| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <bd09a55c777c9342e950f8cfbd9e5f8c1cdecd23.camel@redhat.com>
Date: Wed, 26 Mar 2025 15:38:12 -0400
From: Maxim Levitsky <mlevitsk@...hat.com>
To: James Houghton <jthoughton@...gle.com>
Cc: kvm@...r.kernel.org, Muhammad Usama Anjum <usama.anjum@...labora.com>,
linux-kernel@...r.kernel.org, Sean Christopherson <seanjc@...gle.com>,
Shuah Khan <shuah@...nel.org>, Claudio Imbrenda <imbrenda@...ux.ibm.com>,
Oliver Upton <oliver.upton@...ux.dev>, Paolo Bonzini <pbonzini@...hat.com>,
linux-kselftest@...r.kernel.org, Anup Patel <anup@...infault.org>
Subject: Re: [PATCH v2 2/2] KVM: selftests: access_tracking_perf_test: add
option to skip the sanity check
On Tue, 2025-03-25 at 11:01 -0700, James Houghton wrote:
> On Mon, Mar 24, 2025 at 6:57 PM Maxim Levitsky <mlevitsk@...hat.com> wrote:
> > Add an option to skip sanity check of number of still idle pages,
> > and set it by default to skip, in case hypervisor or NUMA balancing
> > is detected.
> >
> > Signed-off-by: Maxim Levitsky <mlevitsk@...hat.com>
>
> Thanks Maxim! I'm still working on a respin of this test with MGLRU
> integration, like [1]. Sorry it's taking me so long. I'll apply my
> changes on top of yours.
>
> [1]: https://lore.kernel.org/kvm/20241105184333.2305744-12-jthoughton@google.com/
>
> > ---
> > .../selftests/kvm/access_tracking_perf_test.c | 33 ++++++++++++++++---
> > .../testing/selftests/kvm/include/test_util.h | 1 +
> > tools/testing/selftests/kvm/lib/test_util.c | 7 ++++
> > 3 files changed, 37 insertions(+), 4 deletions(-)
> >
> > diff --git a/tools/testing/selftests/kvm/access_tracking_perf_test.c b/tools/testing/selftests/kvm/access_tracking_perf_test.c
> > index 3c7defd34f56..6d50c829f00c 100644
> > --- a/tools/testing/selftests/kvm/access_tracking_perf_test.c
> > +++ b/tools/testing/selftests/kvm/access_tracking_perf_test.c
> > @@ -65,6 +65,8 @@ static int vcpu_last_completed_iteration[KVM_MAX_VCPUS];
> > /* Whether to overlap the regions of memory vCPUs access. */
> > static bool overlap_memory_access;
> >
> > +static int warn_on_too_many_idle_pages = -1;
> > +
> > struct test_params {
> > /* The backing source for the region of memory. */
> > enum vm_mem_backing_src_type backing_src;
> > @@ -184,11 +186,10 @@ static void mark_vcpu_memory_idle(struct kvm_vm *vm,
> > * are cached and the guest won't see the "idle" bit cleared.
> > */
> > if (still_idle >= pages / 10) {
> > -#ifdef __x86_64__
> > - TEST_ASSERT(this_cpu_has(X86_FEATURE_HYPERVISOR),
> > + TEST_ASSERT(warn_on_too_many_idle_pages,
>
> I think this assertion is flipped (or how warn_on_too_many_idle_pages
> is being set is flipped, see below).
Yes it is no doubt about.
I didn't notice this when I flipped the meaning
of the variable as Sean suggested.
Thanks!
