0
br %r14
CFI_ENDPROC
4: .long __CLOCK_REALTIME_RES
5: .long __CLOCK_COARSE_RES
.size __kernel_clock_getres,.-__kernel_clock_getres
+179
View File
@@ -0,0 +1,179 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of clock_gettime() for 32 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
#include <asm/ptrace.h>
.text
.align 4
.globl __kernel_clock_gettime
.type __kernel_clock_gettime,@function
__kernel_clock_gettime:
CFI_STARTPROC
ahi %r15,-16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
basr %r5,0
0: al %r5,21f-0b(%r5) /* get &_vdso_data */
chi %r2,__CLOCK_REALTIME_COARSE
je 10f
chi %r2,__CLOCK_REALTIME
je 11f
chi %r2,__CLOCK_MONOTONIC_COARSE
je 9f
chi %r2,__CLOCK_MONOTONIC
jne 19f
/* CLOCK_MONOTONIC */
1: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */
tml %r4,0x0001 /* pending update ? loop */
jnz 1b
stcke 0(%r15) /* Store TOD clock */
lm %r0,%r1,1(%r15)
s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
sl %r1,__VDSO_XTIME_STAMP+4(%r5)
brc 3,2f
ahi %r0,-1
2: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */
lr %r2,%r0
l %r0,__VDSO_TK_MULT(%r5)
ltr %r1,%r1
mr %r0,%r0
jnm 3f
a %r0,__VDSO_TK_MULT(%r5)
3: alr %r0,%r2
al %r0,__VDSO_WTOM_NSEC(%r5)
al %r1,__VDSO_WTOM_NSEC+4(%r5)
brc 12,5f
ahi %r0,1
5: l %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
srdl %r0,0(%r2) /* >> tk->shift */
l %r2,__VDSO_WTOM_SEC+4(%r5)
cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */
jne 1b
basr %r5,0
6: ltr %r0,%r0
jnz 7f
cl %r1,20f-6b(%r5)
jl 8f
7: ahi %r2,1
sl %r1,20f-6b(%r5)
brc 3,6b
ahi %r0,-1
j 6b
8: st %r2,0(%r3) /* store tp->tv_sec */
st %r1,4(%r3) /* store tp->tv_nsec */
lhi %r2,0
ahi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
/* CLOCK_MONOTONIC_COARSE */
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
9: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */
tml %r4,0x0001 /* pending update ? loop */
jnz 9b
l %r2,__VDSO_WTOM_CRS_SEC+4(%r5)
l %r1,__VDSO_WTOM_CRS_NSEC+4(%r5)
cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */
jne 9b
j 8b
/* CLOCK_REALTIME_COARSE */
10: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */
tml %r4,0x0001 /* pending update ? loop */
jnz 10b
l %r2,__VDSO_XTIME_CRS_SEC+4(%r5)
l %r1,__VDSO_XTIME_CRS_NSEC+4(%r5)
cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */
jne 10b
j 17f
/* CLOCK_REALTIME */
11: l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */
tml %r4,0x0001 /* pending update ? loop */
jnz 11b
stcke 0(%r15) /* Store TOD clock */
lm %r0,%r1,__VDSO_TS_END(%r5) /* TOD steering end time */
s %r0,1(%r15) /* no - ts_steering_end */
sl %r1,5(%r15)
brc 3,22f
ahi %r0,-1
22: ltr %r0,%r0 /* past end of steering? */
jm 24f
srdl %r0,15 /* 1 per 2^16 */
tm __VDSO_TS_DIR+3(%r5),0x01 /* steering direction? */
jz 23f
lcr %r0,%r0 /* negative TOD offset */
lcr %r1,%r1
je 23f
ahi %r0,-1
23: a %r0,1(%r15) /* add TOD timestamp */
al %r1,5(%r15)
brc 12,25f
ahi %r0,1
j 25f
24: lm %r0,%r1,1(%r15) /* load TOD timestamp */
25: s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
sl %r1,__VDSO_XTIME_STAMP+4(%r5)
brc 3,12f
ahi %r0,-1
12: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */
lr %r2,%r0
l %r0,__VDSO_TK_MULT(%r5)
ltr %r1,%r1
mr %r0,%r0
jnm 13f
a %r0,__VDSO_TK_MULT(%r5)
13: alr %r0,%r2
al %r0,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */
al %r1,__VDSO_XTIME_NSEC+4(%r5)
brc 12,14f
ahi %r0,1
14: l %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
srdl %r0,0(%r2) /* >> tk->shift */
l %r2,__VDSO_XTIME_SEC+4(%r5)
cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */
jne 11b
basr %r5,0
15: ltr %r0,%r0
jnz 16f
cl %r1,20f-15b(%r5)
jl 17f
16: ahi %r2,1
sl %r1,20f-15b(%r5)
brc 3,15b
ahi %r0,-1
j 15b
17: st %r2,0(%r3) /* store tp->tv_sec */
st %r1,4(%r3) /* store tp->tv_nsec */
lhi %r2,0
ahi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
/* Fallback to system call */
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
19: lhi %r1,__NR_clock_gettime
svc 0
ahi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
CFI_ENDPROC
20: .long 1000000000
21: .long _vdso_data - 0b
.size __kernel_clock_gettime,.-__kernel_clock_gettime
+33
View File
@@ -0,0 +1,33 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of getcpu() for 32 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2016
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
.text
.align 4
.globl __kernel_getcpu
.type __kernel_getcpu,@function
__kernel_getcpu:
CFI_STARTPROC
la %r4,0
sacf 256
l %r5,__VDSO_CPU_NR(%r4)
l %r4,__VDSO_NODE_ID(%r4)
sacf 0
ltr %r2,%r2
jz 2f
st %r5,0(%r2)
2: ltr %r3,%r3
jz 3f
st %r4,0(%r3)
3: lhi %r2,0
br %r14
CFI_ENDPROC
.size __kernel_getcpu,.-__kernel_getcpu
+103
View File
@@ -0,0 +1,103 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of gettimeofday() for 32 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
#include <asm/ptrace.h>
.text
.align 4
.globl __kernel_gettimeofday
.type __kernel_gettimeofday,@function
__kernel_gettimeofday:
CFI_STARTPROC
ahi %r15,-16
