Initial commit: Linux kernel 4.19 for Unplugged OS

tools/testing/fault-injection/failcmd.sh

@@ -0,0 +1,220 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# NAME
+#	failcmd.sh - run a command with injecting slab/page allocation failures
+#
+# SYNOPSIS
+#	failcmd.sh --help
+#	failcmd.sh [<options>] command [arguments]
+#
+# DESCRIPTION
+#	Run command with injecting slab/page allocation failures by fault
+#	injection.
+#
+#	NOTE: you need to run this script as root.
+#
+
+usage()
+{
+	cat >&2 <<EOF
+Usage: $0 [options] command [arguments]
+
+OPTIONS
+	-p percent
+	--probability=percent
+		likelihood of failure injection, in percent.
+		Default value is 1
+
+	-t value
+	--times=value
+		specifies how many times failures may happen at most.
+		Default value is 1
+
+	--oom-kill-allocating-task=value
+		set /proc/sys/vm/oom_kill_allocating_task to specified value
+		before running the command.
+		Default value is 1
+
+	-h, --help
+		Display a usage message and exit
+
+	--interval=value, --space=value, --verbose=value, --task-filter=value,
+	--stacktrace-depth=value, --require-start=value, --require-end=value,
+	--reject-start=value, --reject-end=value, --ignore-gfp-wait=value
+		See Documentation/fault-injection/fault-injection.txt for more
+		information
+
+	failslab options:
+	--cache-filter=value
+
+	fail_page_alloc options:
+	--ignore-gfp-highmem=value, --min-order=value
+
+ENVIRONMENT
+	FAILCMD_TYPE
+		The following values for FAILCMD_TYPE are recognized:
+
+		failslab
+			inject slab allocation failures
+		fail_page_alloc
+			inject page allocation failures
+
+		If FAILCMD_TYPE is not defined, then failslab is used.
+EOF
+}
+
+if [ $UID != 0 ]; then
+	echo must be run as root >&2
+	exit 1
+fi
+
+DEBUGFS=`mount -t debugfs | head -1 | awk '{ print $3}'`
+
+if [ ! -d "$DEBUGFS" ]; then
+	echo debugfs is not mounted >&2
+	exit 1
+fi
+
+FAILCMD_TYPE=${FAILCMD_TYPE:-failslab}
+FAULTATTR=$DEBUGFS/$FAILCMD_TYPE
+
+if [ ! -d $FAULTATTR ]; then
+	echo $FAILCMD_TYPE is not available >&2
+	exit 1
+fi
+
+LONGOPTS=probability:,interval:,times:,space:,verbose:,task-filter:
+LONGOPTS=$LONGOPTS,stacktrace-depth:,require-start:,require-end:
+LONGOPTS=$LONGOPTS,reject-start:,reject-end:,oom-kill-allocating-task:,help
+
+if [ $FAILCMD_TYPE = failslab ]; then
+	LONGOPTS=$LONGOPTS,ignore-gfp-wait:,cache-filter:
+elif [ $FAILCMD_TYPE = fail_page_alloc ]; then
+	LONGOPTS=$LONGOPTS,ignore-gfp-wait:,ignore-gfp-highmem:,min-order:
+fi
+
+TEMP=`getopt -o p:i:t:s:v:h --long $LONGOPTS -n 'failcmd.sh' -- "$@"`
+
+if [ $? != 0 ]; then
+	usage
+	exit 1
+fi
+
+eval set -- "$TEMP"
+
+fault_attr_default()
+{
+	echo N > $FAULTATTR/task-filter
+	echo 0 > $FAULTATTR/probability
+	echo 1 > $FAULTATTR/times
+}
+
+fault_attr_default
+
+oom_kill_allocating_task_saved=`cat /proc/sys/vm/oom_kill_allocating_task`
+
+restore_values()
+{
+	fault_attr_default
+	echo $oom_kill_allocating_task_saved \
+		> /proc/sys/vm/oom_kill_allocating_task
+}
+
+#
+# Default options
+#
+declare -i oom_kill_allocating_task=1
+declare task_filter=Y
+declare -i probability=1
+declare -i times=1
+
+while true; do
+	case "$1" in
+	-p|--probability)
+		probability=$2
+		shift 2
+		;;
+	-i|--interval)
+		echo $2 > $FAULTATTR/interval
+		shift 2
+		;;
+	-t|--times)
+		times=$2
+		shift 2
+		;;
+	-s|--space)
+		echo $2 > $FAULTATTR/space
+		shift 2
+		;;
+	-v|--verbose)
+		echo $2 > $FAULTATTR/verbose
+		shift 2
+		;;
+	--task-filter)
+		task_filter=$2
+		shift 2
+		;;
+	--stacktrace-depth)
+		echo $2 > $FAULTATTR/stacktrace-depth
+		shift 2
+		;;
+	--require-start)
+		echo $2 > $FAULTATTR/require-start
+		shift 2
+		;;
+	--require-end)
+		echo $2 > $FAULTATTR/require-end
+		shift 2
+		;;
+	--reject-start)
+		echo $2 > $FAULTATTR/reject-start
+		shift 2
+		;;
+	--reject-end)
+		echo $2 > $FAULTATTR/reject-end
+		shift 2
+		;;
+	--oom-kill-allocating-task)
+		oom_kill_allocating_task=$2
+		shift 2
+		;;
+	--ignore-gfp-wait)
+		echo $2 > $FAULTATTR/ignore-gfp-wait
+		shift 2
+		;;
+	--cache-filter)
+		echo $2 > $FAULTATTR/cache_filter
+		shift 2
+		;;
+	--ignore-gfp-highmem)
+		echo $2 > $FAULTATTR/ignore-gfp-highmem
+		shift 2
+		;;
+	--min-order)
+		echo $2 > $FAULTATTR/min-order
+		shift 2
+		;;
+	-h|--help)
+		usage
+		exit 0
+		shift
+		;;
+	--)
+		shift
+		break
+		;;
+	esac
+done
+
+[ -z "$1" ] && exit 0
+
+echo $oom_kill_allocating_task > /proc/sys/vm/oom_kill_allocating_task
+echo $task_filter > $FAULTATTR/task-filter
+echo $probability > $FAULTATTR/probability
+echo $times > $FAULTATTR/times
+
+trap "restore_values" SIGINT SIGTERM EXIT
+
+cmd="echo 1 > /proc/self/make-it-fail && exec $@"
+bash -c "$cmd"

tools/testing/ktest/compare-ktest-sample.pl

@@ -0,0 +1,33 @@
+#!/usr/bin/env perl
+# SPDX-License-Identifier: GPL-2.0
+
+open (IN,"ktest.pl");
+while (<IN>) {
+    # hashes are now used
+    if (/\$opt\{"?([A-Z].*?)(\[.*\])?"?\}/ ||
+	/^\s*"?([A-Z].*?)"?\s*=>\s*/ ||
+	/set_test_option\("(.*?)"/) {
+	$opt{$1} = 1;
+    }
+}
+close IN;
+
+open (IN, "sample.conf");
+while (<IN>) {
+    if (/^\s*#?\s*([A-Z]\S*)\s*=/) {
+	$samp{$1} = 1;
+    }
+}
+close IN;
+
+foreach $opt (keys %opt) {
+    if (!defined($samp{$opt})) {
+	print "opt = $opt\n";
+    }
+}
+
+foreach $samp (keys %samp) {
+    if (!defined($opt{$samp})) {
+	print "samp = $samp\n";
+    }
+}

tools/testing/ktest/config-bisect.pl

@@ -0,0 +1,770 @@
+#!/usr/bin/perl -w
+#
+# Copyright 2015 - Steven Rostedt, Red Hat Inc.
+# Copyright 2017 - Steven Rostedt, VMware, Inc.
+#
+# Licensed under the terms of the GNU GPL License version 2
+#
+
+# usage:
+#  config-bisect.pl [options] good-config bad-config [good|bad]
+#
+
+# Compares a good config to a bad config, then takes half of the diffs
+# and produces a config that is somewhere between the good config and
+# the bad config. That is, the resulting config will start with the
+# good config and will try to make half of the differences of between
+# the good and bad configs match the bad config. It tries because of
+# dependencies between the two configs it may not be able to change
+# exactly half of the configs that are different between the two config
+# files.
+
+# Here's a normal way to use it:
+#
+#  $ cd /path/to/linux/kernel
+#  $ config-bisect.pl /path/to/good/config /path/to/bad/config
+
+# This will now pull in good config (blowing away .config in that directory
+# so do not make that be one of the good or bad configs), and then
+# build the config with "make oldconfig" to make sure it matches the
+# current kernel. It will then store the configs in that result for
+# the good config. It does the same for the bad config as well.
+# The algorithm will run, merging half of the differences between
+# the two configs and building them with "make oldconfig" to make sure
+# the result changes (dependencies may reset changes the tool had made).
+# It then copies the result of its good config to /path/to/good/config.tmp
+# and the bad config to /path/to/bad/config.tmp (just appends ".tmp" to the
+# files passed in). And the ".config" that you should test will be in
+# directory
+
+# After the first run, determine if the result is good or bad then
+# run the same command appending the result
+
+# For good results:
+#  $ config-bisect.pl /path/to/good/config /path/to/bad/config good
+
+# For bad results:
+#  $ config-bisect.pl /path/to/good/config /path/to/bad/config bad
+
+# Do not change the good-config or bad-config, config-bisect.pl will
+# copy the good-config to a temp file with the same name as good-config
+# but with a ".tmp" after it. It will do the same with the bad-config.
+
+# If "good" or "bad" is not stated at the end, it will copy the good and
+# bad configs to the .tmp versions. If a .tmp version already exists, it will
+# warn before writing over them (-r will not warn, and just write over them).
+# If the last config is labeled "good", then it will copy it to the good .tmp
+# version. If the last config is labeled "bad", it will copy it to the bad
+# .tmp version. It will continue this until it can not merge the two any more
+# without the result being equal to either the good or bad .tmp configs.
+
+my $start = 0;
+my $val = "";
+
+my $pwd = `pwd`;
+chomp $pwd;
+my $tree = $pwd;
+my $build;
+
+my $output_config;
+my $reset_bisect;
+
+sub usage {
+    print << "EOF"
+
+usage: config-bisect.pl [-l linux-tree][-b build-dir] good-config bad-config [good|bad]
+  -l [optional] define location of linux-tree (default is current directory)
+  -b [optional] define location to build (O=build-dir) (default is linux-tree)
+  good-config the config that is considered good
+  bad-config the config that does not work
+  "good" add this if the last run produced a good config
+  "bad" add this if the last run produced a bad config
+  If "good" or "bad" is not specified, then it is the start of a new bisect
+
+  Note, each run will create copy of good and bad configs with ".tmp" appended.
+
+EOF
+;
+
+    exit(-1);
+}
+
+sub doprint {
+    print @_;
+}
+
+sub dodie {
+    doprint "CRITICAL FAILURE... ", @_, "\n";
+
+    die @_, "\n";
+}
+
+sub expand_path {
+    my ($file) = @_;
+
+    if ($file =~ m,^/,) {
+	return $file;
+    }
+    return "$pwd/$file";
+}
+
+sub read_prompt {
+    my ($cancel, $prompt) = @_;
+
+    my $ans;
+
+    for (;;) {
+	if ($cancel) {
+	    print "$prompt [y/n/C] ";
+	} else {
+	    print "$prompt [y/N] ";
+	}
+	$ans = <STDIN>;
+	chomp $ans;
+	if ($ans =~ /^\s*$/) {
+	    if ($cancel) {
+		$ans = "c";
+	    } else {
+		$ans = "n";
+	    }
+	}
+	last if ($ans =~ /^y$/i || $ans =~ /^n$/i);
+	if ($cancel) {
+	    last if ($ans =~ /^c$/i);
+	    print "Please answer either 'y', 'n' or 'c'.\n";
+	} else {
+	    print "Please answer either 'y' or 'n'.\n";
+	}
+    }
+    if ($ans =~ /^c/i) {
+	exit;
+    }
+    if ($ans !~ /^y$/i) {
+	return 0;
+    }
+    return 1;
+}
+
+sub read_yn {
+    my ($prompt) = @_;
+
+    return read_prompt 0, $prompt;
+}
+
+sub read_ync {
+    my ($prompt) = @_;
+
+    return read_prompt 1, $prompt;
+}
+
+sub run_command {
+    my ($command, $redirect) = @_;
+    my $start_time;
+    my $end_time;
+    my $dord = 0;
+    my $pid;
+
+    $start_time = time;
+
+    doprint("$command ... ");
+
+    $pid = open(CMD, "$command 2>&1 |") or
+	dodie "unable to exec $command";
+
+    if (defined($redirect)) {
+	open (RD, ">$redirect") or
+	    dodie "failed to write to redirect $redirect";
+	$dord = 1;
+    }
+
+    while (<CMD>) {
+	print RD  if ($dord);
+    }
+
+    waitpid($pid, 0);
+    my $failed = $?;
+
+    close(CMD);
+    close(RD)  if ($dord);
+
+    $end_time = time;
+    my $delta = $end_time - $start_time;
+
+    if ($delta == 1) {
+	doprint "[1 second] ";
+    } else {
+	doprint "[$delta seconds] ";
+    }
+
+    if ($failed) {
+	doprint "FAILED!\n";
+    } else {
+	doprint "SUCCESS\n";
+    }
+
+    return !$failed;
+}
+
+###### CONFIG BISECT ######
+
+# config_ignore holds the configs that were set (or unset) for
+# a good config and we will ignore these configs for the rest
+# of a config bisect. These configs stay as they were.
+my %config_ignore;
+
+# config_set holds what all configs were set as.
+my %config_set;
+
+# config_off holds the set of configs that the bad config had disabled.
+# We need to record them and set them in the .config when running
+# olddefconfig, because olddefconfig keeps the defaults.
+my %config_off;
+
+# config_off_tmp holds a set of configs to turn off for now
+my @config_off_tmp;
+
+# config_list is the set of configs that are being tested
+my %config_list;
+my %null_config;
+
+my %dependency;
+
+my $make;
+
+sub make_oldconfig {
+
+    if (!run_command "$make olddefconfig") {
+	# Perhaps olddefconfig doesn't exist in this version of the kernel
+	# try oldnoconfig
+	doprint "olddefconfig failed, trying make oldnoconfig\n";
+	if (!run_command "$make oldnoconfig") {
+	    doprint "oldnoconfig failed, trying yes '' | make oldconfig\n";
+	    # try a yes '' | oldconfig
+	    run_command "yes '' | $make oldconfig" or
+		dodie "failed make config oldconfig";
+	}
+    }
+}
+
+sub assign_configs {
+    my ($hash, $config) = @_;
+
+    doprint "Reading configs from $config\n";
+
+    open (IN, $config)
+	or dodie "Failed to read $config";
+
+    while (<IN>) {
+	chomp;
+	if (/^((CONFIG\S*)=.*)/) {
+	    ${$hash}{$2} = $1;
+	} elsif (/^(# (CONFIG\S*) is not set)/) {
+	    ${$hash}{$2} = $1;
+	}
+    }
+
+    close(IN);
+}
+
+sub process_config_ignore {
+    my ($config) = @_;
+
+    assign_configs \%config_ignore, $config;
+}
+
+sub get_dependencies {
+    my ($config) = @_;
+
+    my $arr = $dependency{$config};
+    if (!defined($arr)) {
+	return ();
+    }
+
+    my @deps = @{$arr};
+
+    foreach my $dep (@{$arr}) {
+	print "ADD DEP $dep\n";
+	@deps = (@deps, get_dependencies $dep);
+    }
+
+    return @deps;
+}
+
+sub save_config {
+    my ($pc, $file) = @_;
+
+    my %configs = %{$pc};
+
+    doprint "Saving configs into $file\n";
+
+    open(OUT, ">$file") or dodie "Can not write to $file";
+
+    foreach my $config (keys %configs) {
+	print OUT "$configs{$config}\n";
+    }
+    close(OUT);
+}
+
+sub create_config {
+    my ($name, $pc) = @_;
+
+    doprint "Creating old config from $name configs\n";
+
+    save_config $pc, $output_config;
+
+    make_oldconfig;
+}
+
+# compare two config hashes, and return configs with different vals.
+# It returns B's config values, but you can use A to see what A was.
+sub diff_config_vals {
+    my ($pa, $pb) = @_;
+
+    # crappy Perl way to pass in hashes.
+    my %a = %{$pa};
+    my %b = %{$pb};
+
+    my %ret;
+
+    foreach my $item (keys %a) {
+	if (defined($b{$item}) && $b{$item} ne $a{$item}) {
+	    $ret{$item} = $b{$item};
+	}
+    }
+
+    return %ret;
+}
+
+# compare two config hashes and return the configs in B but not A
+sub diff_configs {
+    my ($pa, $pb) = @_;
+
+    my %ret;
+
+    # crappy Perl way to pass in hashes.
+    my %a = %{$pa};
+    my %b = %{$pb};
+
+    foreach my $item (keys %b) {
+	if (!defined($a{$item})) {
+	    $ret{$item} = $b{$item};
+	}
+    }
+
+    return %ret;
+}
+
+# return if two configs are equal or not
+# 0 is equal +1 b has something a does not
+# +1 if a and b have a different item.
+# -1 if a has something b does not
+sub compare_configs {
+    my ($pa, $pb) = @_;
+
+    my %ret;
+
+    # crappy Perl way to pass in hashes.
+    my %a = %{$pa};
+    my %b = %{$pb};
+
+    foreach my $item (keys %b) {
+	if (!defined($a{$item})) {
+	    return 1;
+	}
+	if ($a{$item} ne $b{$item}) {
+	    return 1;
+	}
+    }
+
+    foreach my $item (keys %a) {
+	if (!defined($b{$item})) {
+	    return -1;
+	}
+    }
+
+    return 0;
+}
+
+sub process_failed {
+    my ($config) = @_;
+
+    doprint "\n\n***************************************\n";
+    doprint "Found bad config: $config\n";
+    doprint "***************************************\n\n";
+}
+
+sub process_new_config {
+    my ($tc, $nc, $gc, $bc) = @_;
+
+    my %tmp_config = %{$tc};
+    my %good_configs = %{$gc};
+    my %bad_configs = %{$bc};
+
+    my %new_configs;
+
+    my $runtest = 1;
+    my $ret;
+
+    create_config "tmp_configs", \%tmp_config;
+    assign_configs \%new_configs, $output_config;
+
+    $ret = compare_configs \%new_configs, \%bad_configs;
+    if (!$ret) {
+	doprint "New config equals bad config, try next test\n";
+	$runtest = 0;
+    }
+
+    if ($runtest) {
+	$ret = compare_configs \%new_configs, \%good_configs;
+	if (!$ret) {
+	    doprint "New config equals good config, try next test\n";
+	    $runtest = 0;
+	}
+    }
+
+    %{$nc} = %new_configs;
+
+    return $runtest;
+}
+
+sub convert_config {
+    my ($config) = @_;
+
+    if ($config =~ /^# (.*) is not set/) {
+	$config = "$1=n";
+    }
+
+    $config =~ s/^CONFIG_//;
+    return $config;
+}
+
+sub print_config {
+    my ($sym, $config) = @_;
+
+    $config = convert_config $config;
+    doprint "$sym$config\n";
+}
+
+sub print_config_compare {
+    my ($good_config, $bad_config) = @_;
+
+    $good_config = convert_config $good_config;
+    $bad_config = convert_config $bad_config;
+
+    my $good_value = $good_config;
+    my $bad_value = $bad_config;
+    $good_value =~ s/(.*)=//;
+    my $config = $1;
+
+    $bad_value =~ s/.*=//;
+
+    doprint " $config $good_value -> $bad_value\n";
+}
+
+# Pass in:
+# $phalf: half of the configs names you want to add
+# $oconfigs: The orginial configs to start with
+# $sconfigs: The source to update $oconfigs with (from $phalf)
+# $which: The name of which half that is updating (top / bottom)
+# $type: The name of the source type (good / bad)
+sub make_half {
+    my ($phalf, $oconfigs, $sconfigs, $which, $type) = @_;
+
+    my @half = @{$phalf};
+    my %orig_configs = %{$oconfigs};
+    my %source_configs = %{$sconfigs};
+
+    my %tmp_config = %orig_configs;
+
+    doprint "Settings bisect with $which half of $type configs:\n";
+    foreach my $item (@half) {
+	doprint "Updating $item to $source_configs{$item}\n";
+	$tmp_config{$item} = $source_configs{$item};
+    }
+
+    return %tmp_config;
+}
+
+sub run_config_bisect {
+    my ($pgood, $pbad) = @_;
+
+    my %good_configs = %{$pgood};
+    my %bad_configs = %{$pbad};
+
+    my %diff_configs = diff_config_vals \%good_configs, \%bad_configs;
+    my %b_configs = diff_configs \%good_configs, \%bad_configs;
+    my %g_configs = diff_configs \%bad_configs, \%good_configs;
+
+    # diff_arr is what is in both good and bad but are different (y->n)
+    my @diff_arr = keys %diff_configs;
+    my $len_diff = $#diff_arr + 1;
+
+    # b_arr is what is in bad but not in good (has depends)
+    my @b_arr = keys %b_configs;
+    my $len_b = $#b_arr + 1;
+
+    # g_arr is what is in good but not in bad
+    my @g_arr = keys %g_configs;
+    my $len_g = $#g_arr + 1;
+
+    my $runtest = 0;
+    my %new_configs;
+    my $ret;
+
+    # Look at the configs that are different between good and bad.
+    # This does not include those that depend on other configs
+    #  (configs depending on other configs that are not set would
+    #   not show up even as a "# CONFIG_FOO is not set"
+
+
+    doprint "# of configs to check:             $len_diff\n";
+    doprint "# of configs showing only in good: $len_g\n";
+    doprint "# of configs showing only in bad:  $len_b\n";
+
+    if ($len_diff > 0) {
+	# Now test for different values
+
+	doprint "Configs left to check:\n";
+	doprint "  Good Config\t\t\tBad Config\n";
+	doprint "  -----------\t\t\t----------\n";
+	foreach my $item (@diff_arr) {
+	    doprint "  $good_configs{$item}\t$bad_configs{$item}\n";
+	}
+
+	my $half = int($#diff_arr / 2);
+	my @tophalf = @diff_arr[0 .. $half];
+
+	doprint "Set tmp config to be good config with some bad config values\n";
+
+	my %tmp_config = make_half \@tophalf, \%good_configs,
+	    \%bad_configs, "top", "bad";
+
+	$runtest = process_new_config \%tmp_config, \%new_configs,
+			    \%good_configs, \%bad_configs;
+
+	if (!$runtest) {
+	    doprint "Set tmp config to be bad config with some good config values\n";
+
+	    my %tmp_config = make_half \@tophalf, \%bad_configs,
+		\%good_configs, "top", "good";
+
+	    $runtest = process_new_config \%tmp_config, \%new_configs,
+		\%good_configs, \%bad_configs;
+	}
+    }
+
+    if (!$runtest && $len_diff > 0) {
+	# do the same thing, but this time with bottom half
+
+	my $half = int($#diff_arr / 2);
+	my @bottomhalf = @diff_arr[$half+1 .. $#diff_arr];
+
+	doprint "Set tmp config to be good config with some bad config values\n";
+
+	my %tmp_config = make_half \@bottomhalf, \%good_configs,
+	    \%bad_configs, "bottom", "bad";
+
+	$runtest = process_new_config \%tmp_config, \%new_configs,
+			    \%good_configs, \%bad_configs;
+
+	if (!$runtest) {
+	    doprint "Set tmp config to be bad config with some good config values\n";
+
+	    my %tmp_config = make_half \@bottomhalf, \%bad_configs,
+		\%good_configs, "bottom", "good";
+
+	    $runtest = process_new_config \%tmp_config, \%new_configs,
+		\%good_configs, \%bad_configs;
+	}
+    }
+
+    if ($runtest) {
+	make_oldconfig;
+	doprint "READY TO TEST .config IN $build\n";
+	return 0;
+    }
+
+    doprint "\n%%%%%%%% FAILED TO FIND SINGLE BAD CONFIG %%%%%%%%\n";
+    doprint "Hmm, can't make any more changes without making good == bad?\n";
+    doprint "Difference between good (+) and bad (-)\n";
+
+    foreach my $item (keys %bad_configs) {
+	if (!defined($good_configs{$item})) {
+	    print_config "-", $bad_configs{$item};
+	}
+    }
+
+    foreach my $item (keys %good_configs) {
+	next if (!defined($bad_configs{$item}));
+	if ($good_configs{$item} ne $bad_configs{$item}) {
+	    print_config_compare $good_configs{$item}, $bad_configs{$item};
+	}
+    }
+
+    foreach my $item (keys %good_configs) {
+	if (!defined($bad_configs{$item})) {
+	    print_config "+", $good_configs{$item};
+	}
+    }
+    return -1;
+}
+
+sub config_bisect {
+    my ($good_config, $bad_config) = @_;
+    my $ret;
+
+    my %good_configs;
+    my %bad_configs;
+    my %tmp_configs;
+
+    doprint "Run good configs through make oldconfig\n";
+    assign_configs \%tmp_configs, $good_config;
+    create_config "$good_config", \%tmp_configs;
+    assign_configs \%good_configs, $output_config;
+
+    doprint "Run bad configs through make oldconfig\n";
+    assign_configs \%tmp_configs, $bad_config;
+    create_config "$bad_config", \%tmp_configs;
+    assign_configs \%bad_configs, $output_config;
+
+    save_config \%good_configs, $good_config;
+    save_config \%bad_configs, $bad_config;
+
+    return run_config_bisect \%good_configs, \%bad_configs;
+}
+
+while ($#ARGV >= 0) {
+    if ($ARGV[0] !~ m/^-/) {
+	last;
+    }
+    my $opt = shift @ARGV;
+
+    if ($opt eq "-b") {
+	$val = shift @ARGV;
+	if (!defined($val)) {
+	    die "-b requires value\n";
+	}
+	$build = $val;
+    }
+
+    elsif ($opt eq "-l") {
+	$val = shift @ARGV;
+	if (!defined($val)) {
+	    die "-l requires value\n";
+	}
+	$tree = $val;
+    }
+
+    elsif ($opt eq "-r") {
+	$reset_bisect = 1;
+    }
+
+    elsif ($opt eq "-h") {
+	usage;
+    }
+
+    else {
+	die "Unknow option $opt\n";
+    }
+}
+
+$build = $tree if (!defined($build));
+
+$tree = expand_path $tree;
+$build = expand_path $build;
+
+if ( ! -d $tree ) {
+    die "$tree not a directory\n";
+}
+
+if ( ! -d $build ) {
+    die "$build not a directory\n";
+}
+
+usage if $#ARGV < 1;
+
+if ($#ARGV == 1) {
+    $start = 1;
+} elsif ($#ARGV == 2) {
+    $val = $ARGV[2];
+    if ($val ne "good" && $val ne "bad") {
+	die "Unknown command '$val', bust be either \"good\" or \"bad\"\n";
+    }
+} else {
+    usage;
+}
+
+my $good_start = expand_path $ARGV[0];
+my $bad_start = expand_path $ARGV[1];
+
+my $good = "$good_start.tmp";
+my $bad = "$bad_start.tmp";
+
+$make = "make";
+
+if ($build ne $tree) {
+    $make = "make O=$build"
+}
+
+$output_config = "$build/.config";
+
+if ($start) {
+    if ( ! -f $good_start ) {
+	die "$good_start not found\n";
+    }
+    if ( ! -f $bad_start ) {
+	die "$bad_start not found\n";
+    }
+    if ( -f $good || -f $bad ) {
+	my $p = "";
+
+	if ( -f $good ) {
+	    $p = "$good exists\n";
+	}
+
+	if ( -f $bad ) {
+	    $p = "$p$bad exists\n";
+	}
+
+	if (!defined($reset_bisect)) {
+	    if (!read_yn "${p}Overwrite and start new bisect anyway?") {
+		exit (-1);
+	    }
+	}
+    }
+    run_command "cp $good_start $good" or die "failed to copy to $good\n";
+    run_command "cp $bad_start $bad" or die "faield to copy to $bad\n";
+} else {
+    if ( ! -f $good ) {
+	die "Can not find file $good\n";
+    }
+    if ( ! -f $bad ) {
+	die "Can not find file $bad\n";
+    }
+    if ($val eq "good") {
+	run_command "cp $output_config $good" or die "failed to copy $config to $good\n";
+    } elsif ($val eq "bad") {
+	run_command "cp $output_config $bad" or die "failed to copy $config to $bad\n";
+    }
+}
+
+chdir $tree || die "can't change directory to $tree";
+
+my $ret = config_bisect $good, $bad;
+
+if (!$ret) {
+    exit(0);
+}
+
+if ($ret > 0) {
+    doprint "Cleaning temp files\n";
+    run_command "rm $good";
+    run_command "rm $bad";
+    exit(1);
+} else {
+    doprint "See good and bad configs for details:\n";
+    doprint "good: $good\n";
+    doprint "bad:  $bad\n";
+    doprint "%%%%%%%% FAILED TO FIND SINGLE BAD CONFIG %%%%%%%%\n";
+}
+exit(2);

tools/testing/ktest/examples/README

@@ -0,0 +1,32 @@
+This directory contains example configs to use ktest for various tasks.
+The configs still need to be customized for your environment, but it
+is broken up by task which makes it easier to understand how to set up
+ktest.
+
+The configs are based off of real working configs but have been modified
+and commented to show more generic use cases that are more helpful for
+developers.
+
+crosstests.conf - this config shows an example of testing a git repo against
+    lots of different architectures. It only does build tests, but makes
+    it easy to compile test different archs. You can download the arch
+    cross compilers from:
+  http://kernel.org/pub/tools/crosstool/files/bin/x86_64/
+
+test.conf - A generic example of a config. This is based on an actual config
+     used to perform real testing.
+
+kvm.conf - A example of a config that is used to test a virtual guest running
+     on a host.
+
+snowball.conf - An example config that was used to demo ktest.pl against
+     a snowball ARM board.
+
+include/  -  The include directory holds default configs that can be
+    included into other configs. This is a real use example that shows how
+    to reuse configs for various machines or set ups. The files here
+    are included by other config files, where the other config files define
+    options and variables that will make the included config work for the
+    given environment.
+
+

tools/testing/ktest/examples/crosstests.conf

@@ -0,0 +1,225 @@
+#
+# Example config for cross compiling
+#
+# In this config, it is expected that the tool chains from:
+#
+#   http://kernel.org/pub/tools/crosstool/files/bin/x86_64/
+#
+# running on a x86_64 system have been downloaded and installed into:
+#
+#   /usr/local/
+#
+# such that the compiler binaries are something like:
+#
+#   /usr/local/gcc-4.5.2-nolibc/mips-linux/bin/mips-linux-gcc
+#
+# Some of the archs will use gcc-4.5.1 instead of gcc-4.5.2
+# this config uses variables to differentiate them.
+# 
+# Comments describe some of the options, but full descriptions of
+# options are described in the samples.conf file.
+
+# ${PWD} is defined by ktest.pl to be the directory that the user
+# was in when they executed ktest.pl. It may be better to hardcode the
+# path name here. THIS_DIR is the variable used through out the config file
+# in case you want to change it.
+
+THIS_DIR := ${PWD}
+
+# Update the BUILD_DIR option to the location of your git repo you want to test.
+BUILD_DIR = ${THIS_DIR}/linux.git
+
+# The build will go into this directory. It will be created when you run the test.
+OUTPUT_DIR = ${THIS_DIR}/cross-compile
+
+# The build will be compiled with -j8
+BUILD_OPTIONS = -j8
+
+# The test will not stop when it hits a failure.
+DIE_ON_FAILURE = 0
+
+# If you want to have ktest.pl store the failure somewhere, uncomment this option
+# and change the directory where ktest should store the failures.
+#STORE_FAILURES = ${THIS_DIR}/failures
+
+# The log file is stored in the OUTPUT_DIR called cross.log
+# If you enable this, you need to create the OUTPUT_DIR. It wont be created for you.
+LOG_FILE = ${OUTPUT_DIR}/cross.log
+
+# The log file will be cleared each time you run ktest.
+CLEAR_LOG = 1
+
+# As some archs do not build with the defconfig, they have been marked
+# to be ignored. If you want to test them anyway, change DO_FAILED to 1.
+# If a test that has been marked as DO_FAILED passes, then you should change
+# that test to be DO_DEFAULT
+
+DO_FAILED := 0
+DO_DEFAULT := 1
+
+# By setting both DO_FAILED and DO_DEFAULT to zero, you can pick a single
+# arch that you want to test. (uncomment RUN and chose your arch)
+#RUN := arm
+
+# At the bottom of the config file exists a bisect test. You can update that
+# test and set DO_FAILED and DO_DEFAULT to zero, and uncomment this variable
+# to run the bisect on the arch.
+#RUN := bisect
+
+# By default all tests will be running gcc 4.5.2. Some tests are using 4.5.1
+# and they select that in the test.
+# Note: GCC_VER is declared as on option and not a variable ('=' instead of ':=')
+# This is important. A variable is used only in the config file and if it is set
+# it stays that way for the rest of the config file until it is change again.
+# Here we want GCC_VER to remain persistent and change for each test, as it is used in
+# the MAKE_CMD. By using '=' instead of ':=' we achieve our goal.
+
+GCC_VER = 4.5.2
+MAKE_CMD = PATH=/usr/local/gcc-${GCC_VER}-nolibc/${CROSS}/bin:$PATH CROSS_COMPILE=${CROSS}- make ARCH=${ARCH}
+
+# all tests are only doing builds.
+TEST_TYPE = build
+
+# If you want to add configs on top of the defconfig, you can add those configs into
+# the add-config file and uncomment this option. This is useful if you want to test
+# all cross compiles with PREEMPT set, or TRACING on, etc.
+#ADD_CONFIG = ${THIS_DIR}/add-config
+
+# All tests are using defconfig
+BUILD_TYPE = defconfig
+
+# The test names will have the arch and cross compiler used. This will be shown in
+# the results.
+TEST_NAME = ${ARCH} ${CROSS}
+
+# alpha
+TEST_START IF ${RUN} == alpha || ${DO_DEFAULT}
+# Notice that CROSS and ARCH are also options and not variables (again '=' instead
+# of ':='). This is because TEST_NAME and MAKE_CMD wil use them for each test.
+# Only options are available during runs. Variables are only present in parsing the
+# config file.
+CROSS = alpha-linux
+ARCH = alpha
+
+# arm
+TEST_START IF ${RUN} == arm || ${DO_DEFAULT}
+CROSS = arm-unknown-linux-gnueabi
+ARCH = arm
+
+# ia64
+TEST_START IF ${RUN} == ia64 || ${DO_DEFAULT}
+CROSS = ia64-linux
+ARCH = ia64
+
+# m68k fails with error?
+TEST_START IF ${RUN} == m68k || ${DO_DEFAULT}
+CROSS = m68k-linux
+ARCH = m68k
+
+# mips64
+TEST_START IF ${RUN} == mips || ${RUN} == mips64 || ${DO_DEFAULT}
+CROSS = mips64-linux
+ARCH = mips
+
+# mips32
+TEST_START IF ${RUN} == mips || ${RUN} == mips32 || ${DO_DEFAULT}
+CROSS = mips-linux
+ARCH = mips
+
+# parisc64 failed?
+TEST_START IF ${RUN} == hppa || ${RUN} == hppa64 || ${DO_FAILED}
+CROSS = hppa64-linux
+ARCH = parisc
+
+# parisc
+TEST_START IF ${RUN} == hppa || ${RUN} == hppa32 || ${DO_FAILED}
+CROSS = hppa-linux
+ARCH = parisc
+
+# ppc
+TEST_START IF ${RUN} == ppc || ${RUN} == ppc32 || ${DO_DEFAULT}
+CROSS = powerpc-linux
+ARCH = powerpc
+
+# ppc64
+TEST_START IF ${RUN} == ppc || ${RUN} == ppc64 || ${DO_DEFAULT}
+CROSS = powerpc64-linux
+ARCH = powerpc
+
+# s390
+TEST_START IF ${RUN} == s390 || ${DO_DEFAULT}
+CROSS = s390x-linux
+ARCH = s390
+
+# sh
+TEST_START IF ${RUN} == sh || ${DO_DEFAULT}
+CROSS = sh4-linux
+ARCH = sh
+
+# sparc64
+TEST_START IF ${RUN} == sparc || ${RUN} == sparc64 || ${DO_DEFAULT}
+CROSS = sparc64-linux
+ARCH = sparc64
+
+# sparc
+TEST_START IF ${RUN} == sparc || ${RUN} == sparc32 || ${DO_DEFAULT}
+CROSS = sparc-linux
+ARCH = sparc
+
+# xtensa failed
+TEST_START IF ${RUN} == xtensa || ${DO_FAILED}
+CROSS = xtensa-linux
+ARCH = xtensa
+
+# UML
+TEST_START IF ${RUN} == uml || ${DO_DEFAULT}
+MAKE_CMD = make ARCH=um SUBARCH=x86_64
+ARCH = uml
+CROSS =
+
+TEST_START IF ${RUN} == x86 || ${RUN} == i386 || ${DO_DEFAULT}
+MAKE_CMD = make ARCH=i386
+ARCH = i386
+CROSS = 
+
+TEST_START IF ${RUN} == x86 || ${RUN} == x86_64 || ${DO_DEFAULT}
+MAKE_CMD = make ARCH=x86_64
+ARCH = x86_64
+CROSS = 
+
+#################################
+
+# This is a bisect if needed. You need to give it a MIN_CONFIG that
+# will be the config file it uses. Basically, just copy the created defconfig
+# for the arch someplace and point MIN_CONFIG to it.
+TEST_START IF ${RUN} == bisect
+MIN_CONFIG = ${THIS_DIR}/min-config
+CROSS = s390x-linux
+ARCH = s390
+TEST_TYPE = bisect
+BISECT_TYPE = build
+BISECT_GOOD = v3.1
+BISECT_BAD = v3.2
+CHECKOUT = v3.2
+
+#################################
+
+# These defaults are needed to keep ktest.pl from complaining. They are
+# ignored because the test does not go pass the build. No install or
+# booting of the target images.
+
+DEFAULTS
+MACHINE = crosstest
+SSH_USER = root
+BUILD_TARGET = cross
+TARGET_IMAGE = image
+POWER_CYCLE = cycle
+CONSOLE = console
+LOCALVERSION = version
+GRUB_MENU = grub
+
+REBOOT_ON_ERROR = 0
+POWEROFF_ON_ERROR = 0
+POWEROFF_ON_SUCCESS = 0
+REBOOT_ON_SUCCESS = 0
+

tools/testing/ktest/examples/include/bisect.conf

@@ -0,0 +1,90 @@
+#
+# This example shows the bisect tests (git bisect and config bisect)
+#
+
+
+# The config that includes this file may define a RUN_TEST
+# variable that will tell this config what test to run.
+# (what to set the TEST option to).
+#
+DEFAULTS IF NOT DEFINED RUN_TEST
+# Requires that hackbench is in the PATH
+RUN_TEST := ${SSH} hackbench 50
+
+
+# Set TEST to 'bisect' to do a normal git bisect. You need
+# to modify the options below to make it bisect the exact
+# commits you are interested in.
+#
+TEST_START IF ${TEST} == bisect
+TEST_TYPE = bisect
+# You must set the commit that was considered good (git bisect good)
+BISECT_GOOD = v3.3
+# You must set the commit that was considered bad (git bisect bad)
+BISECT_BAD = HEAD
+# It's best to specify the branch to checkout before starting the bisect.
+CHECKOUT = origin/master
+# This can be build, boot, or test. Here we are doing a bisect
+# that requires to run a test to know if the bisect was good or bad.
+# The test should exit with 0 on good, non-zero for bad. But see
+# the BISECT_RET_* options in samples.conf to override this.
+BISECT_TYPE = test
+TEST = ${RUN_TEST}
+# It is usually a good idea to confirm that the GOOD and the BAD
+# commits are truly good and bad respectively. Having BISECT_CHECK
+# set to 1 will check both that the good commit works and the bad
+# commit fails. If you only want to check one or the other,
+# set BISECT_CHECK to 'good' or to 'bad'.
+BISECT_CHECK = 1
+#BISECT_CHECK = good
+#BISECT_CHECK = bad
+
+# Usually it's a good idea to specify the exact config you
+# want to use throughout the entire bisect. Here we placed
+# it in the directory we called ktest.pl from and named it
+# 'config-bisect'.
+MIN_CONFIG = ${THIS_DIR}/config-bisect
+# By default, if we are doing a BISECT_TYPE = test run but the
+# build or boot fails, ktest.pl will do a 'git bisect skip'.
+# Uncomment the below option to make ktest stop testing on such
+# an error.
+#BISECT_SKIP = 0
+# Now if you had BISECT_SKIP = 0 and the test fails, you can
+# examine what happened and then do 'git bisect log > /tmp/replay'
+# Set BISECT_REPLAY to /tmp/replay and ktest.pl will run the
+# 'git bisect replay /tmp/replay' before continuing the bisect test.
+#BISECT_REPLAY = /tmp/replay
+# If you used BISECT_REPLAY after the bisect test failed, you may
+# not want to continue the bisect on that commit that failed.
+# By setting BISECT_START to a new commit. ktest.pl will checkout
+# that commit after it has performed the 'git bisect replay' but
+# before it continues running the bisect test.
+#BISECT_START = 2545eb6198e7e1ec50daa0cfc64a4cdfecf24ec9
+
+# Now if you don't trust ktest.pl to make the decisions for you, then
+# set BISECT_MANUAL to 1. This will cause ktest.pl not to decide
+# if the commit was good or bad. Instead, it will ask you to tell
+# it if the current commit was good. In the mean time, you could
+# take the result, load it on any machine you want. Run several tests,
+# or whatever you feel like. Then, when you are happy, you can tell
+# ktest if you think it was good or not and ktest.pl will continue
+# the git bisect. You can even change what commit it is currently at.
+#BISECT_MANUAL = 1
+
+
+# One of the unique tests that ktest does is the config bisect.
+# Currently (which hopefully will be fixed soon), the bad config
+# must be a superset of the good config. This is because it only
+# searches for a config that causes the target to fail. If the
+# good config is not a subset of the bad config, or if the target
+# fails because of a lack of a config, then it will not find
+# the config for you.
+TEST_START IF ${TEST} == config-bisect
+TEST_TYPE = config_bisect
+# set to build, boot, test
+CONFIG_BISECT_TYPE = boot
+# Set the config that is considered bad.
+CONFIG_BISECT = ${THIS_DIR}/config-bad
+# This config is optional. By default it uses the
+# MIN_CONFIG as the good config.
+CONFIG_BISECT_GOOD = ${THIS_DIR}/config-good

