From 2c3c1048746a4622d8c89a29670120dc8fab93c4 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Sun, 7 Apr 2024 20:49:45 +0200 Subject: Adding upstream version 6.1.76. Signed-off-by: Daniel Baumann --- tools/testing/selftests/vm/.gitignore | 35 + tools/testing/selftests/vm/Makefile | 157 ++ .../selftests/vm/charge_reserved_hugetlb.sh | 584 ++++++ tools/testing/selftests/vm/compaction_test.c | 231 +++ tools/testing/selftests/vm/config | 8 + tools/testing/selftests/vm/gup_test.c | 271 +++ tools/testing/selftests/vm/hmm-tests.c | 2054 ++++++++++++++++++++ tools/testing/selftests/vm/hugepage-mmap.c | 93 + tools/testing/selftests/vm/hugepage-mremap.c | 191 ++ tools/testing/selftests/vm/hugepage-shm.c | 101 + tools/testing/selftests/vm/hugepage-vmemmap.c | 144 ++ tools/testing/selftests/vm/hugetlb-madvise.c | 411 ++++ .../selftests/vm/hugetlb_reparenting_test.sh | 252 +++ tools/testing/selftests/vm/khugepaged.c | 1558 +++++++++++++++ tools/testing/selftests/vm/ksm_tests.c | 777 ++++++++ tools/testing/selftests/vm/madv_populate.c | 297 +++ tools/testing/selftests/vm/map_fixed_noreplace.c | 231 +++ tools/testing/selftests/vm/map_hugetlb.c | 109 ++ tools/testing/selftests/vm/map_populate.c | 113 ++ tools/testing/selftests/vm/memfd_secret.c | 299 +++ tools/testing/selftests/vm/migration.c | 193 ++ tools/testing/selftests/vm/mlock-random-test.c | 294 +++ tools/testing/selftests/vm/mlock2-tests.c | 520 +++++ tools/testing/selftests/vm/mlock2.h | 63 + tools/testing/selftests/vm/mrelease_test.c | 206 ++ tools/testing/selftests/vm/mremap_dontunmap.c | 364 ++++ tools/testing/selftests/vm/mremap_test.c | 475 +++++ tools/testing/selftests/vm/on-fault-limit.c | 48 + tools/testing/selftests/vm/pkey-helpers.h | 226 +++ tools/testing/selftests/vm/pkey-powerpc.h | 133 ++ tools/testing/selftests/vm/pkey-x86.h | 177 ++ tools/testing/selftests/vm/protection_keys.c | 1788 +++++++++++++++++ tools/testing/selftests/vm/run_vmtests.sh | 200 ++ tools/testing/selftests/vm/settings | 1 + tools/testing/selftests/vm/soft-dirty.c | 210 ++ tools/testing/selftests/vm/split_huge_page_test.c | 309 +++ tools/testing/selftests/vm/test_hmm.sh | 105 + tools/testing/selftests/vm/test_vmalloc.sh | 177 ++ tools/testing/selftests/vm/thuge-gen.c | 257 +++ tools/testing/selftests/vm/transhuge-stress.c | 122 ++ tools/testing/selftests/vm/userfaultfd.c | 1878 ++++++++++++++++++ tools/testing/selftests/vm/util.h | 69 + tools/testing/selftests/vm/va_128TBswitch.c | 289 +++ tools/testing/selftests/vm/va_128TBswitch.sh | 54 + tools/testing/selftests/vm/virtual_address_range.c | 139 ++ tools/testing/selftests/vm/vm_util.c | 126 ++ tools/testing/selftests/vm/vm_util.h | 12 + tools/testing/selftests/vm/write_hugetlb_memory.sh | 23 + tools/testing/selftests/vm/write_to_hugetlbfs.c | 240 +++ 49 files changed, 16614 insertions(+) create mode 100644 tools/testing/selftests/vm/.gitignore create mode 100644 tools/testing/selftests/vm/Makefile create mode 100644 tools/testing/selftests/vm/charge_reserved_hugetlb.sh create mode 100644 tools/testing/selftests/vm/compaction_test.c create mode 100644 tools/testing/selftests/vm/config create mode 100644 tools/testing/selftests/vm/gup_test.c create mode 100644 tools/testing/selftests/vm/hmm-tests.c create mode 100644 tools/testing/selftests/vm/hugepage-mmap.c create mode 100644 tools/testing/selftests/vm/hugepage-mremap.c create mode 100644 tools/testing/selftests/vm/hugepage-shm.c create mode 100644 tools/testing/selftests/vm/hugepage-vmemmap.c create mode 100644 tools/testing/selftests/vm/hugetlb-madvise.c create mode 100644 tools/testing/selftests/vm/hugetlb_reparenting_test.sh create mode 100644 tools/testing/selftests/vm/khugepaged.c create mode 100644 tools/testing/selftests/vm/ksm_tests.c create mode 100644 tools/testing/selftests/vm/madv_populate.c create mode 100644 tools/testing/selftests/vm/map_fixed_noreplace.c create mode 100644 tools/testing/selftests/vm/map_hugetlb.c create mode 100644 tools/testing/selftests/vm/map_populate.c create mode 100644 tools/testing/selftests/vm/memfd_secret.c create mode 100644 tools/testing/selftests/vm/migration.c create mode 100644 tools/testing/selftests/vm/mlock-random-test.c create mode 100644 tools/testing/selftests/vm/mlock2-tests.c create mode 100644 tools/testing/selftests/vm/mlock2.h create mode 100644 tools/testing/selftests/vm/mrelease_test.c create mode 100644 tools/testing/selftests/vm/mremap_dontunmap.c create mode 100644 tools/testing/selftests/vm/mremap_test.c create mode 100644 tools/testing/selftests/vm/on-fault-limit.c create mode 100644 tools/testing/selftests/vm/pkey-helpers.h create mode 100644 tools/testing/selftests/vm/pkey-powerpc.h create mode 100644 tools/testing/selftests/vm/pkey-x86.h create mode 100644 tools/testing/selftests/vm/protection_keys.c create mode 100755 tools/testing/selftests/vm/run_vmtests.sh create mode 100644 tools/testing/selftests/vm/settings create mode 100644 tools/testing/selftests/vm/soft-dirty.c create mode 100644 tools/testing/selftests/vm/split_huge_page_test.c create mode 100755 tools/testing/selftests/vm/test_hmm.sh create mode 100755 tools/testing/selftests/vm/test_vmalloc.sh create mode 100644 tools/testing/selftests/vm/thuge-gen.c create mode 100644 tools/testing/selftests/vm/transhuge-stress.c create mode 100644 tools/testing/selftests/vm/userfaultfd.c create mode 100644 tools/testing/selftests/vm/util.h create mode 100644 tools/testing/selftests/vm/va_128TBswitch.c create mode 100755 tools/testing/selftests/vm/va_128TBswitch.sh create mode 100644 tools/testing/selftests/vm/virtual_address_range.c create mode 100644 tools/testing/selftests/vm/vm_util.c create mode 100644 tools/testing/selftests/vm/vm_util.h create mode 100644 tools/testing/selftests/vm/write_hugetlb_memory.sh create mode 100644 tools/testing/selftests/vm/write_to_hugetlbfs.c (limited to 'tools/testing/selftests/vm') diff --git a/tools/testing/selftests/vm/.gitignore b/tools/testing/selftests/vm/.gitignore new file mode 100644 index 000000000..7b9dc2426 --- /dev/null +++ b/tools/testing/selftests/vm/.gitignore @@ -0,0 +1,35 @@ +# SPDX-License-Identifier: GPL-2.0-only +hugepage-mmap +hugepage-mremap +hugepage-shm +hugepage-vmemmap +hugetlb-madvise +khugepaged +map_hugetlb +map_populate +thuge-gen +compaction_test +migration +mlock2-tests +mrelease_test +mremap_dontunmap +mremap_test +on-fault-limit +transhuge-stress +protection_keys +protection_keys_32 +protection_keys_64 +madv_populate +userfaultfd +mlock-intersect-test +mlock-random-test +virtual_address_range +gup_test +va_128TBswitch +map_fixed_noreplace +write_to_hugetlbfs +hmm-tests +memfd_secret +soft-dirty +split_huge_page_test +ksm_tests diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile new file mode 100644 index 000000000..192ea3725 --- /dev/null +++ b/tools/testing/selftests/vm/Makefile @@ -0,0 +1,157 @@ +# SPDX-License-Identifier: GPL-2.0 +# Makefile for vm selftests + +LOCAL_HDRS += $(top_srcdir)/mm/gup_test.h + +uname_M := $(shell uname -m 2>/dev/null || echo not) +MACHINE ?= $(shell echo $(uname_M) | sed -e 's/aarch64.*/arm64/' -e 's/ppc64.*/ppc64/') + +# Without this, failed build products remain, with up-to-date timestamps, +# thus tricking Make (and you!) into believing that All Is Well, in subsequent +# make invocations: +.DELETE_ON_ERROR: + +# Avoid accidental wrong builds, due to built-in rules working just a little +# bit too well--but not quite as well as required for our situation here. +# +# In other words, "make userfaultfd" is supposed to fail to build at all, +# because this Makefile only supports either "make" (all), or "make /full/path". +# However, the built-in rules, if not suppressed, will pick up CFLAGS and the +# initial LDLIBS (but not the target-specific LDLIBS, because those are only +# set for the full path target!). This causes it to get pretty far into building +# things despite using incorrect values such as an *occasionally* incomplete +# LDLIBS. +MAKEFLAGS += --no-builtin-rules + +CFLAGS = -Wall -I $(top_srcdir) $(EXTRA_CFLAGS) $(KHDR_INCLUDES) +LDLIBS = -lrt -lpthread +TEST_GEN_FILES = compaction_test +TEST_GEN_FILES += gup_test +TEST_GEN_FILES += hmm-tests +TEST_GEN_FILES += hugetlb-madvise +TEST_GEN_FILES += hugepage-mmap +TEST_GEN_FILES += hugepage-mremap +TEST_GEN_FILES += hugepage-shm +TEST_GEN_FILES += hugepage-vmemmap +TEST_GEN_FILES += khugepaged +TEST_GEN_PROGS = madv_populate +TEST_GEN_FILES += map_fixed_noreplace +TEST_GEN_FILES += map_hugetlb +TEST_GEN_FILES += map_populate +TEST_GEN_FILES += memfd_secret +TEST_GEN_FILES += migration +TEST_GEN_FILES += mlock-random-test +TEST_GEN_FILES += mlock2-tests +TEST_GEN_FILES += mrelease_test +TEST_GEN_FILES += mremap_dontunmap +TEST_GEN_FILES += mremap_test +TEST_GEN_FILES += on-fault-limit +TEST_GEN_FILES += thuge-gen +TEST_GEN_FILES += transhuge-stress +TEST_GEN_FILES += userfaultfd +TEST_GEN_PROGS += soft-dirty +TEST_GEN_PROGS += split_huge_page_test +TEST_GEN_FILES += ksm_tests + +ifeq ($(MACHINE),x86_64) +CAN_BUILD_I386 := $(shell ./../x86/check_cc.sh "$(CC)" ../x86/trivial_32bit_program.c -m32) +CAN_BUILD_X86_64 := $(shell ./../x86/check_cc.sh "$(CC)" ../x86/trivial_64bit_program.c) +CAN_BUILD_WITH_NOPIE := $(shell ./../x86/check_cc.sh "$(CC)" ../x86/trivial_program.c -no-pie) + +VMTARGETS := protection_keys +BINARIES_32 := $(VMTARGETS:%=%_32) +BINARIES_64 := $(VMTARGETS:%=%_64) + +ifeq ($(CAN_BUILD_WITH_NOPIE),1) +CFLAGS += -no-pie +endif + +ifeq ($(CAN_BUILD_I386),1) +TEST_GEN_FILES += $(BINARIES_32) +endif + +ifeq ($(CAN_BUILD_X86_64),1) +TEST_GEN_FILES += $(BINARIES_64) +endif +else + +ifneq (,$(findstring $(MACHINE),ppc64)) +TEST_GEN_FILES += protection_keys +endif + +endif + +ifneq (,$(filter $(MACHINE),arm64 ia64 mips64 parisc64 ppc64 riscv64 s390x sh64 sparc64 x86_64)) +TEST_GEN_FILES += va_128TBswitch +TEST_GEN_FILES += virtual_address_range +TEST_GEN_FILES += write_to_hugetlbfs +endif + +TEST_PROGS := run_vmtests.sh + +TEST_FILES := test_vmalloc.sh +TEST_FILES += test_hmm.sh +TEST_FILES += va_128TBswitch.sh + +include ../lib.mk + +$(OUTPUT)/khugepaged: vm_util.c +$(OUTPUT)/madv_populate: vm_util.c +$(OUTPUT)/soft-dirty: vm_util.c +$(OUTPUT)/split_huge_page_test: vm_util.c +$(OUTPUT)/userfaultfd: vm_util.c + +ifeq ($(MACHINE),x86_64) +BINARIES_32 := $(patsubst %,$(OUTPUT)/%,$(BINARIES_32)) +BINARIES_64 := $(patsubst %,$(OUTPUT)/%,$(BINARIES_64)) + +define gen-target-rule-32 +$(1) $(1)_32: $(OUTPUT)/$(1)_32 +.PHONY: $(1) $(1)_32 +endef + +define gen-target-rule-64 +$(1) $(1)_64: $(OUTPUT)/$(1)_64 +.PHONY: $(1) $(1)_64 +endef + +ifeq ($(CAN_BUILD_I386),1) +$(BINARIES_32): CFLAGS += -m32 -mxsave +$(BINARIES_32): LDLIBS += -lrt -ldl -lm +$(BINARIES_32): $(OUTPUT)/%_32: %.c + $(CC) $(CFLAGS) $(EXTRA_CFLAGS) $(notdir $^) $(LDLIBS) -o $@ +$(foreach t,$(VMTARGETS),$(eval $(call gen-target-rule-32,$(t)))) +endif + +ifeq ($(CAN_BUILD_X86_64),1) +$(BINARIES_64): CFLAGS += -m64 -mxsave +$(BINARIES_64): LDLIBS += -lrt -ldl +$(BINARIES_64): $(OUTPUT)/%_64: %.c + $(CC) $(CFLAGS) $(EXTRA_CFLAGS) $(notdir $^) $(LDLIBS) -o $@ +$(foreach t,$(VMTARGETS),$(eval $(call gen-target-rule-64,$(t)))) +endif + +# x86_64 users should be encouraged to install 32-bit libraries +ifeq ($(CAN_BUILD_I386)$(CAN_BUILD_X86_64),01) +all: warn_32bit_failure + +warn_32bit_failure: + @echo "Warning: you seem to have a broken 32-bit build" 2>&1; \ + echo "environment. This will reduce test coverage of 64-bit" 2>&1; \ + echo "kernels. If you are using a Debian-like distribution," 2>&1; \ + echo "try:"; 2>&1; \ + echo ""; \ + echo " apt-get install gcc-multilib libc6-i386 libc6-dev-i386"; \ + echo ""; \ + echo "If you are using a Fedora-like distribution, try:"; \ + echo ""; \ + echo " yum install glibc-devel.*i686"; \ + exit 0; +endif +endif + +$(OUTPUT)/mlock-random-test $(OUTPUT)/memfd_secret: LDLIBS += -lcap + +$(OUTPUT)/ksm_tests: LDLIBS += -lnuma + +$(OUTPUT)/migration: LDLIBS += -lnuma diff --git a/tools/testing/selftests/vm/charge_reserved_hugetlb.sh b/tools/testing/selftests/vm/charge_reserved_hugetlb.sh new file mode 100644 index 000000000..0899019a7 --- /dev/null +++ b/tools/testing/selftests/vm/charge_reserved_hugetlb.sh @@ -0,0 +1,584 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +set -e + +if [[ $(id -u) -ne 0 ]]; then + echo "This test must be run as root. Skipping..." + exit $ksft_skip +fi + +fault_limit_file=limit_in_bytes +reservation_limit_file=rsvd.limit_in_bytes +fault_usage_file=usage_in_bytes +reservation_usage_file=rsvd.usage_in_bytes + +if [[ "$1" == "-cgroup-v2" ]]; then + cgroup2=1 + fault_limit_file=max + reservation_limit_file=rsvd.max + fault_usage_file=current + reservation_usage_file=rsvd.current +fi + +if [[ $cgroup2 ]]; then + cgroup_path=$(mount -t cgroup2 | head -1 | awk '{print $3}') + if [[ -z "$cgroup_path" ]]; then + cgroup_path=/dev/cgroup/memory + mount -t cgroup2 none $cgroup_path + do_umount=1 + fi + echo "+hugetlb" >$cgroup_path/cgroup.subtree_control +else + cgroup_path=$(mount -t cgroup | grep ",hugetlb" | awk '{print $3}') + if [[ -z "$cgroup_path" ]]; then + cgroup_path=/dev/cgroup/memory + mount -t cgroup memory,hugetlb $cgroup_path + do_umount=1 + fi +fi +export cgroup_path + +function cleanup() { + if [[ $cgroup2 ]]; then + echo $$ >$cgroup_path/cgroup.procs + else + echo $$ >$cgroup_path/tasks + fi + + if [[ -e /mnt/huge ]]; then + rm -rf /mnt/huge/* + umount /mnt/huge || echo error + rmdir /mnt/huge + fi + if [[ -e $cgroup_path/hugetlb_cgroup_test ]]; then + rmdir $cgroup_path/hugetlb_cgroup_test + fi + if [[ -e $cgroup_path/hugetlb_cgroup_test1 ]]; then + rmdir $cgroup_path/hugetlb_cgroup_test1 + fi + if [[ -e $cgroup_path/hugetlb_cgroup_test2 ]]; then + rmdir $cgroup_path/hugetlb_cgroup_test2 + fi + echo 0 >/proc/sys/vm/nr_hugepages + echo CLEANUP DONE +} + +function expect_equal() { + local expected="$1" + local actual="$2" + local error="$3" + + if [[ "$expected" != "$actual" ]]; then + echo "expected ($expected) != actual ($actual): $3" + cleanup + exit 1 + fi +} + +function get_machine_hugepage_size() { + hpz=$(grep -i hugepagesize /proc/meminfo) + kb=${hpz:14:-3} + mb=$(($kb / 1024)) + echo $mb +} + +MB=$(get_machine_hugepage_size) + +function setup_cgroup() { + local name="$1" + local cgroup_limit="$2" + local reservation_limit="$3" + + mkdir $cgroup_path/$name + + echo writing cgroup limit: "$cgroup_limit" + echo "$cgroup_limit" >$cgroup_path/$name/hugetlb.${MB}MB.$fault_limit_file + + echo writing reseravation limit: "$reservation_limit" + echo "$reservation_limit" > \ + $cgroup_path/$name/hugetlb.${MB}MB.$reservation_limit_file + + if [ -e "$cgroup_path/$name/cpuset.cpus" ]; then + echo 0 >$cgroup_path/$name/cpuset.cpus + fi + if [ -e "$cgroup_path/$name/cpuset.mems" ]; then + echo 0 >$cgroup_path/$name/cpuset.mems + fi +} + +function wait_for_hugetlb_memory_to_get_depleted() { + local cgroup="$1" + local path="$cgroup_path/$cgroup/hugetlb.${MB}MB.$reservation_usage_file" + # Wait for hugetlbfs memory to get depleted. + while [ $(cat $path) != 0 ]; do + echo Waiting for hugetlb memory to get depleted. + cat $path + sleep 0.5 + done +} + +function wait_for_hugetlb_memory_to_get_reserved() { + local cgroup="$1" + local size="$2" + + local path="$cgroup_path/$cgroup/hugetlb.${MB}MB.$reservation_usage_file" + # Wait for hugetlbfs memory to get written. + while [ $(cat $path) != $size ]; do + echo Waiting for hugetlb memory reservation to reach size $size. + cat $path + sleep 0.5 + done +} + +function wait_for_hugetlb_memory_to_get_written() { + local cgroup="$1" + local size="$2" + + local path="$cgroup_path/$cgroup/hugetlb.${MB}MB.$fault_usage_file" + # Wait for hugetlbfs memory to get written. + while [ $(cat $path) != $size ]; do + echo Waiting for hugetlb memory to reach size $size. + cat $path + sleep 0.5 + done +} + +function write_hugetlbfs_and_get_usage() { + local cgroup="$1" + local size="$2" + local populate="$3" + local write="$4" + local path="$5" + local method="$6" + local private="$7" + local expect_failure="$8" + local reserve="$9" + + # Function return values. + reservation_failed=0 + oom_killed=0 + hugetlb_difference=0 + reserved_difference=0 + + local hugetlb_usage=$cgroup_path/$cgroup/hugetlb.${MB}MB.$fault_usage_file + local reserved_usage=$cgroup_path/$cgroup/hugetlb.${MB}MB.$reservation_usage_file + + local hugetlb_before=$(cat $hugetlb_usage) + local reserved_before=$(cat $reserved_usage) + + echo + echo Starting: + echo hugetlb_usage="$hugetlb_before" + echo reserved_usage="$reserved_before" + echo expect_failure is "$expect_failure" + + output=$(mktemp) + set +e + if [[ "$method" == "1" ]] || [[ "$method" == 2 ]] || + [[ "$private" == "-r" ]] && [[ "$expect_failure" != 1 ]]; then + + bash write_hugetlb_memory.sh "$size" "$populate" "$write" \ + "$cgroup" "$path" "$method" "$private" "-l" "$reserve" 2>&1 | tee $output & + + local write_result=$? + local write_pid=$! + + until grep -q -i "DONE" $output; do + echo waiting for DONE signal. + if ! ps $write_pid > /dev/null + then + echo "FAIL: The write died" + cleanup + exit 1 + fi + sleep 0.5 + done + + echo ================= write_hugetlb_memory.sh output is: + cat $output + echo ================= end output. + + if [[ "$populate" == "-o" ]] || [[ "$write" == "-w" ]]; then + wait_for_hugetlb_memory_to_get_written "$cgroup" "$size" + elif [[ "$reserve" != "-n" ]]; then + wait_for_hugetlb_memory_to_get_reserved "$cgroup" "$size" + else + # This case doesn't produce visible effects, but we still have + # to wait for the async process to start and execute... + sleep 0.5 + fi + + echo write_result is $write_result + else + bash write_hugetlb_memory.sh "$size" "$populate" "$write" \ + "$cgroup" "$path" "$method" "$private" "$reserve" + local write_result=$? + + if [[ "$reserve" != "-n" ]]; then + wait_for_hugetlb_memory_to_get_reserved "$cgroup" "$size" + fi + fi + set -e + + if [[ "$write_result" == 1 ]]; then + reservation_failed=1 + fi + + # On linus/master, the above process gets SIGBUS'd on oomkill, with + # return code 135. On earlier kernels, it gets actual oomkill, with return + # code 137, so just check for both conditions in case we're testing + # against an earlier kernel. + if [[ "$write_result" == 135 ]] || [[ "$write_result" == 137 ]]; then + oom_killed=1 + fi + + local hugetlb_after=$(cat $hugetlb_usage) + local reserved_after=$(cat $reserved_usage) + + echo After write: + echo hugetlb_usage="$hugetlb_after" + echo reserved_usage="$reserved_after" + + hugetlb_difference=$(($hugetlb_after - $hugetlb_before)) + reserved_difference=$(($reserved_after - $reserved_before)) +} + +function cleanup_hugetlb_memory() { + set +e + local cgroup="$1" + if [[ "$(pgrep -f write_to_hugetlbfs)" != "" ]]; then + echo killing write_to_hugetlbfs + killall -2 write_to_hugetlbfs + wait_for_hugetlb_memory_to_get_depleted $cgroup + fi + set -e + + if [[ -e /mnt/huge ]]; then + rm -rf /mnt/huge/* + umount /mnt/huge + rmdir /mnt/huge + fi +} + +function run_test() { + local size=$(($1 * ${MB} * 1024 * 1024)) + local populate="$2" + local write="$3" + local cgroup_limit=$(($4 * ${MB} * 1024 * 1024)) + local reservation_limit=$(($5 * ${MB} * 1024 * 1024)) + local nr_hugepages="$6" + local method="$7" + local private="$8" + local expect_failure="$9" + local reserve="${10}" + + # Function return values. + hugetlb_difference=0 + reserved_difference=0 + reservation_failed=0 + oom_killed=0 + + echo nr hugepages = "$nr_hugepages" + echo "$nr_hugepages" >/proc/sys/vm/nr_hugepages + + setup_cgroup "hugetlb_cgroup_test" "$cgroup_limit" "$reservation_limit" + + mkdir -p /mnt/huge + mount -t hugetlbfs -o pagesize=${MB}M,size=256M none /mnt/huge + + write_hugetlbfs_and_get_usage "hugetlb_cgroup_test" "$size" "$populate" \ + "$write" "/mnt/huge/test" "$method" "$private" "$expect_failure" \ + "$reserve" + + cleanup_hugetlb_memory "hugetlb_cgroup_test" + + local final_hugetlb=$(cat $cgroup_path/hugetlb_cgroup_test/hugetlb.${MB}MB.$fault_usage_file) + local final_reservation=$(cat $cgroup_path/hugetlb_cgroup_test/hugetlb.${MB}MB.$reservation_usage_file) + + echo $hugetlb_difference + echo $reserved_difference + expect_equal "0" "$final_hugetlb" "final hugetlb is not zero" + expect_equal "0" "$final_reservation" "final reservation is not zero" +} + +function run_multiple_cgroup_test() { + local size1="$1" + local populate1="$2" + local write1="$3" + local cgroup_limit1="$4" + local reservation_limit1="$5" + + local size2="$6" + local populate2="$7" + local write2="$8" + local cgroup_limit2="$9" + local reservation_limit2="${10}" + + local nr_hugepages="${11}" + local method="${12}" + local private="${13}" + local expect_failure="${14}" + local reserve="${15}" + + # Function return values. + hugetlb_difference1=0 + reserved_difference1=0 + reservation_failed1=0 + oom_killed1=0 + + hugetlb_difference2=0 + reserved_difference2=0 + reservation_failed2=0 + oom_killed2=0 + + echo nr hugepages = "$nr_hugepages" + echo "$nr_hugepages" >/proc/sys/vm/nr_hugepages + + setup_cgroup "hugetlb_cgroup_test1" "$cgroup_limit1" "$reservation_limit1" + setup_cgroup "hugetlb_cgroup_test2" "$cgroup_limit2" "$reservation_limit2" + + mkdir -p /mnt/huge + mount -t hugetlbfs -o pagesize=${MB}M,size=256M none /mnt/huge + + write_hugetlbfs_and_get_usage "hugetlb_cgroup_test1" "$size1" \ + "$populate1" "$write1" "/mnt/huge/test1" "$method" "$private" \ + "$expect_failure" "$reserve" + + hugetlb_difference1=$hugetlb_difference + reserved_difference1=$reserved_difference + reservation_failed1=$reservation_failed + oom_killed1=$oom_killed + + local cgroup1_hugetlb_usage=$cgroup_path/hugetlb_cgroup_test1/hugetlb.${MB}MB.$fault_usage_file + local cgroup1_reservation_usage=$cgroup_path/hugetlb_cgroup_test1/hugetlb.${MB}MB.$reservation_usage_file + local cgroup2_hugetlb_usage=$cgroup_path/hugetlb_cgroup_test2/hugetlb.${MB}MB.$fault_usage_file + local cgroup2_reservation_usage=$cgroup_path/hugetlb_cgroup_test2/hugetlb.${MB}MB.$reservation_usage_file + + local usage_before_second_write=$(cat $cgroup1_hugetlb_usage) + local reservation_usage_before_second_write=$(cat $cgroup1_reservation_usage) + + write_hugetlbfs_and_get_usage "hugetlb_cgroup_test2" "$size2" \ + "$populate2" "$write2" "/mnt/huge/test2" "$method" "$private" \ + "$expect_failure" "$reserve" + + hugetlb_difference2=$hugetlb_difference + reserved_difference2=$reserved_difference + reservation_failed2=$reservation_failed + oom_killed2=$oom_killed + + expect_equal "$usage_before_second_write" \ + "$(cat $cgroup1_hugetlb_usage)" "Usage changed." + expect_equal "$reservation_usage_before_second_write" \ + "$(cat $cgroup1_reservation_usage)" "Reservation usage changed." + + cleanup_hugetlb_memory + + local final_hugetlb=$(cat $cgroup1_hugetlb_usage) + local final_reservation=$(cat $cgroup1_reservation_usage) + + expect_equal "0" "$final_hugetlb" \ + "hugetlbt_cgroup_test1 final hugetlb is not zero" + expect_equal "0" "$final_reservation" \ + "hugetlbt_cgroup_test1 final reservation is not zero" + + local final_hugetlb=$(cat $cgroup2_hugetlb_usage) + local final_reservation=$(cat $cgroup2_reservation_usage) + + expect_equal "0" "$final_hugetlb" \ + "hugetlb_cgroup_test2 final hugetlb is not zero" + expect_equal "0" "$final_reservation" \ + "hugetlb_cgroup_test2 final reservation is not zero" +} + +cleanup + +for populate in "" "-o"; do + for method in 0 1 2; do + for private in "" "-r"; do + for reserve in "" "-n"; do + + # Skip mmap(MAP_HUGETLB | MAP_SHARED). Doesn't seem to be supported. + if [[ "$method" == 1 ]] && [[ "$private" == "" ]]; then + continue + fi + + # Skip populated shmem tests. Doesn't seem to be supported. + if [[ "$method" == 2"" ]] && [[ "$populate" == "-o" ]]; then + continue + fi + + if [[ "$method" == 2"" ]] && [[ "$reserve" == "-n" ]]; then + continue + fi + + cleanup + echo + echo + echo + echo Test normal case. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + run_test 5 "$populate" "" 10 10 10 "$method" "$private" "0" "$reserve" + + echo Memory charged to hugtlb=$hugetlb_difference + echo Memory charged to reservation=$reserved_difference + + if [[ "$populate" == "-o" ]]; then + expect_equal "$((5 * $MB * 1024 * 1024))" "$hugetlb_difference" \ + "Reserved memory charged to hugetlb cgroup." + else + expect_equal "0" "$hugetlb_difference" \ + "Reserved memory charged to hugetlb cgroup." + fi + + if [[ "$reserve" != "-n" ]] || [[ "$populate" == "-o" ]]; then + expect_equal "$((5 * $MB * 1024 * 1024))" "$reserved_difference" \ + "Reserved memory not charged to reservation usage." + else + expect_equal "0" "$reserved_difference" \ + "Reserved memory not charged to reservation usage." + fi + + echo 'PASS' + + cleanup + echo + echo + echo + echo Test normal case with write. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + run_test 5 "$populate" '-w' 5 5 10 "$method" "$private" "0" "$reserve" + + echo Memory charged to hugtlb=$hugetlb_difference + echo Memory charged to reservation=$reserved_difference + + expect_equal "$((5 * $MB * 1024 * 1024))" "$hugetlb_difference" \ + "Reserved memory charged to hugetlb cgroup." + + expect_equal "$((5 * $MB * 1024 * 1024))" "$reserved_difference" \ + "Reserved memory not charged to reservation usage." + + echo 'PASS' + + cleanup + continue + echo + echo + echo + echo Test more than reservation case. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + + if [ "$reserve" != "-n" ]; then + run_test "5" "$populate" '' "10" "2" "10" "$method" "$private" "1" \ + "$reserve" + + expect_equal "1" "$reservation_failed" "Reservation succeeded." + fi + + echo 'PASS' + + cleanup + + echo + echo + echo + echo Test more than cgroup limit case. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + + # Not sure if shm memory can be cleaned up when the process gets sigbus'd. + if [[ "$method" != 2 ]]; then + run_test 5 "$populate" "-w" 2 10 10 "$method" "$private" "1" "$reserve" + + expect_equal "1" "$oom_killed" "Not oom killed." + fi + echo 'PASS' + + cleanup + + echo + echo + echo + echo Test normal case, multiple cgroups. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + run_multiple_cgroup_test "3" "$populate" "" "10" "10" "5" \ + "$populate" "" "10" "10" "10" \ + "$method" "$private" "0" "$reserve" + + echo Memory charged to hugtlb1=$hugetlb_difference1 + echo Memory charged to reservation1=$reserved_difference1 + echo Memory charged to hugtlb2=$hugetlb_difference2 + echo Memory charged to reservation2=$reserved_difference2 + + if [[ "$reserve" != "-n" ]] || [[ "$populate" == "-o" ]]; then + expect_equal "3" "$reserved_difference1" \ + "Incorrect reservations charged to cgroup 1." + + expect_equal "5" "$reserved_difference2" \ + "Incorrect reservation charged to cgroup 2." + + else + expect_equal "0" "$reserved_difference1" \ + "Incorrect reservations charged to cgroup 1." + + expect_equal "0" "$reserved_difference2" \ + "Incorrect reservation charged to cgroup 2." + fi + + if [[ "$populate" == "-o" ]]; then + expect_equal "3" "$hugetlb_difference1" \ + "Incorrect hugetlb charged to cgroup 1." + + expect_equal "5" "$hugetlb_difference2" \ + "Incorrect hugetlb charged to cgroup 2." + + else + expect_equal "0" "$hugetlb_difference1" \ + "Incorrect hugetlb charged to cgroup 1." + + expect_equal "0" "$hugetlb_difference2" \ + "Incorrect hugetlb charged to cgroup 2." + fi + echo 'PASS' + + cleanup + echo + echo + echo + echo Test normal case with write, multiple cgroups. + echo private=$private, populate=$populate, method=$method, reserve=$reserve + run_multiple_cgroup_test "3" "$populate" "-w" "10" "10" "5" \ + "$populate" "-w" "10" "10" "10" \ + "$method" "$private" "0" "$reserve" + + echo Memory charged to hugtlb1=$hugetlb_difference1 + echo Memory charged to reservation1=$reserved_difference1 + echo Memory charged to hugtlb2=$hugetlb_difference2 + echo Memory charged to reservation2=$reserved_difference2 + + expect_equal "3" "$hugetlb_difference1" \ + "Incorrect hugetlb charged to cgroup 1." + + expect_equal "3" "$reserved_difference1" \ + "Incorrect reservation charged to cgroup 1." + + expect_equal "5" "$hugetlb_difference2" \ + "Incorrect hugetlb charged to cgroup 2." + + expect_equal "5" "$reserved_difference2" \ + "Incorrected reservation charged to cgroup 2." + echo 'PASS' + + cleanup + + done # reserve + done # private + done # populate +done # method + +if [[ $do_umount ]]; then + umount $cgroup_path + rmdir $cgroup_path +fi diff --git a/tools/testing/selftests/vm/compaction_test.c b/tools/testing/selftests/vm/compaction_test.c new file mode 100644 index 000000000..9b420140b --- /dev/null +++ b/tools/testing/selftests/vm/compaction_test.c @@ -0,0 +1,231 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * + * A test for the patch "Allow compaction of unevictable pages". + * With this patch we should be able to allocate at least 1/4 + * of RAM in huge pages. Without the patch much less is + * allocated. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +#define MAP_SIZE_MB 100 +#define MAP_SIZE (MAP_SIZE_MB * 1024 * 1024) + +struct map_list { + void *map; + struct map_list *next; +}; + +int read_memory_info(unsigned long *memfree, unsigned long *hugepagesize) +{ + char buffer[256] = {0}; + char *cmd = "cat /proc/meminfo | grep -i memfree | grep -o '[0-9]*'"; + FILE *cmdfile = popen(cmd, "r"); + + if (!(fgets(buffer, sizeof(buffer), cmdfile))) { + perror("Failed to read meminfo\n"); + return -1; + } + + pclose(cmdfile); + + *memfree = atoll(buffer); + cmd = "cat /proc/meminfo | grep -i hugepagesize | grep -o '[0-9]*'"; + cmdfile = popen(cmd, "r"); + + if (!(fgets(buffer, sizeof(buffer), cmdfile))) { + perror("Failed to read meminfo\n"); + return -1; + } + + pclose(cmdfile); + *hugepagesize = atoll(buffer); + + return 0; +} + +int prereq(void) +{ + char allowed; + int fd; + + fd = open("/proc/sys/vm/compact_unevictable_allowed", + O_RDONLY | O_NONBLOCK); + if (fd < 0) { + perror("Failed to open\n" + "/proc/sys/vm/compact_unevictable_allowed\n"); + return -1; + } + + if (read(fd, &allowed, sizeof(char)) != sizeof(char)) { + perror("Failed to read from\n" + "/proc/sys/vm/compact_unevictable_allowed\n"); + close(fd); + return -1; + } + + close(fd); + if (allowed == '1') + return 0; + + return -1; +} + +int check_compaction(unsigned long mem_free, unsigned int hugepage_size) +{ + int fd; + int compaction_index = 0; + char initial_nr_hugepages[10] = {0}; + char nr_hugepages[10] = {0}; + + /* We want to test with 80% of available memory. Else, OOM killer comes + in to play */ + mem_free = mem_free * 0.8; + + fd = open("/proc/sys/vm/nr_hugepages", O_RDWR | O_NONBLOCK); + if (fd < 0) { + perror("Failed to open /proc/sys/vm/nr_hugepages"); + return -1; + } + + if (read(fd, initial_nr_hugepages, sizeof(initial_nr_hugepages)) <= 0) { + perror("Failed to read from /proc/sys/vm/nr_hugepages"); + goto close_fd; + } + + /* Start with the initial condition of 0 huge pages*/ + if (write(fd, "0", sizeof(char)) != sizeof(char)) { + perror("Failed to write 0 to /proc/sys/vm/nr_hugepages\n"); + goto close_fd; + } + + lseek(fd, 0, SEEK_SET); + + /* Request a large number of huge pages. The Kernel will allocate + as much as it can */ + if (write(fd, "100000", (6*sizeof(char))) != (6*sizeof(char))) { + perror("Failed to write 100000 to /proc/sys/vm/nr_hugepages\n"); + goto close_fd; + } + + lseek(fd, 0, SEEK_SET); + + if (read(fd, nr_hugepages, sizeof(nr_hugepages)) <= 0) { + perror("Failed to re-read from /proc/sys/vm/nr_hugepages\n"); + goto close_fd; + } + + /* We should have been able to request at least 1/3 rd of the memory in + huge pages */ + compaction_index = mem_free/(atoi(nr_hugepages) * hugepage_size); + + if (compaction_index > 3) { + printf("No of huge pages allocated = %d\n", + (atoi(nr_hugepages))); + fprintf(stderr, "ERROR: Less that 1/%d of memory is available\n" + "as huge pages\n", compaction_index); + goto close_fd; + } + + printf("No of huge pages allocated = %d\n", + (atoi(nr_hugepages))); + + lseek(fd, 0, SEEK_SET); + + if (write(fd, initial_nr_hugepages, strlen(initial_nr_hugepages)) + != strlen(initial_nr_hugepages)) { + perror("Failed to write value to /proc/sys/vm/nr_hugepages\n"); + goto close_fd; + } + + close(fd); + return 0; + + close_fd: + close(fd); + printf("Not OK. Compaction test failed."); + return -1; +} + + +int main(int argc, char **argv) +{ + struct rlimit lim; + struct map_list *list, *entry; + size_t page_size, i; + void *map = NULL; + unsigned long mem_free = 0; + unsigned long hugepage_size = 0; + long mem_fragmentable_MB = 0; + + if (prereq() != 0) { + printf("Either the sysctl compact_unevictable_allowed is not\n" + "set to 1 or couldn't read the proc file.\n" + "Skipping the test\n"); + return KSFT_SKIP; + } + + lim.rlim_cur = RLIM_INFINITY; + lim.rlim_max = RLIM_INFINITY; + if (setrlimit(RLIMIT_MEMLOCK, &lim)) { + perror("Failed to set rlimit:\n"); + return -1; + } + + page_size = getpagesize(); + + list = NULL; + + if (read_memory_info(&mem_free, &hugepage_size) != 0) { + printf("ERROR: Cannot read meminfo\n"); + return -1; + } + + mem_fragmentable_MB = mem_free * 0.8 / 1024; + + while (mem_fragmentable_MB > 0) { + map = mmap(NULL, MAP_SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE | MAP_LOCKED, -1, 0); + if (map == MAP_FAILED) + break; + + entry = malloc(sizeof(struct map_list)); + if (!entry) { + munmap(map, MAP_SIZE); + break; + } + entry->map = map; + entry->next = list; + list = entry; + + /* Write something (in this case the address of the map) to + * ensure that KSM can't merge the mapped pages + */ + for (i = 0; i < MAP_SIZE; i += page_size) + *(unsigned long *)(map + i) = (unsigned long)map + i; + + mem_fragmentable_MB -= MAP_SIZE_MB; + } + + for (entry = list; entry != NULL; entry = entry->next) { + munmap(entry->map, MAP_SIZE); + if (!entry->next) + break; + entry = entry->next; + } + + if (check_compaction(mem_free, hugepage_size) == 0) + return 0; + + return -1; +} diff --git a/tools/testing/selftests/vm/config b/tools/testing/selftests/vm/config new file mode 100644 index 000000000..be087c4bc --- /dev/null +++ b/tools/testing/selftests/vm/config @@ -0,0 +1,8 @@ +CONFIG_SYSVIPC=y +CONFIG_USERFAULTFD=y +CONFIG_TEST_VMALLOC=m +CONFIG_DEVICE_PRIVATE=y +CONFIG_TEST_HMM=m +CONFIG_GUP_TEST=y +CONFIG_TRANSPARENT_HUGEPAGE=y +CONFIG_MEM_SOFT_DIRTY=y diff --git a/tools/testing/selftests/vm/gup_test.c b/tools/testing/selftests/vm/gup_test.c new file mode 100644 index 000000000..e43879291 --- /dev/null +++ b/tools/testing/selftests/vm/gup_test.c @@ -0,0 +1,271 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "../kselftest.h" + +#include "util.h" + +#define MB (1UL << 20) + +/* Just the flags we need, copied from mm.h: */ +#define FOLL_WRITE 0x01 /* check pte is writable */ +#define FOLL_TOUCH 0x02 /* mark page accessed */ + +#define GUP_TEST_FILE "/sys/kernel/debug/gup_test" + +static unsigned long cmd = GUP_FAST_BENCHMARK; +static int gup_fd, repeats = 1; +static unsigned long size = 128 * MB; +/* Serialize prints */ +static pthread_mutex_t print_mutex = PTHREAD_MUTEX_INITIALIZER; + +static char *cmd_to_str(unsigned long cmd) +{ + switch (cmd) { + case GUP_FAST_BENCHMARK: + return "GUP_FAST_BENCHMARK"; + case PIN_FAST_BENCHMARK: + return "PIN_FAST_BENCHMARK"; + case PIN_LONGTERM_BENCHMARK: + return "PIN_LONGTERM_BENCHMARK"; + case GUP_BASIC_TEST: + return "GUP_BASIC_TEST"; + case PIN_BASIC_TEST: + return "PIN_BASIC_TEST"; + case DUMP_USER_PAGES_TEST: + return "DUMP_USER_PAGES_TEST"; + } + return "Unknown command"; +} + +void *gup_thread(void *data) +{ + struct gup_test gup = *(struct gup_test *)data; + int i; + + /* Only report timing information on the *_BENCHMARK commands: */ + if ((cmd == PIN_FAST_BENCHMARK) || (cmd == GUP_FAST_BENCHMARK) || + (cmd == PIN_LONGTERM_BENCHMARK)) { + for (i = 0; i < repeats; i++) { + gup.size = size; + if (ioctl(gup_fd, cmd, &gup)) + perror("ioctl"), exit(1); + + pthread_mutex_lock(&print_mutex); + printf("%s: Time: get:%lld put:%lld us", + cmd_to_str(cmd), gup.get_delta_usec, + gup.put_delta_usec); + if (gup.size != size) + printf(", truncated (size: %lld)", gup.size); + printf("\n"); + pthread_mutex_unlock(&print_mutex); + } + } else { + gup.size = size; + if (ioctl(gup_fd, cmd, &gup)) { + perror("ioctl"); + exit(1); + } + + pthread_mutex_lock(&print_mutex); + printf("%s: done\n", cmd_to_str(cmd)); + if (gup.size != size) + printf("Truncated (size: %lld)\n", gup.size); + pthread_mutex_unlock(&print_mutex); + } + + return NULL; +} + +int main(int argc, char **argv) +{ + struct gup_test gup = { 0 }; + int filed, i, opt, nr_pages = 1, thp = -1, write = 1, nthreads = 1, ret; + int flags = MAP_PRIVATE, touch = 0; + char *file = "/dev/zero"; + pthread_t *tid; + char *p; + + while ((opt = getopt(argc, argv, "m:r:n:F:f:abcj:tTLUuwWSHpz")) != -1) { + switch (opt) { + case 'a': + cmd = PIN_FAST_BENCHMARK; + break; + case 'b': + cmd = PIN_BASIC_TEST; + break; + case 'L': + cmd = PIN_LONGTERM_BENCHMARK; + break; + case 'c': + cmd = DUMP_USER_PAGES_TEST; + /* + * Dump page 0 (index 1). May be overridden later, by + * user's non-option arguments. + * + * .which_pages is zero-based, so that zero can mean "do + * nothing". + */ + gup.which_pages[0] = 1; + break; + case 'p': + /* works only with DUMP_USER_PAGES_TEST */ + gup.test_flags |= GUP_TEST_FLAG_DUMP_PAGES_USE_PIN; + break; + case 'F': + /* strtol, so you can pass flags in hex form */ + gup.gup_flags = strtol(optarg, 0, 0); + break; + case 'j': + nthreads = atoi(optarg); + break; + case 'm': + size = atoi(optarg) * MB; + break; + case 'r': + repeats = atoi(optarg); + break; + case 'n': + nr_pages = atoi(optarg); + break; + case 't': + thp = 1; + break; + case 'T': + thp = 0; + break; + case 'U': + cmd = GUP_BASIC_TEST; + break; + case 'u': + cmd = GUP_FAST_BENCHMARK; + break; + case 'w': + write = 1; + break; + case 'W': + write = 0; + break; + case 'f': + file = optarg; + break; + case 'S': + flags &= ~MAP_PRIVATE; + flags |= MAP_SHARED; + break; + case 'H': + flags |= (MAP_HUGETLB | MAP_ANONYMOUS); + break; + case 'z': + /* fault pages in gup, do not fault in userland */ + touch = 1; + break; + default: + return -1; + } + } + + if (optind < argc) { + int extra_arg_count = 0; + /* + * For example: + * + * ./gup_test -c 0 1 0x1001 + * + * ...to dump pages 0, 1, and 4097 + */ + + while ((optind < argc) && + (extra_arg_count < GUP_TEST_MAX_PAGES_TO_DUMP)) { + /* + * Do the 1-based indexing here, so that the user can + * use normal 0-based indexing on the command line. + */ + long page_index = strtol(argv[optind], 0, 0) + 1; + + gup.which_pages[extra_arg_count] = page_index; + extra_arg_count++; + optind++; + } + } + + filed = open(file, O_RDWR|O_CREAT); + if (filed < 0) { + perror("open"); + exit(filed); + } + + gup.nr_pages_per_call = nr_pages; + if (write) + gup.gup_flags |= FOLL_WRITE; + + gup_fd = open(GUP_TEST_FILE, O_RDWR); + if (gup_fd == -1) { + switch (errno) { + case EACCES: + if (getuid()) + printf("Please run this test as root\n"); + break; + case ENOENT: + if (opendir("/sys/kernel/debug") == NULL) { + printf("mount debugfs at /sys/kernel/debug\n"); + break; + } + printf("check if CONFIG_GUP_TEST is enabled in kernel config\n"); + break; + default: + perror("failed to open " GUP_TEST_FILE); + break; + } + exit(KSFT_SKIP); + } + + p = mmap(NULL, size, PROT_READ | PROT_WRITE, flags, filed, 0); + if (p == MAP_FAILED) { + perror("mmap"); + exit(1); + } + gup.addr = (unsigned long)p; + + if (thp == 1) + madvise(p, size, MADV_HUGEPAGE); + else if (thp == 0) + madvise(p, size, MADV_NOHUGEPAGE); + + /* + * FOLL_TOUCH, in gup_test, is used as an either/or case: either + * fault pages in from the kernel via FOLL_TOUCH, or fault them + * in here, from user space. This allows comparison of performance + * between those two cases. + */ + if (touch) { + gup.gup_flags |= FOLL_TOUCH; + } else { + for (; (unsigned long)p < gup.addr + size; p += PAGE_SIZE) + p[0] = 0; + } + + tid = malloc(sizeof(pthread_t) * nthreads); + assert(tid); + for (i = 0; i < nthreads; i++) { + ret = pthread_create(&tid[i], NULL, gup_thread, &gup); + assert(ret == 0); + } + for (i = 0; i < nthreads; i++) { + ret = pthread_join(tid[i], NULL); + assert(ret == 0); + } + free(tid); + + return 0; +} diff --git a/tools/testing/selftests/vm/hmm-tests.c b/tools/testing/selftests/vm/hmm-tests.c new file mode 100644 index 000000000..4adaad1b8 --- /dev/null +++ b/tools/testing/selftests/vm/hmm-tests.c @@ -0,0 +1,2054 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * HMM stands for Heterogeneous Memory Management, it is a helper layer inside + * the linux kernel to help device drivers mirror a process address space in + * the device. This allows the device to use the same address space which + * makes communication and data exchange a lot easier. + * + * This framework's sole purpose is to exercise various code paths inside + * the kernel to make sure that HMM performs as expected and to flush out any + * bugs. + */ + +#include "../kselftest_harness.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +/* + * This is a private UAPI to the kernel test module so it isn't exported + * in the usual include/uapi/... directory. + */ +#include +#include + +struct hmm_buffer { + void *ptr; + void *mirror; + unsigned long size; + int fd; + uint64_t cpages; + uint64_t faults; +}; + +enum { + HMM_PRIVATE_DEVICE_ONE, + HMM_PRIVATE_DEVICE_TWO, + HMM_COHERENCE_DEVICE_ONE, + HMM_COHERENCE_DEVICE_TWO, +}; + +#define TWOMEG (1 << 21) +#define HMM_BUFFER_SIZE (1024 << 12) +#define HMM_PATH_MAX 64 +#define NTIMES 10 + +#define ALIGN(x, a) (((x) + (a - 1)) & (~((a) - 1))) +/* Just the flags we need, copied from mm.h: */ +#define FOLL_WRITE 0x01 /* check pte is writable */ +#define FOLL_LONGTERM 0x10000 /* mapping lifetime is indefinite */ + +FIXTURE(hmm) +{ + int fd; + unsigned int page_size; + unsigned int page_shift; +}; + +FIXTURE_VARIANT(hmm) +{ + int device_number; +}; + +FIXTURE_VARIANT_ADD(hmm, hmm_device_private) +{ + .device_number = HMM_PRIVATE_DEVICE_ONE, +}; + +FIXTURE_VARIANT_ADD(hmm, hmm_device_coherent) +{ + .device_number = HMM_COHERENCE_DEVICE_ONE, +}; + +FIXTURE(hmm2) +{ + int fd0; + int fd1; + unsigned int page_size; + unsigned int page_shift; +}; + +FIXTURE_VARIANT(hmm2) +{ + int device_number0; + int device_number1; +}; + +FIXTURE_VARIANT_ADD(hmm2, hmm2_device_private) +{ + .device_number0 = HMM_PRIVATE_DEVICE_ONE, + .device_number1 = HMM_PRIVATE_DEVICE_TWO, +}; + +FIXTURE_VARIANT_ADD(hmm2, hmm2_device_coherent) +{ + .device_number0 = HMM_COHERENCE_DEVICE_ONE, + .device_number1 = HMM_COHERENCE_DEVICE_TWO, +}; + +static int hmm_open(int unit) +{ + char pathname[HMM_PATH_MAX]; + int fd; + + snprintf(pathname, sizeof(pathname), "/dev/hmm_dmirror%d", unit); + fd = open(pathname, O_RDWR, 0); + if (fd < 0) + fprintf(stderr, "could not open hmm dmirror driver (%s)\n", + pathname); + return fd; +} + +static bool hmm_is_coherent_type(int dev_num) +{ + return (dev_num >= HMM_COHERENCE_DEVICE_ONE); +} + +FIXTURE_SETUP(hmm) +{ + self->page_size = sysconf(_SC_PAGE_SIZE); + self->page_shift = ffs(self->page_size) - 1; + + self->fd = hmm_open(variant->device_number); + if (self->fd < 0 && hmm_is_coherent_type(variant->device_number)) + SKIP(exit(0), "DEVICE_COHERENT not available"); + ASSERT_GE(self->fd, 0); +} + +FIXTURE_SETUP(hmm2) +{ + self->page_size = sysconf(_SC_PAGE_SIZE); + self->page_shift = ffs(self->page_size) - 1; + + self->fd0 = hmm_open(variant->device_number0); + if (self->fd0 < 0 && hmm_is_coherent_type(variant->device_number0)) + SKIP(exit(0), "DEVICE_COHERENT not available"); + ASSERT_GE(self->fd0, 0); + self->fd1 = hmm_open(variant->device_number1); + ASSERT_GE(self->fd1, 0); +} + +FIXTURE_TEARDOWN(hmm) +{ + int ret = close(self->fd); + + ASSERT_EQ(ret, 0); + self->fd = -1; +} + +FIXTURE_TEARDOWN(hmm2) +{ + int ret = close(self->fd0); + + ASSERT_EQ(ret, 0); + self->fd0 = -1; + + ret = close(self->fd1); + ASSERT_EQ(ret, 0); + self->fd1 = -1; +} + +static int hmm_dmirror_cmd(int fd, + unsigned long request, + struct hmm_buffer *buffer, + unsigned long npages) +{ + struct hmm_dmirror_cmd cmd; + int ret; + + /* Simulate a device reading system memory. */ + cmd.addr = (__u64)buffer->ptr; + cmd.ptr = (__u64)buffer->mirror; + cmd.npages = npages; + + for (;;) { + ret = ioctl(fd, request, &cmd); + if (ret == 0) + break; + if (errno == EINTR) + continue; + return -errno; + } + buffer->cpages = cmd.cpages; + buffer->faults = cmd.faults; + + return 0; +} + +static void hmm_buffer_free(struct hmm_buffer *buffer) +{ + if (buffer == NULL) + return; + + if (buffer->ptr) + munmap(buffer->ptr, buffer->size); + free(buffer->mirror); + free(buffer); +} + +/* + * Create a temporary file that will be deleted on close. + */ +static int hmm_create_file(unsigned long size) +{ + char path[HMM_PATH_MAX]; + int fd; + + strcpy(path, "/tmp"); + fd = open(path, O_TMPFILE | O_EXCL | O_RDWR, 0600); + if (fd >= 0) { + int r; + + do { + r = ftruncate(fd, size); + } while (r == -1 && errno == EINTR); + if (!r) + return fd; + close(fd); + } + return -1; +} + +/* + * Return a random unsigned number. + */ +static unsigned int hmm_random(void) +{ + static int fd = -1; + unsigned int r; + + if (fd < 0) { + fd = open("/dev/urandom", O_RDONLY); + if (fd < 0) { + fprintf(stderr, "%s:%d failed to open /dev/urandom\n", + __FILE__, __LINE__); + return ~0U; + } + } + read(fd, &r, sizeof(r)); + return r; +} + +static void hmm_nanosleep(unsigned int n) +{ + struct timespec t; + + t.tv_sec = 0; + t.tv_nsec = n; + nanosleep(&t, NULL); +} + +static int hmm_migrate_sys_to_dev(int fd, + struct hmm_buffer *buffer, + unsigned long npages) +{ + return hmm_dmirror_cmd(fd, HMM_DMIRROR_MIGRATE_TO_DEV, buffer, npages); +} + +static int hmm_migrate_dev_to_sys(int fd, + struct hmm_buffer *buffer, + unsigned long npages) +{ + return hmm_dmirror_cmd(fd, HMM_DMIRROR_MIGRATE_TO_SYS, buffer, npages); +} + +/* + * Simple NULL test of device open/close. + */ +TEST_F(hmm, open_close) +{ +} + +/* + * Read private anonymous memory. + */ +TEST_F(hmm, anon_read) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int val; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* + * Initialize buffer in system memory but leave the first two pages + * zero (pte_none and pfn_zero). + */ + i = 2 * self->page_size / sizeof(*ptr); + for (ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Set buffer permission to read-only. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + + /* Populate the CPU page table with a special zero page. */ + val = *(int *)(buffer->ptr + self->page_size); + ASSERT_EQ(val, 0); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + ptr = buffer->mirror; + for (i = 0; i < 2 * self->page_size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], 0); + for (; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Read private anonymous memory which has been protected with + * mprotect() PROT_NONE. + */ +TEST_F(hmm, anon_read_prot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize mirror buffer so we can verify it isn't written. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + /* Protect buffer from reading. */ + ret = mprotect(buffer->ptr, size, PROT_NONE); + ASSERT_EQ(ret, 0); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, -EFAULT); + + /* Allow CPU to read the buffer so we can check it. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + hmm_buffer_free(buffer); +} + +/* + * Write private anonymous memory. + */ +TEST_F(hmm, anon_write) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Write private anonymous memory which has been protected with + * mprotect() PROT_READ. + */ +TEST_F(hmm, anon_write_prot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device reading a zero page of memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + ASSERT_EQ(buffer->faults, 1); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, -EPERM); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], 0); + + /* Now allow writing and see that the zero page is replaced. */ + ret = mprotect(buffer->ptr, size, PROT_WRITE | PROT_READ); + ASSERT_EQ(ret, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Check that a device writing an anonymous private mapping + * will copy-on-write if a child process inherits the mapping. + */ +TEST_F(hmm, anon_write_child) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + pid_t pid; + int child_fd; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer->ptr so we can tell if it is written. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + pid = fork(); + if (pid == -1) + ASSERT_EQ(pid, 0); + if (pid != 0) { + waitpid(pid, &ret, 0); + ASSERT_EQ(WIFEXITED(ret), 1); + + /* Check that the parent's buffer did not change. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + return; + } + + /* Check that we see the parent's values. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + /* The child process needs its own mirror to its own mm. */ + child_fd = hmm_open(0); + ASSERT_GE(child_fd, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + close(child_fd); + exit(0); +} + +/* + * Check that a device writing an anonymous shared mapping + * will not copy-on-write if a child process inherits the mapping. + */ +TEST_F(hmm, anon_write_child_shared) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + pid_t pid; + int child_fd; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer->ptr so we can tell if it is written. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = -i; + + pid = fork(); + if (pid == -1) + ASSERT_EQ(pid, 0); + if (pid != 0) { + waitpid(pid, &ret, 0); + ASSERT_EQ(WIFEXITED(ret), 1); + + /* Check that the parent's buffer did change. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + return; + } + + /* Check that we see the parent's values. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + /* The child process needs its own mirror to its own mm. */ + child_fd = hmm_open(0); + ASSERT_GE(child_fd, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(child_fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], -i); + + close(child_fd); + exit(0); +} + +/* + * Write private anonymous huge page. + */ +TEST_F(hmm, anon_write_huge) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + void *old_ptr; + void *map; + int *ptr; + int ret; + + size = 2 * TWOMEG; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + size = TWOMEG; + npages = size >> self->page_shift; + map = (void *)ALIGN((uintptr_t)buffer->ptr, size); + ret = madvise(map, size, MADV_HUGEPAGE); + ASSERT_EQ(ret, 0); + old_ptr = buffer->ptr; + buffer->ptr = map; + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + buffer->ptr = old_ptr; + hmm_buffer_free(buffer); +} + +/* + * Read numeric data from raw and tagged kernel status files. Used to read + * /proc and /sys data (without a tag) and from /proc/meminfo (with a tag). + */ +static long file_read_ulong(char *file, const char *tag) +{ + int fd; + char buf[2048]; + int len; + char *p, *q; + long val; + + fd = open(file, O_RDONLY); + if (fd < 0) { + /* Error opening the file */ + return -1; + } + + len = read(fd, buf, sizeof(buf)); + close(fd); + if (len < 0) { + /* Error in reading the file */ + return -1; + } + if (len == sizeof(buf)) { + /* Error file is too large */ + return -1; + } + buf[len] = '\0'; + + /* Search for a tag if provided */ + if (tag) { + p = strstr(buf, tag); + if (!p) + return -1; /* looks like the line we want isn't there */ + p += strlen(tag); + } else + p = buf; + + val = strtol(p, &q, 0); + if (*q != ' ') { + /* Error parsing the file */ + return -1; + } + + return val; +} + +/* + * Write huge TLBFS page. + */ +TEST_F(hmm, anon_write_hugetlbfs) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long default_hsize; + unsigned long i; + int *ptr; + int ret; + + default_hsize = file_read_ulong("/proc/meminfo", "Hugepagesize:"); + if (default_hsize < 0 || default_hsize*1024 < default_hsize) + SKIP(return, "Huge page size could not be determined"); + default_hsize = default_hsize*1024; /* KB to B */ + + size = ALIGN(TWOMEG, default_hsize); + npages = size >> self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, + -1, 0); + if (buffer->ptr == MAP_FAILED) { + free(buffer); + SKIP(return, "Huge page could not be allocated"); + } + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + munmap(buffer->ptr, buffer->size); + buffer->ptr = NULL; + hmm_buffer_free(buffer); +} + +/* + * Read mmap'ed file memory. + */ +TEST_F(hmm, file_read) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int fd; + ssize_t len; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + fd = hmm_create_file(size); + ASSERT_GE(fd, 0); + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = fd; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Write initial contents of the file. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + len = pwrite(fd, buffer->mirror, size, 0); + ASSERT_EQ(len, size); + memset(buffer->mirror, 0, size); + + buffer->ptr = mmap(NULL, size, + PROT_READ, + MAP_SHARED, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Write mmap'ed file memory. + */ +TEST_F(hmm, file_write) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + int fd; + ssize_t len; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + fd = hmm_create_file(size); + ASSERT_GE(fd, 0); + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = fd; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_SHARED, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize data that the device will write to buffer->ptr. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device wrote. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Check that the device also wrote the file. */ + len = pread(fd, buffer->mirror, size, 0); + ASSERT_EQ(len, size); + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device private memory. + */ +TEST_F(hmm, migrate) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device private memory and fault some of it back + * to system memory, then try migrating the resulting mix of system and device + * private memory to the device. + */ +TEST_F(hmm, migrate_fault) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Fault half the pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / (2 * sizeof(*ptr)); ++i) + ASSERT_EQ(ptr[i], i); + + /* Migrate memory to the device again. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +TEST_F(hmm, migrate_release) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Release device memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_RELEASE, buffer, npages); + ASSERT_EQ(ret, 0); + + /* Fault pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / (2 * sizeof(*ptr)); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous shared memory to device private memory. + */ +TEST_F(hmm, migrate_shared) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, -ENOENT); + + hmm_buffer_free(buffer); +} + +/* + * Try to migrate various memory types to device private memory. + */ +TEST_F(hmm2, migrate_mixed) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + int *ptr; + unsigned char *p; + int ret; + int val; + + npages = 6; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_NONE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + p = buffer->ptr; + + /* Migrating a protected area should be an error. */ + ret = hmm_migrate_sys_to_dev(self->fd1, buffer, npages); + ASSERT_EQ(ret, -EINVAL); + + /* Punch a hole after the first page address. */ + ret = munmap(buffer->ptr + self->page_size, self->page_size); + ASSERT_EQ(ret, 0); + + /* We expect an error if the vma doesn't cover the range. */ + ret = hmm_migrate_sys_to_dev(self->fd1, buffer, 3); + ASSERT_EQ(ret, -EINVAL); + + /* Page 2 will be a read-only zero page. */ + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 2 * self->page_size); + val = *ptr + 3; + ASSERT_EQ(val, 3); + + /* Page 3 will be read-only. */ + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 3 * self->page_size); + *ptr = val; + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + + /* Page 4-5 will be read-write. */ + ret = mprotect(buffer->ptr + 4 * self->page_size, 2 * self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 4 * self->page_size); + *ptr = val; + ptr = (int *)(buffer->ptr + 5 * self->page_size); + *ptr = val; + + /* Now try to migrate pages 2-5 to device 1. */ + buffer->ptr = p + 2 * self->page_size; + ret = hmm_migrate_sys_to_dev(self->fd1, buffer, 4); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 4); + + /* Page 5 won't be migrated to device 0 because it's on device 1. */ + buffer->ptr = p + 5 * self->page_size; + ret = hmm_migrate_sys_to_dev(self->fd0, buffer, 1); + ASSERT_EQ(ret, -ENOENT); + buffer->ptr = p; + + buffer->ptr = p; + hmm_buffer_free(buffer); +} + +/* + * Migrate anonymous memory to device memory and back to system memory + * multiple times. In case of private zone configuration, this is done + * through fault pages accessed by CPU. In case of coherent zone configuration, + * the pages from the device should be explicitly migrated back to system memory. + * The reason is Coherent device zone has coherent access by CPU, therefore + * it will not generate any page fault. + */ +TEST_F(hmm, migrate_multiple) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + unsigned long c; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; c++) { + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Migrate back to system memory and check them. */ + if (hmm_is_coherent_type(variant->device_number)) { + ret = hmm_migrate_dev_to_sys(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + } + + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); + } +} + +/* + * Read anonymous memory multiple times. + */ +TEST_F(hmm, anon_read_multiple) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + unsigned long c; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; c++) { + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i + c; + + /* Simulate a device reading system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, + npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i + c); + + hmm_buffer_free(buffer); + } +} + +void *unmap_buffer(void *p) +{ + struct hmm_buffer *buffer = p; + + /* Delay for a bit and then unmap buffer while it is being read. */ + hmm_nanosleep(hmm_random() % 32000); + munmap(buffer->ptr + buffer->size / 2, buffer->size / 2); + buffer->ptr = NULL; + + return NULL; +} + +/* + * Try reading anonymous memory while it is being unmapped. + */ +TEST_F(hmm, anon_teardown) +{ + unsigned long npages; + unsigned long size; + unsigned long c; + void *ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + for (c = 0; c < NTIMES; ++c) { + pthread_t thread; + struct hmm_buffer *buffer; + unsigned long i; + int *ptr; + int rc; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i + c; + + rc = pthread_create(&thread, NULL, unmap_buffer, buffer); + ASSERT_EQ(rc, 0); + + /* Simulate a device reading system memory. */ + rc = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ, buffer, + npages); + if (rc == 0) { + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; + i < size / sizeof(*ptr); + ++i) + ASSERT_EQ(ptr[i], i + c); + } + + pthread_join(thread, &ret); + hmm_buffer_free(buffer); + } +} + +/* + * Test memory snapshot without faulting in pages accessed by the device. + */ +TEST_F(hmm, mixedmap) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned char *m; + int ret; + + npages = 1; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE, + self->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Simulate a device snapshotting CPU pagetables. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_READ); + + hmm_buffer_free(buffer); +} + +/* + * Test memory snapshot without faulting in pages accessed by the device. + */ +TEST_F(hmm2, snapshot) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + int *ptr; + unsigned char *p; + unsigned char *m; + int ret; + int val; + + npages = 7; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_NONE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + p = buffer->ptr; + + /* Punch a hole after the first page address. */ + ret = munmap(buffer->ptr + self->page_size, self->page_size); + ASSERT_EQ(ret, 0); + + /* Page 2 will be read-only zero page. */ + ret = mprotect(buffer->ptr + 2 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 2 * self->page_size); + val = *ptr + 3; + ASSERT_EQ(val, 3); + + /* Page 3 will be read-only. */ + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 3 * self->page_size); + *ptr = val; + ret = mprotect(buffer->ptr + 3 * self->page_size, self->page_size, + PROT_READ); + ASSERT_EQ(ret, 0); + + /* Page 4-6 will be read-write. */ + ret = mprotect(buffer->ptr + 4 * self->page_size, 3 * self->page_size, + PROT_READ | PROT_WRITE); + ASSERT_EQ(ret, 0); + ptr = (int *)(buffer->ptr + 4 * self->page_size); + *ptr = val; + + /* Page 5 will be migrated to device 0. */ + buffer->ptr = p + 5 * self->page_size; + ret = hmm_migrate_sys_to_dev(self->fd0, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + + /* Page 6 will be migrated to device 1. */ + buffer->ptr = p + 6 * self->page_size; + ret = hmm_migrate_sys_to_dev(self->fd1, buffer, 1); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, 1); + + /* Simulate a device snapshotting CPU pagetables. */ + buffer->ptr = p; + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + ASSERT_EQ(m[0], HMM_DMIRROR_PROT_ERROR); + ASSERT_EQ(m[1], HMM_DMIRROR_PROT_ERROR); + ASSERT_EQ(m[2], HMM_DMIRROR_PROT_ZERO | HMM_DMIRROR_PROT_READ); + ASSERT_EQ(m[3], HMM_DMIRROR_PROT_READ); + ASSERT_EQ(m[4], HMM_DMIRROR_PROT_WRITE); + if (!hmm_is_coherent_type(variant->device_number0)) { + ASSERT_EQ(m[5], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL | + HMM_DMIRROR_PROT_WRITE); + ASSERT_EQ(m[6], HMM_DMIRROR_PROT_NONE); + } else { + ASSERT_EQ(m[5], HMM_DMIRROR_PROT_DEV_COHERENT_LOCAL | + HMM_DMIRROR_PROT_WRITE); + ASSERT_EQ(m[6], HMM_DMIRROR_PROT_DEV_COHERENT_REMOTE | + HMM_DMIRROR_PROT_WRITE); + } + + hmm_buffer_free(buffer); +} + +/* + * Test the hmm_range_fault() HMM_PFN_PMD flag for large pages that + * should be mapped by a large page table entry. + */ +TEST_F(hmm, compound) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long default_hsize; + int *ptr; + unsigned char *m; + int ret; + unsigned long i; + + /* Skip test if we can't allocate a hugetlbfs page. */ + + default_hsize = file_read_ulong("/proc/meminfo", "Hugepagesize:"); + if (default_hsize < 0 || default_hsize*1024 < default_hsize) + SKIP(return, "Huge page size could not be determined"); + default_hsize = default_hsize*1024; /* KB to B */ + + size = ALIGN(TWOMEG, default_hsize); + npages = size >> self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, + -1, 0); + if (buffer->ptr == MAP_FAILED) { + free(buffer); + return; + } + + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + /* Initialize the pages the device will snapshot in buffer->ptr. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Simulate a device snapshotting CPU pagetables. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + for (i = 0; i < npages; ++i) + ASSERT_EQ(m[i], HMM_DMIRROR_PROT_WRITE | + HMM_DMIRROR_PROT_PMD); + + /* Make the region read-only. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + + /* Simulate a device snapshotting CPU pagetables. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device saw. */ + m = buffer->mirror; + for (i = 0; i < npages; ++i) + ASSERT_EQ(m[i], HMM_DMIRROR_PROT_READ | + HMM_DMIRROR_PROT_PMD); + + munmap(buffer->ptr, buffer->size); + buffer->ptr = NULL; + hmm_buffer_free(buffer); +} + +/* + * Test two devices reading the same memory (double mapped). + */ +TEST_F(hmm2, double_map) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = 6; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(npages); + ASSERT_NE(buffer->mirror, NULL); + + /* Reserve a range of addresses. */ + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Make region read-only. */ + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + + /* Simulate device 0 reading system memory. */ + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Simulate device 1 reading system memory. */ + ret = hmm_dmirror_cmd(self->fd1, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Migrate pages to device 1 and try to read from device 0. */ + ret = hmm_migrate_sys_to_dev(self->fd1, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + ret = hmm_dmirror_cmd(self->fd0, HMM_DMIRROR_READ, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + ASSERT_EQ(buffer->faults, 1); + + /* Check what device 0 read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + hmm_buffer_free(buffer); +} + +/* + * Basic check of exclusive faulting. + */ +TEST_F(hmm, exclusive) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Map memory exclusively for device access. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_EXCLUSIVE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + /* Fault pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i]++, i); + + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i+1); + + /* Check atomic access revoked */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_CHECK_EXCLUSIVE, buffer, npages); + ASSERT_EQ(ret, 0); + + hmm_buffer_free(buffer); +} + +TEST_F(hmm, exclusive_mprotect) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Map memory exclusively for device access. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_EXCLUSIVE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + ret = mprotect(buffer->ptr, size, PROT_READ); + ASSERT_EQ(ret, 0); + + /* Simulate a device writing system memory. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_WRITE, buffer, npages); + ASSERT_EQ(ret, -EPERM); + + hmm_buffer_free(buffer); +} + +/* + * Check copy-on-write works. + */ +TEST_F(hmm, exclusive_cow) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + + npages = ALIGN(HMM_BUFFER_SIZE, self->page_size) >> self->page_shift; + ASSERT_NE(npages, 0); + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Map memory exclusively for device access. */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_EXCLUSIVE, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + fork(); + + /* Fault pages back to system memory and check them. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i]++, i); + + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i+1); + + hmm_buffer_free(buffer); +} + +static int gup_test_exec(int gup_fd, unsigned long addr, int cmd, + int npages, int size, int flags) +{ + struct gup_test gup = { + .nr_pages_per_call = npages, + .addr = addr, + .gup_flags = FOLL_WRITE | flags, + .size = size, + }; + + if (ioctl(gup_fd, cmd, &gup)) { + perror("ioctl on error\n"); + return errno; + } + + return 0; +} + +/* + * Test get user device pages through gup_test. Setting PIN_LONGTERM flag. + * This should trigger a migration back to system memory for both, private + * and coherent type pages. + * This test makes use of gup_test module. Make sure GUP_TEST_CONFIG is added + * to your configuration before you run it. + */ +TEST_F(hmm, hmm_gup_test) +{ + struct hmm_buffer *buffer; + int gup_fd; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + unsigned char *m; + + gup_fd = open("/sys/kernel/debug/gup_test", O_RDWR); + if (gup_fd == -1) + SKIP(return, "Skipping test, could not find gup_test driver"); + + npages = 4; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + /* Check what the device read. */ + for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + ASSERT_EQ(gup_test_exec(gup_fd, + (unsigned long)buffer->ptr, + GUP_BASIC_TEST, 1, self->page_size, 0), 0); + ASSERT_EQ(gup_test_exec(gup_fd, + (unsigned long)buffer->ptr + 1 * self->page_size, + GUP_FAST_BENCHMARK, 1, self->page_size, 0), 0); + ASSERT_EQ(gup_test_exec(gup_fd, + (unsigned long)buffer->ptr + 2 * self->page_size, + PIN_FAST_BENCHMARK, 1, self->page_size, FOLL_LONGTERM), 0); + ASSERT_EQ(gup_test_exec(gup_fd, + (unsigned long)buffer->ptr + 3 * self->page_size, + PIN_LONGTERM_BENCHMARK, 1, self->page_size, 0), 0); + + /* Take snapshot to CPU pagetables */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + m = buffer->mirror; + if (hmm_is_coherent_type(variant->device_number)) { + ASSERT_EQ(HMM_DMIRROR_PROT_DEV_COHERENT_LOCAL | HMM_DMIRROR_PROT_WRITE, m[0]); + ASSERT_EQ(HMM_DMIRROR_PROT_DEV_COHERENT_LOCAL | HMM_DMIRROR_PROT_WRITE, m[1]); + } else { + ASSERT_EQ(HMM_DMIRROR_PROT_WRITE, m[0]); + ASSERT_EQ(HMM_DMIRROR_PROT_WRITE, m[1]); + } + ASSERT_EQ(HMM_DMIRROR_PROT_WRITE, m[2]); + ASSERT_EQ(HMM_DMIRROR_PROT_WRITE, m[3]); + /* + * Check again the content on the pages. Make sure there's no + * corrupted data. + */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ASSERT_EQ(ptr[i], i); + + close(gup_fd); + hmm_buffer_free(buffer); +} + +/* + * Test copy-on-write in device pages. + * In case of writing to COW private page(s), a page fault will migrate pages + * back to system memory first. Then, these pages will be duplicated. In case + * of COW device coherent type, pages are duplicated directly from device + * memory. + */ +TEST_F(hmm, hmm_cow_in_device) +{ + struct hmm_buffer *buffer; + unsigned long npages; + unsigned long size; + unsigned long i; + int *ptr; + int ret; + unsigned char *m; + pid_t pid; + int status; + + npages = 4; + size = npages << self->page_shift; + + buffer = malloc(sizeof(*buffer)); + ASSERT_NE(buffer, NULL); + + buffer->fd = -1; + buffer->size = size; + buffer->mirror = malloc(size); + ASSERT_NE(buffer->mirror, NULL); + + buffer->ptr = mmap(NULL, size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, + buffer->fd, 0); + ASSERT_NE(buffer->ptr, MAP_FAILED); + + /* Initialize buffer in system memory. */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Migrate memory to device. */ + + ret = hmm_migrate_sys_to_dev(self->fd, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + + pid = fork(); + if (pid == -1) + ASSERT_EQ(pid, 0); + if (!pid) { + /* Child process waitd for SIGTERM from the parent. */ + while (1) { + } + perror("Should not reach this\n"); + exit(0); + } + /* Parent process writes to COW pages(s) and gets a + * new copy in system. In case of device private pages, + * this write causes a migration to system mem first. + */ + for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i) + ptr[i] = i; + + /* Terminate child and wait */ + EXPECT_EQ(0, kill(pid, SIGTERM)); + EXPECT_EQ(pid, waitpid(pid, &status, 0)); + EXPECT_NE(0, WIFSIGNALED(status)); + EXPECT_EQ(SIGTERM, WTERMSIG(status)); + + /* Take snapshot to CPU pagetables */ + ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT, buffer, npages); + ASSERT_EQ(ret, 0); + ASSERT_EQ(buffer->cpages, npages); + m = buffer->mirror; + for (i = 0; i < npages; i++) + ASSERT_EQ(HMM_DMIRROR_PROT_WRITE, m[i]); + + hmm_buffer_free(buffer); +} +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/vm/hugepage-mmap.c b/tools/testing/selftests/vm/hugepage-mmap.c new file mode 100644 index 000000000..93f9e7b81 --- /dev/null +++ b/tools/testing/selftests/vm/hugepage-mmap.c @@ -0,0 +1,93 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * hugepage-mmap: + * + * Example of using huge page memory in a user application using the mmap + * system call. Before running this application, make sure that the + * administrator has mounted the hugetlbfs filesystem (on some directory + * like /mnt) using the command mount -t hugetlbfs nodev /mnt. In this + * example, the app is requesting memory of size 256MB that is backed by + * huge pages. + * + * For the ia64 architecture, the Linux kernel reserves Region number 4 for + * huge pages. That means that if one requires a fixed address, a huge page + * aligned address starting with 0x800000... will be required. If a fixed + * address is not required, the kernel will select an address in the proper + * range. + * Other architectures, such as ppc64, i386 or x86_64 are not so constrained. + */ + +#include +#include +#include +#include +#include + +#define FILE_NAME "huge/hugepagefile" +#define LENGTH (256UL*1024*1024) +#define PROTECTION (PROT_READ | PROT_WRITE) + +/* Only ia64 requires this */ +#ifdef __ia64__ +#define ADDR (void *)(0x8000000000000000UL) +#define FLAGS (MAP_SHARED | MAP_FIXED) +#else +#define ADDR (void *)(0x0UL) +#define FLAGS (MAP_SHARED) +#endif + +static void check_bytes(char *addr) +{ + printf("First hex is %x\n", *((unsigned int *)addr)); +} + +static void write_bytes(char *addr) +{ + unsigned long i; + + for (i = 0; i < LENGTH; i++) + *(addr + i) = (char)i; +} + +static int read_bytes(char *addr) +{ + unsigned long i; + + check_bytes(addr); + for (i = 0; i < LENGTH; i++) + if (*(addr + i) != (char)i) { + printf("Mismatch at %lu\n", i); + return 1; + } + return 0; +} + +int main(void) +{ + void *addr; + int fd, ret; + + fd = open(FILE_NAME, O_CREAT | O_RDWR, 0755); + if (fd < 0) { + perror("Open failed"); + exit(1); + } + + addr = mmap(ADDR, LENGTH, PROTECTION, FLAGS, fd, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + unlink(FILE_NAME); + exit(1); + } + + printf("Returned address is %p\n", addr); + check_bytes(addr); + write_bytes(addr); + ret = read_bytes(addr); + + munmap(addr, LENGTH); + close(fd); + unlink(FILE_NAME); + + return ret; +} diff --git a/tools/testing/selftests/vm/hugepage-mremap.c b/tools/testing/selftests/vm/hugepage-mremap.c new file mode 100644 index 000000000..e63a0214f --- /dev/null +++ b/tools/testing/selftests/vm/hugepage-mremap.c @@ -0,0 +1,191 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * hugepage-mremap: + * + * Example of remapping huge page memory in a user application using the + * mremap system call. The path to a file in a hugetlbfs filesystem must + * be passed as the last argument to this test. The amount of memory used + * by this test in MBs can optionally be passed as an argument. If no memory + * amount is passed, the default amount is 10MB. + * + * To make sure the test triggers pmd sharing and goes through the 'unshare' + * path in the mremap code use 1GB (1024) or more. + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include /* Definition of O_* constants */ +#include /* Definition of SYS_* constants */ +#include +#include + +#define DEFAULT_LENGTH_MB 10UL +#define MB_TO_BYTES(x) (x * 1024 * 1024) + +#define PROTECTION (PROT_READ | PROT_WRITE | PROT_EXEC) +#define FLAGS (MAP_SHARED | MAP_ANONYMOUS) + +static void check_bytes(char *addr) +{ + printf("First hex is %x\n", *((unsigned int *)addr)); +} + +static void write_bytes(char *addr, size_t len) +{ + unsigned long i; + + for (i = 0; i < len; i++) + *(addr + i) = (char)i; +} + +static int read_bytes(char *addr, size_t len) +{ + unsigned long i; + + check_bytes(addr); + for (i = 0; i < len; i++) + if (*(addr + i) != (char)i) { + printf("Mismatch at %lu\n", i); + return 1; + } + return 0; +} + +static void register_region_with_uffd(char *addr, size_t len) +{ + long uffd; /* userfaultfd file descriptor */ + struct uffdio_api uffdio_api; + struct uffdio_register uffdio_register; + + /* Create and enable userfaultfd object. */ + + uffd = syscall(__NR_userfaultfd, O_CLOEXEC | O_NONBLOCK); + if (uffd == -1) { + perror("userfaultfd"); + exit(1); + } + + uffdio_api.api = UFFD_API; + uffdio_api.features = 0; + if (ioctl(uffd, UFFDIO_API, &uffdio_api) == -1) { + perror("ioctl-UFFDIO_API"); + exit(1); + } + + /* Create a private anonymous mapping. The memory will be + * demand-zero paged--that is, not yet allocated. When we + * actually touch the memory, it will be allocated via + * the userfaultfd. + */ + + addr = mmap(NULL, len, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + printf("Address returned by mmap() = %p\n", addr); + + /* Register the memory range of the mapping we just created for + * handling by the userfaultfd object. In mode, we request to track + * missing pages (i.e., pages that have not yet been faulted in). + */ + + uffdio_register.range.start = (unsigned long)addr; + uffdio_register.range.len = len; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register) == -1) { + perror("ioctl-UFFDIO_REGISTER"); + exit(1); + } +} + +int main(int argc, char *argv[]) +{ + size_t length = 0; + + if (argc != 2 && argc != 3) { + printf("Usage: %s [length_in_MB] \n", argv[0]); + exit(1); + } + + /* Read memory length as the first arg if valid, otherwise fallback to + * the default length. + */ + if (argc == 3) + length = argc > 2 ? (size_t)atoi(argv[1]) : 0UL; + + length = length > 0 ? length : DEFAULT_LENGTH_MB; + length = MB_TO_BYTES(length); + + int ret = 0; + + /* last arg is the hugetlb file name */ + int fd = open(argv[argc-1], O_CREAT | O_RDWR, 0755); + + if (fd < 0) { + perror("Open failed"); + exit(1); + } + + /* mmap to a PUD aligned address to hopefully trigger pmd sharing. */ + unsigned long suggested_addr = 0x7eaa40000000; + void *haddr = mmap((void *)suggested_addr, length, PROTECTION, + MAP_HUGETLB | MAP_SHARED | MAP_POPULATE, fd, 0); + printf("Map haddr: Returned address is %p\n", haddr); + if (haddr == MAP_FAILED) { + perror("mmap1"); + exit(1); + } + + /* mmap again to a dummy address to hopefully trigger pmd sharing. */ + suggested_addr = 0x7daa40000000; + void *daddr = mmap((void *)suggested_addr, length, PROTECTION, + MAP_HUGETLB | MAP_SHARED | MAP_POPULATE, fd, 0); + printf("Map daddr: Returned address is %p\n", daddr); + if (daddr == MAP_FAILED) { + perror("mmap3"); + exit(1); + } + + suggested_addr = 0x7faa40000000; + void *vaddr = + mmap((void *)suggested_addr, length, PROTECTION, FLAGS, -1, 0); + printf("Map vaddr: Returned address is %p\n", vaddr); + if (vaddr == MAP_FAILED) { + perror("mmap2"); + exit(1); + } + + register_region_with_uffd(haddr, length); + + void *addr = mremap(haddr, length, length, + MREMAP_MAYMOVE | MREMAP_FIXED, vaddr); + if (addr == MAP_FAILED) { + perror("mremap"); + exit(1); + } + + printf("Mremap: Returned address is %p\n", addr); + check_bytes(addr); + write_bytes(addr, length); + ret = read_bytes(addr, length); + + munmap(addr, length); + + addr = mremap(addr, length, length, 0); + if (addr != MAP_FAILED) { + printf("mremap: Expected failure, but call succeeded\n"); + exit(1); + } + + close(fd); + unlink(argv[argc-1]); + + return ret; +} diff --git a/tools/testing/selftests/vm/hugepage-shm.c b/tools/testing/selftests/vm/hugepage-shm.c new file mode 100644 index 000000000..e2527f320 --- /dev/null +++ b/tools/testing/selftests/vm/hugepage-shm.c @@ -0,0 +1,101 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * hugepage-shm: + * + * Example of using huge page memory in a user application using Sys V shared + * memory system calls. In this example the app is requesting 256MB of + * memory that is backed by huge pages. The application uses the flag + * SHM_HUGETLB in the shmget system call to inform the kernel that it is + * requesting huge pages. + * + * For the ia64 architecture, the Linux kernel reserves Region number 4 for + * huge pages. That means that if one requires a fixed address, a huge page + * aligned address starting with 0x800000... will be required. If a fixed + * address is not required, the kernel will select an address in the proper + * range. + * Other architectures, such as ppc64, i386 or x86_64 are not so constrained. + * + * Note: The default shared memory limit is quite low on many kernels, + * you may need to increase it via: + * + * echo 268435456 > /proc/sys/kernel/shmmax + * + * This will increase the maximum size per shared memory segment to 256MB. + * The other limit that you will hit eventually is shmall which is the + * total amount of shared memory in pages. To set it to 16GB on a system + * with a 4kB pagesize do: + * + * echo 4194304 > /proc/sys/kernel/shmall + */ + +#include +#include +#include +#include +#include +#include + +#ifndef SHM_HUGETLB +#define SHM_HUGETLB 04000 +#endif + +#define LENGTH (256UL*1024*1024) + +#define dprintf(x) printf(x) + +/* Only ia64 requires this */ +#ifdef __ia64__ +#define ADDR (void *)(0x8000000000000000UL) +#define SHMAT_FLAGS (SHM_RND) +#else +#define ADDR (void *)(0x0UL) +#define SHMAT_FLAGS (0) +#endif + +int main(void) +{ + int shmid; + unsigned long i; + char *shmaddr; + + shmid = shmget(2, LENGTH, SHM_HUGETLB | IPC_CREAT | SHM_R | SHM_W); + if (shmid < 0) { + perror("shmget"); + exit(1); + } + printf("shmid: 0x%x\n", shmid); + + shmaddr = shmat(shmid, ADDR, SHMAT_FLAGS); + if (shmaddr == (char *)-1) { + perror("Shared memory attach failure"); + shmctl(shmid, IPC_RMID, NULL); + exit(2); + } + printf("shmaddr: %p\n", shmaddr); + + dprintf("Starting the writes:\n"); + for (i = 0; i < LENGTH; i++) { + shmaddr[i] = (char)(i); + if (!(i % (1024 * 1024))) + dprintf("."); + } + dprintf("\n"); + + dprintf("Starting the Check..."); + for (i = 0; i < LENGTH; i++) + if (shmaddr[i] != (char)i) { + printf("\nIndex %lu mismatched\n", i); + exit(3); + } + dprintf("Done.\n"); + + if (shmdt((const void *)shmaddr) != 0) { + perror("Detach failure"); + shmctl(shmid, IPC_RMID, NULL); + exit(4); + } + + shmctl(shmid, IPC_RMID, NULL); + + return 0; +} diff --git a/tools/testing/selftests/vm/hugepage-vmemmap.c b/tools/testing/selftests/vm/hugepage-vmemmap.c new file mode 100644 index 000000000..557bdbd4f --- /dev/null +++ b/tools/testing/selftests/vm/hugepage-vmemmap.c @@ -0,0 +1,144 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * A test case of using hugepage memory in a user application using the + * mmap system call with MAP_HUGETLB flag. Before running this program + * make sure the administrator has allocated enough default sized huge + * pages to cover the 2 MB allocation. + */ +#include +#include +#include +#include +#include + +#define MAP_LENGTH (2UL * 1024 * 1024) + +#ifndef MAP_HUGETLB +#define MAP_HUGETLB 0x40000 /* arch specific */ +#endif + +#define PAGE_SIZE 4096 + +#define PAGE_COMPOUND_HEAD (1UL << 15) +#define PAGE_COMPOUND_TAIL (1UL << 16) +#define PAGE_HUGE (1UL << 17) + +#define HEAD_PAGE_FLAGS (PAGE_COMPOUND_HEAD | PAGE_HUGE) +#define TAIL_PAGE_FLAGS (PAGE_COMPOUND_TAIL | PAGE_HUGE) + +#define PM_PFRAME_BITS 55 +#define PM_PFRAME_MASK ~((1UL << PM_PFRAME_BITS) - 1) + +/* + * For ia64 architecture, Linux kernel reserves Region number 4 for hugepages. + * That means the addresses starting with 0x800000... will need to be + * specified. Specifying a fixed address is not required on ppc64, i386 + * or x86_64. + */ +#ifdef __ia64__ +#define MAP_ADDR (void *)(0x8000000000000000UL) +#define MAP_FLAGS (MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | MAP_FIXED) +#else +#define MAP_ADDR NULL +#define MAP_FLAGS (MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB) +#endif + +static void write_bytes(char *addr, size_t length) +{ + unsigned long i; + + for (i = 0; i < length; i++) + *(addr + i) = (char)i; +} + +static unsigned long virt_to_pfn(void *addr) +{ + int fd; + unsigned long pagemap; + + fd = open("/proc/self/pagemap", O_RDONLY); + if (fd < 0) + return -1UL; + + lseek(fd, (unsigned long)addr / PAGE_SIZE * sizeof(pagemap), SEEK_SET); + read(fd, &pagemap, sizeof(pagemap)); + close(fd); + + return pagemap & ~PM_PFRAME_MASK; +} + +static int check_page_flags(unsigned long pfn) +{ + int fd, i; + unsigned long pageflags; + + fd = open("/proc/kpageflags", O_RDONLY); + if (fd < 0) + return -1; + + lseek(fd, pfn * sizeof(pageflags), SEEK_SET); + + read(fd, &pageflags, sizeof(pageflags)); + if ((pageflags & HEAD_PAGE_FLAGS) != HEAD_PAGE_FLAGS) { + close(fd); + printf("Head page flags (%lx) is invalid\n", pageflags); + return -1; + } + + /* + * pages other than the first page must be tail and shouldn't be head; + * this also verifies kernel has correctly set the fake page_head to tail + * while hugetlb_free_vmemmap is enabled. + */ + for (i = 1; i < MAP_LENGTH / PAGE_SIZE; i++) { + read(fd, &pageflags, sizeof(pageflags)); + if ((pageflags & TAIL_PAGE_FLAGS) != TAIL_PAGE_FLAGS || + (pageflags & HEAD_PAGE_FLAGS) == HEAD_PAGE_FLAGS) { + close(fd); + printf("Tail page flags (%lx) is invalid\n", pageflags); + return -1; + } + } + + close(fd); + + return 0; +} + +int main(int argc, char **argv) +{ + void *addr; + unsigned long pfn; + + addr = mmap(MAP_ADDR, MAP_LENGTH, PROT_READ | PROT_WRITE, MAP_FLAGS, -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + /* Trigger allocation of HugeTLB page. */ + write_bytes(addr, MAP_LENGTH); + + pfn = virt_to_pfn(addr); + if (pfn == -1UL) { + munmap(addr, MAP_LENGTH); + perror("virt_to_pfn"); + exit(1); + } + + printf("Returned address is %p whose pfn is %lx\n", addr, pfn); + + if (check_page_flags(pfn) < 0) { + munmap(addr, MAP_LENGTH); + perror("check_page_flags"); + exit(1); + } + + /* munmap() length of MAP_HUGETLB memory must be hugepage aligned */ + if (munmap(addr, MAP_LENGTH)) { + perror("munmap"); + exit(1); + } + + return 0; +} diff --git a/tools/testing/selftests/vm/hugetlb-madvise.c b/tools/testing/selftests/vm/hugetlb-madvise.c new file mode 100644 index 000000000..3c9943131 --- /dev/null +++ b/tools/testing/selftests/vm/hugetlb-madvise.c @@ -0,0 +1,411 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * hugepage-madvise: + * + * Basic functional testing of madvise MADV_DONTNEED and MADV_REMOVE + * on hugetlb mappings. + * + * Before running this test, make sure the administrator has pre-allocated + * at least MIN_FREE_PAGES hugetlb pages and they are free. In addition, + * the test takes an argument that is the path to a file in a hugetlbfs + * filesystem. Therefore, a hugetlbfs filesystem must be mounted on some + * directory. + */ + +#include +#include +#include +#include +#define __USE_GNU +#include + +#define USAGE "USAGE: %s \n" +#define MIN_FREE_PAGES 20 +#define NR_HUGE_PAGES 10 /* common number of pages to map/allocate */ + +#define validate_free_pages(exp_free) \ + do { \ + int fhp = get_free_hugepages(); \ + if (fhp != (exp_free)) { \ + printf("Unexpected number of free huge " \ + "pages line %d\n", __LINE__); \ + exit(1); \ + } \ + } while (0) + +unsigned long huge_page_size; +unsigned long base_page_size; + +/* + * default_huge_page_size copied from mlock2-tests.c + */ +unsigned long default_huge_page_size(void) +{ + unsigned long hps = 0; + char *line = NULL; + size_t linelen = 0; + FILE *f = fopen("/proc/meminfo", "r"); + + if (!f) + return 0; + while (getline(&line, &linelen, f) > 0) { + if (sscanf(line, "Hugepagesize: %lu kB", &hps) == 1) { + hps <<= 10; + break; + } + } + + free(line); + fclose(f); + return hps; +} + +unsigned long get_free_hugepages(void) +{ + unsigned long fhp = 0; + char *line = NULL; + size_t linelen = 0; + FILE *f = fopen("/proc/meminfo", "r"); + + if (!f) + return fhp; + while (getline(&line, &linelen, f) > 0) { + if (sscanf(line, "HugePages_Free: %lu", &fhp) == 1) + break; + } + + free(line); + fclose(f); + return fhp; +} + +void write_fault_pages(void *addr, unsigned long nr_pages) +{ + unsigned long i; + + for (i = 0; i < nr_pages; i++) + *((unsigned long *)(addr + (i * huge_page_size))) = i; +} + +void read_fault_pages(void *addr, unsigned long nr_pages) +{ + unsigned long dummy = 0; + unsigned long i; + + for (i = 0; i < nr_pages; i++) + dummy += *((unsigned long *)(addr + (i * huge_page_size))); +} + +int main(int argc, char **argv) +{ + unsigned long free_hugepages; + void *addr, *addr2; + int fd; + int ret; + + if (argc != 2) { + printf(USAGE, argv[0]); + exit(1); + } + + huge_page_size = default_huge_page_size(); + if (!huge_page_size) { + printf("Unable to determine huge page size, exiting!\n"); + exit(1); + } + base_page_size = sysconf(_SC_PAGE_SIZE); + if (!huge_page_size) { + printf("Unable to determine base page size, exiting!\n"); + exit(1); + } + + free_hugepages = get_free_hugepages(); + if (free_hugepages < MIN_FREE_PAGES) { + printf("Not enough free huge pages to test, exiting!\n"); + exit(1); + } + + fd = open(argv[1], O_CREAT | O_RDWR, 0755); + if (fd < 0) { + perror("Open failed"); + exit(1); + } + + /* + * Test validity of MADV_DONTNEED addr and length arguments. mmap + * size is NR_HUGE_PAGES + 2. One page at the beginning and end of + * the mapping will be unmapped so we KNOW there is nothing mapped + * there. + */ + addr = mmap(NULL, (NR_HUGE_PAGES + 2) * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, + -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + if (munmap(addr, huge_page_size) || + munmap(addr + (NR_HUGE_PAGES + 1) * huge_page_size, + huge_page_size)) { + perror("munmap"); + exit(1); + } + addr = addr + huge_page_size; + + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* addr before mapping should fail */ + ret = madvise(addr - base_page_size, NR_HUGE_PAGES * huge_page_size, + MADV_DONTNEED); + if (!ret) { + printf("Unexpected success of madvise call with invalid addr line %d\n", + __LINE__); + exit(1); + } + + /* addr + length after mapping should fail */ + ret = madvise(addr, (NR_HUGE_PAGES * huge_page_size) + base_page_size, + MADV_DONTNEED); + if (!ret) { + printf("Unexpected success of madvise call with invalid length line %d\n", + __LINE__); + exit(1); + } + + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + + /* + * Test alignment of MADV_DONTNEED addr and length arguments + */ + addr = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, + -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* addr is not huge page size aligned and should fail */ + ret = madvise(addr + base_page_size, + NR_HUGE_PAGES * huge_page_size - base_page_size, + MADV_DONTNEED); + if (!ret) { + printf("Unexpected success of madvise call with unaligned start address %d\n", + __LINE__); + exit(1); + } + + /* addr + length should be aligned up to huge page size */ + if (madvise(addr, + ((NR_HUGE_PAGES - 1) * huge_page_size) + base_page_size, + MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + + /* should free all pages in mapping */ + validate_free_pages(free_hugepages); + + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + + /* + * Test MADV_DONTNEED on anonymous private mapping + */ + addr = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB, + -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + + /* should free all pages in mapping */ + validate_free_pages(free_hugepages); + + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + + /* + * Test MADV_DONTNEED on private mapping of hugetlb file + */ + if (fallocate(fd, 0, 0, NR_HUGE_PAGES * huge_page_size)) { + perror("fallocate"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + addr = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE, fd, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + /* read should not consume any pages */ + read_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* madvise should not free any pages */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* writes should allocate private pages */ + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - (2 * NR_HUGE_PAGES)); + + /* madvise should free private pages */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* writes should allocate private pages */ + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - (2 * NR_HUGE_PAGES)); + + /* + * The fallocate below certainly should free the pages associated + * with the file. However, pages in the private mapping are also + * freed. This is not the 'correct' behavior, but is expected + * because this is how it has worked since the initial hugetlb + * implementation. + */ + if (fallocate(fd, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, + 0, NR_HUGE_PAGES * huge_page_size)) { + perror("fallocate"); + exit(1); + } + validate_free_pages(free_hugepages); + + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + + /* + * Test MADV_DONTNEED on shared mapping of hugetlb file + */ + if (fallocate(fd, 0, 0, NR_HUGE_PAGES * huge_page_size)) { + perror("fallocate"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + addr = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_SHARED, fd, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + /* write should not consume any pages */ + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* madvise should not free any pages */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* + * Test MADV_REMOVE on shared mapping of hugetlb file + * + * madvise is same as hole punch and should free all pages. + */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_REMOVE)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages); + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + + /* + * Test MADV_REMOVE on shared and private mapping of hugetlb file + */ + if (fallocate(fd, 0, 0, NR_HUGE_PAGES * huge_page_size)) { + perror("fallocate"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + addr = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_SHARED, fd, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + /* shared write should not consume any additional pages */ + write_fault_pages(addr, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + addr2 = mmap(NULL, NR_HUGE_PAGES * huge_page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE, fd, 0); + if (addr2 == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + /* private read should not consume any pages */ + read_fault_pages(addr2, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* private write should consume additional pages */ + write_fault_pages(addr2, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - (2 * NR_HUGE_PAGES)); + + /* madvise of shared mapping should not free any pages */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages - (2 * NR_HUGE_PAGES)); + + /* madvise of private mapping should free private pages */ + if (madvise(addr2, NR_HUGE_PAGES * huge_page_size, MADV_DONTNEED)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages - NR_HUGE_PAGES); + + /* private write should consume additional pages again */ + write_fault_pages(addr2, NR_HUGE_PAGES); + validate_free_pages(free_hugepages - (2 * NR_HUGE_PAGES)); + + /* + * madvise should free both file and private pages although this is + * not correct. private pages should not be freed, but this is + * expected. See comment associated with FALLOC_FL_PUNCH_HOLE call. + */ + if (madvise(addr, NR_HUGE_PAGES * huge_page_size, MADV_REMOVE)) { + perror("madvise"); + exit(1); + } + validate_free_pages(free_hugepages); + + (void)munmap(addr, NR_HUGE_PAGES * huge_page_size); + (void)munmap(addr2, NR_HUGE_PAGES * huge_page_size); + + close(fd); + unlink(argv[1]); + return 0; +} diff --git a/tools/testing/selftests/vm/hugetlb_reparenting_test.sh b/tools/testing/selftests/vm/hugetlb_reparenting_test.sh new file mode 100644 index 000000000..14d26075c --- /dev/null +++ b/tools/testing/selftests/vm/hugetlb_reparenting_test.sh @@ -0,0 +1,252 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +set -e + +if [[ $(id -u) -ne 0 ]]; then + echo "This test must be run as root. Skipping..." + exit $ksft_skip +fi + +usage_file=usage_in_bytes + +if [[ "$1" == "-cgroup-v2" ]]; then + cgroup2=1 + usage_file=current +fi + + +if [[ $cgroup2 ]]; then + CGROUP_ROOT=$(mount -t cgroup2 | head -1 | awk '{print $3}') + if [[ -z "$CGROUP_ROOT" ]]; then + CGROUP_ROOT=/dev/cgroup/memory + mount -t cgroup2 none $CGROUP_ROOT + do_umount=1 + fi + echo "+hugetlb +memory" >$CGROUP_ROOT/cgroup.subtree_control +else + CGROUP_ROOT=$(mount -t cgroup | grep ",hugetlb" | awk '{print $3}') + if [[ -z "$CGROUP_ROOT" ]]; then + CGROUP_ROOT=/dev/cgroup/memory + mount -t cgroup memory,hugetlb $CGROUP_ROOT + do_umount=1 + fi +fi +MNT='/mnt/huge/' + +function get_machine_hugepage_size() { + hpz=$(grep -i hugepagesize /proc/meminfo) + kb=${hpz:14:-3} + mb=$(($kb / 1024)) + echo $mb +} + +MB=$(get_machine_hugepage_size) + +function cleanup() { + echo cleanup + set +e + rm -rf "$MNT"/* 2>/dev/null + umount "$MNT" 2>/dev/null + rmdir "$MNT" 2>/dev/null + rmdir "$CGROUP_ROOT"/a/b 2>/dev/null + rmdir "$CGROUP_ROOT"/a 2>/dev/null + rmdir "$CGROUP_ROOT"/test1 2>/dev/null + echo 0 >/proc/sys/vm/nr_hugepages + set -e +} + +function assert_state() { + local expected_a="$1" + local expected_a_hugetlb="$2" + local expected_b="" + local expected_b_hugetlb="" + + if [ ! -z ${3:-} ] && [ ! -z ${4:-} ]; then + expected_b="$3" + expected_b_hugetlb="$4" + fi + local tolerance=$((5 * 1024 * 1024)) + + local actual_a + actual_a="$(cat "$CGROUP_ROOT"/a/memory.$usage_file)" + if [[ $actual_a -lt $(($expected_a - $tolerance)) ]] || + [[ $actual_a -gt $(($expected_a + $tolerance)) ]]; then + echo actual a = $((${actual_a%% *} / 1024 / 1024)) MB + echo expected a = $((${expected_a%% *} / 1024 / 1024)) MB + echo fail + + cleanup + exit 1 + fi + + local actual_a_hugetlb + actual_a_hugetlb="$(cat "$CGROUP_ROOT"/a/hugetlb.${MB}MB.$usage_file)" + if [[ $actual_a_hugetlb -lt $(($expected_a_hugetlb - $tolerance)) ]] || + [[ $actual_a_hugetlb -gt $(($expected_a_hugetlb + $tolerance)) ]]; then + echo actual a hugetlb = $((${actual_a_hugetlb%% *} / 1024 / 1024)) MB + echo expected a hugetlb = $((${expected_a_hugetlb%% *} / 1024 / 1024)) MB + echo fail + + cleanup + exit 1 + fi + + if [[ -z "$expected_b" || -z "$expected_b_hugetlb" ]]; then + return + fi + + local actual_b + actual_b="$(cat "$CGROUP_ROOT"/a/b/memory.$usage_file)" + if [[ $actual_b -lt $(($expected_b - $tolerance)) ]] || + [[ $actual_b -gt $(($expected_b + $tolerance)) ]]; then + echo actual b = $((${actual_b%% *} / 1024 / 1024)) MB + echo expected b = $((${expected_b%% *} / 1024 / 1024)) MB + echo fail + + cleanup + exit 1 + fi + + local actual_b_hugetlb + actual_b_hugetlb="$(cat "$CGROUP_ROOT"/a/b/hugetlb.${MB}MB.