Linux-mm Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nico Pache <npache@redhat.com>
To: Baolin Wang <baolin.wang@linux.alibaba.com>,
	akpm@linux-foundation.org, david@kernel.org, ljs@kernel.org,
	hughd@google.com
Cc: willy@infradead.org, ziy@nvidia.com, liam@infradead.org,
	ryan.roberts@arm.com, dev.jain@arm.com, baohua@kernel.org,
	lance.yang@linux.dev, linux-mm@kvack.org,
	linux-kselftest@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH v2 11/11] selftests: mm: add mTHP collapse test cases
Date: Fri, 12 Jun 2026 04:41:18 -0600	[thread overview]
Message-ID: <4406fdcf-098e-4523-a441-5343bb22e667@redhat.com> (raw)
In-Reply-To: <fcf0fcc52aa6c4e18682fcb302fe9371a8d800c3.1781083630.git.baolin.wang@linux.alibaba.com>



On 6/10/26 4:29 AM, Baolin Wang wrote:
> Added a new command 'mthp_khugepaged' for mTHP collapse, along with the '-c'
> parameter to specify the collapse order. Additionally, added mTHP collapse
> test cases for 'collapse_full', 'collapse_empty', and 'collapse_single_mthp'
> for both anonymous pages and shmem. All khugepaged test cases passed.
> 
> Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>

Will rip out the Shmem stuff and test these selftests on top of my mTHP stuff
soon :)

Ill provide the diff so its easier for you and if I come up with more cases ill
add those too.