tools/testing/ktest/examples/include/defaults.conf

@@ -0,0 +1,157 @@
+# This file holds defaults for most the tests. It defines the options that
+# are most common to tests that are likely to be shared.
+#
+# Note, after including this file, a config file may override any option
+# with a DEFAULTS OVERRIDE section.
+#
+
+# For those cases that use the same machine to boot a 64 bit
+# and a 32 bit version. The MACHINE is the DNS name to get to the
+# box (usually different if it was 64 bit or 32 bit) but the
+# BOX here is defined as a variable that will be the name of the box
+# itself. It is useful for calling scripts that will power cycle
+# the box, as only one script needs to be created to power cycle
+# even though the box itself has multiple operating systems on it.
+# By default, BOX and MACHINE are the same.
+
+DEFAULTS IF NOT DEFINED BOX
+BOX := ${MACHINE}
+
+
+# Consider each box as 64 bit box, unless the config including this file
+# has defined BITS = 32
+
+DEFAULTS IF NOT DEFINED BITS
+BITS := 64
+
+
+DEFAULTS
+
+# THIS_DIR is used through out the configs and defaults to ${PWD} which
+# is the directory that ktest.pl was called from.
+
+THIS_DIR := ${PWD}
+
+
+# to organize your configs, having each machine save their configs
+# into a separate directly is useful.
+CONFIG_DIR := ${THIS_DIR}/configs/${MACHINE}
+
+# Reset the log before running each test.
+CLEAR_LOG = 1
+
+# As installing kernels usually requires root privilege, default the
+# user on the target as root. It is also required that the target
+# allows ssh to root from the host without asking for a password.
+
+SSH_USER = root
+
+# For accesing the machine, we will ssh to root@machine.
+SSH := ssh ${SSH_USER}@${MACHINE}
+
+# Update this. The default here is ktest will ssh to the target box
+# and run a script called 'run-test' located on that box.
+TEST = ${SSH} run-test
+
+# Point build dir to the git repo you use
+BUILD_DIR = ${THIS_DIR}/linux.git
+
+# Each machine will have its own output build directory.
+OUTPUT_DIR = ${THIS_DIR}/build/${MACHINE}
+
+# Yes this config is focused on x86 (but ktest works for other archs too)
+BUILD_TARGET = arch/x86/boot/bzImage
+TARGET_IMAGE = /boot/vmlinuz-test
+
+# have directory for the scripts to reboot and power cycle the boxes
+SCRIPTS_DIR := ${THIS_DIR}/scripts
+
+# You can have each box/machine have a script to power cycle it.
+# Name your script <box>-cycle.
+POWER_CYCLE = ${SCRIPTS_DIR}/${BOX}-cycle
+
+# This script is used to power off the box.
+POWER_OFF = ${SCRIPTS_DIR}/${BOX}-poweroff
+
+# Keep your test kernels separate from your other kernels.
+LOCALVERSION = -test
+
+# The /boot/grub/menu.lst is searched for the line:
+#  title Test Kernel
+# and ktest will use that kernel to reboot into.
+# For grub2 or other boot loaders, you need to set BOOT_TYPE
+# to 'script' and define other ways to load the kernel.
+# See snowball.conf example.
+#
+GRUB_MENU = Test Kernel
+
+# The kernel build will use this option.
+BUILD_OPTIONS = -j8
+
+# Keeping the log file with the output dir is convenient.
+LOG_FILE = ${OUTPUT_DIR}/${MACHINE}.log
+
+# Each box should have their own minum configuration
+# See min-config.conf
+MIN_CONFIG = ${CONFIG_DIR}/config-min
+
+# For things like randconfigs, there may be configs you find that
+# are already broken, or there may be some configs that you always
+# want set. Uncomment ADD_CONFIG and point it to the make config files
+# that set the configs you want to keep on (or off) in your build.
+# ADD_CONFIG is usually something to add configs to all machines,
+# where as, MIN_CONFIG is specific per machine.
+#ADD_CONFIG = ${THIS_DIR}/config-broken ${THIS_DIR}/config-general
+
+# To speed up reboots for bisects and patchcheck, instead of
+# waiting 60 seconds for the console to be idle, if this line is
+# seen in the console output, ktest will know the good kernel has
+# finished rebooting and it will be able to continue the tests.
+REBOOT_SUCCESS_LINE = ${MACHINE} login:
+
+# The following is different ways to end the test.
+# by setting the variable REBOOT to: none, error, fail or
+# something else, ktest will power cycle or reboot the target box
+# at the end of the tests.
+#
+# REBOOT := none
+#   Don't do anything at the end of the test.
+#
+# REBOOT := error
+#   Reboot the box if ktest detects an error
+#
+# REBOOT := fail
+#   Do not stop on failure, and after all tests are complete
+#   power off the box (for both success and error)
+#   This is good to run over a weekend and you don't want to waste
+#   electricity.
+#
+
+DEFAULTS IF ${REBOOT} == none
+REBOOT_ON_SUCCESS = 0
+REBOOT_ON_ERROR = 0
+POWEROFF_ON_ERROR = 0
+POWEROFF_ON_SUCCESS = 0
+
+DEFAULTS ELSE IF ${REBOOT} == error
+REBOOT_ON_SUCCESS = 0
+REBOOT_ON_ERROR = 1
+POWEROFF_ON_ERROR = 0
+POWEROFF_ON_SUCCESS = 0
+
+DEFAULTS ELSE IF ${REBOOT} == fail
+REBOOT_ON_SUCCESS = 0
+POWEROFF_ON_ERROR = 1
+POWEROFF_ON_SUCCESS = 1
+POWEROFF_AFTER_HALT = 120
+DIE_ON_FAILURE = 0
+
+# Store the failure information into this directory
+# such as the .config, dmesg, and build log.
+STORE_FAILURES = ${THIS_DIR}/failures
+
+DEFAULTS ELSE
+REBOOT_ON_SUCCESS = 1
+REBOOT_ON_ERROR = 1
+POWEROFF_ON_ERROR = 0
+POWEROFF_ON_SUCCESS = 0

tools/testing/ktest/examples/include/min-config.conf

@@ -0,0 +1,60 @@
+#
+# This file has some examples for creating a MIN_CONFIG.
+# (A .config file that is the minimum for a machine to boot, or
+#  to boot and make a network connection.)
+#
+# A MIN_CONFIG is very useful as it is the minimum configuration
+# needed to boot a given machine. You can debug someone else's
+# .config by only setting the configs in your MIN_CONFIG. The closer
+# your MIN_CONFIG is to the true minimum set of configs needed to
+# boot your machine, the closer the config you test with will be
+# to the users config that had the failure.
+#
+# The make_min_config test allows you to create a MIN_CONFIG that
+# is truly the minimum set of configs needed to boot a box.
+#
+# In this example, the final config will reside in
+# ${CONFIG_DIR}/config-new-min and ${CONFIG_DIR}/config-new-min-net.
+# Just move one to the location you have set for MIN_CONFIG.
+#
+# The first test creates a MIN_CONFIG that will be the minimum
+# configuration to boot ${MACHINE} and be able to ssh to it.
+#
+# The second test creates a MIN_CONFIG that will only boot
+# the target and most likely will not let you ssh to it. (Notice
+# how the second test uses the first test's result to continue with.
+# This is because the second test config is a subset of the first).
+#
+# The ${CONFIG_DIR}/config-skip (and -net) will hold the configs
+# that ktest.pl found would not boot the target without them set.
+# The config-new-min holds configs that ktest.pl could not test
+# directly because another config that was needed to boot the box
+# selected them. Sometimes it is possible that this file will hold
+# the true minimum configuration. You can test to see if this is
+# the case by running the boot test with BOOT_TYPE = allnoconfig and
+# setting setting the MIN_CONFIG to ${CONFIG_DIR}/config-skip. If the
+# machine still boots, then you can use the config-skip as your MIN_CONFIG.
+#
+# These tests can run for several hours (and perhaps days).
+# It's OK to kill the test with a Ctrl^C. By restarting without
+# modifying this config, ktest.pl will notice that the config-new-min(-net)
+# exists, and will use that instead as the starting point.
+# The USE_OUTPUT_MIN_CONFIG is set to 1 to keep ktest.pl from asking
+# you if you want to use the OUTPUT_MIN_CONFIG as the starting point.
+# By using the OUTPUT_MIN_CONFIG as the starting point will allow ktest.pl to
+# start almost where it left off.
+#
+TEST_START IF ${TEST} == min-config
+TEST_TYPE = make_min_config
+OUTPUT_MIN_CONFIG = ${CONFIG_DIR}/config-new-min-net
+IGNORE_CONFIG = ${CONFIG_DIR}/config-skip-net
+MIN_CONFIG_TYPE = test
+TEST = ${SSH} echo hi
+USE_OUTPUT_MIN_CONFIG = 1
+
+TEST_START IF ${TEST} == min-config && ${MULTI}
+TEST_TYPE = make_min_config
+OUTPUT_MIN_CONFIG = ${CONFIG_DIR}/config-new-min
+IGNORE_CONFIG = ${CONFIG_DIR}/config-skip
+MIN_CONFIG = ${CONFIG_DIR}/config-new-min-net
+USE_OUTPUT_MIN_CONFIG = 1

tools/testing/ktest/examples/include/patchcheck.conf

@@ -0,0 +1,111 @@
+# patchcheck.conf
+#
+# This contains a test that takes two git commits and will test each
+# commit between the two. The build test will look at what files the
+# commit has touched, and if any of those files produce a warning, then
+# the build will fail.
+
+
+# PATCH_START is the commit to begin with and PATCH_END is the commit
+# to end with (inclusive). This is similar to doing a git rebase -i PATCH_START~1
+# and then testing each commit and doing a git rebase --continue.
+# You can use a SHA1, a git tag, or anything that git will accept for a checkout
+
+PATCH_START := HEAD~3
+PATCH_END := HEAD
+
+# Use the oldconfig if build_type wasn't defined
+DEFAULTS IF NOT DEFINED BUILD_TYPE
+DO_BUILD_TYPE := oldconfig
+
+DEFAULTS ELSE
+DO_BUILD_TYPE := ${BUILD_TYPE}
+
+DEFAULTS
+
+
+# Change PATCH_CHECKOUT to be the branch you want to test. The test will
+# do a git checkout of this branch before starting. Obviously both
+# PATCH_START and PATCH_END must be in this branch (and PATCH_START must
+# be contained by PATCH_END).
+
+PATCH_CHECKOUT := test/branch
+
+# Usually it's a good idea to have a set config to use for testing individual
+# patches.
+PATCH_CONFIG := ${CONFIG_DIR}/config-patchcheck
+
+# Change PATCH_TEST to run some test for each patch. Each commit that is
+# tested, after it is built and installed on the test machine, this command
+# will be executed. Usually what is done is to ssh to the target box and
+# run some test scripts. If you just want to boot test your patches
+# comment PATCH_TEST out.
+PATCH_TEST := ${SSH} "/usr/local/bin/ktest-test-script"
+
+DEFAULTS IF DEFINED PATCH_TEST
+PATCH_TEST_TYPE := test
+
+DEFAULTS ELSE
+PATCH_TEST_TYPE := boot
+
+# If for some reason a file has a warning that one of your patches touch
+# but you do not care about it, set IGNORE_WARNINGS to that commit(s)
+# (space delimited)
+#IGNORE_WARNINGS = 39eaf7ef884dcc44f7ff1bac803ca2a1dcf43544 6edb2a8a385f0cdef51dae37ff23e74d76d8a6ce
+
+# Instead of just checking for warnings to files that are changed
+# it can be advantageous to check for any new warnings. If a
+# header file is changed, it could cause a warning in a file not
+# touched by the commit. To detect these kinds of warnings, you
+# can use the WARNINGS_FILE option.
+#
+# If the variable CREATE_WARNINGS_FILE is set, this config will
+# enable the WARNINGS_FILE during the patchcheck test. Also,
+# before running the patchcheck test, it will create the
+# warnings file.
+#
+DEFAULTS IF DEFINED CREATE_WARNINGS_FILE
+WARNINGS_FILE = ${OUTPUT_DIR}/warnings_file
+
+TEST_START IF DEFINED CREATE_WARNINGS_FILE
+# WARNINGS_FILE is already set by the DEFAULTS above
+TEST_TYPE = make_warnings_file
+# Checkout the commit before the patches to test,
+# and record all the warnings that exist before the patches
+# to test are added
+CHECKOUT = ${PATCHCHECK_START}~1
+# Force a full build
+BUILD_NOCLEAN = 0
+BUILD_TYPE = ${DO_BUILD_TYPE}
+
+# If you are running a multi test, and the test failed on the first
+# test but on, say the 5th patch. If you want to restart on the
+# fifth patch, set PATCH_START1. This will make the first test start
+# from this commit instead of the PATCH_START commit.
+# Note, do not change this option. Just define PATCH_START1 in the
+# top config (the one you pass to ktest.pl), and this will use it,
+# otherwise it will just use PATCH_START if PATCH_START1 is not defined.
+DEFAULTS IF NOT DEFINED PATCH_START1
+PATCH_START1 := ${PATCH_START}
+
+TEST_START IF ${TEST} == patchcheck
+TEST_TYPE = patchcheck
+MIN_CONFIG = ${PATCH_CONFIG}
+TEST = ${PATCH_TEST}
+PATCHCHECK_TYPE = ${PATCH_TEST_TYPE}
+PATCHCHECK_START = ${PATCH_START1}
+PATCHCHECK_END = ${PATCH_END}
+CHECKOUT = ${PATCH_CHECKOUT}
+BUILD_TYPE = ${DO_BUILD_TYPE}
+
+TEST_START IF ${TEST} == patchcheck && ${MULTI}
+TEST_TYPE = patchcheck
+MIN_CONFIG = ${PATCH_CONFIG}
+TEST = ${PATCH_TEST}
+PATCHCHECK_TYPE = ${PATCH_TEST_TYPE}
+PATCHCHECK_START = ${PATCH_START}
+PATCHCHECK_END = ${PATCH_END}
+CHECKOUT = ${PATCH_CHECKOUT}
+# Use multi to test different compilers?
+MAKE_CMD = CC=gcc-4.5.1 make
+BUILD_TYPE = ${DO_BUILD_TYPE}

tools/testing/ktest/examples/include/tests.conf

@@ -0,0 +1,74 @@
+#
+# This is an example of various tests that you can run
+#
+# The variable TEST can be of boot, build, randconfig, or test.
+#
+# Note that TEST is a variable created with ':=' and only exists
+# throughout the config processing (not during the tests itself).
+#
+# The TEST option (defined with '=') is used to tell ktest.pl
+# what test to run after a successful boot. The TEST option is
+# persistent into the test runs.
+#
+
+# The config that includes this file may define a BOOT_TYPE
+# variable that tells this config what type of boot test to run.
+# If it's not defined, the below DEFAULTS will set the default
+# to 'oldconfig'.
+#
+DEFAULTS IF NOT DEFINED BOOT_TYPE
+BOOT_TYPE := oldconfig
+
+# The config that includes this file may define a RUN_TEST
+# variable that will tell this config what test to run.
+# (what to set the TEST option to).
+#
+DEFAULTS IF NOT DEFINED RUN_TEST
+# Requires that hackbench is in the PATH
+RUN_TEST := ${SSH} hackbench 50
+
+
+# If TEST is set to 'boot' then just build a kernel and boot
+# the target.
+TEST_START IF ${TEST} == boot
+TEST_TYPE = boot
+# Notice how we set the BUILD_TYPE option to the BOOT_TYPE variable.
+BUILD_TYPE = ${BOOT_TYPE}
+# Do not do a make mrproper.
+BUILD_NOCLEAN = 1
+
+# If you only want to build the kernel, and perhaps install
+# and test it yourself, then just set TEST to build.
+TEST_START IF ${TEST} == build
+TEST_TYPE = build
+BUILD_TYPE = ${BOOT_TYPE}
+BUILD_NOCLEAN = 1
+
+# Build, install, boot and test with a randconfg 10 times.
+# It is important that you have set MIN_CONFIG in the config
+# that includes this file otherwise it is likely that the
+# randconfig will not have the necessary configs needed to
+# boot your box. This version of the test requires a min
+# config that has enough to make sure the target has network
+# working.
+TEST_START ITERATE 10 IF ${TEST} == randconfig
+MIN_CONFIG = ${CONFIG_DIR}/config-min-net
+TEST_TYPE = test
+BUILD_TYPE = randconfig
+TEST = ${RUN_TEST}
+
+# This is the same as above, but only tests to a boot prompt.
+# The MIN_CONFIG used here does not need to have networking
+# working.
+TEST_START ITERATE 10 IF ${TEST} == randconfig && ${MULTI}
+TEST_TYPE = boot
+BUILD_TYPE = randconfig
+MIN_CONFIG = ${CONFIG_DIR}/config-min
+MAKE_CMD = make
+
+# This builds, installs, boots and tests the target.
+TEST_START IF ${TEST} == test
+TEST_TYPE = test
+BUILD_TYPE = ${BOOT_TYPE}
+TEST = ${RUN_TEST}
+BUILD_NOCLEAN = 1

tools/testing/ktest/examples/kvm.conf

@@ -0,0 +1,92 @@
+#
+# This config is an example usage of ktest.pl with a kvm guest
+#
+# The guest is called 'Guest' and this would be something that
+# could be run on the host to test a virtual machine target.
+
+MACHINE = Guest
+
+
+# Use virsh to read the serial console of the guest
+CONSOLE =  virsh console ${MACHINE}
+
+# Use SIGKILL to terminate virsh console. We can't kill virsh console
+# by the default signal, SIGINT.
+CLOSE_CONSOLE_SIGNAL = KILL
+
+#*************************************#
+# This part is the same as test.conf  #
+#*************************************#
+
+# The include files will set up the type of test to run. Just set TEST to
+# which test you want to run.
+#
+# TESTS = patchcheck, randconfig, boot, test, config-bisect, bisect, min-config
+#
+# See the include/*.conf files that define these tests
+#
+TEST := patchcheck
+
+# Some tests may have more than one test to run. Define MULTI := 1 to run
+# the extra tests.
+MULTI := 0
+
+# In case you want to differentiate which type of system you are testing
+BITS := 64
+
+# REBOOT = none, error, fail, empty
+#  See include/defaults.conf
+REBOOT := empty
+
+
+# The defaults file will set up various settings that can be used by all
+# machine configs.
+INCLUDE include/defaults.conf
+
+
+#*************************************#
+# Now we are different from test.conf #
+#*************************************#
+
+
+# The example here assumes that Guest is running a Fedora release
+# that uses dracut for its initfs. The POST_INSTALL will be executed
+# after the install of the kernel and modules are complete.
+#
+POST_INSTALL = ${SSH} /sbin/dracut -f /boot/initramfs-test.img $KERNEL_VERSION
+
+# Guests sometimes get stuck on reboot. We wait 3 seconds after running
+# the reboot command and then  do a full power-cycle of the guest.
+# This forces the guest to restart.
+#
+POWERCYCLE_AFTER_REBOOT = 3
+
+# We do the same after the halt command, but this time we wait 20 seconds.
+POWEROFF_AFTER_HALT = 20
+
+
+# As the defaults.conf file has a POWER_CYCLE option already defined,
+# and options can not be defined in the same section more than once
+# (all DEFAULTS sections are considered the same). We use the
+# DEFAULTS OVERRIDE to tell ktest.pl to ignore the previous defined
+# options, for the options set in the OVERRIDE section.
+#
+DEFAULTS OVERRIDE
+
+# Instead of using the default POWER_CYCLE option defined in
+# defaults.conf, we use virsh to cycle it. To do so, we destroy
+# the guest, wait 5 seconds, and then start it up again.
+# Crude, but effective.
+#
+POWER_CYCLE = virsh destroy ${MACHINE}; sleep 5; virsh start ${MACHINE}
+
+
+DEFAULTS
+
+# The following files each handle a different test case.
+# Having them included allows you to set up more than one machine and share
+# the same tests.
+INCLUDE include/patchcheck.conf
+INCLUDE include/tests.conf
+INCLUDE include/bisect.conf
+INCLUDE include/min-config.conf

tools/testing/ktest/examples/snowball.conf

@@ -0,0 +1,53 @@
+# This example was used to boot the snowball ARM board.
+# See http://people.redhat.com/srostedt/ktest-embedded-2012/
+
+# PWD is a ktest.pl variable that will result in the process working
+# directory that ktest.pl is executed in.
+
+# THIS_DIR is automatically assigned the PWD of the path that generated
+# the config file. It is best to use this variable when assigning other
+# directory paths within this directory. This allows you to easily
+# move the test cases to other locations or to other machines.
+#
+THIS_DIR := /home/rostedt/work/demo/ktest-embed
+LOG_FILE = ${OUTPUT_DIR}/snowball.log
+CLEAR_LOG = 1
+MAKE_CMD = PATH=/usr/local/gcc-4.5.2-nolibc/arm-unknown-linux-gnueabi/bin:$PATH CROSS_COMPILE=arm-unknown-linux-gnueabi- make ARCH=arm
+ADD_CONFIG = ${THIS_DIR}/addconfig
+
+SCP_TO_TARGET = echo "don't do scp"
+
+TFTPBOOT := /var/lib/tftpboot
+TFTPDEF := ${TFTPBOOT}/snowball-default
+TFTPTEST := ${OUTPUT_DIR}/${BUILD_TARGET}
+
+SWITCH_TO_GOOD = cp ${TFTPDEF} ${TARGET_IMAGE}
+SWITCH_TO_TEST = cp ${TFTPTEST} ${TARGET_IMAGE}
+
+# Define each test with TEST_START
+# The config options below it will override the defaults
+TEST_START SKIP
+TEST_TYPE = boot
+BUILD_TYPE = u8500_defconfig
+BUILD_NOCLEAN = 1
+
+TEST_START
+TEST_TYPE = make_min_config
+OUTPUT_MIN_CONFIG = ${THIS_DIR}/config.newmin
+START_MIN_CONFIG = ${THIS_DIR}/config.orig
+IGNORE_CONFIG = ${THIS_DIR}/config.ignore
+BUILD_NOCLEAN = 1
+
+
+DEFAULTS
+LOCALVERSION = -test
+POWER_CYCLE = echo use the thumb luke; read a
+CONSOLE = cat ${THIS_DIR}/snowball-cat
+REBOOT_TYPE = script
+SSH_USER = root
+BUILD_OPTIONS = -j8 uImage
+BUILD_DIR = ${THIS_DIR}/linux.git
+OUTPUT_DIR = ${THIS_DIR}/snowball-build
+MACHINE = snowball
+TARGET_IMAGE = /var/lib/tftpboot/snowball-image
+BUILD_TARGET = arch/arm/boot/uImage

tools/testing/ktest/examples/test.conf

@@ -0,0 +1,62 @@
+#
+# Generic config for a machine
+#
+
+# Name your machine (the DNS name, what you ssh to)
+MACHINE = foo
+
+# BOX can be different than foo, if the machine BOX has
+# multiple partitions with different systems installed. For example,
+# you may have a i386 and x86_64 installation on a test box.
+# If this is the case, MACHINE defines the way to connect to the
+# machine, which may be different between which system the machine
+# is booting into. BOX is used for the scripts to reboot and power cycle
+# the machine, where it does not matter which system the machine boots into.
+#
+#BOX := bar
+
+# Define a way to read the console
+CONSOLE = stty -F /dev/ttyS0 115200 parodd; cat /dev/ttyS0
+
+# The include files will set up the type of test to run. Just set TEST to
+# which test you want to run.
+#
+# TESTS = patchcheck, randconfig, boot, test, config-bisect, bisect, min-config
+#
+# See the include/*.conf files that define these tests
+#
+TEST := patchcheck
+
+# Some tests may have more than one test to run. Define MULTI := 1 to run
+# the extra tests.
+MULTI := 0
+
+# In case you want to differentiate which type of system you are testing
+BITS := 64
+
+# REBOOT = none, error, fail, empty
+#  See include/defaults.conf
+REBOOT := empty
+
+# The defaults file will set up various settings that can be used by all
+# machine configs.
+INCLUDE include/defaults.conf
+
+# In case you need to add a patch for a bisect or something
+#PRE_BUILD = patch -p1 < ${THIS_DIR}/fix.patch
+
+# Reset the repo after the build and remove all 'test' modules from the target
+# Notice that DO_POST_BUILD is a variable (defined by ':=') and POST_BUILD
+# is the option (defined by '=')
+
+DO_POST_BUILD := git reset --hard
+POST_BUILD = ${SSH} 'rm -rf /lib/modules/*-test*'; ${DO_POST_BUILD}
+
+# The following files each handle a different test case.
+# Having them included allows you to set up more than one machine and share
+# the same tests.
+INCLUDE include/patchcheck.conf
+INCLUDE include/tests.conf
+INCLUDE include/bisect.conf
+INCLUDE include/min-config.conf
+

tools/testing/nvdimm/Kbuild

@@ -0,0 +1,84 @@
+# SPDX-License-Identifier: GPL-2.0
+ldflags-y += --wrap=ioremap_wc
+ldflags-y += --wrap=memremap
+ldflags-y += --wrap=devm_ioremap_nocache
+ldflags-y += --wrap=devm_memremap
+ldflags-y += --wrap=devm_memunmap
+ldflags-y += --wrap=ioremap_nocache
+ldflags-y += --wrap=iounmap
+ldflags-y += --wrap=memunmap
+ldflags-y += --wrap=__devm_request_region
+ldflags-y += --wrap=__devm_release_region
+ldflags-y += --wrap=__request_region
+ldflags-y += --wrap=__release_region
+ldflags-y += --wrap=devm_memremap_pages
+ldflags-y += --wrap=insert_resource
+ldflags-y += --wrap=remove_resource
+ldflags-y += --wrap=acpi_evaluate_object
+ldflags-y += --wrap=acpi_evaluate_dsm
+
+DRIVERS := ../../../drivers
+NVDIMM_SRC := $(DRIVERS)/nvdimm
+ACPI_SRC := $(DRIVERS)/acpi/nfit
+DAX_SRC := $(DRIVERS)/dax
+ccflags-y := -I$(src)/$(NVDIMM_SRC)/
+
+obj-$(CONFIG_LIBNVDIMM) += libnvdimm.o
+obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o
+obj-$(CONFIG_ND_BTT) += nd_btt.o
+obj-$(CONFIG_ND_BLK) += nd_blk.o
+obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o
+obj-$(CONFIG_ACPI_NFIT) += nfit.o
+ifeq ($(CONFIG_DAX),m)
+obj-$(CONFIG_DAX) += dax.o
+endif
+obj-$(CONFIG_DEV_DAX) += device_dax.o
+obj-$(CONFIG_DEV_DAX_PMEM) += dax_pmem.o
+
+nfit-y := $(ACPI_SRC)/core.o
+nfit-$(CONFIG_X86_MCE) += $(ACPI_SRC)/mce.o
+nfit-y += acpi_nfit_test.o
+nfit-y += config_check.o
+
+nd_pmem-y := $(NVDIMM_SRC)/pmem.o
+nd_pmem-y += pmem-dax.o
+nd_pmem-y += pmem_test.o
+nd_pmem-y += config_check.o
+
+nd_btt-y := $(NVDIMM_SRC)/btt.o
+nd_btt-y += config_check.o
+
+nd_blk-y := $(NVDIMM_SRC)/blk.o
+nd_blk-y += config_check.o
+
+nd_e820-y := $(NVDIMM_SRC)/e820.o
+nd_e820-y += config_check.o
+
+dax-y := $(DAX_SRC)/super.o
+dax-y += config_check.o
+
+device_dax-y := $(DAX_SRC)/device.o
+device_dax-y += dax-dev.o
+device_dax-y += device_dax_test.o
+device_dax-y += config_check.o
+
+dax_pmem-y := $(DAX_SRC)/pmem.o
+dax_pmem-y += config_check.o
+
+libnvdimm-y := $(NVDIMM_SRC)/core.o
+libnvdimm-y += $(NVDIMM_SRC)/bus.o
+libnvdimm-y += $(NVDIMM_SRC)/dimm_devs.o
+libnvdimm-y += $(NVDIMM_SRC)/dimm.o
+libnvdimm-y += $(NVDIMM_SRC)/region_devs.o
+libnvdimm-y += $(NVDIMM_SRC)/region.o
+libnvdimm-y += $(NVDIMM_SRC)/namespace_devs.o
+libnvdimm-y += $(NVDIMM_SRC)/label.o
+libnvdimm-y += $(NVDIMM_SRC)/badrange.o
+libnvdimm-$(CONFIG_ND_CLAIM) += $(NVDIMM_SRC)/claim.o
+libnvdimm-$(CONFIG_BTT) += $(NVDIMM_SRC)/btt_devs.o
+libnvdimm-$(CONFIG_NVDIMM_PFN) += $(NVDIMM_SRC)/pfn_devs.o
+libnvdimm-$(CONFIG_NVDIMM_DAX) += $(NVDIMM_SRC)/dax_devs.o
+libnvdimm-y += libnvdimm_test.o
+libnvdimm-y += config_check.o
+
+obj-m += test/

tools/testing/nvdimm/Makefile

@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+KDIR ?= ../../../
+
+default:
+	$(MAKE) -C $(KDIR) M=$$PWD
+
+install: default
+	$(MAKE) -C $(KDIR) M=$$PWD modules_install

tools/testing/nvdimm/acpi_nfit_test.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright(c) 2018 Intel Corporation. All rights reserved.
+
+#include <linux/module.h>
+#include <linux/printk.h>
+#include "watermark.h"
+
+nfit_test_watermark(acpi_nfit);

tools/testing/nvdimm/config_check.c

@@ -0,0 +1,18 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bug.h>
+
+void check(void)
+{
+	/*
+	 * These kconfig symbols must be set to "m" for nfit_test to
+	 * load and operate.
+	 */
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_LIBNVDIMM));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_BLK_DEV_PMEM));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_ND_BTT));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_ND_PFN));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_ND_BLK));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_ACPI_NFIT));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_DEV_DAX));
+	BUILD_BUG_ON(!IS_MODULE(CONFIG_DEV_DAX_PMEM));
+}

tools/testing/nvdimm/dax-dev.c

@@ -0,0 +1,49 @@
+/*
+ * Copyright (c) 2016, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include "test/nfit_test.h"
+#include <linux/mm.h>
+#include "../../../drivers/dax/dax-private.h"
+
+phys_addr_t dax_pgoff_to_phys(struct dev_dax *dev_dax, pgoff_t pgoff,
+		unsigned long size)
+{
+	struct resource *res;
+	phys_addr_t addr;
+	int i;
+
+	for (i = 0; i < dev_dax->num_resources; i++) {
+		res = &dev_dax->res[i];
+		addr = pgoff * PAGE_SIZE + res->start;
+		if (addr >= res->start && addr <= res->end)
+			break;
+		pgoff -= PHYS_PFN(resource_size(res));
+	}
+
+	if (i < dev_dax->num_resources) {
+		res = &dev_dax->res[i];
+		if (addr + size - 1 <= res->end) {
+			if (get_nfit_res(addr)) {
+				struct page *page;
+
+				if (dev_dax->region->align > PAGE_SIZE)
+					return -1;
+
+				page = vmalloc_to_page((void *)addr);
+				return PFN_PHYS(page_to_pfn(page));
+			} else
+				return addr;
+		}
+	}
+
+	return -1;
+}

tools/testing/nvdimm/device_dax_test.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright(c) 2018 Intel Corporation. All rights reserved.
+
+#include <linux/module.h>
+#include <linux/printk.h>
+#include "watermark.h"
+
+nfit_test_watermark(device_dax);

tools/testing/nvdimm/libnvdimm_test.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright(c) 2018 Intel Corporation. All rights reserved.
+
+#include <linux/module.h>
+#include <linux/printk.h>
+#include "watermark.h"
+
+nfit_test_watermark(libnvdimm);

tools/testing/nvdimm/pmem-dax.c

@@ -0,0 +1,57 @@
+/*
+ * Copyright (c) 2014-2016, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include "test/nfit_test.h"
+#include <linux/blkdev.h>
+#include <pmem.h>
+#include <nd.h>
+
+long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff,
+		long nr_pages, void **kaddr, pfn_t *pfn)
+{
+	resource_size_t offset = PFN_PHYS(pgoff) + pmem->data_offset;
+
+	if (unlikely(is_bad_pmem(&pmem->bb, PFN_PHYS(pgoff) / 512,
+					PFN_PHYS(nr_pages))))
+		return -EIO;
+
+	/*
+	 * Limit dax to a single page at a time given vmalloc()-backed
+	 * in the nfit_test case.
+	 */
+	if (get_nfit_res(pmem->phys_addr + offset)) {
+		struct page *page;
+
+		if (kaddr)
+			*kaddr = pmem->virt_addr + offset;
+		page = vmalloc_to_page(pmem->virt_addr + offset);
+		if (pfn)
+			*pfn = page_to_pfn_t(page);
+		pr_debug_ratelimited("%s: pmem: %p pgoff: %#lx pfn: %#lx\n",
+				__func__, pmem, pgoff, page_to_pfn(page));
+
+		return 1;
+	}
+
+	if (kaddr)
+		*kaddr = pmem->virt_addr + offset;
+	if (pfn)
+		*pfn = phys_to_pfn_t(pmem->phys_addr + offset, pmem->pfn_flags);
+
+	/*
+	 * If badblocks are present, limit known good range to the
+	 * requested range.
+	 */
+	if (unlikely(pmem->bb.count))
+		return nr_pages;
+	return PHYS_PFN(pmem->size - pmem->pfn_pad - offset);
+}

tools/testing/nvdimm/pmem_test.c

@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright(c) 2018 Intel Corporation. All rights reserved.
+
+#include <linux/module.h>
+#include <linux/printk.h>
+#include "watermark.h"
+
+nfit_test_watermark(pmem);

tools/testing/nvdimm/test/Kbuild

@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0
+ccflags-y := -I$(src)/../../../../drivers/nvdimm/
+ccflags-y += -I$(src)/../../../../drivers/acpi/nfit/
+
+obj-m += nfit_test.o
+obj-m += nfit_test_iomap.o
+
+nfit_test-y := nfit.o
+nfit_test_iomap-y := iomap.o