$usage_file)" + if [[ $actual_b_hugetlb -lt $(($expected_b_hugetlb - $tolerance)) ]] || + [[ $actual_b_hugetlb -gt $(($expected_b_hugetlb + $tolerance)) ]]; then + echo actual b hugetlb = $((${actual_b_hugetlb%% *} / 1024 / 1024)) MB + echo expected b hugetlb = $((${expected_b_hugetlb%% *} / 1024 / 1024)) MB + echo fail + + cleanup + exit 1 + fi +} + +function setup() { + echo 100 >/proc/sys/vm/nr_hugepages + mkdir "$CGROUP_ROOT"/a + sleep 1 + if [[ $cgroup2 ]]; then + echo "+hugetlb +memory" >$CGROUP_ROOT/a/cgroup.subtree_control + else + echo 0 >$CGROUP_ROOT/a/cpuset.mems + echo 0 >$CGROUP_ROOT/a/cpuset.cpus + fi + + mkdir "$CGROUP_ROOT"/a/b + + if [[ ! $cgroup2 ]]; then + echo 0 >$CGROUP_ROOT/a/b/cpuset.mems + echo 0 >$CGROUP_ROOT/a/b/cpuset.cpus + fi + + mkdir -p "$MNT" + mount -t hugetlbfs none "$MNT" +} + +write_hugetlbfs() { + local cgroup="$1" + local path="$2" + local size="$3" + + if [[ $cgroup2 ]]; then + echo $$ >$CGROUP_ROOT/$cgroup/cgroup.procs + else + echo 0 >$CGROUP_ROOT/$cgroup/cpuset.mems + echo 0 >$CGROUP_ROOT/$cgroup/cpuset.cpus + echo $$ >"$CGROUP_ROOT/$cgroup/tasks" + fi + ./write_to_hugetlbfs -p "$path" -s "$size" -m 0 -o + if [[ $cgroup2 ]]; then + echo $$ >$CGROUP_ROOT/cgroup.procs + else + echo $$ >"$CGROUP_ROOT/tasks" + fi + echo +} + +set -e + +size=$((${MB} * 1024 * 1024 * 25)) # 50MB = 25 * 2MB hugepages. + +cleanup + +echo +echo +echo Test charge, rmdir, uncharge +setup +echo mkdir +mkdir $CGROUP_ROOT/test1 + +echo write +write_hugetlbfs test1 "$MNT"/test $size + +echo rmdir +rmdir $CGROUP_ROOT/test1 +mkdir $CGROUP_ROOT/test1 + +echo uncharge +rm -rf /mnt/huge/* + +cleanup + +echo done +echo +echo +if [[ ! $cgroup2 ]]; then + echo "Test parent and child hugetlb usage" + setup + + echo write + write_hugetlbfs a "$MNT"/test $size + + echo Assert memory charged correctly for parent use. + assert_state 0 $size 0 0 + + write_hugetlbfs a/b "$MNT"/test2 $size + + echo Assert memory charged correctly for child use. + assert_state 0 $(($size * 2)) 0 $size + + rmdir "$CGROUP_ROOT"/a/b + sleep 5 + echo Assert memory reparent correctly. + assert_state 0 $(($size * 2)) + + rm -rf "$MNT"/* + umount "$MNT" + echo Assert memory uncharged correctly. + assert_state 0 0 + + cleanup +fi + +echo +echo +echo "Test child only hugetlb usage" +echo setup +setup + +echo write +write_hugetlbfs a/b "$MNT"/test2 $size + +echo Assert memory charged correctly for child only use. +assert_state 0 $(($size)) 0 $size + +rmdir "$CGROUP_ROOT"/a/b +echo Assert memory reparent correctly. +assert_state 0 $size + +rm -rf "$MNT"/* +umount "$MNT" +echo Assert memory uncharged correctly. +assert_state 0 0 + +cleanup + +echo ALL PASS + +umount $CGROUP_ROOT +rm -rf $CGROUP_ROOT diff --git a/tools/testing/selftests/vm/khugepaged.c b/tools/testing/selftests/vm/khugepaged.c new file mode 100644 index 000000000..64126c8cd --- /dev/null +++ b/tools/testing/selftests/vm/khugepaged.c @@ -0,0 +1,1558 @@ +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#include "linux/magic.h" + +#include "vm_util.h" + +#ifndef MADV_PAGEOUT +#define MADV_PAGEOUT 21 +#endif +#ifndef MADV_POPULATE_READ +#define MADV_POPULATE_READ 22 +#endif +#ifndef MADV_COLLAPSE +#define MADV_COLLAPSE 25 +#endif + +#define BASE_ADDR ((void *)(1UL << 30)) +static unsigned long hpage_pmd_size; +static unsigned long page_size; +static int hpage_pmd_nr; + +#define THP_SYSFS "/sys/kernel/mm/transparent_hugepage/" +#define PID_SMAPS "/proc/self/smaps" +#define TEST_FILE "collapse_test_file" + +#define MAX_LINE_LENGTH 500 + +enum vma_type { + VMA_ANON, + VMA_FILE, + VMA_SHMEM, +}; + +struct mem_ops { + void *(*setup_area)(int nr_hpages); + void (*cleanup_area)(void *p, unsigned long size); + void (*fault)(void *p, unsigned long start, unsigned long end); + bool (*check_huge)(void *addr, int nr_hpages); + const char *name; +}; + +static struct mem_ops *file_ops; +static struct mem_ops *anon_ops; +static struct mem_ops *shmem_ops; + +struct collapse_context { + void (*collapse)(const char *msg, char *p, int nr_hpages, + struct mem_ops *ops, bool expect); + bool enforce_pte_scan_limits; + const char *name; +}; + +static struct collapse_context *khugepaged_context; +static struct collapse_context *madvise_context; + +struct file_info { + const char *dir; + char path[PATH_MAX]; + enum vma_type type; + int fd; + char dev_queue_read_ahead_path[PATH_MAX]; +}; + +static struct file_info finfo; + +enum thp_enabled { + THP_ALWAYS, + THP_MADVISE, + THP_NEVER, +}; + +static const char *thp_enabled_strings[] = { + "always", + "madvise", + "never", + NULL +}; + +enum thp_defrag { + THP_DEFRAG_ALWAYS, + THP_DEFRAG_DEFER, + THP_DEFRAG_DEFER_MADVISE, + THP_DEFRAG_MADVISE, + THP_DEFRAG_NEVER, +}; + +static const char *thp_defrag_strings[] = { + "always", + "defer", + "defer+madvise", + "madvise", + "never", + NULL +}; + +enum shmem_enabled { + SHMEM_ALWAYS, + SHMEM_WITHIN_SIZE, + SHMEM_ADVISE, + SHMEM_NEVER, + SHMEM_DENY, + SHMEM_FORCE, +}; + +static const char *shmem_enabled_strings[] = { + "always", + "within_size", + "advise", + "never", + "deny", + "force", + NULL +}; + +struct khugepaged_settings { + bool defrag; + unsigned int alloc_sleep_millisecs; + unsigned int scan_sleep_millisecs; + unsigned int max_ptes_none; + unsigned int max_ptes_swap; + unsigned int max_ptes_shared; + unsigned long pages_to_scan; +}; + +struct settings { + enum thp_enabled thp_enabled; + enum thp_defrag thp_defrag; + enum shmem_enabled shmem_enabled; + bool use_zero_page; + struct khugepaged_settings khugepaged; + unsigned long read_ahead_kb; +}; + +static struct settings saved_settings; +static bool skip_settings_restore; + +static int exit_status; + +static void success(const char *msg) +{ + printf(" \e[32m%s\e[0m\n", msg); +} + +static void fail(const char *msg) +{ + printf(" \e[31m%s\e[0m\n", msg); + exit_status++; +} + +static void skip(const char *msg) +{ + printf(" \e[33m%s\e[0m\n", msg); +} + +static int read_file(const char *path, char *buf, size_t buflen) +{ + int fd; + ssize_t numread; + + fd = open(path, O_RDONLY); + if (fd == -1) + return 0; + + numread = read(fd, buf, buflen - 1); + if (numread < 1) { + close(fd); + return 0; + } + + buf[numread] = '\0'; + close(fd); + + return (unsigned int) numread; +} + +static int write_file(const char *path, const char *buf, size_t buflen) +{ + int fd; + ssize_t numwritten; + + fd = open(path, O_WRONLY); + if (fd == -1) { + printf("open(%s)\n", path); + exit(EXIT_FAILURE); + return 0; + } + + numwritten = write(fd, buf, buflen - 1); + close(fd); + if (numwritten < 1) { + printf("write(%s)\n", buf); + exit(EXIT_FAILURE); + return 0; + } + + return (unsigned int) numwritten; +} + +static int read_string(const char *name, const char *strings[]) +{ + char path[PATH_MAX]; + char buf[256]; + char *c; + int ret; + + ret = snprintf(path, PATH_MAX, THP_SYSFS "%s", name); + if (ret >= PATH_MAX) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + + if (!read_file(path, buf, sizeof(buf))) { + perror(path); + exit(EXIT_FAILURE); + } + + c = strchr(buf, '['); + if (!c) { + printf("%s: Parse failure\n", __func__); + exit(EXIT_FAILURE); + } + + c++; + memmove(buf, c, sizeof(buf) - (c - buf)); + + c = strchr(buf, ']'); + if (!c) { + printf("%s: Parse failure\n", __func__); + exit(EXIT_FAILURE); + } + *c = '\0'; + + ret = 0; + while (strings[ret]) { + if (!strcmp(strings[ret], buf)) + return ret; + ret++; + } + + printf("Failed to parse %s\n", name); + exit(EXIT_FAILURE); +} + +static void write_string(const char *name, const char *val) +{ + char path[PATH_MAX]; + int ret; + + ret = snprintf(path, PATH_MAX, THP_SYSFS "%s", name); + if (ret >= PATH_MAX) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + + if (!write_file(path, val, strlen(val) + 1)) { + perror(path); + exit(EXIT_FAILURE); + } +} + +static const unsigned long _read_num(const char *path) +{ + char buf[21]; + + if (read_file(path, buf, sizeof(buf)) < 0) { + perror("read_file(read_num)"); + exit(EXIT_FAILURE); + } + + return strtoul(buf, NULL, 10); +} + +static const unsigned long read_num(const char *name) +{ + char path[PATH_MAX]; + int ret; + + ret = snprintf(path, PATH_MAX, THP_SYSFS "%s", name); + if (ret >= PATH_MAX) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + return _read_num(path); +} + +static void _write_num(const char *path, unsigned long num) +{ + char buf[21]; + + sprintf(buf, "%ld", num); + if (!write_file(path, buf, strlen(buf) + 1)) { + perror(path); + exit(EXIT_FAILURE); + } +} + +static void write_num(const char *name, unsigned long num) +{ + char path[PATH_MAX]; + int ret; + + ret = snprintf(path, PATH_MAX, THP_SYSFS "%s", name); + if (ret >= PATH_MAX) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + _write_num(path, num); +} + +static void write_settings(struct settings *settings) +{ + struct khugepaged_settings *khugepaged = &settings->khugepaged; + + write_string("enabled", thp_enabled_strings[settings->thp_enabled]); + write_string("defrag", thp_defrag_strings[settings->thp_defrag]); + write_string("shmem_enabled", + shmem_enabled_strings[settings->shmem_enabled]); + write_num("use_zero_page", settings->use_zero_page); + + write_num("khugepaged/defrag", khugepaged->defrag); + write_num("khugepaged/alloc_sleep_millisecs", + khugepaged->alloc_sleep_millisecs); + write_num("khugepaged/scan_sleep_millisecs", + khugepaged->scan_sleep_millisecs); + write_num("khugepaged/max_ptes_none", khugepaged->max_ptes_none); + write_num("khugepaged/max_ptes_swap", khugepaged->max_ptes_swap); + write_num("khugepaged/max_ptes_shared", khugepaged->max_ptes_shared); + write_num("khugepaged/pages_to_scan", khugepaged->pages_to_scan); + + if (file_ops && finfo.type == VMA_FILE) + _write_num(finfo.dev_queue_read_ahead_path, + settings->read_ahead_kb); +} + +#define MAX_SETTINGS_DEPTH 4 +static struct settings settings_stack[MAX_SETTINGS_DEPTH]; +static int settings_index; + +static struct settings *current_settings(void) +{ + if (!settings_index) { + printf("Fail: No settings set"); + exit(EXIT_FAILURE); + } + return settings_stack + settings_index - 1; +} + +static void push_settings(struct settings *settings) +{ + if (settings_index >= MAX_SETTINGS_DEPTH) { + printf("Fail: Settings stack exceeded"); + exit(EXIT_FAILURE); + } + settings_stack[settings_index++] = *settings; + write_settings(current_settings()); +} + +static void pop_settings(void) +{ + if (settings_index <= 0) { + printf("Fail: Settings stack empty"); + exit(EXIT_FAILURE); + } + --settings_index; + write_settings(current_settings()); +} + +static void restore_settings(int sig) +{ + if (skip_settings_restore) + goto out; + + printf("Restore THP and khugepaged settings..."); + write_settings(&saved_settings); + success("OK"); + if (sig) + exit(EXIT_FAILURE); +out: + exit(exit_status); +} + +static void save_settings(void) +{ + printf("Save THP and khugepaged settings..."); + saved_settings = (struct settings) { + .thp_enabled = read_string("enabled", thp_enabled_strings), + .thp_defrag = read_string("defrag", thp_defrag_strings), + .shmem_enabled = + read_string("shmem_enabled", shmem_enabled_strings), + .use_zero_page = read_num("use_zero_page"), + }; + saved_settings.khugepaged = (struct khugepaged_settings) { + .defrag = read_num("khugepaged/defrag"), + .alloc_sleep_millisecs = + read_num("khugepaged/alloc_sleep_millisecs"), + .scan_sleep_millisecs = + read_num("khugepaged/scan_sleep_millisecs"), + .max_ptes_none = read_num("khugepaged/max_ptes_none"), + .max_ptes_swap = read_num("khugepaged/max_ptes_swap"), + .max_ptes_shared = read_num("khugepaged/max_ptes_shared"), + .pages_to_scan = read_num("khugepaged/pages_to_scan"), + }; + if (file_ops && finfo.type == VMA_FILE) + saved_settings.read_ahead_kb = + _read_num(finfo.dev_queue_read_ahead_path); + + success("OK"); + + signal(SIGTERM, restore_settings); + signal(SIGINT, restore_settings); + signal(SIGHUP, restore_settings); + signal(SIGQUIT, restore_settings); +} + +static void get_finfo(const char *dir) +{ + struct stat path_stat; + struct statfs fs; + char buf[1 << 10]; + char path[PATH_MAX]; + char *str, *end; + + finfo.dir = dir; + stat(finfo.dir, &path_stat); + if (!S_ISDIR(path_stat.st_mode)) { + printf("%s: Not a directory (%s)\n", __func__, finfo.dir); + exit(EXIT_FAILURE); + } + if (snprintf(finfo.path, sizeof(finfo.path), "%s/" TEST_FILE, + finfo.dir) >= sizeof(finfo.path)) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + if (statfs(finfo.dir, &fs)) { + perror("statfs()"); + exit(EXIT_FAILURE); + } + finfo.type = fs.f_type == TMPFS_MAGIC ? VMA_SHMEM : VMA_FILE; + if (finfo.type == VMA_SHMEM) + return; + + /* Find owning device's queue/read_ahead_kb control */ + if (snprintf(path, sizeof(path), "/sys/dev/block/%d:%d/uevent", + major(path_stat.st_dev), minor(path_stat.st_dev)) + >= sizeof(path)) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + if (read_file(path, buf, sizeof(buf)) < 0) { + perror("read_file(read_num)"); + exit(EXIT_FAILURE); + } + if (strstr(buf, "DEVTYPE=disk")) { + /* Found it */ + if (snprintf(finfo.dev_queue_read_ahead_path, + sizeof(finfo.dev_queue_read_ahead_path), + "/sys/dev/block/%d:%d/queue/read_ahead_kb", + major(path_stat.st_dev), minor(path_stat.st_dev)) + >= sizeof(finfo.dev_queue_read_ahead_path)) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + return; + } + if (!strstr(buf, "DEVTYPE=partition")) { + printf("%s: Unknown device type: %s\n", __func__, path); + exit(EXIT_FAILURE); + } + /* + * Partition of block device - need to find actual device. + * Using naming convention that devnameN is partition of + * device devname. + */ + str = strstr(buf, "DEVNAME="); + if (!str) { + printf("%s: Could not read: %s", __func__, path); + exit(EXIT_FAILURE); + } + str += 8; + end = str; + while (*end) { + if (isdigit(*end)) { + *end = '\0'; + if (snprintf(finfo.dev_queue_read_ahead_path, + sizeof(finfo.dev_queue_read_ahead_path), + "/sys/block/%s/queue/read_ahead_kb", + str) >= sizeof(finfo.dev_queue_read_ahead_path)) { + printf("%s: Pathname is too long\n", __func__); + exit(EXIT_FAILURE); + } + return; + } + ++end; + } + printf("%s: Could not read: %s\n", __func__, path); + exit(EXIT_FAILURE); +} + +static bool check_swap(void *addr, unsigned long size) +{ + bool swap = false; + int ret; + FILE *fp; + char buffer[MAX_LINE_LENGTH]; + char addr_pattern[MAX_LINE_LENGTH]; + + ret = snprintf(addr_pattern, MAX_LINE_LENGTH, "%08lx-", + (unsigned long) addr); + if (ret >= MAX_LINE_LENGTH) { + printf("%s: Pattern is too long\n", __func__); + exit(EXIT_FAILURE); + } + + + fp = fopen(PID_SMAPS, "r"); + if (!fp) { + printf("%s: Failed to open file %s\n", __func__, PID_SMAPS); + exit(EXIT_FAILURE); + } + if (!check_for_pattern(fp, addr_pattern, buffer, sizeof(buffer))) + goto err_out; + + ret = snprintf(addr_pattern, MAX_LINE_LENGTH, "Swap:%19ld kB", + size >> 10); + if (ret >= MAX_LINE_LENGTH) { + printf("%s: Pattern is too long\n", __func__); + exit(EXIT_FAILURE); + } + /* + * Fetch the Swap: in the same block and check whether it got + * the expected number of hugeepages next. + */ + if (!check_for_pattern(fp, "Swap:", buffer, sizeof(buffer))) + goto err_out; + + if (strncmp(buffer, addr_pattern, strlen(addr_pattern))) + goto err_out; + + swap = true; +err_out: + fclose(fp); + return swap; +} + +static void *alloc_mapping(int nr) +{ + void *p; + + p = mmap(BASE_ADDR, nr * hpage_pmd_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + if (p != BASE_ADDR) { + printf("Failed to allocate VMA at %p\n", BASE_ADDR); + exit(EXIT_FAILURE); + } + + return p; +} + +static void fill_memory(int *p, unsigned long start, unsigned long end) +{ + int i; + + for (i = start / page_size; i < end / page_size; i++) + p[i * page_size / sizeof(*p)] = i + 0xdead0000; +} + +/* + * MADV_COLLAPSE is a best-effort request and may fail if an internal + * resource is temporarily unavailable, in which case it will set errno to + * EAGAIN. In such a case, immediately reattempt the operation one more + * time. + */ +static int madvise_collapse_retry(void *p, unsigned long size) +{ + bool retry = true; + int ret; + +retry: + ret = madvise(p, size, MADV_COLLAPSE); + if (ret && errno == EAGAIN && retry) { + retry = false; + goto retry; + } + return ret; +} + +/* + * Returns pmd-mapped hugepage in VMA marked VM_HUGEPAGE, filled with + * validate_memory()'able contents. + */ +static void *alloc_hpage(struct mem_ops *ops) +{ + void *p = ops->setup_area(1); + + ops->fault(p, 0, hpage_pmd_size); + + /* + * VMA should be neither VM_HUGEPAGE nor VM_NOHUGEPAGE. + * The latter is ineligible for collapse by MADV_COLLAPSE + * while the former might cause MADV_COLLAPSE to race with + * khugepaged on low-load system (like a test machine), which + * would cause MADV_COLLAPSE to fail with EAGAIN. + */ + printf("Allocate huge page..."); + if (madvise_collapse_retry(p, hpage_pmd_size)) { + perror("madvise(MADV_COLLAPSE)"); + exit(EXIT_FAILURE); + } + if (!ops->check_huge(p, 1)) { + perror("madvise(MADV_COLLAPSE)"); + exit(EXIT_FAILURE); + } + if (madvise(p, hpage_pmd_size, MADV_HUGEPAGE)) { + perror("madvise(MADV_HUGEPAGE)"); + exit(EXIT_FAILURE); + } + success("OK"); + return p; +} + +static void validate_memory(int *p, unsigned long start, unsigned long end) +{ + int i; + + for (i = start / page_size; i < end / page_size; i++) { + if (p[i * page_size / sizeof(*p)] != i + 0xdead0000) { + printf("Page %d is corrupted: %#x\n", + i, p[i * page_size / sizeof(*p)]); + exit(EXIT_FAILURE); + } + } +} + +static void *anon_setup_area(int nr_hpages) +{ + return alloc_mapping(nr_hpages); +} + +static void anon_cleanup_area(void *p, unsigned long size) +{ + munmap(p, size); +} + +static void anon_fault(void *p, unsigned long start, unsigned long end) +{ + fill_memory(p, start, end); +} + +static bool anon_check_huge(void *addr, int nr_hpages) +{ + return check_huge_anon(addr, nr_hpages, hpage_pmd_size); +} + +static void *file_setup_area(int nr_hpages) +{ + int fd; + void *p; + unsigned long size; + + unlink(finfo.path); /* Cleanup from previous failed tests */ + printf("Creating %s for collapse%s...", finfo.path, + finfo.type == VMA_SHMEM ? " (tmpfs)" : ""); + fd = open(finfo.path, O_DSYNC | O_CREAT | O_RDWR | O_TRUNC | O_EXCL, + 777); + if (fd < 0) { + perror("open()"); + exit(EXIT_FAILURE); + } + + size = nr_hpages * hpage_pmd_size; + p = alloc_mapping(nr_hpages); + fill_memory(p, 0, size); + write(fd, p, size); + close(fd); + munmap(p, size); + success("OK"); + + printf("Opening %s read only for collapse...", finfo.path); + finfo.fd = open(finfo.path, O_RDONLY, 777); + if (finfo.fd < 0) { + perror("open()"); + exit(EXIT_FAILURE); + } + p = mmap(BASE_ADDR, size, PROT_READ | PROT_EXEC, + MAP_PRIVATE, finfo.fd, 0); + if (p == MAP_FAILED || p != BASE_ADDR) { + perror("mmap()"); + exit(EXIT_FAILURE); + } + + /* Drop page cache */ + write_file("/proc/sys/vm/drop_caches", "3", 2); + success("OK"); + return p; +} + +static void file_cleanup_area(void *p, unsigned long size) +{ + munmap(p, size); + close(finfo.fd); + unlink(finfo.path); +} + +static void file_fault(void *p, unsigned long start, unsigned long end) +{ + if (madvise(((char *)p) + start, end - start, MADV_POPULATE_READ)) { + perror("madvise(MADV_POPULATE_READ"); + exit(EXIT_FAILURE); + } +} + +static bool file_check_huge(void *addr, int nr_hpages) +{ + switch (finfo.type) { + case VMA_FILE: + return check_huge_file(addr, nr_hpages, hpage_pmd_size); + case VMA_SHMEM: + return check_huge_shmem(addr, nr_hpages, hpage_pmd_size); + default: + exit(EXIT_FAILURE); + return false; + } +} + +static void *shmem_setup_area(int nr_hpages) +{ + void *p; + unsigned long size = nr_hpages * hpage_pmd_size; + + finfo.fd = memfd_create("khugepaged-selftest-collapse-shmem", 0); + if (finfo.fd < 0) { + perror("memfd_create()"); + exit(EXIT_FAILURE); + } + if (ftruncate(finfo.fd, size)) { + perror("ftruncate()"); + exit(EXIT_FAILURE); + } + p = mmap(BASE_ADDR, size, PROT_READ | PROT_WRITE, MAP_SHARED, finfo.fd, + 0); + if (p != BASE_ADDR) { + perror("mmap()"); + exit(EXIT_FAILURE); + } + return p; +} + +static void shmem_cleanup_area(void *p, unsigned long size) +{ + munmap(p, size); + close(finfo.fd); +} + +static bool shmem_check_huge(void *addr, int nr_hpages) +{ + return check_huge_shmem(addr, nr_hpages, hpage_pmd_size); +} + +static struct mem_ops __anon_ops = { + .setup_area = &anon_setup_area, + .cleanup_area = &anon_cleanup_area, + .fault = &anon_fault, + .check_huge = &anon_check_huge, + .name = "anon", +}; + +static struct mem_ops __file_ops = { + .setup_area = &file_setup_area, + .cleanup_area = &file_cleanup_area, + .fault = &file_fault, + .check_huge = &file_check_huge, + .name = "file", +}; + +static struct mem_ops __shmem_ops = { + .setup_area = &shmem_setup_area, + .cleanup_area = &shmem_cleanup_area, + .fault = &anon_fault, + .check_huge = &shmem_check_huge, + .name = "shmem", +}; + +static void __madvise_collapse(const char *msg, char *p, int nr_hpages, + struct mem_ops *ops, bool expect) +{ + int ret; + struct settings settings = *current_settings(); + + printf("%s...", msg); + + /* + * Prevent khugepaged interference and tests that MADV_COLLAPSE + * ignores /sys/kernel/mm/transparent_hugepage/enabled + */ + settings.thp_enabled = THP_NEVER; + settings.shmem_enabled = SHMEM_NEVER; + push_settings(&settings); + + /* Clear VM_NOHUGEPAGE */ + madvise(p, nr_hpages * hpage_pmd_size, MADV_HUGEPAGE); + ret = madvise_collapse_retry(p, nr_hpages * hpage_pmd_size); + if (((bool)ret) == expect) + fail("Fail: Bad return value"); + else if (!ops->check_huge(p, expect ? nr_hpages : 0)) + fail("Fail: check_huge()"); + else + success("OK"); + + pop_settings(); +} + +static void madvise_collapse(const char *msg, char *p, int nr_hpages, + struct mem_ops *ops, bool expect) +{ + /* Sanity check */ + if (!ops->check_huge(p, 0)) { + printf("Unexpected huge page\n"); + exit(EXIT_FAILURE); + } + __madvise_collapse(msg, p, nr_hpages, ops, expect); +} + +#define TICK 500000 +static bool wait_for_scan(const char *msg, char *p, int nr_hpages, + struct mem_ops *ops) +{ + int full_scans; + int timeout = 6; /* 3 seconds */ + + /* Sanity check */ + if (!ops->check_huge(p, 0)) { + printf("Unexpected huge page\n"); + exit(EXIT_FAILURE); + } + + madvise(p, nr_hpages * hpage_pmd_size, MADV_HUGEPAGE); + + /* Wait until the second full_scan completed */ + full_scans = read_num("khugepaged/full_scans") + 2; + + printf("%s...", msg); + while (timeout--) { + if (ops->check_huge(p, nr_hpages)) + break; + if (read_num("khugepaged/full_scans") >= full_scans) + break; + printf("."); + usleep(TICK); + } + + madvise(p, nr_hpages * hpage_pmd_size, MADV_NOHUGEPAGE); + + return timeout == -1; +} + +static void khugepaged_collapse(const char *msg, char *p, int nr_hpages, + struct mem_ops *ops, bool expect) +{ + if (wait_for_scan(msg, p, nr_hpages, ops)) { + if (expect) + fail("Timeout"); + else + success("OK"); + return; + } + + /* + * For file and shmem memory, khugepaged only retracts pte entries after + * putting the new hugepage in the page cache. The hugepage must be + * subsequently refaulted to install the pmd mapping for the mm. + */ + if (ops != &__anon_ops) + ops->fault(p, 0, nr_hpages * hpage_pmd_size); + + if (ops->check_huge(p, expect ? nr_hpages : 0)) + success("OK"); + else + fail("Fail"); +} + +static struct collapse_context __khugepaged_context = { + .collapse = &khugepaged_collapse, + .enforce_pte_scan_limits = true, + .name = "khugepaged", +}; + +static struct collapse_context __madvise_context = { + .collapse = &madvise_collapse, + .enforce_pte_scan_limits = false, + .name = "madvise", +}; + +static bool is_tmpfs(struct mem_ops *ops) +{ + return ops == &__file_ops && finfo.type == VMA_SHMEM; +} + +static void alloc_at_fault(void) +{ + struct settings settings = *current_settings(); + char *p; + + settings.thp_enabled = THP_ALWAYS; + push_settings(&settings); + + p = alloc_mapping(1); + *p = 1; + printf("Allocate huge page on fault..."); + if (check_huge_anon(p, 1, hpage_pmd_size)) + success("OK"); + else + fail("Fail"); + + pop_settings(); + + madvise(p, page_size, MADV_DONTNEED); + printf("Split huge PMD on MADV_DONTNEED..."); + if (check_huge_anon(p, 0, hpage_pmd_size)) + success("OK"); + else + fail("Fail"); + munmap(p, hpage_pmd_size); +} + +static void collapse_full(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + int nr_hpages = 4; + unsigned long size = nr_hpages * hpage_pmd_size; + + p = ops->setup_area(nr_hpages); + ops->fault(p, 0, size); + c->collapse("Collapse multiple fully populated PTE table", p, nr_hpages, + ops, true); + validate_memory(p, 0, size); + ops->cleanup_area(p, size); +} + +static void collapse_empty(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + + p = ops->setup_area(1); + c->collapse("Do not collapse empty PTE table", p, 1, ops, false); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_single_pte_entry(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + + p = ops->setup_area(1); + ops->fault(p, 0, page_size); + c->collapse("Collapse PTE table with single PTE entry present", p, + 1, ops, true); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_max_ptes_none(struct collapse_context *c, struct mem_ops *ops) +{ + int max_ptes_none = hpage_pmd_nr / 2; + struct settings settings = *current_settings(); + void *p; + + settings.khugepaged.max_ptes_none = max_ptes_none; + push_settings(&settings); + + p = ops->setup_area(1); + + if (is_tmpfs(ops)) { + /* shmem pages always in the page cache */ + printf("tmpfs..."); + skip("Skip"); + goto skip; + } + + ops->fault(p, 0, (hpage_pmd_nr - max_ptes_none - 1) * page_size); + c->collapse("Maybe collapse with max_ptes_none exceeded", p, 1, + ops, !c->enforce_pte_scan_limits); + validate_memory(p, 0, (hpage_pmd_nr - max_ptes_none - 1) * page_size); + + if (c->enforce_pte_scan_limits) { + ops->fault(p, 0, (hpage_pmd_nr - max_ptes_none) * page_size); + c->collapse("Collapse with max_ptes_none PTEs empty", p, 1, ops, + true); + validate_memory(p, 0, + (hpage_pmd_nr - max_ptes_none) * page_size); + } +skip: + ops->cleanup_area(p, hpage_pmd_size); + pop_settings(); +} + +static void collapse_swapin_single_pte(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + + p = ops->setup_area(1); + ops->fault(p, 0, hpage_pmd_size); + + printf("Swapout one page..."); + if (madvise(p, page_size, MADV_PAGEOUT)) { + perror("madvise(MADV_PAGEOUT)"); + exit(EXIT_FAILURE); + } + if (check_swap(p, page_size)) { + success("OK"); + } else { + fail("Fail"); + goto out; + } + + c->collapse("Collapse with swapping in single PTE entry", p, 1, ops, + true); + validate_memory(p, 0, hpage_pmd_size); +out: + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_max_ptes_swap(struct collapse_context *c, struct mem_ops *ops) +{ + int max_ptes_swap = read_num("khugepaged/max_ptes_swap"); + void *p; + + p = ops->setup_area(1); + ops->fault(p, 0, hpage_pmd_size); + + printf("Swapout %d of %d pages...", max_ptes_swap + 1, hpage_pmd_nr); + if (madvise(p, (max_ptes_swap + 1) * page_size, MADV_PAGEOUT)) { + perror("madvise(MADV_PAGEOUT)"); + exit(EXIT_FAILURE); + } + if (check_swap(p, (max_ptes_swap + 1) * page_size)) { + success("OK"); + } else { + fail("Fail"); + goto out; + } + + c->collapse("Maybe collapse with max_ptes_swap exceeded", p, 1, ops, + !c->enforce_pte_scan_limits); + validate_memory(p, 0, hpage_pmd_size); + + if (c->enforce_pte_scan_limits) { + ops->fault(p, 0, hpage_pmd_size); + printf("Swapout %d of %d pages...", max_ptes_swap, + hpage_pmd_nr); + if (madvise(p, max_ptes_swap * page_size, MADV_PAGEOUT)) { + perror("madvise(MADV_PAGEOUT)"); + exit(EXIT_FAILURE); + } + if (check_swap(p, max_ptes_swap * page_size)) { + success("OK"); + } else { + fail("Fail"); + goto out; + } + + c->collapse("Collapse with max_ptes_swap pages swapped out", p, + 1, ops, true); + validate_memory(p, 0, hpage_pmd_size); + } +out: + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_single_pte_entry_compound(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + + p = alloc_hpage(ops); + + if (is_tmpfs(ops)) { + /* MADV_DONTNEED won't evict tmpfs pages */ + printf("tmpfs..."); + skip("Skip"); + goto skip; + } + + madvise(p, hpage_pmd_size, MADV_NOHUGEPAGE); + printf("Split huge page leaving single PTE mapping compound page..."); + madvise(p + page_size, hpage_pmd_size - page_size, MADV_DONTNEED); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + c->collapse("Collapse PTE table with single PTE mapping compound page", + p, 1, ops, true); + validate_memory(p, 0, page_size); +skip: + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_full_of_compound(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + + p = alloc_hpage(ops); + printf("Split huge page leaving single PTE page table full of compound pages..."); + madvise(p, page_size, MADV_NOHUGEPAGE); + madvise(p, hpage_pmd_size, MADV_NOHUGEPAGE); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + c->collapse("Collapse PTE table full of compound pages", p, 1, ops, + true); + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_compound_extreme(struct collapse_context *c, struct mem_ops *ops) +{ + void *p; + int i; + + p = ops->setup_area(1); + for (i = 0; i < hpage_pmd_nr; i++) { + printf("\rConstruct PTE page table full of different PTE-mapped compound pages %3d/%d...", + i + 1, hpage_pmd_nr); + + madvise(BASE_ADDR, hpage_pmd_size, MADV_HUGEPAGE); + ops->fault(BASE_ADDR, 0, hpage_pmd_size); + if (!ops->check_huge(BASE_ADDR, 1)) { + printf("Failed to allocate huge page\n"); + exit(EXIT_FAILURE); + } + madvise(BASE_ADDR, hpage_pmd_size, MADV_NOHUGEPAGE); + + p = mremap(BASE_ADDR - i * page_size, + i * page_size + hpage_pmd_size, + (i + 1) * page_size, + MREMAP_MAYMOVE | MREMAP_FIXED, + BASE_ADDR + 2 * hpage_pmd_size); + if (p == MAP_FAILED) { + perror("mremap+unmap"); + exit(EXIT_FAILURE); + } + + p = mremap(BASE_ADDR + 2 * hpage_pmd_size, + (i + 1) * page_size, + (i + 1) * page_size + hpage_pmd_size, + MREMAP_MAYMOVE | MREMAP_FIXED, + BASE_ADDR - (i + 1) * page_size); + if (p == MAP_FAILED) { + perror("mremap+alloc"); + exit(EXIT_FAILURE); + } + } + + ops->cleanup_area(BASE_ADDR, hpage_pmd_size); + ops->fault(p, 0, hpage_pmd_size); + if (!ops->check_huge(p, 1)) + success("OK"); + else + fail("Fail"); + + c->collapse("Collapse PTE table full of different compound pages", p, 1, + ops, true); + + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_fork(struct collapse_context *c, struct mem_ops *ops) +{ + int wstatus; + void *p; + + p = ops->setup_area(1); + + printf("Allocate small page..."); + ops->fault(p, 0, page_size); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + printf("Share small page over fork()..."); + if (!fork()) { + /* Do not touch settings on child exit */ + skip_settings_restore = true; + exit_status = 0; + + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + ops->fault(p, page_size, 2 * page_size); + c->collapse("Collapse PTE table with single page shared with parent process", + p, 1, ops, true); + + validate_memory(p, 0, page_size); + ops->cleanup_area(p, hpage_pmd_size); + exit(exit_status); + } + + wait(&wstatus); + exit_status += WEXITSTATUS(wstatus); + + printf("Check if parent still has small page..."); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + validate_memory(p, 0, page_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_fork_compound(struct collapse_context *c, struct mem_ops *ops) +{ + int wstatus; + void *p; + + p = alloc_hpage(ops); + printf("Share huge page over fork()..."); + if (!fork()) { + /* Do not touch settings on child exit */ + skip_settings_restore = true; + exit_status = 0; + + if (ops->check_huge(p, 1)) + success("OK"); + else + fail("Fail"); + + printf("Split huge page PMD in child process..."); + madvise(p, page_size, MADV_NOHUGEPAGE); + madvise(p, hpage_pmd_size, MADV_NOHUGEPAGE); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + ops->fault(p, 0, page_size); + + write_num("khugepaged/max_ptes_shared", hpage_pmd_nr - 1); + c->collapse("Collapse PTE table full of compound pages in child", + p, 1, ops, true); + write_num("khugepaged/max_ptes_shared", + current_settings()->khugepaged.max_ptes_shared); + + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); + exit(exit_status); + } + + wait(&wstatus); + exit_status += WEXITSTATUS(wstatus); + + printf("Check if parent still has huge page..."); + if (ops->check_huge(p, 1)) + success("OK"); + else + fail("Fail"); + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void collapse_max_ptes_shared(struct collapse_context *c, struct mem_ops *ops) +{ + int max_ptes_shared = read_num("khugepaged/max_ptes_shared"); + int wstatus; + void *p; + + p = alloc_hpage(ops); + printf("Share huge page over fork()..."); + if (!fork()) { + /* Do not touch settings on child exit */ + skip_settings_restore = true; + exit_status = 0; + + if (ops->check_huge(p, 1)) + success("OK"); + else + fail("Fail"); + + printf("Trigger CoW on page %d of %d...", + hpage_pmd_nr - max_ptes_shared - 1, hpage_pmd_nr); + ops->fault(p, 0, (hpage_pmd_nr - max_ptes_shared - 1) * page_size); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + c->collapse("Maybe collapse with max_ptes_shared exceeded", p, + 1, ops, !c->enforce_pte_scan_limits); + + if (c->enforce_pte_scan_limits) { + printf("Trigger CoW on page %d of %d...", + hpage_pmd_nr - max_ptes_shared, hpage_pmd_nr); + ops->fault(p, 0, (hpage_pmd_nr - max_ptes_shared) * + page_size); + if (ops->check_huge(p, 0)) + success("OK"); + else + fail("Fail"); + + c->collapse("Collapse with max_ptes_shared PTEs shared", + p, 1, ops, true); + } + + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); + exit(exit_status); + } + + wait(&wstatus); + exit_status += WEXITSTATUS(wstatus); + + printf("Check if parent still has huge page..."); + if (ops->check_huge(p, 1)) + success("OK"); + else + fail("Fail"); + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +static void madvise_collapse_existing_thps(struct collapse_context *c, + struct mem_ops *ops) +{ + void *p; + + p = ops->setup_area(1); + ops->fault(p, 0, hpage_pmd_size); + c->collapse("Collapse fully populated PTE table...", p, 1, ops, true); + validate_memory(p, 0, hpage_pmd_size); + + /* c->collapse() will find a hugepage and complain - call directly. */ + __madvise_collapse("Re-collapse PMD-mapped hugepage", p, 1, ops, true); + validate_memory(p, 0, hpage_pmd_size); + ops->cleanup_area(p, hpage_pmd_size); +} + +/* + * Test race with khugepaged where page tables have been retracted and + * pmd cleared. + */ +static void madvise_retracted_page_tables(struct collapse_context *c, + struct mem_ops *ops) +{ + void *p; + int nr_hpages = 1; + unsigned long size = nr_hpages * hpage_pmd_size; + + p = ops->setup_area(nr_hpages); + ops->fault(p, 0, size); + + /* Let khugepaged collapse and leave pmd cleared */ + if (wait_for_scan("Collapse and leave PMD cleared", p, nr_hpages, + ops)) { + fail("Timeout"); + return; + } + success("OK"); + c->collapse("Install huge PMD from page cache", p, nr_hpages, ops, + true); + validate_memory(p, 0, size); + ops->cleanup_area(p, size); +} + +static void usage(void) +{ + fprintf(stderr, "\nUsage: ./khugepaged [dir]\n\n"); + fprintf(stderr, "\t\t: :\n"); + fprintf(stderr, "\t\t: [all|khugepaged|madvise]\n"); + fprintf(stderr, "\t\t: [all|anon|file|shmem]\n"); + fprintf(stderr, "\n\t\"file,all\" mem_type requires [dir] argument\n"); + fprintf(stderr, "\n\t\"file,all\" mem_type requires kernel built with\n"); + fprintf(stderr, "\tCONFIG_READ_ONLY_THP_FOR_FS=y\n"); + fprintf(stderr, "\n\tif [dir] is a (sub)directory of a tmpfs mount, tmpfs must be\n"); + fprintf(stderr, "\tmounted with huge=madvise option for khugepaged tests to work\n"); + exit(1); +} + +static void parse_test_type(int argc, const char **argv) +{ + char *buf; + const char *token; + + if (argc == 1) { + /* Backwards compatibility */ + khugepaged_context = &__khugepaged_context; + madvise_context = &__madvise_context; + anon_ops = &__anon_ops; + return; + } + + buf = strdup(argv[1]); + token = strsep(&buf, ":"); + + if (!strcmp(token, "all")) { + khugepaged_context = &__khugepaged_context; + madvise_context = &__madvise_context; + } else if (!strcmp(token, "khugepaged")) { + khugepaged_context = &__khugepaged_context; + } else if (!strcmp(token, "madvise")) { + madvise_context = &__madvise_context; + } else { + usage(); + } + + if (!buf) + usage(); + + if (!strcmp(buf, "all")) { + file_ops = &__file_ops; + anon_ops = &__anon_ops; + shmem_ops = &__shmem_ops; + } else if (!strcmp(buf, "anon")) { + anon_ops = &__anon_ops; + } else if (!strcmp(buf, "file")) { + file_ops = &__file_ops; + } else if (!strcmp(buf, "shmem")) { + shmem_ops = &__shmem_ops; + } else { + usage(); + } + + if (!file_ops) + return; + + if (argc != 3) + usage(); +} + +int main(int argc, const char **argv) +{ + struct settings default_settings = { + .thp_enabled = THP_MADVISE, + .thp_defrag = THP_DEFRAG_ALWAYS, + .shmem_enabled = SHMEM_ADVISE, + .use_zero_page = 0, + .khugepaged = { + .defrag = 1, + .alloc_sleep_millisecs = 10, + .scan_sleep_millisecs = 10, + }, + /* + * When testing file-backed memory, the collapse path + * looks at how many pages are found in the page cache, not + * what pages are mapped. Disable read ahead optimization so + * pages don't find their way into the page cache unless + * we mem_ops->fault() them in. + */ + .read_ahead_kb = 0, + }; + + parse_test_type(argc, argv); + + if (file_ops) + get_finfo(argv[2]); + + setbuf(stdout, NULL); + + page_size = getpagesize(); + hpage_pmd_size = read_pmd_pagesize(); + hpage_pmd_nr = hpage_pmd_size / page_size; + + default_settings.khugepaged.max_ptes_none = hpage_pmd_nr - 1; + default_settings.khugepaged.max_ptes_swap = hpage_pmd_nr / 8; + default_settings.khugepaged.max_ptes_shared = hpage_pmd_nr / 2; + default_settings.khugepaged.pages_to_scan = hpage_pmd_nr * 8; + + save_settings(); + push_settings(&default_settings); + + alloc_at_fault(); + +#define TEST(t, c, o) do { \ + if (c && o) { \ + printf("\nRun test: " #t " (%s:%s)\n", c->name, o->name); \ + t(c, o); \ + } \ + } while (0) + + TEST(collapse_full, khugepaged_context, anon_ops); + TEST(collapse_full, khugepaged_context, file_ops); + TEST(collapse_full, khugepaged_context, shmem_ops); + TEST(collapse_full, madvise_context, anon_ops); + TEST(collapse_full, madvise_context, file_ops); + TEST(collapse_full, madvise_context, shmem_ops); + + TEST(collapse_empty, khugepaged_context, anon_ops); + TEST(collapse_empty, madvise_context, anon_ops); + + TEST(collapse_single_pte_entry, khugepaged_context, anon_ops); + TEST(collapse_single_pte_entry, khugepaged_context, file_ops); + TEST(collapse_single_pte_entry, khugepaged_context, shmem_ops); + TEST(collapse_single_pte_entry, madvise_context, anon_ops); + TEST(collapse_single_pte_entry, madvise_context, file_ops); + TEST(collapse_single_pte_entry, madvise_context, shmem_ops); + + TEST(collapse_max_ptes_none, khugepaged_context, anon_ops); + TEST(collapse_max_ptes_none, khugepaged_context, file_ops); + TEST(collapse_max_ptes_none, madvise_context, anon_ops); + TEST(collapse_max_ptes_none, madvise_context, file_ops); + + TEST(collapse_single_pte_entry_compound, khugepaged_context, anon_ops); + TEST(collapse_single_pte_entry_compound, khugepaged_context, file_ops); + TEST(collapse_single_pte_entry_compound, madvise_context, anon_ops); + TEST(collapse_single_pte_entry_compound, madvise_context, file_ops); + + TEST(collapse_full_of_compound, khugepaged_context, anon_ops); + TEST(collapse_full_of_compound, khugepaged_context, file_ops); + TEST(collapse_full_of_compound, khugepaged_context, shmem_ops); + TEST(collapse_full_of_compound, madvise_context, anon_ops); + TEST(collapse_full_of_compound, madvise_context, file_ops); + TEST(collapse_full_of_compound, madvise_context, shmem_ops); + + TEST(collapse_compound_extreme, khugepaged_context, anon_ops); + TEST(collapse_compound_extreme, madvise_context, anon_ops); + + TEST(collapse_swapin_single_pte, khugepaged_context, anon_ops); + TEST(collapse_swapin_single_pte, madvise_context, anon_ops); + + TEST(collapse_max_ptes_swap, khugepaged_context, anon_ops); + TEST(collapse_max_ptes_swap, madvise_context, anon_ops); + + TEST(collapse_fork, khugepaged_context, anon_ops); + TEST(collapse_fork, madvise_context, anon_ops); + + TEST(collapse_fork_compound, khugepaged_context, anon_ops); + TEST(collapse_fork_compound, madvise_context, anon_ops); + + TEST(collapse_max_ptes_shared, khugepaged_context, anon_ops); + TEST(collapse_max_ptes_shared, madvise_context, anon_ops); + + TEST(madvise_collapse_existing_thps, madvise_context, anon_ops); + TEST(madvise_collapse_existing_thps, madvise_context, file_ops); + TEST(madvise_collapse_existing_thps, madvise_context, shmem_ops); + + TEST(madvise_retracted_page_tables, madvise_context, file_ops); + TEST(madvise_retracted_page_tables, madvise_context, shmem_ops); + + restore_settings(0); +} diff --git a/tools/testing/selftests/vm/ksm_tests.c b/tools/testing/selftests/vm/ksm_tests.c new file mode 100644 index 000000000..0d85be235 --- /dev/null +++ b/tools/testing/selftests/vm/ksm_tests.c @@ -0,0 +1,777 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" +#include +#include "util.h" + +#define KSM_SYSFS_PATH "/sys/kernel/mm/ksm/" +#define KSM_FP(s) (KSM_SYSFS_PATH s) +#define KSM_SCAN_LIMIT_SEC_DEFAULT 120 +#define KSM_PAGE_COUNT_DEFAULT 10l +#define KSM_PROT_STR_DEFAULT "rw" +#define KSM_USE_ZERO_PAGES_DEFAULT false +#define KSM_MERGE_ACROSS_NODES_DEFAULT true +#define MB (1ul << 20) + +struct ksm_sysfs { + unsigned long max_page_sharing; + unsigned long merge_across_nodes; + unsigned long pages_to_scan; + unsigned long run; + unsigned long sleep_millisecs; + unsigned long stable_node_chains_prune_millisecs; + unsigned long use_zero_pages; +}; + +enum ksm_test_name { + CHECK_KSM_MERGE, + CHECK_KSM_UNMERGE, + CHECK_KSM_ZERO_PAGE_MERGE, + CHECK_KSM_NUMA_MERGE, + KSM_MERGE_TIME, + KSM_MERGE_TIME_HUGE_PAGES, + KSM_COW_TIME +}; + +static int ksm_write_sysfs(const char *file_path, unsigned long val) +{ + FILE *f = fopen(file_path, "w"); + + if (!f) { + fprintf(stderr, "f %s\n", file_path); + perror("fopen"); + return 1; + } + if (fprintf(f, "%lu", val) < 0) { + perror("fprintf"); + fclose(f); + return 1; + } + fclose(f); + + return 0; +} + +static int ksm_read_sysfs(const char *file_path, unsigned long *val) +{ + FILE *f = fopen(file_path, "r"); + + if (!f) { + fprintf(stderr, "f %s\n", file_path); + perror("fopen"); + return 1; + } + if (fscanf(f, "%lu", val) != 1) { + perror("fscanf"); + fclose(f); + return 1; + } + fclose(f); + + return 0; +} + +static int str_to_prot(char *prot_str) +{ + int prot = 0; + + if ((strchr(prot_str, 'r')) != NULL) + prot |= PROT_READ; + if ((strchr(prot_str, 'w')) != NULL) + prot |= PROT_WRITE; + if ((strchr(prot_str, 'x')) != NULL) + prot |= PROT_EXEC; + + return prot; +} + +static void print_help(void) +{ + printf("usage: ksm_tests [-h] [-a prot] [-p page_count] [-l timeout]\n" + "[-z use_zero_pages] [-m merge_across_nodes] [-s size]\n"); + + printf("Supported :\n" + " -M (page merging)\n" + " -Z (zero pages merging)\n" + " -N (merging of pages in different NUMA nodes)\n" + " -U (page unmerging)\n" + " -P evaluate merging time and speed.\n" + " For this test, the size of duplicated memory area (in MiB)\n" + " must be provided using -s option\n" + " -H evaluate merging time and speed of area allocated mostly with huge pages\n" + " For this test, the size of duplicated memory area (in MiB)\n" + " must be provided using -s option\n" + " -C evaluate the time required to break COW of merged pages.\n\n"); + + printf(" -a: specify the access protections of pages.\n" + " must be of the form [rwx].\n" + " Default: %s\n", KSM_PROT_STR_DEFAULT); + printf(" -p: specify the number of pages to test.\n" + " Default: %ld\n", KSM_PAGE_COUNT_DEFAULT); + printf(" -l: limit the maximum running time (in seconds) for a test.\n" + " Default: %d seconds\n", KSM_SCAN_LIMIT_SEC_DEFAULT); + printf(" -z: change use_zero_pages tunable\n" + " Default: %d\n", KSM_USE_ZERO_PAGES_DEFAULT); + printf(" -m: change merge_across_nodes tunable\n" + " Default: %d\n", KSM_MERGE_ACROSS_NODES_DEFAULT); + printf(" -s: the size of duplicated memory area (in MiB)\n"); + + exit(0); +} + +static void *allocate_memory(void *ptr, int prot, int mapping, char data, size_t map_size) +{ + void *map_ptr = mmap(ptr, map_size, PROT_WRITE, mapping, -1, 0); + + if (!map_ptr) { + perror("mmap"); + return NULL; + } + memset(map_ptr, data, map_size); + if (mprotect(map_ptr, map_size, prot)) { + perror("mprotect"); + munmap(map_ptr, map_size); + return NULL; + } + + return map_ptr; +} + +static int ksm_do_scan(int scan_count, struct timespec start_time, int timeout) +{ + struct timespec cur_time; + unsigned long cur_scan, init_scan; + + if (ksm_read_sysfs(KSM_FP("full_scans"), &init_scan)) + return 1; + cur_scan = init_scan; + + while (cur_scan < init_scan + scan_count) { + if (ksm_read_sysfs(KSM_FP("full_scans"), &cur_scan)) + return 1; + if (clock_gettime(CLOCK_MONOTONIC_RAW, &cur_time)) { + perror("clock_gettime"); + return 1; + } + if ((cur_time.tv_sec - start_time.tv_sec) > timeout) { + printf("Scan time limit exceeded\n"); + return 1; + } + } + + return 0; +} + +static int ksm_merge_pages(void *addr, size_t size, struct timespec start_time, int timeout) +{ + if (madvise(addr, size, MADV_MERGEABLE)) { + perror("madvise"); + return 1; + } + if (ksm_write_sysfs(KSM_FP("run"), 1)) + return 1; + + /* Since merging occurs only after 2 scans, make sure to get at least 2 full scans */ + if (ksm_do_scan(2, start_time, timeout)) + return 1; + + return 0; +} + +static bool assert_ksm_pages_count(long dupl_page_count) +{ + unsigned long max_page_sharing, pages_sharing, pages_shared; + + if (ksm_read_sysfs(KSM_FP("pages_shared"), &pages_shared) || + ksm_read_sysfs(KSM_FP("pages_sharing"), &pages_sharing) || + ksm_read_sysfs(KSM_FP("max_page_sharing"), &max_page_sharing)) + return false; + + /* + * Since there must be at least 2 pages for merging and 1 page can be + * shared with the limited number of pages (max_page_sharing), sometimes + * there are 'leftover' pages that cannot be merged. For example, if there + * are 11 pages and max_page_sharing = 10, then only 10 pages will be + * merged and the 11th page won't be affected. As a result, when the number + * of duplicate pages is divided by max_page_sharing and the remainder is 1, + * pages_shared and pages_sharing values will be equal between dupl_page_count + * and dupl_page_count - 1. + */ + if (dupl_page_count % max_page_sharing == 1 || dupl_page_count % max_page_sharing == 0) { + if (pages_shared == dupl_page_count / max_page_sharing && + pages_sharing == pages_shared * (max_page_sharing - 1)) + return true; + } else { + if (pages_shared == (dupl_page_count / max_page_sharing + 1) && + pages_sharing == dupl_page_count - pages_shared) + return true; + } + + return false; +} + +static int ksm_save_def(struct ksm_sysfs *ksm_sysfs) +{ + if (ksm_read_sysfs(KSM_FP("max_page_sharing"), &ksm_sysfs->max_page_sharing) || + numa_available() ? 0 : + ksm_read_sysfs(KSM_FP("merge_across_nodes"), &ksm_sysfs->merge_across_nodes) || + ksm_read_sysfs(KSM_FP("sleep_millisecs"), &ksm_sysfs->sleep_millisecs) || + ksm_read_sysfs(KSM_FP("pages_to_scan"), &ksm_sysfs->pages_to_scan) || + ksm_read_sysfs(KSM_FP("run"), &ksm_sysfs->run) || + ksm_read_sysfs(KSM_FP("stable_node_chains_prune_millisecs"), + &ksm_sysfs->stable_node_chains_prune_millisecs) || + ksm_read_sysfs(KSM_FP("use_zero_pages"), &ksm_sysfs->use_zero_pages)) + return 1; + + return 0; +} + +static int ksm_restore(struct ksm_sysfs *ksm_sysfs) +{ + if (ksm_write_sysfs(KSM_FP("max_page_sharing"), ksm_sysfs->max_page_sharing) || + numa_available() ? 0 : + ksm_write_sysfs(KSM_FP("merge_across_nodes"), ksm_sysfs->merge_across_nodes) || + ksm_write_sysfs(KSM_FP("pages_to_scan"), ksm_sysfs->pages_to_scan) || + ksm_write_sysfs(KSM_FP("run"), ksm_sysfs->run) || + ksm_write_sysfs(KSM_FP("sleep_millisecs"), ksm_sysfs->sleep_millisecs) || + ksm_write_sysfs(KSM_FP("stable_node_chains_prune_millisecs"), + ksm_sysfs->stable_node_chains_prune_millisecs) || + ksm_write_sysfs(KSM_FP("use_zero_pages"), ksm_sysfs->use_zero_pages)) + return 1; + + return 0; +} + +static int check_ksm_merge(int mapping, int prot, long page_count, int timeout, size_t page_size) +{ + void *map_ptr; + struct timespec start_time; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + + /* fill pages with the same data and merge them */ + map_ptr = allocate_memory(NULL, prot, mapping, '*', page_size * page_count); + if (!map_ptr) + return KSFT_FAIL; + + if (ksm_merge_pages(map_ptr, page_size * page_count, start_time, timeout)) + goto err_out; + + /* verify that the right number of pages are merged */ + if (assert_ksm_pages_count(page_count)) { + printf("OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_PASS; + } + +err_out: + printf("Not OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_FAIL; +} + +static int check_ksm_unmerge(int mapping, int prot, int timeout, size_t page_size) +{ + void *map_ptr; + struct timespec start_time; + int page_count = 2; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + + /* fill pages with the same data and merge them */ + map_ptr = allocate_memory(NULL, prot, mapping, '*', page_size * page_count); + if (!map_ptr) + return KSFT_FAIL; + + if (ksm_merge_pages(map_ptr, page_size * page_count, start_time, timeout)) + goto err_out; + + /* change 1 byte in each of the 2 pages -- KSM must automatically unmerge them */ + memset(map_ptr, '-', 1); + memset(map_ptr + page_size, '+', 1); + + /* get at least 1 scan, so KSM can detect that the pages were modified */ + if (ksm_do_scan(1, start_time, timeout)) + goto err_out; + + /* check that unmerging was successful and 0 pages are currently merged */ + if (assert_ksm_pages_count(0)) { + printf("OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_PASS; + } + +err_out: + printf("Not OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_FAIL; +} + +static int check_ksm_zero_page_merge(int mapping, int prot, long page_count, int timeout, + bool use_zero_pages, size_t page_size) +{ + void *map_ptr; + struct timespec start_time; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + + if (ksm_write_sysfs(KSM_FP("use_zero_pages"), use_zero_pages)) + return KSFT_FAIL; + + /* fill pages with zero and try to merge them */ + map_ptr = allocate_memory(NULL, prot, mapping, 0, page_size * page_count); + if (!map_ptr) + return KSFT_FAIL; + + if (ksm_merge_pages(map_ptr, page_size * page_count, start_time, timeout)) + goto err_out; + + /* + * verify that the right number of pages are merged: + * 1) if use_zero_pages is set to 1, empty pages are merged + * with the kernel zero page instead of with each other; + * 2) if use_zero_pages is set to 0, empty pages are not treated specially + * and merged as usual. + */ + if (use_zero_pages && !assert_ksm_pages_count(0)) + goto err_out; + else if (!use_zero_pages && !assert_ksm_pages_count(page_count)) + goto err_out; + + printf("OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_PASS; + +err_out: + printf("Not OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_FAIL; +} + +static int get_next_mem_node(int node) +{ + + long node_size; + int mem_node = 0; + int i, max_node = numa_max_node(); + + for (i = node + 1; i <= max_node + node; i++) { + mem_node = i % (max_node + 1); + node_size = numa_node_size(mem_node, NULL); + if (node_size > 0) + break; + } + return mem_node; +} + +static int get_first_mem_node(void) +{ + return get_next_mem_node(numa_max_node()); +} + +static int check_ksm_numa_merge(int mapping, int prot, int timeout, bool merge_across_nodes, + size_t page_size) +{ + void *numa1_map_ptr, *numa2_map_ptr; + struct timespec start_time; + int page_count = 2; + int first_node; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + + if (numa_available() < 0) { + perror("NUMA support not enabled"); + return KSFT_SKIP; + } + if (numa_num_configured_nodes() <= 1) { + printf("At least 2 NUMA nodes must be available\n"); + return KSFT_SKIP; + } + if (ksm_write_sysfs(KSM_FP("merge_across_nodes"), merge_across_nodes)) + return KSFT_FAIL; + + /* allocate 2 pages in 2 different NUMA nodes and fill them with the same data */ + first_node = get_first_mem_node(); + numa1_map_ptr = numa_alloc_onnode(page_size, first_node); + numa2_map_ptr = numa_alloc_onnode(page_size, get_next_mem_node(first_node)); + if (!numa1_map_ptr || !numa2_map_ptr) { + perror("numa_alloc_onnode"); + return KSFT_FAIL; + } + + memset(numa1_map_ptr, '*', page_size); + memset(numa2_map_ptr, '*', page_size); + + /* try to merge the pages */ + if (ksm_merge_pages(numa1_map_ptr, page_size, start_time, timeout) || + ksm_merge_pages(numa2_map_ptr, page_size, start_time, timeout)) + goto err_out; + + /* + * verify that the right number of pages are merged: + * 1) if merge_across_nodes was enabled, 2 duplicate pages will be merged; + * 2) if merge_across_nodes = 0, there must be 0 merged pages, since there is + * only 1 unique page in each node and they can't be shared. + */ + if (merge_across_nodes && !assert_ksm_pages_count(page_count)) + goto err_out; + else if (!merge_across_nodes && !assert_ksm_pages_count(0)) + goto err_out; + + numa_free(numa1_map_ptr, page_size); + numa_free(numa2_map_ptr, page_size); + printf("OK\n"); + return KSFT_PASS; + +err_out: + numa_free(numa1_map_ptr, page_size); + numa_free(numa2_map_ptr, page_size); + printf("Not OK\n"); + return KSFT_FAIL; +} + +static int ksm_merge_hugepages_time(int mapping, int prot, int timeout, size_t map_size) +{ + void *map_ptr, *map_ptr_orig; + struct timespec start_time, end_time; + unsigned long scan_time_ns; + int pagemap_fd, n_normal_pages, n_huge_pages; + + map_size *= MB; + size_t len = map_size; + + len -= len % HPAGE_SIZE; + map_ptr_orig = mmap(NULL, len + HPAGE_SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_NORESERVE | MAP_PRIVATE, -1, 0); + map_ptr = map_ptr_orig + HPAGE_SIZE - (uintptr_t)map_ptr_orig % HPAGE_SIZE; + + if (map_ptr_orig == MAP_FAILED) + err(2, "initial mmap"); + + if (madvise(map_ptr, len + HPAGE_SIZE, MADV_HUGEPAGE)) + err(2, "MADV_HUGEPAGE"); + + pagemap_fd = open("/proc/self/pagemap", O_RDONLY); + if (pagemap_fd < 0) + err(2, "open pagemap"); + + n_normal_pages = 0; + n_huge_pages = 0; + for (void *p = map_ptr; p < map_ptr + len; p += HPAGE_SIZE) { + if (allocate_transhuge(p, pagemap_fd) < 0) + n_normal_pages++; + else + n_huge_pages++; + } + printf("Number of normal pages: %d\n", n_normal_pages); + printf("Number of huge pages: %d\n", n_huge_pages); + + memset(map_ptr, '*', len); + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + goto err_out; + } + if (ksm_merge_pages(map_ptr, map_size, start_time, timeout)) + goto err_out; + if (clock_gettime(CLOCK_MONOTONIC_RAW, &end_time)) { + perror("clock_gettime"); + goto err_out; + } + + scan_time_ns = (end_time.tv_sec - start_time.tv_sec) * NSEC_PER_SEC + + (end_time.tv_nsec - start_time.tv_nsec); + + printf("Total size: %lu MiB\n", map_size / MB); + printf("Total time: %ld.