Re: [PATCH v3 2/2] selftests: KVM: use dirty logging to check if page stats work correctly

From: Ben Gardon
Date: Mon Aug 30 2021 - 17:10:25 EST


On Sun, Aug 29, 2021 at 9:44 PM Mingwei Zhang <mizhang@xxxxxxxxxx> wrote:
>
> When dirty logging is enabled, KVM splits the hugepage mapping in NPT/EPT
> into the smallest 4K size after guest VM writes to it. This property could
> be used to check if the page stats metrics work properly in KVM x86/mmu. At
> the same time, this logic might be used the other way around: using page
> stats to verify if dirty logging really splits all huge pages after guest
> VM writes to all memory.
>
> So add page stats checking in dirty logging performance selftest. In
> particular, add checks in three locations:
> - just after vm is created;
> - after populating memory into vm without turning on dirty logging;
> - after guest vm writing to all memory again with dirty logging turned on.
>
> Tested using commands:
> - ./dirty_log_perf_test -s anonymous_hugetlb_1gb
> - ./dirty_log_perf_test -s anonymous_hugetlb_2mb
> - ./dirty_log_perf_test -s anonymous_thp
>
> Cc: Sean Christopherson <seanjc@xxxxxxxxxx>
> Cc: David Matlack <dmatlack@xxxxxxxxxx>
> Cc: Jing Zhang <jingzhangos@xxxxxxxxxx>
> Cc: Peter Xu <peterx@xxxxxxxxxx>
>
> Suggested-by: Ben Gardon <bgardon@xxxxxxxxxx>
> Signed-off-by: Mingwei Zhang <mizhang@xxxxxxxxxx>
> ---
> .../selftests/kvm/dirty_log_perf_test.c | 44 +++++++++++++++++++
> .../testing/selftests/kvm/include/test_util.h | 1 +
> .../selftests/kvm/include/x86_64/processor.h | 7 +++
> tools/testing/selftests/kvm/lib/test_util.c | 29 ++++++++++++
> 4 files changed, 81 insertions(+)
>
> diff --git a/tools/testing/selftests/kvm/dirty_log_perf_test.c b/tools/testing/selftests/kvm/dirty_log_perf_test.c
> index 3c30d0045d8d..bc598e07b295 100644
> --- a/tools/testing/selftests/kvm/dirty_log_perf_test.c
> +++ b/tools/testing/selftests/kvm/dirty_log_perf_test.c
> @@ -19,6 +19,10 @@
> #include "perf_test_util.h"
> #include "guest_modes.h"
>
> +#ifdef __x86_64__
> +#include "processor.h"
> +#endif
> +

I know this is only needed for x86_64, but I don't think it needs to
be ifdef in order for everything to work.

> /* How many host loops to run by default (one KVM_GET_DIRTY_LOG for each loop)*/
> #define TEST_HOST_LOOP_N 2UL
>
> @@ -166,6 +170,18 @@ static void run_test(enum vm_guest_mode mode, void *arg)
> vm = perf_test_create_vm(mode, nr_vcpus, guest_percpu_mem_size,
> p->slots, p->backing_src);
>
> +#ifdef __x86_64__
> + /*
> + * No vCPUs have been started yet, so KVM should not have created any
> + * mapping at this moment.
> + */
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_4K) == 0,
> + "4K page is non zero");
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_2M) == 0,
> + "2M page is non zero");
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_1G) == 0,
> + "1G page is non zero");
> +#endif
> perf_test_args.wr_fract = p->wr_fract;
>
> guest_num_pages = (nr_vcpus * guest_percpu_mem_size) >> vm_get_page_shift(vm);
> @@ -211,6 +227,22 @@ static void run_test(enum vm_guest_mode mode, void *arg)
> pr_info("Populate memory time: %ld.%.9lds\n",
> ts_diff.tv_sec, ts_diff.tv_nsec);
>
> +#ifdef __x86_64__
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_4K) != 0,
> + "4K page is zero");
> + /* Ensure THP page stats is non-zero to minimize the flakiness. */
> + if (p->backing_src == VM_MEM_SRC_ANONYMOUS_THP)
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_2M) > 0,
> + "2M page number is zero");

Nice. Thanks for handling this case. It's very frustrating when THP
makes test assertions flaky.