tools/testing/nvdimm/test/iomap.c

@@ -0,0 +1,403 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+#include <linux/memremap.h>
+#include <linux/rculist.h>
+#include <linux/export.h>
+#include <linux/ioport.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/pfn_t.h>
+#include <linux/acpi.h>
+#include <linux/io.h>
+#include <linux/mm.h>
+#include "nfit_test.h"
+
+static LIST_HEAD(iomap_head);
+
+static struct iomap_ops {
+	nfit_test_lookup_fn nfit_test_lookup;
+	nfit_test_evaluate_dsm_fn evaluate_dsm;
+	struct list_head list;
+} iomap_ops = {
+	.list = LIST_HEAD_INIT(iomap_ops.list),
+};
+
+void nfit_test_setup(nfit_test_lookup_fn lookup,
+		nfit_test_evaluate_dsm_fn evaluate)
+{
+	iomap_ops.nfit_test_lookup = lookup;
+	iomap_ops.evaluate_dsm = evaluate;
+	list_add_rcu(&iomap_ops.list, &iomap_head);
+}
+EXPORT_SYMBOL(nfit_test_setup);
+
+void nfit_test_teardown(void)
+{
+	list_del_rcu(&iomap_ops.list);
+	synchronize_rcu();
+}
+EXPORT_SYMBOL(nfit_test_teardown);
+
+static struct nfit_test_resource *__get_nfit_res(resource_size_t resource)
+{
+	struct iomap_ops *ops;
+
+	ops = list_first_or_null_rcu(&iomap_head, typeof(*ops), list);
+	if (ops)
+		return ops->nfit_test_lookup(resource);
+	return NULL;
+}
+
+struct nfit_test_resource *get_nfit_res(resource_size_t resource)
+{
+	struct nfit_test_resource *res;
+
+	rcu_read_lock();
+	res = __get_nfit_res(resource);
+	rcu_read_unlock();
+
+	return res;
+}
+EXPORT_SYMBOL(get_nfit_res);
+
+void __iomem *__nfit_test_ioremap(resource_size_t offset, unsigned long size,
+		void __iomem *(*fallback_fn)(resource_size_t, unsigned long))
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
+
+	if (nfit_res)
+		return (void __iomem *) nfit_res->buf + offset
+			- nfit_res->res.start;
+	return fallback_fn(offset, size);
+}
+
+void __iomem *__wrap_devm_ioremap_nocache(struct device *dev,
+		resource_size_t offset, unsigned long size)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
+
+	if (nfit_res)
+		return (void __iomem *) nfit_res->buf + offset
+			- nfit_res->res.start;
+	return devm_ioremap_nocache(dev, offset, size);
+}
+EXPORT_SYMBOL(__wrap_devm_ioremap_nocache);
+
+void *__wrap_devm_memremap(struct device *dev, resource_size_t offset,
+		size_t size, unsigned long flags)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
+
+	if (nfit_res)
+		return nfit_res->buf + offset - nfit_res->res.start;
+	return devm_memremap(dev, offset, size, flags);
+}
+EXPORT_SYMBOL(__wrap_devm_memremap);
+
+static void nfit_test_kill(void *_pgmap)
+{
+	struct dev_pagemap *pgmap = _pgmap;
+
+	pgmap->kill(pgmap->ref);
+}
+
+void *__wrap_devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap)
+{
+	resource_size_t offset = pgmap->res.start;
+	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
+
+	if (nfit_res) {
+		int rc;
+
+		rc = devm_add_action_or_reset(dev, nfit_test_kill, pgmap);
+		if (rc)
+			return ERR_PTR(rc);
+		return nfit_res->buf + offset - nfit_res->res.start;
+	}
+	return devm_memremap_pages(dev, pgmap);
+}
+EXPORT_SYMBOL_GPL(__wrap_devm_memremap_pages);
+
+pfn_t __wrap_phys_to_pfn_t(phys_addr_t addr, unsigned long flags)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res(addr);
+
+	if (nfit_res)
+		flags &= ~PFN_MAP;
+        return phys_to_pfn_t(addr, flags);
+}
+EXPORT_SYMBOL(__wrap_phys_to_pfn_t);
+
+void *__wrap_memremap(resource_size_t offset, size_t size,
+		unsigned long flags)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res(offset);
+
+	if (nfit_res)
+		return nfit_res->buf + offset - nfit_res->res.start;
+	return memremap(offset, size, flags);
+}
+EXPORT_SYMBOL(__wrap_memremap);
+
+void __wrap_devm_memunmap(struct device *dev, void *addr)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res((long) addr);
+
+	if (nfit_res)
+		return;
+	return devm_memunmap(dev, addr);
+}
+EXPORT_SYMBOL(__wrap_devm_memunmap);
+
+void __iomem *__wrap_ioremap_nocache(resource_size_t offset, unsigned long size)
+{
+	return __nfit_test_ioremap(offset, size, ioremap_nocache);
+}
+EXPORT_SYMBOL(__wrap_ioremap_nocache);
+
+void __iomem *__wrap_ioremap_wc(resource_size_t offset, unsigned long size)
+{
+	return __nfit_test_ioremap(offset, size, ioremap_wc);
+}
+EXPORT_SYMBOL(__wrap_ioremap_wc);
+
+void __wrap_iounmap(volatile void __iomem *addr)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res((long) addr);
+	if (nfit_res)
+		return;
+	return iounmap(addr);
+}
+EXPORT_SYMBOL(__wrap_iounmap);
+
+void __wrap_memunmap(void *addr)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res((long) addr);
+
+	if (nfit_res)
+		return;
+	return memunmap(addr);
+}
+EXPORT_SYMBOL(__wrap_memunmap);
+
+static bool nfit_test_release_region(struct device *dev,
+		struct resource *parent, resource_size_t start,
+		resource_size_t n);
+
+static void nfit_devres_release(struct device *dev, void *data)
+{
+	struct resource *res = *((struct resource **) data);
+
+	WARN_ON(!nfit_test_release_region(NULL, &iomem_resource, res->start,
+			resource_size(res)));
+}
+
+static int match(struct device *dev, void *__res, void *match_data)
+{
+	struct resource *res = *((struct resource **) __res);
+	resource_size_t start = *((resource_size_t *) match_data);
+
+	return res->start == start;
+}
+
+static bool nfit_test_release_region(struct device *dev,
+		struct resource *parent, resource_size_t start,
+		resource_size_t n)
+{
+	if (parent == &iomem_resource) {
+		struct nfit_test_resource *nfit_res = get_nfit_res(start);
+
+		if (nfit_res) {
+			struct nfit_test_request *req;
+			struct resource *res = NULL;
+
+			if (dev) {
+				devres_release(dev, nfit_devres_release, match,
+						&start);
+				return true;
+			}
+
+			spin_lock(&nfit_res->lock);
+			list_for_each_entry(req, &nfit_res->requests, list)
+				if (req->res.start == start) {
+					res = &req->res;
+					list_del(&req->list);
+					break;
+				}
+			spin_unlock(&nfit_res->lock);
+
+			WARN(!res || resource_size(res) != n,
+					"%s: start: %llx n: %llx mismatch: %pr\n",
+						__func__, start, n, res);
+			if (res)
+				kfree(req);
+			return true;
+		}
+	}
+	return false;
+}
+
+static struct resource *nfit_test_request_region(struct device *dev,
+		struct resource *parent, resource_size_t start,
+		resource_size_t n, const char *name, int flags)
+{
+	struct nfit_test_resource *nfit_res;
+
+	if (parent == &iomem_resource) {
+		nfit_res = get_nfit_res(start);
+		if (nfit_res) {
+			struct nfit_test_request *req;
+			struct resource *res = NULL;
+
+			if (start + n > nfit_res->res.start
+					+ resource_size(&nfit_res->res)) {
+				pr_debug("%s: start: %llx n: %llx overflow: %pr\n",
+						__func__, start, n,
+						&nfit_res->res);
+				return NULL;
+			}
+
+			spin_lock(&nfit_res->lock);
+			list_for_each_entry(req, &nfit_res->requests, list)
+				if (start == req->res.start) {
+					res = &req->res;
+					break;
+				}
+			spin_unlock(&nfit_res->lock);
+
+			if (res) {
+				WARN(1, "%pr already busy\n", res);
+				return NULL;
+			}
+
+			req = kzalloc(sizeof(*req), GFP_KERNEL);
+			if (!req)
+				return NULL;
+			INIT_LIST_HEAD(&req->list);
+			res = &req->res;
+
+			res->start = start;
+			res->end = start + n - 1;
+			res->name = name;
+			res->flags = resource_type(parent);
+			res->flags |= IORESOURCE_BUSY | flags;
+			spin_lock(&nfit_res->lock);
+			list_add(&req->list, &nfit_res->requests);
+			spin_unlock(&nfit_res->lock);
+
+			if (dev) {
+				struct resource **d;
+
+				d = devres_alloc(nfit_devres_release,
+						sizeof(struct resource *),
+						GFP_KERNEL);
+				if (!d)
+					return NULL;
+				*d = res;
+				devres_add(dev, d);
+			}
+
+			pr_debug("%s: %pr\n", __func__, res);
+			return res;
+		}
+	}
+	if (dev)
+		return __devm_request_region(dev, parent, start, n, name);
+	return __request_region(parent, start, n, name, flags);
+}
+
+struct resource *__wrap___request_region(struct resource *parent,
+		resource_size_t start, resource_size_t n, const char *name,
+		int flags)
+{
+	return nfit_test_request_region(NULL, parent, start, n, name, flags);
+}
+EXPORT_SYMBOL(__wrap___request_region);
+
+int __wrap_insert_resource(struct resource *parent, struct resource *res)
+{
+	if (get_nfit_res(res->start))
+		return 0;
+	return insert_resource(parent, res);
+}
+EXPORT_SYMBOL(__wrap_insert_resource);
+
+int __wrap_remove_resource(struct resource *res)
+{
+	if (get_nfit_res(res->start))
+		return 0;
+	return remove_resource(res);
+}
+EXPORT_SYMBOL(__wrap_remove_resource);
+
+struct resource *__wrap___devm_request_region(struct device *dev,
+		struct resource *parent, resource_size_t start,
+		resource_size_t n, const char *name)
+{
+	if (!dev)
+		return NULL;
+	return nfit_test_request_region(dev, parent, start, n, name, 0);
+}
+EXPORT_SYMBOL(__wrap___devm_request_region);
+
+void __wrap___release_region(struct resource *parent, resource_size_t start,
+		resource_size_t n)
+{
+	if (!nfit_test_release_region(NULL, parent, start, n))
+		__release_region(parent, start, n);
+}
+EXPORT_SYMBOL(__wrap___release_region);
+
+void __wrap___devm_release_region(struct device *dev, struct resource *parent,
+		resource_size_t start, resource_size_t n)
+{
+	if (!nfit_test_release_region(dev, parent, start, n))
+		__devm_release_region(dev, parent, start, n);
+}
+EXPORT_SYMBOL(__wrap___devm_release_region);
+
+acpi_status __wrap_acpi_evaluate_object(acpi_handle handle, acpi_string path,
+		struct acpi_object_list *p, struct acpi_buffer *buf)
+{
+	struct nfit_test_resource *nfit_res = get_nfit_res((long) handle);
+	union acpi_object **obj;
+
+	if (!nfit_res || strcmp(path, "_FIT") || !buf)
+		return acpi_evaluate_object(handle, path, p, buf);
+
+	obj = nfit_res->buf;
+	buf->length = sizeof(union acpi_object);
+	buf->pointer = *obj;
+	return AE_OK;
+}
+EXPORT_SYMBOL(__wrap_acpi_evaluate_object);
+
+union acpi_object * __wrap_acpi_evaluate_dsm(acpi_handle handle, const guid_t *guid,
+		u64 rev, u64 func, union acpi_object *argv4)
+{
+	union acpi_object *obj = ERR_PTR(-ENXIO);
+	struct iomap_ops *ops;
+
+	rcu_read_lock();
+	ops = list_first_or_null_rcu(&iomap_head, typeof(*ops), list);
+	if (ops)
+		obj = ops->evaluate_dsm(handle, guid, rev, func, argv4);
+	rcu_read_unlock();
+
+	if (IS_ERR(obj))
+		return acpi_evaluate_dsm(handle, guid, rev, func, argv4);
+	return obj;
+}
+EXPORT_SYMBOL(__wrap_acpi_evaluate_dsm);
+
+MODULE_LICENSE("GPL v2");

tools/testing/nvdimm/test/nfit_test.h

@@ -0,0 +1,249 @@
+/*
+ * Copyright(c) 2013-2015 Intel Corporation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+#ifndef __NFIT_TEST_H__
+#define __NFIT_TEST_H__
+#include <linux/acpi.h>
+#include <linux/list.h>
+#include <linux/uuid.h>
+#include <linux/ioport.h>
+#include <linux/spinlock_types.h>
+
+struct nfit_test_request {
+	struct list_head list;
+	struct resource res;
+};
+
+struct nfit_test_resource {
+	struct list_head requests;
+	struct list_head list;
+	struct resource res;
+	struct device *dev;
+	spinlock_t lock;
+	int req_count;
+	void *buf;
+};
+
+#define ND_TRANSLATE_SPA_STATUS_INVALID_SPA  2
+#define NFIT_ARS_INJECT_INVALID 2
+
+enum err_inj_options {
+	ND_ARS_ERR_INJ_OPT_NOTIFY = 0,
+};
+
+/* nfit commands */
+enum nfit_cmd_num {
+	NFIT_CMD_TRANSLATE_SPA = 5,
+	NFIT_CMD_ARS_INJECT_SET = 7,
+	NFIT_CMD_ARS_INJECT_CLEAR = 8,
+	NFIT_CMD_ARS_INJECT_GET = 9,
+};
+
+struct nd_cmd_translate_spa {
+	__u64 spa;
+	__u32 status;
+	__u8  flags;
+	__u8  _reserved[3];
+	__u64 translate_length;
+	__u32 num_nvdimms;
+	struct nd_nvdimm_device {
+		__u32 nfit_device_handle;
+		__u32 _reserved;
+		__u64 dpa;
+	} __packed devices[0];
+
+} __packed;
+
+struct nd_cmd_ars_err_inj {
+	__u64 err_inj_spa_range_base;
+	__u64 err_inj_spa_range_length;
+	__u8  err_inj_options;
+	__u32 status;
+} __packed;
+
+struct nd_cmd_ars_err_inj_clr {
+	__u64 err_inj_clr_spa_range_base;
+	__u64 err_inj_clr_spa_range_length;
+	__u32 status;
+} __packed;
+
+struct nd_cmd_ars_err_inj_stat {
+	__u32 status;
+	__u32 inj_err_rec_count;
+	struct nd_error_stat_query_record {
+		__u64 err_inj_stat_spa_range_base;
+		__u64 err_inj_stat_spa_range_length;
+	} __packed record[0];
+} __packed;
+
+#define ND_INTEL_SMART			 1
+#define ND_INTEL_SMART_THRESHOLD	 2
+#define ND_INTEL_ENABLE_LSS_STATUS	10
+#define ND_INTEL_FW_GET_INFO		12
+#define ND_INTEL_FW_START_UPDATE	13
+#define ND_INTEL_FW_SEND_DATA		14
+#define ND_INTEL_FW_FINISH_UPDATE	15
+#define ND_INTEL_FW_FINISH_QUERY	16
+#define ND_INTEL_SMART_SET_THRESHOLD	17
+#define ND_INTEL_SMART_INJECT		18
+
+#define ND_INTEL_SMART_HEALTH_VALID             (1 << 0)
+#define ND_INTEL_SMART_SPARES_VALID             (1 << 1)
+#define ND_INTEL_SMART_USED_VALID               (1 << 2)
+#define ND_INTEL_SMART_MTEMP_VALID              (1 << 3)
+#define ND_INTEL_SMART_CTEMP_VALID              (1 << 4)
+#define ND_INTEL_SMART_SHUTDOWN_COUNT_VALID     (1 << 5)
+#define ND_INTEL_SMART_AIT_STATUS_VALID         (1 << 6)
+#define ND_INTEL_SMART_PTEMP_VALID              (1 << 7)
+#define ND_INTEL_SMART_ALARM_VALID              (1 << 9)
+#define ND_INTEL_SMART_SHUTDOWN_VALID           (1 << 10)
+#define ND_INTEL_SMART_VENDOR_VALID             (1 << 11)
+#define ND_INTEL_SMART_SPARE_TRIP               (1 << 0)
+#define ND_INTEL_SMART_TEMP_TRIP                (1 << 1)
+#define ND_INTEL_SMART_CTEMP_TRIP               (1 << 2)
+#define ND_INTEL_SMART_NON_CRITICAL_HEALTH      (1 << 0)
+#define ND_INTEL_SMART_CRITICAL_HEALTH          (1 << 1)
+#define ND_INTEL_SMART_FATAL_HEALTH             (1 << 2)
+#define ND_INTEL_SMART_INJECT_MTEMP		(1 << 0)
+#define ND_INTEL_SMART_INJECT_SPARE		(1 << 1)
+#define ND_INTEL_SMART_INJECT_FATAL		(1 << 2)
+#define ND_INTEL_SMART_INJECT_SHUTDOWN		(1 << 3)
+
+struct nd_intel_smart {
+	__u32 status;
+	union {
+		struct {
+			__u32 flags;
+			__u8 reserved0[4];
+			__u8 health;
+			__u8 spares;
+			__u8 life_used;
+			__u8 alarm_flags;
+			__u16 media_temperature;
+			__u16 ctrl_temperature;
+			__u32 shutdown_count;
+			__u8 ait_status;
+			__u16 pmic_temperature;
+			__u8 reserved1[8];
+			__u8 shutdown_state;
+			__u32 vendor_size;
+			__u8 vendor_data[92];
+		} __packed;
+		__u8 data[128];
+	};
+} __packed;
+
+struct nd_intel_smart_threshold {
+	__u32 status;
+	union {
+		struct {
+			__u16 alarm_control;
+			__u8 spares;
+			__u16 media_temperature;
+			__u16 ctrl_temperature;
+			__u8 reserved[1];
+		} __packed;
+		__u8 data[8];
+	};
+} __packed;
+
+struct nd_intel_smart_set_threshold {
+	__u16 alarm_control;
+	__u8 spares;
+	__u16 media_temperature;
+	__u16 ctrl_temperature;
+	__u32 status;
+} __packed;
+
+struct nd_intel_smart_inject {
+	__u64 flags;
+	__u8 mtemp_enable;
+	__u16 media_temperature;
+	__u8 spare_enable;
+	__u8 spares;
+	__u8 fatal_enable;
+	__u8 unsafe_shutdown_enable;
+	__u32 status;
+} __packed;
+
+#define INTEL_FW_STORAGE_SIZE		0x100000
+#define INTEL_FW_MAX_SEND_LEN		0xFFEC
+#define INTEL_FW_QUERY_INTERVAL		250000
+#define INTEL_FW_QUERY_MAX_TIME		3000000
+#define INTEL_FW_FIS_VERSION		0x0105
+#define INTEL_FW_FAKE_VERSION		0xffffffffabcd
+
+enum intel_fw_update_state {
+	FW_STATE_NEW = 0,
+	FW_STATE_IN_PROGRESS,
+	FW_STATE_VERIFY,
+	FW_STATE_UPDATED,
+};
+
+struct nd_intel_fw_info {
+	__u32 status;
+	__u32 storage_size;
+	__u32 max_send_len;
+	__u32 query_interval;
+	__u32 max_query_time;
+	__u8 update_cap;
+	__u8 reserved[3];
+	__u32 fis_version;
+	__u64 run_version;
+	__u64 updated_version;
+} __packed;
+
+struct nd_intel_fw_start {
+	__u32 status;
+	__u32 context;
+} __packed;
+
+/* this one has the output first because the variable input data size */
+struct nd_intel_fw_send_data {
+	__u32 context;
+	__u32 offset;
+	__u32 length;
+	__u8 data[0];
+/* this field is not declared due ot variable data from input */
+/*	__u32 status; */
+} __packed;
+
+struct nd_intel_fw_finish_update {
+	__u8 ctrl_flags;
+	__u8 reserved[3];
+	__u32 context;
+	__u32 status;
+} __packed;
+
+struct nd_intel_fw_finish_query {
+	__u32 context;
+	__u32 status;
+	__u64 updated_fw_rev;
+} __packed;
+
+struct nd_intel_lss {
+	__u8 enable;
+	__u32 status;
+} __packed;
+
+typedef struct nfit_test_resource *(*nfit_test_lookup_fn)(resource_size_t);
+typedef union acpi_object *(*nfit_test_evaluate_dsm_fn)(acpi_handle handle,
+		 const guid_t *guid, u64 rev, u64 func,
+		 union acpi_object *argv4);
+void __iomem *__wrap_ioremap_nocache(resource_size_t offset,
+		unsigned long size);
+void __wrap_iounmap(volatile void __iomem *addr);
+void nfit_test_setup(nfit_test_lookup_fn lookup,
+		nfit_test_evaluate_dsm_fn evaluate);
+void nfit_test_teardown(void);
+struct nfit_test_resource *get_nfit_res(resource_size_t resource);
+#endif

tools/testing/nvdimm/watermark.h

@@ -0,0 +1,21 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright(c) 2018 Intel Corporation. All rights reserved.
+#ifndef _TEST_NVDIMM_WATERMARK_H_
+#define _TEST_NVDIMM_WATERMARK_H_
+int pmem_test(void);
+int libnvdimm_test(void);
+int acpi_nfit_test(void);
+int device_dax_test(void);
+
+/*
+ * dummy routine for nfit_test to validate it is linking to the properly
+ * mocked module and not the standard one from the base tree.
+ */
+#define nfit_test_watermark(x)				\
+int x##_test(void)					\
+{							\
+	pr_debug("%s for nfit_test\n", KBUILD_MODNAME);	\
+	return 0;					\
+}							\
+EXPORT_SYMBOL(x##_test)
+#endif /* _TEST_NVDIMM_WATERMARK_H_ */

tools/testing/radix-tree/Makefile

@@ -0,0 +1,51 @@
+# SPDX-License-Identifier: GPL-2.0
+
+CFLAGS += -I. -I../../include -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address \
+	  -fsanitize=undefined
+LDFLAGS += -fsanitize=address -fsanitize=undefined
+LDLIBS+= -lpthread -lurcu
+TARGETS = main idr-test multiorder
+CORE_OFILES := radix-tree.o idr.o linux.o test.o find_bit.o
+OFILES = main.o $(CORE_OFILES) regression1.o regression2.o regression3.o \
+	 tag_check.o multiorder.o idr-test.o iteration_check.o benchmark.o
+
+ifndef SHIFT
+	SHIFT=3
+endif
+
+ifeq ($(BUILD), 32)
+	CFLAGS += -m32
+	LDFLAGS += -m32
+endif
+
+targets: generated/map-shift.h $(TARGETS)
+
+main:	$(OFILES)
+
+idr-test.o: ../../../lib/test_ida.c
+idr-test: idr-test.o $(CORE_OFILES)
+
+multiorder: multiorder.o $(CORE_OFILES)
+
+clean:
+	$(RM) $(TARGETS) *.o radix-tree.c idr.c generated/map-shift.h
+
+vpath %.c ../../lib
+
+$(OFILES): Makefile *.h */*.h generated/map-shift.h \
+	../../include/linux/*.h \
+	../../include/asm/*.h \
+	../../../include/linux/radix-tree.h \
+	../../../include/linux/idr.h
+
+radix-tree.c: ../../../lib/radix-tree.c
+	sed -e 's/^static //' -e 's/__always_inline //' -e 's/inline //' < $< > $@
+
+idr.c: ../../../lib/idr.c
+	sed -e 's/^static //' -e 's/__always_inline //' -e 's/inline //' < $< > $@
+
+generated/map-shift.h:
+	@if ! grep -qws $(SHIFT) generated/map-shift.h; then		\
+		echo "#define RADIX_TREE_MAP_SHIFT $(SHIFT)" >		\
+				generated/map-shift.h;			\
+	fi

tools/testing/radix-tree/benchmark.c

@@ -0,0 +1,257 @@
+/*
+ * benchmark.c:
+ * Author: Konstantin Khlebnikov <koct9i@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include <linux/radix-tree.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <time.h>
+#include "test.h"
+
+#define for_each_index(i, base, order) \
+	        for (i = base; i < base + (1 << order); i++)
+
+#define NSEC_PER_SEC	1000000000L
+
+static long long benchmark_iter(struct radix_tree_root *root, bool tagged)
+{
+	volatile unsigned long sink = 0;
+	struct radix_tree_iter iter;
+	struct timespec start, finish;
+	long long nsec;
+	int l, loops = 1;
+	void **slot;
+
+#ifdef BENCHMARK
+again:
+#endif
+	clock_gettime(CLOCK_MONOTONIC, &start);
+	for (l = 0; l < loops; l++) {
+		if (tagged) {
+			radix_tree_for_each_tagged(slot, root, &iter, 0, 0)
+				sink ^= (unsigned long)slot;
+		} else {
+			radix_tree_for_each_slot(slot, root, &iter, 0)
+				sink ^= (unsigned long)slot;
+		}
+	}
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+	       (finish.tv_nsec - start.tv_nsec);
+
+#ifdef BENCHMARK
+	if (loops == 1 && nsec * 5 < NSEC_PER_SEC) {
+		loops = NSEC_PER_SEC / nsec / 4 + 1;
+		goto again;
+	}
+#endif
+
+	nsec /= loops;
+	return nsec;
+}
+
+static void benchmark_insert(struct radix_tree_root *root,
+			     unsigned long size, unsigned long step, int order)
+{
+	struct timespec start, finish;
+	unsigned long index;
+	long long nsec;
+
+	clock_gettime(CLOCK_MONOTONIC, &start);
+
+	for (index = 0 ; index < size ; index += step)
+		item_insert_order(root, index, order);
+
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+	       (finish.tv_nsec - start.tv_nsec);
+
+	printv(2, "Size: %8ld, step: %8ld, order: %d, insertion: %15lld ns\n",
+		size, step, order, nsec);
+}
+
+static void benchmark_tagging(struct radix_tree_root *root,
+			     unsigned long size, unsigned long step, int order)
+{
+	struct timespec start, finish;
+	unsigned long index;
+	long long nsec;
+
+	clock_gettime(CLOCK_MONOTONIC, &start);
+
+	for (index = 0 ; index < size ; index += step)
+		radix_tree_tag_set(root, index, 0);
+
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+	       (finish.tv_nsec - start.tv_nsec);
+
+	printv(2, "Size: %8ld, step: %8ld, order: %d, tagging: %17lld ns\n",
+		size, step, order, nsec);
+}
+
+static void benchmark_delete(struct radix_tree_root *root,
+			     unsigned long size, unsigned long step, int order)
+{
+	struct timespec start, finish;
+	unsigned long index, i;
+	long long nsec;
+
+	clock_gettime(CLOCK_MONOTONIC, &start);
+
+	for (index = 0 ; index < size ; index += step)
+		for_each_index(i, index, order)
+			item_delete(root, i);
+
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+	       (finish.tv_nsec - start.tv_nsec);
+
+	printv(2, "Size: %8ld, step: %8ld, order: %d, deletion: %16lld ns\n",
+		size, step, order, nsec);
+}
+
+static void benchmark_size(unsigned long size, unsigned long step, int order)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	long long normal, tagged;
+
+	benchmark_insert(&tree, size, step, order);
+	benchmark_tagging(&tree, size, step, order);
+
+	tagged = benchmark_iter(&tree, true);
+	normal = benchmark_iter(&tree, false);
+
+	printv(2, "Size: %8ld, step: %8ld, order: %d, tagged iteration: %8lld ns\n",
+		size, step, order, tagged);
+	printv(2, "Size: %8ld, step: %8ld, order: %d, normal iteration: %8lld ns\n",
+		size, step, order, normal);
+
+	benchmark_delete(&tree, size, step, order);
+
+	item_kill_tree(&tree);
+	rcu_barrier();
+}
+
+static long long  __benchmark_split(unsigned long index,
+				    int old_order, int new_order)
+{
+	struct timespec start, finish;
+	long long nsec;
+	RADIX_TREE(tree, GFP_ATOMIC);
+
+	item_insert_order(&tree, index, old_order);
+
+	clock_gettime(CLOCK_MONOTONIC, &start);
+	radix_tree_split(&tree, index, new_order);
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+	       (finish.tv_nsec - start.tv_nsec);
+
+	item_kill_tree(&tree);
+
+	return nsec;
+
+}
+
+static void benchmark_split(unsigned long size, unsigned long step)
+{
+	int i, j, idx;
+	long long nsec = 0;
+
+
+	for (idx = 0; idx < size; idx += step) {
+		for (i = 3; i < 11; i++) {
+			for (j = 0; j < i; j++) {
+				nsec += __benchmark_split(idx, i, j);
+			}
+		}
+	}
+
+	printv(2, "Size %8ld, step %8ld, split time %10lld ns\n",
+			size, step, nsec);
+
+}
+
+static long long  __benchmark_join(unsigned long index,
+			     unsigned order1, unsigned order2)
+{
+	unsigned long loc;
+	struct timespec start, finish;
+	long long nsec;
+	void *item, *item2 = item_create(index + 1, order1);
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert_order(&tree, index, order2);
+	item = radix_tree_lookup(&tree, index);
+
+	clock_gettime(CLOCK_MONOTONIC, &start);
+	radix_tree_join(&tree, index + 1, order1, item2);
+	clock_gettime(CLOCK_MONOTONIC, &finish);
+	nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC +
+		(finish.tv_nsec - start.tv_nsec);
+
+	loc = find_item(&tree, item);
+	if (loc == -1)
+		free(item);
+
+	item_kill_tree(&tree);
+
+	return nsec;
+}
+
+static void benchmark_join(unsigned long step)
+{
+	int i, j, idx;
+	long long nsec = 0;
+
+	for (idx = 0; idx < 1 << 10; idx += step) {
+		for (i = 1; i < 15; i++) {
+			for (j = 0; j < i; j++) {
+				nsec += __benchmark_join(idx, i, j);
+			}
+		}
+	}
+
+	printv(2, "Size %8d, step %8ld, join time %10lld ns\n",
+			1 << 10, step, nsec);
+}
+
+void benchmark(void)
+{
+	unsigned long size[] = {1 << 10, 1 << 20, 0};
+	unsigned long step[] = {1, 2, 7, 15, 63, 64, 65,
+				128, 256, 512, 12345, 0};
+	int c, s;
+
+	printv(1, "starting benchmarks\n");
+	printv(1, "RADIX_TREE_MAP_SHIFT = %d\n", RADIX_TREE_MAP_SHIFT);
+
+	for (c = 0; size[c]; c++)
+		for (s = 0; step[s]; s++)
+			benchmark_size(size[c], step[s], 0);
+
+	for (c = 0; size[c]; c++)
+		for (s = 0; step[s]; s++)
+			benchmark_size(size[c], step[s] << 9, 9);
+
+	for (c = 0; size[c]; c++)
+		for (s = 0; step[s]; s++)
+			benchmark_split(size[c], step[s]);
+
+	for (s = 0; step[s]; s++)
+		benchmark_join(step[s]);
+}

tools/testing/radix-tree/generated/autoconf.h

@@ -0,0 +1 @@
+#define CONFIG_RADIX_TREE_MULTIORDER 1

tools/testing/radix-tree/idr-test.c

@@ -0,0 +1,512 @@
+/*
+ * idr-test.c: Test the IDR API
+ * Copyright (c) 2016 Matthew Wilcox <willy@infradead.org>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include <linux/bitmap.h>
+#include <linux/idr.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+
+#include "test.h"
+
+#define DUMMY_PTR	((void *)0x12)
+
+int item_idr_free(int id, void *p, void *data)
+{
+	struct item *item = p;
+	assert(item->index == id);
+	free(p);
+
+	return 0;
+}
+
+void item_idr_remove(struct idr *idr, int id)
+{
+	struct item *item = idr_find(idr, id);
+	assert(item->index == id);
+	idr_remove(idr, id);
+	free(item);
+}
+
+void idr_alloc_test(void)
+{
+	unsigned long i;
+	DEFINE_IDR(idr);
+
+	assert(idr_alloc_cyclic(&idr, DUMMY_PTR, 0, 0x4000, GFP_KERNEL) == 0);
+	assert(idr_alloc_cyclic(&idr, DUMMY_PTR, 0x3ffd, 0x4000, GFP_KERNEL) == 0x3ffd);
+	idr_remove(&idr, 0x3ffd);
+	idr_remove(&idr, 0);
+
+	for (i = 0x3ffe; i < 0x4003; i++) {
+		int id;
+		struct item *item;
+
+		if (i < 0x4000)
+			item = item_create(i, 0);
+		else
+			item = item_create(i - 0x3fff, 0);
+
+		id = idr_alloc_cyclic(&idr, item, 1, 0x4000, GFP_KERNEL);
+		assert(id == item->index);
+	}
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+}
+
+void idr_replace_test(void)
+{
+	DEFINE_IDR(idr);
+
+	idr_alloc(&idr, (void *)-1, 10, 11, GFP_KERNEL);
+	idr_replace(&idr, &idr, 10);
+
+	idr_destroy(&idr);
+}
+
+/*
+ * Unlike the radix tree, you can put a NULL pointer -- with care -- into
+ * the IDR.  Some interfaces, like idr_find() do not distinguish between
+ * "present, value is NULL" and "not present", but that's exactly what some
+ * users want.
+ */
+void idr_null_test(void)
+{
+	int i;
+	DEFINE_IDR(idr);
+
+	assert(idr_is_empty(&idr));
+
+	assert(idr_alloc(&idr, NULL, 0, 0, GFP_KERNEL) == 0);
+	assert(!idr_is_empty(&idr));
+	idr_remove(&idr, 0);
+	assert(idr_is_empty(&idr));
+
+	assert(idr_alloc(&idr, NULL, 0, 0, GFP_KERNEL) == 0);
+	assert(!idr_is_empty(&idr));
+	idr_destroy(&idr);
+	assert(idr_is_empty(&idr));
+
+	for (i = 0; i < 10; i++) {
+		assert(idr_alloc(&idr, NULL, 0, 0, GFP_KERNEL) == i);
+	}
+
+	assert(idr_replace(&idr, DUMMY_PTR, 3) == NULL);
+	assert(idr_replace(&idr, DUMMY_PTR, 4) == NULL);
+	assert(idr_replace(&idr, NULL, 4) == DUMMY_PTR);
+	assert(idr_replace(&idr, DUMMY_PTR, 11) == ERR_PTR(-ENOENT));
+	idr_remove(&idr, 5);
+	assert(idr_alloc(&idr, NULL, 0, 0, GFP_KERNEL) == 5);
+	idr_remove(&idr, 5);
+
+	for (i = 0; i < 9; i++) {
+		idr_remove(&idr, i);
+		assert(!idr_is_empty(&idr));
+	}
+	idr_remove(&idr, 8);
+	assert(!idr_is_empty(&idr));
+	idr_remove(&idr, 9);
+	assert(idr_is_empty(&idr));
+
+	assert(idr_alloc(&idr, NULL, 0, 0, GFP_KERNEL) == 0);
+	assert(idr_replace(&idr, DUMMY_PTR, 3) == ERR_PTR(-ENOENT));
+	assert(idr_replace(&idr, DUMMY_PTR, 0) == NULL);
+	assert(idr_replace(&idr, NULL, 0) == DUMMY_PTR);
+
+	idr_destroy(&idr);
+	assert(idr_is_empty(&idr));
+
+	for (i = 1; i < 10; i++) {
+		assert(idr_alloc(&idr, NULL, 1, 0, GFP_KERNEL) == i);
+	}
+
+	idr_destroy(&idr);
+	assert(idr_is_empty(&idr));
+}
+
+void idr_nowait_test(void)
+{
+	unsigned int i;
+	DEFINE_IDR(idr);
+
+	idr_preload(GFP_KERNEL);
+
+	for (i = 0; i < 3; i++) {
+		struct item *item = item_create(i, 0);
+		assert(idr_alloc(&idr, item, i, i + 1, GFP_NOWAIT) == i);
+	}
+
+	idr_preload_end();
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+}
+
+void idr_get_next_test(int base)
+{
+	unsigned long i;
+	int nextid;
+	DEFINE_IDR(idr);
+	idr_init_base(&idr, base);
+
+	int indices[] = {4, 7, 9, 15, 65, 128, 1000, 99999, 0};
+
+	for(i = 0; indices[i]; i++) {
+		struct item *item = item_create(indices[i], 0);
+		assert(idr_alloc(&idr, item, indices[i], indices[i+1],
+				 GFP_KERNEL) == indices[i]);
+	}
+
+	for(i = 0, nextid = 0; indices[i]; i++) {
+		idr_get_next(&idr, &nextid);
+		assert(nextid == indices[i]);
+		nextid++;
+	}
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+}
+
+int idr_u32_cb(int id, void *ptr, void *data)
+{
+	BUG_ON(id < 0);
+	BUG_ON(ptr != DUMMY_PTR);
+	return 0;
+}
+
+void idr_u32_test1(struct idr *idr, u32 handle)
+{
+	static bool warned = false;
+	u32 id = handle;
+	int sid = 0;
+	void *ptr;
+
+	BUG_ON(idr_alloc_u32(idr, DUMMY_PTR, &id, id, GFP_KERNEL));
+	BUG_ON(id != handle);
+	BUG_ON(idr_alloc_u32(idr, DUMMY_PTR, &id, id, GFP_KERNEL) != -ENOSPC);
+	BUG_ON(id != handle);
+	if (!warned && id > INT_MAX)
+		printk("vvv Ignore these warnings\n");
+	ptr = idr_get_next(idr, &sid);
+	if (id > INT_MAX) {
+		BUG_ON(ptr != NULL);
+		BUG_ON(sid != 0);
+	} else {
+		BUG_ON(ptr != DUMMY_PTR);
+		BUG_ON(sid != id);
+	}
+	idr_for_each(idr, idr_u32_cb, NULL);
+	if (!warned && id > INT_MAX) {
+		printk("^^^ Warnings over\n");
+		warned = true;
+	}
+	BUG_ON(idr_remove(idr, id) != DUMMY_PTR);
+	BUG_ON(!idr_is_empty(idr));
+}
+
+void idr_u32_test(int base)
+{
+	DEFINE_IDR(idr);
+	idr_init_base(&idr, base);
+	idr_u32_test1(&idr, 10);
+	idr_u32_test1(&idr, 0x7fffffff);
+	idr_u32_test1(&idr, 0x80000000);
+	idr_u32_test1(&idr, 0x80000001);
+	idr_u32_test1(&idr, 0xffe00000);
+	idr_u32_test1(&idr, 0xffffffff);
+}
+
+static inline void *idr_mk_value(unsigned long v)
+{
+	BUG_ON((long)v < 0);
+	return (void *)((v & 1) | 2 | (v << 1));
+}
+
+DEFINE_IDR(find_idr);
+
+static void *idr_throbber(void *arg)
+{
+	time_t start = time(NULL);
+	int id = *(int *)arg;
+
+	rcu_register_thread();
+	do {
+		idr_alloc(&find_idr, idr_mk_value(id), id, id + 1, GFP_KERNEL);
+		idr_remove(&find_idr, id);
+	} while (time(NULL) < start + 10);
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+void idr_find_test_1(int anchor_id, int throbber_id)
+{
+	pthread_t throbber;
+	time_t start = time(NULL);
+
+	pthread_create(&throbber, NULL, idr_throbber, &throbber_id);
+
+	BUG_ON(idr_alloc(&find_idr, idr_mk_value(anchor_id), anchor_id,
+				anchor_id + 1, GFP_KERNEL) != anchor_id);
+
+	do {
+		int id = 0;
+		void *entry = idr_get_next(&find_idr, &id);
+		BUG_ON(entry != idr_mk_value(id));
+	} while (time(NULL) < start + 11);
+
+	pthread_join(throbber, NULL);
+
+	idr_remove(&find_idr, anchor_id);
+	BUG_ON(!idr_is_empty(&find_idr));
+}
+
+void idr_find_test(void)
+{
+	idr_find_test_1(100000, 0);
+	idr_find_test_1(0, 100000);
+}
+
+void idr_checks(void)
+{
+	unsigned long i;
+	DEFINE_IDR(idr);
+
+	for (i = 0; i < 10000; i++) {
+		struct item *item = item_create(i, 0);
+		assert(idr_alloc(&idr, item, 0, 20000, GFP_KERNEL) == i);
+	}
+
+	assert(idr_alloc(&idr, DUMMY_PTR, 5, 30, GFP_KERNEL) < 0);
+
+	for (i = 0; i < 5000; i++)
+		item_idr_remove(&idr, i);
+
+	idr_remove(&idr, 3);
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+
+	assert(idr_is_empty(&idr));
+
+	idr_remove(&idr, 3);
+	idr_remove(&idr, 0);
+
+	assert(idr_alloc(&idr, DUMMY_PTR, 0, 0, GFP_KERNEL) == 0);
+	idr_remove(&idr, 1);
+	for (i = 1; i < RADIX_TREE_MAP_SIZE; i++)
+		assert(idr_alloc(&idr, DUMMY_PTR, 0, 0, GFP_KERNEL) == i);
+	idr_remove(&idr, 1 << 30);
+	idr_destroy(&idr);
+
+	for (i = INT_MAX - 3UL; i < INT_MAX + 1UL; i++) {
+		struct item *item = item_create(i, 0);
+		assert(idr_alloc(&idr, item, i, i + 10, GFP_KERNEL) == i);
+	}
+	assert(idr_alloc(&idr, DUMMY_PTR, i - 2, i, GFP_KERNEL) == -ENOSPC);
+	assert(idr_alloc(&idr, DUMMY_PTR, i - 2, i + 10, GFP_KERNEL) == -ENOSPC);
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+	idr_destroy(&idr);
+
+	assert(idr_is_empty(&idr));
+
+	idr_set_cursor(&idr, INT_MAX - 3UL);
+	for (i = INT_MAX - 3UL; i < INT_MAX + 3UL; i++) {
+		struct item *item;
+		unsigned int id;
+		if (i <= INT_MAX)
+			item = item_create(i, 0);
+		else
+			item = item_create(i - INT_MAX - 1, 0);
+
+		id = idr_alloc_cyclic(&idr, item, 0, 0, GFP_KERNEL);
+		assert(id == item->index);
+	}
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+	assert(idr_is_empty(&idr));
+
+	for (i = 1; i < 10000; i++) {
+		struct item *item = item_create(i, 0);
+		assert(idr_alloc(&idr, item, 1, 20000, GFP_KERNEL) == i);
+	}
+
+	idr_for_each(&idr, item_idr_free, &idr);
+	idr_destroy(&idr);
+
+	idr_replace_test();
+	idr_alloc_test();
+	idr_null_test();
+	idr_nowait_test();
+	idr_get_next_test(0);
+	idr_get_next_test(1);
+	idr_get_next_test(4);
+	idr_u32_test(4);
+	idr_u32_test(1);
+	idr_u32_test(0);
+	idr_find_test();
+}
+
+#define module_init(x)
+#define module_exit(x)
+#define MODULE_AUTHOR(x)
+#define MODULE_LICENSE(x)
+#define dump_stack()    assert(0)
+void ida_dump(struct ida *);
+
+#include "../../../lib/test_ida.c"
+
+/*
+ * Check that we get the correct error when we run out of memory doing
+ * allocations.  In userspace, GFP_NOWAIT will always fail an allocation.
+ * The first test is for not having a bitmap available, and the second test
+ * is for not being able to allocate a level of the radix tree.
+ */
+void ida_check_nomem(void)
+{
+	DEFINE_IDA(ida);
+	int id;
+
+	id = ida_alloc_min(&ida, 256, GFP_NOWAIT);
+	IDA_BUG_ON(&ida, id != -ENOMEM);
+	id = ida_alloc_min(&ida, 1UL << 30, GFP_NOWAIT);
+	IDA_BUG_ON(&ida, id != -ENOMEM);
+	IDA_BUG_ON(&ida, !ida_is_empty(&ida));
+}
+
+/*
+ * Check handling of conversions between exceptional entries and full bitmaps.
+ */
+void ida_check_conv_user(void)
+{
+	DEFINE_IDA(ida);
+	unsigned long i;
+
+	radix_tree_cpu_dead(1);
+	for (i = 0; i < 1000000; i++) {
+		int id = ida_alloc(&ida, GFP_NOWAIT);
+		if (id == -ENOMEM) {
+			IDA_BUG_ON(&ida, (i % IDA_BITMAP_BITS) !=
+					BITS_PER_LONG - 2);
+			id = ida_alloc(&ida, GFP_KERNEL);
+		} else {
+			IDA_BUG_ON(&ida, (i % IDA_BITMAP_BITS) ==
+					BITS_PER_LONG - 2);
+		}
+		IDA_BUG_ON(&ida, id != i);
+	}
+	ida_destroy(&ida);
+}
+
+void ida_check_random(void)
+{
+	DEFINE_IDA(ida);
+	DECLARE_BITMAP(bitmap, 2048);
+	unsigned int i;
+	time_t s = time(NULL);
+
+ repeat:
+	memset(bitmap, 0, sizeof(bitmap));
+	for (i = 0; i < 100000; i++) {
+		int i = rand();
+		int bit = i & 2047;
+		if (test_bit(bit, bitmap)) {
+			__clear_bit(bit, bitmap);
+			ida_free(&ida, bit);
+		} else {
+			__set_bit(bit, bitmap);
+			IDA_BUG_ON(&ida, ida_alloc_min(&ida, bit, GFP_KERNEL)
+					!= bit);
+		}
+	}
+	ida_destroy(&ida);
+	if (time(NULL) < s + 10)
+		goto repeat;
+}
+
+void ida_simple_get_remove_test(void)
+{
+	DEFINE_IDA(ida);
+	unsigned long i;
+
+	for (i = 0; i < 10000; i++) {
+		assert(ida_simple_get(&ida, 0, 20000, GFP_KERNEL) == i);
+	}
+	assert(ida_simple_get(&ida, 5, 30, GFP_KERNEL) < 0);
+
+	for (i = 0; i < 10000; i++) {
+		ida_simple_remove(&ida, i);
+	}
+	assert(ida_is_empty(&ida));
+
+	ida_destroy(&ida);
+}
+
+void user_ida_checks(void)
+{
+	radix_tree_cpu_dead(1);
+
+	ida_check_nomem();
+	ida_check_conv_user();
+	ida_check_random();
+	ida_simple_get_remove_test();
+
+	radix_tree_cpu_dead(1);
+}
+
+static void *ida_random_fn(void *arg)
+{
+	rcu_register_thread();
+	ida_check_random();
+	rcu_unregister_thread();
+	return NULL;
+}
+
+void ida_thread_tests(void)
+{
+	pthread_t threads[20];
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(threads); i++)
+		if (pthread_create(&threads[i], NULL, ida_random_fn, NULL)) {
+			perror("creating ida thread");
+			exit(1);
+		}
+
+	while (i--)
+		pthread_join(threads[i], NULL);
+}
+
+void ida_tests(void)
+{
+	user_ida_checks();
+	ida_checks();
+	ida_exit();
+	ida_thread_tests();
+}
+
+int __weak main(void)
+{
+	radix_tree_init();
+	idr_checks();
+	ida_tests();
+	radix_tree_cpu_dead(1);
+	rcu_barrier();
+	if (nr_allocated)
+		printf("nr_allocated = %d\n", nr_allocated);
+	return 0;
+}