%09ld s\n", scan_time_ns / NSEC_PER_SEC, + scan_time_ns % NSEC_PER_SEC); + printf("Average speed: %.3f MiB/s\n", (map_size / MB) / + ((double)scan_time_ns / NSEC_PER_SEC)); + + munmap(map_ptr_orig, len + HPAGE_SIZE); + return KSFT_PASS; + +err_out: + printf("Not OK\n"); + munmap(map_ptr_orig, len + HPAGE_SIZE); + return KSFT_FAIL; +} + +static int ksm_merge_time(int mapping, int prot, int timeout, size_t map_size) +{ + void *map_ptr; + struct timespec start_time, end_time; + unsigned long scan_time_ns; + + map_size *= MB; + + map_ptr = allocate_memory(NULL, prot, mapping, '*', map_size); + if (!map_ptr) + return KSFT_FAIL; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + goto err_out; + } + if (ksm_merge_pages(map_ptr, map_size, start_time, timeout)) + goto err_out; + if (clock_gettime(CLOCK_MONOTONIC_RAW, &end_time)) { + perror("clock_gettime"); + goto err_out; + } + + scan_time_ns = (end_time.tv_sec - start_time.tv_sec) * NSEC_PER_SEC + + (end_time.tv_nsec - start_time.tv_nsec); + + printf("Total size: %lu MiB\n", map_size / MB); + printf("Total time: %ld.%09ld s\n", scan_time_ns / NSEC_PER_SEC, + scan_time_ns % NSEC_PER_SEC); + printf("Average speed: %.3f MiB/s\n", (map_size / MB) / + ((double)scan_time_ns / NSEC_PER_SEC)); + + munmap(map_ptr, map_size); + return KSFT_PASS; + +err_out: + printf("Not OK\n"); + munmap(map_ptr, map_size); + return KSFT_FAIL; +} + +static int ksm_cow_time(int mapping, int prot, int timeout, size_t page_size) +{ + void *map_ptr; + struct timespec start_time, end_time; + unsigned long cow_time_ns; + + /* page_count must be less than 2*page_size */ + size_t page_count = 4000; + + map_ptr = allocate_memory(NULL, prot, mapping, '*', page_size * page_count); + if (!map_ptr) + return KSFT_FAIL; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + for (size_t i = 0; i < page_count - 1; i = i + 2) + memset(map_ptr + page_size * i, '-', 1); + if (clock_gettime(CLOCK_MONOTONIC_RAW, &end_time)) { + perror("clock_gettime"); + return KSFT_FAIL; + } + + cow_time_ns = (end_time.tv_sec - start_time.tv_sec) * NSEC_PER_SEC + + (end_time.tv_nsec - start_time.tv_nsec); + + printf("Total size: %lu MiB\n\n", (page_size * page_count) / MB); + printf("Not merged pages:\n"); + printf("Total time: %ld.%09ld s\n", cow_time_ns / NSEC_PER_SEC, + cow_time_ns % NSEC_PER_SEC); + printf("Average speed: %.3f MiB/s\n\n", ((page_size * (page_count / 2)) / MB) / + ((double)cow_time_ns / NSEC_PER_SEC)); + + /* Create 2000 pairs of duplicate pages */ + for (size_t i = 0; i < page_count - 1; i = i + 2) { + memset(map_ptr + page_size * i, '+', i / 2 + 1); + memset(map_ptr + page_size * (i + 1), '+', i / 2 + 1); + } + if (ksm_merge_pages(map_ptr, page_size * page_count, start_time, timeout)) + goto err_out; + + if (clock_gettime(CLOCK_MONOTONIC_RAW, &start_time)) { + perror("clock_gettime"); + goto err_out; + } + for (size_t i = 0; i < page_count - 1; i = i + 2) + memset(map_ptr + page_size * i, '-', 1); + if (clock_gettime(CLOCK_MONOTONIC_RAW, &end_time)) { + perror("clock_gettime"); + goto err_out; + } + + cow_time_ns = (end_time.tv_sec - start_time.tv_sec) * NSEC_PER_SEC + + (end_time.tv_nsec - start_time.tv_nsec); + + printf("Merged pages:\n"); + printf("Total time: %ld.%09ld s\n", cow_time_ns / NSEC_PER_SEC, + cow_time_ns % NSEC_PER_SEC); + printf("Average speed: %.3f MiB/s\n", ((page_size * (page_count / 2)) / MB) / + ((double)cow_time_ns / NSEC_PER_SEC)); + + munmap(map_ptr, page_size * page_count); + return KSFT_PASS; + +err_out: + printf("Not OK\n"); + munmap(map_ptr, page_size * page_count); + return KSFT_FAIL; +} + +int main(int argc, char *argv[]) +{ + int ret, opt; + int prot = 0; + int ksm_scan_limit_sec = KSM_SCAN_LIMIT_SEC_DEFAULT; + long page_count = KSM_PAGE_COUNT_DEFAULT; + size_t page_size = sysconf(_SC_PAGESIZE); + struct ksm_sysfs ksm_sysfs_old; + int test_name = CHECK_KSM_MERGE; + bool use_zero_pages = KSM_USE_ZERO_PAGES_DEFAULT; + bool merge_across_nodes = KSM_MERGE_ACROSS_NODES_DEFAULT; + long size_MB = 0; + + while ((opt = getopt(argc, argv, "ha:p:l:z:m:s:MUZNPCH")) != -1) { + switch (opt) { + case 'a': + prot = str_to_prot(optarg); + break; + case 'p': + page_count = atol(optarg); + if (page_count <= 0) { + printf("The number of pages must be greater than 0\n"); + return KSFT_FAIL; + } + break; + case 'l': + ksm_scan_limit_sec = atoi(optarg); + if (ksm_scan_limit_sec <= 0) { + printf("Timeout value must be greater than 0\n"); + return KSFT_FAIL; + } + break; + case 'h': + print_help(); + break; + case 'z': + if (strcmp(optarg, "0") == 0) + use_zero_pages = 0; + else + use_zero_pages = 1; + break; + case 'm': + if (strcmp(optarg, "0") == 0) + merge_across_nodes = 0; + else + merge_across_nodes = 1; + break; + case 's': + size_MB = atoi(optarg); + if (size_MB <= 0) { + printf("Size must be greater than 0\n"); + return KSFT_FAIL; + } + case 'M': + break; + case 'U': + test_name = CHECK_KSM_UNMERGE; + break; + case 'Z': + test_name = CHECK_KSM_ZERO_PAGE_MERGE; + break; + case 'N': + test_name = CHECK_KSM_NUMA_MERGE; + break; + case 'P': + test_name = KSM_MERGE_TIME; + break; + case 'H': + test_name = KSM_MERGE_TIME_HUGE_PAGES; + break; + case 'C': + test_name = KSM_COW_TIME; + break; + default: + return KSFT_FAIL; + } + } + + if (prot == 0) + prot = str_to_prot(KSM_PROT_STR_DEFAULT); + + if (access(KSM_SYSFS_PATH, F_OK)) { + printf("Config KSM not enabled\n"); + return KSFT_SKIP; + } + + if (ksm_save_def(&ksm_sysfs_old)) { + printf("Cannot save default tunables\n"); + return KSFT_FAIL; + } + + if (ksm_write_sysfs(KSM_FP("run"), 2) || + ksm_write_sysfs(KSM_FP("sleep_millisecs"), 0) || + numa_available() ? 0 : + ksm_write_sysfs(KSM_FP("merge_across_nodes"), 1) || + ksm_write_sysfs(KSM_FP("pages_to_scan"), page_count)) + return KSFT_FAIL; + + switch (test_name) { + case CHECK_KSM_MERGE: + ret = check_ksm_merge(MAP_PRIVATE | MAP_ANONYMOUS, prot, page_count, + ksm_scan_limit_sec, page_size); + break; + case CHECK_KSM_UNMERGE: + ret = check_ksm_unmerge(MAP_PRIVATE | MAP_ANONYMOUS, prot, ksm_scan_limit_sec, + page_size); + break; + case CHECK_KSM_ZERO_PAGE_MERGE: + ret = check_ksm_zero_page_merge(MAP_PRIVATE | MAP_ANONYMOUS, prot, page_count, + ksm_scan_limit_sec, use_zero_pages, page_size); + break; + case CHECK_KSM_NUMA_MERGE: + ret = check_ksm_numa_merge(MAP_PRIVATE | MAP_ANONYMOUS, prot, ksm_scan_limit_sec, + merge_across_nodes, page_size); + break; + case KSM_MERGE_TIME: + if (size_MB == 0) { + printf("Option '-s' is required.\n"); + return KSFT_FAIL; + } + ret = ksm_merge_time(MAP_PRIVATE | MAP_ANONYMOUS, prot, ksm_scan_limit_sec, + size_MB); + break; + case KSM_MERGE_TIME_HUGE_PAGES: + if (size_MB == 0) { + printf("Option '-s' is required.\n"); + return KSFT_FAIL; + } + ret = ksm_merge_hugepages_time(MAP_PRIVATE | MAP_ANONYMOUS, prot, + ksm_scan_limit_sec, size_MB); + break; + case KSM_COW_TIME: + ret = ksm_cow_time(MAP_PRIVATE | MAP_ANONYMOUS, prot, ksm_scan_limit_sec, + page_size); + break; + } + + if (ksm_restore(&ksm_sysfs_old)) { + printf("Cannot restore default tunables\n"); + return KSFT_FAIL; + } + + return ret; +} diff --git a/tools/testing/selftests/vm/madv_populate.c b/tools/testing/selftests/vm/madv_populate.c new file mode 100644 index 000000000..715a42e8e --- /dev/null +++ b/tools/testing/selftests/vm/madv_populate.c @@ -0,0 +1,297 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * MADV_POPULATE_READ and MADV_POPULATE_WRITE tests + * + * Copyright 2021, Red Hat, Inc. + * + * Author(s): David Hildenbrand + */ +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" +#include "vm_util.h" + +/* + * For now, we're using 2 MiB of private anonymous memory for all tests. + */ +#define SIZE (2 * 1024 * 1024) + +static size_t pagesize; + +static bool pagemap_is_populated(int fd, char *start) +{ + uint64_t entry = pagemap_get_entry(fd, start); + + /* Present or swapped. */ + return entry & 0xc000000000000000ull; +} + +static void sense_support(void) +{ + char *addr; + int ret; + + addr = mmap(0, pagesize, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (!addr) + ksft_exit_fail_msg("mmap failed\n"); + + ret = madvise(addr, pagesize, MADV_POPULATE_READ); + if (ret) + ksft_exit_skip("MADV_POPULATE_READ is not available\n"); + + ret = madvise(addr, pagesize, MADV_POPULATE_WRITE); + if (ret) + ksft_exit_skip("MADV_POPULATE_WRITE is not available\n"); + + munmap(addr, pagesize); +} + +static void test_prot_read(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_READ, MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_READ); + ksft_test_result(!ret, "MADV_POPULATE_READ with PROT_READ\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_WRITE); + ksft_test_result(ret == -1 && errno == EINVAL, + "MADV_POPULATE_WRITE with PROT_READ\n"); + + munmap(addr, SIZE); +} + +static void test_prot_write(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_WRITE, MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_READ); + ksft_test_result(ret == -1 && errno == EINVAL, + "MADV_POPULATE_READ with PROT_WRITE\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_WRITE); + ksft_test_result(!ret, "MADV_POPULATE_WRITE with PROT_WRITE\n"); + + munmap(addr, SIZE); +} + +static void test_holes(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + ret = munmap(addr + pagesize, pagesize); + if (ret) + ksft_exit_fail_msg("munmap failed\n"); + + /* Hole in the middle */ + ret = madvise(addr, SIZE, MADV_POPULATE_READ); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_READ with holes in the middle\n"); + ret = madvise(addr, SIZE, MADV_POPULATE_WRITE); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_WRITE with holes in the middle\n"); + + /* Hole at end */ + ret = madvise(addr, 2 * pagesize, MADV_POPULATE_READ); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_READ with holes at the end\n"); + ret = madvise(addr, 2 * pagesize, MADV_POPULATE_WRITE); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_WRITE with holes at the end\n"); + + /* Hole at beginning */ + ret = madvise(addr + pagesize, pagesize, MADV_POPULATE_READ); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_READ with holes at the beginning\n"); + ret = madvise(addr + pagesize, pagesize, MADV_POPULATE_WRITE); + ksft_test_result(ret == -1 && errno == ENOMEM, + "MADV_POPULATE_WRITE with holes at the beginning\n"); + + munmap(addr, SIZE); +} + +static bool range_is_populated(char *start, ssize_t size) +{ + int fd = open("/proc/self/pagemap", O_RDONLY); + bool ret = true; + + if (fd < 0) + ksft_exit_fail_msg("opening pagemap failed\n"); + for (; size > 0 && ret; size -= pagesize, start += pagesize) + if (!pagemap_is_populated(fd, start)) + ret = false; + close(fd); + return ret; +} + +static bool range_is_not_populated(char *start, ssize_t size) +{ + int fd = open("/proc/self/pagemap", O_RDONLY); + bool ret = true; + + if (fd < 0) + ksft_exit_fail_msg("opening pagemap failed\n"); + for (; size > 0 && ret; size -= pagesize, start += pagesize) + if (pagemap_is_populated(fd, start)) + ret = false; + close(fd); + return ret; +} + +static void test_populate_read(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + ksft_test_result(range_is_not_populated(addr, SIZE), + "range initially not populated\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_READ); + ksft_test_result(!ret, "MADV_POPULATE_READ\n"); + ksft_test_result(range_is_populated(addr, SIZE), + "range is populated\n"); + + munmap(addr, SIZE); +} + +static void test_populate_write(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + ksft_test_result(range_is_not_populated(addr, SIZE), + "range initially not populated\n"); + + ret = madvise(addr, SIZE, MADV_POPULATE_WRITE); + ksft_test_result(!ret, "MADV_POPULATE_WRITE\n"); + ksft_test_result(range_is_populated(addr, SIZE), + "range is populated\n"); + + munmap(addr, SIZE); +} + +static bool range_is_softdirty(char *start, ssize_t size) +{ + int fd = open("/proc/self/pagemap", O_RDONLY); + bool ret = true; + + if (fd < 0) + ksft_exit_fail_msg("opening pagemap failed\n"); + for (; size > 0 && ret; size -= pagesize, start += pagesize) + if (!pagemap_is_softdirty(fd, start)) + ret = false; + close(fd); + return ret; +} + +static bool range_is_not_softdirty(char *start, ssize_t size) +{ + int fd = open("/proc/self/pagemap", O_RDONLY); + bool ret = true; + + if (fd < 0) + ksft_exit_fail_msg("opening pagemap failed\n"); + for (; size > 0 && ret; size -= pagesize, start += pagesize) + if (pagemap_is_softdirty(fd, start)) + ret = false; + close(fd); + return ret; +} + +static void test_softdirty(void) +{ + char *addr; + int ret; + + ksft_print_msg("[RUN] %s\n", __func__); + + addr = mmap(0, SIZE, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + if (addr == MAP_FAILED) + ksft_exit_fail_msg("mmap failed\n"); + + /* Clear any softdirty bits. */ + clear_softdirty(); + ksft_test_result(range_is_not_softdirty(addr, SIZE), + "range is not softdirty\n"); + + /* Populating READ should set softdirty. */ + ret = madvise(addr, SIZE, MADV_POPULATE_READ); + ksft_test_result(!ret, "MADV_POPULATE_READ\n"); + ksft_test_result(range_is_not_softdirty(addr, SIZE), + "range is not softdirty\n"); + + /* Populating WRITE should set softdirty. */ + ret = madvise(addr, SIZE, MADV_POPULATE_WRITE); + ksft_test_result(!ret, "MADV_POPULATE_WRITE\n"); + ksft_test_result(range_is_softdirty(addr, SIZE), + "range is softdirty\n"); + + munmap(addr, SIZE); +} + +int main(int argc, char **argv) +{ + int err; + + pagesize = getpagesize(); + + ksft_print_header(); + ksft_set_plan(21); + + sense_support(); + test_prot_read(); + test_prot_write(); + test_holes(); + test_populate_read(); + test_populate_write(); + test_softdirty(); + + err = ksft_get_fail_cnt(); + if (err) + ksft_exit_fail_msg("%d out of %d tests failed\n", + err, ksft_test_num()); + return ksft_exit_pass(); +} diff --git a/tools/testing/selftests/vm/map_fixed_noreplace.c b/tools/testing/selftests/vm/map_fixed_noreplace.c new file mode 100644 index 000000000..eed44322d --- /dev/null +++ b/tools/testing/selftests/vm/map_fixed_noreplace.c @@ -0,0 +1,231 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Test that MAP_FIXED_NOREPLACE works. + * + * Copyright 2018, Jann Horn + * Copyright 2018, Michael Ellerman, IBM Corporation. + */ + +#include +#include +#include +#include +#include + +#ifndef MAP_FIXED_NOREPLACE +#define MAP_FIXED_NOREPLACE 0x100000 +#endif + +static void dump_maps(void) +{ + char cmd[32]; + + snprintf(cmd, sizeof(cmd), "cat /proc/%d/maps", getpid()); + system(cmd); +} + +static unsigned long find_base_addr(unsigned long size) +{ + void *addr; + unsigned long flags; + + flags = MAP_PRIVATE | MAP_ANONYMOUS; + addr = mmap(NULL, size, PROT_NONE, flags, -1, 0); + if (addr == MAP_FAILED) { + printf("Error: couldn't map the space we need for the test\n"); + return 0; + } + + if (munmap(addr, size) != 0) { + printf("Error: couldn't map the space we need for the test\n"); + return 0; + } + return (unsigned long)addr; +} + +int main(void) +{ + unsigned long base_addr; + unsigned long flags, addr, size, page_size; + char *p; + + page_size = sysconf(_SC_PAGE_SIZE); + + //let's find a base addr that is free before we start the tests + size = 5 * page_size; + base_addr = find_base_addr(size); + if (!base_addr) { + printf("Error: couldn't map the space we need for the test\n"); + return 1; + } + + flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED_NOREPLACE; + + // Check we can map all the areas we need below + errno = 0; + addr = base_addr; + size = 5 * page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p == MAP_FAILED) { + dump_maps(); + printf("Error: couldn't map the space we need for the test\n"); + return 1; + } + + errno = 0; + if (munmap((void *)addr, 5 * page_size) != 0) { + dump_maps(); + printf("Error: munmap failed!?\n"); + return 1; + } + printf("unmap() successful\n"); + + errno = 0; + addr = base_addr + page_size; + size = 3 * page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p == MAP_FAILED) { + dump_maps(); + printf("Error: first mmap() failed unexpectedly\n"); + return 1; + } + + /* + * Exact same mapping again: + * base | free | new + * +1 | mapped | new + * +2 | mapped | new + * +3 | mapped | new + * +4 | free | new + */ + errno = 0; + addr = base_addr; + size = 5 * page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p != MAP_FAILED) { + dump_maps(); + printf("Error:1: mmap() succeeded when it shouldn't have\n"); + return 1; + } + + /* + * Second mapping contained within first: + * + * base | free | + * +1 | mapped | + * +2 | mapped | new + * +3 | mapped | + * +4 | free | + */ + errno = 0; + addr = base_addr + (2 * page_size); + size = page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p != MAP_FAILED) { + dump_maps(); + printf("Error:2: mmap() succeeded when it shouldn't have\n"); + return 1; + } + + /* + * Overlap end of existing mapping: + * base | free | + * +1 | mapped | + * +2 | mapped | + * +3 | mapped | new + * +4 | free | new + */ + errno = 0; + addr = base_addr + (3 * page_size); + size = 2 * page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p != MAP_FAILED) { + dump_maps(); + printf("Error:3: mmap() succeeded when it shouldn't have\n"); + return 1; + } + + /* + * Overlap start of existing mapping: + * base | free | new + * +1 | mapped | new + * +2 | mapped | + * +3 | mapped | + * +4 | free | + */ + errno = 0; + addr = base_addr; + size = 2 * page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p != MAP_FAILED) { + dump_maps(); + printf("Error:4: mmap() succeeded when it shouldn't have\n"); + return 1; + } + + /* + * Adjacent to start of existing mapping: + * base | free | new + * +1 | mapped | + * +2 | mapped | + * +3 | mapped | + * +4 | free | + */ + errno = 0; + addr = base_addr; + size = page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p == MAP_FAILED) { + dump_maps(); + printf("Error:5: mmap() failed when it shouldn't have\n"); + return 1; + } + + /* + * Adjacent to end of existing mapping: + * base | free | + * +1 | mapped | + * +2 | mapped | + * +3 | mapped | + * +4 | free | new + */ + errno = 0; + addr = base_addr + (4 * page_size); + size = page_size; + p = mmap((void *)addr, size, PROT_NONE, flags, -1, 0); + printf("mmap() @ 0x%lx-0x%lx p=%p result=%m\n", addr, addr + size, p); + + if (p == MAP_FAILED) { + dump_maps(); + printf("Error:6: mmap() failed when it shouldn't have\n"); + return 1; + } + + addr = base_addr; + size = 5 * page_size; + if (munmap((void *)addr, size) != 0) { + dump_maps(); + printf("Error: munmap failed!?\n"); + return 1; + } + printf("unmap() successful\n"); + + printf("OK\n"); + return 0; +} diff --git a/tools/testing/selftests/vm/map_hugetlb.c b/tools/testing/selftests/vm/map_hugetlb.c new file mode 100644 index 000000000..312889edb --- /dev/null +++ b/tools/testing/selftests/vm/map_hugetlb.c @@ -0,0 +1,109 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Example of using hugepage memory in a user application using the mmap + * system call with MAP_HUGETLB flag. Before running this program make + * sure the administrator has allocated enough default sized huge pages + * to cover the 256 MB allocation. + * + * For ia64 architecture, Linux kernel reserves Region number 4 for hugepages. + * That means the addresses starting with 0x800000... will need to be + * specified. Specifying a fixed address is not required on ppc64, i386 + * or x86_64. + */ +#include +#include +#include +#include +#include + +#define LENGTH (256UL*1024*1024) +#define PROTECTION (PROT_READ | PROT_WRITE) + +#ifndef MAP_HUGETLB +#define MAP_HUGETLB 0x40000 /* arch specific */ +#endif + +#ifndef MAP_HUGE_SHIFT +#define MAP_HUGE_SHIFT 26 +#endif + +#ifndef MAP_HUGE_MASK +#define MAP_HUGE_MASK 0x3f +#endif + +/* Only ia64 requires this */ +#ifdef __ia64__ +#define ADDR (void *)(0x8000000000000000UL) +#define FLAGS (MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | MAP_FIXED) +#else +#define ADDR (void *)(0x0UL) +#define FLAGS (MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB) +#endif + +static void check_bytes(char *addr) +{ + printf("First hex is %x\n", *((unsigned int *)addr)); +} + +static void write_bytes(char *addr, size_t length) +{ + unsigned long i; + + for (i = 0; i < length; i++) + *(addr + i) = (char)i; +} + +static int read_bytes(char *addr, size_t length) +{ + unsigned long i; + + check_bytes(addr); + for (i = 0; i < length; i++) + if (*(addr + i) != (char)i) { + printf("Mismatch at %lu\n", i); + return 1; + } + return 0; +} + +int main(int argc, char **argv) +{ + void *addr; + int ret; + size_t length = LENGTH; + int flags = FLAGS; + int shift = 0; + + if (argc > 1) + length = atol(argv[1]) << 20; + if (argc > 2) { + shift = atoi(argv[2]); + if (shift) + flags |= (shift & MAP_HUGE_MASK) << MAP_HUGE_SHIFT; + } + + if (shift) + printf("%u kB hugepages\n", 1 << (shift - 10)); + else + printf("Default size hugepages\n"); + printf("Mapping %lu Mbytes\n", (unsigned long)length >> 20); + + addr = mmap(ADDR, length, PROTECTION, flags, -1, 0); + if (addr == MAP_FAILED) { + perror("mmap"); + exit(1); + } + + printf("Returned address is %p\n", addr); + check_bytes(addr); + write_bytes(addr, length); + ret = read_bytes(addr, length); + + /* munmap() length of MAP_HUGETLB memory must be hugepage aligned */ + if (munmap(addr, length)) { + perror("munmap"); + exit(1); + } + + return ret; +} diff --git a/tools/testing/selftests/vm/map_populate.c b/tools/testing/selftests/vm/map_populate.c new file mode 100644 index 000000000..6b8aeaa0b --- /dev/null +++ b/tools/testing/selftests/vm/map_populate.c @@ -0,0 +1,113 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2018 Dmitry Safonov, Arista Networks + * + * MAP_POPULATE | MAP_PRIVATE should COW VMA pages. + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifndef MMAP_SZ +#define MMAP_SZ 4096 +#endif + +#define BUG_ON(condition, description) \ + do { \ + if (condition) { \ + fprintf(stderr, "[FAIL]\t%s:%d\t%s:%s\n", __func__, \ + __LINE__, (description), strerror(errno)); \ + exit(1); \ + } \ + } while (0) + +static int parent_f(int sock, unsigned long *smap, int child) +{ + int status, ret; + + ret = read(sock, &status, sizeof(int)); + BUG_ON(ret <= 0, "read(sock)"); + + *smap = 0x22222BAD; + ret = msync(smap, MMAP_SZ, MS_SYNC); + BUG_ON(ret, "msync()"); + + ret = write(sock, &status, sizeof(int)); + BUG_ON(ret <= 0, "write(sock)"); + + waitpid(child, &status, 0); + BUG_ON(!WIFEXITED(status), "child in unexpected state"); + + return WEXITSTATUS(status); +} + +static int child_f(int sock, unsigned long *smap, int fd) +{ + int ret, buf = 0; + + smap = mmap(0, MMAP_SZ, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_POPULATE, fd, 0); + BUG_ON(smap == MAP_FAILED, "mmap()"); + + BUG_ON(*smap != 0xdeadbabe, "MAP_PRIVATE | MAP_POPULATE changed file"); + + ret = write(sock, &buf, sizeof(int)); + BUG_ON(ret <= 0, "write(sock)"); + + ret = read(sock, &buf, sizeof(int)); + BUG_ON(ret <= 0, "read(sock)"); + + BUG_ON(*smap == 0x22222BAD, "MAP_POPULATE didn't COW private page"); + BUG_ON(*smap != 0xdeadbabe, "mapping was corrupted"); + + return 0; +} + +int main(int argc, char **argv) +{ + int sock[2], child, ret; + FILE *ftmp; + unsigned long *smap; + + ftmp = tmpfile(); + BUG_ON(ftmp == 0, "tmpfile()"); + + ret = ftruncate(fileno(ftmp), MMAP_SZ); + BUG_ON(ret, "ftruncate()"); + + smap = mmap(0, MMAP_SZ, PROT_READ | PROT_WRITE, + MAP_SHARED, fileno(ftmp), 0); + BUG_ON(smap == MAP_FAILED, "mmap()"); + + *smap = 0xdeadbabe; + /* Probably unnecessary, but let it be. */ + ret = msync(smap, MMAP_SZ, MS_SYNC); + BUG_ON(ret, "msync()"); + + ret = socketpair(PF_LOCAL, SOCK_SEQPACKET, 0, sock); + BUG_ON(ret, "socketpair()"); + + child = fork(); + BUG_ON(child == -1, "fork()"); + + if (child) { + ret = close(sock[0]); + BUG_ON(ret, "close()"); + + return parent_f(sock[1], smap, child); + } + + ret = close(sock[1]); + BUG_ON(ret, "close()"); + + return child_f(sock[0], smap, fileno(ftmp)); +} diff --git a/tools/testing/selftests/vm/memfd_secret.c b/tools/testing/selftests/vm/memfd_secret.c new file mode 100644 index 000000000..9b298f6a0 --- /dev/null +++ b/tools/testing/selftests/vm/memfd_secret.c @@ -0,0 +1,299 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright IBM Corporation, 2021 + * + * Author: Mike Rapoport + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +#define fail(fmt, ...) ksft_test_result_fail(fmt, ##__VA_ARGS__) +#define pass(fmt, ...) ksft_test_result_pass(fmt, ##__VA_ARGS__) +#define skip(fmt, ...) ksft_test_result_skip(fmt, ##__VA_ARGS__) + +#ifdef __NR_memfd_secret + +#define PATTERN 0x55 + +static const int prot = PROT_READ | PROT_WRITE; +static const int mode = MAP_SHARED; + +static unsigned long page_size; +static unsigned long mlock_limit_cur; +static unsigned long mlock_limit_max; + +static int memfd_secret(unsigned int flags) +{ + return syscall(__NR_memfd_secret, flags); +} + +static void test_file_apis(int fd) +{ + char buf[64]; + + if ((read(fd, buf, sizeof(buf)) >= 0) || + (write(fd, buf, sizeof(buf)) >= 0) || + (pread(fd, buf, sizeof(buf), 0) >= 0) || + (pwrite(fd, buf, sizeof(buf), 0) >= 0)) + fail("unexpected file IO\n"); + else + pass("file IO is blocked as expected\n"); +} + +static void test_mlock_limit(int fd) +{ + size_t len; + char *mem; + + len = mlock_limit_cur; + if (len % page_size != 0) + len = (len/page_size) * page_size; + + mem = mmap(NULL, len, prot, mode, fd, 0); + if (mem == MAP_FAILED) { + fail("unable to mmap secret memory\n"); + return; + } + munmap(mem, len); + + len = mlock_limit_max * 2; + mem = mmap(NULL, len, prot, mode, fd, 0); + if (mem != MAP_FAILED) { + fail("unexpected mlock limit violation\n"); + munmap(mem, len); + return; + } + + pass("mlock limit is respected\n"); +} + +static void try_process_vm_read(int fd, int pipefd[2]) +{ + struct iovec liov, riov; + char buf[64]; + char *mem; + + if (read(pipefd[0], &mem, sizeof(mem)) < 0) { + fail("pipe write: %s\n", strerror(errno)); + exit(KSFT_FAIL); + } + + liov.iov_len = riov.iov_len = sizeof(buf); + liov.iov_base = buf; + riov.iov_base = mem; + + if (process_vm_readv(getppid(), &liov, 1, &riov, 1, 0) < 0) { + if (errno == ENOSYS) + exit(KSFT_SKIP); + exit(KSFT_PASS); + } + + exit(KSFT_FAIL); +} + +static void try_ptrace(int fd, int pipefd[2]) +{ + pid_t ppid = getppid(); + int status; + char *mem; + long ret; + + if (read(pipefd[0], &mem, sizeof(mem)) < 0) { + perror("pipe write"); + exit(KSFT_FAIL); + } + + ret = ptrace(PTRACE_ATTACH, ppid, 0, 0); + if (ret) { + perror("ptrace_attach"); + exit(KSFT_FAIL); + } + + ret = waitpid(ppid, &status, WUNTRACED); + if ((ret != ppid) || !(WIFSTOPPED(status))) { + fprintf(stderr, "weird waitppid result %ld stat %x\n", + ret, status); + exit(KSFT_FAIL); + } + + if (ptrace(PTRACE_PEEKDATA, ppid, mem, 0)) + exit(KSFT_PASS); + + exit(KSFT_FAIL); +} + +static void check_child_status(pid_t pid, const char *name) +{ + int status; + + waitpid(pid, &status, 0); + + if (WIFEXITED(status) && WEXITSTATUS(status) == KSFT_SKIP) { + skip("%s is not supported\n", name); + return; + } + + if ((WIFEXITED(status) && WEXITSTATUS(status) == KSFT_PASS) || + WIFSIGNALED(status)) { + pass("%s is blocked as expected\n", name); + return; + } + + fail("%s: unexpected memory access\n", name); +} + +static void test_remote_access(int fd, const char *name, + void (*func)(int fd, int pipefd[2])) +{ + int pipefd[2]; + pid_t pid; + char *mem; + + if (pipe(pipefd)) { + fail("pipe failed: %s\n", strerror(errno)); + return; + } + + pid = fork(); + if (pid < 0) { + fail("fork failed: %s\n", strerror(errno)); + return; + } + + if (pid == 0) { + func(fd, pipefd); + return; + } + + mem = mmap(NULL, page_size, prot, mode, fd, 0); + if (mem == MAP_FAILED) { + fail("Unable to mmap secret memory\n"); + return; + } + + ftruncate(fd, page_size); + memset(mem, PATTERN, page_size); + + if (write(pipefd[1], &mem, sizeof(mem)) < 0) { + fail("pipe write: %s\n", strerror(errno)); + return; + } + + check_child_status(pid, name); +} + +static void test_process_vm_read(int fd) +{ + test_remote_access(fd, "process_vm_read", try_process_vm_read); +} + +static void test_ptrace(int fd) +{ + test_remote_access(fd, "ptrace", try_ptrace); +} + +static int set_cap_limits(rlim_t max) +{ + struct rlimit new; + cap_t cap = cap_init(); + + new.rlim_cur = max; + new.rlim_max = max; + if (setrlimit(RLIMIT_MEMLOCK, &new)) { + perror("setrlimit() returns error"); + return -1; + } + + /* drop capabilities including CAP_IPC_LOCK */ + if (cap_set_proc(cap)) { + perror("cap_set_proc() returns error"); + return -2; + } + + return 0; +} + +static void prepare(void) +{ + struct rlimit rlim; + + page_size = sysconf(_SC_PAGE_SIZE); + if (!page_size) + ksft_exit_fail_msg("Failed to get page size %s\n", + strerror(errno)); + + if (getrlimit(RLIMIT_MEMLOCK, &rlim)) + ksft_exit_fail_msg("Unable to detect mlock limit: %s\n", + strerror(errno)); + + mlock_limit_cur = rlim.rlim_cur; + mlock_limit_max = rlim.rlim_max; + + printf("page_size: %ld, mlock.soft: %ld, mlock.hard: %ld\n", + page_size, mlock_limit_cur, mlock_limit_max); + + if (page_size > mlock_limit_cur) + mlock_limit_cur = page_size; + if (page_size > mlock_limit_max) + mlock_limit_max = page_size; + + if (set_cap_limits(mlock_limit_max)) + ksft_exit_fail_msg("Unable to set mlock limit: %s\n", + strerror(errno)); +} + +#define NUM_TESTS 4 + +int main(int argc, char *argv[]) +{ + int fd; + + prepare(); + + ksft_print_header(); + ksft_set_plan(NUM_TESTS); + + fd = memfd_secret(0); + if (fd < 0) { + if (errno == ENOSYS) + ksft_exit_skip("memfd_secret is not supported\n"); + else + ksft_exit_fail_msg("memfd_secret failed: %s\n", + strerror(errno)); + } + + test_mlock_limit(fd); + test_file_apis(fd); + test_process_vm_read(fd); + test_ptrace(fd); + + close(fd); + + ksft_finished(); +} + +#else /* __NR_memfd_secret */ + +int main(int argc, char *argv[]) +{ + printf("skip: skipping memfd_secret test (missing __NR_memfd_secret)\n"); + return KSFT_SKIP; +} + +#endif /* __NR_memfd_secret */ diff --git a/tools/testing/selftests/vm/migration.c b/tools/testing/selftests/vm/migration.c new file mode 100644 index 000000000..1cec8425e --- /dev/null +++ b/tools/testing/selftests/vm/migration.c @@ -0,0 +1,193 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * The main purpose of the tests here is to exercise the migration entry code + * paths in the kernel. + */ + +#include "../kselftest_harness.h" +#include +#include +#include +#include +#include +#include +#include +#include + +#define TWOMEG (2<<20) +#define RUNTIME (60) + +#define ALIGN(x, a) (((x) + (a - 1)) & (~((a) - 1))) + +FIXTURE(migration) +{ + pthread_t *threads; + pid_t *pids; + int nthreads; + int n1; + int n2; +}; + +FIXTURE_SETUP(migration) +{ + int n; + + ASSERT_EQ(numa_available(), 0); + self->nthreads = numa_num_task_cpus() - 1; + self->n1 = -1; + self->n2 = -1; + + for (n = 0; n < numa_max_possible_node(); n++) + if (numa_bitmask_isbitset(numa_all_nodes_ptr, n)) { + if (self->n1 == -1) { + self->n1 = n; + } else { + self->n2 = n; + break; + } + } + + self->threads = malloc(self->nthreads * sizeof(*self->threads)); + ASSERT_NE(self->threads, NULL); + self->pids = malloc(self->nthreads * sizeof(*self->pids)); + ASSERT_NE(self->pids, NULL); +}; + +FIXTURE_TEARDOWN(migration) +{ + free(self->threads); + free(self->pids); +} + +int migrate(uint64_t *ptr, int n1, int n2) +{ + int ret, tmp; + int status = 0; + struct timespec ts1, ts2; + + if (clock_gettime(CLOCK_MONOTONIC, &ts1)) + return -1; + + while (1) { + if (clock_gettime(CLOCK_MONOTONIC, &ts2)) + return -1; + + if (ts2.tv_sec - ts1.tv_sec >= RUNTIME) + return 0; + + ret = move_pages(0, 1, (void **) &ptr, &n2, &status, + MPOL_MF_MOVE_ALL); + if (ret) { + if (ret > 0) + printf("Didn't migrate %d pages\n", ret); + else + perror("Couldn't migrate pages"); + return -2; + } + + tmp = n2; + n2 = n1; + n1 = tmp; + } + + return 0; +} + +void *access_mem(void *ptr) +{ + uint64_t y = 0; + volatile uint64_t *x = ptr; + + while (1) { + pthread_testcancel(); + y += *x; + } + + return NULL; +} + +/* + * Basic migration entry testing. One thread will move pages back and forth + * between nodes whilst other threads try and access them triggering the + * migration entry wait paths in the kernel. + */ +TEST_F_TIMEOUT(migration, private_anon, 2*RUNTIME) +{ + uint64_t *ptr; + int i; + + if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0) + SKIP(return, "Not enough threads or NUMA nodes available"); + + ptr = mmap(NULL, TWOMEG, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + ASSERT_NE(ptr, MAP_FAILED); + + memset(ptr, 0xde, TWOMEG); + for (i = 0; i < self->nthreads - 1; i++) + if (pthread_create(&self->threads[i], NULL, access_mem, ptr)) + perror("Couldn't create thread"); + + ASSERT_EQ(migrate(ptr, self->n1, self->n2), 0); + for (i = 0; i < self->nthreads - 1; i++) + ASSERT_EQ(pthread_cancel(self->threads[i]), 0); +} + +/* + * Same as the previous test but with shared memory. + */ +TEST_F_TIMEOUT(migration, shared_anon, 2*RUNTIME) +{ + pid_t pid; + uint64_t *ptr; + int i; + + if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0) + SKIP(return, "Not enough threads or NUMA nodes available"); + + ptr = mmap(NULL, TWOMEG, PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_ANONYMOUS, -1, 0); + ASSERT_NE(ptr, MAP_FAILED); + + memset(ptr, 0xde, TWOMEG); + for (i = 0; i < self->nthreads - 1; i++) { + pid = fork(); + if (!pid) + access_mem(ptr); + else + self->pids[i] = pid; + } + + ASSERT_EQ(migrate(ptr, self->n1, self->n2), 0); + for (i = 0; i < self->nthreads - 1; i++) + ASSERT_EQ(kill(self->pids[i], SIGTERM), 0); +} + +/* + * Tests the pmd migration entry paths. + */ +TEST_F_TIMEOUT(migration, private_anon_thp, 2*RUNTIME) +{ + uint64_t *ptr; + int i; + + if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0) + SKIP(return, "Not enough threads or NUMA nodes available"); + + ptr = mmap(NULL, 2*TWOMEG, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + ASSERT_NE(ptr, MAP_FAILED); + + ptr = (uint64_t *) ALIGN((uintptr_t) ptr, TWOMEG); + ASSERT_EQ(madvise(ptr, TWOMEG, MADV_HUGEPAGE), 0); + memset(ptr, 0xde, TWOMEG); + for (i = 0; i < self->nthreads - 1; i++) + if (pthread_create(&self->threads[i], NULL, access_mem, ptr)) + perror("Couldn't create thread"); + + ASSERT_EQ(migrate(ptr, self->n1, self->n2), 0); + for (i = 0; i < self->nthreads - 1; i++) + ASSERT_EQ(pthread_cancel(self->threads[i]), 0); +} + +TEST_HARNESS_MAIN diff --git a/tools/testing/selftests/vm/mlock-random-test.c b/tools/testing/selftests/vm/mlock-random-test.c new file mode 100644 index 000000000..782ea94de --- /dev/null +++ b/tools/testing/selftests/vm/mlock-random-test.c @@ -0,0 +1,294 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * It tests the mlock/mlock2() when they are invoked + * on randomly memory region. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "mlock2.h" + +#define CHUNK_UNIT (128 * 1024) +#define MLOCK_RLIMIT_SIZE (CHUNK_UNIT * 2) +#define MLOCK_WITHIN_LIMIT_SIZE CHUNK_UNIT +#define MLOCK_OUTOF_LIMIT_SIZE (CHUNK_UNIT * 3) + +#define TEST_LOOP 100 +#define PAGE_ALIGN(size, ps) (((size) + ((ps) - 1)) & ~((ps) - 1)) + +int set_cap_limits(rlim_t max) +{ + struct rlimit new; + cap_t cap = cap_init(); + + new.rlim_cur = max; + new.rlim_max = max; + if (setrlimit(RLIMIT_MEMLOCK, &new)) { + perror("setrlimit() returns error\n"); + return -1; + } + + /* drop capabilities including CAP_IPC_LOCK */ + if (cap_set_proc(cap)) { + perror("cap_set_proc() returns error\n"); + return -2; + } + + return 0; +} + +int get_proc_locked_vm_size(void) +{ + FILE *f; + int ret = -1; + char line[1024] = {0}; + unsigned long lock_size = 0; + + f = fopen("/proc/self/status", "r"); + if (!f) { + perror("fopen"); + return -1; + } + + while (fgets(line, 1024, f)) { + if (strstr(line, "VmLck")) { + ret = sscanf(line, "VmLck:\t%8lu kB", &lock_size); + if (ret <= 0) { + printf("sscanf() on VmLck error: %s: %d\n", + line, ret); + fclose(f); + return -1; + } + fclose(f); + return (int)(lock_size << 10); + } + } + + perror("cannot parse VmLck in /proc/self/status\n"); + fclose(f); + return -1; +} + +/* + * Get the MMUPageSize of the memory region including input + * address from proc file. + * + * return value: on error case, 0 will be returned. + * Otherwise the page size(in bytes) is returned. + */ +int get_proc_page_size(unsigned long addr) +{ + FILE *smaps; + char *line; + unsigned long mmupage_size = 0; + size_t size; + + smaps = seek_to_smaps_entry(addr); + if (!smaps) { + printf("Unable to parse /proc/self/smaps\n"); + return 0; + } + + while (getline(&line, &size, smaps) > 0) { + if (!strstr(line, "MMUPageSize")) { + free(line); + line = NULL; + size = 0; + continue; + } + + /* found the MMUPageSize of this section */ + if (sscanf(line, "MMUPageSize: %8lu kB", + &mmupage_size) < 1) { + printf("Unable to parse smaps entry for Size:%s\n", + line); + break; + } + + } + free(line); + if (smaps) + fclose(smaps); + return mmupage_size << 10; +} + +/* + * Test mlock/mlock2() on provided memory chunk. + * It expects the mlock/mlock2() to be successful (within rlimit) + * + * With allocated memory chunk [p, p + alloc_size), this + * test will choose start/len randomly to perform mlock/mlock2 + * [start, start + len] memory range. The range is within range + * of the allocated chunk. + * + * The memory region size alloc_size is within the rlimit. + * So we always expect a success of mlock/mlock2. + * + * VmLck is assumed to be 0 before this test. + * + * return value: 0 - success + * else: failure + */ +int test_mlock_within_limit(char *p, int alloc_size) +{ + int i; + int ret = 0; + int locked_vm_size = 0; + struct rlimit cur; + int page_size = 0; + + getrlimit(RLIMIT_MEMLOCK, &cur); + if (cur.rlim_cur < alloc_size) { + printf("alloc_size[%d] < %u rlimit,lead to mlock failure\n", + alloc_size, (unsigned int)cur.rlim_cur); + return -1; + } + + srand(time(NULL)); + for (i = 0; i < TEST_LOOP; i++) { + /* + * - choose mlock/mlock2 randomly + * - choose lock_size randomly but lock_size < alloc_size + * - choose start_offset randomly but p+start_offset+lock_size + * < p+alloc_size + */ + int is_mlock = !!(rand() % 2); + int lock_size = rand() % alloc_size; + int start_offset = rand() % (alloc_size - lock_size); + + if (is_mlock) + ret = mlock(p + start_offset, lock_size); + else + ret = mlock2_(p + start_offset, lock_size, + MLOCK_ONFAULT); + + if (ret) { + printf("%s() failure at |%p(%d)| mlock:|%p(%d)|\n", + is_mlock ? "mlock" : "mlock2", + p, alloc_size, + p + start_offset, lock_size); + return ret; + } + } + + /* + * Check VmLck left by the tests. + */ + locked_vm_size = get_proc_locked_vm_size(); + page_size = get_proc_page_size((unsigned long)p); + if (page_size == 0) { + printf("cannot get proc MMUPageSize\n"); + return -1; + } + + if (locked_vm_size > PAGE_ALIGN(alloc_size, page_size) + page_size) { + printf("test_mlock_within_limit() left VmLck:%d on %d chunk\n", + locked_vm_size, alloc_size); + return -1; + } + + return 0; +} + + +/* + * We expect the mlock/mlock2() to be fail (outof limitation) + * + * With allocated memory chunk [p, p + alloc_size), this + * test will randomly choose start/len and perform mlock/mlock2 + * on [start, start+len] range. + * + * The memory region size alloc_size is above the rlimit. + * And the len to be locked is higher than rlimit. + * So we always expect a failure of mlock/mlock2. + * No locked page number should be increased as a side effect. + * + * return value: 0 - success + * else: failure + */ +int test_mlock_outof_limit(char *p, int alloc_size) +{ + int i; + int ret = 0; + int locked_vm_size = 0, old_locked_vm_size = 0; + struct rlimit cur; + + getrlimit(RLIMIT_MEMLOCK, &cur); + if (cur.rlim_cur >= alloc_size) { + printf("alloc_size[%d] >%u rlimit, violates test condition\n", + alloc_size, (unsigned int)cur.rlim_cur); + return -1; + } + + old_locked_vm_size = get_proc_locked_vm_size(); + srand(time(NULL)); + for (i = 0; i < TEST_LOOP; i++) { + int is_mlock = !!(rand() % 2); + int lock_size = (rand() % (alloc_size - cur.rlim_cur)) + + cur.rlim_cur; + int start_offset = rand() % (alloc_size - lock_size); + + if (is_mlock) + ret = mlock(p + start_offset, lock_size); + else + ret = mlock2_(p + start_offset, lock_size, + MLOCK_ONFAULT); + if (ret == 0) { + printf("%s() succeeds? on %p(%d) mlock%p(%d)\n", + is_mlock ? "mlock" : "mlock2", + p, alloc_size, + p + start_offset, lock_size); + return -1; + } + } + + locked_vm_size = get_proc_locked_vm_size(); + if (locked_vm_size != old_locked_vm_size) { + printf("tests leads to new mlocked page: old[%d], new[%d]\n", + old_locked_vm_size, + locked_vm_size); + return -1; + } + + return 0; +} + +int main(int argc, char **argv) +{ + char *p = NULL; + int ret = 0; + + if (set_cap_limits(MLOCK_RLIMIT_SIZE)) + return -1; + + p = malloc(MLOCK_WITHIN_LIMIT_SIZE); + if (p == NULL) { + perror("malloc() failure\n"); + return -1; + } + ret = test_mlock_within_limit(p, MLOCK_WITHIN_LIMIT_SIZE); + if (ret) + return ret; + munlock(p, MLOCK_WITHIN_LIMIT_SIZE); + free(p); + + + p = malloc(MLOCK_OUTOF_LIMIT_SIZE); + if (p == NULL) { + perror("malloc() failure\n"); + return -1; + } + ret = test_mlock_outof_limit(p, MLOCK_OUTOF_LIMIT_SIZE); + if (ret) + return ret; + munlock(p, MLOCK_OUTOF_LIMIT_SIZE); + free(p); + + return 0; +} diff --git a/tools/testing/selftests/vm/mlock2-tests.c b/tools/testing/selftests/vm/mlock2-tests.c new file mode 100644 index 000000000..11b2301f3 --- /dev/null +++ b/tools/testing/selftests/vm/mlock2-tests.c @@ -0,0 +1,520 @@ +// SPDX-License-Identifier: GPL-2.0 +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include "mlock2.h" + +#include "../kselftest.h" + +struct vm_boundaries { + unsigned long start; + unsigned long end; +}; + +static int get_vm_area(unsigned long addr, struct vm_boundaries *area) +{ + FILE *file; + int ret = 1; + char line[1024] = {0}; + char *end_addr; + char *stop; + unsigned long start; + unsigned long end; + + if (!area) + return ret; + + file = fopen("/proc/self/maps", "r"); + if (!file) { + perror("fopen"); + return ret; + } + + memset(area, 0, sizeof(struct vm_boundaries)); + + while(fgets(line, 1024, file)) { + end_addr = strchr(line, '-'); + if (!end_addr) { + printf("cannot parse /proc/self/maps\n"); + goto out; + } + *end_addr = '\0'; + end_addr++; + stop = strchr(end_addr, ' '); + if (!stop) { + printf("cannot parse /proc/self/maps\n"); + goto out; + } + stop = '\0'; + + sscanf(line, "%lx", &start); + sscanf(end_addr, "%lx", &end); + + if (start <= addr && end > addr) { + area->start = start; + area->end = end; + ret = 0; + goto out; + } + } +out: + fclose(file); + return ret; +} + +#define VMFLAGS "VmFlags:" + +static bool is_vmflag_set(unsigned long addr, const char *vmflag) +{ + char *line = NULL; + char *flags; + size_t size = 0; + bool ret = false; + FILE *smaps; + + smaps = seek_to_smaps_entry(addr); + if (!smaps) { + printf("Unable to parse /proc/self/smaps\n"); + goto out; + } + + while (getline(&line, &size, smaps) > 0) { + if (!strstr(line, VMFLAGS)) { + free(line); + line = NULL; + size = 0; + continue; + } + + flags = line + strlen(VMFLAGS); + ret = (strstr(flags, vmflag) != NULL); + goto out; + } + +out: + free(line); + fclose(smaps); + return ret; +} + +#define SIZE "Size:" +#define RSS "Rss:" +#define LOCKED "lo" + +static unsigned long get_value_for_name(unsigned long addr, const char *name) +{ + char *line = NULL; + size_t size = 0; + char *value_ptr; + FILE *smaps = NULL; + unsigned long value = -1UL; + + smaps = seek_to_smaps_entry(addr); + if (!smaps) { + printf("Unable to parse /proc/self/smaps\n"); + goto out; + } + + while (getline(&line, &size, smaps) > 0) { + if (!strstr(line, name)) { + free(line); + line = NULL; + size = 0; + continue; + } + + value_ptr = line + strlen(name); + if (sscanf(value_ptr, "%lu kB", &value) < 1) { + printf("Unable to parse smaps entry for Size\n"); + goto out; + } + break; + } + +out: + if (smaps) + fclose(smaps); + free(line); + return value; +} + +static bool is_vma_lock_on_fault(unsigned long addr) +{ + bool locked; + unsigned long vma_size, vma_rss; + + locked = is_vmflag_set(addr, LOCKED); + if (!locked) + return false; + + vma_size = get_value_for_name(addr, SIZE); + vma_rss = get_value_for_name(addr, RSS); + + /* only one page is faulted in */ + return (vma_rss < vma_size); +} + +#define PRESENT_BIT 0x8000000000000000ULL +#define PFN_MASK 0x007FFFFFFFFFFFFFULL +#define UNEVICTABLE_BIT (1UL << 18) + +static int lock_check(unsigned long addr) +{ + bool locked; + unsigned long vma_size, vma_rss; + + locked = is_vmflag_set(addr, LOCKED); + if (!locked) + return false; + + vma_size = get_value_for_name(addr, SIZE); + vma_rss = get_value_for_name(addr, RSS); + + return (vma_rss == vma_size); +} + +static int unlock_lock_check(char *map) +{ + if (is_vmflag_set((unsigned long)map, LOCKED)) { + printf("VMA flag %s is present on page 1 after unlock\n", LOCKED); + return 1; + } + + return 0; +} + +static int test_mlock_lock() +{ + char *map; + int ret = 1; + unsigned long page_size = getpagesize(); + + map = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + if (map == MAP_FAILED) { + perror("test_mlock_locked mmap"); + goto out; + } + + if (mlock2_(map, 2 * page_size, 0)) { + if (errno == ENOSYS) { + printf("Cannot call new mlock family, skipping test\n"); + _exit(KSFT_SKIP); + } + perror("mlock2(0)"); + goto unmap; + } + + if (!lock_check((unsigned long)map)) + goto unmap; + + /* Now unlock and recheck attributes */ + if (munlock(map, 2 * page_size)) { + perror("munlock()"); + goto unmap; + } + + ret = unlock_lock_check(map); + +unmap: + munmap(map, 2 * page_size); +out: + return ret; +} + +static int onfault_check(char *map) +{ + *map = 'a'; + if (!is_vma_lock_on_fault((unsigned long)map)) { + printf("VMA is not marked for lock on fault\n"); + return 1; + } + + return 0; +} + +static int unlock_onfault_check(char *map) +{ + unsigned long page_size = getpagesize(); + + if (is_vma_lock_on_fault((unsigned long)map) || + is_vma_lock_on_fault((unsigned long)map + page_size)) { + printf("VMA is still lock on fault after unlock\n"); + return 1; + } + + return 0; +} + +static int test_mlock_onfault() +{ + char *map; + int ret = 1; + unsigned long page_size = getpagesize(); + + map = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + if (map == MAP_FAILED) { + perror("test_mlock_locked mmap"); + goto out; + } + + if (mlock2_(map, 2 * page_size, MLOCK_ONFAULT)) { + if (errno == ENOSYS) { + printf("Cannot call new mlock family, skipping test\n"); + _exit(KSFT_SKIP); + } + perror("mlock2(MLOCK_ONFAULT)"); + goto unmap; + } + + if (onfault_check(map)) + goto unmap; + + /* Now unlock and recheck attributes */ + if (munlock(map, 2 * page_size)) { + if (errno == ENOSYS) { + printf("Cannot call new mlock family, skipping test\n"); + _exit(KSFT_SKIP); + } + perror("munlock()"); + goto unmap; + } + + ret = unlock_onfault_check(map); +unmap: + munmap(map, 2 * page_size); +out: + return ret; +} + +static int test_lock_onfault_of_present() +{ + char *map; + int ret = 1; + unsigned long page_size = getpagesize(); + + map = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + if (map == MAP_FAILED) { + perror("test_mlock_locked mmap"); + goto out; + } + + *map = 'a'; + + if (mlock2_(map, 2 * page_size, MLOCK_ONFAULT)) { + if (errno == ENOSYS) { + printf("Cannot call new mlock family, skipping test\n"); + _exit(KSFT_SKIP); + } + perror("mlock2(MLOCK_ONFAULT)"); + goto unmap; + } + + if (!is_vma_lock_on_fault((unsigned long)map) || + !is_vma_lock_on_fault((unsigned long)map + page_size)) { + printf("VMA with present pages is not marked lock on fault\n"); + goto unmap; + } + ret = 0; +unmap: + munmap(map, 2 * page_size); +out: + return ret; +} + +static int test_munlockall() +{ + char *map; + int ret = 1; + unsigned long page_size = getpagesize(); + + map = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + + if (map == MAP_FAILED) { + perror("test_munlockall mmap"); + goto out; + } + + if (mlockall(MCL_CURRENT)) { + perror("mlockall(MCL_CURRENT)"); + goto out; + } + + if (!lock_check((unsigned long)map)) + goto unmap; + + if (munlockall()) { + perror("munlockall()"); + goto unmap; + } + + if (unlock_lock_check(map)) + goto unmap; + + munmap(map, 2 * page_size); + + map = mmap(NULL, 2 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + + if (map == MAP_FAILED) { + perror("test_munlockall second mmap"); + goto out; + } + + if (mlockall(MCL_CURRENT | MCL_ONFAULT)) { + perror("mlockall(MCL_CURRENT | MCL_ONFAULT)"); + goto unmap; + } + + if (onfault_check(map)) + goto unmap; + + if (munlockall()) { + perror("munlockall()"); + goto unmap; + } + + if (unlock_onfault_check(map)) + goto unmap; + + if (mlockall(MCL_CURRENT | MCL_FUTURE)) { + perror("mlockall(MCL_CURRENT | MCL_FUTURE)"); + goto out; + } + + if (!lock_check((unsigned long)map)) + goto unmap; + + if (munlockall()) { + perror("munlockall()"); + goto unmap; + } + + ret = unlock_lock_check(map); + +unmap: + munmap(map, 2 * page_size); +out: + munlockall(); + return ret; +} + +static int test_vma_management(bool call_mlock) +{ + int ret = 1; + void *map; + unsigned long page_size = getpagesize(); + struct vm_boundaries page1; + struct vm_boundaries page2; + struct vm_boundaries page3; + + map = mmap(NULL, 3 * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + if (map == MAP_FAILED) { + perror("mmap()"); + return ret; + } + + if (call_mlock && mlock2_(map, 3 * page_size, MLOCK_ONFAULT)) { + if (errno == ENOSYS) { + printf("Cannot call new mlock family, skipping test\n"); + _exit(KSFT_SKIP); + } + perror("mlock(ONFAULT)\n"); + goto out; + } + + if (get_vm_area((unsigned