Thanks for doing this!!!
-- Nico

> ---
>  tools/testing/selftests/mm/khugepaged.c   | 135 +++++++++++++++++++---
>  tools/testing/selftests/mm/run_vmtests.sh |   4 +
>  2 files changed, 120 insertions(+), 19 deletions(-)
> 
> diff --git a/tools/testing/selftests/mm/khugepaged.c b/tools/testing/selftests/mm/khugepaged.c
> index f69be6be0ecd..8975be5b7b2f 100644
> --- a/tools/testing/selftests/mm/khugepaged.c
> +++ b/tools/testing/selftests/mm/khugepaged.c
> @@ -26,9 +26,11 @@
>  
>  #define BASE_ADDR ((void *)(1UL << 30))
>  static unsigned long hpage_pmd_size;
> +static int hpage_pmd_order;
>  static unsigned long page_size;
>  static int hpage_pmd_nr;
>  static int anon_order;
> +static int collapse_order;
>  
>  #define PID_SMAPS "/proc/self/smaps"
>  #define TEST_FILE "collapse_test_file"
> @@ -69,6 +71,7 @@ struct collapse_context {
>  };
>  
>  static struct collapse_context *khugepaged_context;
> +static struct collapse_context *mthp_khugepaged_context;
>  static struct collapse_context *madvise_context;
>  
>  struct file_info {
> @@ -554,25 +557,25 @@ static void madvise_collapse(const char *msg, char *p, int nr_hpages,
>  }
>  
>  #define TICK 500000
> -static bool wait_for_scan(const char *msg, char *p, int nr_hpages,
> -			  struct mem_ops *ops)
> +static bool wait_for_scan(const char *msg, char *p, unsigned long size,
> +		int nr_hpages, int collap_order, struct mem_ops *ops)
>  {
> -	unsigned long size = nr_hpages * hpage_pmd_size;
> +	unsigned long hpage_size = page_size << collap_order;
>  	int full_scans;
>  	int timeout = 6; /* 3 seconds */
>  
>  	/* Sanity check */
> -	if (!ops->check_huge(p, size, 0, hpage_pmd_size))
> +	if (!ops->check_huge(p, size, 0, hpage_size))
>  		ksft_exit_fail_msg("Unexpected huge page\n");
>  
> -	madvise(p, nr_hpages * hpage_pmd_size, MADV_HUGEPAGE);
> +	madvise(p, size, MADV_HUGEPAGE);
>  
>  	/* Wait until the second full_scan completed */
>  	full_scans = thp_read_num("khugepaged/full_scans") + 2;
>  
>  	ksft_print_msg("%s...", msg);
>  	while (timeout--) {
> -		if (ops->check_huge(p, size, nr_hpages, hpage_pmd_size))
> +		if (ops->check_huge(p, size, nr_hpages, hpage_size))
>  			break;
>  		if (thp_read_num("khugepaged/full_scans") >= full_scans)
>  			break;
> @@ -595,7 +598,7 @@ static void khugepaged_collapse(const char *msg, char *p, int nr_hpages,
>  	if (!is_tmpfs(ops) && ops == &__read_write_file_write_ops)
>  		expect = false;
>  
> -	if (wait_for_scan(msg, p, nr_hpages, ops)) {
> +	if (wait_for_scan(msg, p, size, nr_hpages, hpage_pmd_order, ops)) {
>  		if (expect)
>  			fail("Timeout");
>  		else
> @@ -617,12 +620,65 @@ static void khugepaged_collapse(const char *msg, char *p, int nr_hpages,
>  		fail("Fail");
>  }
>  
> +static void mthp_khugepaged_collapse(const char *msg, char *p, int nr_hpages,
> +				struct mem_ops *ops, bool expect)
> +{
> +	unsigned long hpage_size = page_size << collapse_order;
> +	struct thp_settings settings = *thp_current_settings();
> +	/* mTHP collpase only allocates PMD sized memory */
> +	unsigned long size = hpage_pmd_size;
> +
> +	/* Set mTHP setting for mTHP collapse */
> +	if (ops == &__anon_ops) {
> +		settings.thp_enabled = THP_NEVER;
> +		settings.hugepages[collapse_order].enabled = THP_ALWAYS;
> +	} else if (ops == &__shmem_ops) {
> +		settings.shmem_enabled = SHMEM_NEVER;
> +		settings.shmem_hugepages[collapse_order].enabled = SHMEM_ALWAYS;
> +	}
> +
> +	thp_push_settings(&settings);
> +
> +	if (wait_for_scan(msg, p, size, nr_hpages, collapse_order, ops)) {
> +		if (expect)
> +			fail("Timeout");
> +		else
> +			success("OK");
> +
> +		/* Restore THP settings for mTHP collapse. */
> +		thp_pop_settings();
> +		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_size);
> +
> +	if (ops->check_huge(p, size, expect ? nr_hpages : 0, hpage_size))
> +		success("OK");
> +	else
> +		fail("Fail");
> +
> +	/* Restore THP settings for mTHP collapse. */
> +	thp_pop_settings();
> +}
> +
>  static struct collapse_context __khugepaged_context = {
>  	.collapse = &khugepaged_collapse,
>  	.enforce_pte_scan_limits = true,
>  	.name = "khugepaged",
>  };
>  
> +static struct collapse_context __mthp_khugepaged_context = {
> +	.collapse = &mthp_khugepaged_collapse,
> +	.enforce_pte_scan_limits = true,
> +	.name = "mthp_khugepaged",
> +};
> +
>  static struct collapse_context __madvise_context = {
>  	.collapse = &madvise_collapse,
>  	.enforce_pte_scan_limits = false,
> @@ -661,10 +717,17 @@ static void alloc_at_fault(void)
>  static void collapse_full(struct collapse_context *c, struct mem_ops *ops)
>  {
>  	void *p;
> -	int nr_hpages = 4;
> +	int nr_pmds = 4, nr_hpages = 4;
>  	unsigned long size = nr_hpages * hpage_pmd_size;
>  
> -	p = ops->setup_area(nr_hpages);
> +	/* Only try 1 PMD sized range for mTHP collapse. */
> +	if (c == &__mthp_khugepaged_context) {
> +		nr_pmds = 1;
> +		nr_hpages = 1 << (hpage_pmd_order - collapse_order);
> +		size = hpage_pmd_size;
> +	}
> +
> +	p = ops->setup_area(nr_pmds);
>  	ops->fault(p, 0, size);
>  	c->collapse("Collapse multiple fully populated PTE table", p, nr_hpages,
>  		    ops, true);
> @@ -676,10 +739,31 @@ static void collapse_full(struct collapse_context *c, struct mem_ops *ops)
>  
>  static void collapse_empty(struct collapse_context *c, struct mem_ops *ops)
>  {
> +	int nr_hpages = 1;
> +	void *p;
> +
> +	if (c == &__mthp_khugepaged_context)
> +		nr_hpages = 1 << (hpage_pmd_order - collapse_order);
> +
> +	p = ops->setup_area(1);