> + else if (p->backing_src == VM_MEM_SRC_ANONYMOUS_HUGETLB_2MB)
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_2M) ==
> + (guest_percpu_mem_size * nr_vcpus) >> X86_PAGE_2M_SHIFT,
> + "2M page number does not match");
> + else if (p->backing_src == VM_MEM_SRC_ANONYMOUS_HUGETLB_1GB)
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_1G) ==
> + (guest_percpu_mem_size * nr_vcpus) >> X86_PAGE_1G_SHIFT,
> + "1G page number does not match");
> +#endif
> /* Enable dirty logging */
> clock_gettime(CLOCK_MONOTONIC, &start);
> enable_dirty_logging(vm, p->slots);
> @@ -256,6 +288,18 @@ static void run_test(enum vm_guest_mode mode, void *arg)
> iteration, ts_diff.tv_sec, ts_diff.tv_nsec);
> }
> }
> +#ifdef __x86_64__
> + /*
> + * When vCPUs writes to all memory again with dirty logging enabled, we
> + * should see only 4K page mappings exist in KVM mmu.
> + */
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_4K) != 0,
> + "4K page is zero after dirtying memory");
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_2M) == 0,
> + "2M page is non-zero after dirtying memory");
> + TEST_ASSERT(get_page_stats(X86_PAGE_SIZE_1G) == 0,
> + "1G page is non-zero after dirtying memory");
> +#endif
>
> /* Disable dirty logging */
> clock_gettime(CLOCK_MONOTONIC, &start);
> diff --git a/tools/testing/selftests/kvm/include/test_util.h b/tools/testing/selftests/kvm/include/test_util.h
> index d79be15dd3d2..dca5fcf7aa87 100644
> --- a/tools/testing/selftests/kvm/include/test_util.h
> +++ b/tools/testing/selftests/kvm/include/test_util.h
> @@ -102,6 +102,7 @@ const struct vm_mem_backing_src_alias *vm_mem_backing_src_alias(uint32_t i);
> size_t get_backing_src_pagesz(uint32_t i);
> void backing_src_help(void);
> enum vm_mem_backing_src_type parse_backing_src_type(const char *type_name);
> +size_t get_page_stats(uint32_t page_level);
>
> /*
> * Whether or not the given source type is shared memory (as opposed to
> diff --git a/tools/testing/selftests/kvm/include/x86_64/processor.h b/tools/testing/selftests/kvm/include/x86_64/processor.h
> index 242ae8e09a65..9749319821a3 100644
> --- a/tools/testing/selftests/kvm/include/x86_64/processor.h
> +++ b/tools/testing/selftests/kvm/include/x86_64/processor.h
> @@ -39,6 +39,13 @@
> #define X86_CR4_SMAP (1ul << 21)
> #define X86_CR4_PKE (1ul << 22)
>
> +#define X86_PAGE_4K_SHIFT 12
> +#define X86_PAGE_4K (1ul << X86_PAGE_4K_SHIFT)
> +#define X86_PAGE_2M_SHIFT 21
> +#define X86_PAGE_2M (1ul << X86_PAGE_2M_SHIFT)
> +#define X86_PAGE_1G_SHIFT 30
> +#define X86_PAGE_1G (1ul << X86_PAGE_1G_SHIFT)
> +

It would be a nice follow up to use these to clean up the magic
numbers in __virt_pg_map:

const uint64_t pg_size = 1ull << ((page_size * 9) + 12);

:(

> /* CPUID.1.ECX */
> #define CPUID_VMX (1ul << 5)
> #define CPUID_SMX (1ul << 6)
> diff --git a/tools/testing/selftests/kvm/lib/test_util.c b/tools/testing/selftests/kvm/lib/test_util.c
> index af1031fed97f..07eb6b5c125e 100644
> --- a/tools/testing/selftests/kvm/lib/test_util.c
> +++ b/tools/testing/selftests/kvm/lib/test_util.c
> @@ -15,6 +15,13 @@
> #include "linux/kernel.h"
>
> #include "test_util.h"
> +#include "processor.h"
> +
> +static const char * const pagestat_filepaths[] = {
> + "/sys/kernel/debug/kvm/pages_4k",
> + "/sys/kernel/debug/kvm/pages_2m",
> + "/sys/kernel/debug/kvm/pages_1g",
> +};

I think these should only be defined for x86_64 too. Is this the right
file for these definitions or is there an arch specific file they
should go in?

>
> /*
> * Parses "[0-9]+[kmgt]?".
> @@ -141,6 +148,28 @@ size_t get_trans_hugepagesz(void)
> return size;
> }
>
> +#ifdef __x86_64__
> +size_t get_stats_from_file(const char *path)
> +{
> + size_t value;
> + FILE *f;
> +
> + f = fopen(path, "r");
> + TEST_ASSERT(f != NULL, "Error in opening file: %s\n", path);
> +
> + fscanf(f, "%ld", &value);
> + fclose(f);
> +
> + return value;
> +}
> +
> +size_t get_page_stats(uint32_t page_level)
> +{
> + TEST_ASSERT(page_level <= X86_PAGE_SIZE_1G, "page type error.");
> + return get_stats_from_file(pagestat_filepaths[page_level]);
> +}
> +#endif
> +
> size_t get_def_hugetlb_pagesz(void)
> {
> char buf[64];
> --
> 2.33.0.259.gc128427fd7-goog
>