long)map, &page1) || + get_vm_area((unsigned long)map + page_size, &page2) || + get_vm_area((unsigned long)map + page_size * 2, &page3)) { + printf("couldn't find mapping in /proc/self/maps\n"); + goto out; + } + + /* + * Before we unlock a portion, we need to that all three pages are in + * the same VMA. If they are not we abort this test (Note that this is + * not a failure) + */ + if (page1.start != page2.start || page2.start != page3.start) { + printf("VMAs are not merged to start, aborting test\n"); + ret = 0; + goto out; + } + + if (munlock(map + page_size, page_size)) { + perror("munlock()"); + goto out; + } + + if (get_vm_area((unsigned long)map, &page1) || + get_vm_area((unsigned long)map + page_size, &page2) || + get_vm_area((unsigned long)map + page_size * 2, &page3)) { + printf("couldn't find mapping in /proc/self/maps\n"); + goto out; + } + + /* All three VMAs should be different */ + if (page1.start == page2.start || page2.start == page3.start) { + printf("failed to split VMA for munlock\n"); + goto out; + } + + /* Now unlock the first and third page and check the VMAs again */ + if (munlock(map, page_size * 3)) { + perror("munlock()"); + goto out; + } + + if (get_vm_area((unsigned long)map, &page1) || + get_vm_area((unsigned long)map + page_size, &page2) || + get_vm_area((unsigned long)map + page_size * 2, &page3)) { + printf("couldn't find mapping in /proc/self/maps\n"); + goto out; + } + + /* Now all three VMAs should be the same */ + if (page1.start != page2.start || page2.start != page3.start) { + printf("failed to merge VMAs after munlock\n"); + goto out; + } + + ret = 0; +out: + munmap(map, 3 * page_size); + return ret; +} + +static int test_mlockall(int (test_function)(bool call_mlock)) +{ + int ret = 1; + + if (mlockall(MCL_CURRENT | MCL_ONFAULT | MCL_FUTURE)) { + perror("mlockall"); + return ret; + } + + ret = test_function(false); + munlockall(); + return ret; +} + +int main(int argc, char **argv) +{ + int ret = 0; + ret += test_mlock_lock(); + ret += test_mlock_onfault(); + ret += test_munlockall(); + ret += test_lock_onfault_of_present(); + ret += test_vma_management(true); + ret += test_mlockall(test_vma_management); + return ret; +} diff --git a/tools/testing/selftests/vm/mlock2.h b/tools/testing/selftests/vm/mlock2.h new file mode 100644 index 000000000..2a6e76c22 --- /dev/null +++ b/tools/testing/selftests/vm/mlock2.h @@ -0,0 +1,63 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include +#include +#include +#include + +#ifndef MLOCK_ONFAULT +#define MLOCK_ONFAULT 1 +#endif + +#ifndef MCL_ONFAULT +#define MCL_ONFAULT (MCL_FUTURE << 1) +#endif + +static int mlock2_(void *start, size_t len, int flags) +{ +#ifdef __NR_mlock2 + return syscall(__NR_mlock2, start, len, flags); +#else + errno = ENOSYS; + return -1; +#endif +} + +static FILE *seek_to_smaps_entry(unsigned long addr) +{ + FILE *file; + char *line = NULL; + size_t size = 0; + unsigned long start, end; + char perms[5]; + unsigned long offset; + char dev[32]; + unsigned long inode; + char path[BUFSIZ]; + + file = fopen("/proc/self/smaps", "r"); + if (!file) { + perror("fopen smaps"); + _exit(1); + } + + while (getline(&line, &size, file) > 0) { + if (sscanf(line, "%lx-%lx %s %lx %s %lu %s\n", + &start, &end, perms, &offset, dev, &inode, path) < 6) + goto next; + + if (start <= addr && addr < end) + goto out; + +next: + free(line); + line = NULL; + size = 0; + } + + fclose(file); + file = NULL; + +out: + free(line); + return file; +} diff --git a/tools/testing/selftests/vm/mrelease_test.c b/tools/testing/selftests/vm/mrelease_test.c new file mode 100644 index 000000000..6c62966ab --- /dev/null +++ b/tools/testing/selftests/vm/mrelease_test.c @@ -0,0 +1,206 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2022 Google LLC + */ +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include + +#include "util.h" + +#include "../kselftest.h" + +#ifndef __NR_pidfd_open +#define __NR_pidfd_open -1 +#endif + +#ifndef __NR_process_mrelease +#define __NR_process_mrelease -1 +#endif + +#define MB(x) (x << 20) +#define MAX_SIZE_MB 1024 + +static int alloc_noexit(unsigned long nr_pages, int pipefd) +{ + int ppid = getppid(); + int timeout = 10; /* 10sec timeout to get killed */ + unsigned long i; + char *buf; + + buf = (char *)mmap(NULL, nr_pages * PAGE_SIZE, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANON, 0, 0); + if (buf == MAP_FAILED) { + perror("mmap failed, halting the test"); + return KSFT_FAIL; + } + + for (i = 0; i < nr_pages; i++) + *((unsigned long *)(buf + (i * PAGE_SIZE))) = i; + + /* Signal the parent that the child is ready */ + if (write(pipefd, "", 1) < 0) { + perror("write"); + return KSFT_FAIL; + } + + /* Wait to be killed (when reparenting happens) */ + while (getppid() == ppid && timeout > 0) { + sleep(1); + timeout--; + } + + munmap(buf, nr_pages * PAGE_SIZE); + + return (timeout > 0) ? KSFT_PASS : KSFT_FAIL; +} + +/* The process_mrelease calls in this test are expected to fail */ +static void run_negative_tests(int pidfd) +{ + int res; + /* Test invalid flags. Expect to fail with EINVAL error code. */ + if (!syscall(__NR_process_mrelease, pidfd, (unsigned int)-1) || + errno != EINVAL) { + res = (errno == ENOSYS ? KSFT_SKIP : KSFT_FAIL); + perror("process_mrelease with wrong flags"); + exit(res); + } + /* + * Test reaping while process is alive with no pending SIGKILL. + * Expect to fail with EINVAL error code. + */ + if (!syscall(__NR_process_mrelease, pidfd, 0) || errno != EINVAL) { + res = (errno == ENOSYS ? KSFT_SKIP : KSFT_FAIL); + perror("process_mrelease on a live process"); + exit(res); + } +} + +static int child_main(int pipefd[], size_t size) +{ + int res; + + /* Allocate and fault-in memory and wait to be killed */ + close(pipefd[0]); + res = alloc_noexit(MB(size) / PAGE_SIZE, pipefd[1]); + close(pipefd[1]); + return res; +} + +int main(void) +{ + int pipefd[2], pidfd; + bool success, retry; + size_t size; + pid_t pid; + char byte; + int res; + + /* Test a wrong pidfd */ + if (!syscall(__NR_process_mrelease, -1, 0) || errno != EBADF) { + res = (errno == ENOSYS ? KSFT_SKIP : KSFT_FAIL); + perror("process_mrelease with wrong pidfd"); + exit(res); + } + + /* Start the test with 1MB child memory allocation */ + size = 1; +retry: + /* + * Pipe for the child to signal when it's done allocating + * memory + */ + if (pipe(pipefd)) { + perror("pipe"); + exit(KSFT_FAIL); + } + pid = fork(); + if (pid < 0) { + perror("fork"); + close(pipefd[0]); + close(pipefd[1]); + exit(KSFT_FAIL); + } + + if (pid == 0) { + /* Child main routine */ + res = child_main(pipefd, size); + exit(res); + } + + /* + * Parent main routine: + * Wait for the child to finish allocations, then kill and reap + */ + close(pipefd[1]); + /* Block until the child is ready */ + res = read(pipefd[0], &byte, 1); + close(pipefd[0]); + if (res < 0) { + perror("read"); + if (!kill(pid, SIGKILL)) + waitpid(pid, NULL, 0); + exit(KSFT_FAIL); + } + + pidfd = syscall(__NR_pidfd_open, pid, 0); + if (pidfd < 0) { + perror("pidfd_open"); + if (!kill(pid, SIGKILL)) + waitpid(pid, NULL, 0); + exit(KSFT_FAIL); + } + + /* Run negative tests which require a live child */ + run_negative_tests(pidfd); + + if (kill(pid, SIGKILL)) { + res = (errno == ENOSYS ? KSFT_SKIP : KSFT_FAIL); + perror("kill"); + exit(res); + } + + success = (syscall(__NR_process_mrelease, pidfd, 0) == 0); + if (!success) { + /* + * If we failed to reap because the child exited too soon, + * before we could call process_mrelease. Double child's memory + * which causes it to spend more time on cleanup and increases + * our chances of reaping its memory before it exits. + * Retry until we succeed or reach MAX_SIZE_MB. + */ + if (errno == ESRCH) { + retry = (size <= MAX_SIZE_MB); + } else { + res = (errno == ENOSYS ? KSFT_SKIP : KSFT_FAIL); + perror("process_mrelease"); + waitpid(pid, NULL, 0); + exit(res); + } + } + + /* Cleanup to prevent zombies */ + if (waitpid(pid, NULL, 0) < 0) { + perror("waitpid"); + exit(KSFT_FAIL); + } + close(pidfd); + + if (!success) { + if (retry) { + size *= 2; + goto retry; + } + printf("All process_mrelease attempts failed!\n"); + exit(KSFT_FAIL); + } + + printf("Success reaping a child with %zuMB of memory allocations\n", + size); + return KSFT_PASS; +} diff --git a/tools/testing/selftests/vm/mremap_dontunmap.c b/tools/testing/selftests/vm/mremap_dontunmap.c new file mode 100644 index 000000000..f01dc4a85 --- /dev/null +++ b/tools/testing/selftests/vm/mremap_dontunmap.c @@ -0,0 +1,364 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Tests for mremap w/ MREMAP_DONTUNMAP. + * + * Copyright 2020, Brian Geffon + */ +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +#ifndef MREMAP_DONTUNMAP +#define MREMAP_DONTUNMAP 4 +#endif + +unsigned long page_size; +char *page_buffer; + +static void dump_maps(void) +{ + char cmd[32]; + + snprintf(cmd, sizeof(cmd), "cat /proc/%d/maps", getpid()); + system(cmd); +} + +#define BUG_ON(condition, description) \ + do { \ + if (condition) { \ + fprintf(stderr, "[FAIL]\t%s():%d\t%s:%s\n", __func__, \ + __LINE__, (description), strerror(errno)); \ + dump_maps(); \ + exit(1); \ + } \ + } while (0) + +// Try a simple operation for to "test" for kernel support this prevents +// reporting tests as failed when it's run on an older kernel. +static int kernel_support_for_mremap_dontunmap() +{ + int ret = 0; + unsigned long num_pages = 1; + void *source_mapping = mmap(NULL, num_pages * page_size, PROT_NONE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + + // This simple remap should only fail if MREMAP_DONTUNMAP isn't + // supported. + void *dest_mapping = + mremap(source_mapping, num_pages * page_size, num_pages * page_size, + MREMAP_DONTUNMAP | MREMAP_MAYMOVE, 0); + if (dest_mapping == MAP_FAILED) { + ret = errno; + } else { + BUG_ON(munmap(dest_mapping, num_pages * page_size) == -1, + "unable to unmap destination mapping"); + } + + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); + return ret; +} + +// This helper will just validate that an entire mapping contains the expected +// byte. +static int check_region_contains_byte(void *addr, unsigned long size, char byte) +{ + BUG_ON(size & (page_size - 1), + "check_region_contains_byte expects page multiples"); + BUG_ON((unsigned long)addr & (page_size - 1), + "check_region_contains_byte expects page alignment"); + + memset(page_buffer, byte, page_size); + + unsigned long num_pages = size / page_size; + unsigned long i; + + // Compare each page checking that it contains our expected byte. + for (i = 0; i < num_pages; ++i) { + int ret = + memcmp(addr + (i * page_size), page_buffer, page_size); + if (ret) { + return ret; + } + } + + return 0; +} + +// this test validates that MREMAP_DONTUNMAP moves the pagetables while leaving +// the source mapping mapped. +static void mremap_dontunmap_simple() +{ + unsigned long num_pages = 5; + + void *source_mapping = + mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + + memset(source_mapping, 'a', num_pages * page_size); + + // Try to just move the whole mapping anywhere (not fixed). + void *dest_mapping = + mremap(source_mapping, num_pages * page_size, num_pages * page_size, + MREMAP_DONTUNMAP | MREMAP_MAYMOVE, NULL); + BUG_ON(dest_mapping == MAP_FAILED, "mremap"); + + // Validate that the pages have been moved, we know they were moved if + // the dest_mapping contains a's. + BUG_ON(check_region_contains_byte + (dest_mapping, num_pages * page_size, 'a') != 0, + "pages did not migrate"); + BUG_ON(check_region_contains_byte + (source_mapping, num_pages * page_size, 0) != 0, + "source should have no ptes"); + + BUG_ON(munmap(dest_mapping, num_pages * page_size) == -1, + "unable to unmap destination mapping"); + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); +} + +// This test validates that MREMAP_DONTUNMAP on a shared mapping works as expected. +static void mremap_dontunmap_simple_shmem() +{ + unsigned long num_pages = 5; + + int mem_fd = memfd_create("memfd", MFD_CLOEXEC); + BUG_ON(mem_fd < 0, "memfd_create"); + + BUG_ON(ftruncate(mem_fd, num_pages * page_size) < 0, + "ftruncate"); + + void *source_mapping = + mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE, + MAP_FILE | MAP_SHARED, mem_fd, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + + BUG_ON(close(mem_fd) < 0, "close"); + + memset(source_mapping, 'a', num_pages * page_size); + + // Try to just move the whole mapping anywhere (not fixed). + void *dest_mapping = + mremap(source_mapping, num_pages * page_size, num_pages * page_size, + MREMAP_DONTUNMAP | MREMAP_MAYMOVE, NULL); + if (dest_mapping == MAP_FAILED && errno == EINVAL) { + // Old kernel which doesn't support MREMAP_DONTUNMAP on shmem. + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); + return; + } + + BUG_ON(dest_mapping == MAP_FAILED, "mremap"); + + // Validate that the pages have been moved, we know they were moved if + // the dest_mapping contains a's. + BUG_ON(check_region_contains_byte + (dest_mapping, num_pages * page_size, 'a') != 0, + "pages did not migrate"); + + // Because the region is backed by shmem, we will actually see the same + // memory at the source location still. + BUG_ON(check_region_contains_byte + (source_mapping, num_pages * page_size, 'a') != 0, + "source should have no ptes"); + + BUG_ON(munmap(dest_mapping, num_pages * page_size) == -1, + "unable to unmap destination mapping"); + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); +} + +// This test validates MREMAP_DONTUNMAP will move page tables to a specific +// destination using MREMAP_FIXED, also while validating that the source +// remains intact. +static void mremap_dontunmap_simple_fixed() +{ + unsigned long num_pages = 5; + + // Since we want to guarantee that we can remap to a point, we will + // create a mapping up front. + void *dest_mapping = + mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(dest_mapping == MAP_FAILED, "mmap"); + memset(dest_mapping, 'X', num_pages * page_size); + + void *source_mapping = + mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + memset(source_mapping, 'a', num_pages * page_size); + + void *remapped_mapping = + mremap(source_mapping, num_pages * page_size, num_pages * page_size, + MREMAP_FIXED | MREMAP_DONTUNMAP | MREMAP_MAYMOVE, + dest_mapping); + BUG_ON(remapped_mapping == MAP_FAILED, "mremap"); + BUG_ON(remapped_mapping != dest_mapping, + "mremap should have placed the remapped mapping at dest_mapping"); + + // The dest mapping will have been unmap by mremap so we expect the Xs + // to be gone and replaced with a's. + BUG_ON(check_region_contains_byte + (dest_mapping, num_pages * page_size, 'a') != 0, + "pages did not migrate"); + + // And the source mapping will have had its ptes dropped. + BUG_ON(check_region_contains_byte + (source_mapping, num_pages * page_size, 0) != 0, + "source should have no ptes"); + + BUG_ON(munmap(dest_mapping, num_pages * page_size) == -1, + "unable to unmap destination mapping"); + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); +} + +// This test validates that we can MREMAP_DONTUNMAP for a portion of an +// existing mapping. +static void mremap_dontunmap_partial_mapping() +{ + /* + * source mapping: + * -------------- + * | aaaaaaaaaa | + * -------------- + * to become: + * -------------- + * | aaaaa00000 | + * -------------- + * With the destination mapping containing 5 pages of As. + * --------- + * | aaaaa | + * --------- + */ + unsigned long num_pages = 10; + void *source_mapping = + mmap(NULL, num_pages * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + memset(source_mapping, 'a', num_pages * page_size); + + // We will grab the last 5 pages of the source and move them. + void *dest_mapping = + mremap(source_mapping + (5 * page_size), 5 * page_size, + 5 * page_size, + MREMAP_DONTUNMAP | MREMAP_MAYMOVE, NULL); + BUG_ON(dest_mapping == MAP_FAILED, "mremap"); + + // We expect the first 5 pages of the source to contain a's and the + // final 5 pages to contain zeros. + BUG_ON(check_region_contains_byte(source_mapping, 5 * page_size, 'a') != + 0, "first 5 pages of source should have original pages"); + BUG_ON(check_region_contains_byte + (source_mapping + (5 * page_size), 5 * page_size, 0) != 0, + "final 5 pages of source should have no ptes"); + + // Finally we expect the destination to have 5 pages worth of a's. + BUG_ON(check_region_contains_byte(dest_mapping, 5 * page_size, 'a') != + 0, "dest mapping should contain ptes from the source"); + + BUG_ON(munmap(dest_mapping, 5 * page_size) == -1, + "unable to unmap destination mapping"); + BUG_ON(munmap(source_mapping, num_pages * page_size) == -1, + "unable to unmap source mapping"); +} + +// This test validates that we can remap over only a portion of a mapping. +static void mremap_dontunmap_partial_mapping_overwrite(void) +{ + /* + * source mapping: + * --------- + * |aaaaa| + * --------- + * dest mapping initially: + * ----------- + * |XXXXXXXXXX| + * ------------ + * Source to become: + * --------- + * |00000| + * --------- + * With the destination mapping containing 5 pages of As. + * ------------ + * |aaaaaXXXXX| + * ------------ + */ + void *source_mapping = + mmap(NULL, 5 * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(source_mapping == MAP_FAILED, "mmap"); + memset(source_mapping, 'a', 5 * page_size); + + void *dest_mapping = + mmap(NULL, 10 * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(dest_mapping == MAP_FAILED, "mmap"); + memset(dest_mapping, 'X', 10 * page_size); + + // We will grab the last 5 pages of the source and move them. + void *remapped_mapping = + mremap(source_mapping, 5 * page_size, + 5 * page_size, + MREMAP_DONTUNMAP | MREMAP_MAYMOVE | MREMAP_FIXED, dest_mapping); + BUG_ON(dest_mapping == MAP_FAILED, "mremap"); + BUG_ON(dest_mapping != remapped_mapping, "expected to remap to dest_mapping"); + + BUG_ON(check_region_contains_byte(source_mapping, 5 * page_size, 0) != + 0, "first 5 pages of source should have no ptes"); + + // Finally we expect the destination to have 5 pages worth of a's. + BUG_ON(check_region_contains_byte(dest_mapping, 5 * page_size, 'a') != 0, + "dest mapping should contain ptes from the source"); + + // Finally the last 5 pages shouldn't have been touched. + BUG_ON(check_region_contains_byte(dest_mapping + (5 * page_size), + 5 * page_size, 'X') != 0, + "dest mapping should have retained the last 5 pages"); + + BUG_ON(munmap(dest_mapping, 10 * page_size) == -1, + "unable to unmap destination mapping"); + BUG_ON(munmap(source_mapping, 5 * page_size) == -1, + "unable to unmap source mapping"); +} + +int main(void) +{ + page_size = sysconf(_SC_PAGE_SIZE); + + // test for kernel support for MREMAP_DONTUNMAP skipping the test if + // not. + if (kernel_support_for_mremap_dontunmap() != 0) { + printf("No kernel support for MREMAP_DONTUNMAP\n"); + return KSFT_SKIP; + } + + // Keep a page sized buffer around for when we need it. + page_buffer = + mmap(NULL, page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + BUG_ON(page_buffer == MAP_FAILED, "unable to mmap a page."); + + mremap_dontunmap_simple(); + mremap_dontunmap_simple_shmem(); + mremap_dontunmap_simple_fixed(); + mremap_dontunmap_partial_mapping(); + mremap_dontunmap_partial_mapping_overwrite(); + + BUG_ON(munmap(page_buffer, page_size) == -1, + "unable to unmap page buffer"); + + printf("OK\n"); + return 0; +} diff --git a/tools/testing/selftests/vm/mremap_test.c b/tools/testing/selftests/vm/mremap_test.c new file mode 100644 index 000000000..949634697 --- /dev/null +++ b/tools/testing/selftests/vm/mremap_test.c @@ -0,0 +1,475 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright 2020 Google LLC + */ +#define _GNU_SOURCE + +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +#define EXPECT_SUCCESS 0 +#define EXPECT_FAILURE 1 +#define NON_OVERLAPPING 0 +#define OVERLAPPING 1 +#define NS_PER_SEC 1000000000ULL +#define VALIDATION_DEFAULT_THRESHOLD 4 /* 4MB */ +#define VALIDATION_NO_THRESHOLD 0 /* Verify the entire region */ + +#define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) + +struct config { + unsigned long long src_alignment; + unsigned long long dest_alignment; + unsigned long long region_size; + int overlapping; +}; + +struct test { + const char *name; + struct config config; + int expect_failure; +}; + +enum { + _1KB = 1ULL << 10, /* 1KB -> not page aligned */ + _4KB = 4ULL << 10, + _8KB = 8ULL << 10, + _1MB = 1ULL << 20, + _2MB = 2ULL << 20, + _4MB = 4ULL << 20, + _1GB = 1ULL << 30, + _2GB = 2ULL << 30, + PMD = _2MB, + PUD = _1GB, +}; + +#define PTE page_size + +#define MAKE_TEST(source_align, destination_align, size, \ + overlaps, should_fail, test_name) \ +(struct test){ \ + .name = test_name, \ + .config = { \ + .src_alignment = source_align, \ + .dest_alignment = destination_align, \ + .region_size = size, \ + .overlapping = overlaps, \ + }, \ + .expect_failure = should_fail \ +} + +/* + * Returns false if the requested remap region overlaps with an + * existing mapping (e.g text, stack) else returns true. + */ +static bool is_remap_region_valid(void *addr, unsigned long long size) +{ + void *remap_addr = NULL; + bool ret = true; + + /* Use MAP_FIXED_NOREPLACE flag to ensure region is not mapped */ + remap_addr = mmap(addr, size, PROT_READ | PROT_WRITE, + MAP_FIXED_NOREPLACE | MAP_ANONYMOUS | MAP_SHARED, + -1, 0); + + if (remap_addr == MAP_FAILED) { + if (errno == EEXIST) + ret = false; + } else { + munmap(remap_addr, size); + } + + return ret; +} + +/* Returns mmap_min_addr sysctl tunable from procfs */ +static unsigned long long get_mmap_min_addr(void) +{ + FILE *fp; + int n_matched; + static unsigned long long addr; + + if (addr) + return addr; + + fp = fopen("/proc/sys/vm/mmap_min_addr", "r"); + if (fp == NULL) { + ksft_print_msg("Failed to open /proc/sys/vm/mmap_min_addr: %s\n", + strerror(errno)); + exit(KSFT_SKIP); + } + + n_matched = fscanf(fp, "%llu", &addr); + if (n_matched != 1) { + ksft_print_msg("Failed to read /proc/sys/vm/mmap_min_addr: %s\n", + strerror(errno)); + fclose(fp); + exit(KSFT_SKIP); + } + + fclose(fp); + return addr; +} + +/* + * This test validates that merge is called when expanding a mapping. + * Mapping containing three pages is created, middle page is unmapped + * and then the mapping containing the first page is expanded so that + * it fills the created hole. The two parts should merge creating + * single mapping with three pages. + */ +static void mremap_expand_merge(unsigned long page_size) +{ + char *test_name = "mremap expand merge"; + FILE *fp; + char *line = NULL; + size_t len = 0; + bool success = false; + char *start = mmap(NULL, 3 * page_size, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + + munmap(start + page_size, page_size); + mremap(start, page_size, 2 * page_size, 0); + + fp = fopen("/proc/self/maps", "r"); + if (fp == NULL) { + ksft_test_result_fail("%s\n", test_name); + return; + } + + while (getline(&line, &len, fp) != -1) { + char *first = strtok(line, "- "); + void *first_val = (void *)strtol(first, NULL, 16); + char *second = strtok(NULL, "- "); + void *second_val = (void *) strtol(second, NULL, 16); + + if (first_val == start && second_val == start + 3 * page_size) { + success = true; + break; + } + } + if (success) + ksft_test_result_pass("%s\n", test_name); + else + ksft_test_result_fail("%s\n", test_name); + fclose(fp); +} + +/* + * Returns the start address of the mapping on success, else returns + * NULL on failure. + */ +static void *get_source_mapping(struct config c) +{ + unsigned long long addr = 0ULL; + void *src_addr = NULL; + unsigned long long mmap_min_addr; + + mmap_min_addr = get_mmap_min_addr(); + +retry: + addr += c.src_alignment; + if (addr < mmap_min_addr) + goto retry; + + src_addr = mmap((void *) addr, c.region_size, PROT_READ | PROT_WRITE, + MAP_FIXED_NOREPLACE | MAP_ANONYMOUS | MAP_SHARED, + -1, 0); + if (src_addr == MAP_FAILED) { + if (errno == EPERM || errno == EEXIST) + goto retry; + goto error; + } + /* + * Check that the address is aligned to the specified alignment. + * Addresses which have alignments that are multiples of that + * specified are not considered valid. For instance, 1GB address is + * 2MB-aligned, however it will not be considered valid for a + * requested alignment of 2MB. This is done to reduce coincidental + * alignment in the tests. + */ + if (((unsigned long long) src_addr & (c.src_alignment - 1)) || + !((unsigned long long) src_addr & c.src_alignment)) { + munmap(src_addr, c.region_size); + goto retry; + } + + if (!src_addr) + goto error; + + return src_addr; +error: + ksft_print_msg("Failed to map source region: %s\n", + strerror(errno)); + return NULL; +} + +/* Returns the time taken for the remap on success else returns -1. */ +static long long remap_region(struct config c, unsigned int threshold_mb, + char pattern_seed) +{ + void *addr, *src_addr, *dest_addr; + unsigned long long i; + struct timespec t_start = {0, 0}, t_end = {0, 0}; + long long start_ns, end_ns, align_mask, ret, offset; + unsigned long long threshold; + + if (threshold_mb == VALIDATION_NO_THRESHOLD) + threshold = c.region_size; + else + threshold = MIN(threshold_mb * _1MB, c.region_size); + + src_addr = get_source_mapping(c); + if (!src_addr) { + ret = -1; + goto out; + } + + /* Set byte pattern */ + srand(pattern_seed); + for (i = 0; i < threshold; i++) + memset((char *) src_addr + i, (char) rand(), 1); + + /* Mask to zero out lower bits of address for alignment */ + align_mask = ~(c.dest_alignment - 1); + /* Offset of destination address from the end of the source region */ + offset = (c.overlapping) ? -c.dest_alignment : c.dest_alignment; + addr = (void *) (((unsigned long long) src_addr + c.region_size + + offset) & align_mask); + + /* See comment in get_source_mapping() */ + if (!((unsigned long long) addr & c.dest_alignment)) + addr = (void *) ((unsigned long long) addr | c.dest_alignment); + + /* Don't destroy existing mappings unless expected to overlap */ + while (!is_remap_region_valid(addr, c.region_size) && !c.overlapping) { + /* Check for unsigned overflow */ + if (addr + c.dest_alignment < addr) { + ksft_print_msg("Couldn't find a valid region to remap to\n"); + ret = -1; + goto out; + } + addr += c.dest_alignment; + } + + clock_gettime(CLOCK_MONOTONIC, &t_start); + dest_addr = mremap(src_addr, c.region_size, c.region_size, + MREMAP_MAYMOVE|MREMAP_FIXED, (char *) addr); + clock_gettime(CLOCK_MONOTONIC, &t_end); + + if (dest_addr == MAP_FAILED) { + ksft_print_msg("mremap failed: %s\n", strerror(errno)); + ret = -1; + goto clean_up_src; + } + + /* Verify byte pattern after remapping */ + srand(pattern_seed); + for (i = 0; i < threshold; i++) { + char c = (char) rand(); + + if (((char *) dest_addr)[i] != c) { + ksft_print_msg("Data after remap doesn't match at offset %d\n", + i); + ksft_print_msg("Expected: %#x\t Got: %#x\n", c & 0xff, + ((char *) dest_addr)[i] & 0xff); + ret = -1; + goto clean_up_dest; + } + } + + start_ns = t_start.tv_sec * NS_PER_SEC + t_start.tv_nsec; + end_ns = t_end.tv_sec * NS_PER_SEC + t_end.tv_nsec; + ret = end_ns - start_ns; + +/* + * Since the destination address is specified using MREMAP_FIXED, subsequent + * mremap will unmap any previous mapping at the address range specified by + * dest_addr and region_size. This significantly affects the remap time of + * subsequent tests. So we clean up mappings after each test. + */ +clean_up_dest: + munmap(dest_addr, c.region_size); +clean_up_src: + munmap(src_addr, c.region_size); +out: + return ret; +} + +static void run_mremap_test_case(struct test test_case, int *failures, + unsigned int threshold_mb, + unsigned int pattern_seed) +{ + long long remap_time = remap_region(test_case.config, threshold_mb, + pattern_seed); + + if (remap_time < 0) { + if (test_case.expect_failure) + ksft_test_result_xfail("%s\n\tExpected mremap failure\n", + test_case.name); + else { + ksft_test_result_fail("%s\n", test_case.name); + *failures += 1; + } + } else { + /* + * Comparing mremap time is only applicable if entire region + * was faulted in. + */ + if (threshold_mb == VALIDATION_NO_THRESHOLD || + test_case.config.region_size <= threshold_mb * _1MB) + ksft_test_result_pass("%s\n\tmremap time: %12lldns\n", + test_case.name, remap_time); + else + ksft_test_result_pass("%s\n", test_case.name); + } +} + +static void usage(const char *cmd) +{ + fprintf(stderr, + "Usage: %s [[-t ] [-p ]]\n" + "-t\t only validate threshold_mb of the remapped region\n" + " \t if 0 is supplied no threshold is used; all tests\n" + " \t are run and remapped regions validated fully.\n" + " \t The default threshold used is 4MB.\n" + "-p\t provide a seed to generate the random pattern for\n" + " \t validating the remapped region.\n", cmd); +} + +static int parse_args(int argc, char **argv, unsigned int *threshold_mb, + unsigned int *pattern_seed) +{ + const char *optstr = "t:p:"; + int opt; + + while ((opt = getopt(argc, argv, optstr)) != -1) { + switch (opt) { + case 't': + *threshold_mb = atoi(optarg); + break; + case 'p': + *pattern_seed = atoi(optarg); + break; + default: + usage(argv[0]); + return -1; + } + } + + if (optind < argc) { + usage(argv[0]); + return -1; + } + + return 0; +} + +#define MAX_TEST 13 +#define MAX_PERF_TEST 3 +int main(int argc, char **argv) +{ + int failures = 0; + int i, run_perf_tests; + unsigned int threshold_mb = VALIDATION_DEFAULT_THRESHOLD; + unsigned int pattern_seed; + int num_expand_tests = 1; + struct test test_cases[MAX_TEST]; + struct test perf_test_cases[MAX_PERF_TEST]; + int page_size; + time_t t; + + pattern_seed = (unsigned int) time(&t); + + if (parse_args(argc, argv, &threshold_mb, &pattern_seed) < 0) + exit(EXIT_FAILURE); + + ksft_print_msg("Test configs:\n\tthreshold_mb=%u\n\tpattern_seed=%u\n\n", + threshold_mb, pattern_seed); + + page_size = sysconf(_SC_PAGESIZE); + + /* Expected mremap failures */ + test_cases[0] = MAKE_TEST(page_size, page_size, page_size, + OVERLAPPING, EXPECT_FAILURE, + "mremap - Source and Destination Regions Overlapping"); + + test_cases[1] = MAKE_TEST(page_size, page_size/4, page_size, + NON_OVERLAPPING, EXPECT_FAILURE, + "mremap - Destination Address Misaligned (1KB-aligned)"); + test_cases[2] = MAKE_TEST(page_size/4, page_size, page_size, + NON_OVERLAPPING, EXPECT_FAILURE, + "mremap - Source Address Misaligned (1KB-aligned)"); + + /* Src addr PTE aligned */ + test_cases[3] = MAKE_TEST(PTE, PTE, PTE * 2, + NON_OVERLAPPING, EXPECT_SUCCESS, + "8KB mremap - Source PTE-aligned, Destination PTE-aligned"); + + /* Src addr 1MB aligned */ + test_cases[4] = MAKE_TEST(_1MB, PTE, _2MB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2MB mremap - Source 1MB-aligned, Destination PTE-aligned"); + test_cases[5] = MAKE_TEST(_1MB, _1MB, _2MB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2MB mremap - Source 1MB-aligned, Destination 1MB-aligned"); + + /* Src addr PMD aligned */ + test_cases[6] = MAKE_TEST(PMD, PTE, _4MB, NON_OVERLAPPING, EXPECT_SUCCESS, + "4MB mremap - Source PMD-aligned, Destination PTE-aligned"); + test_cases[7] = MAKE_TEST(PMD, _1MB, _4MB, NON_OVERLAPPING, EXPECT_SUCCESS, + "4MB mremap - Source PMD-aligned, Destination 1MB-aligned"); + test_cases[8] = MAKE_TEST(PMD, PMD, _4MB, NON_OVERLAPPING, EXPECT_SUCCESS, + "4MB mremap - Source PMD-aligned, Destination PMD-aligned"); + + /* Src addr PUD aligned */ + test_cases[9] = MAKE_TEST(PUD, PTE, _2GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2GB mremap - Source PUD-aligned, Destination PTE-aligned"); + test_cases[10] = MAKE_TEST(PUD, _1MB, _2GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2GB mremap - Source PUD-aligned, Destination 1MB-aligned"); + test_cases[11] = MAKE_TEST(PUD, PMD, _2GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2GB mremap - Source PUD-aligned, Destination PMD-aligned"); + test_cases[12] = MAKE_TEST(PUD, PUD, _2GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "2GB mremap - Source PUD-aligned, Destination PUD-aligned"); + + perf_test_cases[0] = MAKE_TEST(page_size, page_size, _1GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "1GB mremap - Source PTE-aligned, Destination PTE-aligned"); + /* + * mremap 1GB region - Page table level aligned time + * comparison. + */ + perf_test_cases[1] = MAKE_TEST(PMD, PMD, _1GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "1GB mremap - Source PMD-aligned, Destination PMD-aligned"); + perf_test_cases[2] = MAKE_TEST(PUD, PUD, _1GB, NON_OVERLAPPING, EXPECT_SUCCESS, + "1GB mremap - Source PUD-aligned, Destination PUD-aligned"); + + run_perf_tests = (threshold_mb == VALIDATION_NO_THRESHOLD) || + (threshold_mb * _1MB >= _1GB); + + ksft_set_plan(ARRAY_SIZE(test_cases) + (run_perf_tests ? + ARRAY_SIZE(perf_test_cases) : 0) + num_expand_tests); + + for (i = 0; i < ARRAY_SIZE(test_cases); i++) + run_mremap_test_case(test_cases[i], &failures, threshold_mb, + pattern_seed); + + mremap_expand_merge(page_size); + + if (run_perf_tests) { + ksft_print_msg("\n%s\n", + "mremap HAVE_MOVE_PMD/PUD optimization time comparison for 1GB region:"); + for (i = 0; i < ARRAY_SIZE(perf_test_cases); i++) + run_mremap_test_case(perf_test_cases[i], &failures, + threshold_mb, pattern_seed); + } + + if (failures > 0) + ksft_exit_fail(); + else + ksft_exit_pass(); +} diff --git a/tools/testing/selftests/vm/on-fault-limit.c b/tools/testing/selftests/vm/on-fault-limit.c new file mode 100644 index 000000000..634d87dfb --- /dev/null +++ b/tools/testing/selftests/vm/on-fault-limit.c @@ -0,0 +1,48 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include + +#ifndef MCL_ONFAULT +#define MCL_ONFAULT (MCL_FUTURE << 1) +#endif + +static int test_limit(void) +{ + int ret = 1; + struct rlimit lims; + void *map; + + if (getrlimit(RLIMIT_MEMLOCK, &lims)) { + perror("getrlimit"); + return ret; + } + + if (mlockall(MCL_ONFAULT | MCL_FUTURE)) { + perror("mlockall"); + return ret; + } + + map = mmap(NULL, 2 * lims.rlim_max, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_POPULATE, -1, 0); + if (map != MAP_FAILED) + printf("mmap should have failed, but didn't\n"); + else { + ret = 0; + munmap(map, 2 * lims.rlim_max); + } + + munlockall(); + return ret; +} + +int main(int argc, char **argv) +{ + int ret = 0; + + ret += test_limit(); + return ret; +} diff --git a/tools/testing/selftests/vm/pkey-helpers.h b/tools/testing/selftests/vm/pkey-helpers.h new file mode 100644 index 000000000..92f3be3dd --- /dev/null +++ b/tools/testing/selftests/vm/pkey-helpers.h @@ -0,0 +1,226 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _PKEYS_HELPER_H +#define _PKEYS_HELPER_H +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +/* Define some kernel-like types */ +#define u8 __u8 +#define u16 __u16 +#define u32 __u32 +#define u64 __u64 + +#define PTR_ERR_ENOTSUP ((void *)-ENOTSUP) + +#ifndef DEBUG_LEVEL +#define DEBUG_LEVEL 0 +#endif +#define DPRINT_IN_SIGNAL_BUF_SIZE 4096 +extern int dprint_in_signal; +extern char dprint_in_signal_buffer[DPRINT_IN_SIGNAL_BUF_SIZE]; + +extern int test_nr; +extern int iteration_nr; + +#ifdef __GNUC__ +__attribute__((format(printf, 1, 2))) +#endif +static inline void sigsafe_printf(const char *format, ...) +{ + va_list ap; + + if (!dprint_in_signal) { + va_start(ap, format); + vprintf(format, ap); + va_end(ap); + } else { + int ret; + /* + * No printf() functions are signal-safe. + * They deadlock easily. Write the format + * string to get some output, even if + * incomplete. + */ + ret = write(1, format, strlen(format)); + if (ret < 0) + exit(1); + } +} +#define dprintf_level(level, args...) do { \ + if (level <= DEBUG_LEVEL) \ + sigsafe_printf(args); \ +} while (0) +#define dprintf0(args...) dprintf_level(0, args) +#define dprintf1(args...) dprintf_level(1, args) +#define dprintf2(args...) dprintf_level(2, args) +#define dprintf3(args...) dprintf_level(3, args) +#define dprintf4(args...) dprintf_level(4, args) + +extern void abort_hooks(void); +#define pkey_assert(condition) do { \ + if (!(condition)) { \ + dprintf0("assert() at %s::%d test_nr: %d iteration: %d\n", \ + __FILE__, __LINE__, \ + test_nr, iteration_nr); \ + dprintf0("errno at assert: %d", errno); \ + abort_hooks(); \ + exit(__LINE__); \ + } \ +} while (0) + +__attribute__((noinline)) int read_ptr(int *ptr); +void expected_pkey_fault(int pkey); +int sys_pkey_alloc(unsigned long flags, unsigned long init_val); +int sys_pkey_free(unsigned long pkey); +int mprotect_pkey(void *ptr, size_t size, unsigned long orig_prot, + unsigned long pkey); +void record_pkey_malloc(void *ptr, long size, int prot); + +#if defined(__i386__) || defined(__x86_64__) /* arch */ +#include "pkey-x86.h" +#elif defined(__powerpc64__) /* arch */ +#include "pkey-powerpc.h" +#else /* arch */ +#error Architecture not supported +#endif /* arch */ + +#define PKEY_MASK (PKEY_DISABLE_ACCESS | PKEY_DISABLE_WRITE) + +static inline u64 set_pkey_bits(u64 reg, int pkey, u64 flags) +{ + u32 shift = pkey_bit_position(pkey); + /* mask out bits from pkey in old value */ + reg &= ~((u64)PKEY_MASK << shift); + /* OR in new bits for pkey */ + reg |= (flags & PKEY_MASK) << shift; + return reg; +} + +static inline u64 get_pkey_bits(u64 reg, int pkey) +{ + u32 shift = pkey_bit_position(pkey); + /* + * shift down the relevant bits to the lowest two, then + * mask off all the other higher bits + */ + return ((reg >> shift) & PKEY_MASK); +} + +extern u64 shadow_pkey_reg; + +static inline u64 _read_pkey_reg(int line) +{ + u64 pkey_reg = __read_pkey_reg(); + + dprintf4("read_pkey_reg(line=%d) pkey_reg: %016llx" + " shadow: %016llx\n", + line, pkey_reg, shadow_pkey_reg); + assert(pkey_reg == shadow_pkey_reg); + + return pkey_reg; +} + +#define read_pkey_reg() _read_pkey_reg(__LINE__) + +static inline void write_pkey_reg(u64 pkey_reg) +{ + dprintf4("%s() changing %016llx to %016llx\n", __func__, + __read_pkey_reg(), pkey_reg); + /* will do the shadow check for us: */ + read_pkey_reg(); + __write_pkey_reg(pkey_reg); + shadow_pkey_reg = pkey_reg; + dprintf4("%s(%016llx) pkey_reg: %016llx\n", __func__, + pkey_reg, __read_pkey_reg()); +} + +/* + * These are technically racy. since something could + * change PKEY register between the read and the write. + */ +static inline void __pkey_access_allow(int pkey, int do_allow) +{ + u64 pkey_reg = read_pkey_reg(); + int bit = pkey * 2; + + if (do_allow) + pkey_reg &= (1<si_pkey; +#else + return (u32 *)(((u8 *)si) + si_pkey_offset); +#endif +} + +static inline int kernel_has_pkeys(void) +{ + /* try allocating a key and see if it succeeds */ + int ret = sys_pkey_alloc(0, 0); + if (ret <= 0) { + return 0; + } + sys_pkey_free(ret); + return 1; +} + +static inline int is_pkeys_supported(void) +{ + /* check if the cpu supports pkeys */ + if (!cpu_has_pkeys()) { + dprintf1("SKIP: %s: no CPU support\n", __func__); + return 0; + } + + /* check if the kernel supports pkeys */ + if (!kernel_has_pkeys()) { + dprintf1("SKIP: %s: no kernel support\n", __func__); + return 0; + } + + return 1; +} + +#endif /* _PKEYS_HELPER_H */ diff --git a/tools/testing/selftests/vm/pkey-powerpc.h b/tools/testing/selftests/vm/pkey-powerpc.h new file mode 100644 index 000000000..1ebb586b2 --- /dev/null +++ b/tools/testing/selftests/vm/pkey-powerpc.h @@ -0,0 +1,133 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _PKEYS_POWERPC_H +#define _PKEYS_POWERPC_H + +#ifndef SYS_mprotect_key +# define SYS_mprotect_key 386 +#endif +#ifndef SYS_pkey_alloc +# define SYS_pkey_alloc 384 +# define SYS_pkey_free 385 +#endif +#define REG_IP_IDX PT_NIP +#define REG_TRAPNO PT_TRAP +#define gregs gp_regs +#define fpregs fp_regs +#define si_pkey_offset 0x20 + +#undef PKEY_DISABLE_ACCESS +#define PKEY_DISABLE_ACCESS 0x3 /* disable read and write */ + +#undef PKEY_DISABLE_WRITE +#define PKEY_DISABLE_WRITE 0x2 + +#define NR_PKEYS 32 +#define NR_RESERVED_PKEYS_4K 27 /* pkey-0, pkey-1, exec-only-pkey + and 24 other keys that cannot be + represented in the PTE */ +#define NR_RESERVED_PKEYS_64K_3KEYS 3 /* PowerNV and KVM: pkey-0, + pkey-1 and exec-only key */ +#define NR_RESERVED_PKEYS_64K_4KEYS 4 /* PowerVM: pkey-0, pkey-1, + pkey-31 and exec-only key */ +#define PKEY_BITS_PER_PKEY 2 +#define HPAGE_SIZE (1UL << 24) +#define PAGE_SIZE sysconf(_SC_PAGESIZE) + +static inline u32 pkey_bit_position(int pkey) +{ + return (NR_PKEYS - pkey - 1) * PKEY_BITS_PER_PKEY; +} + +static inline u64 __read_pkey_reg(void) +{ + u64 pkey_reg; + + asm volatile("mfspr %0, 0xd" : "=r" (pkey_reg)); + + return pkey_reg; +} + +static inline void __write_pkey_reg(u64 pkey_reg) +{ + u64 amr = pkey_reg; + + dprintf4("%s() changing %016llx to %016llx\n", + __func__, __read_pkey_reg(), pkey_reg); + + asm volatile("isync; mtspr 0xd, %0; isync" + : : "r" ((unsigned long)(amr)) : "memory"); + + dprintf4("%s() pkey register after changing %016llx to %016llx\n", + __func__, __read_pkey_reg(), pkey_reg); +} + +static inline int cpu_has_pkeys(void) +{ + /* No simple way to determine this */ + return 1; +} + +static inline bool arch_is_powervm() +{ + struct stat buf; + + if ((stat("/sys/firmware/devicetree/base/ibm,partition-name", &buf) == 0) && + (stat("/sys/firmware/devicetree/base/hmc-managed?", &buf) == 0) && + (stat("/sys/firmware/devicetree/base/chosen/qemu,graphic-width", &buf) == -1) ) + return true; + + return false; +} + +static inline int get_arch_reserved_keys(void) +{ + if (sysconf(_SC_PAGESIZE) == 4096) + return NR_RESERVED_PKEYS_4K; + else + if (arch_is_powervm()) + return NR_RESERVED_PKEYS_64K_4KEYS; + else + return NR_RESERVED_PKEYS_64K_3KEYS; +} + +void expect_fault_on_read_execonly_key(void *p1, int pkey) +{ + /* + * powerpc does not allow userspace to change permissions of exec-only + * keys since those keys are not allocated by userspace. The signal + * handler wont be able to reset the permissions, which means the code + * will infinitely continue to segfault here. + */ + return; +} + +/* 4-byte instructions * 16384 = 64K page */ +#define __page_o_noops() asm(".rept 16384 ; nop; .endr") + +void *malloc_pkey_with_mprotect_subpage(long size, int prot, u16 pkey) +{ + void *ptr; + int ret; + + dprintf1("doing %s(size=%ld, prot=0x%x, pkey=%d)\n", __func__, + size, prot, pkey); + pkey_assert(pkey < NR_PKEYS); + ptr = mmap(NULL, size, prot, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + pkey_assert(ptr != (void *)-1); + + ret = syscall(__NR_subpage_prot, ptr, size, NULL); + if (ret) { + perror("subpage_perm"); + return PTR_ERR_ENOTSUP; + } + + ret = mprotect_pkey((void *)ptr, PAGE_SIZE, prot, pkey); + pkey_assert(!ret); + record_pkey_malloc(ptr, size, prot); + + dprintf1("%s() for pkey %d @ %p\n", __func__, pkey, ptr); + return ptr; +} + +#endif /* _PKEYS_POWERPC_H */ diff --git a/tools/testing/selftests/vm/pkey-x86.h b/tools/testing/selftests/vm/pkey-x86.h new file mode 100644 index 000000000..72c14cd3d --- /dev/null +++ b/tools/testing/selftests/vm/pkey-x86.h @@ -0,0 +1,177 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _PKEYS_X86_H +#define _PKEYS_X86_H + +#ifdef __i386__ + +#ifndef SYS_mprotect_key +# define SYS_mprotect_key 380 +#endif + +#ifndef SYS_pkey_alloc +# define SYS_pkey_alloc 381 +# define SYS_pkey_free 382 +#endif + +#define REG_IP_IDX REG_EIP +#define si_pkey_offset 0x14 + +#else + +#ifndef SYS_mprotect_key +# define SYS_mprotect_key 329 +#endif + +#ifndef SYS_pkey_alloc +# define SYS_pkey_alloc 330 +# define SYS_pkey_free 331 +#endif + +#define REG_IP_IDX REG_RIP +#define si_pkey_offset 0x20 + +#endif + +#ifndef PKEY_DISABLE_ACCESS +# define PKEY_DISABLE_ACCESS 0x1 +#endif + +#ifndef PKEY_DISABLE_WRITE +# define PKEY_DISABLE_WRITE 0x2 +#endif + +#define NR_PKEYS 16 +#define NR_RESERVED_PKEYS 2 /* pkey-0 and exec-only-pkey */ +#define PKEY_BITS_PER_PKEY 2 +#define HPAGE_SIZE (1UL<<21) +#define PAGE_SIZE 4096 +#define MB (1<<20) + +static inline void __page_o_noops(void) +{ + /* 8-bytes of instruction * 512 bytes = 1 page */ + asm(".rept 512 ; nopl 0x7eeeeeee(%eax) ; .endr"); +} + +static inline u64 __read_pkey_reg(void) +{ + unsigned int eax, edx; + unsigned int ecx = 0; + unsigned pkey_reg; + + asm volatile(".byte 0x0f,0x01,0xee\n\t" + : "=a" (eax), "=d" (edx) + : "c" (ecx)); + pkey_reg = eax; + return pkey_reg; +} + +static inline void __write_pkey_reg(u64 pkey_reg) +{ + unsigned int eax = pkey_reg; + unsigned int ecx = 0; + unsigned int edx = 0; + + dprintf4("%s() changing %016llx to %016llx\n", __func__, + __read_pkey_reg(), pkey_reg); + asm volatile(".byte 0x0f,0x01,0xef\n\t" + : : "a" (eax), "c" (ecx), "d" (edx)); + assert(pkey_reg == __read_pkey_reg()); +} + +/* Intel-defined CPU features, CPUID level 0x00000007:0 (ecx) */ +#define X86_FEATURE_PKU (1<<3) /* Protection Keys for Userspace */ +#define X86_FEATURE_OSPKE (1<<4) /* OS Protection Keys Enable */ + +static inline int cpu_has_pkeys(void) +{ + unsigned int eax; + unsigned int ebx; + unsigned int ecx; + unsigned int edx; + + __cpuid_count(0x7, 0x0, eax, ebx, ecx, edx); + + if (!(ecx & X86_FEATURE_PKU)) { + dprintf2("cpu does not have PKU\n"); + return 0; + } + if (!(ecx & X86_FEATURE_OSPKE)) { + dprintf2("cpu does not have OSPKE\n"); + return 0; + } + return 1; +} + +static inline int cpu_max_xsave_size(void) +{ + unsigned long XSTATE_CPUID = 0xd; + unsigned int eax; + unsigned int ebx; + unsigned int ecx; + unsigned int edx; + + __cpuid_count(XSTATE_CPUID, 0, eax, ebx, ecx, edx); + return ecx; +} + +static inline u32 pkey_bit_position(int pkey) +{ + return pkey * PKEY_BITS_PER_PKEY; +} + +#define XSTATE_PKEY_BIT (9) +#define XSTATE_PKEY 0x200 +#define XSTATE_BV_OFFSET 512 + +int pkey_reg_xstate_offset(void) +{ + unsigned int eax; + unsigned int ebx; + unsigned int ecx; + unsigned int edx; + int xstate_offset; + int xstate_size; + unsigned long XSTATE_CPUID = 0xd; + int leaf; + + /* assume that XSTATE_PKEY is set in XCR0 */ + leaf = XSTATE_PKEY_BIT; + { + __cpuid_count(XSTATE_CPUID, leaf, eax, ebx, ecx, edx); + + if (leaf == XSTATE_PKEY_BIT) { + xstate_offset = ebx; + xstate_size = eax; + } + } + + if (xstate_size == 0) { + printf("could not find size/offset of PKEY in xsave state\n"); + return 0; + } + + return xstate_offset; +} + +static inline int get_arch_reserved_keys(void) +{ + return NR_RESERVED_PKEYS; +} + +void expect_fault_on_read_execonly_key(void *p1, int pkey) +{ + int ptr_contents; + + ptr_contents = read_ptr(p1); + dprintf2("ptr (%p) contents@%d: %x\n", p1, __LINE__, ptr_contents); + expected_pkey_fault(pkey); +} + +void *malloc_pkey_with_mprotect_subpage(long size, int prot, u16 pkey) +{ + return PTR_ERR_ENOTSUP; +} + +#endif /* _PKEYS_X86_H */ diff --git a/tools/testing/selftests/vm/protection_keys.c b/tools/testing/selftests/vm/protection_keys.c new file mode 100644 index 000000000..95f403a0c --- /dev/null +++ b/tools/testing/selftests/vm/protection_keys.c @@ -0,0 +1,1788 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Tests Memory Protection Keys (see Documentation/core-api/protection-keys.rst) + * + * There are examples in here of: + * * how to set protection keys on memory + * * how to set/clear bits in pkey registers (the rights register) + * * how to handle SEGV_PKUERR signals and extract pkey-relevant + * information from the siginfo + * + * Things to add: + * make sure KSM and KSM COW breaking works + * prefault pages in at malloc, or not + * protect MPX bounds tables with protection keys? + * make sure VMA splitting/merging is working correctly + * OOMs can destroy mm->mmap (see exit_mmap()), so make sure it is immune to pkeys + * look for pkey "leaks" where it is still set on a VMA but "freed" back to the kernel + * do a plain mprotect() to a mprotect_pkey() area and make sure the pkey sticks + * + * Compile like this: + * gcc -mxsave -o protection_keys -O2 -g -std=gnu99 -pthread -Wall protection_keys.c -lrt -ldl -lm + * gcc -mxsave -m32 -o protection_keys_32 -O2 -g -std=gnu99 -pthread -Wall protection_keys.c -lrt -ldl -lm + */ +#define _GNU_SOURCE +#define __SANE_USERSPACE_TYPES__ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "pkey-helpers.h" + +int iteration_nr = 1; +int test_nr; + +u64 shadow_pkey_reg; +int dprint_in_signal; +char dprint_in_signal_buffer[DPRINT_IN_SIGNAL_BUF_SIZE]; + +void cat_into_file(char *str, char *file) +{ + int fd = open(file, O_RDWR); + int ret; + + dprintf2("%s(): writing '%s' to '%s'\n", __func__, str, file); + /* + * these need to be raw because they are called under + * pkey_assert() + */ + if (fd < 0) { + fprintf(stderr, "error opening '%s'\n", str); + perror("error: "); + exit(__LINE__); + } + + ret = write(fd, str, strlen(str)); + if (ret != strlen(str)) { + perror("write to file failed"); + fprintf(stderr, "filename: '%s' str: '%s'\n", file, str); + exit(__LINE__); + } + close(fd); +} + +#if CONTROL_TRACING > 0 +static int warned_tracing; +int tracing_root_ok(void) +{ + if (geteuid() != 0) { + if (!warned_tracing) + fprintf(stderr, "WARNING: not run as root, " + "can not do tracing control\n"); + warned_tracing = 1; + return 0; + } + return 1; +} +#endif + +void tracing_on(void) +{ +#if CONTROL_TRACING > 0 +#define TRACEDIR "/sys/kernel/debug/tracing" + char pidstr[32]; + + if (!tracing_root_ok()) + return; + + sprintf(pidstr, "%d", getpid()); + cat_into_file("0", TRACEDIR "/tracing_on"); + cat_into_file("\n", TRACEDIR "/trace"); + if (1) { + cat_into_file("function_graph", TRACEDIR "/current_tracer"); + cat_into_file("1", TRACEDIR "/options/funcgraph-proc"); + } else { + cat_into_file("nop", TRACEDIR "/current_tracer"); + } + cat_into_file(pidstr, TRACEDIR "/set_ftrace_pid"); + cat_into_file("1", TRACEDIR "/tracing_on"); + dprintf1("enabled tracing\n"); +#endif +} + +void tracing_off(void) +{ +#if CONTROL_TRACING > 0 + if (!tracing_root_ok()) + return; + cat_into_file("0", "/sys/kernel/debug/tracing/tracing_on"); +#endif +} + +void abort_hooks(void) +{ + fprintf(stderr, "running %s()...