CFI_ADJUST_CFA_OFFSET 16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
basr %r5,0
0: al %r5,13f-0b(%r5) /* get &_vdso_data */
1: ltr %r3,%r3 /* check if tz is NULL */
je 2f
mvc 0(8,%r3),__VDSO_TIMEZONE(%r5)
2: ltr %r2,%r2 /* check if tv is NULL */
je 10f
l %r4,__VDSO_UPD_COUNT+4(%r5) /* load update counter */
tml %r4,0x0001 /* pending update ? loop */
jnz 1b
stcke 0(%r15) /* Store TOD clock */
lm %r0,%r1,__VDSO_TS_END(%r5) /* TOD steering end time */
s %r0,1(%r15)
sl %r1,5(%r15)
brc 3,14f
ahi %r0,-1
14: ltr %r0,%r0 /* past end of steering? */
jm 16f
srdl %r0,15 /* 1 per 2^16 */
tm __VDSO_TS_DIR+3(%r5),0x01 /* steering direction? */
jz 15f
lcr %r0,%r0 /* negative TOD offset */
lcr %r1,%r1
je 15f
ahi %r0,-1
15: a %r0,1(%r15) /* add TOD timestamp */
al %r1,5(%r15)
brc 12,17f
ahi %r0,1
j 17f
16: lm %r0,%r1,1(%r15) /* load TOD timestamp */
17: s %r0,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
sl %r1,__VDSO_XTIME_STAMP+4(%r5)
brc 3,3f
ahi %r0,-1
3: ms %r0,__VDSO_TK_MULT(%r5) /* * tk->mult */
st %r0,0(%r15)
l %r0,__VDSO_TK_MULT(%r5)
ltr %r1,%r1
mr %r0,%r0
jnm 4f
a %r0,__VDSO_TK_MULT(%r5)
4: al %r0,0(%r15)
al %r0,__VDSO_XTIME_NSEC(%r5) /* + xtime */
al %r1,__VDSO_XTIME_NSEC+4(%r5)
brc 12,5f
ahi %r0,1
5: mvc 0(4,%r15),__VDSO_XTIME_SEC+4(%r5)
cl %r4,__VDSO_UPD_COUNT+4(%r5) /* check update counter */
jne 1b
l %r4,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
srdl %r0,0(%r4) /* >> tk->shift */
l %r4,0(%r15) /* get tv_sec from stack */
basr %r5,0
6: ltr %r0,%r0
jnz 7f
cl %r1,11f-6b(%r5)
jl 8f
7: ahi %r4,1
sl %r1,11f-6b(%r5)
brc 3,6b
ahi %r0,-1
j 6b
8: st %r4,0(%r2) /* store tv->tv_sec */
ltr %r1,%r1
m %r0,12f-6b(%r5)
jnm 9f
al %r0,12f-6b(%r5)
9: srl %r0,6
st %r0,4(%r2) /* store tv->tv_usec */
10: slr %r2,%r2
ahi %r15,16
CFI_ADJUST_CFA_OFFSET -16
CFI_RESTORE 15
br %r14
CFI_ENDPROC
11: .long 1000000000
12: .long 274877907
13: .long _vdso_data - 0b
.size __kernel_gettimeofday,.-__kernel_gettimeofday
+13
View File
@@ -0,0 +1,13 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
* Here we can supply some information useful to userland.
*/
#include <linux/uts.h>
#include <linux/version.h>
#include <linux/elfnote.h>
ELFNOTE_START(Linux, 0, "a")
.long LINUX_VERSION_CODE
ELFNOTE_END
+142
View File
@@ -0,0 +1,142 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* This is the infamous ld script for the 32 bits vdso
* library
*/
#include <asm/page.h>
#include <asm/vdso.h>
OUTPUT_FORMAT("elf32-s390", "elf32-s390", "elf32-s390")
OUTPUT_ARCH(s390:31-bit)
ENTRY(_start)
SECTIONS
{
. = VDSO32_LBASE + SIZEOF_HEADERS;
.hash : { *(.hash) } :text
.gnu.hash : { *(.gnu.hash) }
.dynsym : { *(.dynsym) }
.dynstr : { *(.dynstr) }
.gnu.version : { *(.gnu.version) }
.gnu.version_d : { *(.gnu.version_d) }
.gnu.version_r : { *(.gnu.version_r) }
.note : { *(.note.*) } :text :note
. = ALIGN(16);
.text : {
*(.text .stub .text.* .gnu.linkonce.t.*)
} :text
PROVIDE(__etext = .);
PROVIDE(_etext = .);
PROVIDE(etext = .);
/*
* Other stuff is appended to the text segment:
*/
.rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) }
.rodata1 : { *(.rodata1) }
.dynamic : { *(.dynamic) } :text :dynamic
.eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr
.eh_frame : { KEEP (*(.eh_frame)) } :text
.gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) }
.rela.dyn ALIGN(8) : { *(.rela.dyn) }
.got ALIGN(8) : { *(.got .toc) }
_end = .;
PROVIDE(end = .);
/*
* Stabs debugging sections are here too.
*/
.stab 0 : { *(.stab) }
.stabstr 0 : { *(.stabstr) }
.stab.excl 0 : { *(.stab.excl) }
.stab.exclstr 0 : { *(.stab.exclstr) }
.stab.index 0 : { *(.stab.index) }
.stab.indexstr 0 : { *(.stab.indexstr) }
.comment 0 : { *(.comment) }
/*
* DWARF debug sections.
* Symbols in the DWARF debugging sections are relative to the
* beginning of the section so we begin them at 0.
*/
/* DWARF 1 */
.debug 0 : { *(.debug) }
.line 0 : { *(.line) }
/* GNU DWARF 1 extensions */
.debug_srcinfo 0 : { *(.debug_srcinfo) }
.debug_sfnames 0 : { *(.debug_sfnames) }
/* DWARF 1.1 and DWARF 2 */
.debug_aranges 0 : { *(.debug_aranges) }
.debug_pubnames 0 : { *(.debug_pubnames) }
/* DWARF 2 */
.debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) }
.debug_abbrev 0 : { *(.debug_abbrev) }
.debug_line 0 : { *(.debug_line) }
.debug_frame 0 : { *(.debug_frame) }
.debug_str 0 : { *(.debug_str) }
.debug_loc 0 : { *(.debug_loc) }
.debug_macinfo 0 : { *(.debug_macinfo) }
/* SGI/MIPS DWARF 2 extensions */
.debug_weaknames 0 : { *(.debug_weaknames) }
.debug_funcnames 0 : { *(.debug_funcnames) }
.debug_typenames 0 : { *(.debug_typenames) }
.debug_varnames 0 : { *(.debug_varnames) }
/* DWARF 3 */
.debug_pubtypes 0 : { *(.debug_pubtypes) }
.debug_ranges 0 : { *(.debug_ranges) }
.gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }
. = ALIGN(PAGE_SIZE);
PROVIDE(_vdso_data = .);
/DISCARD/ : {
*(.note.GNU-stack)
*(.branch_lt)
*(.data .data.* .gnu.linkonce.d.* .sdata*)
*(.bss .sbss .dynbss .dynsbss)
}
}
/*
* Very old versions of ld do not recognize this name token; use the constant.