> +	c->collapse("Do not collapse empty PTE table", p, nr_hpages, ops, false);
> +	ops->cleanup_area(p, hpage_pmd_size);
> +	ksft_test_result_report(exit_status, "%s\n", __func__);
> +}
> +
> +static void collapse_single_mthp(struct collapse_context *c, struct mem_ops *ops)
> +{
> +	unsigned long hpage_size = page_size << collapse_order;
>  	void *p;
>  
>  	p = ops->setup_area(1);
> -	c->collapse("Do not collapse empty PTE table", p, 1, ops, false);
> +	/*
> +	 * Only fault collapse_order sized ranges, and only check 1
> +	 * collapse_order sized huge page.
> +	 */
> +	ops->fault(p, 0, hpage_size);
> +	c->collapse("Collapse PTE table with half PTE entries present",
> +		p, 1, ops, true);
>  	ops->cleanup_area(p, hpage_pmd_size);
>  	ksft_test_result_report(exit_status, "%s\n", __func__);
>  }
> @@ -1081,8 +1165,8 @@ static void madvise_retracted_page_tables(struct collapse_context *c,
>  	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)) {
> +	if (wait_for_scan("Collapse and leave PMD cleared", p, size, nr_hpages,
> +			  hpage_pmd_order, ops)) {
>  		fail("Timeout");
>  		return;
>  	}
> @@ -1098,7 +1182,7 @@ static void usage(void)
>  {
>  	fprintf(stderr, "\nUsage: ./khugepaged [OPTIONS] <test type> [dir]\n\n");
>  	fprintf(stderr, "\t<test type>\t: <context>:<mem_type>\n");
> -	fprintf(stderr, "\t<context>\t: [all|khugepaged|madvise]\n");
> +	fprintf(stderr, "\t<context>\t: [all|khugepaged|mthp_khugepaged|madvise]\n");
>  	fprintf(stderr, "\t<mem_type>\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 a file system\n");
> @@ -1109,6 +1193,7 @@ static void usage(void)
>  	fprintf(stderr,	"\t\t-h: This help message.\n");
>  	fprintf(stderr,	"\t\t-s: mTHP size, expressed as page order.\n");
>  	fprintf(stderr,	"\t\t    Defaults to 0. Use this size for anon or shmem allocations.\n");
> +	fprintf(stderr,	"\t\t-c: collapse order for mTHP collapse, expressed as page order.\n");
>  	exit(1);
>  }
>  
> @@ -1118,11 +1203,14 @@ static void parse_test_type(int argc, char **argv)
>  	char *buf;
>  	const char *token;
>  
> -	while ((opt = getopt(argc, argv, "s:h")) != -1) {
> +	while ((opt = getopt(argc, argv, "s:c:h")) != -1) {
>  		switch (opt) {
>  		case 's':
>  			anon_order = atoi(optarg);
>  			break;
> +		case 'c':
> +			collapse_order = atoi(optarg);
> +			break;
>  		case 'h':
>  		default:
>  			usage();
> @@ -1148,6 +1236,10 @@ static void parse_test_type(int argc, char **argv)
>  		madvise_context =  &__madvise_context;
>  	} else if (!strcmp(token, "khugepaged")) {
>  		khugepaged_context =  &__khugepaged_context;
> +	} else if (!strcmp(token, "mthp_khugepaged")) {
> +		mthp_khugepaged_context =  &__mthp_khugepaged_context;
> +		if (collapse_order == 0 || collapse_order >= hpage_pmd_order)
> +			usage();
>  	} else if (!strcmp(token, "madvise")) {
>  		madvise_context =  &__madvise_context;
>  	} else {
> @@ -1213,7 +1305,6 @@ static int nr_test_cases;
>  
>  int main(int argc, char **argv)
>  {
> -	int hpage_pmd_order;
>  	struct thp_settings default_settings = {
>  		.thp_enabled = THP_MADVISE,
>  		.thp_defrag = THP_DEFRAG_ALWAYS,
> @@ -1239,10 +1330,6 @@ int main(int argc, char **argv)
>  	if (!thp_is_enabled())
>  		ksft_exit_skip("Transparent Hugepages not available\n");
>  
> -	parse_test_type(argc, argv);
> -
> -	setbuf(stdout, NULL);
> -
>  	page_size = getpagesize();
>  	hpage_pmd_size = read_pmd_pagesize();
>  	if (!hpage_pmd_size)
> @@ -1250,6 +1337,10 @@ int main(int argc, char **argv)
>  	hpage_pmd_nr = hpage_pmd_size / page_size;
>  	hpage_pmd_order = __builtin_ctz(hpage_pmd_nr);
>  
> +	parse_test_type(argc, argv);
> +
> +	setbuf(stdout, NULL);
> +
>  	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;
> @@ -1267,6 +1358,8 @@ int main(int argc, char **argv)
>  	TEST(collapse_full, khugepaged_context, read_write_file_read_ops);
>  	TEST(collapse_full, khugepaged_context, read_write_file_write_ops);
>  	TEST(collapse_full, khugepaged_context, shmem_ops);
> +	TEST(collapse_full, mthp_khugepaged_context, anon_ops);
> +	TEST(collapse_full, mthp_khugepaged_context, shmem_ops);
>  	TEST(collapse_full, madvise_context, anon_ops);
>  	TEST(collapse_full, madvise_context, read_only_file_ops);
>  	TEST(collapse_full, madvise_context, read_write_file_read_ops);
> @@ -1274,8 +1367,12 @@ int main(int argc, char **argv)
>  	TEST(collapse_full, madvise_context, shmem_ops);
>  
>  	TEST(collapse_empty, khugepaged_context, anon_ops);
> +	TEST(collapse_empty, mthp_khugepaged_context, anon_ops);
>  	TEST(collapse_empty, madvise_context, anon_ops);
>  
> +	TEST(collapse_single_mthp, mthp_khugepaged_context, anon_ops);
> +	TEST(collapse_single_mthp, mthp_khugepaged_context, shmem_ops);
> +
>  	TEST(collapse_single_pte_entry, khugepaged_context, anon_ops);
>  	TEST(collapse_single_pte_entry, khugepaged_context, read_only_file_ops);
>  	TEST(collapse_single_pte_entry, khugepaged_context, read_write_file_read_ops);
> diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh
> index 8c296dedf047..c0f4f3e5f1f1 100755
> --- a/tools/testing/selftests/mm/run_vmtests.sh
> +++ b/tools/testing/selftests/mm/run_vmtests.sh
> @@ -411,6 +411,10 @@ CATEGORY="thp" run_test ./khugepaged all:shmem
>  
>  CATEGORY="thp" run_test ./khugepaged -s 4 all:shmem
>  
> +CATEGORY="thp" run_test ./khugepaged -c 4 mthp_khugepaged:anon
> +
> +CATEGORY="thp" run_test ./khugepaged -c 4 mthp_khugepaged:shmem
> +
>  # Try to create XFS if not provided
>  if [ -z "${SPLIT_HUGE_PAGE_TEST_XFS_PATH}" ]; then
>      if test_selected "thp"; then