\n", __func__); + tracing_off(); +#ifdef SLEEP_ON_ABORT + sleep(SLEEP_ON_ABORT); +#endif +} + +/* + * This attempts to have roughly a page of instructions followed by a few + * instructions that do a write, and another page of instructions. That + * way, we are pretty sure that the write is in the second page of + * instructions and has at least a page of padding behind it. + * + * *That* lets us be sure to madvise() away the write instruction, which + * will then fault, which makes sure that the fault code handles + * execute-only memory properly. + */ +#ifdef __powerpc64__ +/* This way, both 4K and 64K alignment are maintained */ +__attribute__((__aligned__(65536))) +#else +__attribute__((__aligned__(PAGE_SIZE))) +#endif +void lots_o_noops_around_write(int *write_to_me) +{ + dprintf3("running %s()\n", __func__); + __page_o_noops(); + /* Assume this happens in the second page of instructions: */ + *write_to_me = __LINE__; + /* pad out by another page: */ + __page_o_noops(); + dprintf3("%s() done\n", __func__); +} + +void dump_mem(void *dumpme, int len_bytes) +{ + char *c = (void *)dumpme; + int i; + + for (i = 0; i < len_bytes; i += sizeof(u64)) { + u64 *ptr = (u64 *)(c + i); + dprintf1("dump[%03d][@%p]: %016llx\n", i, ptr, *ptr); + } +} + +static u32 hw_pkey_get(int pkey, unsigned long flags) +{ + u64 pkey_reg = __read_pkey_reg(); + + dprintf1("%s(pkey=%d, flags=%lx) = %x / %d\n", + __func__, pkey, flags, 0, 0); + dprintf2("%s() raw pkey_reg: %016llx\n", __func__, pkey_reg); + + return (u32) get_pkey_bits(pkey_reg, pkey); +} + +static int hw_pkey_set(int pkey, unsigned long rights, unsigned long flags) +{ + u32 mask = (PKEY_DISABLE_ACCESS|PKEY_DISABLE_WRITE); + u64 old_pkey_reg = __read_pkey_reg(); + u64 new_pkey_reg; + + /* make sure that 'rights' only contains the bits we expect: */ + assert(!(rights & ~mask)); + + /* modify bits accordingly in old pkey_reg and assign it */ + new_pkey_reg = set_pkey_bits(old_pkey_reg, pkey, rights); + + __write_pkey_reg(new_pkey_reg); + + dprintf3("%s(pkey=%d, rights=%lx, flags=%lx) = %x" + " pkey_reg now: %016llx old_pkey_reg: %016llx\n", + __func__, pkey, rights, flags, 0, __read_pkey_reg(), + old_pkey_reg); + return 0; +} + +void pkey_disable_set(int pkey, int flags) +{ + unsigned long syscall_flags = 0; + int ret; + int pkey_rights; + u64 orig_pkey_reg = read_pkey_reg(); + + dprintf1("START->%s(%d, 0x%x)\n", __func__, + pkey, flags); + pkey_assert(flags & (PKEY_DISABLE_ACCESS | PKEY_DISABLE_WRITE)); + + pkey_rights = hw_pkey_get(pkey, syscall_flags); + + dprintf1("%s(%d) hw_pkey_get(%d): %x\n", __func__, + pkey, pkey, pkey_rights); + + pkey_assert(pkey_rights >= 0); + + pkey_rights |= flags; + + ret = hw_pkey_set(pkey, pkey_rights, syscall_flags); + assert(!ret); + /* pkey_reg and flags have the same format */ + shadow_pkey_reg = set_pkey_bits(shadow_pkey_reg, pkey, pkey_rights); + dprintf1("%s(%d) shadow: 0x%016llx\n", + __func__, pkey, shadow_pkey_reg); + + pkey_assert(ret >= 0); + + pkey_rights = hw_pkey_get(pkey, syscall_flags); + dprintf1("%s(%d) hw_pkey_get(%d): %x\n", __func__, + pkey, pkey, pkey_rights); + + dprintf1("%s(%d) pkey_reg: 0x%016llx\n", + __func__, pkey, read_pkey_reg()); + if (flags) + pkey_assert(read_pkey_reg() >= orig_pkey_reg); + dprintf1("END<---%s(%d, 0x%x)\n", __func__, + pkey, flags); +} + +void pkey_disable_clear(int pkey, int flags) +{ + unsigned long syscall_flags = 0; + int ret; + int pkey_rights = hw_pkey_get(pkey, syscall_flags); + u64 orig_pkey_reg = read_pkey_reg(); + + pkey_assert(flags & (PKEY_DISABLE_ACCESS | PKEY_DISABLE_WRITE)); + + dprintf1("%s(%d) hw_pkey_get(%d): %x\n", __func__, + pkey, pkey, pkey_rights); + pkey_assert(pkey_rights >= 0); + + pkey_rights &= ~flags; + + ret = hw_pkey_set(pkey, pkey_rights, 0); + shadow_pkey_reg = set_pkey_bits(shadow_pkey_reg, pkey, pkey_rights); + pkey_assert(ret >= 0); + + pkey_rights = hw_pkey_get(pkey, syscall_flags); + dprintf1("%s(%d) hw_pkey_get(%d): %x\n", __func__, + pkey, pkey, pkey_rights); + + dprintf1("%s(%d) pkey_reg: 0x%016llx\n", __func__, + pkey, read_pkey_reg()); + if (flags) + assert(read_pkey_reg() <= orig_pkey_reg); +} + +void pkey_write_allow(int pkey) +{ + pkey_disable_clear(pkey, PKEY_DISABLE_WRITE); +} +void pkey_write_deny(int pkey) +{ + pkey_disable_set(pkey, PKEY_DISABLE_WRITE); +} +void pkey_access_allow(int pkey) +{ + pkey_disable_clear(pkey, PKEY_DISABLE_ACCESS); +} +void pkey_access_deny(int pkey) +{ + pkey_disable_set(pkey, PKEY_DISABLE_ACCESS); +} + +/* Failed address bound checks: */ +#ifndef SEGV_BNDERR +# define SEGV_BNDERR 3 +#endif + +#ifndef SEGV_PKUERR +# define SEGV_PKUERR 4 +#endif + +static char *si_code_str(int si_code) +{ + if (si_code == SEGV_MAPERR) + return "SEGV_MAPERR"; + if (si_code == SEGV_ACCERR) + return "SEGV_ACCERR"; + if (si_code == SEGV_BNDERR) + return "SEGV_BNDERR"; + if (si_code == SEGV_PKUERR) + return "SEGV_PKUERR"; + return "UNKNOWN"; +} + +int pkey_faults; +int last_si_pkey = -1; +void signal_handler(int signum, siginfo_t *si, void *vucontext) +{ + ucontext_t *uctxt = vucontext; + int trapno; + unsigned long ip; + char *fpregs; +#if defined(__i386__) || defined(__x86_64__) /* arch */ + u32 *pkey_reg_ptr; + int pkey_reg_offset; +#endif /* arch */ + u64 siginfo_pkey; + u32 *si_pkey_ptr; + + dprint_in_signal = 1; + dprintf1(">>>>===============SIGSEGV============================\n"); + dprintf1("%s()::%d, pkey_reg: 0x%016llx shadow: %016llx\n", + __func__, __LINE__, + __read_pkey_reg(), shadow_pkey_reg); + + trapno = uctxt->uc_mcontext.gregs[REG_TRAPNO]; + ip = uctxt->uc_mcontext.gregs[REG_IP_IDX]; + fpregs = (char *) uctxt->uc_mcontext.fpregs; + + dprintf2("%s() trapno: %d ip: 0x%016lx info->si_code: %s/%d\n", + __func__, trapno, ip, si_code_str(si->si_code), + si->si_code); + +#if defined(__i386__) || defined(__x86_64__) /* arch */ +#ifdef __i386__ + /* + * 32-bit has some extra padding so that userspace can tell whether + * the XSTATE header is present in addition to the "legacy" FPU + * state. We just assume that it is here. + */ + fpregs += 0x70; +#endif /* i386 */ + pkey_reg_offset = pkey_reg_xstate_offset(); + pkey_reg_ptr = (void *)(&fpregs[pkey_reg_offset]); + + /* + * If we got a PKEY fault, we *HAVE* to have at least one bit set in + * here. + */ + dprintf1("pkey_reg_xstate_offset: %d\n", pkey_reg_xstate_offset()); + if (DEBUG_LEVEL > 4) + dump_mem(pkey_reg_ptr - 128, 256); + pkey_assert(*pkey_reg_ptr); +#endif /* arch */ + + dprintf1("siginfo: %p\n", si); + dprintf1(" fpregs: %p\n", fpregs); + + if ((si->si_code == SEGV_MAPERR) || + (si->si_code == SEGV_ACCERR) || + (si->si_code == SEGV_BNDERR)) { + printf("non-PK si_code, exiting...\n"); + exit(4); + } + + si_pkey_ptr = siginfo_get_pkey_ptr(si); + dprintf1("si_pkey_ptr: %p\n", si_pkey_ptr); + dump_mem((u8 *)si_pkey_ptr - 8, 24); + siginfo_pkey = *si_pkey_ptr; + pkey_assert(siginfo_pkey < NR_PKEYS); + last_si_pkey = siginfo_pkey; + + /* + * need __read_pkey_reg() version so we do not do shadow_pkey_reg + * checking + */ + dprintf1("signal pkey_reg from pkey_reg: %016llx\n", + __read_pkey_reg()); + dprintf1("pkey from siginfo: %016llx\n", siginfo_pkey); +#if defined(__i386__) || defined(__x86_64__) /* arch */ + dprintf1("signal pkey_reg from xsave: %08x\n", *pkey_reg_ptr); + *(u64 *)pkey_reg_ptr = 0x00000000; + dprintf1("WARNING: set PKEY_REG=0 to allow faulting instruction to continue\n"); +#elif defined(__powerpc64__) /* arch */ + /* restore access and let the faulting instruction continue */ + pkey_access_allow(siginfo_pkey); +#endif /* arch */ + pkey_faults++; + dprintf1("<<<<==================================================\n"); + dprint_in_signal = 0; +} + +int wait_all_children(void) +{ + int status; + return waitpid(-1, &status, 0); +} + +void sig_chld(int x) +{ + dprint_in_signal = 1; + dprintf2("[%d] SIGCHLD: %d\n", getpid(), x); + dprint_in_signal = 0; +} + +void setup_sigsegv_handler(void) +{ + int r, rs; + struct sigaction newact; + struct sigaction oldact; + + /* #PF is mapped to sigsegv */ + int signum = SIGSEGV; + + newact.sa_handler = 0; + newact.sa_sigaction = signal_handler; + + /*sigset_t - signals to block while in the handler */ + /* get the old signal mask. */ + rs = sigprocmask(SIG_SETMASK, 0, &newact.sa_mask); + pkey_assert(rs == 0); + + /* call sa_sigaction, not sa_handler*/ + newact.sa_flags = SA_SIGINFO; + + newact.sa_restorer = 0; /* void(*)(), obsolete */ + r = sigaction(signum, &newact, &oldact); + r = sigaction(SIGALRM, &newact, &oldact); + pkey_assert(r == 0); +} + +void setup_handlers(void) +{ + signal(SIGCHLD, &sig_chld); + setup_sigsegv_handler(); +} + +pid_t fork_lazy_child(void) +{ + pid_t forkret; + + forkret = fork(); + pkey_assert(forkret >= 0); + dprintf3("[%d] fork() ret: %d\n", getpid(), forkret); + + if (!forkret) { + /* in the child */ + while (1) { + dprintf1("child sleeping...\n"); + sleep(30); + } + } + return forkret; +} + +int sys_mprotect_pkey(void *ptr, size_t size, unsigned long orig_prot, + unsigned long pkey) +{ + int sret; + + dprintf2("%s(0x%p, %zx, prot=%lx, pkey=%lx)\n", __func__, + ptr, size, orig_prot, pkey); + + errno = 0; + sret = syscall(SYS_mprotect_key, ptr, size, orig_prot, pkey); + if (errno) { + dprintf2("SYS_mprotect_key sret: %d\n", sret); + dprintf2("SYS_mprotect_key prot: 0x%lx\n", orig_prot); + dprintf2("SYS_mprotect_key failed, errno: %d\n", errno); + if (DEBUG_LEVEL >= 2) + perror("SYS_mprotect_pkey"); + } + return sret; +} + +int sys_pkey_alloc(unsigned long flags, unsigned long init_val) +{ + int ret = syscall(SYS_pkey_alloc, flags, init_val); + dprintf1("%s(flags=%lx, init_val=%lx) syscall ret: %d errno: %d\n", + __func__, flags, init_val, ret, errno); + return ret; +} + +int alloc_pkey(void) +{ + int ret; + unsigned long init_val = 0x0; + + dprintf1("%s()::%d, pkey_reg: 0x%016llx shadow: %016llx\n", + __func__, __LINE__, __read_pkey_reg(), shadow_pkey_reg); + ret = sys_pkey_alloc(0, init_val); + /* + * pkey_alloc() sets PKEY register, so we need to reflect it in + * shadow_pkey_reg: + */ + dprintf4("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + if (ret > 0) { + /* clear both the bits: */ + shadow_pkey_reg = set_pkey_bits(shadow_pkey_reg, ret, + ~PKEY_MASK); + dprintf4("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, + __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + /* + * move the new state in from init_val + * (remember, we cheated and init_val == pkey_reg format) + */ + shadow_pkey_reg = set_pkey_bits(shadow_pkey_reg, ret, + init_val); + } + dprintf4("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + dprintf1("%s()::%d errno: %d\n", __func__, __LINE__, errno); + /* for shadow checking: */ + read_pkey_reg(); + dprintf4("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + return ret; +} + +int sys_pkey_free(unsigned long pkey) +{ + int ret = syscall(SYS_pkey_free, pkey); + dprintf1("%s(pkey=%ld) syscall ret: %d\n", __func__, pkey, ret); + return ret; +} + +/* + * I had a bug where pkey bits could be set by mprotect() but + * not cleared. This ensures we get lots of random bit sets + * and clears on the vma and pte pkey bits. + */ +int alloc_random_pkey(void) +{ + int max_nr_pkey_allocs; + int ret; + int i; + int alloced_pkeys[NR_PKEYS]; + int nr_alloced = 0; + int random_index; + memset(alloced_pkeys, 0, sizeof(alloced_pkeys)); + + /* allocate every possible key and make a note of which ones we got */ + max_nr_pkey_allocs = NR_PKEYS; + for (i = 0; i < max_nr_pkey_allocs; i++) { + int new_pkey = alloc_pkey(); + if (new_pkey < 0) + break; + alloced_pkeys[nr_alloced++] = new_pkey; + } + + pkey_assert(nr_alloced > 0); + /* select a random one out of the allocated ones */ + random_index = rand() % nr_alloced; + ret = alloced_pkeys[random_index]; + /* now zero it out so we don't free it next */ + alloced_pkeys[random_index] = 0; + + /* go through the allocated ones that we did not want and free them */ + for (i = 0; i < nr_alloced; i++) { + int free_ret; + if (!alloced_pkeys[i]) + continue; + free_ret = sys_pkey_free(alloced_pkeys[i]); + pkey_assert(!free_ret); + } + dprintf1("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", __func__, + __LINE__, ret, __read_pkey_reg(), shadow_pkey_reg); + return ret; +} + +int mprotect_pkey(void *ptr, size_t size, unsigned long orig_prot, + unsigned long pkey) +{ + int nr_iterations = random() % 100; + int ret; + + while (0) { + int rpkey = alloc_random_pkey(); + ret = sys_mprotect_pkey(ptr, size, orig_prot, pkey); + dprintf1("sys_mprotect_pkey(%p, %zx, prot=0x%lx, pkey=%ld) ret: %d\n", + ptr, size, orig_prot, pkey, ret); + if (nr_iterations-- < 0) + break; + + dprintf1("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + sys_pkey_free(rpkey); + dprintf1("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, ret, __read_pkey_reg(), + shadow_pkey_reg); + } + pkey_assert(pkey < NR_PKEYS); + + ret = sys_mprotect_pkey(ptr, size, orig_prot, pkey); + dprintf1("mprotect_pkey(%p, %zx, prot=0x%lx, pkey=%ld) ret: %d\n", + ptr, size, orig_prot, pkey, ret); + pkey_assert(!ret); + dprintf1("%s()::%d, ret: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", __func__, + __LINE__, ret, __read_pkey_reg(), shadow_pkey_reg); + return ret; +} + +struct pkey_malloc_record { + void *ptr; + long size; + int prot; +}; +struct pkey_malloc_record *pkey_malloc_records; +struct pkey_malloc_record *pkey_last_malloc_record; +long nr_pkey_malloc_records; +void record_pkey_malloc(void *ptr, long size, int prot) +{ + long i; + struct pkey_malloc_record *rec = NULL; + + for (i = 0; i < nr_pkey_malloc_records; i++) { + rec = &pkey_malloc_records[i]; + /* find a free record */ + if (rec) + break; + } + if (!rec) { + /* every record is full */ + size_t old_nr_records = nr_pkey_malloc_records; + size_t new_nr_records = (nr_pkey_malloc_records * 2 + 1); + size_t new_size = new_nr_records * sizeof(struct pkey_malloc_record); + dprintf2("new_nr_records: %zd\n", new_nr_records); + dprintf2("new_size: %zd\n", new_size); + pkey_malloc_records = realloc(pkey_malloc_records, new_size); + pkey_assert(pkey_malloc_records != NULL); + rec = &pkey_malloc_records[nr_pkey_malloc_records]; + /* + * realloc() does not initialize memory, so zero it from + * the first new record all the way to the end. + */ + for (i = 0; i < new_nr_records - old_nr_records; i++) + memset(rec + i, 0, sizeof(*rec)); + } + dprintf3("filling malloc record[%d/%p]: {%p, %ld}\n", + (int)(rec - pkey_malloc_records), rec, ptr, size); + rec->ptr = ptr; + rec->size = size; + rec->prot = prot; + pkey_last_malloc_record = rec; + nr_pkey_malloc_records++; +} + +void free_pkey_malloc(void *ptr) +{ + long i; + int ret; + dprintf3("%s(%p)\n", __func__, ptr); + for (i = 0; i < nr_pkey_malloc_records; i++) { + struct pkey_malloc_record *rec = &pkey_malloc_records[i]; + dprintf4("looking for ptr %p at record[%ld/%p]: {%p, %ld}\n", + ptr, i, rec, rec->ptr, rec->size); + if ((ptr < rec->ptr) || + (ptr >= rec->ptr + rec->size)) + continue; + + dprintf3("found ptr %p at record[%ld/%p]: {%p, %ld}\n", + ptr, i, rec, rec->ptr, rec->size); + nr_pkey_malloc_records--; + ret = munmap(rec->ptr, rec->size); + dprintf3("munmap ret: %d\n", ret); + pkey_assert(!ret); + dprintf3("clearing rec->ptr, rec: %p\n", rec); + rec->ptr = NULL; + dprintf3("done clearing rec->ptr, rec: %p\n", rec); + return; + } + pkey_assert(false); +} + + +void *malloc_pkey_with_mprotect(long size, int prot, u16 pkey) +{ + void *ptr; + int ret; + + read_pkey_reg(); + dprintf1("doing %s(size=%ld, prot=0x%x, pkey=%d)\n", __func__, + size, prot, pkey); + pkey_assert(pkey < NR_PKEYS); + ptr = mmap(NULL, size, prot, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + pkey_assert(ptr != (void *)-1); + ret = mprotect_pkey((void *)ptr, PAGE_SIZE, prot, pkey); + pkey_assert(!ret); + record_pkey_malloc(ptr, size, prot); + read_pkey_reg(); + + dprintf1("%s() for pkey %d @ %p\n", __func__, pkey, ptr); + return ptr; +} + +void *malloc_pkey_anon_huge(long size, int prot, u16 pkey) +{ + int ret; + void *ptr; + + dprintf1("doing %s(size=%ld, prot=0x%x, pkey=%d)\n", __func__, + size, prot, pkey); + /* + * Guarantee we can fit at least one huge page in the resulting + * allocation by allocating space for 2: + */ + size = ALIGN_UP(size, HPAGE_SIZE * 2); + ptr = mmap(NULL, size, PROT_NONE, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + pkey_assert(ptr != (void *)-1); + record_pkey_malloc(ptr, size, prot); + mprotect_pkey(ptr, size, prot, pkey); + + dprintf1("unaligned ptr: %p\n", ptr); + ptr = ALIGN_PTR_UP(ptr, HPAGE_SIZE); + dprintf1(" aligned ptr: %p\n", ptr); + ret = madvise(ptr, HPAGE_SIZE, MADV_HUGEPAGE); + dprintf1("MADV_HUGEPAGE ret: %d\n", ret); + ret = madvise(ptr, HPAGE_SIZE, MADV_WILLNEED); + dprintf1("MADV_WILLNEED ret: %d\n", ret); + memset(ptr, 0, HPAGE_SIZE); + + dprintf1("mmap()'d thp for pkey %d @ %p\n", pkey, ptr); + return ptr; +} + +int hugetlb_setup_ok; +#define SYSFS_FMT_NR_HUGE_PAGES "/sys/kernel/mm/hugepages/hugepages-%ldkB/nr_hugepages" +#define GET_NR_HUGE_PAGES 10 +void setup_hugetlbfs(void) +{ + int err; + int fd; + char buf[256]; + long hpagesz_kb; + long hpagesz_mb; + + if (geteuid() != 0) { + fprintf(stderr, "WARNING: not run as root, can not do hugetlb test\n"); + return; + } + + cat_into_file(__stringify(GET_NR_HUGE_PAGES), "/proc/sys/vm/nr_hugepages"); + + /* + * Now go make sure that we got the pages and that they + * are PMD-level pages. Someone might have made PUD-level + * pages the default. + */ + hpagesz_kb = HPAGE_SIZE / 1024; + hpagesz_mb = hpagesz_kb / 1024; + sprintf(buf, SYSFS_FMT_NR_HUGE_PAGES, hpagesz_kb); + fd = open(buf, O_RDONLY); + if (fd < 0) { + fprintf(stderr, "opening sysfs %ldM hugetlb config: %s\n", + hpagesz_mb, strerror(errno)); + return; + } + + /* -1 to guarantee leaving the trailing \0 */ + err = read(fd, buf, sizeof(buf)-1); + close(fd); + if (err <= 0) { + fprintf(stderr, "reading sysfs %ldM hugetlb config: %s\n", + hpagesz_mb, strerror(errno)); + return; + } + + if (atoi(buf) != GET_NR_HUGE_PAGES) { + fprintf(stderr, "could not confirm %ldM pages, got: '%s' expected %d\n", + hpagesz_mb, buf, GET_NR_HUGE_PAGES); + return; + } + + hugetlb_setup_ok = 1; +} + +void *malloc_pkey_hugetlb(long size, int prot, u16 pkey) +{ + void *ptr; + int flags = MAP_ANONYMOUS|MAP_PRIVATE|MAP_HUGETLB; + + if (!hugetlb_setup_ok) + return PTR_ERR_ENOTSUP; + + dprintf1("doing %s(%ld, %x, %x)\n", __func__, size, prot, pkey); + size = ALIGN_UP(size, HPAGE_SIZE * 2); + pkey_assert(pkey < NR_PKEYS); + ptr = mmap(NULL, size, PROT_NONE, flags, -1, 0); + pkey_assert(ptr != (void *)-1); + mprotect_pkey(ptr, size, prot, pkey); + + record_pkey_malloc(ptr, size, prot); + + dprintf1("mmap()'d hugetlbfs for pkey %d @ %p\n", pkey, ptr); + return ptr; +} + +void *malloc_pkey_mmap_dax(long size, int prot, u16 pkey) +{ + void *ptr; + int fd; + + dprintf1("doing %s(size=%ld, prot=0x%x, pkey=%d)\n", __func__, + size, prot, pkey); + pkey_assert(pkey < NR_PKEYS); + fd = open("/dax/foo", O_RDWR); + pkey_assert(fd >= 0); + + ptr = mmap(0, size, prot, MAP_SHARED, fd, 0); + pkey_assert(ptr != (void *)-1); + + mprotect_pkey(ptr, size, prot, pkey); + + record_pkey_malloc(ptr, size, prot); + + dprintf1("mmap()'d for pkey %d @ %p\n", pkey, ptr); + close(fd); + return ptr; +} + +void *(*pkey_malloc[])(long size, int prot, u16 pkey) = { + + malloc_pkey_with_mprotect, + malloc_pkey_with_mprotect_subpage, + malloc_pkey_anon_huge, + malloc_pkey_hugetlb +/* can not do direct with the pkey_mprotect() API: + malloc_pkey_mmap_direct, + malloc_pkey_mmap_dax, +*/ +}; + +void *malloc_pkey(long size, int prot, u16 pkey) +{ + void *ret; + static int malloc_type; + int nr_malloc_types = ARRAY_SIZE(pkey_malloc); + + pkey_assert(pkey < NR_PKEYS); + + while (1) { + pkey_assert(malloc_type < nr_malloc_types); + + ret = pkey_malloc[malloc_type](size, prot, pkey); + pkey_assert(ret != (void *)-1); + + malloc_type++; + if (malloc_type >= nr_malloc_types) + malloc_type = (random()%nr_malloc_types); + + /* try again if the malloc_type we tried is unsupported */ + if (ret == PTR_ERR_ENOTSUP) + continue; + + break; + } + + dprintf3("%s(%ld, prot=%x, pkey=%x) returning: %p\n", __func__, + size, prot, pkey, ret); + return ret; +} + +int last_pkey_faults; +#define UNKNOWN_PKEY -2 +void expected_pkey_fault(int pkey) +{ + dprintf2("%s(): last_pkey_faults: %d pkey_faults: %d\n", + __func__, last_pkey_faults, pkey_faults); + dprintf2("%s(%d): last_si_pkey: %d\n", __func__, pkey, last_si_pkey); + pkey_assert(last_pkey_faults + 1 == pkey_faults); + + /* + * For exec-only memory, we do not know the pkey in + * advance, so skip this check. + */ + if (pkey != UNKNOWN_PKEY) + pkey_assert(last_si_pkey == pkey); + +#if defined(__i386__) || defined(__x86_64__) /* arch */ + /* + * The signal handler shold have cleared out PKEY register to let the + * test program continue. We now have to restore it. + */ + if (__read_pkey_reg() != 0) +#else /* arch */ + if (__read_pkey_reg() != shadow_pkey_reg) +#endif /* arch */ + pkey_assert(0); + + __write_pkey_reg(shadow_pkey_reg); + dprintf1("%s() set pkey_reg=%016llx to restore state after signal " + "nuked it\n", __func__, shadow_pkey_reg); + last_pkey_faults = pkey_faults; + last_si_pkey = -1; +} + +#define do_not_expect_pkey_fault(msg) do { \ + if (last_pkey_faults != pkey_faults) \ + dprintf0("unexpected PKey fault: %s\n", msg); \ + pkey_assert(last_pkey_faults == pkey_faults); \ +} while (0) + +int test_fds[10] = { -1 }; +int nr_test_fds; +void __save_test_fd(int fd) +{ + pkey_assert(fd >= 0); + pkey_assert(nr_test_fds < ARRAY_SIZE(test_fds)); + test_fds[nr_test_fds] = fd; + nr_test_fds++; +} + +int get_test_read_fd(void) +{ + int test_fd = open("/etc/passwd", O_RDONLY); + __save_test_fd(test_fd); + return test_fd; +} + +void close_test_fds(void) +{ + int i; + + for (i = 0; i < nr_test_fds; i++) { + if (test_fds[i] < 0) + continue; + close(test_fds[i]); + test_fds[i] = -1; + } + nr_test_fds = 0; +} + +#define barrier() __asm__ __volatile__("": : :"memory") +__attribute__((noinline)) int read_ptr(int *ptr) +{ + /* + * Keep GCC from optimizing this away somehow + */ + barrier(); + return *ptr; +} + +void test_pkey_alloc_free_attach_pkey0(int *ptr, u16 pkey) +{ + int i, err; + int max_nr_pkey_allocs; + int alloced_pkeys[NR_PKEYS]; + int nr_alloced = 0; + long size; + + pkey_assert(pkey_last_malloc_record); + size = pkey_last_malloc_record->size; + /* + * This is a bit of a hack. But mprotect() requires + * huge-page-aligned sizes when operating on hugetlbfs. + * So, make sure that we use something that's a multiple + * of a huge page when we can. + */ + if (size >= HPAGE_SIZE) + size = HPAGE_SIZE; + + /* allocate every possible key and make sure key-0 never got allocated */ + max_nr_pkey_allocs = NR_PKEYS; + for (i = 0; i < max_nr_pkey_allocs; i++) { + int new_pkey = alloc_pkey(); + pkey_assert(new_pkey != 0); + + if (new_pkey < 0) + break; + alloced_pkeys[nr_alloced++] = new_pkey; + } + /* free all the allocated keys */ + for (i = 0; i < nr_alloced; i++) { + int free_ret; + + if (!alloced_pkeys[i]) + continue; + free_ret = sys_pkey_free(alloced_pkeys[i]); + pkey_assert(!free_ret); + } + + /* attach key-0 in various modes */ + err = sys_mprotect_pkey(ptr, size, PROT_READ, 0); + pkey_assert(!err); + err = sys_mprotect_pkey(ptr, size, PROT_WRITE, 0); + pkey_assert(!err); + err = sys_mprotect_pkey(ptr, size, PROT_EXEC, 0); + pkey_assert(!err); + err = sys_mprotect_pkey(ptr, size, PROT_READ|PROT_WRITE, 0); + pkey_assert(!err); + err = sys_mprotect_pkey(ptr, size, PROT_READ|PROT_WRITE|PROT_EXEC, 0); + pkey_assert(!err); +} + +void test_read_of_write_disabled_region(int *ptr, u16 pkey) +{ + int ptr_contents; + + dprintf1("disabling write access to PKEY[1], doing read\n"); + pkey_write_deny(pkey); + ptr_contents = read_ptr(ptr); + dprintf1("*ptr: %d\n", ptr_contents); + dprintf1("\n"); +} +void test_read_of_access_disabled_region(int *ptr, u16 pkey) +{ + int ptr_contents; + + dprintf1("disabling access to PKEY[%02d], doing read @ %p\n", pkey, ptr); + read_pkey_reg(); + pkey_access_deny(pkey); + ptr_contents = read_ptr(ptr); + dprintf1("*ptr: %d\n", ptr_contents); + expected_pkey_fault(pkey); +} + +void test_read_of_access_disabled_region_with_page_already_mapped(int *ptr, + u16 pkey) +{ + int ptr_contents; + + dprintf1("disabling access to PKEY[%02d], doing read @ %p\n", + pkey, ptr); + ptr_contents = read_ptr(ptr); + dprintf1("reading ptr before disabling the read : %d\n", + ptr_contents); + read_pkey_reg(); + pkey_access_deny(pkey); + ptr_contents = read_ptr(ptr); + dprintf1("*ptr: %d\n", ptr_contents); + expected_pkey_fault(pkey); +} + +void test_write_of_write_disabled_region_with_page_already_mapped(int *ptr, + u16 pkey) +{ + *ptr = __LINE__; + dprintf1("disabling write access; after accessing the page, " + "to PKEY[%02d], doing write\n", pkey); + pkey_write_deny(pkey); + *ptr = __LINE__; + expected_pkey_fault(pkey); +} + +void test_write_of_write_disabled_region(int *ptr, u16 pkey) +{ + dprintf1("disabling write access to PKEY[%02d], doing write\n", pkey); + pkey_write_deny(pkey); + *ptr = __LINE__; + expected_pkey_fault(pkey); +} +void test_write_of_access_disabled_region(int *ptr, u16 pkey) +{ + dprintf1("disabling access to PKEY[%02d], doing write\n", pkey); + pkey_access_deny(pkey); + *ptr = __LINE__; + expected_pkey_fault(pkey); +} + +void test_write_of_access_disabled_region_with_page_already_mapped(int *ptr, + u16 pkey) +{ + *ptr = __LINE__; + dprintf1("disabling access; after accessing the page, " + " to PKEY[%02d], doing write\n", pkey); + pkey_access_deny(pkey); + *ptr = __LINE__; + expected_pkey_fault(pkey); +} + +void test_kernel_write_of_access_disabled_region(int *ptr, u16 pkey) +{ + int ret; + int test_fd = get_test_read_fd(); + + dprintf1("disabling access to PKEY[%02d], " + "having kernel read() to buffer\n", pkey); + pkey_access_deny(pkey); + ret = read(test_fd, ptr, 1); + dprintf1("read ret: %d\n", ret); + pkey_assert(ret); +} +void test_kernel_write_of_write_disabled_region(int *ptr, u16 pkey) +{ + int ret; + int test_fd = get_test_read_fd(); + + pkey_write_deny(pkey); + ret = read(test_fd, ptr, 100); + dprintf1("read ret: %d\n", ret); + if (ret < 0 && (DEBUG_LEVEL > 0)) + perror("verbose read result (OK for this to be bad)"); + pkey_assert(ret); +} + +void test_kernel_gup_of_access_disabled_region(int *ptr, u16 pkey) +{ + int pipe_ret, vmsplice_ret; + struct iovec iov; + int pipe_fds[2]; + + pipe_ret = pipe(pipe_fds); + + pkey_assert(pipe_ret == 0); + dprintf1("disabling access to PKEY[%02d], " + "having kernel vmsplice from buffer\n", pkey); + pkey_access_deny(pkey); + iov.iov_base = ptr; + iov.iov_len = PAGE_SIZE; + vmsplice_ret = vmsplice(pipe_fds[1], &iov, 1, SPLICE_F_GIFT); + dprintf1("vmsplice() ret: %d\n", vmsplice_ret); + pkey_assert(vmsplice_ret == -1); + + close(pipe_fds[0]); + close(pipe_fds[1]); +} + +void test_kernel_gup_write_to_write_disabled_region(int *ptr, u16 pkey) +{ + int ignored = 0xdada; + int futex_ret; + int some_int = __LINE__; + + dprintf1("disabling write to PKEY[%02d], " + "doing futex gunk in buffer\n", pkey); + *ptr = some_int; + pkey_write_deny(pkey); + futex_ret = syscall(SYS_futex, ptr, FUTEX_WAIT, some_int-1, NULL, + &ignored, ignored); + if (DEBUG_LEVEL > 0) + perror("futex"); + dprintf1("futex() ret: %d\n", futex_ret); +} + +/* Assumes that all pkeys other than 'pkey' are unallocated */ +void test_pkey_syscalls_on_non_allocated_pkey(int *ptr, u16 pkey) +{ + int err; + int i; + + /* Note: 0 is the default pkey, so don't mess with it */ + for (i = 1; i < NR_PKEYS; i++) { + if (pkey == i) + continue; + + dprintf1("trying get/set/free to non-allocated pkey: %2d\n", i); + err = sys_pkey_free(i); + pkey_assert(err); + + err = sys_pkey_free(i); + pkey_assert(err); + + err = sys_mprotect_pkey(ptr, PAGE_SIZE, PROT_READ, i); + pkey_assert(err); + } +} + +/* Assumes that all pkeys other than 'pkey' are unallocated */ +void test_pkey_syscalls_bad_args(int *ptr, u16 pkey) +{ + int err; + int bad_pkey = NR_PKEYS+99; + + /* pass a known-invalid pkey in: */ + err = sys_mprotect_pkey(ptr, PAGE_SIZE, PROT_READ, bad_pkey); + pkey_assert(err); +} + +void become_child(void) +{ + pid_t forkret; + + forkret = fork(); + pkey_assert(forkret >= 0); + dprintf3("[%d] fork() ret: %d\n", getpid(), forkret); + + if (!forkret) { + /* in the child */ + return; + } + exit(0); +} + +/* Assumes that all pkeys other than 'pkey' are unallocated */ +void test_pkey_alloc_exhaust(int *ptr, u16 pkey) +{ + int err; + int allocated_pkeys[NR_PKEYS] = {0}; + int nr_allocated_pkeys = 0; + int i; + + for (i = 0; i < NR_PKEYS*3; i++) { + int new_pkey; + dprintf1("%s() alloc loop: %d\n", __func__, i); + new_pkey = alloc_pkey(); + dprintf4("%s()::%d, err: %d pkey_reg: 0x%016llx" + " shadow: 0x%016llx\n", + __func__, __LINE__, err, __read_pkey_reg(), + shadow_pkey_reg); + read_pkey_reg(); /* for shadow checking */ + dprintf2("%s() errno: %d ENOSPC: %d\n", __func__, errno, ENOSPC); + if ((new_pkey == -1) && (errno == ENOSPC)) { + dprintf2("%s() failed to allocate pkey after %d tries\n", + __func__, nr_allocated_pkeys); + } else { + /* + * Ensure the number of successes never + * exceeds the number of keys supported + * in the hardware. + */ + pkey_assert(nr_allocated_pkeys < NR_PKEYS); + allocated_pkeys[nr_allocated_pkeys++] = new_pkey; + } + + /* + * Make sure that allocation state is properly + * preserved across fork(). + */ + if (i == NR_PKEYS*2) + become_child(); + } + + dprintf3("%s()::%d\n", __func__, __LINE__); + + /* + * On x86: + * There are 16 pkeys supported in hardware. Three are + * allocated by the time we get here: + * 1. The default key (0) + * 2. One possibly consumed by an execute-only mapping. + * 3. One allocated by the test code and passed in via + * 'pkey' to this function. + * Ensure that we can allocate at least another 13 (16-3). + * + * On powerpc: + * There are either 5, 28, 29 or 32 pkeys supported in + * hardware depending on the page size (4K or 64K) and + * platform (powernv or powervm). Four are allocated by + * the time we get here. These include pkey-0, pkey-1, + * exec-only pkey and the one allocated by the test code. + * Ensure that we can allocate the remaining. + */ + pkey_assert(i >= (NR_PKEYS - get_arch_reserved_keys() - 1)); + + for (i = 0; i < nr_allocated_pkeys; i++) { + err = sys_pkey_free(allocated_pkeys[i]); + pkey_assert(!err); + read_pkey_reg(); /* for shadow checking */ + } +} + +void arch_force_pkey_reg_init(void) +{ +#if defined(__i386__) || defined(__x86_64__) /* arch */ + u64 *buf; + + /* + * All keys should be allocated and set to allow reads and + * writes, so the register should be all 0. If not, just + * skip the test. + */ + if (read_pkey_reg()) + return; + + /* + * Just allocate an absurd about of memory rather than + * doing the XSAVE size enumeration dance. + */ + buf = mmap(NULL, 1*MB, PROT_READ|PROT_WRITE, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + + /* These __builtins require compiling with -mxsave */ + + /* XSAVE to build a valid buffer: */ + __builtin_ia32_xsave(buf, XSTATE_PKEY); + /* Clear XSTATE_BV[PKRU]: */ + buf[XSTATE_BV_OFFSET/sizeof(u64)] &= ~XSTATE_PKEY; + /* XRSTOR will likely get PKRU back to the init state: */ + __builtin_ia32_xrstor(buf, XSTATE_PKEY); + + munmap(buf, 1*MB); +#endif +} + + +/* + * This is mostly useless on ppc for now. But it will not + * hurt anything and should give some better coverage as + * a long-running test that continually checks the pkey + * register. + */ +void test_pkey_init_state(int *ptr, u16 pkey) +{ + int err; + int allocated_pkeys[NR_PKEYS] = {0}; + int nr_allocated_pkeys = 0; + int i; + + for (i = 0; i < NR_PKEYS; i++) { + int new_pkey = alloc_pkey(); + + if (new_pkey < 0) + continue; + allocated_pkeys[nr_allocated_pkeys++] = new_pkey; + } + + dprintf3("%s()::%d\n", __func__, __LINE__); + + arch_force_pkey_reg_init(); + + /* + * Loop for a bit, hoping to get exercise the kernel + * context switch code. + */ + for (i = 0; i < 1000000; i++) + read_pkey_reg(); + + for (i = 0; i < nr_allocated_pkeys; i++) { + err = sys_pkey_free(allocated_pkeys[i]); + pkey_assert(!err); + read_pkey_reg(); /* for shadow checking */ + } +} + +/* + * pkey 0 is special. It is allocated by default, so you do not + * have to call pkey_alloc() to use it first. Make sure that it + * is usable. + */ +void test_mprotect_with_pkey_0(int *ptr, u16 pkey) +{ + long size; + int prot; + + assert(pkey_last_malloc_record); + size = pkey_last_malloc_record->size; + /* + * This is a bit of a hack. But mprotect() requires + * huge-page-aligned sizes when operating on hugetlbfs. + * So, make sure that we use something that's a multiple + * of a huge page when we can. + */ + if (size >= HPAGE_SIZE) + size = HPAGE_SIZE; + prot = pkey_last_malloc_record->prot; + + /* Use pkey 0 */ + mprotect_pkey(ptr, size, prot, 0); + + /* Make sure that we can set it back to the original pkey. */ + mprotect_pkey(ptr, size, prot, pkey); +} + +void test_ptrace_of_child(int *ptr, u16 pkey) +{ + __attribute__((__unused__)) int peek_result; + pid_t child_pid; + void *ignored = 0; + long ret; + int status; + /* + * This is the "control" for our little expermient. Make sure + * we can always access it when ptracing. + */ + int *plain_ptr_unaligned = malloc(HPAGE_SIZE); + int *plain_ptr = ALIGN_PTR_UP(plain_ptr_unaligned, PAGE_SIZE); + + /* + * Fork a child which is an exact copy of this process, of course. + * That means we can do all of our tests via ptrace() and then plain + * memory access and ensure they work differently. + */ + child_pid = fork_lazy_child(); + dprintf1("[%d] child pid: %d\n", getpid(), child_pid); + + ret = ptrace(PTRACE_ATTACH, child_pid, ignored, ignored); + if (ret) + perror("attach"); + dprintf1("[%d] attach ret: %ld %d\n", getpid(), ret, __LINE__); + pkey_assert(ret != -1); + ret = waitpid(child_pid, &status, WUNTRACED); + if ((ret != child_pid) || !(WIFSTOPPED(status))) { + fprintf(stderr, "weird waitpid result %ld stat %x\n", + ret, status); + pkey_assert(0); + } + dprintf2("waitpid ret: %ld\n", ret); + dprintf2("waitpid status: %d\n", status); + + pkey_access_deny(pkey); + pkey_write_deny(pkey); + + /* Write access, untested for now: + ret = ptrace(PTRACE_POKEDATA, child_pid, peek_at, data); + pkey_assert(ret != -1); + dprintf1("poke at %p: %ld\n", peek_at, ret); + */ + + /* + * Try to access the pkey-protected "ptr" via ptrace: + */ + ret = ptrace(PTRACE_PEEKDATA, child_pid, ptr, ignored); + /* expect it to work, without an error: */ + pkey_assert(ret != -1); + /* Now access from the current task, and expect an exception: */ + peek_result = read_ptr(ptr); + expected_pkey_fault(pkey); + + /* + * Try to access the NON-pkey-protected "plain_ptr" via ptrace: + */ + ret = ptrace(PTRACE_PEEKDATA, child_pid, plain_ptr, ignored); + /* expect it to work, without an error: */ + pkey_assert(ret != -1); + /* Now access from the current task, and expect NO exception: */ + peek_result = read_ptr(plain_ptr); + do_not_expect_pkey_fault("read plain pointer after ptrace"); + + ret = ptrace(PTRACE_DETACH, child_pid, ignored, 0); + pkey_assert(ret != -1); + + ret = kill(child_pid, SIGKILL); + pkey_assert(ret != -1); + + wait(&status); + + free(plain_ptr_unaligned); +} + +void *get_pointer_to_instructions(void) +{ + void *p1; + + p1 = ALIGN_PTR_UP(&lots_o_noops_around_write, PAGE_SIZE); + dprintf3("&lots_o_noops: %p\n", &lots_o_noops_around_write); + /* lots_o_noops_around_write should be page-aligned already */ + assert(p1 == &lots_o_noops_around_write); + + /* Point 'p1' at the *second* page of the function: */ + p1 += PAGE_SIZE; + + /* + * Try to ensure we fault this in on next touch to ensure + * we get an instruction fault as opposed to a data one + */ + madvise(p1, PAGE_SIZE, MADV_DONTNEED); + + return p1; +} + +void test_executing_on_unreadable_memory(int *ptr, u16 pkey) +{ + void *p1; + int scratch; + int ptr_contents; + int ret; + + p1 = get_pointer_to_instructions(); + lots_o_noops_around_write(&scratch); + ptr_contents = read_ptr(p1); + dprintf2("ptr (%p) contents@%d: %x\n", p1, __LINE__, ptr_contents); + + ret = mprotect_pkey(p1, PAGE_SIZE, PROT_EXEC, (u64)pkey); + pkey_assert(!ret); + pkey_access_deny(pkey); + + dprintf2("pkey_reg: %016llx\n", read_pkey_reg()); + + /* + * Make sure this is an *instruction* fault + */ + madvise(p1, PAGE_SIZE, MADV_DONTNEED); + lots_o_noops_around_write(&scratch); + do_not_expect_pkey_fault("executing on PROT_EXEC memory"); + expect_fault_on_read_execonly_key(p1, pkey); +} + +void test_implicit_mprotect_exec_only_memory(int *ptr, u16 pkey) +{ + void *p1; + int scratch; + int ptr_contents; + int ret; + + dprintf1("%s() start\n", __func__); + + p1 = get_pointer_to_instructions(); + lots_o_noops_around_write(&scratch); + ptr_contents = read_ptr(p1); + dprintf2("ptr (%p) contents@%d: %x\n", p1, __LINE__, ptr_contents); + + /* Use a *normal* mprotect(), not mprotect_pkey(): */ + ret = mprotect(p1, PAGE_SIZE, PROT_EXEC); + pkey_assert(!ret); + + /* + * Reset the shadow, assuming that the above mprotect() + * correctly changed PKRU, but to an unknown value since + * the actual allocated pkey is unknown. + */ + shadow_pkey_reg = __read_pkey_reg(); + + dprintf2("pkey_reg: %016llx\n", read_pkey_reg()); + + /* Make sure this is an *instruction* fault */ + madvise(p1, PAGE_SIZE, MADV_DONTNEED); + lots_o_noops_around_write(&scratch); + do_not_expect_pkey_fault("executing on PROT_EXEC memory"); + expect_fault_on_read_execonly_key(p1, UNKNOWN_PKEY); + + /* + * Put the memory back to non-PROT_EXEC. Should clear the + * exec-only pkey off the VMA and allow it to be readable + * again. Go to PROT_NONE first to check for a kernel bug + * that did not clear the pkey when doing PROT_NONE. + */ + ret = mprotect(p1, PAGE_SIZE, PROT_NONE); + pkey_assert(!ret); + + ret = mprotect(p1, PAGE_SIZE, PROT_READ|PROT_EXEC); + pkey_assert(!ret); + ptr_contents = read_ptr(p1); + do_not_expect_pkey_fault("plain read on recently PROT_EXEC area"); +} + +#if defined(__i386__) || defined(__x86_64__) +void test_ptrace_modifies_pkru(int *ptr, u16 pkey) +{ + u32 new_pkru; + pid_t child; + int status, ret; + int pkey_offset = pkey_reg_xstate_offset(); + size_t xsave_size = cpu_max_xsave_size(); + void *xsave; + u32 *pkey_register; + u64 *xstate_bv; + struct iovec iov; + + new_pkru = ~read_pkey_reg(); + /* Don't make PROT_EXEC mappings inaccessible */ + new_pkru &= ~3; + + child = fork(); + pkey_assert(child >= 0); + dprintf3("[%d] fork() ret: %d\n", getpid(), child); + if (!child) { + ptrace(PTRACE_TRACEME, 0, 0, 0); + /* Stop and allow the tracer to modify PKRU directly */ + raise(SIGSTOP); + + /* + * need __read_pkey_reg() version so we do not do shadow_pkey_reg + * checking + */ + if (__read_pkey_reg() != new_pkru) + exit(1); + + /* Stop and allow the tracer to clear XSTATE_BV for PKRU */ + raise(SIGSTOP); + + if (__read_pkey_reg() != 0) + exit(1); + + /* Stop and allow the tracer to examine PKRU */ + raise(SIGSTOP); + + exit(0); + } + + pkey_assert(child == waitpid(child, &status, 0)); + dprintf3("[%d] waitpid(%d) status: %x\n", getpid(), child, status); + pkey_assert(WIFSTOPPED(status) && WSTOPSIG(status) == SIGSTOP); + + xsave = (void *)malloc(xsave_size); + pkey_assert(xsave > 0); + + /* Modify the PKRU register directly */ + iov.iov_base = xsave; + iov.iov_len = xsave_size; + ret = ptrace(PTRACE_GETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + + pkey_register = (u32 *)(xsave + pkey_offset); + pkey_assert(*pkey_register == read_pkey_reg()); + + *pkey_register = new_pkru; + + ret = ptrace(PTRACE_SETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + + /* Test that the modification is visible in ptrace before any execution */ + memset(xsave, 0xCC, xsave_size); + ret = ptrace(PTRACE_GETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + pkey_assert(*pkey_register == new_pkru); + + /* Execute the tracee */ + ret = ptrace(PTRACE_CONT, child, 0, 0); + pkey_assert(ret == 0); + + /* Test that the tracee saw the PKRU value change */ + pkey_assert(child == waitpid(child, &status, 0)); + dprintf3("[%d] waitpid(%d) status: %x\n", getpid(), child, status); + pkey_assert(WIFSTOPPED(status) && WSTOPSIG(status) == SIGSTOP); + + /* Test that the modification is visible in ptrace after execution */ + memset(xsave, 0xCC, xsave_size); + ret = ptrace(PTRACE_GETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + pkey_assert(*pkey_register == new_pkru); + + /* Clear the PKRU bit from XSTATE_BV */ + xstate_bv = (u64 *)(xsave + 512); + *xstate_bv &= ~(1 << 9); + + ret = ptrace(PTRACE_SETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + + /* Test that the modification is visible in ptrace before any execution */ + memset(xsave, 0xCC, xsave_size); + ret = ptrace(PTRACE_GETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + pkey_assert(*pkey_register == 0); + + ret = ptrace(PTRACE_CONT, child, 0, 0); + pkey_assert(ret == 0); + + /* Test that the tracee saw the PKRU value go to 0 */ + pkey_assert(child == waitpid(child, &status, 0)); + dprintf3("[%d] waitpid(%d) status: %x\n", getpid(), child, status); + pkey_assert(WIFSTOPPED(status) && WSTOPSIG(status) == SIGSTOP); + + /* Test that the modification is visible in ptrace after execution */ + memset(xsave, 0xCC, xsave_size); + ret = ptrace(PTRACE_GETREGSET, child, (void *)NT_X86_XSTATE, &iov); + pkey_assert(ret == 0); + pkey_assert(*pkey_register == 0); + + ret = ptrace(PTRACE_CONT, child, 0, 0); + pkey_assert(ret == 0); + pkey_assert(child == waitpid(child, &status, 0)); + dprintf3("[%d] waitpid(%d) status: %x\n", getpid(), child, status); + pkey_assert(WIFEXITED(status)); + pkey_assert(WEXITSTATUS(status) == 0); + free(xsave); +} +#endif + +void test_mprotect_pkey_on_unsupported_cpu(int *ptr, u16 pkey) +{ + int size = PAGE_SIZE; + int sret; + + if (cpu_has_pkeys()) { + dprintf1("SKIP: %s: no CPU support\n", __func__); + return; + } + + sret = syscall(SYS_mprotect_key, ptr, size, PROT_READ, pkey); + pkey_assert(sret < 0); +} + +void (*pkey_tests[])(int *ptr, u16 pkey) = { + test_read_of_write_disabled_region, + test_read_of_access_disabled_region, + test_read_of_access_disabled_region_with_page_already_mapped, + test_write_of_write_disabled_region, + test_write_of_write_disabled_region_with_page_already_mapped, + test_write_of_access_disabled_region, + test_write_of_access_disabled_region_with_page_already_mapped, + test_kernel_write_of_access_disabled_region, + test_kernel_write_of_write_disabled_region, + test_kernel_gup_of_access_disabled_region, + test_kernel_gup_write_to_write_disabled_region, + test_executing_on_unreadable_memory, + test_implicit_mprotect_exec_only_memory, + test_mprotect_with_pkey_0, + test_ptrace_of_child, + test_pkey_init_state, + test_pkey_syscalls_on_non_allocated_pkey, + test_pkey_syscalls_bad_args, + test_pkey_alloc_exhaust, + test_pkey_alloc_free_attach_pkey0, +#if defined(__i386__) || defined(__x86_64__) + test_ptrace_modifies_pkru, +#endif +}; + +void run_tests_once(void) +{ + int *ptr; + int prot = PROT_READ|PROT_WRITE; + + for (test_nr = 0; test_nr < ARRAY_SIZE(pkey_tests); test_nr++) { + int pkey; + int orig_pkey_faults = pkey_faults; + + dprintf1("======================\n"); + dprintf1("test %d preparing...\n", test_nr); + + tracing_on(); + pkey = alloc_random_pkey(); + dprintf1("test %d starting with pkey: %d\n", test_nr, pkey); + ptr = malloc_pkey(PAGE_SIZE, prot, pkey); + dprintf1("test %d starting...\n", test_nr); + pkey_tests[test_nr](ptr, pkey); + dprintf1("freeing test memory: %p\n", ptr); + free_pkey_malloc(ptr); + sys_pkey_free(pkey); + + dprintf1("pkey_faults: %d\n", pkey_faults); + dprintf1("orig_pkey_faults: %d\n", orig_pkey_faults); + + tracing_off(); + close_test_fds(); + + printf("test %2d PASSED (iteration %d)\n", test_nr, iteration_nr); + dprintf1("======================\n\n"); + } + iteration_nr++; +} + +void pkey_setup_shadow(void) +{ + shadow_pkey_reg = __read_pkey_reg(); +} + +int main(void) +{ + int nr_iterations = 22; + int pkeys_supported = is_pkeys_supported(); + + srand((unsigned int)time(NULL)); + + setup_handlers(); + + printf("has pkeys: %d\n", pkeys_supported); + + if (!pkeys_supported) { + int size = PAGE_SIZE; + int *ptr; + + printf("running PKEY tests for unsupported CPU/OS\n"); + + ptr = mmap(NULL, size, PROT_NONE, MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + assert(ptr != (void *)-1); + test_mprotect_pkey_on_unsupported_cpu(ptr, 1); + exit(0); + } + + pkey_setup_shadow(); + printf("startup pkey_reg: %016llx\n", read_pkey_reg()); + setup_hugetlbfs(); + + while (nr_iterations-- > 0) + run_tests_once(); + + printf("done (all tests OK)\n"); + return 0; +} diff --git a/tools/testing/selftests/vm/run_vmtests.sh b/tools/testing/selftests/vm/run_vmtests.sh new file mode 100755 index 000000000..e780e76c2 --- /dev/null +++ b/tools/testing/selftests/vm/run_vmtests.sh @@ -0,0 +1,200 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +#please run as root + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +mnt=./huge +exitcode=0 + +#get huge pagesize and freepages from /proc/meminfo +while read -r name size unit; do + if [ "$name" = "HugePages_Free:" ]; then + freepgs="$size" + fi + if [ "$name" = "Hugepagesize:" ]; then + hpgsize_KB="$size" + fi +done < /proc/meminfo + +# Simple hugetlbfs tests have a hardcoded minimum requirement of +# huge pages totaling 256MB (262144KB) in size. The userfaultfd +# hugetlb test requires a minimum of 2 * nr_cpus huge pages. Take +# both of these requirements into account and attempt to increase +# number of huge pages available. +nr_cpus=$(nproc) +hpgsize_MB=$((hpgsize_KB / 1024)) +half_ufd_size_MB=$((((nr_cpus * hpgsize_MB + 127) / 128) * 128)) +needmem_KB=$((half_ufd_size_MB * 2 * 1024)) + +#set proper nr_hugepages +if [ -n "$freepgs" ] && [ -n "$hpgsize_KB" ]; then + nr_hugepgs=$(cat /proc/sys/vm/nr_hugepages) + needpgs=$((needmem_KB / hpgsize_KB)) + tries=2 + while [ "$tries" -gt 0 ] && [ "$freepgs" -lt "$needpgs" ]; do + lackpgs=$((needpgs - freepgs)) + echo 3 > /proc/sys/vm/drop_caches + if ! echo $((lackpgs + nr_hugepgs)) > /proc/sys/vm/nr_hugepages; then + echo "Please run this test as root" + exit $ksft_skip + fi + while read -r name size unit; do + if [ "$name" = "HugePages_Free:" ]; then + freepgs=$size + fi + done < /proc/meminfo + tries=$((tries - 1)) + done + if [ "$freepgs" -lt "$needpgs" ]; then + printf "Not enough huge pages available (%d < %d)\n" \ + "$freepgs" "$needpgs" + exit 1 + fi +else + echo "no hugetlbfs support in kernel?" + exit 1 +fi + +#filter 64bit architectures +ARCH64STR="arm64 ia64 mips64 parisc64 ppc64 ppc64le riscv64 s390x sh64 sparc64 x86_64" +if [ -z "$ARCH" ]; then + ARCH=$(uname -m 2>/dev/null | sed -e 's/aarch64.