tools/testing/radix-tree/iteration_check.c

@@ -0,0 +1,221 @@
+/*
+ * iteration_check.c: test races having to do with radix tree iteration
+ * Copyright (c) 2016 Intel Corporation
+ * Author: Ross Zwisler <ross.zwisler@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include <linux/radix-tree.h>
+#include <pthread.h>
+#include "test.h"
+
+#define NUM_THREADS	5
+#define MAX_IDX		100
+#define TAG		0
+#define NEW_TAG		1
+
+static pthread_mutex_t tree_lock = PTHREAD_MUTEX_INITIALIZER;
+static pthread_t threads[NUM_THREADS];
+static unsigned int seeds[3];
+static RADIX_TREE(tree, GFP_KERNEL);
+static bool test_complete;
+static int max_order;
+
+/* relentlessly fill the tree with tagged entries */
+static void *add_entries_fn(void *arg)
+{
+	rcu_register_thread();
+
+	while (!test_complete) {
+		unsigned long pgoff;
+		int order;
+
+		for (pgoff = 0; pgoff < MAX_IDX; pgoff++) {
+			pthread_mutex_lock(&tree_lock);
+			for (order = max_order; order >= 0; order--) {
+				if (item_insert_order(&tree, pgoff, order)
+						== 0) {
+					item_tag_set(&tree, pgoff, TAG);
+					break;
+				}
+			}
+			pthread_mutex_unlock(&tree_lock);
+		}
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+/*
+ * Iterate over the tagged entries, doing a radix_tree_iter_retry() as we find
+ * things that have been removed and randomly resetting our iteration to the
+ * next chunk with radix_tree_iter_resume().  Both radix_tree_iter_retry() and
+ * radix_tree_iter_resume() cause radix_tree_next_slot() to be called with a
+ * NULL 'slot' variable.
+ */
+static void *tagged_iteration_fn(void *arg)
+{
+	struct radix_tree_iter iter;
+	void **slot;
+
+	rcu_register_thread();
+
+	while (!test_complete) {
+		rcu_read_lock();
+		radix_tree_for_each_tagged(slot, &tree, &iter, 0, TAG) {
+			void *entry = radix_tree_deref_slot(slot);
+			if (unlikely(!entry))
+				continue;
+
+			if (radix_tree_deref_retry(entry)) {
+				slot = radix_tree_iter_retry(&iter);
+				continue;
+			}
+
+			if (rand_r(&seeds[0]) % 50 == 0) {
+				slot = radix_tree_iter_resume(slot, &iter);
+				rcu_read_unlock();
+				rcu_barrier();
+				rcu_read_lock();
+			}
+		}
+		rcu_read_unlock();
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+/*
+ * Iterate over the entries, doing a radix_tree_iter_retry() as we find things
+ * that have been removed and randomly resetting our iteration to the next
+ * chunk with radix_tree_iter_resume().  Both radix_tree_iter_retry() and
+ * radix_tree_iter_resume() cause radix_tree_next_slot() to be called with a
+ * NULL 'slot' variable.
+ */
+static void *untagged_iteration_fn(void *arg)
+{
+	struct radix_tree_iter iter;
+	void **slot;
+
+	rcu_register_thread();
+
+	while (!test_complete) {
+		rcu_read_lock();
+		radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+			void *entry = radix_tree_deref_slot(slot);
+			if (unlikely(!entry))
+				continue;
+
+			if (radix_tree_deref_retry(entry)) {
+				slot = radix_tree_iter_retry(&iter);
+				continue;
+			}
+
+			if (rand_r(&seeds[1]) % 50 == 0) {
+				slot = radix_tree_iter_resume(slot, &iter);
+				rcu_read_unlock();
+				rcu_barrier();
+				rcu_read_lock();
+			}
+		}
+		rcu_read_unlock();
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+/*
+ * Randomly remove entries to help induce radix_tree_iter_retry() calls in the
+ * two iteration functions.
+ */
+static void *remove_entries_fn(void *arg)
+{
+	rcu_register_thread();
+
+	while (!test_complete) {
+		int pgoff;
+
+		pgoff = rand_r(&seeds[2]) % MAX_IDX;
+
+		pthread_mutex_lock(&tree_lock);
+		item_delete(&tree, pgoff);
+		pthread_mutex_unlock(&tree_lock);
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+static void *tag_entries_fn(void *arg)
+{
+	rcu_register_thread();
+
+	while (!test_complete) {
+		tag_tagged_items(&tree, &tree_lock, 0, MAX_IDX, 10, TAG,
+					NEW_TAG);
+	}
+	rcu_unregister_thread();
+	return NULL;
+}
+
+/* This is a unit test for a bug found by the syzkaller tester */
+void iteration_test(unsigned order, unsigned test_duration)
+{
+	int i;
+
+	printv(1, "Running %siteration tests for %d seconds\n",
+			order > 0 ? "multiorder " : "", test_duration);
+
+	max_order = order;
+	test_complete = false;
+
+	for (i = 0; i < 3; i++)
+		seeds[i] = rand();
+
+	if (pthread_create(&threads[0], NULL, tagged_iteration_fn, NULL)) {
+		perror("create tagged iteration thread");
+		exit(1);
+	}
+	if (pthread_create(&threads[1], NULL, untagged_iteration_fn, NULL)) {
+		perror("create untagged iteration thread");
+		exit(1);
+	}
+	if (pthread_create(&threads[2], NULL, add_entries_fn, NULL)) {
+		perror("create add entry thread");
+		exit(1);
+	}
+	if (pthread_create(&threads[3], NULL, remove_entries_fn, NULL)) {
+		perror("create remove entry thread");
+		exit(1);
+	}
+	if (pthread_create(&threads[4], NULL, tag_entries_fn, NULL)) {
+		perror("create tag entry thread");
+		exit(1);
+	}
+
+	sleep(test_duration);
+	test_complete = true;
+
+	for (i = 0; i < NUM_THREADS; i++) {
+		if (pthread_join(threads[i], NULL)) {
+			perror("pthread_join");
+			exit(1);
+		}
+	}
+
+	item_kill_tree(&tree);
+}

tools/testing/radix-tree/linux.c

@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdlib.h>
+#include <string.h>
+#include <malloc.h>
+#include <pthread.h>
+#include <unistd.h>
+#include <assert.h>
+
+#include <linux/gfp.h>
+#include <linux/poison.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <urcu/uatomic.h>
+
+int nr_allocated;
+int preempt_count;
+int kmalloc_verbose;
+int test_verbose;
+
+struct kmem_cache {
+	pthread_mutex_t lock;
+	int size;
+	int nr_objs;
+	void *objs;
+	void (*ctor)(void *);
+};
+
+void *kmem_cache_alloc(struct kmem_cache *cachep, int flags)
+{
+	struct radix_tree_node *node;
+
+	if (!(flags & __GFP_DIRECT_RECLAIM))
+		return NULL;
+
+	pthread_mutex_lock(&cachep->lock);
+	if (cachep->nr_objs) {
+		cachep->nr_objs--;
+		node = cachep->objs;
+		cachep->objs = node->parent;
+		pthread_mutex_unlock(&cachep->lock);
+		node->parent = NULL;
+	} else {
+		pthread_mutex_unlock(&cachep->lock);
+		node = malloc(cachep->size);
+		if (cachep->ctor)
+			cachep->ctor(node);
+	}
+
+	uatomic_inc(&nr_allocated);
+	if (kmalloc_verbose)
+		printf("Allocating %p from slab\n", node);
+	return node;
+}
+
+void kmem_cache_free(struct kmem_cache *cachep, void *objp)
+{
+	assert(objp);
+	uatomic_dec(&nr_allocated);
+	if (kmalloc_verbose)
+		printf("Freeing %p to slab\n", objp);
+	pthread_mutex_lock(&cachep->lock);
+	if (cachep->nr_objs > 10) {
+		memset(objp, POISON_FREE, cachep->size);
+		free(objp);
+	} else {
+		struct radix_tree_node *node = objp;
+		cachep->nr_objs++;
+		node->parent = cachep->objs;
+		cachep->objs = node;
+	}
+	pthread_mutex_unlock(&cachep->lock);
+}
+
+void *kmalloc(size_t size, gfp_t gfp)
+{
+	void *ret;
+
+	if (!(gfp & __GFP_DIRECT_RECLAIM))
+		return NULL;
+
+	ret = malloc(size);
+	uatomic_inc(&nr_allocated);
+	if (kmalloc_verbose)
+		printf("Allocating %p from malloc\n", ret);
+	if (gfp & __GFP_ZERO)
+		memset(ret, 0, size);
+	return ret;
+}
+
+void kfree(void *p)
+{
+	if (!p)
+		return;
+	uatomic_dec(&nr_allocated);
+	if (kmalloc_verbose)
+		printf("Freeing %p to malloc\n", p);
+	free(p);
+}
+
+struct kmem_cache *
+kmem_cache_create(const char *name, size_t size, size_t offset,
+	unsigned long flags, void (*ctor)(void *))
+{
+	struct kmem_cache *ret = malloc(sizeof(*ret));
+
+	pthread_mutex_init(&ret->lock, NULL);
+	ret->size = size;
+	ret->nr_objs = 0;
+	ret->objs = NULL;
+	ret->ctor = ctor;
+	return ret;
+}

tools/testing/radix-tree/linux/bug.h

@@ -0,0 +1 @@
+#include "asm/bug.h"

tools/testing/radix-tree/linux/cpu.h

@@ -0,0 +1 @@
+#define cpuhp_setup_state_nocalls(a, b, c, d)	(0)

tools/testing/radix-tree/linux/gfp.h

@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _GFP_H
+#define _GFP_H
+
+#include <linux/types.h>
+
+#define __GFP_BITS_SHIFT 26
+#define __GFP_BITS_MASK ((gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
+
+#define __GFP_HIGH		0x20u
+#define __GFP_IO		0x40u
+#define __GFP_FS		0x80u
+#define __GFP_NOWARN		0x200u
+#define __GFP_ZERO		0x8000u
+#define __GFP_ATOMIC		0x80000u
+#define __GFP_ACCOUNT		0x100000u
+#define __GFP_DIRECT_RECLAIM	0x400000u
+#define __GFP_KSWAPD_RECLAIM	0x2000000u
+
+#define __GFP_RECLAIM	(__GFP_DIRECT_RECLAIM|__GFP_KSWAPD_RECLAIM)
+
+#define GFP_ZONEMASK	0x0fu
+#define GFP_ATOMIC	(__GFP_HIGH|__GFP_ATOMIC|__GFP_KSWAPD_RECLAIM)
+#define GFP_KERNEL	(__GFP_RECLAIM | __GFP_IO | __GFP_FS)
+#define GFP_NOWAIT	(__GFP_KSWAPD_RECLAIM)
+
+
+static inline bool gfpflags_allow_blocking(const gfp_t gfp_flags)
+{
+	return !!(gfp_flags & __GFP_DIRECT_RECLAIM);
+}
+
+#endif

tools/testing/radix-tree/linux/idr.h

@@ -0,0 +1 @@
+#include "../../../../include/linux/idr.h"

tools/testing/radix-tree/linux/init.h

@@ -0,0 +1 @@
+#define __init

tools/testing/radix-tree/linux/kernel.h

@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _KERNEL_H
+#define _KERNEL_H
+
+#include "../../include/linux/kernel.h"
+#include <string.h>
+#include <stdio.h>
+#include <limits.h>
+
+#include <linux/compiler.h>
+#include <linux/err.h>
+#include <linux/bitops.h>
+#include <linux/log2.h>
+#include "../../../include/linux/kconfig.h"
+
+#define printk printf
+#define pr_debug printk
+#define pr_cont printk
+
+#endif /* _KERNEL_H */

tools/testing/radix-tree/linux/kmemleak.h

@@ -0,0 +1 @@
+static inline void kmemleak_update_trace(const void *ptr) { }

tools/testing/radix-tree/linux/percpu.h

@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#define DECLARE_PER_CPU(type, val) extern type val
+#define DEFINE_PER_CPU(type, val) type val
+
+#define __get_cpu_var(var)	var
+#define this_cpu_ptr(var)	var
+#define this_cpu_read(var)	var
+#define this_cpu_xchg(var, val)		uatomic_xchg(&var, val)
+#define this_cpu_cmpxchg(var, old, new)	uatomic_cmpxchg(&var, old, new)
+#define per_cpu_ptr(ptr, cpu)   ({ (void)(cpu); (ptr); })
+#define per_cpu(var, cpu)	(*per_cpu_ptr(&(var), cpu))

tools/testing/radix-tree/linux/preempt.h

@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __LINUX_PREEMPT_H
+#define __LINUX_PREEMPT_H
+
+extern int preempt_count;
+
+#define preempt_disable()	uatomic_inc(&preempt_count)
+#define preempt_enable()	uatomic_dec(&preempt_count)
+
+static inline int in_interrupt(void)
+{
+	return 0;
+}
+
+#endif /* __LINUX_PREEMPT_H */

tools/testing/radix-tree/linux/radix-tree.h

@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _TEST_RADIX_TREE_H
+#define _TEST_RADIX_TREE_H
+
+#include "generated/map-shift.h"
+#include "../../../../include/linux/radix-tree.h"
+
+extern int kmalloc_verbose;
+extern int test_verbose;
+
+static inline void trace_call_rcu(struct rcu_head *head,
+		void (*func)(struct rcu_head *head))
+{
+	if (kmalloc_verbose)
+		printf("Delaying free of %p to slab\n", (char *)head -
+				offsetof(struct radix_tree_node, rcu_head));
+	call_rcu(head, func);
+}
+
+#define printv(verbosity_level, fmt, ...) \
+	if(test_verbose >= verbosity_level) \
+		printf(fmt, ##__VA_ARGS__)
+
+#undef call_rcu
+#define call_rcu(x, y) trace_call_rcu(x, y)
+
+#endif /* _TEST_RADIX_TREE_H */

tools/testing/radix-tree/linux/rcupdate.h

@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _RCUPDATE_H
+#define _RCUPDATE_H
+
+#include <urcu.h>
+
+#define rcu_dereference_raw(p) rcu_dereference(p)
+#define rcu_dereference_protected(p, cond) rcu_dereference(p)
+
+#endif

tools/testing/radix-tree/linux/slab.h

@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef SLAB_H
+#define SLAB_H
+
+#include <linux/types.h>
+#include <linux/gfp.h>
+
+#define SLAB_HWCACHE_ALIGN 1
+#define SLAB_PANIC 2
+#define SLAB_RECLAIM_ACCOUNT    0x00020000UL            /* Objects are reclaimable */
+
+void *kmalloc(size_t size, gfp_t);
+void kfree(void *);
+
+static inline void *kzalloc(size_t size, gfp_t gfp)
+{
+        return kmalloc(size, gfp | __GFP_ZERO);
+}
+
+void *kmem_cache_alloc(struct kmem_cache *cachep, int flags);
+void kmem_cache_free(struct kmem_cache *cachep, void *objp);
+
+struct kmem_cache *
+kmem_cache_create(const char *name, size_t size, size_t offset,
+	unsigned long flags, void (*ctor)(void *));
+
+#endif		/* SLAB_H */

tools/testing/radix-tree/linux/xarray.h

@@ -0,0 +1,2 @@
+#include "generated/map-shift.h"
+#include "../../../../include/linux/xarray.h"

tools/testing/radix-tree/main.c

@@ -0,0 +1,388 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <time.h>
+#include <assert.h>
+#include <limits.h>
+
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+
+#include "test.h"
+#include "regression.h"
+
+void __gang_check(unsigned long middle, long down, long up, int chunk, int hop)
+{
+	long idx;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	middle = 1 << 30;
+
+	for (idx = -down; idx < up; idx++)
+		item_insert(&tree, middle + idx);
+
+	item_check_absent(&tree, middle - down - 1);
+	for (idx = -down; idx < up; idx++)
+		item_check_present(&tree, middle + idx);
+	item_check_absent(&tree, middle + up);
+
+	if (chunk > 0) {
+		item_gang_check_present(&tree, middle - down, up + down,
+				chunk, hop);
+		item_full_scan(&tree, middle - down, down + up, chunk);
+	}
+	item_kill_tree(&tree);
+}
+
+void gang_check(void)
+{
+	__gang_check(1UL << 30, 128, 128, 35, 2);
+	__gang_check(1UL << 31, 128, 128, 32, 32);
+	__gang_check(1UL << 31, 128, 128, 32, 100);
+	__gang_check(1UL << 31, 128, 128, 17, 7);
+	__gang_check(0xffff0000UL, 0, 65536, 17, 7);
+	__gang_check(0xfffffffeUL, 1, 1, 17, 7);
+}
+
+void __big_gang_check(void)
+{
+	unsigned long start;
+	int wrapped = 0;
+
+	start = 0;
+	do {
+		unsigned long old_start;
+
+//		printf("0x%08lx\n", start);
+		__gang_check(start, rand() % 113 + 1, rand() % 71,
+				rand() % 157, rand() % 91 + 1);
+		old_start = start;
+		start += rand() % 1000000;
+		start %= 1ULL << 33;
+		if (start < old_start)
+			wrapped = 1;
+	} while (!wrapped);
+}
+
+void big_gang_check(bool long_run)
+{
+	int i;
+
+	for (i = 0; i < (long_run ? 1000 : 3); i++) {
+		__big_gang_check();
+		printv(2, "%d ", i);
+		fflush(stdout);
+	}
+}
+
+void add_and_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 44);
+	item_check_present(&tree, 44);
+	item_check_absent(&tree, 43);
+	item_kill_tree(&tree);
+}
+
+void dynamic_height_check(void)
+{
+	int i;
+	RADIX_TREE(tree, GFP_KERNEL);
+	tree_verify_min_height(&tree, 0);
+
+	item_insert(&tree, 42);
+	tree_verify_min_height(&tree, 42);
+
+	item_insert(&tree, 1000000);
+	tree_verify_min_height(&tree, 1000000);
+
+	assert(item_delete(&tree, 1000000));
+	tree_verify_min_height(&tree, 42);
+
+	assert(item_delete(&tree, 42));
+	tree_verify_min_height(&tree, 0);
+
+	for (i = 0; i < 1000; i++) {
+		item_insert(&tree, i);
+		tree_verify_min_height(&tree, i);
+	}
+
+	i--;
+	for (;;) {
+		assert(item_delete(&tree, i));
+		if (i == 0) {
+			tree_verify_min_height(&tree, 0);
+			break;
+		}
+		i--;
+		tree_verify_min_height(&tree, i);
+	}
+
+	item_kill_tree(&tree);
+}
+
+void check_copied_tags(struct radix_tree_root *tree, unsigned long start, unsigned long end, unsigned long *idx, int count, int fromtag, int totag)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+/*		if (i % 1000 == 0)
+			putchar('.'); */
+		if (idx[i] < start || idx[i] > end) {
+			if (item_tag_get(tree, idx[i], totag)) {
+				printv(2, "%lu-%lu: %lu, tags %d-%d\n", start,
+				       end, idx[i], item_tag_get(tree, idx[i],
+								 fromtag),
+				       item_tag_get(tree, idx[i], totag));
+			}
+			assert(!item_tag_get(tree, idx[i], totag));
+			continue;
+		}
+		if (item_tag_get(tree, idx[i], fromtag) ^
+			item_tag_get(tree, idx[i], totag)) {
+			printv(2, "%lu-%lu: %lu, tags %d-%d\n", start, end,
+			       idx[i], item_tag_get(tree, idx[i], fromtag),
+			       item_tag_get(tree, idx[i], totag));
+		}
+		assert(!(item_tag_get(tree, idx[i], fromtag) ^
+			 item_tag_get(tree, idx[i], totag)));
+	}
+}
+
+#define ITEMS 50000
+
+void copy_tag_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	unsigned long idx[ITEMS];
+	unsigned long start, end, count = 0, tagged, cur, tmp;
+	int i;
+
+//	printf("generating radix tree indices...\n");
+	start = rand();
+	end = rand();
+	if (start > end && (rand() % 10)) {
+		cur = start;
+		start = end;
+		end = cur;
+	}
+	/* Specifically create items around the start and the end of the range
+	 * with high probability to check for off by one errors */
+	cur = rand();
+	if (cur & 1) {
+		item_insert(&tree, start);
+		if (cur & 2) {
+			if (start <= end)
+				count++;
+			item_tag_set(&tree, start, 0);
+		}
+	}
+	if (cur & 4) {
+		item_insert(&tree, start-1);
+		if (cur & 8)
+			item_tag_set(&tree, start-1, 0);
+	}
+	if (cur & 16) {
+		item_insert(&tree, end);
+		if (cur & 32) {
+			if (start <= end)
+				count++;
+			item_tag_set(&tree, end, 0);
+		}
+	}
+	if (cur & 64) {
+		item_insert(&tree, end+1);
+		if (cur & 128)
+			item_tag_set(&tree, end+1, 0);
+	}
+
+	for (i = 0; i < ITEMS; i++) {
+		do {
+			idx[i] = rand();
+		} while (item_lookup(&tree, idx[i]));
+
+		item_insert(&tree, idx[i]);
+		if (rand() & 1) {
+			item_tag_set(&tree, idx[i], 0);
+			if (idx[i] >= start && idx[i] <= end)
+				count++;
+		}
+/*		if (i % 1000 == 0)
+			putchar('.'); */
+	}
+
+//	printf("\ncopying tags...\n");
+	tagged = tag_tagged_items(&tree, NULL, start, end, ITEMS, 0, 1);
+
+//	printf("checking copied tags\n");
+	assert(tagged == count);
+	check_copied_tags(&tree, start, end, idx, ITEMS, 0, 1);
+
+	/* Copy tags in several rounds */
+//	printf("\ncopying tags...\n");
+	tmp = rand() % (count / 10 + 2);
+	tagged = tag_tagged_items(&tree, NULL, start, end, tmp, 0, 2);
+	assert(tagged == count);
+
+//	printf("%lu %lu %lu\n", tagged, tmp, count);
+//	printf("checking copied tags\n");
+	check_copied_tags(&tree, start, end, idx, ITEMS, 0, 2);
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	verify_tag_consistency(&tree, 2);
+//	printf("\n");
+	item_kill_tree(&tree);
+}
+
+static void __locate_check(struct radix_tree_root *tree, unsigned long index,
+			unsigned order)
+{
+	struct item *item;
+	unsigned long index2;
+
+	item_insert_order(tree, index, order);
+	item = item_lookup(tree, index);
+	index2 = find_item(tree, item);
+	if (index != index2) {
+		printv(2, "index %ld order %d inserted; found %ld\n",
+			index, order, index2);
+		abort();
+	}
+}
+
+static void __order_0_locate_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	int i;
+
+	for (i = 0; i < 50; i++)
+		__locate_check(&tree, rand() % INT_MAX, 0);
+
+	item_kill_tree(&tree);
+}
+
+static void locate_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	unsigned order;
+	unsigned long offset, index;
+
+	__order_0_locate_check();
+
+	for (order = 0; order < 20; order++) {
+		for (offset = 0; offset < (1 << (order + 3));
+		     offset += (1UL << order)) {
+			for (index = 0; index < (1UL << (order + 5));
+			     index += (1UL << order)) {
+				__locate_check(&tree, index + offset, order);
+			}
+			if (find_item(&tree, &tree) != -1)
+				abort();
+
+			item_kill_tree(&tree);
+		}
+	}
+
+	if (find_item(&tree, &tree) != -1)
+		abort();
+	__locate_check(&tree, -1, 0);
+	if (find_item(&tree, &tree) != -1)
+		abort();
+	item_kill_tree(&tree);
+}
+
+static void single_thread_tests(bool long_run)
+{
+	int i;
+
+	printv(1, "starting single_thread_tests: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	multiorder_checks();
+	rcu_barrier();
+	printv(2, "after multiorder_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	locate_check();
+	rcu_barrier();
+	printv(2, "after locate_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	tag_check();
+	rcu_barrier();
+	printv(2, "after tag_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	gang_check();
+	rcu_barrier();
+	printv(2, "after gang_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	add_and_check();
+	rcu_barrier();
+	printv(2, "after add_and_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	dynamic_height_check();
+	rcu_barrier();
+	printv(2, "after dynamic_height_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	idr_checks();
+	ida_tests();
+	rcu_barrier();
+	printv(2, "after idr_checks: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	big_gang_check(long_run);
+	rcu_barrier();
+	printv(2, "after big_gang_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	for (i = 0; i < (long_run ? 2000 : 3); i++) {
+		copy_tag_check();
+		printv(2, "%d ", i);
+		fflush(stdout);
+	}
+	rcu_barrier();
+	printv(2, "after copy_tag_check: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+}
+
+int main(int argc, char **argv)
+{
+	bool long_run = false;
+	int opt;
+	unsigned int seed = time(NULL);
+
+	while ((opt = getopt(argc, argv, "ls:v")) != -1) {
+		if (opt == 'l')
+			long_run = true;
+		else if (opt == 's')
+			seed = strtoul(optarg, NULL, 0);
+		else if (opt == 'v')
+			test_verbose++;
+	}
+
+	printf("random seed %u\n", seed);
+	srand(seed);
+
+	printf("running tests\n");
+
+	rcu_register_thread();
+	radix_tree_init();
+
+	regression1_test();
+	regression2_test();
+	regression3_test();
+	iteration_test(0, 10 + 90 * long_run);
+	iteration_test(7, 10 + 90 * long_run);
+	single_thread_tests(long_run);
+
+	/* Free any remaining preallocated nodes */
+	radix_tree_cpu_dead(0);
+
+	benchmark();
+
+	rcu_barrier();
+	printv(2, "after rcu_barrier: %d allocated, preempt %d\n",
+		nr_allocated, preempt_count);
+	rcu_unregister_thread();
+
+	printf("tests completed\n");
+
+	exit(0);
+}

tools/testing/radix-tree/multiorder.c

@@ -0,0 +1,719 @@
+/*
+ * multiorder.c: Multi-order radix tree entry testing
+ * Copyright (c) 2016 Intel Corporation
+ * Author: Ross Zwisler <ross.zwisler@linux.intel.com>
+ * Author: Matthew Wilcox <matthew.r.wilcox@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ */
+#include <linux/radix-tree.h>
+#include <linux/slab.h>
+#include <linux/errno.h>
+#include <pthread.h>
+
+#include "test.h"
+
+#define for_each_index(i, base, order) \
+	for (i = base; i < base + (1 << order); i++)
+
+static void __multiorder_tag_test(int index, int order)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	int base, err, i;
+
+	/* our canonical entry */
+	base = index & ~((1 << order) - 1);
+
+	printv(2, "Multiorder tag test with index %d, canonical entry %d\n",
+			index, base);
+
+	err = item_insert_order(&tree, index, order);
+	assert(!err);
+
+	/*
+	 * Verify we get collisions for covered indices.  We try and fail to
+	 * insert an exceptional entry so we don't leak memory via
+	 * item_insert_order().
+	 */
+	for_each_index(i, base, order) {
+		err = __radix_tree_insert(&tree, i, order,
+				(void *)(0xA0 | RADIX_TREE_EXCEPTIONAL_ENTRY));
+		assert(err == -EEXIST);
+	}
+
+	for_each_index(i, base, order) {
+		assert(!radix_tree_tag_get(&tree, i, 0));
+		assert(!radix_tree_tag_get(&tree, i, 1));
+	}
+
+	assert(radix_tree_tag_set(&tree, index, 0));
+
+	for_each_index(i, base, order) {
+		assert(radix_tree_tag_get(&tree, i, 0));
+		assert(!radix_tree_tag_get(&tree, i, 1));
+	}
+
+	assert(tag_tagged_items(&tree, NULL, 0, ~0UL, 10, 0, 1) == 1);
+	assert(radix_tree_tag_clear(&tree, index, 0));
+
+	for_each_index(i, base, order) {
+		assert(!radix_tree_tag_get(&tree, i, 0));
+		assert(radix_tree_tag_get(&tree, i, 1));
+	}
+
+	assert(radix_tree_tag_clear(&tree, index, 1));
+
+	assert(!radix_tree_tagged(&tree, 0));
+	assert(!radix_tree_tagged(&tree, 1));
+
+	item_kill_tree(&tree);
+}
+
+static void __multiorder_tag_test2(unsigned order, unsigned long index2)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	unsigned long index = (1 << order);
+	index2 += index;
+
+	assert(item_insert_order(&tree, 0, order) == 0);
+	assert(item_insert(&tree, index2) == 0);
+
+	assert(radix_tree_tag_set(&tree, 0, 0));
+	assert(radix_tree_tag_set(&tree, index2, 0));
+
+	assert(tag_tagged_items(&tree, NULL, 0, ~0UL, 10, 0, 1) == 2);
+
+	item_kill_tree(&tree);
+}
+
+static void multiorder_tag_tests(void)
+{
+	int i, j;
+
+	/* test multi-order entry for indices 0-7 with no sibling pointers */
+	__multiorder_tag_test(0, 3);
+	__multiorder_tag_test(5, 3);
+
+	/* test multi-order entry for indices 8-15 with no sibling pointers */
+	__multiorder_tag_test(8, 3);
+	__multiorder_tag_test(15, 3);
+
+	/*
+	 * Our order 5 entry covers indices 0-31 in a tree with height=2.
+	 * This is broken up as follows:
+	 * 0-7:		canonical entry
+	 * 8-15:	sibling 1
+	 * 16-23:	sibling 2
+	 * 24-31:	sibling 3
+	 */
+	__multiorder_tag_test(0, 5);
+	__multiorder_tag_test(29, 5);
+
+	/* same test, but with indices 32-63 */
+	__multiorder_tag_test(32, 5);
+	__multiorder_tag_test(44, 5);
+
+	/*
+	 * Our order 8 entry covers indices 0-255 in a tree with height=3.
+	 * This is broken up as follows:
+	 * 0-63:	canonical entry
+	 * 64-127:	sibling 1
+	 * 128-191:	sibling 2
+	 * 192-255:	sibling 3
+	 */
+	__multiorder_tag_test(0, 8);
+	__multiorder_tag_test(190, 8);
+
+	/* same test, but with indices 256-511 */
+	__multiorder_tag_test(256, 8);
+	__multiorder_tag_test(300, 8);
+
+	__multiorder_tag_test(0x12345678UL, 8);
+
+	for (i = 1; i < 10; i++)
+		for (j = 0; j < (10 << i); j++)
+			__multiorder_tag_test2(i, j);
+}
+
+static void multiorder_check(unsigned long index, int order)
+{
+	unsigned long i;
+	unsigned long min = index & ~((1UL << order) - 1);
+	unsigned long max = min + (1UL << order);
+	void **slot;
+	struct item *item2 = item_create(min, order);
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	printv(2, "Multiorder index %ld, order %d\n", index, order);
+
+	assert(item_insert_order(&tree, index, order) == 0);
+
+	for (i = min; i < max; i++) {
+		struct item *item = item_lookup(&tree, i);
+		assert(item != 0);
+		assert(item->index == index);
+	}
+	for (i = 0; i < min; i++)
+		item_check_absent(&tree, i);
+	for (i = max; i < 2*max; i++)
+		item_check_absent(&tree, i);
+	for (i = min; i < max; i++)
+		assert(radix_tree_insert(&tree, i, item2) == -EEXIST);
+
+	slot = radix_tree_lookup_slot(&tree, index);
+	free(*slot);
+	radix_tree_replace_slot(&tree, slot, item2);
+	for (i = min; i < max; i++) {
+		struct item *item = item_lookup(&tree, i);
+		assert(item != 0);
+		assert(item->index == min);
+	}
+
+	assert(item_delete(&tree, min) != 0);
+
+	for (i = 0; i < 2*max; i++)
+		item_check_absent(&tree, i);
+}
+
+static void multiorder_shrink(unsigned long index, int order)
+{
+	unsigned long i;
+	unsigned long max = 1 << order;
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_node *node;
+
+	printv(2, "Multiorder shrink index %ld, order %d\n", index, order);
+
+	assert(item_insert_order(&tree, 0, order) == 0);
+
+	node = tree.rnode;
+
+	assert(item_insert(&tree, index) == 0);
+	assert(node != tree.rnode);
+
+	assert(item_delete(&tree, index) != 0);
+	assert(node == tree.rnode);
+
+	for (i = 0; i < max; i++) {
+		struct item *item = item_lookup(&tree, i);
+		assert(item != 0);
+		assert(item->index == 0);
+	}
+	for (i = max; i < 2*max; i++)
+		item_check_absent(&tree, i);
+
+	if (!item_delete(&tree, 0)) {
+		printv(2, "failed to delete index %ld (order %d)\n", index, order);
+		abort();
+	}
+
+	for (i = 0; i < 2*max; i++)
+		item_check_absent(&tree, i);
+}
+
+static void multiorder_insert_bug(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 0);
+	radix_tree_tag_set(&tree, 0, 0);
+	item_insert_order(&tree, 3 << 6, 6);
+
+	item_kill_tree(&tree);
+}
+
+void multiorder_iteration(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_iter iter;
+	void **slot;
+	int i, j, err;
+
+	printv(1, "Multiorder iteration test\n");
+
+#define NUM_ENTRIES 11
+	int index[NUM_ENTRIES] = {0, 2, 4, 8, 16, 32, 34, 36, 64, 72, 128};
+	int order[NUM_ENTRIES] = {1, 1, 2, 3,  4,  1,  0,  1,  3,  0, 7};
+
+	for (i = 0; i < NUM_ENTRIES; i++) {
+		err = item_insert_order(&tree, index[i], order[i]);
+		assert(!err);
+	}
+
+	for (j = 0; j < 256; j++) {
+		for (i = 0; i < NUM_ENTRIES; i++)
+			if (j <= (index[i] | ((1 << order[i]) - 1)))
+				break;
+
+		radix_tree_for_each_slot(slot, &tree, &iter, j) {
+			int height = order[i] / RADIX_TREE_MAP_SHIFT;
+			int shift = height * RADIX_TREE_MAP_SHIFT;
+			unsigned long mask = (1UL << order[i]) - 1;
+			struct item *item = *slot;
+
+			assert((iter.index | mask) == (index[i] | mask));
+			assert(iter.shift == shift);
+			assert(!radix_tree_is_internal_node(item));
+			assert((item->index | mask) == (index[i] | mask));
+			assert(item->order == order[i]);
+			i++;
+		}
+	}
+
+	item_kill_tree(&tree);
+}
+
+void multiorder_tagged_iteration(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_iter iter;
+	void **slot;
+	int i, j;
+
+	printv(1, "Multiorder tagged iteration test\n");
+
+#define MT_NUM_ENTRIES 9
+	int index[MT_NUM_ENTRIES] = {0, 2, 4, 16, 32, 40, 64, 72, 128};
+	int order[MT_NUM_ENTRIES] = {1, 0, 2, 4,  3,  1,  3,  0,   7};
+
+#define TAG_ENTRIES 7
+	int tag_index[TAG_ENTRIES] = {0, 4, 16, 40, 64, 72, 128};
+
+	for (i = 0; i < MT_NUM_ENTRIES; i++)
+		assert(!item_insert_order(&tree, index[i], order[i]));
+
+	assert(!radix_tree_tagged(&tree, 1));
+
+	for (i = 0; i < TAG_ENTRIES; i++)
+		assert(radix_tree_tag_set(&tree, tag_index[i], 1));
+
+	for (j = 0; j < 256; j++) {
+		int k;
+
+		for (i = 0; i < TAG_ENTRIES; i++) {
+			for (k = i; index[k] < tag_index[i]; k++)
+				;
+			if (j <= (index[k] | ((1 << order[k]) - 1)))
+				break;
+		}
+
+		radix_tree_for_each_tagged(slot, &tree, &iter, j, 1) {
+			unsigned long mask;
+			struct item *item = *slot;
+			for (k = i; index[k] < tag_index[i]; k++)
+				;
+			mask = (1UL << order[k]) - 1;
+
+			assert((iter.index | mask) == (tag_index[i] | mask));
+			assert(!radix_tree_is_internal_node(item));
+			assert((item->index | mask) == (tag_index[i] | mask));
+			assert(item->order == order[k]);
+			i++;
+		}
+	}
+
+	assert(tag_tagged_items(&tree, NULL, 0, ~0UL, TAG_ENTRIES, 1, 2) ==
+				TAG_ENTRIES);
+
+	for (j = 0; j < 256; j++) {
+		int mask, k;
+
+		for (i = 0; i < TAG_ENTRIES; i++) {
+			for (k = i; index[k] < tag_index[i]; k++)
+				;
+			if (j <= (index[k] | ((1 << order[k]) - 1)))
+				break;
+		}
+
+		radix_tree_for_each_tagged(slot, &tree, &iter, j, 2) {
+			struct item *item = *slot;
+			for (k = i; index[k] < tag_index[i]; k++)
+				;
+			mask = (1 << order[k]) - 1;
+
+			assert((iter.index | mask) == (tag_index[i] | mask));
+			assert(!radix_tree_is_internal_node(item));
+			assert((item->index | mask) == (tag_index[i] | mask));
+			assert(item->order == order[k]);
+			i++;
+		}
+	}
+
+	assert(tag_tagged_items(&tree, NULL, 1, ~0UL, MT_NUM_ENTRIES * 2, 1, 0)
+			== TAG_ENTRIES);
+	i = 0;
+	radix_tree_for_each_tagged(slot, &tree, &iter, 0, 0) {
+		assert(iter.index == tag_index[i]);
+		i++;
+	}
+
+	item_kill_tree(&tree);
+}
+
+/*
+ * Basic join checks: make sure we can't find an entry in the tree after
+ * a larger entry has replaced it
+ */
+static void multiorder_join1(unsigned long index,
+				unsigned order1, unsigned order2)
+{
+	unsigned long loc;
+	void *item, *item2 = item_create(index + 1, order1);
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert_order(&tree, index, order2);
+	item = radix_tree_lookup(&tree, index);
+	radix_tree_join(&tree, index + 1, order1, item2);
+	loc = find_item(&tree, item);
+	if (loc == -1)
+		free(item);
+	item = radix_tree_lookup(&tree, index + 1);
+	assert(item == item2);
+	item_kill_tree(&tree);
+}
+
+/*
+ * Check that the accounting of exceptional entries is handled correctly
+ * by joining an exceptional entry to a normal pointer.
+ */
+static void multiorder_join2(unsigned order1, unsigned order2)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_node *node;
+	void *item1 = item_create(0, order1);
+	void *item2;
+
+	item_insert_order(&tree, 0, order2);
+	radix_tree_insert(&tree, 1 << order2, (void *)0x12UL);
+	item2 = __radix_tree_lookup(&tree, 1 << order2, &node, NULL);
+	assert(item2 == (void *)0x12UL);
+	assert(node->exceptional == 1);
+
+	item2 = radix_tree_lookup(&tree, 0);
+	free(item2);
+
+	radix_tree_join(&tree, 0, order1, item1);
+	item2 = __radix_tree_lookup(&tree, 1 << order2, &node, NULL);
+	assert(item2 == item1);
+	assert(node->exceptional == 0);
+	item_kill_tree(&tree);
+}
+
+/*
+ * This test revealed an accounting bug for exceptional entries at one point.
+ * Nodes were being freed back into the pool with an elevated exception count
+ * by radix_tree_join() and then radix_tree_split() was failing to zero the
+ * count of exceptional entries.
+ */
+static void multiorder_join3(unsigned int order)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_node *node;
+	void **slot;
+	struct radix_tree_iter iter;
+	unsigned long i;
+
+	for (i = 0; i < (1 << order); i++) {
+		radix_tree_insert(&tree, i, (void *)0x12UL);
+	}
+
+	radix_tree_join(&tree, 0, order, (void *)0x16UL);
+	rcu_barrier();
+
+	radix_tree_split(&tree, 0, 0);
+
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		radix_tree_iter_replace(&tree, &iter, slot, (void *)0x12UL);
+	}
+
+	__radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(node->exceptional == node->count);
+
+	item_kill_tree(&tree);
+}
+
+static void multiorder_join(void)
+{
+	int i, j, idx;
+
+	for (idx = 0; idx < 1024; idx = idx * 2 + 3) {
+		for (i = 1; i < 15; i++) {
+			for (j = 0; j < i; j++) {
+				multiorder_join1(idx, i, j);
+			}
+		}
+	}
+
+	for (i = 1; i < 15; i++) {
+		for (j = 0; j < i; j++) {
+			multiorder_join2(i, j);
+		}
+	}
+
+	for (i = 3; i < 10; i++) {
+		multiorder_join3(i);
+	}
+}
+
+static void check_mem(unsigned old_order, unsigned new_order, unsigned alloc)
+{
+	struct radix_tree_preload *rtp = &radix_tree_preloads;
+	if (rtp->nr != 0)
+		printv(2, "split(%u %u) remaining %u\n", old_order, new_order,
+							rtp->nr);
+	/*
+	 * Can't check for equality here as some nodes may have been
+	 * RCU-freed while we ran.  But we should never finish with more
+	 * nodes allocated since they should have all been preloaded.
+	 */
+	if (nr_allocated > alloc)
+		printv(2, "split(%u %u) allocated %u %u\n", old_order, new_order,
+							alloc, nr_allocated);
+}
+
+static void __multiorder_split(int old_order, int new_order)
+{
+	RADIX_TREE(tree, GFP_ATOMIC);
+	void **slot;
+	struct radix_tree_iter iter;
+	unsigned alloc;
+	struct item *item;
+
+	radix_tree_preload(GFP_KERNEL);
+	assert(item_insert_order(&tree, 0, old_order) == 0);
+	radix_tree_preload_end();
+
+	/* Wipe out the preloaded cache or it'll confuse check_mem() */
+	radix_tree_cpu_dead(0);
+
+	item = radix_tree_tag_set(&tree, 0, 2);
+
+	radix_tree_split_preload(old_order, new_order, GFP_KERNEL);
+	alloc = nr_allocated;
+	radix_tree_split(&tree, 0, new_order);
+	check_mem(old_order, new_order, alloc);
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		radix_tree_iter_replace(&tree, &iter, slot,
+					item_create(iter.index, new_order));
+	}
+	radix_tree_preload_end();
+
+	item_kill_tree(&tree);
+	free(item);
+}
+
+static void __multiorder_split2(int old_order, int new_order)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	void **slot;
+	struct radix_tree_iter iter;
+	struct radix_tree_node *node;
+	void *item;
+
+	__radix_tree_insert(&tree, 0, old_order, (void *)0x12);
+
+	item = __radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(item == (void *)0x12);
+	assert(node->exceptional > 0);
+
+	radix_tree_split(&tree, 0, new_order);
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		radix_tree_iter_replace(&tree, &iter, slot,
+					item_create(iter.index, new_order));
+	}
+
+	item = __radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(item != (void *)0x12);
+	assert(node->exceptional == 0);
+
+	item_kill_tree(&tree);
+}
+
+static void __multiorder_split3(int old_order, int new_order)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	void **slot;
+	struct radix_tree_iter iter;
+	struct radix_tree_node *node;
+	void *item;
+
+	__radix_tree_insert(&tree, 0, old_order, (void *)0x12);
+
+	item = __radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(item == (void *)0x12);
+	assert(node->exceptional > 0);
+
+	radix_tree_split(&tree, 0, new_order);
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		radix_tree_iter_replace(&tree, &iter, slot, (void *)0x16);
+	}
+
+	item = __radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(item == (void *)0x16);
+	assert(node->exceptional > 0);
+
+	item_kill_tree(&tree);
+
+	__radix_tree_insert(&tree, 0, old_order, (void *)0x12);
+
+	item = __radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(item == (void *)0x12);
+	assert(node->exceptional > 0);
+
+	radix_tree_split(&tree, 0, new_order);
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		if (iter.index == (1 << new_order))
+			radix_tree_iter_replace(&tree, &iter, slot,
+						(void *)0x16);
+		else
+			radix_tree_iter_replace(&tree, &iter, slot, NULL);
+	}
+
+	item = __radix_tree_lookup(&tree, 1 << new_order, &node, NULL);
+	assert(item == (void *)0x16);
+	assert(node->count == node->exceptional);
+	do {
+		node = node->parent;
+		if (!node)
+			break;
+		assert(node->count == 1);
+		assert(node->exceptional == 0);
+	} while (1);
+
+	item_kill_tree(&tree);
+}
+
+static void multiorder_split(void)
+{
+	int i, j;
+
+	for (i = 3; i < 11; i++)
+		for (j = 0; j < i; j++) {
+			__multiorder_split(i, j);
+			__multiorder_split2(i, j);
+			__multiorder_split3(i, j);
+		}
+}
+
+static void multiorder_account(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	struct radix_tree_node *node;
+	void **slot;
+
+	item_insert_order(&tree, 0, 5);
+
+	__radix_tree_insert(&tree, 1 << 5, 5, (void *)0x12);
+	__radix_tree_lookup(&tree, 0, &node, NULL);
+	assert(node->count == node->exceptional * 2);
+	radix_tree_delete(&tree, 1 << 5);
+	assert(node->exceptional == 0);
+
+	__radix_tree_insert(&tree, 1 << 5, 5, (void *)0x12);
+	__radix_tree_lookup(&tree, 1 << 5, &node, &slot);
+	assert(node->count == node->exceptional * 2);
+	__radix_tree_replace(&tree, node, slot, NULL, NULL);
+	assert(node->exceptional == 0);
+
+	item_kill_tree(&tree);
+}
+
+bool stop_iteration = false;
+
+static void *creator_func(void *ptr)
+{
+	/* 'order' is set up to ensure we have sibling entries */
+	unsigned int order = RADIX_TREE_MAP_SHIFT - 1;
+	struct radix_tree_root *tree = ptr;
+	int i;
+
+	for (i = 0; i < 10000; i++) {
+		item_insert_order(tree, 0, order);
+		item_delete_rcu(tree, 0);
+	}
+
+	stop_iteration = true;
+	return NULL;
+}
+
+static void *iterator_func(void *ptr)
+{
+	struct radix_tree_root *tree = ptr;
+	struct radix_tree_iter iter;
+	struct item *item;
+	void **slot;
+
+	while (!stop_iteration) {
+		rcu_read_lock();
+		radix_tree_for_each_slot(slot, tree, &iter, 0) {
+			item = radix_tree_deref_slot(slot);
+
+			if (!item)
+				continue;
+			if (radix_tree_deref_retry(item)) {
+				slot = radix_tree_iter_retry(&iter);
+				continue;
+			}
+
+			item_sanity(item, iter.index);
+		}
+		rcu_read_unlock();
+	}
+	return NULL;
+}
+
+static void multiorder_iteration_race(void)
+{
+	const int num_threads = sysconf(_SC_NPROCESSORS_ONLN);
+	pthread_t worker_thread[num_threads];
+	RADIX_TREE(tree, GFP_KERNEL);
+	int i;
+
+	pthread_create(&worker_thread[0], NULL, &creator_func, &tree);
+	for (i = 1; i < num_threads; i++)
+		pthread_create(&worker_thread[i], NULL, &iterator_func, &tree);
+
+	for (i = 0; i < num_threads; i++)
+		pthread_join(worker_thread[i], NULL);
+
+	item_kill_tree(&tree);
+}
+
+void multiorder_checks(void)
+{
+	int i;
+
+	for (i = 0; i < 20; i++) {
+		multiorder_check(200, i);
+		multiorder_check(0, i);
+		multiorder_check((1UL << i) + 1, i);
+	}
+
+	for (i = 0; i < 15; i++)
+		multiorder_shrink((1UL << (i + RADIX_TREE_MAP_SHIFT)), i);
+
+	multiorder_insert_bug();
+	multiorder_tag_tests();
+	multiorder_iteration();
+	multiorder_tagged_iteration();
+	multiorder_join();
+	multiorder_split();
+	multiorder_account();
+	multiorder_iteration_race();
+
+	radix_tree_cpu_dead(0);
+}
+
+int __weak main(void)
+{
+	radix_tree_init();
+	multiorder_checks();
+	return 0;
+}