*/
#define PT_GNU_EH_FRAME 0x6474e550
/*
* We must supply the ELF program headers explicitly to get just one
* PT_LOAD segment, and set the flags explicitly to make segments read-only.
*/
PHDRS
{
text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
note PT_NOTE FLAGS(4); /* PF_R */
eh_frame_hdr PT_GNU_EH_FRAME;
}
/*
* This controls what symbols we export from the DSO.
*/
VERSION
{
VDSO_VERSION_STRING {
global:
/*
* Has to be there for the kernel to find
*/
__kernel_gettimeofday;
__kernel_clock_gettime;
__kernel_clock_getres;
__kernel_getcpu;
local: *;
};
}
+15
View File
@@ -0,0 +1,15 @@
/* SPDX-License-Identifier: GPL-2.0 */
#include <linux/init.h>
#include <linux/linkage.h>
#include <asm/page.h>
__PAGE_ALIGNED_DATA
.globl vdso32_start, vdso32_end
.balign PAGE_SIZE
vdso32_start:
.incbin "arch/s390/kernel/vdso32/vdso32.so"
.balign PAGE_SIZE
vdso32_end:
.previous
+66
View File
@@ -0,0 +1,66 @@
# SPDX-License-Identifier: GPL-2.0
# List of files in the vdso, has to be asm only for now
KCOV_INSTRUMENT := n
obj-vdso64 = gettimeofday.o clock_getres.o clock_gettime.o note.o getcpu.o
# Build rules
targets := $(obj-vdso64) vdso64.so vdso64.so.dbg
obj-vdso64 := $(addprefix $(obj)/, $(obj-vdso64))
KBUILD_AFLAGS += -DBUILD_VDSO
KBUILD_CFLAGS += -DBUILD_VDSO
KBUILD_AFLAGS_64 := $(filter-out -m64,$(KBUILD_AFLAGS))
KBUILD_AFLAGS_64 += -m64 -s
KBUILD_CFLAGS_64 := $(filter-out -m64,$(KBUILD_CFLAGS))
KBUILD_CFLAGS_64 += -m64 -fPIC -shared -fno-common -fno-builtin
KBUILD_CFLAGS_64 += -nostdlib -Wl,-soname=linux-vdso64.so.1 \
$(call cc-ldoption, -Wl$(comma)--hash-style=both)
$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_64)
$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_64)
obj-y += vdso64_wrapper.o
extra-y += vdso64.lds
CPPFLAGS_vdso64.lds += -P -C -U$(ARCH)
# Disable gcov profiling, ubsan and kasan for VDSO code
GCOV_PROFILE := n
UBSAN_SANITIZE := n
KASAN_SANITIZE := n
# Force dependency (incbin is bad)
$(obj)/vdso64_wrapper.o : $(obj)/vdso64.so
# link rule for the .so file, .lds has to be first
$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) FORCE
$(call if_changed,vdso64ld)
# strip rule for the .so file
$(obj)/%.so: OBJCOPYFLAGS := -S
$(obj)/%.so: $(obj)/%.so.dbg FORCE
$(call if_changed,objcopy)
# assembly rules for the .S files
$(obj-vdso64): %.o: %.S FORCE
$(call if_changed_dep,vdso64as)
# actual build commands
quiet_cmd_vdso64ld = VDSO64L $@
cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $(filter %.lds %.o,$^) -o $@
quiet_cmd_vdso64as = VDSO64A $@
cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $<
# install commands for the unstripped file
quiet_cmd_vdso_install = INSTALL $@
cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@
vdso64.so: $(obj)/vdso64.so.dbg
@mkdir -p $(MODLIB)/vdso
$(call cmd,vdso_install)
vdso_install: vdso64.so
+50
View File
@@ -0,0 +1,50 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of clock_getres() for 64 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
.text
.align 4
.globl __kernel_clock_getres
.type __kernel_clock_getres,@function
__kernel_clock_getres:
CFI_STARTPROC
larl %r1,3f
lg %r0,0(%r1)
cghi %r2,__CLOCK_REALTIME_COARSE
je 0f
cghi %r2,__CLOCK_MONOTONIC_COARSE
je 0f
larl %r1,_vdso_data
llgf %r0,__VDSO_CLOCK_REALTIME_RES(%r1)
cghi %r2,__CLOCK_REALTIME
je 0f
cghi %r2,__CLOCK_MONOTONIC
je 0f
cghi %r2,__CLOCK_THREAD_CPUTIME_ID
je 0f
cghi %r2,-2 /* Per-thread CPUCLOCK with PID=0, VIRT=1 */
jne 2f
larl %r5,_vdso_data
icm %r0,15,__LC_ECTG_OK(%r5)
jz 2f
0: ltgr %r3,%r3
jz 1f /* res == NULL */
xc 0(8,%r3),0(%r3) /* set tp->tv_sec to zero */
stg %r0,8(%r3) /* store tp->tv_usec */
1: lghi %r2,0
br %r14
2: lghi %r1,__NR_clock_getres /* fallback to svc */
svc 0
br %r14
CFI_ENDPROC
3: .quad __CLOCK_COARSE_RES
.size __kernel_clock_getres,.-__kernel_clock_getres
+163
View File
@@ -0,0 +1,163 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of clock_gettime() for 64 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
#include <asm/ptrace.h>
.text
.align 4
.globl __kernel_clock_gettime
.type __kernel_clock_gettime,@function
__kernel_clock_gettime:
CFI_STARTPROC
aghi %r15,-16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
larl %r5,_vdso_data
cghi %r2,__CLOCK_REALTIME_COARSE
je 4f
cghi %r2,__CLOCK_REALTIME
je 5f
cghi %r2,-3 /* Per-thread CPUCLOCK with PID=0, VIRT=1 */
je 9f
cghi %r2,__CLOCK_MONOTONIC_COARSE
je 3f
cghi %r2,__CLOCK_MONOTONIC
jne 12f
/* CLOCK_MONOTONIC */
0: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */
tmll %r4,0x0001 /* pending update ? loop */
jnz 0b