*/arm64/') +fi +VADDR64=0 +echo "$ARCH64STR" | grep "$ARCH" && VADDR64=1 + +# Usage: run_test [test binary] [arbitrary test arguments...] +run_test() { + local title="running $*" + local sep=$(echo -n "$title" | tr "[:graph:][:space:]" -) + printf "%s\n%s\n%s\n" "$sep" "$title" "$sep" + + "$@" + local ret=$? + if [ $ret -eq 0 ]; then + echo "[PASS]" + elif [ $ret -eq $ksft_skip ]; then + echo "[SKIP]" + exitcode=$ksft_skip + else + echo "[FAIL]" + exitcode=1 + fi +} + +mkdir "$mnt" +mount -t hugetlbfs none "$mnt" + +run_test ./hugepage-mmap + +shmmax=$(cat /proc/sys/kernel/shmmax) +shmall=$(cat /proc/sys/kernel/shmall) +echo 268435456 > /proc/sys/kernel/shmmax +echo 4194304 > /proc/sys/kernel/shmall +run_test ./hugepage-shm +echo "$shmmax" > /proc/sys/kernel/shmmax +echo "$shmall" > /proc/sys/kernel/shmall + +run_test ./map_hugetlb + +run_test ./hugepage-mremap "$mnt"/huge_mremap +rm -f "$mnt"/huge_mremap + +run_test ./hugepage-vmemmap + +run_test ./hugetlb-madvise "$mnt"/madvise-test +rm -f "$mnt"/madvise-test + +echo "NOTE: The above hugetlb tests provide minimal coverage. Use" +echo " https://github.com/libhugetlbfs/libhugetlbfs.git for" +echo " hugetlb regression testing." + +run_test ./map_fixed_noreplace + +# get_user_pages_fast() benchmark +run_test ./gup_test -u +# pin_user_pages_fast() benchmark +run_test ./gup_test -a +# Dump pages 0, 19, and 4096, using pin_user_pages: +run_test ./gup_test -ct -F 0x1 0 19 0x1000 + +uffd_mods=("" ":dev") +for mod in "${uffd_mods[@]}"; do + run_test ./userfaultfd anon${mod} 20 16 + # Hugetlb tests require source and destination huge pages. Pass in half + # the size ($half_ufd_size_MB), which is used for *each*. + run_test ./userfaultfd hugetlb${mod} "$half_ufd_size_MB" 32 + run_test ./userfaultfd hugetlb_shared${mod} "$half_ufd_size_MB" 32 "$mnt"/uffd-test + rm -f "$mnt"/uffd-test + run_test ./userfaultfd shmem${mod} 20 16 +done + +#cleanup +umount "$mnt" +rm -rf "$mnt" +echo "$nr_hugepgs" > /proc/sys/vm/nr_hugepages + +run_test ./compaction_test + +run_test sudo -u nobody ./on-fault-limit + +run_test ./map_populate + +run_test ./mlock-random-test + +run_test ./mlock2-tests + +run_test ./mrelease_test + +run_test ./mremap_test + +run_test ./thuge-gen + +if [ $VADDR64 -ne 0 ]; then + run_test ./virtual_address_range + + # virtual address 128TB switch test + run_test ./va_128TBswitch.sh +fi # VADDR64 + +# vmalloc stability smoke test +run_test ./test_vmalloc.sh smoke + +run_test ./mremap_dontunmap + +run_test ./test_hmm.sh smoke + +# MADV_POPULATE_READ and MADV_POPULATE_WRITE tests +run_test ./madv_populate + +run_test ./memfd_secret + +# KSM MADV_MERGEABLE test with 10 identical pages +run_test ./ksm_tests -M -p 10 +# KSM unmerge test +run_test ./ksm_tests -U +# KSM test with 10 zero pages and use_zero_pages = 0 +run_test ./ksm_tests -Z -p 10 -z 0 +# KSM test with 10 zero pages and use_zero_pages = 1 +run_test ./ksm_tests -Z -p 10 -z 1 +# KSM test with 2 NUMA nodes and merge_across_nodes = 1 +run_test ./ksm_tests -N -m 1 +# KSM test with 2 NUMA nodes and merge_across_nodes = 0 +run_test ./ksm_tests -N -m 0 + +# protection_keys tests +if [ -x ./protection_keys_32 ] +then + run_test ./protection_keys_32 +fi + +if [ -x ./protection_keys_64 ] +then + run_test ./protection_keys_64 +fi + +run_test ./soft-dirty + +exit $exitcode diff --git a/tools/testing/selftests/vm/settings b/tools/testing/selftests/vm/settings new file mode 100644 index 000000000..9abfc60e9 --- /dev/null +++ b/tools/testing/selftests/vm/settings @@ -0,0 +1 @@ +timeout=45 diff --git a/tools/testing/selftests/vm/soft-dirty.c b/tools/testing/selftests/vm/soft-dirty.c new file mode 100644 index 000000000..21d8830c5 --- /dev/null +++ b/tools/testing/selftests/vm/soft-dirty.c @@ -0,0 +1,210 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include +#include +#include "../kselftest.h" +#include "vm_util.h" + +#define PAGEMAP_FILE_PATH "/proc/self/pagemap" +#define TEST_ITERATIONS 10000 + +static void test_simple(int pagemap_fd, int pagesize) +{ + int i; + char *map; + + map = aligned_alloc(pagesize, pagesize); + if (!map) + ksft_exit_fail_msg("mmap failed\n"); + + clear_softdirty(); + + for (i = 0 ; i < TEST_ITERATIONS; i++) { + if (pagemap_is_softdirty(pagemap_fd, map) == 1) { + ksft_print_msg("dirty bit was 1, but should be 0 (i=%d)\n", i); + break; + } + + clear_softdirty(); + // Write something to the page to get the dirty bit enabled on the page + map[0]++; + + if (pagemap_is_softdirty(pagemap_fd, map) == 0) { + ksft_print_msg("dirty bit was 0, but should be 1 (i=%d)\n", i); + break; + } + + clear_softdirty(); + } + free(map); + + ksft_test_result(i == TEST_ITERATIONS, "Test %s\n", __func__); +} + +static void test_vma_reuse(int pagemap_fd, int pagesize) +{ + char *map, *map2; + + map = mmap(NULL, pagesize, (PROT_READ | PROT_WRITE), (MAP_PRIVATE | MAP_ANON), -1, 0); + if (map == MAP_FAILED) + ksft_exit_fail_msg("mmap failed"); + + // The kernel always marks new regions as soft dirty + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 1, + "Test %s dirty bit of allocated page\n", __func__); + + clear_softdirty(); + munmap(map, pagesize); + + map2 = mmap(NULL, pagesize, (PROT_READ | PROT_WRITE), (MAP_PRIVATE | MAP_ANON), -1, 0); + if (map2 == MAP_FAILED) + ksft_exit_fail_msg("mmap failed"); + + // Dirty bit is set for new regions even if they are reused + if (map == map2) + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map2) == 1, + "Test %s dirty bit of reused address page\n", __func__); + else + ksft_test_result_skip("Test %s dirty bit of reused address page\n", __func__); + + munmap(map2, pagesize); +} + +static void test_hugepage(int pagemap_fd, int pagesize) +{ + char *map; + int i, ret; + size_t hpage_len = read_pmd_pagesize(); + + map = memalign(hpage_len, hpage_len); + if (!map) + ksft_exit_fail_msg("memalign failed\n"); + + ret = madvise(map, hpage_len, MADV_HUGEPAGE); + if (ret) + ksft_exit_fail_msg("madvise failed %d\n", ret); + + for (i = 0; i < hpage_len; i++) + map[i] = (char)i; + + if (check_huge_anon(map, 1, hpage_len)) { + ksft_test_result_pass("Test %s huge page allocation\n", __func__); + + clear_softdirty(); + for (i = 0 ; i < TEST_ITERATIONS ; i++) { + if (pagemap_is_softdirty(pagemap_fd, map) == 1) { + ksft_print_msg("dirty bit was 1, but should be 0 (i=%d)\n", i); + break; + } + + clear_softdirty(); + // Write something to the page to get the dirty bit enabled on the page + map[0]++; + + if (pagemap_is_softdirty(pagemap_fd, map) == 0) { + ksft_print_msg("dirty bit was 0, but should be 1 (i=%d)\n", i); + break; + } + clear_softdirty(); + } + + ksft_test_result(i == TEST_ITERATIONS, "Test %s huge page dirty bit\n", __func__); + } else { + // hugepage allocation failed. skip these tests + ksft_test_result_skip("Test %s huge page allocation\n", __func__); + ksft_test_result_skip("Test %s huge page dirty bit\n", __func__); + } + free(map); +} + +static void test_mprotect(int pagemap_fd, int pagesize, bool anon) +{ + const char *type[] = {"file", "anon"}; + const char *fname = "./soft-dirty-test-file"; + int test_fd; + char *map; + + if (anon) { + map = mmap(NULL, pagesize, PROT_READ|PROT_WRITE, + MAP_ANONYMOUS|MAP_PRIVATE, -1, 0); + if (!map) + ksft_exit_fail_msg("anon mmap failed\n"); + } else { + test_fd = open(fname, O_RDWR | O_CREAT); + if (test_fd < 0) { + ksft_test_result_skip("Test %s open() file failed\n", __func__); + return; + } + unlink(fname); + ftruncate(test_fd, pagesize); + map = mmap(NULL, pagesize, PROT_READ|PROT_WRITE, + MAP_SHARED, test_fd, 0); + if (!map) + ksft_exit_fail_msg("file mmap failed\n"); + } + + *map = 1; + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 1, + "Test %s-%s dirty bit of new written page\n", + __func__, type[anon]); + clear_softdirty(); + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 0, + "Test %s-%s soft-dirty clear after clear_refs\n", + __func__, type[anon]); + mprotect(map, pagesize, PROT_READ); + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 0, + "Test %s-%s soft-dirty clear after marking RO\n", + __func__, type[anon]); + mprotect(map, pagesize, PROT_READ|PROT_WRITE); + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 0, + "Test %s-%s soft-dirty clear after marking RW\n", + __func__, type[anon]); + *map = 2; + ksft_test_result(pagemap_is_softdirty(pagemap_fd, map) == 1, + "Test %s-%s soft-dirty after rewritten\n", + __func__, type[anon]); + + munmap(map, pagesize); + + if (!anon) + close(test_fd); +} + +static void test_mprotect_anon(int pagemap_fd, int pagesize) +{ + test_mprotect(pagemap_fd, pagesize, true); +} + +static void test_mprotect_file(int pagemap_fd, int pagesize) +{ + test_mprotect(pagemap_fd, pagesize, false); +} + +int main(int argc, char **argv) +{ + int pagemap_fd; + int pagesize; + + ksft_print_header(); + ksft_set_plan(15); + + pagemap_fd = open(PAGEMAP_FILE_PATH, O_RDONLY); + if (pagemap_fd < 0) + ksft_exit_fail_msg("Failed to open %s\n", PAGEMAP_FILE_PATH); + + pagesize = getpagesize(); + + test_simple(pagemap_fd, pagesize); + test_vma_reuse(pagemap_fd, pagesize); + test_hugepage(pagemap_fd, pagesize); + test_mprotect_anon(pagemap_fd, pagesize); + test_mprotect_file(pagemap_fd, pagesize); + + close(pagemap_fd); + + return ksft_exit_pass(); +} diff --git a/tools/testing/selftests/vm/split_huge_page_test.c b/tools/testing/selftests/vm/split_huge_page_test.c new file mode 100644 index 000000000..76e1c36dd --- /dev/null +++ b/tools/testing/selftests/vm/split_huge_page_test.c @@ -0,0 +1,309 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * A test of splitting PMD THPs and PTE-mapped THPs from a specified virtual + * address range in a process via /split_huge_pages interface. + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "vm_util.h" + +uint64_t pagesize; +unsigned int pageshift; +uint64_t pmd_pagesize; + +#define SPLIT_DEBUGFS "/sys/kernel/debug/split_huge_pages" +#define INPUT_MAX 80 + +#define PID_FMT "%d,0x%lx,0x%lx" +#define PATH_FMT "%s,0x%lx,0x%lx" + +#define PFN_MASK ((1UL<<55)-1) +#define KPF_THP (1UL<<22) + +int is_backed_by_thp(char *vaddr, int pagemap_file, int kpageflags_file) +{ + uint64_t paddr; + uint64_t page_flags; + + if (pagemap_file) { + pread(pagemap_file, &paddr, sizeof(paddr), + ((long)vaddr >> pageshift) * sizeof(paddr)); + + if (kpageflags_file) { + pread(kpageflags_file, &page_flags, sizeof(page_flags), + (paddr & PFN_MASK) * sizeof(page_flags)); + + return !!(page_flags & KPF_THP); + } + } + return 0; +} + +static int write_file(const char *path, const char *buf, size_t buflen) +{ + int fd; + ssize_t numwritten; + + fd = open(path, O_WRONLY); + if (fd == -1) + return 0; + + numwritten = write(fd, buf, buflen - 1); + close(fd); + if (numwritten < 1) + return 0; + + return (unsigned int) numwritten; +} + +static void write_debugfs(const char *fmt, ...) +{ + char input[INPUT_MAX]; + int ret; + va_list argp; + + va_start(argp, fmt); + ret = vsnprintf(input, INPUT_MAX, fmt, argp); + va_end(argp); + + if (ret >= INPUT_MAX) { + printf("%s: Debugfs input is too long\n", __func__); + exit(EXIT_FAILURE); + } + + if (!write_file(SPLIT_DEBUGFS, input, ret + 1)) { + perror(SPLIT_DEBUGFS); + exit(EXIT_FAILURE); + } +} + +void split_pmd_thp(void) +{ + char *one_page; + size_t len = 4 * pmd_pagesize; + size_t i; + + one_page = memalign(pmd_pagesize, len); + + if (!one_page) { + printf("Fail to allocate memory\n"); + exit(EXIT_FAILURE); + } + + madvise(one_page, len, MADV_HUGEPAGE); + + for (i = 0; i < len; i++) + one_page[i] = (char)i; + + if (!check_huge_anon(one_page, 1, pmd_pagesize)) { + printf("No THP is allocated\n"); + exit(EXIT_FAILURE); + } + + /* split all THPs */ + write_debugfs(PID_FMT, getpid(), (uint64_t)one_page, + (uint64_t)one_page + len); + + for (i = 0; i < len; i++) + if (one_page[i] != (char)i) { + printf("%ld byte corrupted\n", i); + exit(EXIT_FAILURE); + } + + + if (check_huge_anon(one_page, 0, pmd_pagesize)) { + printf("Still AnonHugePages not split\n"); + exit(EXIT_FAILURE); + } + + printf("Split huge pages successful\n"); + free(one_page); +} + +void split_pte_mapped_thp(void) +{ + char *one_page, *pte_mapped, *pte_mapped2; + size_t len = 4 * pmd_pagesize; + uint64_t thp_size; + size_t i; + const char *pagemap_template = "/proc/%d/pagemap"; + const char *kpageflags_proc = "/proc/kpageflags"; + char pagemap_proc[255]; + int pagemap_fd; + int kpageflags_fd; + + if (snprintf(pagemap_proc, 255, pagemap_template, getpid()) < 0) { + perror("get pagemap proc error"); + exit(EXIT_FAILURE); + } + pagemap_fd = open(pagemap_proc, O_RDONLY); + + if (pagemap_fd == -1) { + perror("read pagemap:"); + exit(EXIT_FAILURE); + } + + kpageflags_fd = open(kpageflags_proc, O_RDONLY); + + if (kpageflags_fd == -1) { + perror("read kpageflags:"); + exit(EXIT_FAILURE); + } + + one_page = mmap((void *)(1UL << 30), len, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); + + madvise(one_page, len, MADV_HUGEPAGE); + + for (i = 0; i < len; i++) + one_page[i] = (char)i; + + if (!check_huge_anon(one_page, 1, pmd_pagesize)) { + printf("No THP is allocated\n"); + exit(EXIT_FAILURE); + } + + /* remap the first pagesize of first THP */ + pte_mapped = mremap(one_page, pagesize, pagesize, MREMAP_MAYMOVE); + + /* remap the Nth pagesize of Nth THP */ + for (i = 1; i < 4; i++) { + pte_mapped2 = mremap(one_page + pmd_pagesize * i + pagesize * i, + pagesize, pagesize, + MREMAP_MAYMOVE|MREMAP_FIXED, + pte_mapped + pagesize * i); + if (pte_mapped2 == (char *)-1) { + perror("mremap failed"); + exit(EXIT_FAILURE); + } + } + + /* smap does not show THPs after mremap, use kpageflags instead */ + thp_size = 0; + for (i = 0; i < pagesize * 4; i++) + if (i % pagesize == 0 && + is_backed_by_thp(&pte_mapped[i], pagemap_fd, kpageflags_fd)) + thp_size++; + + if (thp_size != 4) { + printf("Some THPs are missing during mremap\n"); + exit(EXIT_FAILURE); + } + + /* split all remapped THPs */ + write_debugfs(PID_FMT, getpid(), (uint64_t)pte_mapped, + (uint64_t)pte_mapped + pagesize * 4); + + /* smap does not show THPs after mremap, use kpageflags instead */ + thp_size = 0; + for (i = 0; i < pagesize * 4; i++) { + if (pte_mapped[i] != (char)i) { + printf("%ld byte corrupted\n", i); + exit(EXIT_FAILURE); + } + if (i % pagesize == 0 && + is_backed_by_thp(&pte_mapped[i], pagemap_fd, kpageflags_fd)) + thp_size++; + } + + if (thp_size) { + printf("Still %ld THPs not split\n", thp_size); + exit(EXIT_FAILURE); + } + + printf("Split PTE-mapped huge pages successful\n"); + munmap(one_page, len); + close(pagemap_fd); + close(kpageflags_fd); +} + +void split_file_backed_thp(void) +{ + int status; + int fd; + ssize_t num_written; + char tmpfs_template[] = "/tmp/thp_split_XXXXXX"; + const char *tmpfs_loc = mkdtemp(tmpfs_template); + char testfile[INPUT_MAX]; + uint64_t pgoff_start = 0, pgoff_end = 1024; + + printf("Please enable pr_debug in split_huge_pages_in_file() if you need more info.\n"); + + status = mount("tmpfs", tmpfs_loc, "tmpfs", 0, "huge=always,size=4m"); + + if (status) { + printf("Unable to create a tmpfs for testing\n"); + exit(EXIT_FAILURE); + } + + status = snprintf(testfile, INPUT_MAX, "%s/thp_file", tmpfs_loc); + if (status >= INPUT_MAX) { + printf("Fail to create file-backed THP split testing file\n"); + goto cleanup; + } + + fd = open(testfile, O_CREAT|O_WRONLY); + if (fd == -1) { + perror("Cannot open testing file\n"); + goto cleanup; + } + + /* write something to the file, so a file-backed THP can be allocated */ + num_written = write(fd, tmpfs_loc, strlen(tmpfs_loc) + 1); + close(fd); + + if (num_written < 1) { + printf("Fail to write data to testing file\n"); + goto cleanup; + } + + /* split the file-backed THP */ + write_debugfs(PATH_FMT, testfile, pgoff_start, pgoff_end); + + status = unlink(testfile); + if (status) + perror("Cannot remove testing file\n"); + +cleanup: + status = umount(tmpfs_loc); + if (status) { + printf("Unable to umount %s\n", tmpfs_loc); + exit(EXIT_FAILURE); + } + status = rmdir(tmpfs_loc); + if (status) { + perror("cannot remove tmp dir"); + exit(EXIT_FAILURE); + } + + printf("file-backed THP split test done, please check dmesg for more information\n"); +} + +int main(int argc, char **argv) +{ + if (geteuid() != 0) { + printf("Please run the benchmark as root\n"); + exit(EXIT_FAILURE); + } + + pagesize = getpagesize(); + pageshift = ffs(pagesize) - 1; + pmd_pagesize = read_pmd_pagesize(); + + split_pmd_thp(); + split_pte_mapped_thp(); + split_file_backed_thp(); + + return 0; +} diff --git a/tools/testing/selftests/vm/test_hmm.sh b/tools/testing/selftests/vm/test_hmm.sh new file mode 100755 index 000000000..46e19b5d6 --- /dev/null +++ b/tools/testing/selftests/vm/test_hmm.sh @@ -0,0 +1,105 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (C) 2018 Uladzislau Rezki (Sony) +# +# This is a test script for the kernel test driver to analyse vmalloc +# allocator. Therefore it is just a kernel module loader. You can specify +# and pass different parameters in order to: +# a) analyse performance of vmalloc allocations; +# b) stressing and stability check of vmalloc subsystem. + +TEST_NAME="test_hmm" +DRIVER="test_hmm" + +# 1 if fails +exitcode=1 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +check_test_requirements() +{ + uid=$(id -u) + if [ $uid -ne 0 ]; then + echo "$0: Must be run as root" + exit $ksft_skip + fi + + if ! which modprobe > /dev/null 2>&1; then + echo "$0: You need modprobe installed" + exit $ksft_skip + fi + + if ! modinfo $DRIVER > /dev/null 2>&1; then + echo "$0: You must have the following enabled in your kernel:" + echo "CONFIG_TEST_HMM=m" + exit $ksft_skip + fi +} + +load_driver() +{ + if [ $# -eq 0 ]; then + modprobe $DRIVER > /dev/null 2>&1 + else + if [ $# -eq 2 ]; then + modprobe $DRIVER spm_addr_dev0=$1 spm_addr_dev1=$2 + > /dev/null 2>&1 + else + echo "Missing module parameters. Make sure pass"\ + "spm_addr_dev0 and spm_addr_dev1" + usage + fi + fi +} + +unload_driver() +{ + modprobe -r $DRIVER > /dev/null 2>&1 +} + +run_smoke() +{ + echo "Running smoke test. Note, this test provides basic coverage." + + load_driver $1 $2 + $(dirname "${BASH_SOURCE[0]}")/hmm-tests + unload_driver +} + +usage() +{ + echo -n "Usage: $0" + echo + echo "Example usage:" + echo + echo "# Shows help message" + echo "./${TEST_NAME}.sh" + echo + echo "# Smoke testing" + echo "./${TEST_NAME}.sh smoke" + echo + echo "# Smoke testing with SPM enabled" + echo "./${TEST_NAME}.sh smoke " + echo + exit 0 +} + +function run_test() +{ + if [ $# -eq 0 ]; then + usage + else + if [ "$1" = "smoke" ]; then + run_smoke $2 $3 + else + usage + fi + fi +} + +check_test_requirements +run_test $@ + +exit 0 diff --git a/tools/testing/selftests/vm/test_vmalloc.sh b/tools/testing/selftests/vm/test_vmalloc.sh new file mode 100755 index 000000000..d73b84673 --- /dev/null +++ b/tools/testing/selftests/vm/test_vmalloc.sh @@ -0,0 +1,177 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (C) 2018 Uladzislau Rezki (Sony) +# +# This is a test script for the kernel test driver to analyse vmalloc +# allocator. Therefore it is just a kernel module loader. You can specify +# and pass different parameters in order to: +# a) analyse performance of vmalloc allocations; +# b) stressing and stability check of vmalloc subsystem. + +TEST_NAME="vmalloc" +DRIVER="test_${TEST_NAME}" +NUM_CPUS=`grep -c ^processor /proc/cpuinfo` + +# 1 if fails +exitcode=1 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +# +# Static templates for performance, stressing and smoke tests. +# Also it is possible to pass any supported parameters manualy. +# +PERF_PARAM="sequential_test_order=1 test_repeat_count=3" +SMOKE_PARAM="test_loop_count=10000 test_repeat_count=10" +STRESS_PARAM="nr_threads=$NUM_CPUS test_repeat_count=20" + +check_test_requirements() +{ + uid=$(id -u) + if [ $uid -ne 0 ]; then + echo "$0: Must be run as root" + exit $ksft_skip + fi + + if ! which modprobe > /dev/null 2>&1; then + echo "$0: You need modprobe installed" + exit $ksft_skip + fi + + if ! modinfo $DRIVER > /dev/null 2>&1; then + echo "$0: You must have the following enabled in your kernel:" + echo "CONFIG_TEST_VMALLOC=m" + exit $ksft_skip + fi +} + +run_perfformance_check() +{ + echo "Run performance tests to evaluate how fast vmalloc allocation is." + echo "It runs all test cases on one single CPU with sequential order." + + modprobe $DRIVER $PERF_PARAM > /dev/null 2>&1 + echo "Done." + echo "Ccheck the kernel message buffer to see the summary." +} + +run_stability_check() +{ + echo "Run stability tests. In order to stress vmalloc subsystem all" + echo "available test cases are run by NUM_CPUS workers simultaneously." + echo "It will take time, so be patient." + + modprobe $DRIVER $STRESS_PARAM > /dev/null 2>&1 + echo "Done." + echo "Check the kernel ring buffer to see the summary." +} + +run_smoke_check() +{ + echo "Run smoke test. Note, this test provides basic coverage." + echo "Please check $0 output how it can be used" + echo "for deep performance analysis as well as stress testing." + + modprobe $DRIVER $SMOKE_PARAM > /dev/null 2>&1 + echo "Done." + echo "Check the kernel ring buffer to see the summary." +} + +usage() +{ + echo -n "Usage: $0 [ performance ] | [ stress ] | | [ smoke ] | " + echo "manual parameters" + echo + echo "Valid tests and parameters:" + echo + modinfo $DRIVER + echo + echo "Example usage:" + echo + echo "# Shows help message" + echo "./${DRIVER}.sh" + echo + echo "# Runs 1 test(id_1), repeats it 5 times by NUM_CPUS workers" + echo "./${DRIVER}.sh nr_threads=$NUM_CPUS run_test_mask=1 test_repeat_count=5" + echo + echo -n "# Runs 4 tests(id_1|id_2|id_4|id_16) on one CPU with " + echo "sequential order" + echo -n "./${DRIVER}.sh sequential_test_order=1 " + echo "run_test_mask=23" + echo + echo -n "# Runs all tests by NUM_CPUS workers, shuffled order, repeats " + echo "20 times" + echo "./${DRIVER}.sh nr_threads=$NUM_CPUS test_repeat_count=20" + echo + echo "# Performance analysis" + echo "./${DRIVER}.sh performance" + echo + echo "# Stress testing" + echo "./${DRIVER}.sh stress" + echo + exit 0 +} + +function validate_passed_args() +{ + VALID_ARGS=`modinfo $DRIVER | awk '/parm:/ {print $2}' | sed 's/:.*//'` + + # + # Something has been passed, check it. + # + for passed_arg in $@; do + key=${passed_arg//=*/} + val="${passed_arg:$((${#key}+1))}" + valid=0 + + for valid_arg in $VALID_ARGS; do + if [[ $key = $valid_arg ]] && [[ $val -gt 0 ]]; then + valid=1 + break + fi + done + + if [[ $valid -ne 1 ]]; then + echo "Error: key or value is not correct: ${key} $val" + exit $exitcode + fi + done +} + +function run_manual_check() +{ + # + # Validate passed parameters. If there is wrong one, + # the script exists and does not execute further. + # + validate_passed_args $@ + + echo "Run the test with following parameters: $@" + modprobe $DRIVER $@ > /dev/null 2>&1 + echo "Done." + echo "Check the kernel ring buffer to see the summary." +} + +function run_test() +{ + if [ $# -eq 0 ]; then + usage + else + if [[ "$1" = "performance" ]]; then + run_perfformance_check + elif [[ "$1" = "stress" ]]; then + run_stability_check + elif [[ "$1" = "smoke" ]]; then + run_smoke_check + else + run_manual_check $@ + fi + fi +} + +check_test_requirements +run_test $@ + +exit 0 diff --git a/tools/testing/selftests/vm/thuge-gen.c b/tools/testing/selftests/vm/thuge-gen.c new file mode 100644 index 000000000..361ef7192 --- /dev/null +++ b/tools/testing/selftests/vm/thuge-gen.c @@ -0,0 +1,257 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Test selecting other page sizes for mmap/shmget. + + Before running this huge pages for each huge page size must have been + reserved. + For large pages beyond MAX_ORDER (like 1GB on x86) boot options must be used. + Also shmmax must be increased. + And you need to run as root to work around some weird permissions in shm. + And nothing using huge pages should run in parallel. + When the program aborts you may need to clean up the shm segments with + ipcrm -m by hand, like this + sudo ipcs | awk '$1 == "0x00000000" {print $2}' | xargs -n1 sudo ipcrm -m + (warning this will remove all if someone else uses them) */ + +#define _GNU_SOURCE 1 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define err(x) perror(x), exit(1) + +#define MAP_HUGE_2MB (21 << MAP_HUGE_SHIFT) +#define MAP_HUGE_1GB (30 << MAP_HUGE_SHIFT) +#define MAP_HUGE_SHIFT 26 +#define MAP_HUGE_MASK 0x3f +#if !defined(MAP_HUGETLB) +#define MAP_HUGETLB 0x40000 +#endif + +#define SHM_HUGETLB 04000 /* segment will use huge TLB pages */ +#define SHM_HUGE_SHIFT 26 +#define SHM_HUGE_MASK 0x3f +#define SHM_HUGE_2MB (21 << SHM_HUGE_SHIFT) +#define SHM_HUGE_1GB (30 << SHM_HUGE_SHIFT) + +#define NUM_PAGESIZES 5 + +#define NUM_PAGES 4 + +#define Dprintf(fmt...) // printf(fmt) + +unsigned long page_sizes[NUM_PAGESIZES]; +int num_page_sizes; + +int ilog2(unsigned long v) +{ + int l = 0; + while ((1UL << l) < v) + l++; + return l; +} + +void find_pagesizes(void) +{ + glob_t g; + int i; + glob("/sys/kernel/mm/hugepages/hugepages-*kB", 0, NULL, &g); + assert(g.gl_pathc <= NUM_PAGESIZES); + for (i = 0; i < g.gl_pathc; i++) { + sscanf(g.gl_pathv[i], "/sys/kernel/mm/hugepages/hugepages-%lukB", + &page_sizes[i]); + page_sizes[i] <<= 10; + printf("Found %luMB\n", page_sizes[i] >> 20); + } + num_page_sizes = g.gl_pathc; + globfree(&g); +} + +unsigned long default_huge_page_size(void) +{ + unsigned long hps = 0; + char *line = NULL; + size_t linelen = 0; + FILE *f = fopen("/proc/meminfo", "r"); + if (!f) + return 0; + while (getline(&line, &linelen, f) > 0) { + if (sscanf(line, "Hugepagesize: %lu kB", &hps) == 1) { + hps <<= 10; + break; + } + } + free(line); + return hps; +} + +void show(unsigned long ps) +{ + char buf[100]; + if (ps == getpagesize()) + return; + printf("%luMB: ", ps >> 20); + fflush(stdout); + snprintf(buf, sizeof buf, + "cat /sys/kernel/mm/hugepages/hugepages-%lukB/free_hugepages", + ps >> 10); + system(buf); +} + +unsigned long read_sysfs(int warn, char *fmt, ...) +{ + char *line = NULL; + size_t linelen = 0; + char buf[100]; + FILE *f; + va_list ap; + unsigned long val = 0; + + va_start(ap, fmt); + vsnprintf(buf, sizeof buf, fmt, ap); + va_end(ap); + + f = fopen(buf, "r"); + if (!f) { + if (warn) + printf("missing %s\n", buf); + return 0; + } + if (getline(&line, &linelen, f) > 0) { + sscanf(line, "%lu", &val); + } + fclose(f); + free(line); + return val; +} + +unsigned long read_free(unsigned long ps) +{ + return read_sysfs(ps != getpagesize(), + "/sys/kernel/mm/hugepages/hugepages-%lukB/free_hugepages", + ps >> 10); +} + +void test_mmap(unsigned long size, unsigned flags) +{ + char *map; + unsigned long before, after; + int err; + + before = read_free(size); + map = mmap(NULL, size*NUM_PAGES, PROT_READ|PROT_WRITE, + MAP_PRIVATE|MAP_ANONYMOUS|MAP_HUGETLB|flags, -1, 0); + + if (map == (char *)-1) err("mmap"); + memset(map, 0xff, size*NUM_PAGES); + after = read_free(size); + Dprintf("before %lu after %lu diff %ld size %lu\n", + before, after, before - after, size); + assert(size == getpagesize() || (before - after) == NUM_PAGES); + show(size); + err = munmap(map, size); + assert(!err); +} + +void test_shmget(unsigned long size, unsigned flags) +{ + int id; + unsigned long before, after; + int err; + + before = read_free(size); + id = shmget(IPC_PRIVATE, size * NUM_PAGES, IPC_CREAT|0600|flags); + if (id < 0) err("shmget"); + + struct shm_info i; + if (shmctl(id, SHM_INFO, (void *)&i) < 0) err("shmctl"); + Dprintf("alloc %lu res %lu\n", i.shm_tot, i.shm_rss); + + + Dprintf("id %d\n", id); + char *map = shmat(id, NULL, 0600); + if (map == (char*)-1) err("shmat"); + + shmctl(id, IPC_RMID, NULL); + + memset(map, 0xff, size*NUM_PAGES); + after = read_free(size); + + Dprintf("before %lu after %lu diff %ld size %lu\n", + before, after, before - after, size); + assert(size == getpagesize() || (before - after) == NUM_PAGES); + show(size); + err = shmdt(map); + assert(!err); +} + +void sanity_checks(void) +{ + int i; + unsigned long largest = getpagesize(); + + for (i = 0; i < num_page_sizes; i++) { + if (page_sizes[i] > largest) + largest = page_sizes[i]; + + if (read_free(page_sizes[i]) < NUM_PAGES) { + printf("Not enough huge pages for page size %lu MB, need %u\n", + page_sizes[i] >> 20, + NUM_PAGES); + exit(0); + } + } + + if (read_sysfs(0, "/proc/sys/kernel/shmmax") < NUM_PAGES * largest) { + printf("Please do echo %lu > /proc/sys/kernel/shmmax", largest * NUM_PAGES); + exit(0); + } + +#if defined(__x86_64__) + if (largest != 1U<<30) { + printf("No GB pages available on x86-64\n" + "Please boot with hugepagesz=1G hugepages=%d\n", NUM_PAGES); + exit(0); + } +#endif +} + +int main(void) +{ + int i; + unsigned default_hps = default_huge_page_size(); + + find_pagesizes(); + + sanity_checks(); + + for (i = 0; i < num_page_sizes; i++) { + unsigned long ps = page_sizes[i]; + int arg = ilog2(ps) << MAP_HUGE_SHIFT; + printf("Testing %luMB mmap with shift %x\n", ps >> 20, arg); + test_mmap(ps, MAP_HUGETLB | arg); + } + printf("Testing default huge mmap\n"); + test_mmap(default_hps, SHM_HUGETLB); + + puts("Testing non-huge shmget"); + test_shmget(getpagesize(), 0); + + for (i = 0; i < num_page_sizes; i++) { + unsigned long ps = page_sizes[i]; + int arg = ilog2(ps) << SHM_HUGE_SHIFT; + printf("Testing %luMB shmget with shift %x\n", ps >> 20, arg); + test_shmget(ps, SHM_HUGETLB | arg); + } + puts("default huge shmget"); + test_shmget(default_hps, SHM_HUGETLB); + + return 0; +} diff --git a/tools/testing/selftests/vm/transhuge-stress.c b/tools/testing/selftests/vm/transhuge-stress.c new file mode 100644 index 000000000..e3f00adb1 --- /dev/null +++ b/tools/testing/selftests/vm/transhuge-stress.c @@ -0,0 +1,122 @@ +/* + * Stress test for transparent huge pages, memory compaction and migration. + * + * Authors: Konstantin Khlebnikov + * + * This is free and unencumbered software released into the public domain. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "util.h" + +int backing_fd = -1; +int mmap_flags = MAP_ANONYMOUS | MAP_NORESERVE | MAP_PRIVATE; +#define PROT_RW (PROT_READ | PROT_WRITE) + +int main(int argc, char **argv) +{ + size_t ram, len; + void *ptr, *p; + struct timespec a, b; + int i = 0; + char *name = NULL; + double s; + uint8_t *map; + size_t map_len; + int pagemap_fd; + + ram = sysconf(_SC_PHYS_PAGES); + if (ram > SIZE_MAX / sysconf(_SC_PAGESIZE) / 4) + ram = SIZE_MAX / 4; + else + ram *= sysconf(_SC_PAGESIZE); + len = ram; + + while (++i < argc) { + if (!strcmp(argv[i], "-h")) + errx(1, "usage: %s [size in MiB]", argv[0]); + else if (!strcmp(argv[i], "-f")) + name = argv[++i]; + else + len = atoll(argv[i]) << 20; + } + + if (name) { + backing_fd = open(name, O_RDWR); + if (backing_fd == -1) + errx(2, "open %s", name); + mmap_flags = MAP_SHARED; + } + + warnx("allocate %zd transhuge pages, using %zd MiB virtual memory" + " and %zd MiB of ram", len >> HPAGE_SHIFT, len >> 20, + ram >> (20 + HPAGE_SHIFT - PAGE_SHIFT - 1)); + + pagemap_fd = open("/proc/self/pagemap", O_RDONLY); + if (pagemap_fd < 0) + err(2, "open pagemap"); + + len -= len % HPAGE_SIZE; + ptr = mmap(NULL, len + HPAGE_SIZE, PROT_RW, mmap_flags, backing_fd, 0); + if (ptr == MAP_FAILED) + err(2, "initial mmap"); + ptr += HPAGE_SIZE - (uintptr_t)ptr % HPAGE_SIZE; + + if (madvise(ptr, len, MADV_HUGEPAGE)) + err(2, "MADV_HUGEPAGE"); + + map_len = ram >> (HPAGE_SHIFT - 1); + map = malloc(map_len); + if (!map) + errx(2, "map malloc"); + + while (1) { + int nr_succeed = 0, nr_failed = 0, nr_pages = 0; + + memset(map, 0, map_len); + + clock_gettime(CLOCK_MONOTONIC, &a); + for (p = ptr; p < ptr + len; p += HPAGE_SIZE) { + int64_t pfn; + + pfn = allocate_transhuge(p, pagemap_fd); + + if (pfn < 0) { + nr_failed++; + } else { + size_t idx = pfn >> (HPAGE_SHIFT - PAGE_SHIFT); + + nr_succeed++; + if (idx >= map_len) { + map = realloc(map, idx + 1); + if (!map) + errx(2, "map realloc"); + memset(map + map_len, 0, idx + 1 - map_len); + map_len = idx + 1; + } + if (!map[idx]) + nr_pages++; + map[idx] = 1; + } + + /* split transhuge page, keep last page */ + if (madvise(p, HPAGE_SIZE - PAGE_SIZE, MADV_DONTNEED)) + err(2, "MADV_DONTNEED"); + } + clock_gettime(CLOCK_MONOTONIC, &b); + s = b.tv_sec - a.tv_sec + (b.tv_nsec - a.tv_nsec) / 1000000000.; + + warnx("%.3f s/loop, %.3f ms/page, %10.3f MiB/s\t" + "%4d succeed, %4d failed, %4d different pages", + s, s * 1000 / (len >> HPAGE_SHIFT), len / s / (1 << 20), + nr_succeed, nr_failed, nr_pages); + } +} diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c new file mode 100644 index 000000000..297f250c1 --- /dev/null +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -0,0 +1,1878 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Stress userfaultfd syscall. + * + * Copyright (C) 2015 Red Hat, Inc. + * + * This test allocates two virtual areas and bounces the physical + * memory across the two virtual areas (from area_src to area_dst) + * using userfaultfd. + * + * There are three threads running per CPU: + * + * 1) one per-CPU thread takes a per-page pthread_mutex in a random + * page of the area_dst (while the physical page may still be in + * area_src), and increments a per-page counter in the same page, + * and checks its value against a verification region. + * + * 2) another per-CPU thread handles the userfaults generated by + * thread 1 above. userfaultfd blocking reads or poll() modes are + * exercised interleaved. + * + * 3) one last per-CPU thread transfers the memory in the background + * at maximum bandwidth (if not already transferred by thread + * 2). Each cpu thread takes cares of transferring a portion of the + * area. + * + * When all threads of type 3 completed the transfer, one bounce is + * complete. area_src and area_dst are then swapped. All threads are + * respawned and so the bounce is immediately restarted in the + * opposite direction. + * + * per-CPU threads 1 by triggering userfaults inside + * pthread_mutex_lock will also verify the atomicity of the memory + * transfer (UFFDIO_COPY). + */ + +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" +#include "vm_util.h" + +#ifdef __NR_userfaultfd + +static unsigned long nr_cpus, nr_pages, nr_pages_per_cpu, page_size, hpage_size; + +#define BOUNCE_RANDOM (1<<0) +#define BOUNCE_RACINGFAULTS (1<<1) +#define BOUNCE_VERIFY (1<<2) +#define BOUNCE_POLL (1<<3) +static int bounces; + +#define TEST_ANON 1 +#define TEST_HUGETLB 2 +#define TEST_SHMEM 3 +static int test_type; + +#define UFFD_FLAGS (O_CLOEXEC | O_NONBLOCK | UFFD_USER_MODE_ONLY) + +#define BASE_PMD_ADDR ((void *)(1UL << 30)) + +/* test using /dev/userfaultfd, instead of userfaultfd(2) */ +static bool test_dev_userfaultfd; + +/* exercise the test_uffdio_*_eexist every ALARM_INTERVAL_SECS */ +#define ALARM_INTERVAL_SECS 10 +static volatile bool test_uffdio_copy_eexist = true; +static volatile bool test_uffdio_zeropage_eexist = true; +/* Whether to test uffd write-protection */ +static bool test_uffdio_wp = true; +/* Whether to test uffd minor faults */ +static bool test_uffdio_minor = false; + +static bool map_shared; +static int shm_fd; +static int huge_fd; +static unsigned long long *count_verify; +static int uffd = -1; +static int uffd_flags, finished, *pipefd; +static char *area_src, *area_src_alias, *area_dst, *area_dst_alias, *area_remap; +static char *zeropage; +pthread_attr_t attr; +static bool test_collapse; + +/* Userfaultfd test statistics */ +struct uffd_stats { + int cpu; + unsigned long missing_faults; + unsigned long wp_faults; + unsigned long minor_faults; +}; + +/* pthread_mutex_t starts at page offset 0 */ +#define area_mutex(___area, ___nr) \ + ((pthread_mutex_t *) ((___area) + (___nr)*page_size)) +/* + * count is placed in the page after pthread_mutex_t naturally aligned + * to avoid non alignment faults on non-x86 archs. + */ +#define area_count(___area, ___nr) \ + ((volatile unsigned long long *) ((unsigned long) \ + ((___area) + (___nr)*page_size + \ + sizeof(pthread_mutex_t) + \ + sizeof(unsigned long long) - 1) & \ + ~(unsigned long)(sizeof(unsigned long long) \ + - 1))) + +#define swap(a, b) \ + do { typeof(a) __tmp = (a); (a) = (b); (b) = __tmp; } while (0) + +#define factor_of_2(x) ((x) ^ ((x) & ((x) - 1))) + +const char *examples = + "# Run anonymous memory test on 100MiB region with 99999 bounces:\n" + "./userfaultfd anon 100 99999\n\n" + "# Run the same anonymous memory test, but using /dev/userfaultfd:\n" + "./userfaultfd anon:dev 100 99999\n\n" + "# Run share memory test on 1GiB region with 99 bounces:\n" + "./userfaultfd shmem 1000 99\n\n" + "# Run hugetlb memory test on 256MiB region with 50 bounces:\n" + "./userfaultfd hugetlb 256 50\n\n" + "# Run the same hugetlb test but using shared file:\n" + "./userfaultfd hugetlb_shared 256 50 /dev/hugepages/hugefile\n\n" + "# 10MiB-~6GiB 999 bounces anonymous test, " + "continue forever unless an error triggers\n" + "while ./userfaultfd anon $[RANDOM % 6000 + 10] 999; do true; done\n\n"; + +static void usage(void) +{ + fprintf(stderr, "\nUsage: ./userfaultfd " + "[hugetlbfs_file]\n\n"); + fprintf(stderr, "Supported : anon, hugetlb, " + "hugetlb_shared, shmem\n\n"); + fprintf(stderr, "'Test mods' can be joined to the test type string with a ':'. " + "Supported mods:\n"); + fprintf(stderr, "\tsyscall - Use userfaultfd(2) (default)\n"); + fprintf(stderr, "\tdev - Use /dev/userfaultfd instead of userfaultfd(2)\n"); + fprintf(stderr, "\tcollapse - Test MADV_COLLAPSE of UFFDIO_REGISTER_MODE_MINOR\n" + "memory\n"); + fprintf(stderr, "\nExample test mod usage:\n"); + fprintf(stderr, "# Run anonymous memory test with /dev/userfaultfd:\n"); + fprintf(stderr, "./userfaultfd anon:dev 100 99999\n\n"); + + fprintf(stderr, "Examples:\n\n"); + fprintf(stderr, "%s", examples); + exit(1); +} + +#define _err(fmt, ...) \ + do { \ + int ret = errno; \ + fprintf(stderr, "ERROR: " fmt, ##__VA_ARGS__); \ + fprintf(stderr, " (errno=%d, line=%d)\n", \ + ret, __LINE__); \ + } while (0) + +#define errexit(exitcode, fmt, ...) \ + do { \ + _err(fmt, ##__VA_ARGS__); \ + exit(exitcode); \ + } while (0) + +#define err(fmt, ...) errexit(1, fmt, ##__VA_ARGS__) + +static void uffd_stats_reset(struct uffd_stats *uffd_stats, + unsigned long n_cpus) +{ + int i; + + for (i = 0; i < n_cpus; i++) { + uffd_stats[i].cpu = i; + uffd_stats[i].missing_faults = 0; + uffd_stats[i].wp_faults = 0; + uffd_stats[i].minor_faults = 0; + } +} + +static void uffd_stats_report(struct uffd_stats *stats, int n_cpus) +{ + int i; + unsigned long long miss_total = 0, wp_total = 0, minor_total = 0; + + for (i = 0; i < n_cpus; i++) { + miss_total += stats[i].missing_faults; + wp_total += stats[i].wp_faults; + minor_total += stats[i].minor_faults; + } + + printf("userfaults: "); + if (miss_total) { + printf("%llu missing (", miss_total); + for (i = 0; i < n_cpus; i++) + printf("%lu+", stats[i].missing_faults); + printf("\b) "); + } + if (wp_total) { + printf("%llu wp (", wp_total); + for (i = 0; i < n_cpus; i++) + printf("%lu+", stats[i].wp_faults); + printf("\b) "); + } + if (minor_total) { + printf("%llu minor (", minor_total); + for (i = 0; i < n_cpus; i++) + printf("%lu+", stats[i].minor_faults); + printf("\b)"); + } + printf("\n"); +} + +static void anon_release_pages(char *rel_area) +{ + if (madvise(rel_area, nr_pages * page_size, MADV_DONTNEED)) + err("madvise(MADV_DONTNEED) failed"); +} + +static void anon_allocate_area(void **alloc_area, bool is_src) +{ + *alloc_area = mmap(NULL, nr_pages * page_size, PROT_READ | PROT_WRITE, + MAP_ANONYMOUS | MAP_PRIVATE, -1, 0); +} + +static void noop_alias_mapping(__u64 *start, size_t len, unsigned long offset) +{ +} + +static void hugetlb_release_pages(char *rel_area) +{ + if (!map_shared) { + if (madvise(rel_area, nr_pages * page_size, MADV_DONTNEED)) + err("madvise(MADV_DONTNEED) failed"); + } else { + if (madvise(rel_area, nr_pages * page_size, MADV_REMOVE)) + err("madvise(MADV_REMOVE) failed"); + } +} + +static void hugetlb_allocate_area(void **alloc_area, bool is_src) +{ + void *area_alias = NULL; + char **alloc_area_alias; + + if (!map_shared) + *alloc_area = mmap(NULL, + nr_pages * page_size, + PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB | + (is_src ? 0 : MAP_NORESERVE), + -1, + 0); + else + *alloc_area = mmap(NULL, + nr_pages * page_size, + PROT_READ | PROT_WRITE, + MAP_SHARED | + (is_src ? 0 : MAP_NORESERVE), + huge_fd, + is_src ? 0 : nr_pages * page_size); + if (*alloc_area == MAP_FAILED) + err("mmap of hugetlbfs file failed"); + + if (map_shared) { + area_alias = mmap(NULL, + nr_pages * page_size, + PROT_READ | PROT_WRITE, + MAP_SHARED, + huge_fd, + is_src ? 0 : nr_pages * page_size); + if (area_alias == MAP_FAILED) + err("mmap of hugetlb file alias failed"); + } + + if (is_src) { + alloc_area_alias = &area_src_alias; + } else { + alloc_area_alias = &area_dst_alias; + } + if (area_alias) + *alloc_area_alias = area_alias; +} + +static void hugetlb_alias_mapping(__u64 *start, size_t len, unsigned long offset) +{ + if (!map_shared) + return; + + *start = (unsigned long) area_dst_alias + offset; +} + +static void shmem_release_pages(char *rel_area) +{ + if (madvise(rel_area, nr_pages * page_size, MADV_REMOVE)) + err("madvise(MADV_REMOVE) failed"); +} + +static void shmem_allocate_area(void **alloc_area, bool is_src) +{ + void *area_alias = NULL; + size_t bytes = nr_pages * page_size; + unsigned long offset = is_src ? 0 : bytes; + char *p = NULL, *p_alias = NULL; + + if (test_collapse) { + p = BASE_PMD_ADDR; + if (!is_src) + /* src map + alias + interleaved hpages */ + p += 2 * (bytes + hpage_size); + p_alias = p; + p_alias += bytes; + p_alias += hpage_size; /* Prevent src/dst VMA merge */ + } + + *alloc_area = mmap(p, bytes, PROT_READ | PROT_WRITE, MAP_SHARED, + shm_fd, offset); + if (*alloc_area == MAP_FAILED) + err("mmap of memfd failed"); + if (test_collapse && *alloc_area != p) + err("mmap of memfd failed at %p", p); + + area_alias = mmap(p_alias, bytes, PROT_READ | PROT_WRITE, MAP_SHARED, + shm_fd, offset); + if (area_alias == MAP_FAILED) + err("mmap of memfd alias failed"); + if (test_collapse && area_alias != p_alias) + err("mmap of anonymous memory failed at %p", p_alias); + + if (is_src) + area_src_alias = area_alias; + else + area_dst_alias = area_alias; +} + +static void shmem_alias_mapping(__u64 *start, size_t len, unsigned long offset) +{ + *start = (unsigned long)area_dst_alias + offset; +} + +static void shmem_check_pmd_mapping(void *p, int expect_nr_hpages) +{ + if (!check_huge_shmem(area_dst_alias, expect_nr_hpages, hpage_size)) + err("Did not find expected %d number of hugepages", + expect_nr_hpages); +} + +struct uffd_test_ops { + void (*allocate_area)(void **alloc_area, bool is_src); + void (*release_pages)(char *rel_area); + void (*alias_mapping)(__u64 *start, size_t len, unsigned long offset); + void (*check_pmd_mapping)(void *p, int expect_nr_hpages); +}; + +static struct uffd_test_ops anon_uffd_test_ops = { + .allocate_area = anon_allocate_area, + .release_pages = anon_release_pages, + .alias_mapping = noop_alias_mapping, + .check_pmd_mapping = NULL, +}; + +static struct uffd_test_ops shmem_uffd_test_ops = { + .allocate_area = shmem_allocate_area, + .release_pages = shmem_release_pages, + .alias_mapping = shmem_alias_mapping, + .check_pmd_mapping = shmem_check_pmd_mapping, +}; + +static struct uffd_test_ops hugetlb_uffd_test_ops = { + .allocate_area = hugetlb_allocate_area, + .release_pages = hugetlb_release_pages, + .alias_mapping = hugetlb_alias_mapping, + .check_pmd_mapping = NULL, +}; + +static struct uffd_test_ops *uffd_test_ops; + +static inline uint64_t uffd_minor_feature(void) +{ + if (test_type == TEST_HUGETLB && map_shared) + return UFFD_FEATURE_MINOR_HUGETLBFS; + else if (test_type == TEST_SHMEM) + return UFFD_FEATURE_MINOR_SHMEM; + else + return 0; +} + +static uint64_t get_expected_ioctls(uint64_t mode) +{ + uint64_t ioctls = UFFD_API_RANGE_IOCTLS; + + if (test_type == TEST_HUGETLB) + ioctls &= ~(1 << _UFFDIO_ZEROPAGE); + + if (!((mode & UFFDIO_REGISTER_MODE_WP) && test_uffdio_wp)) + ioctls &= ~(1 << _UFFDIO_WRITEPROTECT); + + if (!((mode & UFFDIO_REGISTER_MODE_MINOR) && test_uffdio_minor)) + ioctls &= ~(1 << _UFFDIO_CONTINUE); + + return ioctls; +} + +static void assert_expected_ioctls_present(uint64_t mode, uint64_t ioctls) +{ + uint64_t expected = get_expected_ioctls(mode); + uint64_t actual = ioctls & expected; + + if (actual != expected) { + err("missing ioctl(s): expected %"PRIx64" actual: %"PRIx64, + expected, actual); + } +} + +static int __userfaultfd_open_dev(void) +{ + int fd, _uffd; + + fd = open("/dev/userfaultfd", O_RDWR | O_CLOEXEC); + if (fd < 0) + errexit(KSFT_SKIP, "opening /dev/userfaultfd failed"); + + _uffd = ioctl(fd, USERFAULTFD_IOC_NEW, UFFD_FLAGS); + if (_uffd < 0) + errexit(errno == ENOTTY ? KSFT_SKIP : 1, + "creating userfaultfd failed"); + close(fd); + return _uffd; +} + +static void userfaultfd_open(uint64_t *features) +{ + struct uffdio_api uffdio_api; + + if (test_dev_userfaultfd) + uffd = __userfaultfd_open_dev(); + else { + uffd = syscall(__NR_userfaultfd, UFFD_FLAGS); + if (uffd < 0) + errexit(errno == ENOSYS ? KSFT_SKIP : 1, + "creating userfaultfd failed"); + } + uffd_flags = fcntl(uffd, F_GETFD, NULL); + + uffdio_api.api = UFFD_API; + uffdio_api.features = *features; + if (ioctl(uffd, UFFDIO_API, &uffdio_api)) + err("UFFDIO_API failed.\nPlease make sure to " + "run with either root or ptrace capability."); + if (uffdio_api.api != UFFD_API) + err("UFFDIO_API error: %" PRIu64, (uint64_t)uffdio_api.api); + + *features = uffdio_api.features; +} + +static inline void munmap_area(void **area) +{ + if (*area) + if (munmap(*area, nr_pages * page_size)) + err("munmap"); + + *area = NULL; +} + +static void uffd_test_ctx_clear(void) +{ + size_t i; + + if (pipefd) { + for (i = 0; i < nr_cpus * 2; ++i) { + if (close(pipefd[i])) + err("close pipefd"); + } + free(pipefd); + pipefd = NULL; + } + + if (count_verify) { + free(count_verify); + count_verify = NULL; + } + + if (uffd != -1) { + if (close(uffd)) + err("close uffd"); + uffd = -1; + } + + munmap_area((void **)&area_src); + munmap_area((void **)&area_src_alias); + munmap_area((void **)&area_dst); + munmap_area((void **)&area_dst_alias); + munmap_area((void **)&area_remap); +} + +static void uffd_test_ctx_init(uint64_t features) +{ + unsigned long nr, cpu; + + uffd_test_ctx_clear(); + + uffd_test_ops->allocate_area((void **)&area_src, true); + uffd_test_ops->allocate_area((void **)&area_dst, false); + + userfaultfd_open(&features); + + count_verify = malloc(nr_pages * sizeof(unsigned long long)); + if (!count_verify) + err("count_verify"); + + for (nr = 0; nr < nr_pages; nr++) { + *area_mutex(area_src, nr) = + (pthread_mutex_t)PTHREAD_MUTEX_INITIALIZER; + count_verify[nr] = *area_count(area_src, nr) = 1; + /* + * In the transition between 255 to 256, powerpc will + * read out of order in my_bcmp and see both bytes as + * zero, so leave a placeholder below always non-zero + * after the count, to avoid my_bcmp to trigger false + * positives. + */ + *(area_count(area_src, nr) + 1) = 1; + } + + /* + * After initialization of area_src, we must explicitly release pages + * for area_dst to make sure it's fully empty. Otherwise we could have + * some area_dst pages be errornously initialized with zero pages, + * hence we could hit memory corruption later in the test. + * + * One example is when THP is globally enabled, above allocate_area() + * calls could have the two areas merged into a single VMA (as they + * will have the same VMA flags so they're mergeable). When we + * initialize the area_src above, it's possible that some part of + * area_dst could have been faulted in via one huge THP that will be + * shared between area_src and area_dst. It could cause some of the + * area_dst won't be trapped by missing userfaults. + * + * This release_pages() will guarantee even if that happened, we'll + * proactively split the thp and drop any accidentally initialized + * pages within area_dst. + */ + uffd_test_ops->release_pages(area_dst); + + pipefd = malloc(sizeof(int) * nr_cpus * 2); + if (!pipefd) + err("pipefd"); + for (cpu = 0; cpu < nr_cpus; cpu++) + if (pipe2(&pipefd[cpu * 2], O_CLOEXEC | O_NONBLOCK)) + err("pipe"); +} + +static int my_bcmp(char *str1, char *str2, size_t n) +{ + unsigned long i; + for (i = 0; i < n; i++) + if (str1[i] != str2[i]) + return 1; + return 0; +} + +static void wp_range(int ufd, __u64 start, __u64 len, bool wp) +{ + struct uffdio_writeprotect prms; + + /* Write protection page faults */ + prms.range.start = start; + prms.range.len = len; + /* Undo write-protect, do wakeup after that */ + prms.mode = wp ? UFFDIO_WRITEPROTECT_MODE_WP : 0; + + if (ioctl(ufd, UFFDIO_WRITEPROTECT, &prms)) + err("clear WP failed: address=0x%"PRIx64, (uint64_t)start); +} + +static void continue_range(int ufd, __u64 start, __u64 len) +{ + struct uffdio_continue req; + int ret; + + req.range.start = start; + req.range.len = len; + req.mode = 0; + + if (ioctl(ufd, UFFDIO_CONTINUE, &req)) + err("UFFDIO_CONTINUE failed for address 0x%" PRIx64, + (uint64_t)start); + + /* + * Error handling within the kernel for continue is subtly different + * from copy or zeropage, so it may be a source of bugs. Trigger an + * error (-EEXIST) on purpose, to verify doing so doesn't cause a BUG. + */ + req.mapped = 0; + ret = ioctl(ufd, UFFDIO_CONTINUE, &req); + if (ret >= 0 || req.mapped != -EEXIST) + err("failed to exercise UFFDIO_CONTINUE error handling, ret=%d, mapped=%" PRId64, + ret, (int64_t) req.mapped); +} + +static void *locking_thread(void *arg) +{ + unsigned long cpu = (unsigned long) arg; + unsigned long page_nr; + unsigned long long count; + + if (!(bounces & BOUNCE_RANDOM)) { + page_nr = -bounces; + if (!(bounces & BOUNCE_RACINGFAULTS)) + page_nr += cpu * nr_pages_per_cpu; + } + + while (!finished) { + if (bounces & BOUNCE_RANDOM) { + if (getrandom(&page_nr, sizeof(page_nr), 0) != sizeof(page_nr)) + err("getrandom failed"); + } else + page_nr += 1; + page_nr %= nr_pages; + pthread_mutex_lock(area_mutex(area_dst, page_nr)); + count = *area_count(area_dst, page_nr); + if (count != count_verify[page_nr]) + err("page_nr %lu memory corruption %llu %llu", + page_nr, count, count_verify[page_nr]); + count++; + *area_count(area_dst, page_nr) = count_verify[page_nr] = count; + pthread_mutex_unlock(area_mutex(area_dst, page_nr)); + } + + return NULL; +} + +static void retry_copy_page(int ufd, struct uffdio_copy *uffdio_copy, + unsigned long offset) +{ + uffd_test_ops->alias_mapping(&uffdio_copy->dst, + uffdio_copy->len, + offset); + if (ioctl(ufd, UFFDIO_COPY, uffdio_copy)) { + /* real retval in ufdio_copy.copy */ + if (uffdio_copy->copy != -EEXIST) + err("UFFDIO_COPY retry error: %"PRId64, + (int64_t)uffdio_copy->copy); + } else { + err("UFFDIO_COPY retry unexpected: %"PRId64, + (int64_t)uffdio_copy->copy); + } +} + +static void wake_range(int ufd, unsigned long addr, unsigned long len) +{ + struct uffdio_range uffdio_wake; + + uffdio_wake.start = addr; + uffdio_wake.len = len; + + if (ioctl(ufd, UFFDIO_WAKE, &uffdio_wake)) + fprintf(stderr, "error waking %lu\n", + addr), exit(1); +} + +static int __copy_page(int ufd, unsigned long offset, bool retry) +{ + struct uffdio_copy uffdio_copy; + + if (offset >= nr_pages * page_size) + err("unexpected offset %lu\n", offset); + uffdio_copy.dst = (unsigned long) area_dst + offset; + uffdio_copy.src = (unsigned long) area_src + offset; + uffdio_copy.len = page_size; + if (test_uffdio_wp) + uffdio_copy.mode = UFFDIO_COPY_MODE_WP; + else + uffdio_copy.mode = 0; + uffdio_copy.copy = 0; + if (ioctl(ufd, UFFDIO_COPY, &uffdio_copy)) { + /* real retval in ufdio_copy.copy */ + if (uffdio_copy.copy != -EEXIST) + err("UFFDIO_COPY error: %"PRId64, + (int64_t)uffdio_copy.copy); + wake_range(ufd, uffdio_copy.dst, page_size); + } else if (uffdio_copy.copy != page_size) { + err("UFFDIO_COPY error: %"PRId64, (int64_t)uffdio_copy.copy); + } else { + if (test_uffdio_copy_eexist && retry) { + test_uffdio_copy_eexist = false; + retry_copy_page(ufd, &uffdio_copy, offset); + } + return 1; + } + return 0; +} + +static int copy_page_retry(int ufd, unsigned long offset) +{ + return __copy_page(ufd, offset, true); +} + +static int copy_page(int ufd, unsigned long offset) +{ + return __copy_page(ufd, offset, false); +} + +static int uffd_read_msg(int ufd, struct uffd_msg *msg) +{ + int ret = read(uffd, msg, sizeof(*msg)); + + if (ret != sizeof(*msg)) { + if (ret < 0) { + if (errno == EAGAIN || errno == EINTR) + return 1; + err("blocking read error"); + } else { + err("short read"); + } + } + + return 0; +} + +static void uffd_handle_page_fault(struct uffd_msg *msg, + struct uffd_stats *stats) +{ + unsigned long offset; + + if (msg->event != UFFD_EVENT_PAGEFAULT) + err("unexpected msg event %u", msg->event); + + if (msg->arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_WP) { + /* Write protect page faults */ + wp_range(uffd, msg->arg.pagefault.address, page_size, false); + stats->wp_faults++; + } else if (msg->arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_MINOR) { + uint8_t *area; + int b; + + /* + * Minor page faults + * + * To prove we can modify the original range for testing + * purposes, we're going to bit flip this range before + * continuing. + * + * Note that this requires all minor page fault tests operate on + * area_dst (non-UFFD-registered) and area_dst_alias + * (UFFD-registered). + */ + + area = (uint8_t *)(area_dst + + ((char *)msg->arg.pagefault.address - + area_dst_alias)); + for (b = 0; b < page_size; ++b) + area[b] = ~area[b]; + continue_range(uffd, msg->arg.pagefault.address, page_size); + stats->minor_faults++; + } else { + /* + * Missing page faults. + * + * Here we force a write check for each of the missing mode + * faults. It's guaranteed because the only threads that + * will trigger uffd faults are the locking threads, and + * their first instruction to touch the missing page will + * always be pthread_mutex_lock(). + * + * Note that here we relied on an NPTL glibc impl detail to + * always read the lock type at the entry of the lock op + * (pthread_mutex_t.__data.__type, offset 0x10) before + * doing any locking operations to guarantee that. It's + * actually not good to rely on this impl detail because + * logically a pthread-compatible lib can implement the + * locks without types and we can fail when linking with + * them. However since we used to find bugs with this + * strict check we still keep it around. Hopefully this + * could be a good hint when it fails again. If one day + * it'll break on some other impl of glibc we'll revisit. + */ + if (msg->arg.pagefault.flags & UFFD_PAGEFAULT_FLAG_WRITE) + err("unexpected write fault"); + + offset = (char *)(unsigned long)msg->arg.pagefault.address - area_dst; + offset &= ~(page_size-1); + + if (copy_page(uffd, offset)) + stats->missing_faults++; + } +} + +static void *uffd_poll_thread(void *arg) +{ + struct uffd_stats *stats = (struct uffd_stats *)arg; + unsigned long cpu = stats->cpu; + struct pollfd pollfd[2]; + struct uffd_msg msg; + struct uffdio_register uffd_reg; + int ret; + char tmp_chr; + + pollfd[0].fd = uffd; + pollfd[0].events = POLLIN; + pollfd[1].fd = pipefd[cpu*2]; + pollfd[1].events = POLLIN; + + for (;;) { + ret = poll(pollfd, 2, -1); + if (ret <= 0) { + if (errno == EINTR || errno == EAGAIN) + continue; + err("poll error: %d", ret); + } + if (pollfd[1].revents & POLLIN) { + if (read(pollfd[1].fd, &tmp_chr, 1) != 1) + err("read pipefd error"); + break; + } + if (!(pollfd[0].revents & POLLIN)) + err("pollfd[0].revents %d", pollfd[0].revents); + if (uffd_read_msg(uffd, &msg)) + continue; + switch (msg.event) { + default: + err("unexpected msg event %u\n", msg.event); + break; + case UFFD_EVENT_PAGEFAULT: + uffd_handle_page_fault(&msg, stats); + break; + case UFFD_EVENT_FORK: + close(uffd); + uffd = msg.arg.fork.ufd; + pollfd[0].fd = uffd; + break; + case UFFD_EVENT_REMOVE: + uffd_reg.range.start = msg.arg.remove.start; + uffd_reg.range.len = msg.arg.remove.end - + msg.arg.remove.start; + if (ioctl(uffd, UFFDIO_UNREGISTER, &uffd_reg.range)) + err("remove failure"); + break; + case UFFD_EVENT_REMAP: + area_remap = area_dst; /* save for later unmap */ + area_dst = (char *)(unsigned long)msg.arg.remap.to; + break; + } + } + + return NULL; +} + +pthread_mutex_t uffd_read_mutex = PTHREAD_MUTEX_INITIALIZER; + +static void *uffd_read_thread(void *arg) +{ + struct uffd_stats *stats = (struct uffd_stats *)arg; + struct uffd_msg msg; + + pthread_mutex_unlock(&uffd_read_mutex); + /* from here cancellation is ok */ + + for (;;) { + if (uffd_read_msg(uffd, &msg)) + continue; + uffd_handle_page_fault(&msg, stats); + } + + return NULL; +} + +static void *background_thread(void *arg) +{ + unsigned long cpu = (unsigned long) arg; + unsigned long page_nr, start_nr, mid_nr, end_nr; + + start_nr = cpu * nr_pages_per_cpu; + end_nr = (cpu+1) * nr_pages_per_cpu; + mid_nr = (start_nr + end_nr) / 2; + + /* Copy the first half of the pages */ + for (page_nr = start_nr; page_nr < mid_nr; page_nr++) + copy_page_retry(uffd, page_nr * page_size); + + /* + * If we need to test uffd-wp, set it up now. Then we'll have + * at least the first half of the pages mapped already which + * can be write-protected for testing + */ + if (test_uffdio_wp) + wp_range(uffd, (unsigned long)area_dst + start_nr * page_size, + nr_pages_per_cpu * page_size, true); + + /* + * Continue the 2nd half of the page copying, handling write + * protection faults if any + */ + for (page_nr = mid_nr; page_nr < end_nr; page_nr++) + copy_page_retry(uffd, page_nr * page_size); + + return NULL; +} + +static int stress(struct uffd_stats *uffd_stats) +{ + unsigned long cpu; + pthread_t locking_threads[nr_cpus]; + pthread_t uffd_threads[nr_cpus]; + pthread_t background_threads[nr_cpus]; + + finished = 0; + for (cpu = 0; cpu < nr_cpus; cpu++) { + if (pthread_create(&locking_threads[cpu], &attr, + locking_thread, (void *)cpu)) + return 1; + if (bounces & BOUNCE_POLL) { + if (pthread_create(&uffd_threads[cpu], &attr, + uffd_poll_thread, + (void *)&uffd_stats[cpu])) + return 1; + } else { + if (pthread_create(&uffd_threads[cpu], &attr, + uffd_read_thread, + (void *)&uffd_stats[cpu])) + return 1; + pthread_mutex_lock(&uffd_read_mutex); + } + if (pthread_create(&background_threads[cpu], &attr, + background_thread, (void *)cpu)) + return 1; + } + for (cpu = 0; cpu < nr_cpus; cpu++) + if (pthread_join(background_threads[cpu], NULL)) + return 1; + + /* + * Be strict and immediately zap area_src, the whole area has + * been transferred already by the background treads. The + * area_src could then be faulted in a racy way by still + * running uffdio_threads reading zeropages after we zapped + * area_src (but they're guaranteed to get -EEXIST from + * UFFDIO_COPY without writing zero pages into area_dst + * because the background threads already completed). + */ + uffd_test_ops->release_pages(area_src); + + finished = 1; + for (cpu = 0; cpu < nr_cpus; cpu++) + if (pthread_join(locking_threads[cpu], NULL)) + return 1; + + for (cpu = 0; cpu < nr_cpus; cpu++) { + char c; + if (bounces & BOUNCE_POLL) { + if (write(pipefd[cpu*2+1], &c, 1) != 1) + err("pipefd write error"); + if (pthread_join(uffd_threads[cpu], + (void *)&uffd_stats[cpu])) + return 1; + } else { + if (pthread_cancel(uffd_threads[cpu])) + return 1; + if (pthread_join(uffd_threads[cpu], NULL)) + return 1; + } + } + + return 0; +} + +sigjmp_buf jbuf, *sigbuf; + +static void sighndl(int sig, siginfo_t *siginfo, void *ptr) +{ + if (sig == SIGBUS) { + if (sigbuf) + siglongjmp(*sigbuf, 1); + abort(); + } +} + +/* + * For non-cooperative userfaultfd test we fork() a process that will + * generate pagefaults, will mremap the area monitored by the + * userfaultfd and at last this process will release the monitored + * area. + * For the anonymous and shared memory the area is divided into two + * parts, the first part is accessed before mremap, and the second + * part is accessed after mremap. Since hugetlbfs does not support + * mremap, the entire monitored area is accessed in a single pass for + * HUGETLB_TEST. + * The release of the pages currently generates event for shmem and + * anonymous memory (UFFD_EVENT_REMOVE), hence it is not checked + * for hugetlb. + * For signal test(UFFD_FEATURE_SIGBUS), signal_test = 1, we register + * monitored area, generate pagefaults and test that signal is delivered. + * Use UFFDIO_COPY to allocate missing page and retry. For signal_test = 2 + * test robustness use case - we release monitored area, fork a process + * that will generate pagefaults and verify signal is generated. + * This also tests UFFD_FEATURE_EVENT_FORK event along with the signal + * feature. Using monitor thread, verify no userfault events are generated. + */ +static int faulting_process(int signal_test) +{ + unsigned long nr; + unsigned long long count; + unsigned long split_nr_pages; + unsigned long lastnr; + struct sigaction act; + volatile unsigned long signalled = 0; + + split_nr_pages = (nr_pages + 1) / 2; + + if (signal_test) { + sigbuf = &jbuf; + memset(&act, 0, sizeof(act)); + act.sa_sigaction = sighndl; + act.sa_flags = SA_SIGINFO; + if (sigaction(SIGBUS, &act, 0)) + err("sigaction"); + lastnr = (unsigned long)-1; + } + + for (nr = 0; nr < split_nr_pages; nr++) { + volatile int steps = 1; + unsigned long offset = nr * page_size; + + if (signal_test) { + if (sigsetjmp(*sigbuf, 1) != 0) { + if (steps == 1 && nr == lastnr) + err("Signal repeated"); + + lastnr = nr; + if (signal_test == 1) { + if (steps == 1) { + /* This is a MISSING request */ + steps++; + if (copy_page(uffd, offset)) + signalled++; + } else { + /* This is a WP request */ + assert(steps == 2); + wp_range(uffd, + (__u64)area_dst + + offset, + page_size, false); + } + } else { + signalled++; + continue; + } + } + } + + count = *area_count(area_dst, nr); + if (count != count_verify[nr]) + err("nr %lu memory corruption %llu %llu\n", + nr, count, count_verify[nr]); + /* + * Trigger write protection if there is by writing + * the same value back. + */ + *area_count(area_dst, nr) = count; + } + + if (signal_test) + return signalled != split_nr_pages; + + area_dst = mremap(area_dst, nr_pages * page_size, nr_pages * page_size, + MREMAP_MAYMOVE | MREMAP_FIXED, area_src); + if (area_dst == MAP_FAILED) + err("mremap"); + /* Reset area_src since we just clobbered it */ + area_src = NULL; + + for (; nr < nr_pages; nr++) { + count = *area_count(area_dst, nr); + if (count != count_verify[nr]) { + err("nr %lu memory corruption %llu %llu\n", + nr, count, count_verify[nr]); + } + /* + * Trigger write protection if there is by writing + * the same value back. + */ + *area_count(area_dst, nr) = count; + } + + uffd_test_ops->release_pages(area_dst); + + for (nr = 0; nr < nr_pages; nr++) + if (my_bcmp(area_dst + nr * page_size, zeropage, page_size)) + err("nr %lu is not zero", nr); + + return 0; +} + +static void retry_uffdio_zeropage(int ufd, + struct uffdio_zeropage *uffdio_zeropage, + unsigned long offset) +{ + uffd_test_ops->alias_mapping(&uffdio_zeropage->range.start, + uffdio_zeropage->range.len, + offset); + if (ioctl(ufd, UFFDIO_ZEROPAGE, uffdio_zeropage)) { + if (uffdio_zeropage->zeropage != -EEXIST) + err("UFFDIO_ZEROPAGE error: %"PRId64, + (int64_t)uffdio_zeropage->zeropage); + } else { + err("UFFDIO_ZEROPAGE error: %"PRId64, + (int64_t)uffdio_zeropage->zeropage); + } +} + +static int __uffdio_zeropage(int ufd, unsigned long offset, bool retry) +{ + struct uffdio_zeropage uffdio_zeropage; + int ret; + bool has_zeropage = get_expected_ioctls(0) & (1 << _UFFDIO_ZEROPAGE); + __s64 res; + + if (offset >= nr_pages * page_size) + err("unexpected offset %lu", offset); + uffdio_zeropage.range.start = (unsigned long) area_dst + offset; + uffdio_zeropage.range.len = page_size; + uffdio_zeropage.mode = 0; + ret = ioctl(ufd, UFFDIO_ZEROPAGE, &uffdio_zeropage); + res = uffdio_zeropage.zeropage; + if (ret) { + /* real retval in ufdio_zeropage.zeropage */ + if (has_zeropage) + err("UFFDIO_ZEROPAGE error: %"PRId64, (int64_t)res); + else if (res != -EINVAL) + err("UFFDIO_ZEROPAGE not -EINVAL"); + } else if (has_zeropage) { + if (res != page_size) { + err("UFFDIO_ZEROPAGE unexpected size"); + } else { + if (test_uffdio_zeropage_eexist && retry) { + test_uffdio_zeropage_eexist = false; + retry_uffdio_zeropage(ufd, &uffdio_zeropage, + offset); + } + return 1; + } + } else + err("UFFDIO_ZEROPAGE succeeded"); + + return 0; +} + +static int uffdio_zeropage(int ufd, unsigned long offset) +{ + return __uffdio_zeropage(ufd, offset, false); +} + +/* exercise UFFDIO_ZEROPAGE */ +static int userfaultfd_zeropage_test(void) +{ + struct uffdio_register uffdio_register; + + printf("testing UFFDIO_ZEROPAGE: "); + fflush(stdout); + + uffd_test_ctx_init(0); + + uffdio_register.range.start = (unsigned long) area_dst; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING; + if (test_uffdio_wp) + uffdio_register.mode |= UFFDIO_REGISTER_MODE_WP; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure"); + + assert_expected_ioctls_present( + uffdio_register.mode, uffdio_register.ioctls); + + if (uffdio_zeropage(uffd, 0)) + if (my_bcmp(area_dst, zeropage, page_size)) + err("zeropage is not zero"); + + printf("done.\n"); + return 0; +} + +static int userfaultfd_events_test(void) +{ + struct uffdio_register uffdio_register; + pthread_t uffd_mon; + int err, features; + pid_t pid; + char c; + struct uffd_stats stats = { 0 }; + + printf("testing events (fork, remap, remove): "); + fflush(stdout); + + features = UFFD_FEATURE_EVENT_FORK | UFFD_FEATURE_EVENT_REMAP | + UFFD_FEATURE_EVENT_REMOVE; + uffd_test_ctx_init(features); + + fcntl(uffd, F_SETFL, uffd_flags | O_NONBLOCK); + + uffdio_register.range.start = (unsigned long) area_dst; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING; + if (test_uffdio_wp) + uffdio_register.mode |= UFFDIO_REGISTER_MODE_WP; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure"); + + assert_expected_ioctls_present( + uffdio_register.mode, uffdio_register.ioctls); + + if (pthread_create(&uffd_mon, &attr, uffd_poll_thread, &stats)) + err("uffd_poll_thread create"); + + pid = fork(); + if (pid < 0) + err("fork"); + + if (!pid) + exit(faulting_process(0)); + + waitpid(pid, &err, 0); + if (err) + err("faulting process failed"); + if (write(pipefd[1], &c, sizeof(c)) != sizeof(c)) + err("pipe write"); + if (pthread_join(uffd_mon, NULL)) + return 1; + + uffd_stats_report(&stats, 1); + + return stats.missing_faults != nr_pages; +} + +static int userfaultfd_sig_test(void) +{ + struct uffdio_register uffdio_register; + unsigned long userfaults; + pthread_t uffd_mon; + int err, features; + pid_t pid; + char c; + struct uffd_stats stats = { 0 }; + + printf("testing signal delivery: "); + fflush(stdout); + + features = UFFD_FEATURE_EVENT_FORK|UFFD_FEATURE_SIGBUS; + uffd_test_ctx_init(features); + + fcntl(uffd, F_SETFL, uffd_flags | O_NONBLOCK); + + uffdio_register.range.start = (unsigned long) area_dst; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING; + if (test_uffdio_wp) + uffdio_register.mode |= UFFDIO_REGISTER_MODE_WP; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure"); + + assert_expected_ioctls_present( + uffdio_register.mode, uffdio_register.ioctls); + + if (faulting_process(1)) + err("faulting process failed"); + + uffd_test_ops->release_pages(area_dst); + + if (pthread_create(&uffd_mon, &attr, uffd_poll_thread, &stats)) + err("uffd_poll_thread create"); + + pid = fork(); + if (pid < 0) + err("fork"); + + if (!pid) + exit(faulting_process(2)); + + waitpid(pid, &err, 0); + if (err) + err("faulting process failed"); + if (write(pipefd[1], &c, sizeof(c)) != sizeof(c)) + err("pipe write"); + if (pthread_join(uffd_mon, (void **)&userfaults)) + return 1; + + printf("done.\n"); + if (userfaults) + err("Signal test failed, userfaults: %ld", userfaults); + + return userfaults != 0; +} + +void check_memory_contents(char *p) +{ + unsigned long i; + uint8_t expected_byte; + void *expected_page; + + if (posix_memalign(&expected_page, page_size, page_size)) + err("out of memory"); + + for (i = 0; i < nr_pages; ++i) { + expected_byte = ~((uint8_t)(i % ((uint8_t)-1))); + memset(expected_page, expected_byte, page_size); + if (my_bcmp(expected_page, p + (i * page_size), page_size)) + err("unexpected page contents after minor fault"); + } + + free(expected_page); +} + +static int userfaultfd_minor_test(void) +{ + unsigned long p; + struct uffdio_register uffdio_register; + pthread_t uffd_mon; + char c; + struct uffd_stats stats = { 0 }; + + if (!test_uffdio_minor) + return 0; + + printf("testing minor faults: "); + fflush(stdout); + + uffd_test_ctx_init(uffd_minor_feature()); + + uffdio_register.range.start = (unsigned long)area_dst_alias; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MINOR; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure"); + + assert_expected_ioctls_present( + uffdio_register.mode, uffdio_register.ioctls); + + /* + * After registering with UFFD, populate the non-UFFD-registered side of + * the shared mapping. This should *not* trigger any UFFD minor faults. + */ + for (p = 0; p < nr_pages; ++p) { + memset(area_dst + (p * page_size), p % ((uint8_t)-1), + page_size); + } + + if (pthread_create(&uffd_mon, &attr, uffd_poll_thread, &stats)) + err("uffd_poll_thread create"); + + /* + * Read each of the pages back using the UFFD-registered mapping. We + * expect that the first time we touch a page, it will result in a minor + * fault. uffd_poll_thread will resolve the fault by bit-flipping the + * page's contents, and then issuing a CONTINUE ioctl. + */ + check_memory_contents(area_dst_alias); + + if (write(pipefd[1], &c, sizeof(c)) != sizeof(c)) + err("pipe write"); + if (pthread_join(uffd_mon, NULL)) + return 1; + + uffd_stats_report(&stats, 1); + + if (test_collapse) { + printf("testing collapse of uffd memory into PMD-mapped THPs:"); + if (madvise(area_dst_alias, nr_pages * page_size, + MADV_COLLAPSE)) + err("madvise(MADV_COLLAPSE)"); + + uffd_test_ops->check_pmd_mapping(area_dst, + nr_pages * page_size / + hpage_size); + /* + * This won't cause uffd-fault - it purely just makes sure there + * was no corruption. + */ + check_memory_contents(area_dst_alias); + printf(" done.\n"); + } + + return stats.missing_faults != 0 || stats.minor_faults != nr_pages; +} + +#define BIT_ULL(nr) (1ULL << (nr)) +#define PM_SOFT_DIRTY BIT_ULL(55) +#define PM_MMAP_EXCLUSIVE BIT_ULL(56) +#define PM_UFFD_WP BIT_ULL(57) +#define PM_FILE BIT_ULL(61) +#define PM_SWAP BIT_ULL(62) +#define PM_PRESENT BIT_ULL(63) + +static int pagemap_open(void) +{ + int fd = open("/proc/self/pagemap", O_RDONLY); + + if (fd < 0) + err("open pagemap"); + + return fd; +} + +static uint64_t pagemap_read_vaddr(int fd, void *vaddr) +{ + uint64_t value; + int ret; + + ret = pread(fd, &value, sizeof(uint64_t), + ((uint64_t)vaddr >> 12) * sizeof(uint64_t)); + if (ret != sizeof(uint64_t)) + err("pread() on pagemap failed"); + + return value; +} + +/* This macro let __LINE__ works in err() */ +#define pagemap_check_wp(value, wp) do { \ + if (!!(value & PM_UFFD_WP) != wp) \ + err("pagemap uffd-wp bit error: 0x%"PRIx64, value); \ + } while (0) + +static int pagemap_test_fork(bool present) +{ + pid_t child = fork(); + uint64_t value; + int fd, result; + + if (!child) { + /* Open the pagemap fd of the child itself */ + fd = pagemap_open(); + value = pagemap_read_vaddr(fd, area_dst); + /* + * After fork() uffd-wp bit should be gone as long as we're + * without UFFD_FEATURE_EVENT_FORK + */ + pagemap_check_wp(value, false); + /* Succeed */ + exit(0); + } + waitpid(child, &result, 0); + return result; +} + +static void userfaultfd_pagemap_test(unsigned int test_pgsize) +{ + struct uffdio_register uffdio_register; + int pagemap_fd; + uint64_t value; + + /* Pagemap tests uffd-wp only */ + if (!test_uffdio_wp) + return; + + /* Not enough memory to test this page size */ + if (test_pgsize > nr_pages * page_size) + return; + + printf("testing uffd-wp with pagemap (pgsize=%u): ", test_pgsize); + /* Flush so it doesn't flush twice in parent/child later */ + fflush(stdout); + + uffd_test_ctx_init(0); + + if (test_pgsize > page_size) { + /* This is a thp test */ + if (madvise(area_dst, nr_pages * page_size, MADV_HUGEPAGE)) + err("madvise(MADV_HUGEPAGE) failed"); + } else if (test_pgsize == page_size) { + /* This is normal page test; force no thp */ + if (madvise(area_dst, nr_pages * page_size, MADV_NOHUGEPAGE)) + err("madvise(MADV_NOHUGEPAGE) failed"); + } + + uffdio_register.range.start = (unsigned long) area_dst; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_WP; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failed"); + + pagemap_fd = pagemap_open(); + + /* Touch the page */ + *area_dst = 1; + wp_range(uffd, (uint64_t)area_dst, test_pgsize, true); + value = pagemap_read_vaddr(pagemap_fd, area_dst); + pagemap_check_wp(value, true); + /* Make sure uffd-wp bit dropped when fork */ + if (pagemap_test_fork(true)) + err("Detected stall uffd-wp bit in child"); + + /* Exclusive required or PAGEOUT won't work */ + if (!(value & PM_MMAP_EXCLUSIVE)) + err("multiple mapping detected: 0x%"PRIx64, value); + + if (madvise(area_dst, test_pgsize, MADV_PAGEOUT)) + err("madvise(MADV_PAGEOUT) failed"); + + /* Uffd-wp should persist even swapped out */ + value = pagemap_read_vaddr(pagemap_fd, area_dst); + pagemap_check_wp(value, true); + /* Make sure uffd-wp bit dropped when fork */ + if (pagemap_test_fork(false)) + err("Detected stall uffd-wp bit in child"); + + /* Unprotect; this tests swap pte modifications */ + wp_range(uffd, (uint64_t)area_dst, page_size, false); + value = pagemap_read_vaddr(pagemap_fd, area_dst); + pagemap_check_wp(value, false); + + /* Fault in the page from disk */ + *area_dst = 2; + value = pagemap_read_vaddr(pagemap_fd, area_dst); + pagemap_check_wp(value, false); + + close(pagemap_fd); + printf("done\n"); +} + +static int userfaultfd_stress(void) +{ + void *area; + unsigned long nr; + struct uffdio_register uffdio_register; + struct uffd_stats uffd_stats[nr_cpus]; + + uffd_test_ctx_init(0); + + if (posix_memalign(&area, page_size, page_size)) + err("out of memory"); + zeropage = area; + bzero(zeropage, page_size); + + pthread_mutex_lock(&uffd_read_mutex); + + pthread_attr_init(&attr); + pthread_attr_setstacksize(&attr, 16*1024*1024); + + while (bounces--) { + printf("bounces: %d, mode:", bounces); + if (bounces & BOUNCE_RANDOM) + printf(" rnd"); + if (bounces & BOUNCE_RACINGFAULTS) + printf(" racing"); + if (bounces & BOUNCE_VERIFY) + printf(" ver"); + if (bounces & BOUNCE_POLL) + printf(" poll"); + else + printf(" read"); + printf(", "); + fflush(stdout); + + if (bounces & BOUNCE_POLL) + fcntl(uffd, F_SETFL, uffd_flags | O_NONBLOCK); + else + fcntl(uffd, F_SETFL, uffd_flags & ~O_NONBLOCK); + + /* register */ + uffdio_register.range.start = (unsigned long) area_dst; + uffdio_register.range.len = nr_pages * page_size; + uffdio_register.mode = UFFDIO_REGISTER_MODE_MISSING; + if (test_uffdio_wp) + uffdio_register.mode |= UFFDIO_REGISTER_MODE_WP; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure"); + assert_expected_ioctls_present( + uffdio_register.mode, uffdio_register.ioctls); + + if (area_dst_alias) { + uffdio_register.range.start = (unsigned long) + area_dst_alias; + if (ioctl(uffd, UFFDIO_REGISTER, &uffdio_register)) + err("register failure alias"); + } + + /* + * The madvise done previously isn't enough: some + * uffd_thread could have read userfaults (one of + * those already resolved by the background thread) + * and it may be in the process of calling + * UFFDIO_COPY. UFFDIO_COPY will read the zapped + * area_src and it would map a zero page in it (of + * course such a UFFDIO_COPY is perfectly safe as it'd + * return -EEXIST). The problem comes at the next + * bounce though: that racing UFFDIO_COPY would + * generate zeropages in the area_src, so invalidating + * the previous MADV_DONTNEED. Without this additional + * MADV_DONTNEED those zeropages leftovers in the + * area_src would lead to -EEXIST failure during the + * next bounce, effectively leaving a zeropage in the + * area_dst. + * + * Try to comment this out madvise to see the memory + * corruption being caught pretty quick. + * + * khugepaged is also inhibited to collapse THP after + * MADV_DONTNEED only after the UFFDIO_REGISTER, so it's + * required to MADV_DONTNEED here. + */ + uffd_test_ops->release_pages(area_dst); + + uffd_stats_reset(uffd_stats, nr_cpus); + + /* bounce pass */ + if (stress(uffd_stats)) + return 1; + + /* Clear all the write protections if there is any */ + if (test_uffdio_wp) + wp_range(uffd, (unsigned long)area_dst, + nr_pages * page_size, false); + + /* unregister */ + if (ioctl(uffd, UFFDIO_UNREGISTER, &uffdio_register.range)) + err("unregister failure"); + if (area_dst_alias) { + uffdio_register.range.start = (unsigned long) area_dst; + if (ioctl(uffd, UFFDIO_UNREGISTER, + &uffdio_register.range)) + err("unregister failure alias"); + } + + /* verification */ + if (bounces & BOUNCE_VERIFY) + for (nr = 0; nr < nr_pages; nr++) + if (*area_count(area_dst, nr) != count_verify[nr]) + err("error area_count %llu %llu %lu\n", + *area_count(area_src, nr), + count_verify[nr], nr); + + /* prepare next bounce */ + swap(area_src, area_dst); + + swap(area_src_alias, area_dst_alias); + + uffd_stats_report(uffd_stats, nr_cpus); + } + + if (test_type == TEST_ANON) { + /* + * shmem/hugetlb won't be able to run since they have different + * behavior on fork() (file-backed memory normally drops ptes + * directly when fork), meanwhile the pagemap test will verify + * pgtable entry of fork()ed child. + */ + userfaultfd_pagemap_test(page_size); + /* + * Hard-code for x86_64 for now for 2M THP, as x86_64 is + * currently the only one that supports uffd-wp + */ + userfaultfd_pagemap_test(page_size * 512); + } + + return userfaultfd_zeropage_test() || userfaultfd_sig_test() + || userfaultfd_events_test() || userfaultfd_minor_test(); +} + +/* + * Copied from mlock2-tests.c + */ +unsigned long default_huge_page_size(void) +{ + unsigned long hps = 0; + char *line = NULL; + size_t linelen = 0; + FILE *f = fopen("/proc/meminfo", "r"); + + if (!f) + return 0; + while (getline(&line, &linelen, f) > 0) { + if (sscanf(line, "Hugepagesize: %lu kB", &hps) == 1) { + hps <<= 10; + break; + } + } + + free(line); + fclose(f); + return hps; +} + +static void set_test_type(const char *type) +{ + if (!strcmp(type, "anon")) { + test_type = TEST_ANON; + uffd_test_ops = &anon_uffd_test_ops; + } else if (!strcmp(type, "hugetlb")) { + test_type = TEST_HUGETLB; + uffd_test_ops = &hugetlb_uffd_test_ops; + } else if (!strcmp(type, "hugetlb_shared")) { + map_shared = true; + test_type = TEST_HUGETLB; + uffd_test_ops = &hugetlb_uffd_test_ops; + /* Minor faults require shared hugetlb; only enable here. */ + test_uffdio_minor = true; + } else if (!strcmp(type, "shmem")) { + map_shared = true; + test_type = TEST_SHMEM; + uffd_test_ops = &shmem_uffd_test_ops; + test_uffdio_minor = true; + } +} + +static void parse_test_type_arg(const char *raw_type) +{ + char *buf = strdup(raw_type); + uint64_t features = UFFD_API_FEATURES; + + while (buf) { + const char *token = strsep(&buf, ":"); + + if (!test_type) + set_test_type(token); + else if (!strcmp(token, "dev")) + test_dev_userfaultfd = true; + else if (!strcmp(token, "syscall")) + test_dev_userfaultfd = false; + else if (!strcmp(token, "collapse")) + test_collapse = true; + else + err("unrecognized test mod '%s'", token); + } + + if (!test_type) + err("failed to parse test type argument: '%s'", raw_type); + + if (test_collapse && test_type != TEST_SHMEM) + err("Unsupported test: %s", raw_type); + + if (test_type == TEST_HUGETLB) + page_size = hpage_size; + else + page_size = sysconf(_SC_PAGE_SIZE); + + if (!page_size) + err("Unable to determine page size"); + if ((unsigned long) area_count(NULL, 0) + sizeof(unsigned long long) * 2 + > page_size) + err("Impossible to run this test"); + + /* + * Whether we can test certain features depends not just on test type, + * but also on whether or not this particular kernel supports the + * feature. + */ + + userfaultfd_open(&features); + + test_uffdio_wp = test_uffdio_wp && + (features & UFFD_FEATURE_PAGEFAULT_FLAG_WP); + test_uffdio_minor = test_uffdio_minor && + (features & uffd_minor_feature()); + + close(uffd); + uffd = -1; +} + +static void sigalrm(int sig) +{ + if (sig != SIGALRM) + abort(); + test_uffdio_copy_eexist = true; + test_uffdio_zeropage_eexist = true; + alarm(ALARM_INTERVAL_SECS); +} + +int main(int argc, char **argv) +{ + size_t bytes; + + if (argc < 4) + usage(); + + if (signal(SIGALRM, sigalrm) == SIG_ERR) + err("failed to arm SIGALRM"); + alarm(ALARM_INTERVAL_SECS); + + hpage_size = default_huge_page_size(); + parse_test_type_arg(argv[1]); + bytes = atol(argv[2]) * 1024 * 1024; + + if (test_collapse && bytes & (hpage_size - 1)) + err("MiB must be multiple of %lu if :collapse mod set", + hpage_size >> 20); + + nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); + + if (test_collapse) { + /* nr_cpus must divide (bytes / page_size), otherwise, + * area allocations of (nr_pages * paze_size) won't be a + * multiple of hpage_size, even if bytes is a multiple of + * hpage_size. + * + * This means that nr_cpus must divide (N * (2 << (H-P)) + * where: + * bytes = hpage_size * N + * hpage_size = 2 << H + * page_size = 2 << P + * + * And we want to chose nr_cpus to be the largest value + * satisfying this constraint, not larger than the number + * of online CPUs. Unfortunately, prime factorization of + * N and nr_cpus may be arbitrary, so have to search for it. + * Instead, just use the highest power of 2 dividing both + * nr_cpus and (bytes / page_size). + */ + int x = factor_of_2(nr_cpus); + int y = factor_of_2(bytes / page_size); + + nr_cpus = x < y ? x : y; + } + nr_pages_per_cpu = bytes / page_size / nr_cpus; + if (!nr_pages_per_cpu) { + _err("invalid MiB"); + usage(); + } + + bounces = atoi(argv[3]); + if (bounces <= 0) { + _err("invalid bounces"); + usage(); + } + nr_pages = nr_pages_per_cpu * nr_cpus; + + if (test_type == TEST_HUGETLB && map_shared) { + if (argc < 5) + usage(); + huge_fd = open(argv[4], O_CREAT | O_RDWR, 0755); + if (huge_fd < 0) + err("Open of %s failed", argv[4]); + if (ftruncate(huge_fd, 0)) + err("ftruncate %s to size 0 failed", argv[4]); + } else if (test_type == TEST_SHMEM) { + shm_fd = memfd_create(argv[0], 0); + if (shm_fd < 0) + err("memfd_create"); + if (ftruncate(shm_fd, nr_pages * page_size * 2)) + err("ftruncate"); + if (fallocate(shm_fd, + FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, 0, + nr_pages * page_size * 2)) + err("fallocate"); + } + printf("nr_pages: %lu, nr_pages_per_cpu: %lu\n", + nr_pages, nr_pages_per_cpu); + return userfaultfd_stress(); +} + +#else /* __NR_userfaultfd */ + +#warning "missing __NR_userfaultfd definition" + +int main(void) +{ + printf("skip: Skipping userfaultfd test (missing __NR_userfaultfd)\n"); + return KSFT_SKIP; +} + +#endif /* __NR_userfaultfd */ diff --git a/tools/testing/selftests/vm/util.h b/tools/testing/selftests/vm/util.h new file mode 100644 index 000000000..b27d26199 --- /dev/null +++ b/tools/testing/selftests/vm/util.h @@ -0,0 +1,69 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef __KSELFTEST_VM_UTIL_H +#define __KSELFTEST_VM_UTIL_H + +#include +#include +#include +#include /* ffsl() */ +#include /* _SC_PAGESIZE */ + +static unsigned int __page_size; +static unsigned int __page_shift; + +static inline unsigned int page_size(void) +{ + if (!__page_size) + __page_size = sysconf(_SC_PAGESIZE); + return __page_size; +} + +static inline unsigned int page_shift(void) +{ + if (!__page_shift) + __page_shift = (ffsl(page_size()) - 1); + return __page_shift; +} + +#define PAGE_SHIFT (page_shift()) +#define PAGE_SIZE (page_size()) +/* + * On ppc64 this will only work with radix 2M hugepage size + */ +#define HPAGE_SHIFT 21 +#define HPAGE_SIZE (1 << HPAGE_SHIFT) + +#define PAGEMAP_PRESENT(ent) (((ent) & (1ull << 63)) != 0) +#define PAGEMAP_PFN(ent) ((ent) & ((1ull << 55) - 1)) + + +static inline int64_t allocate_transhuge(void *ptr, int pagemap_fd) +{ + uint64_t ent[2]; + + /* drop pmd */ + if (mmap(ptr, HPAGE_SIZE, PROT_READ | PROT_WRITE, + MAP_FIXED | MAP_ANONYMOUS | + MAP_NORESERVE | MAP_PRIVATE, -1, 0) != ptr) + errx(2, "mmap transhuge"); + + if (madvise(ptr, HPAGE_SIZE, MADV_HUGEPAGE)) + err(2, "MADV_HUGEPAGE"); + + /* allocate transparent huge page */ + *(volatile void **)ptr = ptr; + + if (pread(pagemap_fd, ent, sizeof(ent), + (uintptr_t)ptr >> (PAGE_SHIFT - 3)) != sizeof(ent)) + err(2, "read pagemap"); + + if (PAGEMAP_PRESENT(ent[0]) && PAGEMAP_PRESENT(ent[1]) && + PAGEMAP_PFN(ent[0]) + 1 == PAGEMAP_PFN(ent[1]) && + !(PAGEMAP_PFN(ent[0]) & ((1 << (HPAGE_SHIFT - PAGE_SHIFT)) - 1))) + return PAGEMAP_PFN(ent[0]); + + return -1; +} + +#endif diff --git a/tools/testing/selftests/vm/va_128TBswitch.c b/tools/testing/selftests/vm/va_128TBswitch.c new file mode 100644 index 000000000..1d2068989 --- /dev/null +++ b/tools/testing/selftests/vm/va_128TBswitch.c @@ -0,0 +1,289 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * + * Authors: Kirill A. Shutemov + * Authors: Aneesh Kumar K.V + */ + +#include +#include +#include + +#include "../kselftest.h" + +#ifdef __powerpc64__ +#define PAGE_SIZE (64 << 10) +/* + * This will work with 16M and 2M hugepage size + */ +#define HUGETLB_SIZE (16 << 20) +#else +#define PAGE_SIZE (4 << 10) +#define HUGETLB_SIZE (2 << 20) +#endif + +/* + * >= 128TB is the hint addr value we used to select + * large address space. + */ +#define ADDR_SWITCH_HINT (1UL << 47) +#define LOW_ADDR ((void *) (1UL << 30)) +#define HIGH_ADDR ((void *) (1UL << 48)) + +struct testcase { + void *addr; + unsigned long size; + unsigned long flags; + const char *msg; + unsigned int low_addr_required:1; + unsigned int keep_mapped:1; +}; + +static struct testcase testcases[] = { + { + /* + * If stack is moved, we could possibly allocate + * this at the requested address. + */ + .addr = ((void *)(ADDR_SWITCH_HINT - PAGE_SIZE)), + .size = PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE, PAGE_SIZE)", + .low_addr_required = 1, + }, + { + /* + * We should never allocate at the requested address or above it + * The len cross the 128TB boundary. Without MAP_FIXED + * we will always search in the lower address space. + */ + .addr = ((void *)(ADDR_SWITCH_HINT - PAGE_SIZE)), + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE, (2 * PAGE_SIZE))", + .low_addr_required = 1, + }, + { + /* + * Exact mapping at 128TB, the area is free we should get that + * even without MAP_FIXED. + */ + .addr = ((void *)(ADDR_SWITCH_HINT)), + .size = PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT, PAGE_SIZE)", + .keep_mapped = 1, + }, + { + .addr = (void *)(ADDR_SWITCH_HINT), + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED, + .msg = "mmap(ADDR_SWITCH_HINT, 2 * PAGE_SIZE, MAP_FIXED)", + }, + { + .addr = NULL, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(NULL)", + .low_addr_required = 1, + }, + { + .addr = LOW_ADDR, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(LOW_ADDR)", + .low_addr_required = 1, + }, + { + .addr = HIGH_ADDR, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(HIGH_ADDR)", + .keep_mapped = 1, + }, + { + .addr = HIGH_ADDR, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(HIGH_ADDR) again", + .keep_mapped = 1, + }, + { + .addr = HIGH_ADDR, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED, + .msg = "mmap(HIGH_ADDR, MAP_FIXED)", + }, + { + .addr = (void *) -1, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(-1)", + .keep_mapped = 1, + }, + { + .addr = (void *) -1, + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(-1) again", + }, + { + .addr = ((void *)(ADDR_SWITCH_HINT - PAGE_SIZE)), + .size = PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE, PAGE_SIZE)", + .low_addr_required = 1, + }, + { + .addr = (void *)(ADDR_SWITCH_HINT - PAGE_SIZE), + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE, 2 * PAGE_SIZE)", + .low_addr_required = 1, + .keep_mapped = 1, + }, + { + .addr = (void *)(ADDR_SWITCH_HINT - PAGE_SIZE / 2), + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE/2 , 2 * PAGE_SIZE)", + .low_addr_required = 1, + .keep_mapped = 1, + }, + { + .addr = ((void *)(ADDR_SWITCH_HINT)), + .size = PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT, PAGE_SIZE)", + }, + { + .addr = (void *)(ADDR_SWITCH_HINT), + .size = 2 * PAGE_SIZE, + .flags = MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED, + .msg = "mmap(ADDR_SWITCH_HINT, 2 * PAGE_SIZE, MAP_FIXED)", + }, +}; + +static struct testcase hugetlb_testcases[] = { + { + .addr = NULL, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(NULL, MAP_HUGETLB)", + .low_addr_required = 1, + }, + { + .addr = LOW_ADDR, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(LOW_ADDR, MAP_HUGETLB)", + .low_addr_required = 1, + }, + { + .addr = HIGH_ADDR, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(HIGH_ADDR, MAP_HUGETLB)", + .keep_mapped = 1, + }, + { + .addr = HIGH_ADDR, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(HIGH_ADDR, MAP_HUGETLB) again", + .keep_mapped = 1, + }, + { + .addr = HIGH_ADDR, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED, + .msg = "mmap(HIGH_ADDR, MAP_FIXED | MAP_HUGETLB)", + }, + { + .addr = (void *) -1, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(-1, MAP_HUGETLB)", + .keep_mapped = 1, + }, + { + .addr = (void *) -1, + .size = HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(-1, MAP_HUGETLB) again", + }, + { + .addr = (void *)(ADDR_SWITCH_HINT - PAGE_SIZE), + .size = 2 * HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS, + .msg = "mmap(ADDR_SWITCH_HINT - PAGE_SIZE, 2*HUGETLB_SIZE, MAP_HUGETLB)", + .low_addr_required = 1, + .keep_mapped = 1, + }, + { + .addr = (void *)(ADDR_SWITCH_HINT), + .size = 2 * HUGETLB_SIZE, + .flags = MAP_HUGETLB | MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED, + .msg = "mmap(ADDR_SWITCH_HINT , 2*HUGETLB_SIZE, MAP_FIXED | MAP_HUGETLB)", + }, +}; + +static int run_test(struct testcase *test, int count) +{ + void *p; + int i, ret = KSFT_PASS; + + for (i = 0; i < count; i++) { + struct testcase *t = test + i; + + p = mmap(t->addr, t->size, PROT_READ | PROT_WRITE, t->flags, -1, 0); + + printf("%s: %p - ", t->msg, p); + + if (p == MAP_FAILED) { + printf("FAILED\n"); + ret = KSFT_FAIL; + continue; + } + + if (t->low_addr_required && p >= (void *)(ADDR_SWITCH_HINT)) { + printf("FAILED\n"); + ret = KSFT_FAIL; + } else { + /* + * Do a dereference of the address returned so that we catch + * bugs in page fault handling + */ + memset(p, 0, t->size); + printf("OK\n"); + } + if (!t->keep_mapped) + munmap(p, t->size); + } + + return ret; +} + +static int supported_arch(void) +{ +#if defined(__powerpc64__) + return 1; +#elif defined(__x86_64__) + return 1; +#else + return 0; +#endif +} + +int main(int argc, char **argv) +{ + int ret; + + if (!supported_arch()) + return KSFT_SKIP; + + ret = run_test(testcases, ARRAY_SIZE(testcases)); + if (argc == 2 && !strcmp(argv[1], "--run-hugetlb")) + ret = run_test(hugetlb_testcases, ARRAY_SIZE(hugetlb_testcases)); + return ret; +} diff --git a/tools/testing/selftests/vm/va_128TBswitch.sh b/tools/testing/selftests/vm/va_128TBswitch.sh new file mode 100755 index 000000000..41580751d --- /dev/null +++ b/tools/testing/selftests/vm/va_128TBswitch.sh @@ -0,0 +1,54 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# +# Copyright (C) 2022 Adam Sindelar (Meta) +# +# This is a test for mmap behavior with 5-level paging. This script wraps the +# real test to check that the kernel is configured to support at least 5 +# pagetable levels. + +# 1 means the test failed +exitcode=1 + +# Kselftest framework requirement - SKIP code is 4. +ksft_skip=4 + +fail() +{ + echo "$1" + exit $exitcode +} + +check_supported_x86_64() +{ + local config="/proc/config.gz" + [[ -f "${config}" ]] || config="/boot/config-$(uname -r)" + [[ -f "${config}" ]] || fail "Cannot find kernel config in /proc or /boot" + + # gzip -dcfq automatically handles both compressed and plaintext input. + # See man 1 gzip under '-f'. + local pg_table_levels=$(gzip -dcfq "${config}" | grep PGTABLE_LEVELS | cut -d'=' -f 2) + + if [[ "${pg_table_levels}" -lt 5 ]]; then + echo "$0: PGTABLE_LEVELS=${pg_table_levels}, must be >= 5 to run this test" + exit $ksft_skip + fi +} + +check_test_requirements() +{ + # The test supports x86_64 and powerpc64. We currently have no useful + # eligibility check for powerpc64, and the test itself will reject other + # architectures. + case `uname -m` in + "x86_64") + check_supported_x86_64 + ;; + *) + return 0 + ;; + esac +} + +check_test_requirements +./va_128TBswitch diff --git a/tools/testing/selftests/vm/virtual_address_range.c b/tools/testing/selftests/vm/virtual_address_range.c new file mode 100644 index 000000000..c0592646e --- /dev/null +++ b/tools/testing/selftests/vm/virtual_address_range.c @@ -0,0 +1,139 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright 2017, Anshuman Khandual, IBM Corp. + * + * Works on architectures which support 128TB virtual + * address range and beyond. + */ +#include +#include +#include +#include +#include +#include +#include + +/* + * Maximum address range mapped with a single mmap() + * call is little bit more than 16GB. Hence 16GB is + * chosen as the single chunk size for address space + * mapping. + */ +#define MAP_CHUNK_SIZE 17179869184UL /* 16GB */ + +/* + * Address space till 128TB is mapped without any hint + * and is enabled by default. Address space beyond 128TB + * till 512TB is obtained by passing hint address as the + * first argument into mmap() system call. + * + * The process heap address space is divided into two + * different areas one below 128TB and one above 128TB + * till it reaches 512TB. One with size 128TB and the + * other being 384TB. + * + * On Arm64 the address space is 256TB and no high mappings + * are supported so far. + */ + +#define NR_CHUNKS_128TB 8192UL /* Number of 16GB chunks for 128TB */ +#define NR_CHUNKS_256TB (NR_CHUNKS_128TB * 2UL) +#define NR_CHUNKS_384TB (NR_CHUNKS_128TB * 3UL) + +#define ADDR_MARK_128TB (1UL << 47) /* First address beyond 128TB */ +#define ADDR_MARK_256TB (1UL << 48) /* First address beyond 256TB */ + +#ifdef __aarch64__ +#define HIGH_ADDR_MARK ADDR_MARK_256TB +#define HIGH_ADDR_SHIFT 49 +#define NR_CHUNKS_LOW NR_CHUNKS_256TB +#define NR_CHUNKS_HIGH 0 +#else +#define HIGH_ADDR_MARK ADDR_MARK_128TB +#define HIGH_ADDR_SHIFT 48 +#define NR_CHUNKS_LOW NR_CHUNKS_128TB +#define NR_CHUNKS_HIGH NR_CHUNKS_384TB +#endif + +static char *hind_addr(void) +{ + int bits = HIGH_ADDR_SHIFT + rand() % (63 - HIGH_ADDR_SHIFT); + + return (char *) (1UL << bits); +} + +static int validate_addr(char *ptr, int high_addr) +{ + unsigned long addr = (unsigned long) ptr; + + if (high_addr) { + if (addr < HIGH_ADDR_MARK) { + printf("Bad address %lx\n", addr); + return 1; + } + return 0; + } + + if (addr > HIGH_ADDR_MARK) { + printf("Bad address %lx\n", addr); + return 1; + } + return 0; +} + +static int validate_lower_address_hint(void) +{ + char *ptr; + + ptr = mmap((void *) (1UL << 45), MAP_CHUNK_SIZE, PROT_READ | + PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + + if (ptr == MAP_FAILED) + return 0; + + return 1; +} + +int main(int argc, char *argv[]) +{ + char *ptr[NR_CHUNKS_LOW]; + char *hptr[NR_CHUNKS_HIGH]; + char *hint; + unsigned long i, lchunks, hchunks; + + for (i = 0; i < NR_CHUNKS_LOW; i++) { + ptr[i] = mmap(NULL, MAP_CHUNK_SIZE, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + + if (ptr[i] == MAP_FAILED) { + if (validate_lower_address_hint()) + return 1; + break; + } + + if (validate_addr(ptr[i], 0)) + return 1; + } + lchunks = i; + + for (i = 0; i < NR_CHUNKS_HIGH; i++) { + hint = hind_addr(); + hptr[i] = mmap(hint, MAP_CHUNK_SIZE, PROT_READ | PROT_WRITE, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + + if (hptr[i] == MAP_FAILED) + break; + + if (validate_addr(hptr[i], 1)) + return 1; + } + hchunks = i; + + for (i = 0; i < lchunks; i++) + munmap(ptr[i], MAP_CHUNK_SIZE); + + for (i = 0; i < hchunks; i++) + munmap(hptr[i], MAP_CHUNK_SIZE); + + return 0; +} diff --git a/tools/testing/selftests/vm/vm_util.c b/tools/testing/selftests/vm/vm_util.c new file mode 100644 index 000000000..f11f8adda --- /dev/null +++ b/tools/testing/selftests/vm/vm_util.c @@ -0,0 +1,126 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include "../kselftest.h" +#include "vm_util.h" + +#define PMD_SIZE_FILE_PATH "/sys/kernel/mm/transparent_hugepage/hpage_pmd_size" +#define SMAP_FILE_PATH "/proc/self/smaps" +#define MAX_LINE_LENGTH 500 + +uint64_t pagemap_get_entry(int fd, char *start) +{ + const unsigned long pfn = (unsigned long)start / getpagesize(); + uint64_t entry; + int ret; + + ret = pread(fd, &entry, sizeof(entry), pfn * sizeof(entry)); + if (ret != sizeof(entry)) + ksft_exit_fail_msg("reading pagemap failed\n"); + return entry; +} + +bool pagemap_is_softdirty(int fd, char *start) +{ + uint64_t entry = pagemap_get_entry(fd, start); + + // Check if dirty bit (55th bit) is set + return entry & 0x0080000000000000ull; +} + +void clear_softdirty(void) +{ + int ret; + const char *ctrl = "4"; + int fd = open("/proc/self/clear_refs", O_WRONLY); + + if (fd < 0) + ksft_exit_fail_msg("opening clear_refs failed\n"); + ret = write(fd, ctrl, strlen(ctrl)); + close(fd); + if (ret != strlen(ctrl)) + ksft_exit_fail_msg("writing clear_refs failed\n"); +} + +bool check_for_pattern(FILE *fp, const char *pattern, char *buf, size_t len) +{ + while (fgets(buf, len, fp)) { + if (!strncmp(buf, pattern, strlen(pattern))) + return true; + } + return false; +} + +uint64_t read_pmd_pagesize(void) +{ + int fd; + char buf[20]; + ssize_t num_read; + + fd = open(PMD_SIZE_FILE_PATH, O_RDONLY); + if (fd == -1) + ksft_exit_fail_msg("Open hpage_pmd_size failed\n"); + + num_read = read(fd, buf, 19); + if (num_read < 1) { + close(fd); + ksft_exit_fail_msg("Read hpage_pmd_size failed\n"); + } + buf[num_read] = '\0'; + close(fd); + + return strtoul(buf, NULL, 10); +} + +bool __check_huge(void *addr, char *pattern, int nr_hpages, + uint64_t hpage_size) +{ + uint64_t thp = -1; + int ret; + FILE *fp; + char buffer[MAX_LINE_LENGTH]; + char addr_pattern[MAX_LINE_LENGTH]; + + ret = snprintf(addr_pattern, MAX_LINE_LENGTH, "%08lx-", + (unsigned long) addr); + if (ret >= MAX_LINE_LENGTH) + ksft_exit_fail_msg("%s: Pattern is too long\n", __func__); + + fp = fopen(SMAP_FILE_PATH, "r"); + if (!fp) + ksft_exit_fail_msg("%s: Failed to open file %s\n", __func__, SMAP_FILE_PATH); + + if (!check_for_pattern(fp, addr_pattern, buffer, sizeof(buffer))) + goto err_out; + + /* + * Fetch the pattern in the same block and check the number of + * hugepages. + */ + if (!check_for_pattern(fp, pattern, buffer, sizeof(buffer))) + goto err_out; + + snprintf(addr_pattern, MAX_LINE_LENGTH, "%s%%9ld kB", pattern); + + if (sscanf(buffer, addr_pattern, &thp) != 1) + ksft_exit_fail_msg("Reading smap error\n"); + +err_out: + fclose(fp); + return thp == (nr_hpages * (hpage_size >> 10)); +} + +bool check_huge_anon(void *addr, int nr_hpages, uint64_t hpage_size) +{ + return __check_huge(addr, "AnonHugePages: ", nr_hpages, hpage_size); +} + +bool check_huge_file(void *addr, int nr_hpages, uint64_t hpage_size) +{ + return __check_huge(addr, "FilePmdMapped:", nr_hpages, hpage_size); +} + +bool check_huge_shmem(void *addr, int nr_hpages, uint64_t hpage_size) +{ + return __check_huge(addr, "ShmemPmdMapped:", nr_hpages, hpage_size); +} diff --git a/tools/testing/selftests/vm/vm_util.h b/tools/testing/selftests/vm/vm_util.h new file mode 100644 index 000000000..5c35de454 --- /dev/null +++ b/tools/testing/selftests/vm/vm_util.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include +#include + +uint64_t pagemap_get_entry(int fd, char *start); +bool pagemap_is_softdirty(int fd, char *start); +void clear_softdirty(void); +bool check_for_pattern(FILE *fp, const char *pattern, char *buf, size_t len); +uint64_t read_pmd_pagesize(void); +bool check_huge_anon(void *addr, int nr_hpages, uint64_t hpage_size); +bool check_huge_file(void *addr, int nr_hpages, uint64_t hpage_size); +bool check_huge_shmem(void *addr, int nr_hpages, uint64_t hpage_size); diff --git a/tools/testing/selftests/vm/write_hugetlb_memory.sh b/tools/testing/selftests/vm/write_hugetlb_memory.sh new file mode 100644 index 000000000..70a02301f --- /dev/null +++ b/tools/testing/selftests/vm/write_hugetlb_memory.sh @@ -0,0 +1,23 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 + +set -e + +size=$1 +populate=$2 +write=$3 +cgroup=$4 +path=$5 +method=$6 +private=$7 +want_sleep=$8 +reserve=$9 + +echo "Putting task in cgroup '$cgroup'" +echo $$ > ${cgroup_path:-/dev/cgroup/memory}/"$cgroup"/cgroup.procs + +echo "Method is $method" + +set +e +./write_to_hugetlbfs -p "$path" -s "$size" "$write" "$populate" -m "$method" \ + "$private" "$want_sleep" "$reserve" diff --git a/tools/testing/selftests/vm/write_to_hugetlbfs.c b/tools/testing/selftests/vm/write_to_hugetlbfs.c new file mode 100644 index 000000000..6a2caba19 --- /dev/null +++ b/tools/testing/selftests/vm/write_to_hugetlbfs.c @@ -0,0 +1,240 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * This program reserves and uses hugetlb memory, supporting a bunch of + * scenarios needed by the charged_reserved_hugetlb.sh test. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* Global definitions. */ +enum method { + HUGETLBFS, + MMAP_MAP_HUGETLB, + SHM, + MAX_METHOD +}; + + +/* Global variables. */ +static const char *self; +static char *shmaddr; +static int shmid; + +/* + * Show usage and exit. + */ +static void exit_usage(void) +{ + printf("Usage: %s -p -s " + "[-m <0=hugetlbfs | 1=mmap(MAP_HUGETLB)>] [-l] [-r] " + "[-o] [-w] [-n]\n", + self); + exit(EXIT_FAILURE); +} + +void sig_handler(int signo) +{ + printf("Received %d.\n", signo); + if (signo == SIGINT) { + printf("Deleting the memory\n"); + if (shmdt((const void *)shmaddr) != 0) { + perror("Detach failure"); + shmctl(shmid, IPC_RMID, NULL); + exit(4); + } + + shmctl(shmid, IPC_RMID, NULL); + printf("Done deleting the memory\n"); + } + exit(2); +} + +int main(int argc, char **argv) +{ + int fd = 0; + int key = 0; + int *ptr = NULL; + int c = 0; + int size = 0; + char path[256] = ""; + enum method method = MAX_METHOD; + int want_sleep = 0, private = 0; + int populate = 0; + int write = 0; + int reserve = 1; + + if (signal(SIGINT, sig_handler) == SIG_ERR) + err(1, "\ncan't catch SIGINT\n"); + + /* Parse command-line arguments. */ + setvbuf(stdout, NULL, _IONBF, 0); + self = argv[0]; + + while ((c = getopt(argc, argv, "s:p:m:owlrn")) != -1) { + switch (c) { + case 's': + size = atoi(optarg); + break; + case 'p': + strncpy(path, optarg, sizeof(path)); + break; + case 'm': + if (atoi(optarg) >= MAX_METHOD) { + errno = EINVAL; + perror("Invalid -m."); + exit_usage(); + } + method = atoi(optarg); + break; + case 'o': + populate = 1; + break; + case 'w': + write = 1; + break; + case 'l': + want_sleep = 1; + break; + case 'r': + private + = 1; + break; + case 'n': + reserve = 0; + break; + default: + errno = EINVAL; + perror("Invalid arg"); + exit_usage(); + } + } + + if (strncmp(path, "", sizeof(path)) != 0) { + printf("Writing to this path: %s\n", path); + } else { + errno = EINVAL; + perror("path not found"); + exit_usage(); + } + + if (size != 0) { + printf("Writing this size: %d\n", size); + } else { + errno = EINVAL; + perror("size not found"); + exit_usage(); + } + + if (!populate) + printf("Not populating.\n"); + else + printf("Populating.\n"); + + if (!write) + printf("Not writing to memory.\n"); + + if (method == MAX_METHOD) { + errno = EINVAL; + perror("-m Invalid"); + exit_usage(); + } else + printf("Using method=%d\n", method); + + if (!private) + printf("Shared mapping.\n"); + else + printf("Private mapping.\n"); + + if (!reserve) + printf("NO_RESERVE mapping.\n"); + else + printf("RESERVE mapping.\n"); + + switch (method) { + case HUGETLBFS: + printf("Allocating using HUGETLBFS.\n"); + fd = open(path, O_CREAT | O_RDWR, 0777); + if (fd == -1) + err(1, "Failed to open file."); + + ptr = mmap(NULL, size, PROT_READ | PROT_WRITE, + (private ? MAP_PRIVATE : MAP_SHARED) | + (populate ? MAP_POPULATE : 0) | + (reserve ? 0 : MAP_NORESERVE), + fd, 0); + + if (ptr == MAP_FAILED) { + close(fd); + err(1, "Error mapping the file"); + } + break; + case MMAP_MAP_HUGETLB: + printf("Allocating using MAP_HUGETLB.\n"); + ptr = mmap(NULL, size, PROT_READ | PROT_WRITE, + (private ? (MAP_PRIVATE | MAP_ANONYMOUS) : + MAP_SHARED) | + MAP_HUGETLB | (populate ? MAP_POPULATE : 0) | + (reserve ? 0 : MAP_NORESERVE), + -1, 0); + + if (ptr == MAP_FAILED) + err(1, "mmap"); + + printf("Returned address is %p\n", ptr); + break; + case SHM: + printf("Allocating using SHM.\n"); + shmid = shmget(key, size, + SHM_HUGETLB | IPC_CREAT | SHM_R | SHM_W); + if (shmid < 0) { + shmid = shmget(++key, size, + SHM_HUGETLB | IPC_CREAT | SHM_R | SHM_W); + if (shmid < 0) + err(1, "shmget"); + } + printf("shmid: 0x%x, shmget key:%d\n", shmid, key); + + ptr = shmat(shmid, NULL, 0); + if (ptr == (int *)-1) { + perror("Shared memory attach failure"); + shmctl(shmid, IPC_RMID, NULL); + exit(2); + } + printf("shmaddr: %p\n", ptr); + + break; + default: + errno = EINVAL; + err(1, "Invalid method."); + } + + if (write) { + printf("Writing to memory.\n"); + memset(ptr, 1, size); + } + + if (want_sleep) { + /* Signal to caller that we're done. */ + printf("DONE\n"); + + /* Hold memory until external kill signal is delivered. */ + while (1) + sleep(100); + } + + if (method == HUGETLBFS) + close(fd); + + return 0; +} -- cgit v1.2.3