tools/testing/radix-tree/regression.h

@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __REGRESSION_H__
+#define __REGRESSION_H__
+
+void regression1_test(void);
+void regression2_test(void);
+void regression3_test(void);
+
+#endif

tools/testing/radix-tree/regression1.c

@@ -0,0 +1,221 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Regression1
+ * Description:
+ * Salman Qazi describes the following radix-tree bug:
+ *
+ * In the following case, we get can get a deadlock:
+ *
+ * 0.  The radix tree contains two items, one has the index 0.
+ * 1.  The reader (in this case find_get_pages) takes the rcu_read_lock.
+ * 2.  The reader acquires slot(s) for item(s) including the index 0 item.
+ * 3.  The non-zero index item is deleted, and as a consequence the other item
+ *     is moved to the root of the tree. The place where it used to be is queued
+ *     for deletion after the readers finish.
+ * 3b. The zero item is deleted, removing it from the direct slot, it remains in
+ *     the rcu-delayed indirect node.
+ * 4.  The reader looks at the index 0 slot, and finds that the page has 0 ref
+ *     count
+ * 5.  The reader looks at it again, hoping that the item will either be freed
+ *     or the ref count will increase. This never happens, as the slot it is
+ *     looking at will never be updated. Also, this slot can never be reclaimed
+ *     because the reader is holding rcu_read_lock and is in an infinite loop.
+ *
+ * The fix is to re-use the same "indirect" pointer case that requires a slot
+ * lookup retry into a general "retry the lookup" bit.
+ *
+ * Running:
+ * This test should run to completion in a few seconds. The above bug would
+ * cause it to hang indefinitely.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <linux/rcupdate.h>
+#include <stdlib.h>
+#include <pthread.h>
+#include <stdio.h>
+#include <assert.h>
+
+#include "regression.h"
+
+static RADIX_TREE(mt_tree, GFP_KERNEL);
+static pthread_mutex_t mt_lock = PTHREAD_MUTEX_INITIALIZER;
+
+struct page {
+	pthread_mutex_t lock;
+	struct rcu_head rcu;
+	int count;
+	unsigned long index;
+};
+
+static struct page *page_alloc(void)
+{
+	struct page *p;
+	p = malloc(sizeof(struct page));
+	p->count = 1;
+	p->index = 1;
+	pthread_mutex_init(&p->lock, NULL);
+
+	return p;
+}
+
+static void page_rcu_free(struct rcu_head *rcu)
+{
+	struct page *p = container_of(rcu, struct page, rcu);
+	assert(!p->count);
+	pthread_mutex_destroy(&p->lock);
+	free(p);
+}
+
+static void page_free(struct page *p)
+{
+	call_rcu(&p->rcu, page_rcu_free);
+}
+
+static unsigned find_get_pages(unsigned long start,
+			    unsigned int nr_pages, struct page **pages)
+{
+	unsigned int i;
+	unsigned int ret;
+	unsigned int nr_found;
+
+	rcu_read_lock();
+restart:
+	nr_found = radix_tree_gang_lookup_slot(&mt_tree,
+				(void ***)pages, NULL, start, nr_pages);
+	ret = 0;
+	for (i = 0; i < nr_found; i++) {
+		struct page *page;
+repeat:
+		page = radix_tree_deref_slot((void **)pages[i]);
+		if (unlikely(!page))
+			continue;
+
+		if (radix_tree_exception(page)) {
+			if (radix_tree_deref_retry(page)) {
+				/*
+				 * Transient condition which can only trigger
+				 * when entry at index 0 moves out of or back
+				 * to root: none yet gotten, safe to restart.
+				 */
+				assert((start | i) == 0);
+				goto restart;
+			}
+			/*
+			 * No exceptional entries are inserted in this test.
+			 */
+			assert(0);
+		}
+
+		pthread_mutex_lock(&page->lock);
+		if (!page->count) {
+			pthread_mutex_unlock(&page->lock);
+			goto repeat;
+		}
+		/* don't actually update page refcount */
+		pthread_mutex_unlock(&page->lock);
+
+		/* Has the page moved? */
+		if (unlikely(page != *((void **)pages[i]))) {
+			goto repeat;
+		}
+
+		pages[ret] = page;
+		ret++;
+	}
+	rcu_read_unlock();
+	return ret;
+}
+
+static pthread_barrier_t worker_barrier;
+
+static void *regression1_fn(void *arg)
+{
+	rcu_register_thread();
+
+	if (pthread_barrier_wait(&worker_barrier) ==
+			PTHREAD_BARRIER_SERIAL_THREAD) {
+		int j;
+
+		for (j = 0; j < 1000000; j++) {
+			struct page *p;
+
+			p = page_alloc();
+			pthread_mutex_lock(&mt_lock);
+			radix_tree_insert(&mt_tree, 0, p);
+			pthread_mutex_unlock(&mt_lock);
+
+			p = page_alloc();
+			pthread_mutex_lock(&mt_lock);
+			radix_tree_insert(&mt_tree, 1, p);
+			pthread_mutex_unlock(&mt_lock);
+
+			pthread_mutex_lock(&mt_lock);
+			p = radix_tree_delete(&mt_tree, 1);
+			pthread_mutex_lock(&p->lock);
+			p->count--;
+			pthread_mutex_unlock(&p->lock);
+			pthread_mutex_unlock(&mt_lock);
+			page_free(p);
+
+			pthread_mutex_lock(&mt_lock);
+			p = radix_tree_delete(&mt_tree, 0);
+			pthread_mutex_lock(&p->lock);
+			p->count--;
+			pthread_mutex_unlock(&p->lock);
+			pthread_mutex_unlock(&mt_lock);
+			page_free(p);
+		}
+	} else {
+		int j;
+
+		for (j = 0; j < 100000000; j++) {
+			struct page *pages[10];
+
+			find_get_pages(0, 10, pages);
+		}
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+static pthread_t *threads;
+void regression1_test(void)
+{
+	int nr_threads;
+	int i;
+	long arg;
+
+	/* Regression #1 */
+	printv(1, "running regression test 1, should finish in under a minute\n");
+	nr_threads = 2;
+	pthread_barrier_init(&worker_barrier, NULL, nr_threads);
+
+	threads = malloc(nr_threads * sizeof(pthread_t *));
+
+	for (i = 0; i < nr_threads; i++) {
+		arg = i;
+		if (pthread_create(&threads[i], NULL, regression1_fn, (void *)arg)) {
+			perror("pthread_create");
+			exit(1);
+		}
+	}
+
+	for (i = 0; i < nr_threads; i++) {
+		if (pthread_join(threads[i], NULL)) {
+			perror("pthread_join");
+			exit(1);
+		}
+	}
+
+	free(threads);
+
+	printv(1, "regression test 1, done\n");
+}

tools/testing/radix-tree/regression2.c

@@ -0,0 +1,123 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Regression2
+ * Description:
+ * Toshiyuki Okajima describes the following radix-tree bug:
+ *
+ * In the following case, we can get a hangup on
+ *   radix_radix_tree_gang_lookup_tag_slot.
+ *
+ * 0.  The radix tree contains RADIX_TREE_MAP_SIZE items. And the tag of
+ *     a certain item has PAGECACHE_TAG_DIRTY.
+ * 1.  radix_tree_range_tag_if_tagged(, start, end, , PAGECACHE_TAG_DIRTY,
+ *     PAGECACHE_TAG_TOWRITE) is called to add PAGECACHE_TAG_TOWRITE tag
+ *     for the tag which has PAGECACHE_TAG_DIRTY. However, there is no tag with
+ *     PAGECACHE_TAG_DIRTY within the range from start to end. As the result,
+ *     There is no tag with PAGECACHE_TAG_TOWRITE but the root tag has
+ *     PAGECACHE_TAG_TOWRITE.
+ * 2.  An item is added into the radix tree and then the level of it is
+ *     extended into 2 from 1. At that time, the new radix tree node succeeds
+ *     the tag status of the root tag. Therefore the tag of the new radix tree
+ *     node has PAGECACHE_TAG_TOWRITE but there is not slot with
+ *     PAGECACHE_TAG_TOWRITE tag in the child node of the new radix tree node.
+ * 3.  The tag of a certain item is cleared with PAGECACHE_TAG_DIRTY.
+ * 4.  All items within the index range from 0 to RADIX_TREE_MAP_SIZE - 1 are
+ *     released. (Only the item which index is RADIX_TREE_MAP_SIZE exist in the
+ *     radix tree.) As the result, the slot of the radix tree node is NULL but
+ *     the tag which corresponds to the slot has PAGECACHE_TAG_TOWRITE.
+ * 5.  radix_tree_gang_lookup_tag_slot(PAGECACHE_TAG_TOWRITE) calls
+ *     __lookup_tag. __lookup_tag returns with 0. And __lookup_tag doesn't
+ *     change the index that is the input and output parameter. Because the 1st
+ *     slot of the radix tree node is NULL, but the tag which corresponds to
+ *     the slot has PAGECACHE_TAG_TOWRITE.
+ *     Therefore radix_tree_gang_lookup_tag_slot tries to get some items by
+ *     calling __lookup_tag, but it cannot get any items forever.
+ *
+ * The fix is to change that radix_tree_tag_if_tagged doesn't tag the root tag
+ * if it doesn't set any tags within the specified range.
+ *
+ * Running:
+ * This test should run to completion immediately. The above bug would cause it
+ * to hang indefinitely.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "regression.h"
+#include "test.h"
+
+#define PAGECACHE_TAG_DIRTY     0
+#define PAGECACHE_TAG_WRITEBACK 1
+#define PAGECACHE_TAG_TOWRITE   2
+
+static RADIX_TREE(mt_tree, GFP_KERNEL);
+unsigned long page_count = 0;
+
+struct page {
+	unsigned long index;
+};
+
+static struct page *page_alloc(void)
+{
+	struct page *p;
+	p = malloc(sizeof(struct page));
+	p->index = page_count++;
+
+	return p;
+}
+
+void regression2_test(void)
+{
+	int i;
+	struct page *p;
+	int max_slots = RADIX_TREE_MAP_SIZE;
+	unsigned long int start, end;
+	struct page *pages[1];
+
+	printv(1, "running regression test 2 (should take milliseconds)\n");
+	/* 0. */
+	for (i = 0; i <= max_slots - 1; i++) {
+		p = page_alloc();
+		radix_tree_insert(&mt_tree, i, p);
+	}
+	radix_tree_tag_set(&mt_tree, max_slots - 1, PAGECACHE_TAG_DIRTY);
+
+	/* 1. */
+	start = 0;
+	end = max_slots - 2;
+	tag_tagged_items(&mt_tree, NULL, start, end, 1,
+				PAGECACHE_TAG_DIRTY, PAGECACHE_TAG_TOWRITE);
+
+	/* 2. */
+	p = page_alloc();
+	radix_tree_insert(&mt_tree, max_slots, p);
+
+	/* 3. */
+	radix_tree_tag_clear(&mt_tree, max_slots - 1, PAGECACHE_TAG_DIRTY);
+
+	/* 4. */
+	for (i = max_slots - 1; i >= 0; i--)
+		free(radix_tree_delete(&mt_tree, i));
+
+	/* 5. */
+	// NOTE: start should not be 0 because radix_tree_gang_lookup_tag_slot
+	//       can return.
+	start = 1;
+	end = max_slots - 2;
+	radix_tree_gang_lookup_tag_slot(&mt_tree, (void ***)pages, start, end,
+		PAGECACHE_TAG_TOWRITE);
+
+	/* We remove all the remained nodes */
+	free(radix_tree_delete(&mt_tree, max_slots));
+
+	BUG_ON(!radix_tree_empty(&mt_tree));
+
+	printv(1, "regression test 2, done\n");
+}

tools/testing/radix-tree/regression3.c

@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Regression3
+ * Description:
+ * Helper radix_tree_iter_retry resets next_index to the current index.
+ * In following radix_tree_next_slot current chunk size becomes zero.
+ * This isn't checked and it tries to dereference null pointer in slot.
+ *
+ * Helper radix_tree_iter_resume reset slot to NULL and next_index to index + 1,
+ * for tagger iteraction it also must reset cached tags in iterator to abort
+ * next radix_tree_next_slot and go to slow-path into radix_tree_next_chunk.
+ *
+ * Running:
+ * This test should run to completion immediately. The above bug would
+ * cause it to segfault.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "regression.h"
+
+void regression3_test(void)
+{
+	RADIX_TREE(root, GFP_KERNEL);
+	void *ptr0 = (void *)4ul;
+	void *ptr = (void *)8ul;
+	struct radix_tree_iter iter;
+	void **slot;
+	bool first;
+
+	printv(1, "running regression test 3 (should take milliseconds)\n");
+
+	radix_tree_insert(&root, 0, ptr0);
+	radix_tree_tag_set(&root, 0, 0);
+
+	first = true;
+	radix_tree_for_each_tagged(slot, &root, &iter, 0, 0) {
+		printv(2, "tagged %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			radix_tree_tag_set(&root, 1, 0);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printv(2, "retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+	radix_tree_delete(&root, 1);
+
+	first = true;
+	radix_tree_for_each_slot(slot, &root, &iter, 0) {
+		printv(2, "slot %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printv(2, "retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+	radix_tree_delete(&root, 1);
+
+	first = true;
+	radix_tree_for_each_contig(slot, &root, &iter, 0) {
+		printv(2, "contig %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printv(2, "retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+
+	radix_tree_for_each_slot(slot, &root, &iter, 0) {
+		printv(2, "slot %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printv(2, "next at %ld\n", iter.index);
+			slot = radix_tree_iter_resume(slot, &iter);
+		}
+	}
+
+	radix_tree_for_each_contig(slot, &root, &iter, 0) {
+		printv(2, "contig %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printv(2, "next at %ld\n", iter.index);
+			slot = radix_tree_iter_resume(slot, &iter);
+		}
+	}
+
+	radix_tree_tag_set(&root, 0, 0);
+	radix_tree_tag_set(&root, 1, 0);
+	radix_tree_for_each_tagged(slot, &root, &iter, 0, 0) {
+		printv(2, "tagged %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printv(2, "next at %ld\n", iter.index);
+			slot = radix_tree_iter_resume(slot, &iter);
+		}
+	}
+
+	radix_tree_delete(&root, 0);
+	radix_tree_delete(&root, 1);
+
+	printv(1, "regression test 3 passed\n");
+}

tools/testing/radix-tree/tag_check.c

@@ -0,0 +1,380 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdlib.h>
+#include <assert.h>
+#include <stdio.h>
+#include <string.h>
+
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+
+#include "test.h"
+
+
+static void
+__simple_checks(struct radix_tree_root *tree, unsigned long index, int tag)
+{
+	unsigned long first = 0;
+	int ret;
+
+	item_check_absent(tree, index);
+	assert(item_tag_get(tree, index, tag) == 0);
+
+	item_insert(tree, index);
+	assert(item_tag_get(tree, index, tag) == 0);
+	item_tag_set(tree, index, tag);
+	ret = item_tag_get(tree, index, tag);
+	assert(ret != 0);
+	ret = tag_tagged_items(tree, NULL, first, ~0UL, 10, tag, !tag);
+	assert(ret == 1);
+	ret = item_tag_get(tree, index, !tag);
+	assert(ret != 0);
+	ret = item_delete(tree, index);
+	assert(ret != 0);
+	item_insert(tree, index);
+	ret = item_tag_get(tree, index, tag);
+	assert(ret == 0);
+	ret = item_delete(tree, index);
+	assert(ret != 0);
+	ret = item_delete(tree, index);
+	assert(ret == 0);
+}
+
+void simple_checks(void)
+{
+	unsigned long index;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	for (index = 0; index < 10000; index++) {
+		__simple_checks(&tree, index, 0);
+		__simple_checks(&tree, index, 1);
+	}
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	printv(2, "before item_kill_tree: %d allocated\n", nr_allocated);
+	item_kill_tree(&tree);
+	rcu_barrier();
+	printv(2, "after item_kill_tree: %d allocated\n", nr_allocated);
+}
+
+/*
+ * Check that tags propagate correctly when extending a tree.
+ */
+static void extend_checks(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 43);
+	assert(item_tag_get(&tree, 43, 0) == 0);
+	item_tag_set(&tree, 43, 0);
+	assert(item_tag_get(&tree, 43, 0) == 1);
+	item_insert(&tree, 1000000);
+	assert(item_tag_get(&tree, 43, 0) == 1);
+
+	item_insert(&tree, 0);
+	item_tag_set(&tree, 0, 0);
+	item_delete(&tree, 1000000);
+	assert(item_tag_get(&tree, 43, 0) != 0);
+	item_delete(&tree, 43);
+	assert(item_tag_get(&tree, 43, 0) == 0);	/* crash */
+	assert(item_tag_get(&tree, 0, 0) == 1);
+
+	verify_tag_consistency(&tree, 0);
+
+	item_kill_tree(&tree);
+}
+
+/*
+ * Check that tags propagate correctly when contracting a tree.
+ */
+static void contract_checks(void)
+{
+	struct item *item;
+	int tmp;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	tmp = 1<<RADIX_TREE_MAP_SHIFT;
+	item_insert(&tree, tmp);
+	item_insert(&tree, tmp+1);
+	item_tag_set(&tree, tmp, 0);
+	item_tag_set(&tree, tmp, 1);
+	item_tag_set(&tree, tmp+1, 0);
+	item_delete(&tree, tmp+1);
+	item_tag_clear(&tree, tmp, 1);
+
+	assert(radix_tree_gang_lookup_tag(&tree, (void **)&item, 0, 1, 0) == 1);
+	assert(radix_tree_gang_lookup_tag(&tree, (void **)&item, 0, 1, 1) == 0);
+
+	assert(item_tag_get(&tree, tmp, 0) == 1);
+	assert(item_tag_get(&tree, tmp, 1) == 0);
+
+	verify_tag_consistency(&tree, 0);
+	item_kill_tree(&tree);
+}
+
+/*
+ * Stupid tag thrasher
+ *
+ * Create a large linear array corresponding to the tree.   Each element in
+ * the array is coherent with each node in the tree
+ */
+
+enum {
+	NODE_ABSENT = 0,
+	NODE_PRESENT = 1,
+	NODE_TAGGED = 2,
+};
+
+#define THRASH_SIZE		(1000 * 1000)
+#define N 127
+#define BATCH	33
+
+static void gang_check(struct radix_tree_root *tree,
+			char *thrash_state, int tag)
+{
+	struct item *items[BATCH];
+	int nr_found;
+	unsigned long index = 0;
+	unsigned long last_index = 0;
+
+	while ((nr_found = radix_tree_gang_lookup_tag(tree, (void **)items,
+					index, BATCH, tag))) {
+		int i;
+
+		for (i = 0; i < nr_found; i++) {
+			struct item *item = items[i];
+
+			while (last_index < item->index) {
+				assert(thrash_state[last_index] != NODE_TAGGED);
+				last_index++;
+			}
+			assert(thrash_state[last_index] == NODE_TAGGED);
+			last_index++;
+		}
+		index = items[nr_found - 1]->index + 1;
+	}
+}
+
+static void do_thrash(struct radix_tree_root *tree, char *thrash_state, int tag)
+{
+	int insert_chunk;
+	int delete_chunk;
+	int tag_chunk;
+	int untag_chunk;
+	int total_tagged = 0;
+	int total_present = 0;
+
+	for (insert_chunk = 1; insert_chunk < THRASH_SIZE; insert_chunk *= N)
+	for (delete_chunk = 1; delete_chunk < THRASH_SIZE; delete_chunk *= N)
+	for (tag_chunk = 1; tag_chunk < THRASH_SIZE; tag_chunk *= N)
+	for (untag_chunk = 1; untag_chunk < THRASH_SIZE; untag_chunk *= N) {
+		int i;
+		unsigned long index;
+		int nr_inserted = 0;
+		int nr_deleted = 0;
+		int nr_tagged = 0;
+		int nr_untagged = 0;
+		int actual_total_tagged;
+		int actual_total_present;
+
+		for (i = 0; i < insert_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_ABSENT)
+				continue;
+			item_check_absent(tree, index);
+			item_insert(tree, index);
+			assert(thrash_state[index] != NODE_PRESENT);
+			thrash_state[index] = NODE_PRESENT;
+			nr_inserted++;
+			total_present++;
+		}
+
+		for (i = 0; i < delete_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] == NODE_ABSENT)
+				continue;
+			item_check_present(tree, index);
+			if (item_tag_get(tree, index, tag)) {
+				assert(thrash_state[index] == NODE_TAGGED);
+				total_tagged--;
+			} else {
+				assert(thrash_state[index] == NODE_PRESENT);
+			}
+			item_delete(tree, index);
+			assert(thrash_state[index] != NODE_ABSENT);
+			thrash_state[index] = NODE_ABSENT;
+			nr_deleted++;
+			total_present--;
+		}
+
+		for (i = 0; i < tag_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_PRESENT) {
+				if (item_lookup(tree, index))
+					assert(item_tag_get(tree, index, tag));
+				continue;
+			}
+			item_tag_set(tree, index, tag);
+			item_tag_set(tree, index, tag);
+			assert(thrash_state[index] != NODE_TAGGED);
+			thrash_state[index] = NODE_TAGGED;
+			nr_tagged++;
+			total_tagged++;
+		}
+
+		for (i = 0; i < untag_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_TAGGED)
+				continue;
+			item_check_present(tree, index);
+			assert(item_tag_get(tree, index, tag));
+			item_tag_clear(tree, index, tag);
+			item_tag_clear(tree, index, tag);
+			assert(thrash_state[index] != NODE_PRESENT);
+			thrash_state[index] = NODE_PRESENT;
+			nr_untagged++;
+			total_tagged--;
+		}
+
+		actual_total_tagged = 0;
+		actual_total_present = 0;
+		for (index = 0; index < THRASH_SIZE; index++) {
+			switch (thrash_state[index]) {
+			case NODE_ABSENT:
+				item_check_absent(tree, index);
+				break;
+			case NODE_PRESENT:
+				item_check_present(tree, index);
+				assert(!item_tag_get(tree, index, tag));
+				actual_total_present++;
+				break;
+			case NODE_TAGGED:
+				item_check_present(tree, index);
+				assert(item_tag_get(tree, index, tag));
+				actual_total_present++;
+				actual_total_tagged++;
+				break;
+			}
+		}
+
+		gang_check(tree, thrash_state, tag);
+
+		printv(2, "%d(%d) %d(%d) %d(%d) %d(%d) / "
+				"%d(%d) present, %d(%d) tagged\n",
+			insert_chunk, nr_inserted,
+			delete_chunk, nr_deleted,
+			tag_chunk, nr_tagged,
+			untag_chunk, nr_untagged,
+			total_present, actual_total_present,
+			total_tagged, actual_total_tagged);
+	}
+}
+
+static void thrash_tags(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	char *thrash_state;
+
+	thrash_state = malloc(THRASH_SIZE);
+	memset(thrash_state, 0, THRASH_SIZE);
+
+	do_thrash(&tree, thrash_state, 0);
+
+	verify_tag_consistency(&tree, 0);
+	item_kill_tree(&tree);
+	free(thrash_state);
+}
+
+static void leak_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 1000000);
+	item_delete(&tree, 1000000);
+	item_kill_tree(&tree);
+}
+
+static void __leak_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	printv(2, "%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_insert(&tree, 1000000);
+	printv(2, "%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_delete(&tree, 1000000);
+	printv(2, "%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_kill_tree(&tree);
+	printv(2, "%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+}
+
+static void single_check(void)
+{
+	struct item *items[BATCH];
+	RADIX_TREE(tree, GFP_KERNEL);
+	int ret;
+	unsigned long first = 0;
+
+	item_insert(&tree, 0);
+	item_tag_set(&tree, 0, 0);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 0, BATCH, 0);
+	assert(ret == 1);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 1, BATCH, 0);
+	assert(ret == 0);
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	ret = tag_tagged_items(&tree, NULL, first, 10, 10, 0, 1);
+	assert(ret == 1);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 0, BATCH, 1);
+	assert(ret == 1);
+	item_tag_clear(&tree, 0, 0);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 0, BATCH, 0);
+	assert(ret == 0);
+	item_kill_tree(&tree);
+}
+
+void radix_tree_clear_tags_test(void)
+{
+	unsigned long index;
+	struct radix_tree_node *node;
+	struct radix_tree_iter iter;
+	void **slot;
+
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 0);
+	item_tag_set(&tree, 0, 0);
+	__radix_tree_lookup(&tree, 0, &node, &slot);
+	radix_tree_clear_tags(&tree, node, slot);
+	assert(item_tag_get(&tree, 0, 0) == 0);
+
+	for (index = 0; index < 1000; index++) {
+		item_insert(&tree, index);
+		item_tag_set(&tree, index, 0);
+	}
+
+	radix_tree_for_each_slot(slot, &tree, &iter, 0) {
+		radix_tree_clear_tags(&tree, iter.node, slot);
+		assert(item_tag_get(&tree, iter.index, 0) == 0);
+	}
+
+	item_kill_tree(&tree);
+}
+
+void tag_check(void)
+{
+	single_check();
+	extend_checks();
+	contract_checks();
+	rcu_barrier();
+	printv(2, "after extend_checks: %d allocated\n", nr_allocated);
+	__leak_check();
+	leak_check();
+	rcu_barrier();
+	printv(2, "after leak_check: %d allocated\n", nr_allocated);
+	simple_checks();
+	rcu_barrier();
+	printv(2, "after simple_checks: %d allocated\n", nr_allocated);
+	thrash_tags();
+	rcu_barrier();
+	printv(2, "after thrash_tags: %d allocated\n", nr_allocated);
+	radix_tree_clear_tags_test();
+}

tools/testing/radix-tree/test.c

@@ -0,0 +1,334 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdlib.h>
+#include <assert.h>
+#include <stdio.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+
+#include "test.h"
+
+struct item *
+item_tag_set(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_set(root, index, tag);
+}
+
+struct item *
+item_tag_clear(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_clear(root, index, tag);
+}
+
+int item_tag_get(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_get(root, index, tag);
+}
+
+int __item_insert(struct radix_tree_root *root, struct item *item)
+{
+	return __radix_tree_insert(root, item->index, item->order, item);
+}
+
+struct item *item_create(unsigned long index, unsigned int order)
+{
+	struct item *ret = malloc(sizeof(*ret));
+
+	ret->index = index;
+	ret->order = order;
+	return ret;
+}
+
+int item_insert_order(struct radix_tree_root *root, unsigned long index,
+			unsigned order)
+{
+	struct item *item = item_create(index, order);
+	int err = __item_insert(root, item);
+	if (err)
+		free(item);
+	return err;
+}
+
+int item_insert(struct radix_tree_root *root, unsigned long index)
+{
+	return item_insert_order(root, index, 0);
+}
+
+void item_sanity(struct item *item, unsigned long index)
+{
+	unsigned long mask;
+	assert(!radix_tree_is_internal_node(item));
+	assert(item->order < BITS_PER_LONG);
+	mask = (1UL << item->order) - 1;
+	assert((item->index | mask) == (index | mask));
+}
+
+int item_delete(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item = radix_tree_delete(root, index);
+
+	if (item) {
+		item_sanity(item, index);
+		free(item);
+		return 1;
+	}
+	return 0;
+}
+
+static void item_free_rcu(struct rcu_head *head)
+{
+	struct item *item = container_of(head, struct item, rcu_head);
+
+	free(item);
+}
+
+int item_delete_rcu(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item = radix_tree_delete(root, index);
+
+	if (item) {
+		item_sanity(item, index);
+		call_rcu(&item->rcu_head, item_free_rcu);
+		return 1;
+	}
+	return 0;
+}
+
+void item_check_present(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item;
+
+	item = radix_tree_lookup(root, index);
+	assert(item != NULL);
+	item_sanity(item, index);
+}
+
+struct item *item_lookup(struct radix_tree_root *root, unsigned long index)
+{
+	return radix_tree_lookup(root, index);
+}
+
+void item_check_absent(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item;
+
+	item = radix_tree_lookup(root, index);
+	assert(item == NULL);
+}
+
+/*
+ * Scan only the passed (start, start+nr] for present items
+ */
+void item_gang_check_present(struct radix_tree_root *root,
+			unsigned long start, unsigned long nr,
+			int chunk, int hop)
+{
+	struct item *items[chunk];
+	unsigned long into;
+
+	for (into = 0; into < nr; ) {
+		int nfound;
+		int nr_to_find = chunk;
+		int i;
+
+		if (nr_to_find > (nr - into))
+			nr_to_find = nr - into;
+
+		nfound = radix_tree_gang_lookup(root, (void **)items,
+						start + into, nr_to_find);
+		assert(nfound == nr_to_find);
+		for (i = 0; i < nfound; i++)
+			assert(items[i]->index == start + into + i);
+		into += hop;
+	}
+}
+
+/*
+ * Scan the entire tree, only expecting present items (start, start+nr]
+ */
+void item_full_scan(struct radix_tree_root *root, unsigned long start,
+			unsigned long nr, int chunk)
+{
+	struct item *items[chunk];
+	unsigned long into = 0;
+	unsigned long this_index = start;
+	int nfound;
+	int i;
+
+//	printf("%s(0x%08lx, 0x%08lx, %d)\n", __FUNCTION__, start, nr, chunk);
+
+	while ((nfound = radix_tree_gang_lookup(root, (void **)items, into,
+					chunk))) {
+//		printf("At 0x%08lx, nfound=%d\n", into, nfound);
+		for (i = 0; i < nfound; i++) {
+			assert(items[i]->index == this_index);
+			this_index++;
+		}
+//		printf("Found 0x%08lx->0x%08lx\n",
+//			items[0]->index, items[nfound-1]->index);
+		into = this_index;
+	}
+	if (chunk)
+		assert(this_index == start + nr);
+	nfound = radix_tree_gang_lookup(root, (void **)items,
+					this_index, chunk);
+	assert(nfound == 0);
+}
+
+/* Use the same pattern as tag_pages_for_writeback() in mm/page-writeback.c */
+int tag_tagged_items(struct radix_tree_root *root, pthread_mutex_t *lock,
+			unsigned long start, unsigned long end, unsigned batch,
+			unsigned iftag, unsigned thentag)
+{
+	unsigned long tagged = 0;
+	struct radix_tree_iter iter;
+	void **slot;
+
+	if (batch == 0)
+		batch = 1;
+
+	if (lock)
+		pthread_mutex_lock(lock);
+	radix_tree_for_each_tagged(slot, root, &iter, start, iftag) {
+		if (iter.index > end)
+			break;
+		radix_tree_iter_tag_set(root, &iter, thentag);
+		tagged++;
+		if ((tagged % batch) != 0)
+			continue;
+		slot = radix_tree_iter_resume(slot, &iter);
+		if (lock) {
+			pthread_mutex_unlock(lock);
+			rcu_barrier();
+			pthread_mutex_lock(lock);
+		}
+	}
+	if (lock)
+		pthread_mutex_unlock(lock);
+
+	return tagged;
+}
+
+/* Use the same pattern as find_swap_entry() in mm/shmem.c */
+unsigned long find_item(struct radix_tree_root *root, void *item)
+{
+	struct radix_tree_iter iter;
+	void **slot;
+	unsigned long found = -1;
+	unsigned long checked = 0;
+
+	radix_tree_for_each_slot(slot, root, &iter, 0) {
+		if (*slot == item) {
+			found = iter.index;
+			break;
+		}
+		checked++;
+		if ((checked % 4) != 0)
+			continue;
+		slot = radix_tree_iter_resume(slot, &iter);
+	}
+
+	return found;
+}
+
+static int verify_node(struct radix_tree_node *slot, unsigned int tag,
+			int tagged)
+{
+	int anyset = 0;
+	int i;
+	int j;
+
+	slot = entry_to_node(slot);
+
+	/* Verify consistency at this level */
+	for (i = 0; i < RADIX_TREE_TAG_LONGS; i++) {
+		if (slot->tags[tag][i]) {
+			anyset = 1;
+			break;
+		}
+	}
+	if (tagged != anyset) {
+		printf("tag: %u, shift %u, tagged: %d, anyset: %d\n",
+			tag, slot->shift, tagged, anyset);
+		for (j = 0; j < RADIX_TREE_MAX_TAGS; j++) {
+			printf("tag %d: ", j);
+			for (i = 0; i < RADIX_TREE_TAG_LONGS; i++)
+				printf("%016lx ", slot->tags[j][i]);
+			printf("\n");
+		}
+		return 1;
+	}
+	assert(tagged == anyset);
+
+	/* Go for next level */
+	if (slot->shift > 0) {
+		for (i = 0; i < RADIX_TREE_MAP_SIZE; i++)
+			if (slot->slots[i])
+				if (verify_node(slot->slots[i], tag,
+					    !!test_bit(i, slot->tags[tag]))) {
+					printf("Failure at off %d\n", i);
+					for (j = 0; j < RADIX_TREE_MAX_TAGS; j++) {
+						printf("tag %d: ", j);
+						for (i = 0; i < RADIX_TREE_TAG_LONGS; i++)
+							printf("%016lx ", slot->tags[j][i]);
+						printf("\n");
+					}
+					return 1;
+				}
+	}
+	return 0;
+}
+
+void verify_tag_consistency(struct radix_tree_root *root, unsigned int tag)
+{
+	struct radix_tree_node *node = root->rnode;
+	if (!radix_tree_is_internal_node(node))
+		return;
+	verify_node(node, tag, !!root_tag_get(root, tag));
+}
+
+void item_kill_tree(struct radix_tree_root *root)
+{
+	struct radix_tree_iter iter;
+	void **slot;
+	struct item *items[32];
+	int nfound;
+
+	radix_tree_for_each_slot(slot, root, &iter, 0) {
+		if (radix_tree_exceptional_entry(*slot))
+			radix_tree_delete(root, iter.index);
+	}
+
+	while ((nfound = radix_tree_gang_lookup(root, (void **)items, 0, 32))) {
+		int i;
+
+		for (i = 0; i < nfound; i++) {
+			void *ret;
+
+			ret = radix_tree_delete(root, items[i]->index);
+			assert(ret == items[i]);
+			free(items[i]);
+		}
+	}
+	assert(radix_tree_gang_lookup(root, (void **)items, 0, 32) == 0);
+	assert(root->rnode == NULL);
+}
+
+void tree_verify_min_height(struct radix_tree_root *root, int maxindex)
+{
+	unsigned shift;
+	struct radix_tree_node *node = root->rnode;
+	if (!radix_tree_is_internal_node(node)) {
+		assert(maxindex == 0);
+		return;
+	}
+
+	node = entry_to_node(node);
+	assert(maxindex <= node_maxindex(node));
+
+	shift = node->shift;
+	if (shift > 0)
+		assert(maxindex > shift_maxindex(shift - RADIX_TREE_MAP_SHIFT));
+	else
+		assert(maxindex > 0);
+}