Best regards,
Maxim Levitsky
>
> > "vCPU%d: Too many pages still idle (%lu out of %lu)",
> > vcpu_idx, still_idle, pages);
> > -#endif
> > +
> > printf("WARNING: vCPU%d: Too many pages still idle (%lu out of %lu), "
> > "this will affect performance results.\n",
> > vcpu_idx, still_idle, pages);
> > @@ -342,6 +343,8 @@ static void help(char *name)
> > printf(" -v: specify the number of vCPUs to run.\n");
> > printf(" -o: Overlap guest memory accesses instead of partitioning\n"
> > " them into a separate region of memory for each vCPU.\n");
> > + printf(" -w: Skip or force enable the check that after dirtying the guest memory, most (90%%) of \n"
> > + "it is reported as dirty again (0/1)");
> > backing_src_help("-s");
> > puts("");
> > exit(0);
> > @@ -359,7 +362,7 @@ int main(int argc, char *argv[])
> >
> > guest_modes_append_default();
> >
> > - while ((opt = getopt(argc, argv, "hm:b:v:os:")) != -1) {
> > + while ((opt = getopt(argc, argv, "hm:b:v:os:w:")) != -1) {
> > switch (opt) {
> > case 'm':
> > guest_modes_cmdline(optarg);
> > @@ -376,6 +379,11 @@ int main(int argc, char *argv[])
> > case 's':
> > params.backing_src = parse_backing_src_type(optarg);
> > break;
> > + case 'w':
> > + warn_on_too_many_idle_pages =
> > + atoi_non_negative("1 - enable warning, 0 - disable",
> > + optarg);
>
> We still get a "warning" either way, right? Maybe this should be
> called "fail_on_too_many_idle_pages" (in which case the above
> assertion is indeed flipped). Or "warn_on_too_many_idle_pages" should
> mean *only* warn, i.e., *don't* fail, in which case, below we need to
> flip how we set it below.
>
> > + break;
> > case 'h':
> > default:
> > help(argv[0]);
> > @@ -386,6 +394,23 @@ int main(int argc, char *argv[])
> > page_idle_fd = open("/sys/kernel/mm/page_idle/bitmap", O_RDWR);
> > __TEST_REQUIRE(page_idle_fd >= 0,
> > "CONFIG_IDLE_PAGE_TRACKING is not enabled");
> > + if (warn_on_too_many_idle_pages == -1) {
> > +#ifdef __x86_64__
> > + if (this_cpu_has(X86_FEATURE_HYPERVISOR)) {
> > + printf("Skipping idle page count sanity check, because the test is run nested\n");
> > + warn_on_too_many_idle_pages = 0;
> > + } else
> > +#endif
> > + if (is_numa_balancing_enabled()) {
> > + printf("Skipping idle page count sanity check, because NUMA balance is enabled\n");
> > + warn_on_too_many_idle_pages = 0;
> > + } else {
> > + warn_on_too_many_idle_pages = 1;
> > + }
> > + } else if (!warn_on_too_many_idle_pages) {
> > + printf("Skipping idle page count sanity check, because this was requested by the user\n");
> > + }
> > +
> > close(page_idle_fd);
> >
> > for_each_guest_mode(run_test, ¶ms);
> > diff --git a/tools/testing/selftests/kvm/include/test_util.h b/tools/testing/selftests/kvm/include/test_util.h
> > index 3e473058849f..1bc9b0a92427 100644
> > --- a/tools/testing/selftests/kvm/include/test_util.h
> > +++ b/tools/testing/selftests/kvm/include/test_util.h
> > @@ -153,6 +153,7 @@ bool is_backing_src_hugetlb(uint32_t i);
> > void backing_src_help(const char *flag);
> > enum vm_mem_backing_src_type parse_backing_src_type(const char *type_name);
> > long get_run_delay(void);
> > +bool is_numa_balancing_enabled(void);
> >
> > /*
> > * Whether or not the given source type is shared memory (as opposed to
> > diff --git a/tools/testing/selftests/kvm/lib/test_util.c b/tools/testing/selftests/kvm/lib/test_util.c
> > index 3dc8538f5d69..03eb99af9b8d 100644
> > --- a/tools/testing/selftests/kvm/lib/test_util.c
> > +++ b/tools/testing/selftests/kvm/lib/test_util.c
> > @@ -176,6 +176,13 @@ size_t get_trans_hugepagesz(void)
> > return get_sysfs_val("/sys/kernel/mm/transparent_hugepage/hpage_pmd_size");
> > }
> >
> > +bool is_numa_balancing_enabled(void)
> > +{
> > + if (!test_sysfs_path("/proc/sys/kernel/numa_balancing"))
> > + return false;
> > + return get_sysfs_val("/proc/sys/kernel/numa_balancing") == 1;
> > +}
> > +
> > size_t get_def_hugetlb_pagesz(void)
> > {
> > char buf[64];
> > --
> > 2.26.3
> >
Powered by blists - more mailing lists