stcke 0(%r15) /* Store TOD clock */
lgf %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
lg %r0,__VDSO_WTOM_SEC(%r5)
lg %r1,1(%r15)
sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */
alg %r1,__VDSO_WTOM_NSEC(%r5)
srlg %r1,%r1,0(%r2) /* >> tk->shift */
clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */
jne 0b
larl %r5,13f
1: clg %r1,0(%r5)
jl 2f
slg %r1,0(%r5)
aghi %r0,1
j 1b
2: stg %r0,0(%r3) /* store tp->tv_sec */
stg %r1,8(%r3) /* store tp->tv_nsec */
lghi %r2,0
aghi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
/* CLOCK_MONOTONIC_COARSE */
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
3: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */
tmll %r4,0x0001 /* pending update ? loop */
jnz 3b
lg %r0,__VDSO_WTOM_CRS_SEC(%r5)
lg %r1,__VDSO_WTOM_CRS_NSEC(%r5)
clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */
jne 3b
j 2b
/* CLOCK_REALTIME_COARSE */
4: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */
tmll %r4,0x0001 /* pending update ? loop */
jnz 4b
lg %r0,__VDSO_XTIME_CRS_SEC(%r5)
lg %r1,__VDSO_XTIME_CRS_NSEC(%r5)
clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */
jne 4b
j 7f
/* CLOCK_REALTIME */
5: lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */
tmll %r4,0x0001 /* pending update ? loop */
jnz 5b
stcke 0(%r15) /* Store TOD clock */
lg %r1,1(%r15)
lg %r0,__VDSO_TS_END(%r5) /* TOD steering end time */
slgr %r0,%r1 /* now - ts_steering_end */
ltgr %r0,%r0 /* past end of steering ? */
jm 17f
srlg %r0,%r0,15 /* 1 per 2^16 */
tm __VDSO_TS_DIR+3(%r5),0x01 /* steering direction? */
jz 18f
lcgr %r0,%r0 /* negative TOD offset */
18: algr %r1,%r0 /* add steering offset */
17: lgf %r2,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */
alg %r1,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */
srlg %r1,%r1,0(%r2) /* >> tk->shift */
lg %r0,__VDSO_XTIME_SEC(%r5) /* tk->xtime_sec */
clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */
jne 5b
larl %r5,13f
6: clg %r1,0(%r5)
jl 7f
slg %r1,0(%r5)
aghi %r0,1
j 6b
7: stg %r0,0(%r3) /* store tp->tv_sec */
stg %r1,8(%r3) /* store tp->tv_nsec */
lghi %r2,0
aghi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
/* CPUCLOCK_VIRT for this thread */
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
9: lghi %r4,0
icm %r0,15,__VDSO_ECTG_OK(%r5)
jz 12f
sacf 256 /* Magic ectg instruction */
.insn ssf,0xc80100000000,__VDSO_ECTG_BASE(4),__VDSO_ECTG_USER(4),4
sacf 0
algr %r1,%r0 /* r1 = cputime as TOD value */
mghi %r1,1000 /* convert to nanoseconds */
srlg %r1,%r1,12 /* r1 = cputime in nanosec */
lgr %r4,%r1
larl %r5,13f
srlg %r1,%r1,9 /* divide by 1000000000 */
mlg %r0,8(%r5)
srlg %r0,%r0,11 /* r0 = tv_sec */
stg %r0,0(%r3)
msg %r0,0(%r5) /* calculate tv_nsec */
slgr %r4,%r0 /* r4 = tv_nsec */
stg %r4,8(%r3)
lghi %r2,0
aghi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
/* Fallback to system call */
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD+16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
12: lghi %r1,__NR_clock_gettime
svc 0
aghi %r15,16
CFI_DEF_CFA_OFFSET STACK_FRAME_OVERHEAD
CFI_RESTORE 15
br %r14
CFI_ENDPROC
13: .quad 1000000000
14: .quad 19342813113834067
.size __kernel_clock_gettime,.-__kernel_clock_gettime
+33
View File
@@ -0,0 +1,33 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of getcpu() for 64 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2016
* Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
.text
.align 4
.globl __kernel_getcpu
.type __kernel_getcpu,@function
__kernel_getcpu:
CFI_STARTPROC
la %r4,0
sacf 256
l %r5,__VDSO_CPU_NR(%r4)
l %r4,__VDSO_NODE_ID(%r4)
sacf 0
ltgr %r2,%r2
jz 2f
st %r5,0(%r2)
2: ltgr %r3,%r3
jz 3f
st %r4,0(%r3)
3: lghi %r2,0
br %r14
CFI_ENDPROC
.size __kernel_getcpu,.-__kernel_getcpu
+71
View File
@@ -0,0 +1,71 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Userland implementation of gettimeofday() for 64 bits processes in a
* s390 kernel for use in the vDSO
*
* Copyright IBM Corp. 2008
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/vdso.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/dwarf.h>
#include <asm/ptrace.h>
.text
.align 4
.globl __kernel_gettimeofday
.type __kernel_gettimeofday,@function
__kernel_gettimeofday:
CFI_STARTPROC
aghi %r15,-16
CFI_ADJUST_CFA_OFFSET 16
CFI_VAL_OFFSET 15, -STACK_FRAME_OVERHEAD
larl %r5,_vdso_data
0: ltgr %r3,%r3 /* check if tz is NULL */
je 1f
mvc 0(8,%r3),__VDSO_TIMEZONE(%r5)
1: ltgr %r2,%r2 /* check if tv is NULL */
je 4f
lg %r4,__VDSO_UPD_COUNT(%r5) /* load update counter */
tmll %r4,0x0001 /* pending update ? loop */
jnz 0b
stcke 0(%r15) /* Store TOD clock */
lg %r1,1(%r15)
lg %r0,__VDSO_TS_END(%r5) /* TOD steering end time */
slgr %r0,%r1 /* now - ts_steering_end */
ltgr %r0,%r0 /* past end of steering ? */
jm 6f
srlg %r0,%r0,15 /* 1 per 2^16 */
tm __VDSO_TS_DIR+3(%r5),0x01 /* steering direction? */
jz 7f
lcgr %r0,%r0 /* negative TOD offset */
7: algr %r1,%r0 /* add steering offset */
6: sg %r1,__VDSO_XTIME_STAMP(%r5) /* TOD - cycle_last */
msgf %r1,__VDSO_TK_MULT(%r5) /* * tk->mult */
alg %r1,__VDSO_XTIME_NSEC(%r5) /* + tk->xtime_nsec */
lg %r0,__VDSO_XTIME_SEC(%r5) /* tk->xtime_sec */
clg %r4,__VDSO_UPD_COUNT(%r5) /* check update counter */
jne 0b
lgf %r5,__VDSO_TK_SHIFT(%r5) /* Timekeeper shift */
srlg %r1,%r1,0(%r5) /* >> tk->shift */
larl %r5,5f
2: clg %r1,0(%r5)
jl 3f
slg %r1,0(%r5)
aghi %r0,1
j 2b
3: stg %r0,0(%r2) /* store tv->tv_sec */
slgr %r0,%r0 /* tv_nsec -> tv_usec */
ml %r0,8(%r5)
srlg %r0,%r0,6
stg %r0,8(%r2) /* store tv->tv_usec */
4: lghi %r2,0
aghi %r15,16
CFI_ADJUST_CFA_OFFSET -16
CFI_RESTORE 15
br %r14
CFI_ENDPROC
5: .quad 1000000000
.long 274877907
.size __kernel_gettimeofday,.-__kernel_gettimeofday
+13
View File
@@ -0,0 +1,13 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
* Here we can supply some information useful to userland.
*/
#include <linux/uts.h>
#include <linux/version.h>
#include <linux/elfnote.h>
ELFNOTE_START(Linux, 0, "a")
.long LINUX_VERSION_CODE
ELFNOTE_END
+142
View File
@@ -0,0 +1,142 @@
/* SPDX-License-Identifier: GPL-2.0 */
/*
* This is the infamous ld script for the 64 bits vdso
* library
*/
#include <asm/page.h>
#include <asm/vdso.h>
OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390")
OUTPUT_ARCH(s390:64-bit)
ENTRY(_start)
SECTIONS
{
. = VDSO64_LBASE + SIZEOF_HEADERS;
.hash : { *(.hash) } :text
.gnu.hash : { *(.gnu.hash) }
.dynsym : { *(.dynsym) }
.dynstr : { *(.dynstr) }
.gnu.version : { *(.gnu.version) }
.gnu.version_d : { *(.gnu.version_d) }
.gnu.version_r : { *(.gnu.version_r) }
.note : { *(.note.*) } :text :note
. = ALIGN(16);
.text : {
*(.text .stub .text.* .gnu.linkonce.t.*)
} :text
PROVIDE(__etext = .);
PROVIDE(_etext = .);
PROVIDE(etext = .);
/*
* Other stuff is appended to the text segment:
*/
.rodata : { *(.rodata .rodata.* .gnu.linkonce.r.*) }
.rodata1 : { *(.rodata1) }
.dynamic : { *(.dynamic) } :text :dynamic
.eh_frame_hdr : { *(.eh_frame_hdr) } :text :eh_frame_hdr
.eh_frame : { KEEP (*(.eh_frame)) } :text
.gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) }
.rela.dyn ALIGN(8) : { *(.rela.dyn) }
.got ALIGN(8) : { *(.got .toc) }
_end = .;
PROVIDE(end = .);
/*
* Stabs debugging sections are here too.
*/
.stab 0 : { *(.stab) }
.stabstr 0 : { *(.stabstr) }
.stab.excl 0 : { *(.stab.excl) }
.stab.exclstr 0 : { *(.stab.exclstr) }
.stab.index 0 : { *(.stab.index) }
.stab.indexstr 0 : { *(.stab.indexstr) }
.comment 0 : { *(.comment) }
/*
* DWARF debug sections.
* Symbols in the DWARF debugging sections are relative to the
* beginning of the section so we begin them at 0.
*/
/* DWARF 1 */
.debug 0 : { *(.debug) }
.line 0 : { *(.line) }
/* GNU DWARF 1 extensions */
.debug_srcinfo 0 : { *(.debug_srcinfo) }
.debug_sfnames 0 : { *(.debug_sfnames) }
/* DWARF 1.1 and DWARF 2 */
.debug_aranges 0 : { *(.debug_aranges) }
.debug_pubnames 0 : { *(.debug_pubnames) }
/* DWARF 2 */
.debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) }
.debug_abbrev 0 : { *(.debug_abbrev) }
.debug_line 0 : { *(.debug_line) }
.debug_frame 0 : { *(.debug_frame) }
.debug_str 0 : { *(.debug_str) }
.debug_loc 0 : { *(.debug_loc) }
.debug_macinfo 0 : { *(.debug_macinfo) }
/* SGI/MIPS DWARF 2 extensions */
.debug_weaknames 0 : { *(.debug_weaknames) }
.debug_funcnames 0 : { *(.debug_funcnames) }
.debug_typenames 0 : { *(.debug_typenames) }
.debug_varnames 0 : { *(.debug_varnames) }
/* DWARF 3 */
.debug_pubtypes 0 : { *(.debug_pubtypes) }
.debug_ranges 0 : { *(.debug_ranges) }
.gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }
. = ALIGN(PAGE_SIZE);
PROVIDE(_vdso_data = .);
/DISCARD/ : {
*(.note.GNU-stack)
*(.branch_lt)
*(.data .data.* .gnu.linkonce.d.* .sdata*)
*(.bss .sbss .dynbss .dynsbss)
}
}
/*
* Very old versions of ld do not recognize this name token; use the constant.
*/
#define PT_GNU_EH_FRAME 0x6474e550
/*
* We must supply the ELF program headers explicitly to get just one
* PT_LOAD segment, and set the flags explicitly to make segments read-only.
*/
PHDRS
{
text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
note PT_NOTE FLAGS(4); /* PF_R */
eh_frame_hdr PT_GNU_EH_FRAME;
}
/*
* This controls what symbols we export from the DSO.
*/
VERSION
{
VDSO_VERSION_STRING {
global:
/*
* Has to be there for the kernel to find
*/
__kernel_gettimeofday;
__kernel_clock_gettime;
__kernel_clock_getres;
__kernel_getcpu;
local: *;
};
}
+15
View File
@@ -0,0 +1,15 @@
/* SPDX-License-Identifier: GPL-2.0 */
#include <linux/init.h>
#include <linux/linkage.h>
#include <asm/page.h>
__PAGE_ALIGNED_DATA
.globl vdso64_start, vdso64_end
.balign PAGE_SIZE
vdso64_start:
.incbin "arch/s390/kernel/vdso64/vdso64.so"
.balign PAGE_SIZE
vdso64_end:
.previous
+156
View File
@@ -0,0 +1,156 @@
/* SPDX-License-Identifier: GPL-2.0 */
/* ld script to make s390 Linux kernel
* Written by Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <asm/thread_info.h>
#include <asm/page.h>
/*
* Put .bss..swapper_pg_dir as the first thing in .bss. This will
* make sure it has 16k alignment.
*/
#define BSS_FIRST_SECTIONS *(.bss..swapper_pg_dir)
/* Handle ro_after_init data on our own. */
#define RO_AFTER_INIT_DATA
#include <asm-generic/vmlinux.lds.h>
OUTPUT_FORMAT("elf64-s390", "elf64-s390", "elf64-s390")
OUTPUT_ARCH(s390:64-bit)
ENTRY(startup_continue)
jiffies = jiffies_64;
PHDRS {
text PT_LOAD FLAGS(5); /* R_E */
data PT_LOAD FLAGS(7); /* RWE */
note PT_NOTE FLAGS(0); /* ___ */
}
SECTIONS
{
. = 0x100000;
.text : {
_stext = .; /* Start of text section */
_text = .; /* Text and read-only data */
HEAD_TEXT
TEXT_TEXT
SCHED_TEXT
CPUIDLE_TEXT
LOCK_TEXT
KPROBES_TEXT
IRQENTRY_TEXT
SOFTIRQENTRY_TEXT
*(.text.*_indirect_*)
*(.fixup)
*(.gnu.warning)
. = ALIGN(PAGE_SIZE);
_etext = .; /* End of text section */
} :text = 0x0700
NOTES :text :note
.dummy : { *(.dummy) } :data
RO_DATA_SECTION(PAGE_SIZE)
. = ALIGN(PAGE_SIZE);
_sdata = .; /* Start of data section */
. = ALIGN(PAGE_SIZE);
__start_ro_after_init = .;
.data..ro_after_init : {
*(.data..ro_after_init)
}
EXCEPTION_TABLE(16)
. = ALIGN(PAGE_SIZE);
__end_ro_after_init = .;
RW_DATA_SECTION(0x100, PAGE_SIZE, THREAD_SIZE)
_edata = .; /* End of data section */
/* will be freed after init */
. = ALIGN(PAGE_SIZE); /* Init code and data */
__init_begin = .;
. = ALIGN(PAGE_SIZE);
.init.text : AT(ADDR(.init.text) - LOAD_OFFSET) {
_sinittext = .;
INIT_TEXT
. = ALIGN(PAGE_SIZE);
_einittext = .;
}
/*
* .exit.text is discarded at runtime, not link time,
* to deal with references from __bug_table
*/
.exit.text : {
EXIT_TEXT
}
.exit.data : {
EXIT_DATA
}
/*
* struct alt_inst entries. From the header (alternative.h):
* "Alternative instructions for different CPU types or capabilities"
* Think locking instructions on spinlocks.
* Note, that it is a part of __init region.
*/
. = ALIGN(8);
.altinstructions : {
__alt_instructions = .;
*(.altinstructions)
__alt_instructions_end = .;
}
/*
* And here are the replacement instructions. The linker sticks
* them as binary blobs. The .altinstructions has enough data to
* get the address and the length of them to patch the kernel safely.
* Note, that it is a part of __init region.
*/
.altinstr_replacement : {
*(.altinstr_replacement)
}
/*
* Table with the patch locations to undo expolines
*/
.nospec_call_table : {
__nospec_call_start = . ;
*(.s390_indirect*)
__nospec_call_end = . ;
}
.nospec_return_table : {
__nospec_return_start = . ;
*(.s390_return*)
__nospec_return_end = . ;
}
/* early.c uses stsi, which requires page aligned data. */
. = ALIGN(PAGE_SIZE);
INIT_DATA_SECTION(0x100)
PERCPU_SECTION(0x100)
. = ALIGN(PAGE_SIZE);
__init_end = .; /* freed after init ends here */
BSS_SECTION(PAGE_SIZE, 4 * PAGE_SIZE, PAGE_SIZE)
_end = . ;
/* Debugging sections. */
STABS_DEBUG
DWARF_DEBUG
/* Sections to be discarded */
DISCARDS
/DISCARD/ : {
*(.eh_frame)
}
}
+437
View File
@@ -0,0 +1,437 @@
// SPDX-License-Identifier: GPL-2.0
/*
* Virtual cpu timer based timer functions.
*
* Copyright IBM Corp. 2004, 2012
* Author(s): Jan Glauber <jan.glauber@de.ibm.com>
*/
#include <linux/kernel_stat.h>
#include <linux/sched/cputime.h>
#include <linux/export.h>
#include <linux/kernel.h>
#include <linux/timex.h>
#include <linux/types.h>
#include <linux/time.h>
#include <asm/vtimer.h>
#include <asm/vtime.h>
#include <asm/cpu_mf.h>
#include <asm/smp.h>
#include "entry.h"
static void virt_timer_expire(void);
static LIST_HEAD(virt_timer_list);
static DEFINE_SPINLOCK(virt_timer_lock);
static atomic64_t virt_timer_current;
static atomic64_t virt_timer_elapsed;
DEFINE_PER_CPU(u64, mt_cycles[8]);
static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 };
static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 };
static DEFINE_PER_CPU(u64, mt_scaling_jiffies);
static inline u64 get_vtimer(void)
{
u64 timer;
asm volatile("stpt %0" : "=m" (timer));
return timer;
}
static inline void set_vtimer(u64 expires)
{
u64 timer;
asm volatile(
" stpt %0\n" /* Store current cpu timer value */
" spt %1" /* Set new value imm. afterwards */
: "=m" (timer) : "m" (expires));
S390_lowcore.system_timer += S390_lowcore.last_update_timer - timer;
S390_lowcore.last_update_timer = expires;
}
static inline int virt_timer_forward(u64 elapsed)
{
BUG_ON(!irqs_disabled());
if (list_empty(&virt_timer_list))
return 0;
elapsed = atomic64_add_return(elapsed, &virt_timer_elapsed);
return elapsed >= atomic64_read(&virt_timer_current);
}
static void update_mt_scaling(void)
{
u64 cycles_new[8], *cycles_old;
u64 delta, fac, mult, div;
int i;
stcctm5(smp_cpu_mtid + 1, cycles_new);
cycles_old = this_cpu_ptr(mt_cycles);
fac = 1;
mult = div = 0;
for (i = 0; i <= smp_cpu_mtid; i++) {
delta = cycles_new[i] - cycles_old[i];
div += delta;
mult *= i + 1;
mult += delta * fac;
fac *= i + 1;
}
div *= fac;
if (div > 0) {
/* Update scaling factor */
__this_cpu_write(mt_scaling_mult, mult);
__this_cpu_write(mt_scaling_div, div);
memcpy(cycles_old, cycles_new,
sizeof(u64) * (smp_cpu_mtid + 1));
}
__this_cpu_write(mt_scaling_jiffies, jiffies_64);
}
static inline u64 update_tsk_timer(unsigned long *tsk_vtime, u64 new)
{
u64 delta;
delta = new - *tsk_vtime;
*tsk_vtime = new;
return delta;
}
static inline u64 scale_vtime(u64 vtime)
{
u64 mult = __this_cpu_read(mt_scaling_mult);
u64 div = __this_cpu_read(mt_scaling_div);
if (smp_cpu_mtid)
return vtime * mult / div;
return vtime;
}
static void account_system_index_scaled(struct task_struct *p, u64 cputime,
enum cpu_usage_stat index)
{
p->stimescaled += cputime_to_nsecs(scale_vtime(cputime));
account_system_index_time(p, cputime_to_nsecs(cputime), index);
}
/*
* Update process times based on virtual cpu times stored by entry.S
* to the lowcore fields user_timer, system_timer & steal_clock.
*/
static int do_account_vtime(struct task_struct *tsk)
{
u64 timer, clock, user, guest, system, hardirq, softirq, steal;
timer = S390_lowcore.last_update_timer;
clock = S390_lowcore.last_update_clock;
asm volatile(
" stpt %0\n" /* Store current cpu timer value */
#ifdef CONFIG_HAVE_MARCH_Z9_109_FEATURES
" stckf %1" /* Store current tod clock value */
#else
" stck %1" /* Store current tod clock value */
#endif
: "=m" (S390_lowcore.last_update_timer),
"=m" (S390_lowcore.last_update_clock));
clock = S390_lowcore.last_update_clock - clock;
timer -= S390_lowcore.last_update_timer;
if (hardirq_count())
S390_lowcore.hardirq_timer += timer;
else
S390_lowcore.system_timer += timer;
/* Update MT utilization calculation */
if (smp_cpu_mtid &&
time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies)))
update_mt_scaling();
/* Calculate cputime delta */
user = update_tsk_timer(&tsk->thread.user_timer,
READ_ONCE(S390_lowcore.user_timer));
guest = update_tsk_timer(&tsk->thread.guest_timer,
READ_ONCE(S390_lowcore.guest_timer));
system = update_tsk_timer(&tsk->thread.system_timer,
READ_ONCE(S390_lowcore.system_timer));
hardirq = update_tsk_timer(&tsk->thread.hardirq_timer,
READ_ONCE(S390_lowcore.hardirq_timer));
softirq = update_tsk_timer(&tsk->thread.softirq_timer,
READ_ONCE(S390_lowcore.softirq_timer));
S390_lowcore.steal_timer +=
clock - user - guest - system - hardirq - softirq;
/* Push account value */
if (user) {
account_user_time(tsk, cputime_to_nsecs(user));
tsk->utimescaled += cputime_to_nsecs(scale_vtime(user));
}
if (guest) {
account_guest_time(tsk, cputime_to_nsecs(guest));
tsk->utimescaled += cputime_to_nsecs(scale_vtime(guest));
}
if (system)
account_system_index_scaled(tsk, system, CPUTIME_SYSTEM);
if (hardirq)
account_system_index_scaled(tsk, hardirq, CPUTIME_IRQ);
if (softirq)
account_system_index_scaled(tsk, softirq, CPUTIME_SOFTIRQ);
steal = S390_lowcore.steal_timer;
if ((s64) steal > 0) {
S390_lowcore.steal_timer = 0;
account_steal_time(cputime_to_nsecs(steal));
}
return virt_timer_forward(user + guest + system + hardirq + softirq);
}
void vtime_task_switch(struct task_struct *prev)
{
do_account_vtime(prev);
prev->thread.user_timer = S390_lowcore.user_timer;
prev->thread.guest_timer = S390_lowcore.guest_timer;
prev->thread.system_timer = S390_lowcore.system_timer;
prev->thread.hardirq_timer = S390_lowcore.hardirq_timer;
prev->thread.softirq_timer = S390_lowcore.softirq_timer;
S390_lowcore.user_timer = current->thread.user_timer;
S390_lowcore.guest_timer = current->thread.guest_timer;
S390_lowcore.system_timer = current->thread.system_timer;
S390_lowcore.hardirq_timer = current->thread.hardirq_timer;
S390_lowcore.softirq_timer = current->thread.softirq_timer;
}
/*
* In s390, accounting pending user time also implies
* accounting system time in order to correctly compute
* the stolen time accounting.
*/
void vtime_flush(struct task_struct *tsk)
{
if (do_account_vtime(tsk))
virt_timer_expire();
}
/*
* Update process times based on virtual cpu times stored by entry.S
* to the lowcore fields user_timer, system_timer & steal_clock.
*/
void vtime_account_irq_enter(struct task_struct *tsk)
{
u64 timer;
timer = S390_lowcore.last_update_timer;
S390_lowcore.last_update_timer = get_vtimer();
timer -= S390_lowcore.last_update_timer;
if ((tsk->flags & PF_VCPU) && (irq_count() == 0))
S390_lowcore.guest_timer += timer;
else if (hardirq_count())
S390_lowcore.hardirq_timer += timer;
else if (in_serving_softirq())
S390_lowcore.softirq_timer += timer;
else
S390_lowcore.system_timer += timer;
virt_timer_forward(timer);
}
EXPORT_SYMBOL_GPL(vtime_account_irq_enter);
void vtime_account_system(struct task_struct *tsk)
__attribute__((alias("vtime_account_irq_enter")));
EXPORT_SYMBOL_GPL(vtime_account_system);
/*
* Sorted add to a list. List is linear searched until first bigger
* element is found.
*/
static void list_add_sorted(struct vtimer_list *timer, struct list_head *head)
{
struct vtimer_list *tmp;
list_for_each_entry(tmp, head, entry) {
if (tmp->expires > timer->expires) {
list_add_tail(&timer->entry, &tmp->entry);
return;
}
}
list_add_tail(&timer->entry, head);
}
/*
* Handler for expired virtual CPU timer.
*/
static void virt_timer_expire(void)
{
struct vtimer_list *timer, *tmp;
unsigned long elapsed;
LIST_HEAD(cb_list);
/* walk timer list, fire all expired timers */
spin_lock(&virt_timer_lock);
elapsed = atomic64_read(&virt_timer_elapsed);
list_for_each_entry_safe(timer, tmp, &virt_timer_list, entry) {
if (timer->expires < elapsed)
/* move expired timer to the callback queue */
list_move_tail(&timer->entry, &cb_list);
else
timer->expires -= elapsed;
}
if (!list_empty(&virt_timer_list)) {
timer = list_first_entry(&virt_timer_list,
struct vtimer_list, entry);
atomic64_set(&virt_timer_current, timer->expires);
}
atomic64_sub(elapsed, &virt_timer_elapsed);
spin_unlock(&virt_timer_lock);
/* Do callbacks and recharge periodic timers */
list_for_each_entry_safe(timer, tmp, &cb_list, entry) {
list_del_init(&timer->entry);
timer->function(timer->data);
if (timer->interval) {
/* Recharge interval timer */
timer->expires = timer->interval +
atomic64_read(&virt_timer_elapsed);
spin_lock(&virt_timer_lock);
list_add_sorted(timer, &virt_timer_list);
spin_unlock(&virt_timer_lock);
}
}
}
void init_virt_timer(struct vtimer_list *timer)
{
timer->function = NULL;
INIT_LIST_HEAD(&timer->entry);
}
EXPORT_SYMBOL(init_virt_timer);
static inline int vtimer_pending(struct vtimer_list *timer)
{
return !list_empty(&timer->entry);
}
static void internal_add_vtimer(struct vtimer_list *timer)
{
if (list_empty(&virt_timer_list)) {
/* First timer, just program it. */
atomic64_set(&virt_timer_current, timer->expires);
atomic64_set(&virt_timer_elapsed, 0);
list_add(&timer->entry, &virt_timer_list);
} else {
/* Update timer against current base. */
timer->expires += atomic64_read(&virt_timer_elapsed);
if (likely((s64) timer->expires <
(s64) atomic64_read(&virt_timer_current)))
/* The new timer expires before the current timer. */
atomic64_set(&virt_timer_current, timer->expires);
/* Insert new timer into the list. */
list_add_sorted(timer, &virt_timer_list);
}
}
static void __add_vtimer(struct vtimer_list *timer, int periodic)
{
unsigned long flags;
timer->interval = periodic ? timer->expires : 0;
spin_lock_irqsave(&virt_timer_lock, flags);
internal_add_vtimer(timer);
spin_unlock_irqrestore(&virt_timer_lock, flags);
}
/*
* add_virt_timer - add a oneshot virtual CPU timer
*/
void add_virt_timer(struct vtimer_list *timer)
{
__add_vtimer(timer, 0);
}
EXPORT_SYMBOL(add_virt_timer);
/*
* add_virt_timer_int - add an interval virtual CPU timer
*/
void add_virt_timer_periodic(struct vtimer_list *timer)
{
__add_vtimer(timer, 1);
}
EXPORT_SYMBOL(add_virt_timer_periodic);
static int __mod_vtimer(struct vtimer_list *timer, u64 expires, int periodic)
{
unsigned long flags;
int rc;
BUG_ON(!timer->function);
if (timer->expires == expires && vtimer_pending(timer))
return 1;
spin_lock_irqsave(&virt_timer_lock, flags);
rc = vtimer_pending(timer);
if (rc)
list_del_init(&timer->entry);
timer->interval = periodic ? expires : 0;
timer->expires = expires;
internal_add_vtimer(timer);
spin_unlock_irqrestore(&virt_timer_lock, flags);
return rc;
}
/*
* returns whether it has modified a pending timer (1) or not (0)
*/
int mod_virt_timer(struct vtimer_list *timer, u64 expires)
{
return __mod_vtimer(timer, expires, 0);
}
EXPORT_SYMBOL(mod_virt_timer);
/*
* returns whether it has modified a pending timer (1) or not (0)
*/
int mod_virt_timer_periodic(struct vtimer_list *timer, u64 expires)
{
return __mod_vtimer(timer, expires, 1);
}
EXPORT_SYMBOL(mod_virt_timer_periodic);
/*
* Delete a virtual timer.
*
* returns whether the deleted timer was pending (1) or not (0)
*/
int del_virt_timer(struct vtimer_list *timer)
{
unsigned long flags;
if (!vtimer_pending(timer))
return 0;
spin_lock_irqsave(&virt_timer_lock, flags);
list_del_init(&timer->entry);
spin_unlock_irqrestore(&virt_timer_lock, flags);
return 1;
}
EXPORT_SYMBOL(del_virt_timer);
/*
* Start the virtual CPU timer on the current CPU.
*/
void vtime_init(void)
{
/* set initial cpu timer */
set_vtimer(VTIMER_MAX_SLICE);
/* Setup initial MT scaling values */
if (smp_cpu_mtid) {
__this_cpu_write(mt_scaling_jiffies, jiffies);
__this_cpu_write(mt_scaling_mult, 1);
__this_cpu_write(mt_scaling_div, 1);
stcctm5(smp_cpu_mtid + 1, this_cpu_ptr(mt_cycles));
}
}