tools/testing/radix-tree/test.h

@@ -0,0 +1,65 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/gfp.h>
+#include <linux/types.h>
+#include <linux/radix-tree.h>
+#include <linux/rcupdate.h>
+
+struct item {
+	struct rcu_head	rcu_head;
+	unsigned long index;
+	unsigned int order;
+};
+
+struct item *item_create(unsigned long index, unsigned int order);
+int __item_insert(struct radix_tree_root *root, struct item *item);
+int item_insert(struct radix_tree_root *root, unsigned long index);
+void item_sanity(struct item *item, unsigned long index);
+int item_insert_order(struct radix_tree_root *root, unsigned long index,
+			unsigned order);
+int item_delete(struct radix_tree_root *root, unsigned long index);
+int item_delete_rcu(struct radix_tree_root *root, unsigned long index);
+struct item *item_lookup(struct radix_tree_root *root, unsigned long index);
+
+void item_check_present(struct radix_tree_root *root, unsigned long index);
+void item_check_absent(struct radix_tree_root *root, unsigned long index);
+void item_gang_check_present(struct radix_tree_root *root,
+			unsigned long start, unsigned long nr,
+			int chunk, int hop);
+void item_full_scan(struct radix_tree_root *root, unsigned long start,
+			unsigned long nr, int chunk);
+void item_kill_tree(struct radix_tree_root *root);
+
+int tag_tagged_items(struct radix_tree_root *, pthread_mutex_t *,
+			unsigned long start, unsigned long end, unsigned batch,
+			unsigned iftag, unsigned thentag);
+unsigned long find_item(struct radix_tree_root *, void *item);
+
+void tag_check(void);
+void multiorder_checks(void);
+void iteration_test(unsigned order, unsigned duration);
+void benchmark(void);
+void idr_checks(void);
+void ida_tests(void);
+
+struct item *
+item_tag_set(struct radix_tree_root *root, unsigned long index, int tag);
+struct item *
+item_tag_clear(struct radix_tree_root *root, unsigned long index, int tag);
+int item_tag_get(struct radix_tree_root *root, unsigned long index, int tag);
+void tree_verify_min_height(struct radix_tree_root *root, int maxindex);
+void verify_tag_consistency(struct radix_tree_root *root, unsigned int tag);
+
+extern int nr_allocated;
+
+/* Normally private parts of lib/radix-tree.c */
+struct radix_tree_node *entry_to_node(void *ptr);
+void radix_tree_dump(struct radix_tree_root *root);
+int root_tag_get(struct radix_tree_root *root, unsigned int tag);
+unsigned long node_maxindex(struct radix_tree_node *);
+unsigned long shift_maxindex(unsigned int shift);
+int radix_tree_cpu_dead(unsigned int cpu);
+struct radix_tree_preload {
+	unsigned nr;
+	struct radix_tree_node *nodes;
+};
+extern struct radix_tree_preload radix_tree_preloads;

tools/testing/scatterlist/Makefile

@@ -0,0 +1,30 @@
+CFLAGS += -I. -I../../include -g -O2 -Wall -fsanitize=address
+LDFLAGS += -fsanitize=address -fsanitize=undefined
+TARGETS = main
+OFILES = main.o scatterlist.o
+
+ifeq ($(BUILD), 32)
+        CFLAGS += -m32
+        LDFLAGS += -m32
+endif
+
+targets: include $(TARGETS)
+
+main: $(OFILES)
+
+clean:
+	$(RM) $(TARGETS) $(OFILES) scatterlist.c linux/scatterlist.h linux/highmem.h linux/kmemleak.h asm/io.h
+	@rmdir asm
+
+scatterlist.c: ../../../lib/scatterlist.c
+	@sed -e 's/^static //' -e 's/__always_inline //' -e 's/inline //' < $< > $@
+
+.PHONY: include
+
+include: ../../../include/linux/scatterlist.h
+	@mkdir -p linux
+	@mkdir -p asm
+	@touch asm/io.h
+	@touch linux/highmem.h
+	@touch linux/kmemleak.h
+	@cp $< linux/scatterlist.h

tools/testing/scatterlist/linux/mm.h

@@ -0,0 +1,125 @@
+#ifndef _LINUX_MM_H
+#define _LINUX_MM_H
+
+#include <assert.h>
+#include <string.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <limits.h>
+#include <stdio.h>
+
+typedef unsigned long dma_addr_t;
+
+#define unlikely
+
+#define BUG_ON(x) assert(!(x))
+
+#define WARN_ON(condition) ({                                           \
+	int __ret_warn_on = !!(condition);                              \
+	unlikely(__ret_warn_on);                                        \
+})
+
+#define WARN_ON_ONCE(condition) ({                              \
+	int __ret_warn_on = !!(condition);                      \
+	if (unlikely(__ret_warn_on))                            \
+		assert(0);                                      \
+	unlikely(__ret_warn_on);                                \
+})
+
+#define PAGE_SIZE	(4096)
+#define PAGE_SHIFT	(12)
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+
+#define __ALIGN_KERNEL(x, a)		__ALIGN_KERNEL_MASK(x, (typeof(x))(a) - 1)
+#define __ALIGN_KERNEL_MASK(x, mask)	(((x) + (mask)) & ~(mask))
+#define ALIGN(x, a)			__ALIGN_KERNEL((x), (a))
+
+#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)
+
+#define offset_in_page(p)	((unsigned long)(p) & ~PAGE_MASK)
+
+#define virt_to_page(x)	((void *)x)
+#define page_address(x)	((void *)x)
+
+static inline unsigned long page_to_phys(struct page *page)
+{
+	assert(0);
+
+	return 0;
+}
+
+#define page_to_pfn(page) ((unsigned long)(page) / PAGE_SIZE)
+#define pfn_to_page(pfn) (void *)((pfn) * PAGE_SIZE)
+#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
+
+#define __min(t1, t2, min1, min2, x, y) ({              \
+	t1 min1 = (x);                                  \
+	t2 min2 = (y);                                  \
+	(void) (&min1 == &min2);                        \
+	min1 < min2 ? min1 : min2; })
+
+#define ___PASTE(a,b) a##b
+#define __PASTE(a,b) ___PASTE(a,b)
+
+#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__)
+
+#define min(x, y)                                       \
+	__min(typeof(x), typeof(y),                     \
+	      __UNIQUE_ID(min1_), __UNIQUE_ID(min2_),   \
+	      x, y)
+
+#define min_t(type, x, y)                               \
+	__min(type, type,                               \
+	      __UNIQUE_ID(min1_), __UNIQUE_ID(min2_),   \
+	      x, y)
+
+#define preemptible() (1)
+
+static inline void *kmap(struct page *page)
+{
+	assert(0);
+
+	return NULL;
+}
+
+static inline void *kmap_atomic(struct page *page)
+{
+	assert(0);
+
+	return NULL;
+}
+
+static inline void kunmap(void *addr)
+{
+	assert(0);
+}
+
+static inline void kunmap_atomic(void *addr)
+{
+	assert(0);
+}
+
+static inline unsigned long __get_free_page(unsigned int flags)
+{
+	return (unsigned long)malloc(PAGE_SIZE);
+}
+
+static inline void free_page(unsigned long page)
+{
+	free((void *)page);
+}
+
+static inline void *kmalloc(unsigned int size, unsigned int flags)
+{
+	return malloc(size);
+}
+
+#define kfree(x) free(x)
+
+#define kmemleak_alloc(a, b, c, d)
+#define kmemleak_free(a)
+
+#define PageSlab(p) (0)
+#define flush_kernel_dcache_page(p)
+
+#endif

tools/testing/scatterlist/main.c

@@ -0,0 +1,79 @@
+#include <stdio.h>
+#include <assert.h>
+
+#include <linux/scatterlist.h>
+
+#define MAX_PAGES (64)
+
+static void set_pages(struct page **pages, const unsigned *array, unsigned num)
+{
+	unsigned int i;
+
+	assert(num < MAX_PAGES);
+	for (i = 0; i < num; i++)
+		pages[i] = (struct page *)(unsigned long)
+			   ((1 + array[i]) * PAGE_SIZE);
+}
+
+#define pfn(...) (unsigned []){ __VA_ARGS__ }
+
+int main(void)
+{
+	const unsigned int sgmax = SCATTERLIST_MAX_SEGMENT;
+	struct test {
+		int alloc_ret;
+		unsigned num_pages;
+		unsigned *pfn;
+		unsigned size;
+		unsigned int max_seg;
+		unsigned int expected_segments;
+	} *test, tests[] = {
+		{ -EINVAL, 1, pfn(0), PAGE_SIZE, PAGE_SIZE + 1, 1 },
+		{ -EINVAL, 1, pfn(0), PAGE_SIZE, 0, 1 },
+		{ -EINVAL, 1, pfn(0), PAGE_SIZE, sgmax + 1, 1 },
+		{ 0, 1, pfn(0), PAGE_SIZE, sgmax, 1 },
+		{ 0, 1, pfn(0), 1, sgmax, 1 },
+		{ 0, 2, pfn(0, 1), 2 * PAGE_SIZE, sgmax, 1 },
+		{ 0, 2, pfn(1, 0), 2 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 3, pfn(0, 1, 2), 3 * PAGE_SIZE, sgmax, 1 },
+		{ 0, 3, pfn(0, 2, 1), 3 * PAGE_SIZE, sgmax, 3 },
+		{ 0, 3, pfn(0, 1, 3), 3 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 3, pfn(1, 2, 4), 3 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 3, pfn(1, 3, 4), 3 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 4, pfn(0, 1, 3, 4), 4 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 5, pfn(0, 1, 3, 4, 5), 5 * PAGE_SIZE, sgmax, 2 },
+		{ 0, 5, pfn(0, 1, 3, 4, 6), 5 * PAGE_SIZE, sgmax, 3 },
+		{ 0, 5, pfn(0, 1, 2, 3, 4), 5 * PAGE_SIZE, sgmax, 1 },
+		{ 0, 5, pfn(0, 1, 2, 3, 4), 5 * PAGE_SIZE, 2 * PAGE_SIZE, 3 },
+		{ 0, 6, pfn(0, 1, 2, 3, 4, 5), 6 * PAGE_SIZE, 2 * PAGE_SIZE, 3 },
+		{ 0, 6, pfn(0, 2, 3, 4, 5, 6), 6 * PAGE_SIZE, 2 * PAGE_SIZE, 4 },
+		{ 0, 6, pfn(0, 1, 3, 4, 5, 6), 6 * PAGE_SIZE, 2 * PAGE_SIZE, 3 },
+		{ 0, 0, NULL, 0, 0, 0 },
+	};
+	unsigned int i;
+
+	for (i = 0, test = tests; test->expected_segments; test++, i++) {
+		struct page *pages[MAX_PAGES];
+		struct sg_table st;
+		int ret;
+
+		set_pages(pages, test->pfn, test->num_pages);
+
+		ret = __sg_alloc_table_from_pages(&st, pages, test->num_pages,
+						  0, test->size, test->max_seg,
+						  GFP_KERNEL);
+		assert(ret == test->alloc_ret);
+
+		if (test->alloc_ret)
+			continue;
+
+		assert(st.nents == test->expected_segments);
+		assert(st.orig_nents == test->expected_segments);
+
+		sg_free_table(&st);
+	}
+
+	assert(i == (sizeof(tests) / sizeof(tests[0])) - 1);
+
+	return 0;
+}

tools/testing/selftests/Makefile

@@ -0,0 +1,167 @@
+# SPDX-License-Identifier: GPL-2.0
+TARGETS = android
+TARGETS += bpf
+TARGETS += breakpoints
+TARGETS += capabilities
+TARGETS += cgroup
+TARGETS += cpufreq
+TARGETS += cpu-hotplug
+TARGETS += efivarfs
+TARGETS += exec
+TARGETS += filesystems
+TARGETS += firmware
+TARGETS += ftrace
+TARGETS += futex
+TARGETS += gpio
+TARGETS += intel_pstate
+TARGETS += ipc
+TARGETS += kcmp
+TARGETS += kvm
+TARGETS += lib
+TARGETS += membarrier
+TARGETS += memfd
+TARGETS += memory-hotplug
+TARGETS += mount
+TARGETS += mqueue
+TARGETS += net
+TARGETS += netfilter
+TARGETS += nsfs
+TARGETS += powerpc
+TARGETS += proc
+TARGETS += pstore
+TARGETS += ptrace
+TARGETS += rseq
+TARGETS += rtc
+TARGETS += seccomp
+TARGETS += sigaltstack
+TARGETS += size
+TARGETS += sparc64
+TARGETS += splice
+TARGETS += static_keys
+TARGETS += sync
+TARGETS += sysctl
+ifneq (1, $(quicktest))
+TARGETS += timers
+endif
+TARGETS += user
+TARGETS += vm
+TARGETS += x86
+TARGETS += zram
+#Please keep the TARGETS list alphabetically sorted
+# Run "make quicktest=1 run_tests" or
+# "make quicktest=1 kselftest" from top level Makefile
+
+TARGETS_HOTPLUG = cpu-hotplug
+TARGETS_HOTPLUG += memory-hotplug
+
+# Clear LDFLAGS and MAKEFLAGS if called from main
+# Makefile to avoid test build failures when test
+# Makefile doesn't have explicit build rules.
+ifeq (1,$(MAKELEVEL))
+override LDFLAGS =
+override MAKEFLAGS =
+endif
+
+ifneq ($(KBUILD_SRC),)
+override LDFLAGS =
+endif
+
+BUILD := $(O)
+ifndef BUILD
+  BUILD := $(KBUILD_OUTPUT)
+endif
+ifndef BUILD
+  BUILD := $(shell pwd)
+endif
+
+# KSFT_TAP_LEVEL is used from KSFT framework to prevent nested TAP header
+# printing from tests. Applicable to run_tests case where run_tests adds
+# TAP header prior running tests and when a test program invokes another
+# with system() call. Export it here to cover override RUN_TESTS defines.
+export KSFT_TAP_LEVEL=`echo 1`
+
+export BUILD
+all:
+	@for TARGET in $(TARGETS); do		\
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		mkdir $$BUILD_TARGET  -p;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET;\
+	done;
+
+run_tests: all
+	@for TARGET in $(TARGETS); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET run_tests;\
+	done;
+
+hotplug:
+	@for TARGET in $(TARGETS_HOTPLUG); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET;\
+	done;
+
+run_hotplug: hotplug
+	@for TARGET in $(TARGETS_HOTPLUG); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET run_full_test;\
+	done;
+
+clean_hotplug:
+	@for TARGET in $(TARGETS_HOTPLUG); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET clean;\
+	done;
+
+run_pstore_crash:
+	make -C pstore run_crash
+
+INSTALL_PATH ?= install
+INSTALL_PATH := $(abspath $(INSTALL_PATH))
+ALL_SCRIPT := $(INSTALL_PATH)/run_kselftest.sh
+
+install:
+ifdef INSTALL_PATH
+	@# Ask all targets to install their files
+	mkdir -p $(INSTALL_PATH)
+	@for TARGET in $(TARGETS); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET INSTALL_PATH=$(INSTALL_PATH)/$$TARGET install; \
+	done;
+
+	@# Ask all targets to emit their test scripts
+	echo "#!/bin/sh" > $(ALL_SCRIPT)
+	echo "BASE_DIR=\$$(realpath \$$(dirname \$$0))" >> $(ALL_SCRIPT)
+	echo "cd \$$BASE_DIR" >> $(ALL_SCRIPT)
+	echo "ROOT=\$$PWD" >> $(ALL_SCRIPT)
+	echo "if [ \"\$$1\" = \"--summary\" ]; then" >> $(ALL_SCRIPT)
+	echo "  OUTPUT=\$$BASE_DIR/output.log" >> $(ALL_SCRIPT)
+	echo "  cat /dev/null > \$$OUTPUT" >> $(ALL_SCRIPT)
+	echo "else" >> $(ALL_SCRIPT)
+	echo "  OUTPUT=/dev/stdout" >> $(ALL_SCRIPT)
+	echo "fi" >> $(ALL_SCRIPT)
+	echo "export KSFT_TAP_LEVEL=1" >> $(ALL_SCRIPT)
+	echo "export skip=4" >> $(ALL_SCRIPT)
+
+	for TARGET in $(TARGETS); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		echo "echo ; echo TAP version 13" >> $(ALL_SCRIPT);	\
+		echo "echo Running tests in $$TARGET" >> $(ALL_SCRIPT); \
+		echo "echo ========================================" >> $(ALL_SCRIPT); \
+		echo "[ -w /dev/kmsg ] && echo \"kselftest: Running tests in $$TARGET\" >> /dev/kmsg" >> $(ALL_SCRIPT); \
+		echo "cd $$TARGET" >> $(ALL_SCRIPT); \
+		make -s --no-print-directory OUTPUT=$$BUILD_TARGET -C $$TARGET emit_tests >> $(ALL_SCRIPT); \
+		echo "cd \$$ROOT" >> $(ALL_SCRIPT); \
+	done;
+
+	chmod u+x $(ALL_SCRIPT)
+else
+	$(error Error: set INSTALL_PATH to use install)
+endif
+
+clean:
+	@for TARGET in $(TARGETS); do \
+		BUILD_TARGET=$$BUILD/$$TARGET;	\
+		make OUTPUT=$$BUILD_TARGET -C $$TARGET clean;\
+	done;
+
+.PHONY: all run_tests hotplug run_hotplug clean_hotplug run_pstore_crash install clean

tools/testing/selftests/android/Makefile

@@ -0,0 +1,38 @@
+SUBDIRS := ion
+
+TEST_PROGS := run.sh
+
+.PHONY: all clean
+
+include ../lib.mk
+
+all:
+	@for DIR in $(SUBDIRS); do		\
+		BUILD_TARGET=$(OUTPUT)/$$DIR;	\
+		mkdir $$BUILD_TARGET  -p;	\
+		make OUTPUT=$$BUILD_TARGET -C $$DIR $@;\
+		#SUBDIR test prog name should be in the form: SUBDIR_test.sh \
+		TEST=$$DIR"_test.sh"; \
+		if [ -e $$DIR/$$TEST ]; then \
+			rsync -a $$DIR/$$TEST $$BUILD_TARGET/; \
+		fi \
+	done
+
+override define INSTALL_RULE
+	mkdir -p $(INSTALL_PATH)
+	install -t $(INSTALL_PATH) $(TEST_PROGS) $(TEST_PROGS_EXTENDED) $(TEST_FILES)
+
+	@for SUBDIR in $(SUBDIRS); do \
+		BUILD_TARGET=$(OUTPUT)/$$SUBDIR;	\
+		mkdir $$BUILD_TARGET  -p;	\
+		$(MAKE) OUTPUT=$$BUILD_TARGET -C $$SUBDIR INSTALL_PATH=$(INSTALL_PATH)/$$SUBDIR install; \
+	done;
+endef
+
+override define CLEAN
+	@for DIR in $(SUBDIRS); do		\
+		BUILD_TARGET=$(OUTPUT)/$$DIR;	\
+		mkdir $$BUILD_TARGET  -p;	\
+		make OUTPUT=$$BUILD_TARGET -C $$DIR $@;\
+	done
+endef

tools/testing/selftests/android/config

@@ -0,0 +1,5 @@
+CONFIG_ANDROID=y
+CONFIG_STAGING=y
+CONFIG_ION=y
+CONFIG_ION_SYSTEM_HEAP=y
+CONFIG_DRM_VGEM=y

tools/testing/selftests/android/ion/Makefile

@@ -0,0 +1,19 @@
+
+INCLUDEDIR := -I. -I../../../../../drivers/staging/android/uapi/ -I../../../../../usr/include/
+CFLAGS := $(CFLAGS) $(INCLUDEDIR) -Wall -O2 -g
+
+TEST_GEN_FILES := ionapp_export ionapp_import ionmap_test
+
+all: $(TEST_GEN_FILES)
+
+$(TEST_GEN_FILES): ipcsocket.c ionutils.c
+
+TEST_PROGS := ion_test.sh
+
+KSFT_KHDR_INSTALL := 1
+top_srcdir = ../../../../..
+include ../../lib.mk
+
+$(OUTPUT)/ionapp_export: ionapp_export.c ipcsocket.c ionutils.c
+$(OUTPUT)/ionapp_import: ionapp_import.c ipcsocket.c ionutils.c
+$(OUTPUT)/ionmap_test: ionmap_test.c ionutils.c

tools/testing/selftests/android/ion/README

@@ -0,0 +1,101 @@
+ION BUFFER SHARING UTILITY
+==========================
+File: ion_test.sh : Utility to test ION driver buffer sharing mechanism.
+Author: Pintu Kumar <pintu.ping@gmail.com>
+
+Introduction:
+-------------
+This is a test utility to verify ION buffer sharing in user space
+between 2 independent processes.
+It uses unix domain socket (with SCM_RIGHTS) as IPC to transfer an FD to
+another process to share the same buffer.
+This utility demonstrates how ION buffer sharing can be implemented between
+two user space processes, using various heap types.
+The following heap types are supported by ION driver.
+ION_HEAP_TYPE_SYSTEM (0)
+ION_HEAP_TYPE_SYSTEM_CONTIG (1)
+ION_HEAP_TYPE_CARVEOUT (2)
+ION_HEAP_TYPE_CHUNK (3)
+ION_HEAP_TYPE_DMA (4)
+
+By default only the SYSTEM and SYSTEM_CONTIG heaps are supported.
+Each heap is associated with the respective heap id.
+This utility is designed in the form of client/server program.
+The server part (ionapp_export) is the exporter of the buffer.
+It is responsible for creating an ION client, allocating the buffer based on
+the heap id, writing some data to this buffer and then exporting the FD
+(associated with this buffer) to another process using socket IPC.
+This FD is called as buffer FD (which is different than the ION client FD).
+
+The client part (ionapp_import) is the importer of the buffer.
+It retrives the FD from the socket data and installs into its address space.
+This new FD internally points to the same kernel buffer.
+So first it reads the data that is stored in this buffer and prints it.
+Then it writes the different size of data (it could be different data) to the
+same buffer.
+Finally the buffer FD must be closed by both the exporter and importer.
+Thus the same kernel buffer is shared among two user space processes using
+ION driver and only one time allocation.
+
+Prerequisite:
+-------------
+This utility works only if /dev/ion interface is present.
+The following configs needs to be enabled in kernel to include ion driver.
+CONFIG_ANDROID=y
+CONFIG_STAGING=y
+CONFIG_ION=y
+CONFIG_ION_SYSTEM_HEAP=y
+
+This utility requires to be run as root user.
+
+
+Compile and test:
+-----------------
+This utility is made to be run as part of kselftest framework in kernel.
+To compile and run using kselftest you can simply do the following from the
+kernel top directory.
+linux$ make TARGETS=android kselftest
+Or you can also use:
+linux$ make -C tools/testing/selftests TARGETS=android run_tests
+Using the selftest it can directly execute the ion_test.sh script to test the
+buffer sharing using ion system heap.
+Currently the heap size is hard coded as just 10 bytes inside this script.
+You need to be a root user to run under selftest.
+
+You can also compile and test manually using the following steps:
+ion$ make
+These will generate 2 executable: ionapp_export, ionapp_import
+Now you can run the export and import manually by specifying the heap type
+and the heap size.
+You can also directly execute the shell script to run the test automatically.
+Simply use the following command to run the test.
+ion$ sudo ./ion_test.sh
+
+Test Results:
+-------------
+The utility is verified on Ubuntu-32 bit system with Linux Kernel 4.14.
+Here is the snapshot of the test result using kselftest.
+
+linux# make TARGETS=android kselftest
+heap_type: 0, heap_size: 10
+--------------------------------------
+heap type: 0
+  heap id: 1
+heap name: ion_system_heap
+--------------------------------------
+Fill buffer content:
+0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd
+Sharing fd: 6, Client fd: 5
+<ion_close_buffer_fd>: buffer release successfully....
+Received buffer fd: 4
+Read buffer content:
+0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0x0 0x0 0x0 0x0 0x0 0x0
+0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0
+Fill buffer content:
+0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd
+0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd 0xfd
+0xfd 0xfd
+<ion_close_buffer_fd>: buffer release successfully....
+ion_test.sh: heap_type: 0 - [PASS]
+
+ion_test.sh: done

tools/testing/selftests/android/ion/ion.h

@@ -0,0 +1,143 @@
+/*
+ * ion.h
+ *
+ * Copyright (C) 2011 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+/* This file is copied from drivers/staging/android/uapi/ion.h
+ * This local copy is required for the selftest to pass, when build
+ * outside the kernel source tree.
+ * Please keep this file in sync with its original file until the
+ * ion driver is moved outside the staging tree.
+ */
+
+#ifndef _UAPI_LINUX_ION_H
+#define _UAPI_LINUX_ION_H
+
+#include <linux/ioctl.h>
+#include <linux/types.h>
+
+/**
+ * enum ion_heap_types - list of all possible types of heaps
+ * @ION_HEAP_TYPE_SYSTEM:	 memory allocated via vmalloc
+ * @ION_HEAP_TYPE_SYSTEM_CONTIG: memory allocated via kmalloc
+ * @ION_HEAP_TYPE_CARVEOUT:	 memory allocated from a prereserved
+ *				 carveout heap, allocations are physically
+ *				 contiguous
+ * @ION_HEAP_TYPE_DMA:		 memory allocated via DMA API
+ * @ION_NUM_HEAPS:		 helper for iterating over heaps, a bit mask
+ *				 is used to identify the heaps, so only 32
+ *				 total heap types are supported
+ */
+enum ion_heap_type {
+	ION_HEAP_TYPE_SYSTEM,
+	ION_HEAP_TYPE_SYSTEM_CONTIG,
+	ION_HEAP_TYPE_CARVEOUT,
+	ION_HEAP_TYPE_CHUNK,
+	ION_HEAP_TYPE_DMA,
+	ION_HEAP_TYPE_CUSTOM, /*
+			       * must be last so device specific heaps always
+			       * are at the end of this enum
+			       */
+};
+
+#define ION_NUM_HEAP_IDS		(sizeof(unsigned int) * 8)
+
+/**
+ * allocation flags - the lower 16 bits are used by core ion, the upper 16
+ * bits are reserved for use by the heaps themselves.
+ */
+
+/*
+ * mappings of this buffer should be cached, ion will do cache maintenance
+ * when the buffer is mapped for dma
+ */
+#define ION_FLAG_CACHED 1
+
+/**
+ * DOC: Ion Userspace API
+ *
+ * create a client by opening /dev/ion
+ * most operations handled via following ioctls
+ *
+ */
+
+/**
+ * struct ion_allocation_data - metadata passed from userspace for allocations
+ * @len:		size of the allocation
+ * @heap_id_mask:	mask of heap ids to allocate from
+ * @flags:		flags passed to heap
+ * @handle:		pointer that will be populated with a cookie to use to
+ *			refer to this allocation
+ *
+ * Provided by userspace as an argument to the ioctl
+ */
+struct ion_allocation_data {
+	__u64 len;
+	__u32 heap_id_mask;
+	__u32 flags;
+	__u32 fd;
+	__u32 unused;
+};
+
+#define MAX_HEAP_NAME			32
+
+/**
+ * struct ion_heap_data - data about a heap
+ * @name - first 32 characters of the heap name
+ * @type - heap type
+ * @heap_id - heap id for the heap
+ */
+struct ion_heap_data {
+	char name[MAX_HEAP_NAME];
+	__u32 type;
+	__u32 heap_id;
+	__u32 reserved0;
+	__u32 reserved1;
+	__u32 reserved2;
+};
+
+/**
+ * struct ion_heap_query - collection of data about all heaps
+ * @cnt - total number of heaps to be copied
+ * @heaps - buffer to copy heap data
+ */
+struct ion_heap_query {
+	__u32 cnt; /* Total number of heaps to be copied */
+	__u32 reserved0; /* align to 64bits */
+	__u64 heaps; /* buffer to be populated */
+	__u32 reserved1;
+	__u32 reserved2;
+};
+
+#define ION_IOC_MAGIC		'I'
+
+/**
+ * DOC: ION_IOC_ALLOC - allocate memory
+ *
+ * Takes an ion_allocation_data struct and returns it with the handle field
+ * populated with the opaque handle for the allocation.
+ */
+#define ION_IOC_ALLOC		_IOWR(ION_IOC_MAGIC, 0, \
+				      struct ion_allocation_data)
+
+/**
+ * DOC: ION_IOC_HEAP_QUERY - information about available heaps
+ *
+ * Takes an ion_heap_query structure and populates information about
+ * available Ion heaps.
+ */
+#define ION_IOC_HEAP_QUERY     _IOWR(ION_IOC_MAGIC, 8, \
+					struct ion_heap_query)
+
+#endif /* _UAPI_LINUX_ION_H */

tools/testing/selftests/android/ion/ion_test.sh

@@ -0,0 +1,58 @@
+#!/bin/bash
+
+heapsize=4096
+TCID="ion_test.sh"
+errcode=0
+
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+run_test()
+{
+	heaptype=$1
+	./ionapp_export -i $heaptype -s $heapsize &
+	sleep 1
+	./ionapp_import
+	if [ $? -ne 0 ]; then
+		echo "$TCID: heap_type: $heaptype - [FAIL]"
+		errcode=1
+	else
+		echo "$TCID: heap_type: $heaptype - [PASS]"
+	fi
+	sleep 1
+	echo ""
+}
+
+check_root()
+{
+	uid=$(id -u)
+	if [ $uid -ne 0 ]; then
+		echo $TCID: must be run as root >&2
+		exit $ksft_skip
+	fi
+}
+
+check_device()
+{
+	DEVICE=/dev/ion
+	if [ ! -e $DEVICE ]; then
+		echo $TCID: No $DEVICE device found >&2
+		echo $TCID: May be CONFIG_ION is not set >&2
+		exit $ksft_skip
+	fi
+}
+
+main_function()
+{
+	check_device
+	check_root
+
+	# ION_SYSTEM_HEAP TEST
+	run_test 0
+	# ION_SYSTEM_CONTIG_HEAP TEST
+	run_test 1
+}
+
+main_function
+echo "$TCID: done"
+exit $errcode

tools/testing/selftests/android/ion/ionapp_export.c

@@ -0,0 +1,136 @@
+/*
+ * ionapp_export.c
+ *
+ * It is a user space utility to create and export android
+ * ion memory buffer fd to another process using unix domain socket as IPC.
+ * This acts like a server for ionapp_import(client).
+ * So, this server has to be started first before the client.
+ *
+ * Copyright (C) 2017 Pintu Kumar <pintu.ping@gmail.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <errno.h>
+#include <sys/time.h>
+#include "ionutils.h"
+#include "ipcsocket.h"
+
+
+void print_usage(int argc, char *argv[])
+{
+	printf("Usage: %s [-h <help>] [-i <heap id>] [-s <size in bytes>]\n",
+		argv[0]);
+}
+
+int main(int argc, char *argv[])
+{
+	int opt, ret, status, heapid;
+	int sockfd, client_fd, shared_fd;
+	unsigned char *map_buf;
+	unsigned long map_len, heap_type, heap_size, flags;
+	struct ion_buffer_info info;
+	struct socket_info skinfo;
+
+	if (argc < 2) {
+		print_usage(argc, argv);
+		return -1;
+	}
+
+	heap_size = 0;
+	flags = 0;
+	heap_type = ION_HEAP_TYPE_SYSTEM;
+
+	while ((opt = getopt(argc, argv, "hi:s:")) != -1) {
+		switch (opt) {
+		case 'h':
+			print_usage(argc, argv);
+			exit(0);
+			break;
+		case 'i':
+			heapid = atoi(optarg);
+			switch (heapid) {
+			case 0:
+				heap_type = ION_HEAP_TYPE_SYSTEM;
+				break;
+			case 1:
+				heap_type = ION_HEAP_TYPE_SYSTEM_CONTIG;
+				break;
+			default:
+				printf("ERROR: heap type not supported\n");
+				exit(1);
+			}
+			break;
+		case 's':
+			heap_size = atoi(optarg);
+			break;
+		default:
+			print_usage(argc, argv);
+			exit(1);
+			break;
+		}
+	}
+
+	if (heap_size <= 0) {
+		printf("heap_size cannot be 0\n");
+		print_usage(argc, argv);
+		exit(1);
+	}
+
+	printf("heap_type: %ld, heap_size: %ld\n", heap_type, heap_size);
+	info.heap_type = heap_type;
+	info.heap_size = heap_size;
+	info.flag_type = flags;
+
+	/* This is server: open the socket connection first */
+	/* Here; 1 indicates server or exporter */
+	status = opensocket(&sockfd, SOCKET_NAME, 1);
+	if (status < 0) {
+		fprintf(stderr, "<%s>: Failed opensocket.\n", __func__);
+		goto err_socket;
+	}
+	skinfo.sockfd = sockfd;
+
+	ret = ion_export_buffer_fd(&info);
+	if (ret < 0) {
+		fprintf(stderr, "FAILED: ion_get_buffer_fd\n");
+		goto err_export;
+	}
+	client_fd = info.ionfd;
+	shared_fd = info.buffd;
+	map_buf = info.buffer;
+	map_len = info.buflen;
+	write_buffer(map_buf, map_len);
+
+	/* share ion buf fd with other user process */
+	printf("Sharing fd: %d, Client fd: %d\n", shared_fd, client_fd);
+	skinfo.datafd = shared_fd;
+	skinfo.buflen = map_len;
+
+	ret = socket_send_fd(&skinfo);
+	if (ret < 0) {
+		fprintf(stderr, "FAILED: socket_send_fd\n");
+		goto err_send;
+	}
+
+err_send:
+err_export:
+	ion_close_buffer_fd(&info);
+
+err_socket:
+	closesocket(sockfd, SOCKET_NAME);
+
+	return 0;
+}

tools/testing/selftests/android/ion/ionapp_import.c

@@ -0,0 +1,88 @@
+/*
+ * ionapp_import.c
+ *
+ * It is a user space utility to receive android ion memory buffer fd
+ * over unix domain socket IPC that can be exported by ionapp_export.
+ * This acts like a client for ionapp_export.
+ *
+ * Copyright (C) 2017 Pintu Kumar <pintu.ping@gmail.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+#include "ionutils.h"
+#include "ipcsocket.h"
+
+
+int main(void)
+{
+	int ret, status;
+	int sockfd, shared_fd;
+	unsigned char *map_buf;
+	unsigned long map_len;
+	struct ion_buffer_info info;
+	struct socket_info skinfo;
+
+	/* This is the client part. Here 0 means client or importer */
+	status = opensocket(&sockfd, SOCKET_NAME, 0);
+	if (status < 0) {
+		fprintf(stderr, "No exporter exists...\n");
+		ret = status;
+		goto err_socket;
+	}
+
+	skinfo.sockfd = sockfd;
+
+	ret = socket_receive_fd(&skinfo);
+	if (ret < 0) {
+		fprintf(stderr, "Failed: socket_receive_fd\n");
+		goto err_recv;
+	}
+
+	shared_fd = skinfo.datafd;
+	printf("Received buffer fd: %d\n", shared_fd);
+	if (shared_fd <= 0) {
+		fprintf(stderr, "ERROR: improper buf fd\n");
+		ret = -1;
+		goto err_fd;
+	}
+
+	memset(&info, 0, sizeof(info));
+	info.buffd = shared_fd;
+	info.buflen = ION_BUFFER_LEN;
+
+	ret = ion_import_buffer_fd(&info);
+	if (ret < 0) {
+		fprintf(stderr, "Failed: ion_use_buffer_fd\n");
+		goto err_import;
+	}
+
+	map_buf = info.buffer;
+	map_len = info.buflen;
+	read_buffer(map_buf, map_len);
+
+	/* Write probably new data to the same buffer again */
+	map_len = ION_BUFFER_LEN;
+	write_buffer(map_buf, map_len);
+
+err_import:
+	ion_close_buffer_fd(&info);
+err_fd:
+err_recv:
+err_socket:
+	closesocket(sockfd, SOCKET_NAME);
+
+	return ret;
+}

tools/testing/selftests/android/ion/ionmap_test.c

@@ -0,0 +1,136 @@
+#include <errno.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <string.h>
+#include <unistd.h>
+
+#include <sys/ioctl.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#include <linux/dma-buf.h>
+
+#include <drm/drm.h>
+
+#include "ion.h"
+#include "ionutils.h"
+
+int check_vgem(int fd)
+{
+	drm_version_t version = { 0 };
+	char name[5];
+	int ret;
+
+	version.name_len = 4;
+	version.name = name;
+
+	ret = ioctl(fd, DRM_IOCTL_VERSION, &version);
+	if (ret)
+		return 1;
+
+	return strcmp(name, "vgem");
+}
+
+int open_vgem(void)
+{
+	int i, fd;
+	const char *drmstr = "/dev/dri/card";
+
+	fd = -1;
+	for (i = 0; i < 16; i++) {
+		char name[80];
+
+		sprintf(name, "%s%u", drmstr, i);
+
+		fd = open(name, O_RDWR);
+		if (fd < 0)
+			continue;
+
+		if (check_vgem(fd)) {
+			close(fd);
+			continue;
+		} else {
+			break;
+		}
+
+	}
+	return fd;
+}
+
+int import_vgem_fd(int vgem_fd, int dma_buf_fd, uint32_t *handle)
+{
+	struct drm_prime_handle import_handle = { 0 };
+	int ret;
+
+	import_handle.fd = dma_buf_fd;
+	import_handle.flags = 0;
+	import_handle.handle = 0;
+
+	ret = ioctl(vgem_fd, DRM_IOCTL_PRIME_FD_TO_HANDLE, &import_handle);
+	if (ret == 0)
+		*handle = import_handle.handle;
+	return ret;
+}
+
+void close_handle(int vgem_fd, uint32_t handle)
+{
+	struct drm_gem_close close = { 0 };
+
+	close.handle = handle;
+	ioctl(vgem_fd, DRM_IOCTL_GEM_CLOSE, &close);
+}
+
+int main()
+{
+	int ret, vgem_fd;
+	struct ion_buffer_info info;
+	uint32_t handle = 0;
+	struct dma_buf_sync sync = { 0 };
+
+	info.heap_type = ION_HEAP_TYPE_SYSTEM;
+	info.heap_size = 4096;
+	info.flag_type = ION_FLAG_CACHED;
+
+	ret = ion_export_buffer_fd(&info);
+	if (ret < 0) {
+		printf("ion buffer alloc failed\n");
+		return -1;
+	}
+
+	vgem_fd = open_vgem();
+	if (vgem_fd < 0) {
+		ret = vgem_fd;
+		printf("Failed to open vgem\n");
+		goto out_ion;
+	}
+
+	ret = import_vgem_fd(vgem_fd, info.buffd, &handle);
+
+	if (ret < 0) {
+		printf("Failed to import buffer\n");
+		goto out_vgem;
+	}
+
+	sync.flags = DMA_BUF_SYNC_START | DMA_BUF_SYNC_RW;
+	ret = ioctl(info.buffd, DMA_BUF_IOCTL_SYNC, &sync);
+	if (ret)
+		printf("sync start failed %d\n", errno);
+
+	memset(info.buffer, 0xff, 4096);
+
+	sync.flags = DMA_BUF_SYNC_END | DMA_BUF_SYNC_RW;
+	ret = ioctl(info.buffd, DMA_BUF_IOCTL_SYNC, &sync);
+	if (ret)
+		printf("sync end failed %d\n", errno);
+
+	close_handle(vgem_fd, handle);
+	ret = 0;
+
+out_vgem:
+	close(vgem_fd);
+out_ion:
+	ion_close_buffer_fd(&info);
+	printf("done.\n");
+	return ret;
+}

tools/testing/selftests/android/ion/ionutils.c

@@ -0,0 +1,253 @@
+#include <stdio.h>
+#include <string.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <errno.h>
+//#include <stdint.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+#include "ionutils.h"
+#include "ipcsocket.h"
+
+
+void write_buffer(void *buffer, unsigned long len)
+{
+	int i;
+	unsigned char *ptr = (unsigned char *)buffer;
+
+	if (!ptr) {
+		fprintf(stderr, "<%s>: Invalid buffer...\n", __func__);
+		return;
+	}
+
+	printf("Fill buffer content:\n");
+	memset(ptr, 0xfd, len);
+	for (i = 0; i < len; i++)
+		printf("0x%x ", ptr[i]);
+	printf("\n");
+}
+
+void read_buffer(void *buffer, unsigned long len)
+{
+	int i;
+	unsigned char *ptr = (unsigned char *)buffer;
+
+	if (!ptr) {
+		fprintf(stderr, "<%s>: Invalid buffer...\n", __func__);
+		return;
+	}
+
+	printf("Read buffer content:\n");
+	for (i = 0; i < len; i++)
+		printf("0x%x ", ptr[i]);
+	printf("\n");
+}
+
+int ion_export_buffer_fd(struct ion_buffer_info *ion_info)
+{
+	int i, ret, ionfd, buffer_fd;
+	unsigned int heap_id;
+	unsigned long maplen;
+	unsigned char *map_buffer;
+	struct ion_allocation_data alloc_data;
+	struct ion_heap_query query;
+	struct ion_heap_data heap_data[MAX_HEAP_COUNT];
+
+	if (!ion_info) {
+		fprintf(stderr, "<%s>: Invalid ion info\n", __func__);
+		return -1;
+	}
+
+	/* Create an ION client */
+	ionfd = open(ION_DEVICE, O_RDWR);
+	if (ionfd < 0) {
+		fprintf(stderr, "<%s>: Failed to open ion client: %s\n",
+			__func__, strerror(errno));
+		return -1;
+	}
+
+	memset(&query, 0, sizeof(query));
+	query.cnt = MAX_HEAP_COUNT;
+	query.heaps = (unsigned long int)&heap_data[0];
+	/* Query ION heap_id_mask from ION heap */
+	ret = ioctl(ionfd, ION_IOC_HEAP_QUERY, &query);
+	if (ret < 0) {
+		fprintf(stderr, "<%s>: Failed: ION_IOC_HEAP_QUERY: %s\n",
+			__func__, strerror(errno));
+		goto err_query;
+	}
+
+	heap_id = MAX_HEAP_COUNT + 1;
+	for (i = 0; i < query.cnt; i++) {
+		if (heap_data[i].type == ion_info->heap_type) {
+			heap_id = heap_data[i].heap_id;
+			break;
+		}
+	}
+
+	if (heap_id > MAX_HEAP_COUNT) {
+		fprintf(stderr, "<%s>: ERROR: heap type does not exists\n",
+			__func__);
+		goto err_heap;
+	}
+
+	alloc_data.len = ion_info->heap_size;
+	alloc_data.heap_id_mask = 1 << heap_id;
+	alloc_data.flags = ion_info->flag_type;
+
+	/* Allocate memory for this ION client as per heap_type */
+	ret = ioctl(ionfd, ION_IOC_ALLOC, &alloc_data);
+	if (ret < 0) {
+		fprintf(stderr, "<%s>: Failed: ION_IOC_ALLOC: %s\n",
+			__func__, strerror(errno));
+		goto err_alloc;
+	}
+
+	/* This will return a valid buffer fd */
+	buffer_fd = alloc_data.fd;
+	maplen = alloc_data.len;
+
+	if (buffer_fd < 0 || maplen <= 0) {
+		fprintf(stderr, "<%s>: Invalid map data, fd: %d, len: %ld\n",
+			__func__, buffer_fd, maplen);
+		goto err_fd_data;
+	}
+
+	/* Create memory mapped buffer for the buffer fd */
+	map_buffer = (unsigned char *)mmap(NULL, maplen, PROT_READ|PROT_WRITE,
+			MAP_SHARED, buffer_fd, 0);
+	if (map_buffer == MAP_FAILED) {
+		fprintf(stderr, "<%s>: Failed: mmap: %s\n",
+			__func__, strerror(errno));
+		goto err_mmap;
+	}
+
+	ion_info->ionfd = ionfd;
+	ion_info->buffd = buffer_fd;
+	ion_info->buffer = map_buffer;
+	ion_info->buflen = maplen;
+
+	return 0;
+
+	munmap(map_buffer, maplen);
+
+err_fd_data:
+err_mmap:
+	/* in case of error: close the buffer fd */
+	if (buffer_fd)
+		close(buffer_fd);
+
+err_query:
+err_heap:
+err_alloc:
+	/* In case of error: close the ion client fd */
+	if (ionfd)
+		close(ionfd);
+
+	return -1;
+}
+
+int ion_import_buffer_fd(struct ion_buffer_info *ion_info)
+{
+	int buffd;
+	unsigned char *map_buf;
+	unsigned long map_len;
+
+	if (!ion_info) {
+		fprintf(stderr, "<%s>: Invalid ion info\n", __func__);
+		return -1;
+	}
+
+	map_len = ion_info->buflen;
+	buffd = ion_info->buffd;
+
+	if (buffd < 0 || map_len <= 0) {
+		fprintf(stderr, "<%s>: Invalid map data, fd: %d, len: %ld\n",
+			__func__, buffd, map_len);
+		goto err_buffd;
+	}
+
+	map_buf = (unsigned char *)mmap(NULL, map_len, PROT_READ|PROT_WRITE,
+			MAP_SHARED, buffd, 0);
+	if (map_buf == MAP_FAILED) {
+		printf("<%s>: Failed - mmap: %s\n",
+			__func__, strerror(errno));
+		goto err_mmap;
+	}
+
+	ion_info->buffer = map_buf;
+	ion_info->buflen = map_len;
+
+	return 0;
+
+err_mmap:
+	if (buffd)
+		close(buffd);
+
+err_buffd:
+	return -1;
+}
+
+void ion_close_buffer_fd(struct ion_buffer_info *ion_info)
+{
+	if (ion_info) {
+		/* unmap the buffer properly in the end */
+		munmap(ion_info->buffer, ion_info->buflen);
+		/* close the buffer fd */
+		if (ion_info->buffd > 0)
+			close(ion_info->buffd);
+		/* Finally, close the client fd */
+		if (ion_info->ionfd > 0)
+			close(ion_info->ionfd);
+	}
+}
+
+int socket_send_fd(struct socket_info *info)
+{
+	int status;
+	int fd, sockfd;
+	struct socketdata skdata;
+
+	if (!info) {
+		fprintf(stderr, "<%s>: Invalid socket info\n", __func__);
+		return -1;
+	}
+
+	sockfd = info->sockfd;
+	fd = info->datafd;
+	memset(&skdata, 0, sizeof(skdata));
+	skdata.data = fd;
+	skdata.len = sizeof(skdata.data);
+	status = sendtosocket(sockfd, &skdata);
+	if (status < 0) {
+		fprintf(stderr, "<%s>: Failed: sendtosocket\n", __func__);
+		return -1;
+	}
+
+	return 0;
+}
+
+int socket_receive_fd(struct socket_info *info)
+{
+	int status;
+	int fd, sockfd;
+	struct socketdata skdata;
+
+	if (!info) {
+		fprintf(stderr, "<%s>: Invalid socket info\n", __func__);
+		return -1;
+	}
+
+	sockfd = info->sockfd;
+	memset(&skdata, 0, sizeof(skdata));
+	status = receivefromsocket(sockfd, &skdata);
+	if (status < 0) {
+		fprintf(stderr, "<%s>: Failed: receivefromsocket\n", __func__);
+		return -1;
+	}
+
+	fd = (int)skdata.data;
+	info->datafd = fd;
+
+	return status;
+}

tools/testing/selftests/android/ion/ionutils.h

@@ -0,0 +1,55 @@
+#ifndef __ION_UTILS_H
+#define __ION_UTILS_H
+
+#include "ion.h"
+
+#define SOCKET_NAME "ion_socket"
+#define ION_DEVICE "/dev/ion"
+
+#define ION_BUFFER_LEN	4096
+#define MAX_HEAP_COUNT	ION_HEAP_TYPE_CUSTOM
+
+struct socket_info {
+	int sockfd;
+	int datafd;
+	unsigned long buflen;
+};
+
+struct ion_buffer_info {
+	int ionfd;
+	int buffd;
+	unsigned int heap_type;
+	unsigned int flag_type;
+	unsigned long heap_size;
+	unsigned long buflen;
+	unsigned char *buffer;
+};
+
+
+/* This is used to fill the data into the mapped buffer */
+void write_buffer(void *buffer, unsigned long len);
+
+/* This is used to read the data from the exported buffer */
+void read_buffer(void *buffer, unsigned long len);
+
+/* This is used to create an ION buffer FD for the kernel buffer
+ * So you can export this same buffer to others in the form of FD
+ */
+int ion_export_buffer_fd(struct ion_buffer_info *ion_info);
+
+/* This is used to import or map an exported FD.
+ * So we point to same buffer without making a copy. Hence zero-copy.
+ */
+int ion_import_buffer_fd(struct ion_buffer_info *ion_info);
+
+/* This is used to close all references for the ION client */
+void ion_close_buffer_fd(struct ion_buffer_info *ion_info);
+
+/* This is used to send FD to another process using socket IPC */
+int socket_send_fd(struct socket_info *skinfo);
+
+/* This is used to receive FD from another process using socket IPC */
+int socket_receive_fd(struct socket_info *skinfo);
+
+
+#endif

tools/testing/selftests/android/ion/ipcsocket.c

@@ -0,0 +1,227 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/time.h>
+#include <sys/un.h>
+#include <errno.h>
+
+#include "ipcsocket.h"
+
+
+int opensocket(int *sockfd, const char *name, int connecttype)
+{
+	int ret, temp = 1;
+
+	if (!name || strlen(name) > MAX_SOCK_NAME_LEN) {
+		fprintf(stderr, "<%s>: Invalid socket name.\n", __func__);
+		return -1;
+	}
+
+	ret = socket(PF_LOCAL, SOCK_STREAM, 0);
+	if (ret < 0) {
+		fprintf(stderr, "<%s>: Failed socket: <%s>\n",
+			__func__, strerror(errno));
+		return ret;
+	}
+
+	*sockfd = ret;
+	if (setsockopt(*sockfd, SOL_SOCKET, SO_REUSEADDR,
+		(char *)&temp, sizeof(int)) < 0) {
+		fprintf(stderr, "<%s>: Failed setsockopt: <%s>\n",
+		__func__, strerror(errno));
+		goto err;
+	}
+
+	sprintf(sock_name, "/tmp/%s", name);
+
+	if (connecttype == 1) {
+		/* This is for Server connection */
+		struct sockaddr_un skaddr;
+		int clientfd;
+		socklen_t sklen;
+
+		unlink(sock_name);
+		memset(&skaddr, 0, sizeof(skaddr));
+		skaddr.sun_family = AF_LOCAL;
+		strcpy(skaddr.sun_path, sock_name);
+
+		ret = bind(*sockfd, (struct sockaddr *)&skaddr,
+			SUN_LEN(&skaddr));
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed bind: <%s>\n",
+			__func__, strerror(errno));
+			goto err;
+		}
+
+		ret = listen(*sockfd, 5);
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed listen: <%s>\n",
+			__func__, strerror(errno));
+			goto err;
+		}
+
+		memset(&skaddr, 0, sizeof(skaddr));
+		sklen = sizeof(skaddr);
+
+		ret = accept(*sockfd, (struct sockaddr *)&skaddr,
+			(socklen_t *)&sklen);
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed accept: <%s>\n",
+			__func__, strerror(errno));
+			goto err;
+		}
+
+		clientfd = ret;
+		*sockfd = clientfd;
+	} else {
+		/* This is for client connection */
+		struct sockaddr_un skaddr;
+
+		memset(&skaddr, 0, sizeof(skaddr));
+		skaddr.sun_family = AF_LOCAL;
+		strcpy(skaddr.sun_path, sock_name);
+
+		ret = connect(*sockfd, (struct sockaddr *)&skaddr,
+			SUN_LEN(&skaddr));
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed connect: <%s>\n",
+			__func__, strerror(errno));
+			goto err;
+		}
+	}
+
+	return 0;
+
+err:
+	if (*sockfd)
+		close(*sockfd);
+
+	return ret;
+}
+
+int sendtosocket(int sockfd, struct socketdata *skdata)
+{
+	int ret, buffd;
+	unsigned int len;
+	char cmsg_b[CMSG_SPACE(sizeof(int))];
+	struct cmsghdr *cmsg;
+	struct msghdr msgh;
+	struct iovec iov;
+	struct timeval timeout;
+	fd_set selFDs;
+
+	if (!skdata) {
+		fprintf(stderr, "<%s>: socketdata is NULL\n", __func__);
+		return -1;
+	}
+
+	FD_ZERO(&selFDs);
+	FD_SET(0, &selFDs);
+	FD_SET(sockfd, &selFDs);
+	timeout.tv_sec = 20;
+	timeout.tv_usec = 0;
+
+	ret = select(sockfd+1, NULL, &selFDs, NULL, &timeout);
+	if (ret < 0) {
+		fprintf(stderr, "<%s>: Failed select: <%s>\n",
+		__func__, strerror(errno));
+		return -1;
+	}
+
+	if (FD_ISSET(sockfd, &selFDs)) {
+		buffd = skdata->data;
+		len = skdata->len;
+		memset(&msgh, 0, sizeof(msgh));
+		msgh.msg_control = &cmsg_b;
+		msgh.msg_controllen = CMSG_LEN(len);
+		iov.iov_base = "OK";
+		iov.iov_len = 2;
+		msgh.msg_iov = &iov;
+		msgh.msg_iovlen = 1;
+		cmsg = CMSG_FIRSTHDR(&msgh);
+		cmsg->cmsg_level = SOL_SOCKET;
+		cmsg->cmsg_type = SCM_RIGHTS;
+		cmsg->cmsg_len = CMSG_LEN(len);
+		memcpy(CMSG_DATA(cmsg), &buffd, len);
+
+		ret = sendmsg(sockfd, &msgh, MSG_DONTWAIT);
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed sendmsg: <%s>\n",
+			__func__, strerror(errno));
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+int receivefromsocket(int sockfd, struct socketdata *skdata)
+{
+	int ret, buffd;
+	unsigned int len = 0;
+	char cmsg_b[CMSG_SPACE(sizeof(int))];
+	struct cmsghdr *cmsg;
+	struct msghdr msgh;
+	struct iovec iov;
+	fd_set recvFDs;
+	char data[32];
+
+	if (!skdata) {
+		fprintf(stderr, "<%s>: socketdata is NULL\n", __func__);
+		return -1;
+	}
+
+	FD_ZERO(&recvFDs);
+	FD_SET(0, &recvFDs);
+	FD_SET(sockfd, &recvFDs);
+
+	ret = select(sockfd+1, &recvFDs, NULL, NULL, NULL);
+	if (ret < 0) {
+		fprintf(stderr, "<%s>: Failed select: <%s>\n",
+		__func__, strerror(errno));
+		return -1;
+	}
+
+	if (FD_ISSET(sockfd, &recvFDs)) {
+		len = sizeof(buffd);
+		memset(&msgh, 0, sizeof(msgh));
+		msgh.msg_control = &cmsg_b;
+		msgh.msg_controllen = CMSG_LEN(len);
+		iov.iov_base = data;
+		iov.iov_len = sizeof(data)-1;
+		msgh.msg_iov = &iov;
+		msgh.msg_iovlen = 1;
+		cmsg = CMSG_FIRSTHDR(&msgh);
+		cmsg->cmsg_level = SOL_SOCKET;
+		cmsg->cmsg_type = SCM_RIGHTS;
+		cmsg->cmsg_len = CMSG_LEN(len);
+
+		ret = recvmsg(sockfd, &msgh, MSG_DONTWAIT);
+		if (ret < 0) {
+			fprintf(stderr, "<%s>: Failed recvmsg: <%s>\n",
+			__func__, strerror(errno));
+			return -1;
+		}
+
+		memcpy(&buffd, CMSG_DATA(cmsg), len);
+		skdata->data = buffd;
+		skdata->len = len;
+	}
+	return 0;
+}
+
+int closesocket(int sockfd, char *name)
+{
+	char sockname[MAX_SOCK_NAME_LEN];
+
+	if (sockfd)
+		close(sockfd);
+	sprintf(sockname, "/tmp/%s", name);
+	unlink(sockname);
+	shutdown(sockfd, 2);
+
+	return 0;
+}

tools/testing/selftests/android/ion/ipcsocket.h

@@ -0,0 +1,35 @@
+
+#ifndef _IPCSOCKET_H
+#define _IPCSOCKET_H
+
+
+#define MAX_SOCK_NAME_LEN	64
+
+char sock_name[MAX_SOCK_NAME_LEN];
+
+/* This structure is responsible for holding the IPC data
+ * data: hold the buffer fd
+ * len: just the length of 32-bit integer fd
+ */
+struct socketdata {
+	int data;
+	unsigned int len;
+};
+
+/* This API is used to open the IPC socket connection
+ * name: implies a unique socket name in the system
+ * connecttype: implies server(0) or client(1)
+ */
+int opensocket(int *sockfd, const char *name, int connecttype);
+
+/* This is the API to send socket data over IPC socket */
+int sendtosocket(int sockfd, struct socketdata *data);
+
+/* This is the API to receive socket data over IPC socket */
+int receivefromsocket(int sockfd, struct socketdata *data);
+
+/* This is the API to close the socket connection */
+int closesocket(int sockfd, char *name);
+
+
+#endif

tools/testing/selftests/android/run.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+(cd ion; ./ion_test.sh)

tools/testing/selftests/arm64/Makefile

@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0
+
+# ARCH can be overridden by the user for cross compiling
+ARCH ?= $(shell uname -m 2>/dev/null || echo not)
+
+ifneq (,$(filter $(ARCH),aarch64 arm64))
+CFLAGS += -I../../../../usr/include/
+TEST_GEN_PROGS := tags_test
+TEST_PROGS := run_tags_test.sh
+endif
+
+include ../lib.mk

tools/testing/selftests/arm64/run_tags_test.sh

@@ -0,0 +1,12 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+
+echo "--------------------"
+echo "running tags test"
+echo "--------------------"
+./tags_test
+if [ $? -ne 0 ]; then
+	echo "[FAIL]"
+else
+	echo "[PASS]"
+fi

tools/testing/selftests/arm64/tags_test.c

@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <stdint.h>
+#include <sys/prctl.h>
+#include <sys/utsname.h>
+
+#define SHIFT_TAG(tag)		((uint64_t)(tag) << 56)
+#define SET_TAG(ptr, tag)	(((uint64_t)(ptr) & ~SHIFT_TAG(0xff)) | \
+					SHIFT_TAG(tag))
+
+int main(void)
+{
+	static int tbi_enabled = 0;
+	unsigned long tag = 0;
+	struct utsname *ptr;
+	int err;
+
+	if (prctl(PR_SET_TAGGED_ADDR_CTRL, PR_TAGGED_ADDR_ENABLE, 0, 0, 0) == 0)
+		tbi_enabled = 1;
+	ptr = (struct utsname *)malloc(sizeof(*ptr));
+	if (tbi_enabled)
+		tag = 0x42;
+	ptr = (struct utsname *)SET_TAG(ptr, tag);
+	err = uname(ptr);
+	free(ptr);
+
+	return err;
+}

tools/testing/selftests/bpf/Makefile

@@ -0,0 +1,145 @@
+# SPDX-License-Identifier: GPL-2.0
+
+LIBDIR := ../../../lib
+BPFDIR := $(LIBDIR)/bpf
+APIDIR := ../../../include/uapi
+GENDIR := ../../../../include/generated
+GENHDR := $(GENDIR)/autoconf.h
+
+ifneq ($(wildcard $(GENHDR)),)
+  GENFLAGS := -DHAVE_GENHDR
+endif
+
+CFLAGS += -Wall -O2 -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR) -I$(GENDIR) $(GENFLAGS) -I../../../include
+LDLIBS += -lcap -lelf -lrt -lpthread
+
+TEST_CUSTOM_PROGS = $(OUTPUT)/urandom_read
+all: $(TEST_CUSTOM_PROGS)
+
+$(TEST_CUSTOM_PROGS): $(OUTPUT)/%: %.c
+	$(CC) -o $(TEST_CUSTOM_PROGS) -static $< -Wl,--build-id
+
+# Order correspond to 'make run_tests' order
+TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
+	test_align test_verifier_log test_dev_cgroup test_tcpbpf_user \
+	test_sock test_btf test_sockmap test_lirc_mode2_user get_cgroup_id_user \
+	test_socket_cookie test_cgroup_storage test_select_reuseport
+
+TEST_GEN_FILES = test_pkt_access.o test_xdp.o test_l4lb.o test_tcp_estats.o test_obj_id.o \
+	test_pkt_md_access.o test_xdp_redirect.o test_xdp_meta.o sockmap_parse_prog.o     \
+	sockmap_verdict_prog.o dev_cgroup.o sample_ret0.o test_tracepoint.o \
+	test_l4lb_noinline.o test_xdp_noinline.o test_stacktrace_map.o \
+	sample_map_ret0.o test_tcpbpf_kern.o test_stacktrace_build_id.o \
+	sockmap_tcp_msg_prog.o connect4_prog.o connect6_prog.o test_adjust_tail.o \
+	test_btf_haskv.o test_btf_nokv.o test_sockmap_kern.o test_tunnel_kern.o \
+	test_get_stack_rawtp.o test_sockmap_kern.o test_sockhash_kern.o \
+	test_lwt_seg6local.o sendmsg4_prog.o sendmsg6_prog.o test_lirc_mode2_kern.o \
+	get_cgroup_id_kern.o socket_cookie_prog.o test_select_reuseport_kern.o \
+	test_skb_cgroup_id_kern.o
+
+# Order correspond to 'make run_tests' order
+TEST_PROGS := test_kmod.sh \
+	test_libbpf.sh \
+	test_xdp_redirect.sh \
+	test_xdp_meta.sh \
+	test_offload.py \
+	test_sock_addr.sh \
+	test_tunnel.sh \
+	test_lwt_seg6local.sh \
+	test_lirc_mode2.sh \
+	test_skb_cgroup_id.sh
+
+# Compile but not part of 'make run_tests'
+TEST_GEN_PROGS_EXTENDED = test_libbpf_open test_sock_addr test_skb_cgroup_id_user
+
+include ../lib.mk
+
+BPFOBJ := $(OUTPUT)/libbpf.a
+
+$(TEST_GEN_PROGS): $(BPFOBJ)
+
+$(TEST_GEN_PROGS_EXTENDED): $(OUTPUT)/libbpf.a
+
+$(OUTPUT)/test_dev_cgroup: cgroup_helpers.c
+$(OUTPUT)/test_skb_cgroup_id_user: cgroup_helpers.c
+$(OUTPUT)/test_sock: cgroup_helpers.c
+$(OUTPUT)/test_sock_addr: cgroup_helpers.c
+$(OUTPUT)/test_socket_cookie: cgroup_helpers.c
+$(OUTPUT)/test_sockmap: cgroup_helpers.c
+$(OUTPUT)/test_tcpbpf_user: cgroup_helpers.c
+$(OUTPUT)/test_progs: trace_helpers.c
+$(OUTPUT)/get_cgroup_id_user: cgroup_helpers.c
+$(OUTPUT)/test_cgroup_storage: cgroup_helpers.c
+
+.PHONY: force
+
+# force a rebuild of BPFOBJ when its dependencies are updated
+force:
+
+$(BPFOBJ): force
+	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
+
+CLANG ?= clang
+LLC   ?= llc
+LLVM_OBJCOPY ?= llvm-objcopy
+BTF_PAHOLE ?= pahole
+
+PROBE := $(shell $(LLC) -march=bpf -mcpu=probe -filetype=null /dev/null 2>&1)
+
+# Let newer LLVM versions transparently probe the kernel for availability
+# of full BPF instruction set.
+ifeq ($(PROBE),)
+  CPU ?= probe
+else
+  CPU ?= generic
+endif
+
+# Get Clang's default includes on this system, as opposed to those seen by
+# '-target bpf'. This fixes "missing" files on some architectures/distros,
+# such as asm/byteorder.h, asm/socket.h, asm/sockios.h, sys/cdefs.h etc.
+#
+# Use '-idirafter': Don't interfere with include mechanics except where the
+# build would have failed anyways.
+CLANG_SYS_INCLUDES := $(shell $(CLANG) -v -E - </dev/null 2>&1 \
+	| sed -n '/<...> search starts here:/,/End of search list./{ s| \(/.*\)|-idirafter \1|p }')
+
+CLANG_FLAGS = -I. -I./include/uapi -I../../../include/uapi \
+	      $(CLANG_SYS_INCLUDES) \
+	      -Wno-compare-distinct-pointer-types
+
+$(OUTPUT)/test_l4lb_noinline.o: CLANG_FLAGS += -fno-inline
+$(OUTPUT)/test_xdp_noinline.o: CLANG_FLAGS += -fno-inline
+
+BTF_LLC_PROBE := $(shell $(LLC) -march=bpf -mattr=help 2>&1 | grep dwarfris)
+BTF_PAHOLE_PROBE := $(shell $(BTF_PAHOLE) --help 2>&1 | grep BTF)
+BTF_OBJCOPY_PROBE := $(shell $(LLVM_OBJCOPY) --help 2>&1 | grep -i 'usage.*llvm')
+
+ifneq ($(BTF_LLC_PROBE),)
+ifneq ($(BTF_PAHOLE_PROBE),)
+ifneq ($(BTF_OBJCOPY_PROBE),)
+	CLANG_FLAGS += -g
+	LLC_FLAGS += -mattr=dwarfris
+	DWARF2BTF = y
+endif
+endif
+endif
+
+# Have one program compiled without "-target bpf" to test whether libbpf loads
+# it successfully
+$(OUTPUT)/test_xdp.o: test_xdp.c
+	$(CLANG) $(CLANG_FLAGS) \
+		-O2 -emit-llvm -c $< -o - | \
+	$(LLC) -march=bpf -mcpu=$(CPU) $(LLC_FLAGS) -filetype=obj -o $@
+ifeq ($(DWARF2BTF),y)
+	$(BTF_PAHOLE) -J $@
+endif
+
+$(OUTPUT)/%.o: %.c
+	$(CLANG) $(CLANG_FLAGS) \
+		 -O2 -target bpf -emit-llvm -c $< -o - |      \
+	$(LLC) -march=bpf -mcpu=$(CPU) $(LLC_FLAGS) -filetype=obj -o $@
+ifeq ($(DWARF2BTF),y)
+	$(BTF_PAHOLE) -J $@
+endif
+
+EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)

tools/testing/selftests/bpf/bpf_endian.h

@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_ENDIAN__
+#define __BPF_ENDIAN__
+
+#include <linux/swab.h>
+
+/* LLVM's BPF target selects the endianness of the CPU
+ * it compiles on, or the user specifies (bpfel/bpfeb),
+ * respectively. The used __BYTE_ORDER__ is defined by
+ * the compiler, we cannot rely on __BYTE_ORDER from
+ * libc headers, since it doesn't reflect the actual
+ * requested byte order.
+ *
+ * Note, LLVM's BPF target has different __builtin_bswapX()
+ * semantics. It does map to BPF_ALU | BPF_END | BPF_TO_BE
+ * in bpfel and bpfeb case, which means below, that we map
+ * to cpu_to_be16(). We could use it unconditionally in BPF
+ * case, but better not rely on it, so that this header here
+ * can be used from application and BPF program side, which
+ * use different targets.
+ */
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+# define __bpf_ntohs(x)			__builtin_bswap16(x)
+# define __bpf_htons(x)			__builtin_bswap16(x)
+# define __bpf_constant_ntohs(x)	___constant_swab16(x)
+# define __bpf_constant_htons(x)	___constant_swab16(x)
+# define __bpf_ntohl(x)			__builtin_bswap32(x)
+# define __bpf_htonl(x)			__builtin_bswap32(x)
+# define __bpf_constant_ntohl(x)	___constant_swab32(x)
+# define __bpf_constant_htonl(x)	___constant_swab32(x)
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+# define __bpf_ntohs(x)			(x)
+# define __bpf_htons(x)			(x)
+# define __bpf_constant_ntohs(x)	(x)
+# define __bpf_constant_htons(x)	(x)
+# define __bpf_ntohl(x)			(x)
+# define __bpf_htonl(x)			(x)
+# define __bpf_constant_ntohl(x)	(x)
+# define __bpf_constant_htonl(x)	(x)
+#else
+# error "Fix your compiler's __BYTE_ORDER__?!"
+#endif
+
+#define bpf_htons(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_htons(x) : __bpf_htons(x))
+#define bpf_ntohs(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_ntohs(x) : __bpf_ntohs(x))
+#define bpf_htonl(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_htonl(x) : __bpf_htonl(x))
+#define bpf_ntohl(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_ntohl(x) : __bpf_ntohl(x))
+
+#endif /* __BPF_ENDIAN__ */

tools/testing/selftests/bpf/bpf_helpers.h

@@ -0,0 +1,336 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_HELPERS_H
+#define __BPF_HELPERS_H
+
+/* helper macro to place programs, maps, license in
+ * different sections in elf_bpf file. Section names
+ * are interpreted by elf_bpf loader
+ */
+#define SEC(NAME) __attribute__((section(NAME), used))
+
+/* helper functions called from eBPF programs written in C */
+static void *(*bpf_map_lookup_elem)(void *map, void *key) =
+	(void *) BPF_FUNC_map_lookup_elem;
+static int (*bpf_map_update_elem)(void *map, void *key, void *value,
+				  unsigned long long flags) =
+	(void *) BPF_FUNC_map_update_elem;
+static int (*bpf_map_delete_elem)(void *map, void *key) =
+	(void *) BPF_FUNC_map_delete_elem;
+static int (*bpf_probe_read)(void *dst, int size, void *unsafe_ptr) =
+	(void *) BPF_FUNC_probe_read;
+static unsigned long long (*bpf_ktime_get_ns)(void) =
+	(void *) BPF_FUNC_ktime_get_ns;
+static int (*bpf_trace_printk)(const char *fmt, int fmt_size, ...) =
+	(void *) BPF_FUNC_trace_printk;
+static void (*bpf_tail_call)(void *ctx, void *map, int index) =
+	(void *) BPF_FUNC_tail_call;
+static unsigned long long (*bpf_get_smp_processor_id)(void) =
+	(void *) BPF_FUNC_get_smp_processor_id;
+static unsigned long long (*bpf_get_current_pid_tgid)(void) =
+	(void *) BPF_FUNC_get_current_pid_tgid;
+static unsigned long long (*bpf_get_current_uid_gid)(void) =
+	(void *) BPF_FUNC_get_current_uid_gid;
+static int (*bpf_get_current_comm)(void *buf, int buf_size) =
+	(void *) BPF_FUNC_get_current_comm;
+static unsigned long long (*bpf_perf_event_read)(void *map,
+						 unsigned long long flags) =
+	(void *) BPF_FUNC_perf_event_read;
+static int (*bpf_clone_redirect)(void *ctx, int ifindex, int flags) =
+	(void *) BPF_FUNC_clone_redirect;
+static int (*bpf_redirect)(int ifindex, int flags) =
+	(void *) BPF_FUNC_redirect;
+static int (*bpf_redirect_map)(void *map, int key, int flags) =
+	(void *) BPF_FUNC_redirect_map;
+static int (*bpf_perf_event_output)(void *ctx, void *map,
+				    unsigned long long flags, void *data,
+				    int size) =
+	(void *) BPF_FUNC_perf_event_output;
+static int (*bpf_get_stackid)(void *ctx, void *map, int flags) =
+	(void *) BPF_FUNC_get_stackid;
+static int (*bpf_probe_write_user)(void *dst, void *src, int size) =
+	(void *) BPF_FUNC_probe_write_user;
+static int (*bpf_current_task_under_cgroup)(void *map, int index) =
+	(void *) BPF_FUNC_current_task_under_cgroup;
+static int (*bpf_skb_get_tunnel_key)(void *ctx, void *key, int size, int flags) =
+	(void *) BPF_FUNC_skb_get_tunnel_key;
+static int (*bpf_skb_set_tunnel_key)(void *ctx, void *key, int size, int flags) =
+	(void *) BPF_FUNC_skb_set_tunnel_key;
+static int (*bpf_skb_get_tunnel_opt)(void *ctx, void *md, int size) =
+	(void *) BPF_FUNC_skb_get_tunnel_opt;
+static int (*bpf_skb_set_tunnel_opt)(void *ctx, void *md, int size) =
+	(void *) BPF_FUNC_skb_set_tunnel_opt;
+static unsigned long long (*bpf_get_prandom_u32)(void) =
+	(void *) BPF_FUNC_get_prandom_u32;
+static int (*bpf_xdp_adjust_head)(void *ctx, int offset) =
+	(void *) BPF_FUNC_xdp_adjust_head;
+static int (*bpf_xdp_adjust_meta)(void *ctx, int offset) =
+	(void *) BPF_FUNC_xdp_adjust_meta;
+static int (*bpf_get_socket_cookie)(void *ctx) =
+	(void *) BPF_FUNC_get_socket_cookie;
+static int (*bpf_setsockopt)(void *ctx, int level, int optname, void *optval,
+			     int optlen) =
+	(void *) BPF_FUNC_setsockopt;
+static int (*bpf_getsockopt)(void *ctx, int level, int optname, void *optval,
+			     int optlen) =
+	(void *) BPF_FUNC_getsockopt;
+static int (*bpf_sock_ops_cb_flags_set)(void *ctx, int flags) =
+	(void *) BPF_FUNC_sock_ops_cb_flags_set;
+static int (*bpf_sk_redirect_map)(void *ctx, void *map, int key, int flags) =
+	(void *) BPF_FUNC_sk_redirect_map;
+static int (*bpf_sk_redirect_hash)(void *ctx, void *map, void *key, int flags) =
+	(void *) BPF_FUNC_sk_redirect_hash;
+static int (*bpf_sock_map_update)(void *map, void *key, void *value,
+				  unsigned long long flags) =
+	(void *) BPF_FUNC_sock_map_update;
+static int (*bpf_sock_hash_update)(void *map, void *key, void *value,
+				   unsigned long long flags) =
+	(void *) BPF_FUNC_sock_hash_update;
+static int (*bpf_perf_event_read_value)(void *map, unsigned long long flags,
+					void *buf, unsigned int buf_size) =
+	(void *) BPF_FUNC_perf_event_read_value;
+static int (*bpf_perf_prog_read_value)(void *ctx, void *buf,
+				       unsigned int buf_size) =
+	(void *) BPF_FUNC_perf_prog_read_value;
+static int (*bpf_override_return)(void *ctx, unsigned long rc) =
+	(void *) BPF_FUNC_override_return;
+static int (*bpf_msg_redirect_map)(void *ctx, void *map, int key, int flags) =
+	(void *) BPF_FUNC_msg_redirect_map;
+static int (*bpf_msg_redirect_hash)(void *ctx,
+				    void *map, void *key, int flags) =
+	(void *) BPF_FUNC_msg_redirect_hash;
+static int (*bpf_msg_apply_bytes)(void *ctx, int len) =
+	(void *) BPF_FUNC_msg_apply_bytes;
+static int (*bpf_msg_cork_bytes)(void *ctx, int len) =
+	(void *) BPF_FUNC_msg_cork_bytes;
+static int (*bpf_msg_pull_data)(void *ctx, int start, int end, int flags) =
+	(void *) BPF_FUNC_msg_pull_data;
+static int (*bpf_bind)(void *ctx, void *addr, int addr_len) =
+	(void *) BPF_FUNC_bind;
+static int (*bpf_xdp_adjust_tail)(void *ctx, int offset) =
+	(void *) BPF_FUNC_xdp_adjust_tail;
+static int (*bpf_skb_get_xfrm_state)(void *ctx, int index, void *state,
+				     int size, int flags) =
+	(void *) BPF_FUNC_skb_get_xfrm_state;
+static int (*bpf_sk_select_reuseport)(void *ctx, void *map, void *key, __u32 flags) =
+	(void *) BPF_FUNC_sk_select_reuseport;
+static int (*bpf_get_stack)(void *ctx, void *buf, int size, int flags) =
+	(void *) BPF_FUNC_get_stack;
+static int (*bpf_fib_lookup)(void *ctx, struct bpf_fib_lookup *params,
+			     int plen, __u32 flags) =
+	(void *) BPF_FUNC_fib_lookup;
+static int (*bpf_lwt_push_encap)(void *ctx, unsigned int type, void *hdr,
+				 unsigned int len) =
+	(void *) BPF_FUNC_lwt_push_encap;
+static int (*bpf_lwt_seg6_store_bytes)(void *ctx, unsigned int offset,
+				       void *from, unsigned int len) =
+	(void *) BPF_FUNC_lwt_seg6_store_bytes;
+static int (*bpf_lwt_seg6_action)(void *ctx, unsigned int action, void *param,
+				  unsigned int param_len) =
+	(void *) BPF_FUNC_lwt_seg6_action;
+static int (*bpf_lwt_seg6_adjust_srh)(void *ctx, unsigned int offset,
+				      unsigned int len) =
+	(void *) BPF_FUNC_lwt_seg6_adjust_srh;
+static int (*bpf_rc_repeat)(void *ctx) =
+	(void *) BPF_FUNC_rc_repeat;
+static int (*bpf_rc_keydown)(void *ctx, unsigned int protocol,
+			     unsigned long long scancode, unsigned int toggle) =
+	(void *) BPF_FUNC_rc_keydown;
+static unsigned long long (*bpf_get_current_cgroup_id)(void) =
+	(void *) BPF_FUNC_get_current_cgroup_id;
+static void *(*bpf_get_local_storage)(void *map, unsigned long long flags) =
+	(void *) BPF_FUNC_get_local_storage;
+static unsigned long long (*bpf_skb_cgroup_id)(void *ctx) =
+	(void *) BPF_FUNC_skb_cgroup_id;
+static unsigned long long (*bpf_skb_ancestor_cgroup_id)(void *ctx, int level) =
+	(void *) BPF_FUNC_skb_ancestor_cgroup_id;
+
+/* llvm builtin functions that eBPF C program may use to
+ * emit BPF_LD_ABS and BPF_LD_IND instructions
+ */
+struct sk_buff;
+unsigned long long load_byte(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.byte");
+unsigned long long load_half(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.half");
+unsigned long long load_word(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.word");
+
+/* a helper structure used by eBPF C program
+ * to describe map attributes to elf_bpf loader
+ */
+struct bpf_map_def {
+	unsigned int type;
+	unsigned int key_size;
+	unsigned int value_size;
+	unsigned int max_entries;
+	unsigned int map_flags;
+	unsigned int inner_map_idx;
+	unsigned int numa_node;
+};
+
+#define BPF_ANNOTATE_KV_PAIR(name, type_key, type_val)		\
+	struct ____btf_map_##name {				\
+		type_key key;					\
+		type_val value;					\
+	};							\
+	struct ____btf_map_##name				\
+	__attribute__ ((section(".maps." #name), used))		\
+		____btf_map_##name = { }
+
+static int (*bpf_skb_load_bytes)(void *ctx, int off, void *to, int len) =
+	(void *) BPF_FUNC_skb_load_bytes;
+static int (*bpf_skb_load_bytes_relative)(void *ctx, int off, void *to, int len, __u32 start_header) =
+	(void *) BPF_FUNC_skb_load_bytes_relative;
+static int (*bpf_skb_store_bytes)(void *ctx, int off, void *from, int len, int flags) =
+	(void *) BPF_FUNC_skb_store_bytes;
+static int (*bpf_l3_csum_replace)(void *ctx, int off, int from, int to, int flags) =
+	(void *) BPF_FUNC_l3_csum_replace;
+static int (*bpf_l4_csum_replace)(void *ctx, int off, int from, int to, int flags) =
+	(void *) BPF_FUNC_l4_csum_replace;
+static int (*bpf_csum_diff)(void *from, int from_size, void *to, int to_size, int seed) =
+	(void *) BPF_FUNC_csum_diff;
+static int (*bpf_skb_under_cgroup)(void *ctx, void *map, int index) =
+	(void *) BPF_FUNC_skb_under_cgroup;
+static int (*bpf_skb_change_head)(void *, int len, int flags) =
+	(void *) BPF_FUNC_skb_change_head;
+static int (*bpf_skb_pull_data)(void *, int len) =
+	(void *) BPF_FUNC_skb_pull_data;
+
+/* Scan the ARCH passed in from ARCH env variable (see Makefile) */
+#if defined(__TARGET_ARCH_x86)
+	#define bpf_target_x86
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_s930x)
+	#define bpf_target_s930x
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_arm64)
+	#define bpf_target_arm64
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_mips)
+	#define bpf_target_mips
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_powerpc)
+	#define bpf_target_powerpc
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_sparc)
+	#define bpf_target_sparc
+	#define bpf_target_defined
+#else
+	#undef bpf_target_defined
+#endif
+
+/* Fall back to what the compiler says */
+#ifndef bpf_target_defined
+#if defined(__x86_64__)
+	#define bpf_target_x86
+#elif defined(__s390x__)
+	#define bpf_target_s930x
+#elif defined(__aarch64__)
+	#define bpf_target_arm64
+#elif defined(__mips__)
+	#define bpf_target_mips
+#elif defined(__powerpc__)
+	#define bpf_target_powerpc
+#elif defined(__sparc__)
+	#define bpf_target_sparc
+#endif
+#endif
+
+#if defined(bpf_target_x86)
+
+#define PT_REGS_PARM1(x) ((x)->di)
+#define PT_REGS_PARM2(x) ((x)->si)
+#define PT_REGS_PARM3(x) ((x)->dx)
+#define PT_REGS_PARM4(x) ((x)->cx)
+#define PT_REGS_PARM5(x) ((x)->r8)
+#define PT_REGS_RET(x) ((x)->sp)
+#define PT_REGS_FP(x) ((x)->bp)
+#define PT_REGS_RC(x) ((x)->ax)
+#define PT_REGS_SP(x) ((x)->sp)
+#define PT_REGS_IP(x) ((x)->ip)
+
+#elif defined(bpf_target_s390x)
+
+#define PT_REGS_PARM1(x) ((x)->gprs[2])
+#define PT_REGS_PARM2(x) ((x)->gprs[3])
+#define PT_REGS_PARM3(x) ((x)->gprs[4])
+#define PT_REGS_PARM4(x) ((x)->gprs[5])
+#define PT_REGS_PARM5(x) ((x)->gprs[6])
+#define PT_REGS_RET(x) ((x)->gprs[14])
+#define PT_REGS_FP(x) ((x)->gprs[11]) /* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_RC(x) ((x)->gprs[2])
+#define PT_REGS_SP(x) ((x)->gprs[15])
+#define PT_REGS_IP(x) ((x)->psw.addr)
+
+#elif defined(bpf_target_arm64)
+
+#define PT_REGS_PARM1(x) ((x)->regs[0])
+#define PT_REGS_PARM2(x) ((x)->regs[1])
+#define PT_REGS_PARM3(x) ((x)->regs[2])
+#define PT_REGS_PARM4(x) ((x)->regs[3])
+#define PT_REGS_PARM5(x) ((x)->regs[4])
+#define PT_REGS_RET(x) ((x)->regs[30])
+#define PT_REGS_FP(x) ((x)->regs[29]) /* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_RC(x) ((x)->regs[0])
+#define PT_REGS_SP(x) ((x)->sp)
+#define PT_REGS_IP(x) ((x)->pc)
+
+#elif defined(bpf_target_mips)
+
+#define PT_REGS_PARM1(x) ((x)->regs[4])
+#define PT_REGS_PARM2(x) ((x)->regs[5])
+#define PT_REGS_PARM3(x) ((x)->regs[6])
+#define PT_REGS_PARM4(x) ((x)->regs[7])
+#define PT_REGS_PARM5(x) ((x)->regs[8])
+#define PT_REGS_RET(x) ((x)->regs[31])
+#define PT_REGS_FP(x) ((x)->regs[30]) /* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_RC(x) ((x)->regs[1])
+#define PT_REGS_SP(x) ((x)->regs[29])
+#define PT_REGS_IP(x) ((x)->cp0_epc)
+
+#elif defined(bpf_target_powerpc)
+
+#define PT_REGS_PARM1(x) ((x)->gpr[3])
+#define PT_REGS_PARM2(x) ((x)->gpr[4])
+#define PT_REGS_PARM3(x) ((x)->gpr[5])
+#define PT_REGS_PARM4(x) ((x)->gpr[6])
+#define PT_REGS_PARM5(x) ((x)->gpr[7])
+#define PT_REGS_RC(x) ((x)->gpr[3])
+#define PT_REGS_SP(x) ((x)->sp)
+#define PT_REGS_IP(x) ((x)->nip)
+
+#elif defined(bpf_target_sparc)
+
+#define PT_REGS_PARM1(x) ((x)->u_regs[UREG_I0])
+#define PT_REGS_PARM2(x) ((x)->u_regs[UREG_I1])
+#define PT_REGS_PARM3(x) ((x)->u_regs[UREG_I2])
+#define PT_REGS_PARM4(x) ((x)->u_regs[UREG_I3])
+#define PT_REGS_PARM5(x) ((x)->u_regs[UREG_I4])
+#define PT_REGS_RET(x) ((x)->u_regs[UREG_I7])
+#define PT_REGS_RC(x) ((x)->u_regs[UREG_I0])
+#define PT_REGS_SP(x) ((x)->u_regs[UREG_FP])
+
+/* Should this also be a bpf_target check for the sparc case? */
+#if defined(__arch64__)
+#define PT_REGS_IP(x) ((x)->tpc)
+#else
+#define PT_REGS_IP(x) ((x)->pc)
+#endif
+
+#endif
+
+#ifdef bpf_target_powerpc
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({ (ip) = (ctx)->link; })
+#define BPF_KRETPROBE_READ_RET_IP		BPF_KPROBE_READ_RET_IP
+#elif bpf_target_sparc
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({ (ip) = PT_REGS_RET(ctx); })
+#define BPF_KRETPROBE_READ_RET_IP		BPF_KPROBE_READ_RET_IP
+#else
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({				\
+		bpf_probe_read(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); })
+#define BPF_KRETPROBE_READ_RET_IP(ip, ctx)	({				\
+		bpf_probe_read(&(ip), sizeof(ip),				\
+				(void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
+#endif
+
+#endif

tools/testing/selftests/bpf/bpf_rand.h

@@ -0,0 +1,80 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_RAND__
+#define __BPF_RAND__
+
+#include <stdint.h>
+#include <stdlib.h>
+#include <time.h>
+
+static inline uint64_t bpf_rand_mask(uint64_t mask)
+{
+	return (((uint64_t)(uint32_t)rand()) |
+	        ((uint64_t)(uint32_t)rand() << 32)) & mask;
+}
+
+#define bpf_rand_ux(x, m)			\
+static inline uint64_t bpf_rand_u##x(int shift)	\
+{						\
+	return bpf_rand_mask((m)) << shift;	\
+}
+
+bpf_rand_ux( 8,               0xffULL)
+bpf_rand_ux(16,             0xffffULL)
+bpf_rand_ux(24,           0xffffffULL)
+bpf_rand_ux(32,         0xffffffffULL)
+bpf_rand_ux(40,       0xffffffffffULL)
+bpf_rand_ux(48,     0xffffffffffffULL)
+bpf_rand_ux(56,   0xffffffffffffffULL)
+bpf_rand_ux(64, 0xffffffffffffffffULL)
+
+static inline void bpf_semi_rand_init(void)
+{
+	srand(time(NULL));
+}
+
+static inline uint64_t bpf_semi_rand_get(void)
+{
+	switch (rand() % 39) {
+	case  0: return 0x000000ff00000000ULL | bpf_rand_u8(0);
+	case  1: return 0xffffffff00000000ULL | bpf_rand_u16(0);
+	case  2: return 0x00000000ffff0000ULL | bpf_rand_u16(0);
+	case  3: return 0x8000000000000000ULL | bpf_rand_u32(0);
+	case  4: return 0x00000000f0000000ULL | bpf_rand_u32(0);
+	case  5: return 0x0000000100000000ULL | bpf_rand_u24(0);
+	case  6: return 0x800ff00000000000ULL | bpf_rand_u32(0);
+	case  7: return 0x7fffffff00000000ULL | bpf_rand_u32(0);
+	case  8: return 0xffffffffffffff00ULL ^ bpf_rand_u32(24);
+	case  9: return 0xffffffffffffff00ULL | bpf_rand_u8(0);
+	case 10: return 0x0000000010000000ULL | bpf_rand_u32(0);
+	case 11: return 0xf000000000000000ULL | bpf_rand_u8(0);
+	case 12: return 0x0000f00000000000ULL | bpf_rand_u8(8);
+	case 13: return 0x000000000f000000ULL | bpf_rand_u8(16);
+	case 14: return 0x0000000000000f00ULL | bpf_rand_u8(32);
+	case 15: return 0x00fff00000000f00ULL | bpf_rand_u8(48);
+	case 16: return 0x00007fffffffffffULL ^ bpf_rand_u32(1);
+	case 17: return 0xffff800000000000ULL | bpf_rand_u8(4);
+	case 18: return 0xffff800000000000ULL | bpf_rand_u8(20);
+	case 19: return (0xffffffc000000000ULL + 0x80000ULL) | bpf_rand_u32(0);
+	case 20: return (0xffffffc000000000ULL - 0x04000000ULL) | bpf_rand_u32(0);
+	case 21: return 0x0000000000000000ULL | bpf_rand_u8(55) | bpf_rand_u32(20);
+	case 22: return 0xffffffffffffffffULL ^ bpf_rand_u8(3) ^ bpf_rand_u32(40);
+	case 23: return 0x0000000000000000ULL | bpf_rand_u8(bpf_rand_u8(0) % 64);
+	case 24: return 0x0000000000000000ULL | bpf_rand_u16(bpf_rand_u8(0) % 64);
+	case 25: return 0xffffffffffffffffULL ^ bpf_rand_u8(bpf_rand_u8(0) % 64);
+	case 26: return 0xffffffffffffffffULL ^ bpf_rand_u40(bpf_rand_u8(0) % 64);
+	case 27: return 0x0000800000000000ULL;
+	case 28: return 0x8000000000000000ULL;
+	case 29: return 0x0000000000000000ULL;
+	case 30: return 0xffffffffffffffffULL;
+	case 31: return bpf_rand_u16(bpf_rand_u8(0) % 64);
+	case 32: return bpf_rand_u24(bpf_rand_u8(0) % 64);
+	case 33: return bpf_rand_u32(bpf_rand_u8(0) % 64);
+	case 34: return bpf_rand_u40(bpf_rand_u8(0) % 64);
+	case 35: return bpf_rand_u48(bpf_rand_u8(0) % 64);
+	case 36: return bpf_rand_u56(bpf_rand_u8(0) % 64);
+	case 37: return bpf_rand_u64(bpf_rand_u8(0) % 64);
+	default: return bpf_rand_u64(0);
+	}
+}
+
+#endif /* __BPF_RAND__ */

tools/testing/selftests/bpf/bpf_rlimit.h

@@ -0,0 +1,28 @@
+#include <sys/resource.h>
+#include <stdio.h>
+
+static  __attribute__((constructor)) void bpf_rlimit_ctor(void)
+{
+	struct rlimit rlim_old, rlim_new = {
+		.rlim_cur	= RLIM_INFINITY,
+		.rlim_max	= RLIM_INFINITY,
+	};
+
+	getrlimit(RLIMIT_MEMLOCK, &rlim_old);
+	/* For the sake of running the test cases, we temporarily
+	 * set rlimit to infinity in order for kernel to focus on
+	 * errors from actual test cases and not getting noise
+	 * from hitting memlock limits. The limit is on per-process
+	 * basis and not a global one, hence destructor not really
+	 * needed here.
+	 */
+	if (setrlimit(RLIMIT_MEMLOCK, &rlim_new) < 0) {
+		perror("Unable to lift memlock rlimit");
+		/* Trying out lower limit, but expect potential test
+		 * case failures from this!
+		 */
+		rlim_new.rlim_cur = rlim_old.rlim_cur + (1UL << 20);
+		rlim_new.rlim_max = rlim_old.rlim_max + (1UL << 20);
+		setrlimit(RLIMIT_MEMLOCK, &rlim_new);
+	}
+}

tools/testing/selftests/bpf/bpf_util.h

@@ -0,0 +1,61 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_UTIL__
+#define __BPF_UTIL__
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+
+static inline unsigned int bpf_num_possible_cpus(void)
+{
+	static const char *fcpu = "/sys/devices/system/cpu/possible";
+	unsigned int start, end, possible_cpus = 0;
+	char buff[128];
+	FILE *fp;
+	int len, n, i, j = 0;
+
+	fp = fopen(fcpu, "r");
+	if (!fp) {
+		printf("Failed to open %s: '%s'!\n", fcpu, strerror(errno));
+		exit(1);
+	}
+
+	if (!fgets(buff, sizeof(buff), fp)) {
+		printf("Failed to read %s!\n", fcpu);
+		exit(1);
+	}
+
+	len = strlen(buff);
+	for (i = 0; i <= len; i++) {
+		if (buff[i] == ',' || buff[i] == '\0') {
+			buff[i] = '\0';
+			n = sscanf(&buff[j], "%u-%u", &start, &end);
+			if (n <= 0) {
+				printf("Failed to retrieve # possible CPUs!\n");
+				exit(1);
+			} else if (n == 1) {
+				end = start;
+			}
+			possible_cpus += end - start + 1;
+			j = i + 1;
+		}
+	}
+
+	fclose(fp);
+
+	return possible_cpus;
+}
+
+#define __bpf_percpu_val_align	__attribute__((__aligned__(8)))
+
+#define BPF_DECLARE_PERCPU(type, name)				\
+	struct { type v; /* padding */ } __bpf_percpu_val_align	\
+		name[bpf_num_possible_cpus()]
+#define bpf_percpu(name, cpu) name[(cpu)].v
+
+#ifndef ARRAY_SIZE
+# define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+#endif
+
+#endif /* __BPF_UTIL__ */

tools/testing/selftests/bpf/cgroup_helpers.c

@@ -0,0 +1,235 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+#include <sched.h>
+#include <sys/mount.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <linux/limits.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <linux/sched.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <ftw.h>
+
+
+#include "cgroup_helpers.h"
+
+/*
+ * To avoid relying on the system setup, when setup_cgroup_env is called
+ * we create a new mount namespace, and cgroup namespace. The cgroup2
+ * root is mounted at CGROUP_MOUNT_PATH
+ *
+ * Unfortunately, most people don't have cgroupv2 enabled at this point in time.
+ * It's easier to create our own mount namespace and manage it ourselves.
+ *
+ * We assume /mnt exists.
+ */
+
+#define WALK_FD_LIMIT			16
+#define CGROUP_MOUNT_PATH		"/mnt"
+#define CGROUP_WORK_DIR			"/cgroup-test-work-dir"
+#define format_cgroup_path(buf, path) \
+	snprintf(buf, sizeof(buf), "%s%s%s", CGROUP_MOUNT_PATH, \
+		 CGROUP_WORK_DIR, path)
+
+/**
+ * setup_cgroup_environment() - Setup the cgroup environment
+ *
+ * After calling this function, cleanup_cgroup_environment should be called
+ * once testing is complete.
+ *
+ * This function will print an error to stderr and return 1 if it is unable
+ * to setup the cgroup environment. If setup is successful, 0 is returned.
+ */
+int setup_cgroup_environment(void)
+{
+	char cgroup_workdir[PATH_MAX - 24];
+
+	format_cgroup_path(cgroup_workdir, "");
+
+	if (unshare(CLONE_NEWNS)) {
+		log_err("unshare");
+		return 1;
+	}
+
+	if (mount("none", "/", NULL, MS_REC | MS_PRIVATE, NULL)) {
+		log_err("mount fakeroot");
+		return 1;
+	}
+
+	if (mount("none", CGROUP_MOUNT_PATH, "cgroup2", 0, NULL) && errno != EBUSY) {
+		log_err("mount cgroup2");
+		return 1;
+	}
+
+	/* Cleanup existing failed runs, now that the environment is setup */
+	cleanup_cgroup_environment();
+
+	if (mkdir(cgroup_workdir, 0777) && errno != EEXIST) {
+		log_err("mkdir cgroup work dir");
+		return 1;
+	}
+
+	return 0;
+}
+
+static int nftwfunc(const char *filename, const struct stat *statptr,
+		    int fileflags, struct FTW *pfwt)
+{
+	if ((fileflags & FTW_D) && rmdir(filename))
+		log_err("Removing cgroup: %s", filename);
+	return 0;
+}
+
+
+static int join_cgroup_from_top(char *cgroup_path)
+{
+	char cgroup_procs_path[PATH_MAX + 1];
+	pid_t pid = getpid();
+	int fd, rc = 0;
+
+	snprintf(cgroup_procs_path, sizeof(cgroup_procs_path),
+		 "%s/cgroup.procs", cgroup_path);
+
+	fd = open(cgroup_procs_path, O_WRONLY);
+	if (fd < 0) {
+		log_err("Opening Cgroup Procs: %s", cgroup_procs_path);
+		return 1;
+	}
+
+	if (dprintf(fd, "%d\n", pid) < 0) {
+		log_err("Joining Cgroup");
+		rc = 1;
+	}
+
+	close(fd);
+	return rc;
+}
+
+/**
+ * join_cgroup() - Join a cgroup
+ * @path: The cgroup path, relative to the workdir, to join
+ *
+ * This function expects a cgroup to already be created, relative to the cgroup
+ * work dir, and it joins it. For example, passing "/my-cgroup" as the path
+ * would actually put the calling process into the cgroup
+ * "/cgroup-test-work-dir/my-cgroup"
+ *
+ * On success, it returns 0, otherwise on failure it returns 1.
+ */
+int join_cgroup(const char *path)
+{
+	char cgroup_path[PATH_MAX + 1];
+
+	format_cgroup_path(cgroup_path, path);
+	return join_cgroup_from_top(cgroup_path);
+}
+
+/**
+ * cleanup_cgroup_environment() - Cleanup Cgroup Testing Environment
+ *
+ * This is an idempotent function to delete all temporary cgroups that
+ * have been created during the test, including the cgroup testing work
+ * directory.
+ *
+ * At call time, it moves the calling process to the root cgroup, and then
+ * runs the deletion process. It is idempotent, and should not fail, unless
+ * a process is lingering.
+ *
+ * On failure, it will print an error to stderr, and try to continue.
+ */
+void cleanup_cgroup_environment(void)
+{
+	char cgroup_workdir[PATH_MAX + 1];
+
+	format_cgroup_path(cgroup_workdir, "");
+	join_cgroup_from_top(CGROUP_MOUNT_PATH);
+	nftw(cgroup_workdir, nftwfunc, WALK_FD_LIMIT, FTW_DEPTH | FTW_MOUNT);
+}
+
+/**
+ * create_and_get_cgroup() - Create a cgroup, relative to workdir, and get the FD
+ * @path: The cgroup path, relative to the workdir, to join
+ *
+ * This function creates a cgroup under the top level workdir and returns the
+ * file descriptor. It is idempotent.
+ *
+ * On success, it returns the file descriptor. On failure it returns 0.
+ * If there is a failure, it prints the error to stderr.
+ */
+int create_and_get_cgroup(const char *path)
+{
+	char cgroup_path[PATH_MAX + 1];
+	int fd;
+
+	format_cgroup_path(cgroup_path, path);
+	if (mkdir(cgroup_path, 0777) && errno != EEXIST) {
+		log_err("mkdiring cgroup %s .. %s", path, cgroup_path);
+		return 0;
+	}
+
+	fd = open(cgroup_path, O_RDONLY);
+	if (fd < 0) {
+		log_err("Opening Cgroup");
+		return 0;
+	}
+
+	return fd;
+}
+
+/**
+ * get_cgroup_id() - Get cgroup id for a particular cgroup path
+ * @path: The cgroup path, relative to the workdir, to join
+ *
+ * On success, it returns the cgroup id. On failure it returns 0,
+ * which is an invalid cgroup id.
+ * If there is a failure, it prints the error to stderr.
+ */
+unsigned long long get_cgroup_id(const char *path)
+{
+	int dirfd, err, flags, mount_id, fhsize;
+	union {
+		unsigned long long cgid;
+		unsigned char raw_bytes[8];
+	} id;
+	char cgroup_workdir[PATH_MAX + 1];
+	struct file_handle *fhp, *fhp2;
+	unsigned long long ret = 0;
+
+	format_cgroup_path(cgroup_workdir, path);
+
+	dirfd = AT_FDCWD;
+	flags = 0;
+	fhsize = sizeof(*fhp);
+	fhp = calloc(1, fhsize);
+	if (!fhp) {
+		log_err("calloc");
+		return 0;
+	}
+	err = name_to_handle_at(dirfd, cgroup_workdir, fhp, &mount_id, flags);
+	if (err >= 0 || fhp->handle_bytes != 8) {
+		log_err("name_to_handle_at");
+		goto free_mem;
+	}
+
+	fhsize = sizeof(struct file_handle) + fhp->handle_bytes;
+	fhp2 = realloc(fhp, fhsize);
+	if (!fhp2) {
+		log_err("realloc");
+		goto free_mem;
+	}
+	err = name_to_handle_at(dirfd, cgroup_workdir, fhp2, &mount_id, flags);
+	fhp = fhp2;
+	if (err < 0) {
+		log_err("name_to_handle_at");
+		goto free_mem;
+	}
+
+	memcpy(id.raw_bytes, fhp->f_handle, 8);
+	ret = id.cgid;
+
+free_mem:
+	free(fhp);
+	return ret;
+}

tools/testing/selftests/bpf/cgroup_helpers.h

@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __CGROUP_HELPERS_H
+#define __CGROUP_HELPERS_H
+#include <errno.h>
+#include <string.h>
+
+#define clean_errno() (errno == 0 ? "None" : strerror(errno))
+#define log_err(MSG, ...) fprintf(stderr, "(%s:%d: errno: %s) " MSG "\n", \
+	__FILE__, __LINE__, clean_errno(), ##__VA_ARGS__)
+
+
+int create_and_get_cgroup(const char *path);
+int join_cgroup(const char *path);
+int setup_cgroup_environment(void);
+void cleanup_cgroup_environment(void);
+unsigned long long get_cgroup_id(const char *path);
+
+#endif

tools/testing/selftests/bpf/config

@@ -0,0 +1,20 @@
+CONFIG_BPF=y
+CONFIG_BPF_SYSCALL=y
+CONFIG_NET_CLS_BPF=m
+CONFIG_BPF_EVENTS=y
+CONFIG_TEST_BPF=m
+CONFIG_CGROUP_BPF=y
+CONFIG_NETDEVSIM=m
+CONFIG_NET_CLS_ACT=y
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_INGRESS=y
+CONFIG_NET_IPIP=y
+CONFIG_IPV6=y
+CONFIG_NET_IPGRE_DEMUX=y
+CONFIG_NET_IPGRE=y
+CONFIG_IPV6_GRE=y
+CONFIG_CRYPTO_USER_API_HASH=m
+CONFIG_CRYPTO_HMAC=m
+CONFIG_CRYPTO_SHA256=m
+CONFIG_VXLAN=y
+CONFIG_GENEVE=y

tools/testing/selftests/bpf/connect4_prog.c

@@ -0,0 +1,45 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <string.h>
+
+#include <linux/stddef.h>
+#include <linux/bpf.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <sys/socket.h>
+
+#include "bpf_helpers.h"
+#include "bpf_endian.h"
+
+#define SRC_REWRITE_IP4		0x7f000004U
+#define DST_REWRITE_IP4		0x7f000001U
+#define DST_REWRITE_PORT4	4444
+
+int _version SEC("version") = 1;
+
+SEC("cgroup/connect4")
+int connect_v4_prog(struct bpf_sock_addr *ctx)
+{
+	struct sockaddr_in sa;
+
+	/* Rewrite destination. */
+	ctx->user_ip4 = bpf_htonl(DST_REWRITE_IP4);
+	ctx->user_port = bpf_htons(DST_REWRITE_PORT4);
+
+	if (ctx->type == SOCK_DGRAM || ctx->type == SOCK_STREAM) {
+		///* Rewrite source. */
+		memset(&sa, 0, sizeof(sa));
+
+		sa.sin_family = AF_INET;
+		sa.sin_port = bpf_htons(0);
+		sa.sin_addr.s_addr = bpf_htonl(SRC_REWRITE_IP4);
+
+		if (bpf_bind(ctx, (struct sockaddr *)&sa, sizeof(sa)) != 0)
+			return 0;
+	}
+
+	return 1;
+}
+
+char _license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/connect6_prog.c

@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <string.h>
+
+#include <linux/stddef.h>
+#include <linux/bpf.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <sys/socket.h>
+
+#include "bpf_helpers.h"
+#include "bpf_endian.h"
+
+#define SRC_REWRITE_IP6_0	0
+#define SRC_REWRITE_IP6_1	0
+#define SRC_REWRITE_IP6_2	0
+#define SRC_REWRITE_IP6_3	6
+
+#define DST_REWRITE_IP6_0	0
+#define DST_REWRITE_IP6_1	0
+#define DST_REWRITE_IP6_2	0
+#define DST_REWRITE_IP6_3	1
+
+#define DST_REWRITE_PORT6	6666
+
+int _version SEC("version") = 1;
+
+SEC("cgroup/connect6")
+int connect_v6_prog(struct bpf_sock_addr *ctx)
+{
+	struct sockaddr_in6 sa;
+
+	/* Rewrite destination. */
+	ctx->user_ip6[0] = bpf_htonl(DST_REWRITE_IP6_0);
+	ctx->user_ip6[1] = bpf_htonl(DST_REWRITE_IP6_1);
+	ctx->user_ip6[2] = bpf_htonl(DST_REWRITE_IP6_2);
+	ctx->user_ip6[3] = bpf_htonl(DST_REWRITE_IP6_3);
+
+	ctx->user_port = bpf_htons(DST_REWRITE_PORT6);
+
+	if (ctx->type == SOCK_DGRAM || ctx->type == SOCK_STREAM) {
+		/* Rewrite source. */
+		memset(&sa, 0, sizeof(sa));
+
+		sa.sin6_family = AF_INET6;
+		sa.sin6_port = bpf_htons(0);
+
+		sa.sin6_addr.s6_addr32[0] = bpf_htonl(SRC_REWRITE_IP6_0);
+		sa.sin6_addr.s6_addr32[1] = bpf_htonl(SRC_REWRITE_IP6_1);
+		sa.sin6_addr.s6_addr32[2] = bpf_htonl(SRC_REWRITE_IP6_2);
+		sa.sin6_addr.s6_addr32[3] = bpf_htonl(SRC_REWRITE_IP6_3);
+
+		if (bpf_bind(ctx, (struct sockaddr *)&sa, sizeof(sa)) != 0)
+			return 0;
+	}
+
+	return 1;
+}
+
+char _license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/dev_cgroup.c

@@ -0,0 +1,60 @@
+/* Copyright (c) 2017 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+
+#include <linux/bpf.h>
+#include <linux/version.h>
+#include "bpf_helpers.h"
+
+SEC("cgroup/dev")
+int bpf_prog1(struct bpf_cgroup_dev_ctx *ctx)
+{
+	short type = ctx->access_type & 0xFFFF;
+#ifdef DEBUG
+	short access = ctx->access_type >> 16;
+	char fmt[] = "  %d:%d    \n";
+
+	switch (type) {
+	case BPF_DEVCG_DEV_BLOCK:
+		fmt[0] = 'b';
+		break;
+	case BPF_DEVCG_DEV_CHAR:
+		fmt[0] = 'c';
+		break;
+	default:
+		fmt[0] = '?';
+		break;
+	}
+
+	if (access & BPF_DEVCG_ACC_READ)
+		fmt[8] = 'r';
+
+	if (access & BPF_DEVCG_ACC_WRITE)
+		fmt[9] = 'w';
+
+	if (access & BPF_DEVCG_ACC_MKNOD)
+		fmt[10] = 'm';
+
+	bpf_trace_printk(fmt, sizeof(fmt), ctx->major, ctx->minor);
+#endif
+
+	/* Allow access to /dev/zero and /dev/random.
+	 * Forbid everything else.
+	 */
+	if (ctx->major != 1 || type != BPF_DEVCG_DEV_CHAR)
+		return 0;
+
+	switch (ctx->minor) {
+	case 5: /* 1:5 /dev/zero */
+	case 9: /* 1:9 /dev/urandom */
+		return 1;
+	}
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
+__u32 _version SEC("version") = LINUX_VERSION_CODE;

tools/testing/selftests/bpf/get_cgroup_id_kern.c

@@ -0,0 +1,40 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+struct bpf_map_def SEC("maps") cg_ids = {
+	.type = BPF_MAP_TYPE_ARRAY,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(__u64),
+	.max_entries = 1,
+};
+
+struct bpf_map_def SEC("maps") pidmap = {
+	.type = BPF_MAP_TYPE_ARRAY,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(__u32),
+	.max_entries = 1,
+};
+
+SEC("tracepoint/syscalls/sys_enter_nanosleep")
+int trace(void *ctx)
+{
+	__u32 pid = bpf_get_current_pid_tgid();
+	__u32 key = 0, *expected_pid;
+	__u64 *val;
+
+	expected_pid = bpf_map_lookup_elem(&pidmap, &key);
+	if (!expected_pid || *expected_pid != pid)
+		return 0;
+
+	val = bpf_map_lookup_elem(&cg_ids, &key);
+	if (val)
+		*val = bpf_get_current_cgroup_id();
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
+__u32 _version SEC("version") = 1; /* ignored by tracepoints, required by libbpf.a */

tools/testing/selftests/bpf/get_cgroup_id_user.c

@@ -0,0 +1,149 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <syscall.h>
+#include <unistd.h>
+#include <linux/perf_event.h>
+#include <sys/ioctl.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#include <linux/bpf.h>
+#include <bpf/bpf.h>
+#include <bpf/libbpf.h>
+
+#include "cgroup_helpers.h"
+#include "bpf_rlimit.h"
+
+#define CHECK(condition, tag, format...) ({		\
+	int __ret = !!(condition);			\
+	if (__ret) {					\
+		printf("%s:FAIL:%s ", __func__, tag);	\
+		printf(format);				\
+	} else {					\
+		printf("%s:PASS:%s\n", __func__, tag);	\
+	}						\
+	__ret;						\
+})
+
+static int bpf_find_map(const char *test, struct bpf_object *obj,
+			const char *name)
+{
+	struct bpf_map *map;
+
+	map = bpf_object__find_map_by_name(obj, name);
+	if (!map)
+		return -1;
+	return bpf_map__fd(map);
+}
+
+#define TEST_CGROUP "/test-bpf-get-cgroup-id/"
+
+int main(int argc, char **argv)
+{
+	const char *probe_name = "syscalls/sys_enter_nanosleep";
+	const char *file = "get_cgroup_id_kern.o";
+	int err, bytes, efd, prog_fd, pmu_fd;
+	int cgroup_fd, cgidmap_fd, pidmap_fd;
+	struct perf_event_attr attr = {};
+	struct bpf_object *obj;
+	__u64 kcgid = 0, ucgid;
+	__u32 key = 0, pid;
+	int exit_code = 1;
+	char buf[256];
+
+	err = setup_cgroup_environment();
+	if (CHECK(err, "setup_cgroup_environment", "err %d errno %d\n", err,
+		  errno))
+		return 1;
+
+	cgroup_fd = create_and_get_cgroup(TEST_CGROUP);
+	if (CHECK(cgroup_fd < 0, "create_and_get_cgroup", "err %d errno %d\n",
+		  cgroup_fd, errno))
+		goto cleanup_cgroup_env;
+
+	err = join_cgroup(TEST_CGROUP);
+	if (CHECK(err, "join_cgroup", "err %d errno %d\n", err, errno))
+		goto cleanup_cgroup_env;
+
+	err = bpf_prog_load(file, BPF_PROG_TYPE_TRACEPOINT, &obj, &prog_fd);
+	if (CHECK(err, "bpf_prog_load", "err %d errno %d\n", err, errno))
+		goto cleanup_cgroup_env;
+
+	cgidmap_fd = bpf_find_map(__func__, obj, "cg_ids");
+	if (CHECK(cgidmap_fd < 0, "bpf_find_map", "err %d errno %d\n",
+		  cgidmap_fd, errno))
+		goto close_prog;
+
+	pidmap_fd = bpf_find_map(__func__, obj, "pidmap");
+	if (CHECK(pidmap_fd < 0, "bpf_find_map", "err %d errno %d\n",
+		  pidmap_fd, errno))
+		goto close_prog;
+
+	pid = getpid();
+	bpf_map_update_elem(pidmap_fd, &key, &pid, 0);
+
+	snprintf(buf, sizeof(buf),
+		 "/sys/kernel/debug/tracing/events/%s/id", probe_name);
+	efd = open(buf, O_RDONLY, 0);
+	if (CHECK(efd < 0, "open", "err %d errno %d\n", efd, errno))
+		goto close_prog;
+	bytes = read(efd, buf, sizeof(buf));
+	close(efd);
+	if (CHECK(bytes <= 0 || bytes >= sizeof(buf), "read",
+		  "bytes %d errno %d\n", bytes, errno))
+		goto close_prog;
+
+	attr.config = strtol(buf, NULL, 0);
+	attr.type = PERF_TYPE_TRACEPOINT;
+	attr.sample_type = PERF_SAMPLE_RAW;
+	attr.sample_period = 1;
+	attr.wakeup_events = 1;
+
+	/* attach to this pid so the all bpf invocations will be in the
+	 * cgroup associated with this pid.
+	 */
+	pmu_fd = syscall(__NR_perf_event_open, &attr, getpid(), -1, -1, 0);
+	if (CHECK(pmu_fd < 0, "perf_event_open", "err %d errno %d\n", pmu_fd,
+		  errno))
+		goto close_prog;
+
+	err = ioctl(pmu_fd, PERF_EVENT_IOC_ENABLE, 0);
+	if (CHECK(err, "perf_event_ioc_enable", "err %d errno %d\n", err,
+		  errno))
+		goto close_pmu;
+
+	err = ioctl(pmu_fd, PERF_EVENT_IOC_SET_BPF, prog_fd);
+	if (CHECK(err, "perf_event_ioc_set_bpf", "err %d errno %d\n", err,
+		  errno))
+		goto close_pmu;
+
+	/* trigger some syscalls */
+	sleep(1);
+
+	err = bpf_map_lookup_elem(cgidmap_fd, &key, &kcgid);
+	if (CHECK(err, "bpf_map_lookup_elem", "err %d errno %d\n", err, errno))
+		goto close_pmu;
+
+	ucgid = get_cgroup_id(TEST_CGROUP);
+	if (CHECK(kcgid != ucgid, "compare_cgroup_id",
+		  "kern cgid %llx user cgid %llx", kcgid, ucgid))
+		goto close_pmu;
+
+	exit_code = 0;
+	printf("%s:PASS\n", argv[0]);
+
+close_pmu:
+	close(pmu_fd);
+close_prog:
+	bpf_object__close(obj);
+cleanup_cgroup_env:
+	cleanup_cgroup_environment();
+	return exit_code;
+}

tools/testing/selftests/bpf/gnu/stubs.h

@@ -0,0 +1 @@
+/* dummy .h to trick /usr/include/features.h to work with 'clang -target bpf' */

tools/testing/selftests/bpf/include/uapi/linux/types.h

@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _UAPI_LINUX_TYPES_H
+#define _UAPI_LINUX_TYPES_H
+
+#include <asm-generic/int-ll64.h>
+
+/* copied from linux:include/uapi/linux/types.h */
+#define __bitwise
+typedef __u16 __bitwise __le16;
+typedef __u16 __bitwise __be16;
+typedef __u32 __bitwise __le32;
+typedef __u32 __bitwise __be32;
+typedef __u64 __bitwise __le64;
+typedef __u64 __bitwise __be64;
+
+typedef __u16 __bitwise __sum16;
+typedef __u32 __bitwise __wsum;
+
+#define __aligned_u64 __u64 __attribute__((aligned(8)))
+#define __aligned_be64 __be64 __attribute__((aligned(8)))
+#define __aligned_le64 __le64 __attribute__((aligned(8)))
+
+#endif /* _UAPI_LINUX_TYPES_H */

tools/testing/selftests/bpf/sample_map_ret0.c

@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) */
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+struct bpf_map_def SEC("maps") htab = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(long),
+	.max_entries = 2,
+};
+
+struct bpf_map_def SEC("maps") array = {
+	.type = BPF_MAP_TYPE_ARRAY,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(long),
+	.max_entries = 2,
+};
+
+/* Sample program which should always load for testing control paths. */
+SEC(".text") int func()
+{
+	__u64 key64 = 0;
+	__u32 key = 0;
+	long *value;
+
+	value = bpf_map_lookup_elem(&htab, &key);
+	if (!value)
+		return 1;
+	value = bpf_map_lookup_elem(&array, &key64);
+	if (!value)
+		return 1;
+
+	return 0;
+}

tools/testing/selftests/bpf/sample_ret0.c

@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) */
+
+/* Sample program which should always load for testing control paths. */
+int func()
+{
+	return 0;
+}

tools/testing/selftests/bpf/sendmsg4_prog.c

@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <linux/stddef.h>
+#include <linux/bpf.h>
+#include <sys/socket.h>
+
+#include "bpf_helpers.h"
+#include "bpf_endian.h"
+
+#define SRC1_IP4		0xAC100001U /* 172.16.0.1 */
+#define SRC2_IP4		0x00000000U
+#define SRC_REWRITE_IP4		0x7f000004U
+#define DST_IP4			0xC0A801FEU /* 192.168.1.254 */
+#define DST_REWRITE_IP4		0x7f000001U
+#define DST_PORT		4040
+#define DST_REWRITE_PORT4	4444
+
+int _version SEC("version") = 1;
+
+SEC("cgroup/sendmsg4")
+int sendmsg_v4_prog(struct bpf_sock_addr *ctx)
+{
+	if (ctx->type != SOCK_DGRAM)
+		return 0;
+
+	/* Rewrite source. */
+	if (ctx->msg_src_ip4 == bpf_htonl(SRC1_IP4) ||
+	    ctx->msg_src_ip4 == bpf_htonl(SRC2_IP4)) {
+		ctx->msg_src_ip4 = bpf_htonl(SRC_REWRITE_IP4);
+	} else {
+		/* Unexpected source. Reject sendmsg. */
+		return 0;
+	}
+
+	/* Rewrite destination. */
+	if ((ctx->user_ip4 >> 24) == (bpf_htonl(DST_IP4) >> 24) &&
+	     ctx->user_port == bpf_htons(DST_PORT)) {
+		ctx->user_ip4 = bpf_htonl(DST_REWRITE_IP4);
+		ctx->user_port = bpf_htons(DST_REWRITE_PORT4);
+	} else {
+		/* Unexpected source. Reject sendmsg. */
+		return 0;
+	}
+
+	return 1;
+}
+
+char _license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/sendmsg6_prog.c

@@ -0,0 +1,59 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <linux/stddef.h>
+#include <linux/bpf.h>
+#include <sys/socket.h>
+
+#include "bpf_helpers.h"
+#include "bpf_endian.h"
+
+#define SRC_REWRITE_IP6_0	0
+#define SRC_REWRITE_IP6_1	0
+#define SRC_REWRITE_IP6_2	0
+#define SRC_REWRITE_IP6_3	6
+
+#define DST_REWRITE_IP6_0	0
+#define DST_REWRITE_IP6_1	0
+#define DST_REWRITE_IP6_2	0
+#define DST_REWRITE_IP6_3	1
+
+#define DST_REWRITE_PORT6	6666
+
+int _version SEC("version") = 1;
+
+SEC("cgroup/sendmsg6")
+int sendmsg_v6_prog(struct bpf_sock_addr *ctx)
+{
+	if (ctx->type != SOCK_DGRAM)
+		return 0;
+
+	/* Rewrite source. */
+	if (ctx->msg_src_ip6[3] == bpf_htonl(1) ||
+	    ctx->msg_src_ip6[3] == bpf_htonl(0)) {
+		ctx->msg_src_ip6[0] = bpf_htonl(SRC_REWRITE_IP6_0);
+		ctx->msg_src_ip6[1] = bpf_htonl(SRC_REWRITE_IP6_1);
+		ctx->msg_src_ip6[2] = bpf_htonl(SRC_REWRITE_IP6_2);
+		ctx->msg_src_ip6[3] = bpf_htonl(SRC_REWRITE_IP6_3);
+	} else {
+		/* Unexpected source. Reject sendmsg. */
+		return 0;
+	}
+
+	/* Rewrite destination. */
+	if (ctx->user_ip6[0] == bpf_htonl(0xFACEB00C)) {
+		ctx->user_ip6[0] = bpf_htonl(DST_REWRITE_IP6_0);
+		ctx->user_ip6[1] = bpf_htonl(DST_REWRITE_IP6_1);
+		ctx->user_ip6[2] = bpf_htonl(DST_REWRITE_IP6_2);
+		ctx->user_ip6[3] = bpf_htonl(DST_REWRITE_IP6_3);
+
+		ctx->user_port = bpf_htons(DST_REWRITE_PORT6);
+	} else {
+		/* Unexpected destination. Reject sendmsg. */
+		return 0;
+	}
+
+	return 1;
+}
+
+char _license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/socket_cookie_prog.c

@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Facebook
+
+#include <linux/bpf.h>
+#include <sys/socket.h>
+
+#include "bpf_helpers.h"
+#include "bpf_endian.h"
+
+struct bpf_map_def SEC("maps") socket_cookies = {
+	.type = BPF_MAP_TYPE_HASH,
+	.key_size = sizeof(__u64),
+	.value_size = sizeof(__u32),
+	.max_entries = 1 << 8,
+};
+
+SEC("cgroup/connect6")
+int set_cookie(struct bpf_sock_addr *ctx)
+{
+	__u32 cookie_value = 0xFF;
+	__u64 cookie_key;
+
+	if (ctx->family != AF_INET6 || ctx->user_family != AF_INET6)
+		return 1;
+
+	cookie_key = bpf_get_socket_cookie(ctx);
+	if (bpf_map_update_elem(&socket_cookies, &cookie_key, &cookie_value, 0))
+		return 0;
+
+	return 1;
+}
+
+SEC("sockops")
+int update_cookie(struct bpf_sock_ops *ctx)
+{
+	__u32 new_cookie_value;
+	__u32 *cookie_value;
+	__u64 cookie_key;
+
+	if (ctx->family != AF_INET6)
+		return 1;
+
+	if (ctx->op != BPF_SOCK_OPS_TCP_CONNECT_CB)
+		return 1;
+
+	cookie_key = bpf_get_socket_cookie(ctx);
+
+	cookie_value = bpf_map_lookup_elem(&socket_cookies, &cookie_key);
+	if (!cookie_value)
+		return 1;
+
+	new_cookie_value = (ctx->local_port << 8) | *cookie_value;
+	bpf_map_update_elem(&socket_cookies, &cookie_key, &new_cookie_value, 0);
+
+	return 1;
+}
+
+int _version SEC("version") = 1;
+
+char _license[] SEC("license") = "GPL";