  reply	other threads:[~2026-06-12 10:40 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-10 10:29 [RFC PATCH v2 00/11] add shmem mTHP collapse support Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 01/11] mm: khugepaged: add max_ptes_none check in collapse_file() Baolin Wang
2026-06-12 10:10   ` Nico Pache
2026-06-10 10:29 ` [RFC PATCH v2 02/11] mm: khugepaged: generalize collapse_file() for shmem mTHP support Baolin Wang
2026-06-12 10:16   ` Nico Pache
2026-06-15  6:39     ` Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 03/11] mm: khugepaged: add an order check for PMD-sized THP statistics Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 04/11] mm: khugepaged: add shmem mTHP collapse support Baolin Wang
2026-06-10 12:13   ` Lance Yang
2026-06-11  0:31     ` Baolin Wang
2026-06-10 12:44   ` Lance Yang
2026-06-11  0:42     ` Baolin Wang
2026-06-11  2:47       ` Lance Yang
2026-06-11 11:15         ` Baolin Wang
2026-06-12 10:22     ` Nico Pache
2026-06-15  6:46       ` Baolin Wang
2026-06-11  6:06   ` Lance Yang
2026-06-11 12:11     ` Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 05/11] mm: shmem: run khugepaged for all shmem mTHP orders Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 06/11] mm: khugepaged: allow khugepaged to check all shmem mTHP-sized orders Baolin Wang
2026-06-10 11:33   ` Lance Yang
2026-06-11  0:47     ` Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 07/11] mm: khugepaged: skip large folios that don't need to be collapsed Baolin Wang
2026-06-11  4:59   ` Lance Yang
2026-06-11 11:29     ` Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 08/11] selftests: mm: extend the check_huge() to support mTHP check Baolin Wang
2026-06-12 10:32   ` Nico Pache
2026-06-15  6:51     ` Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 09/11] selftests: mm: move gather_after_split_folio_orders() into vm_util.c file Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 10/11] selftests: mm: implement the mTHP-sized hugepage check helpers Baolin Wang
2026-06-10 10:29 ` [RFC PATCH v2 11/11] selftests: mm: add mTHP collapse test cases Baolin Wang
2026-06-12 10:41   ` Nico Pache [this message]
2026-06-10 16:28 ` [RFC PATCH v2 00/11] add shmem mTHP collapse support Nico Pache
2026-06-11  0:52   ` Baolin Wang
2026-06-11  6:18 ` Lorenzo Stoakes
2026-06-11 11:46   ` Baolin Wang
2026-06-11 12:15     ` Lorenzo Stoakes

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4406fdcf-098e-4523-a441-5343bb22e667@redhat.com \
    --to=npache@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=baohua@kernel.org \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=david@kernel.org \
    --cc=dev.jain@arm.com \
    --cc=hughd@google.com \
    --cc=lance.yang@linux.dev \
    --cc=liam@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=ljs@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=willy@infradead.org \
    --cc=ziy@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox