* [PATCH] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled
@ 2025-06-21 0:48 Li Wang
2025-06-23 9:01 ` Lorenzo Stoakes
2025-06-24 3:27 ` [PATCH v2] " Li Wang
0 siblings, 2 replies; 4+ messages in thread
From: Li Wang @ 2025-06-21 0:48 UTC (permalink / raw)
To: akpm, linux-kselftest, linux-mm, linux-kernel
Cc: Aruna Ramakrishna, Bagas Sanjaya, Catalin Marinas, Dave Hansen,
David Hildenbrand, Joey Gouly, Johannes Weiner, Keith Lucas,
Ryan Roberts, Shuah Khan
Some systems (e.g. minimal or real-time kernels) may not enable
Transparent Hugepages (THP), causing MADV_HUGEPAGE to return EINVAL.
This patch introduces a runtime check using the existing THP sysfs
interface and skips the hugepage merging test (`-H`) when THP is
not available.
To avoid those failures:
# -----------------------------
# running ./ksm_tests -H -s 100
# -----------------------------
# ksm_tests: MADV_HUGEPAGE: Invalid argument
# [FAIL]
not ok 1 ksm_tests -H -s 100 # exit=2
# --------------------
# running ./khugepaged
# --------------------
# Reading PMD pagesize failed# [FAIL]
not ok 1 khugepaged # exit=1
# --------------------
# running ./soft-dirty
# --------------------
# TAP version 13
# 1..15
# ok 1 Test test_simple
# ok 2 Test test_vma_reuse dirty bit of allocated page
# ok 3 Test test_vma_reuse dirty bit of reused address page
# Bail out! Reading PMD pagesize failed# Planned tests != run tests (15 != 3)
# # Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 1 soft-dirty # exit=1
# SUMMARY: PASS=0 SKIP=0 FAIL=1
# -------------------
# running ./migration
# -------------------
# TAP version 13
# 1..3
# # Starting 3 tests from 1 test cases.
# # RUN migration.private_anon ...
# # OK migration.private_anon
# ok 1 migration.private_anon
# # RUN migration.shared_anon ...
# # OK migration.shared_anon
# ok 2 migration.shared_anon
# # RUN migration.private_anon_thp ...
# # migration.c:196:private_anon_thp:Expected madvise(ptr, TWOMEG, MADV_HUGEPAGE) (-1) == 0 (0)
# # private_anon_thp: Test terminated by assertion
# # FAIL migration.private_anon_thp
# not ok 3 migration.private_anon_thp
# # FAILED: 2 / 3 tests passed.
# # Totals: pass:2 fail:1 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 1 migration # exit=1
Signed-off-by: Li Wang <liwang@redhat.com>
Cc: Aruna Ramakrishna <aruna.ramakrishna@oracle.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Keith Lucas <keith.lucas@oracle.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Shuah Khan <shuah@kernel.org>
---
tools/testing/selftests/mm/khugepaged.c | 5 +++++
tools/testing/selftests/mm/ksm_tests.c | 6 ++++++
tools/testing/selftests/mm/migration.c | 8 ++++++++
tools/testing/selftests/mm/soft-dirty.c | 9 ++++++++-
tools/testing/selftests/mm/thp_settings.c | 11 +++++++++++
tools/testing/selftests/mm/thp_settings.h | 2 ++
6 files changed, 40 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/mm/khugepaged.c b/tools/testing/selftests/mm/khugepaged.c
index 8a4d34cce36b..6822eb7ea58e 100644
--- a/tools/testing/selftests/mm/khugepaged.c
+++ b/tools/testing/selftests/mm/khugepaged.c
@@ -1190,6 +1190,11 @@ int main(int argc, char **argv)
.read_ahead_kb = 0,
};
+ if (!thp_is_enabled()) {
+ printf("Transparent Hugepages not available\n");
+ return KSFT_SKIP;
+ }
+
parse_test_type(argc, argv);
setbuf(stdout, NULL);
diff --git a/tools/testing/selftests/mm/ksm_tests.c b/tools/testing/selftests/mm/ksm_tests.c
index dcdd5bb20f3d..1fb2263faa8d 100644
--- a/tools/testing/selftests/mm/ksm_tests.c
+++ b/tools/testing/selftests/mm/ksm_tests.c
@@ -15,6 +15,7 @@
#include "../kselftest.h"
#include <include/vdso/time64.h>
#include "vm_util.h"
+#include "thp_settings.h"
#define KSM_SYSFS_PATH "/sys/kernel/mm/ksm/"
#define KSM_FP(s) (KSM_SYSFS_PATH s)
@@ -555,6 +556,11 @@ static int ksm_merge_hugepages_time(int merge_type, int mapping, int prot,
unsigned long scan_time_ns;
int pagemap_fd, n_normal_pages, n_huge_pages;
+ if (!thp_is_enabled()) {
+ printf("Transparent Hugepages not available\n");
+ return KSFT_SKIP;
+ }
+
map_size *= MB;
size_t len = map_size;
diff --git a/tools/testing/selftests/mm/migration.c b/tools/testing/selftests/mm/migration.c
index 1e3a595fbf01..a306f8bab087 100644
--- a/tools/testing/selftests/mm/migration.c
+++ b/tools/testing/selftests/mm/migration.c
@@ -5,6 +5,8 @@
*/
#include "../kselftest_harness.h"
+#include "thp_settings.h"
+
#include <strings.h>
#include <pthread.h>
#include <numa.h>
@@ -185,6 +187,9 @@ TEST_F_TIMEOUT(migration, private_anon_thp, 2*RUNTIME)
uint64_t *ptr;
int i;
+ if (!thp_is_enabled())
+ SKIP(return, "Transparent Hugepages not available");
+
if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0)
SKIP(return, "Not enough threads or NUMA nodes available");
@@ -214,6 +219,9 @@ TEST_F_TIMEOUT(migration, shared_anon_thp, 2*RUNTIME)
uint64_t *ptr;
int i;
+ if (!thp_is_enabled())
+ SKIP(return, "Transparent Hugepages not available");
+
if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0)
SKIP(return, "Not enough threads or NUMA nodes available");
diff --git a/tools/testing/selftests/mm/soft-dirty.c b/tools/testing/selftests/mm/soft-dirty.c
index 8e1462ce0532..72d8ded87756 100644
--- a/tools/testing/selftests/mm/soft-dirty.c
+++ b/tools/testing/selftests/mm/soft-dirty.c
@@ -6,8 +6,10 @@
#include <stdint.h>
#include <malloc.h>
#include <sys/mman.h>
+
#include "../kselftest.h"
#include "vm_util.h"
+#include "thp_settings.h"
#define PAGEMAP_FILE_PATH "/proc/self/pagemap"
#define TEST_ITERATIONS 10000
@@ -78,8 +80,13 @@ static void test_hugepage(int pagemap_fd, int pagesize)
{
char *map;
int i, ret;
- size_t hpage_len = read_pmd_pagesize();
+ if (!thp_is_enabled()) {
+ printf("Skipping test: Transparent Hugepages not available\n");
+ return KSFT_SKIP;
+ }
+
+ size_t hpage_len = read_pmd_pagesize();
if (!hpage_len)
ksft_exit_fail_msg("Reading PMD pagesize failed");
diff --git a/tools/testing/selftests/mm/thp_settings.c b/tools/testing/selftests/mm/thp_settings.c
index ad872af1c81a..bad60ac52874 100644
--- a/tools/testing/selftests/mm/thp_settings.c
+++ b/tools/testing/selftests/mm/thp_settings.c
@@ -381,3 +381,14 @@ unsigned long thp_shmem_supported_orders(void)
{
return __thp_supported_orders(true);
}
+
+bool thp_is_enabled(void)
+{
+ if (access(THP_SYSFS, F_OK) != 0)
+ return false;
+
+ int mode = thp_read_string("enabled", thp_enabled_strings);
+
+ /* THP is considered enabled if it's either "always" or "madvise" */
+ return mode == 1 || mode == 3;
+}
diff --git a/tools/testing/selftests/mm/thp_settings.h b/tools/testing/selftests/mm/thp_settings.h
index fc131d23d593..6c07f70beee9 100644
--- a/tools/testing/selftests/mm/thp_settings.h
+++ b/tools/testing/selftests/mm/thp_settings.h
@@ -84,4 +84,6 @@ void thp_set_read_ahead_path(char *path);
unsigned long thp_supported_orders(void);
unsigned long thp_shmem_supported_orders(void);
+bool thp_is_enabled(void);
+
#endif /* __THP_SETTINGS_H__ */
--
2.49.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled
2025-06-21 0:48 [PATCH] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled Li Wang
@ 2025-06-23 9:01 ` Lorenzo Stoakes
2025-06-24 3:27 ` [PATCH v2] " Li Wang
1 sibling, 0 replies; 4+ messages in thread
From: Lorenzo Stoakes @ 2025-06-23 9:01 UTC (permalink / raw)
To: Li Wang
Cc: akpm, linux-kselftest, linux-mm, linux-kernel, Aruna Ramakrishna,
Bagas Sanjaya, Catalin Marinas, Dave Hansen, David Hildenbrand,
Joey Gouly, Johannes Weiner, Keith Lucas, Ryan Roberts,
Shuah Khan
Hi Li,
This is breaking the mm-new mm selftest test build, did you have some local
changes you didn't submit by mistake?
You seem to be returning KSFT_SKIP from a void function.
I enclose a simple fix-patch that fixes the issue, but obviously you should
check to see if this is still doing what you want :)
Cheers, Lorenzo
On Sat, Jun 21, 2025 at 08:48:08AM +0800, Li Wang wrote:
[snip]
> diff --git a/tools/testing/selftests/mm/soft-dirty.c b/tools/testing/selftests/mm/soft-dirty.c
> index 8e1462ce0532..72d8ded87756 100644
> --- a/tools/testing/selftests/mm/soft-dirty.c
> +++ b/tools/testing/selftests/mm/soft-dirty.c
> @@ -6,8 +6,10 @@
> #include <stdint.h>
> #include <malloc.h>
> #include <sys/mman.h>
> +
> #include "../kselftest.h"
> #include "vm_util.h"
> +#include "thp_settings.h"
>
> #define PAGEMAP_FILE_PATH "/proc/self/pagemap"
> #define TEST_ITERATIONS 10000
> @@ -78,8 +80,13 @@ static void test_hugepage(int pagemap_fd, int pagesize)
> {
> char *map;
> int i, ret;
> - size_t hpage_len = read_pmd_pagesize();
>
> + if (!thp_is_enabled()) {
> + printf("Skipping test: Transparent Hugepages not available\n");
> + return KSFT_SKIP;
^---- Returning a value in a void function?
----8<----
From cf6643f4232da4a7014082363896a2cf56eaa88e Mon Sep 17 00:00:00 2001
From: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Date: Mon, 23 Jun 2025 09:59:28 +0100
Subject: [PATCH] fix
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
---
tools/testing/selftests/mm/soft-dirty.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/mm/soft-dirty.c b/tools/testing/selftests/mm/soft-dirty.c
index 72d8ded87756..c0b34cf75205 100644
--- a/tools/testing/selftests/mm/soft-dirty.c
+++ b/tools/testing/selftests/mm/soft-dirty.c
@@ -82,8 +82,8 @@ static void test_hugepage(int pagemap_fd, int pagesize)
int i, ret;
if (!thp_is_enabled()) {
- printf("Skipping test: Transparent Hugepages not available\n");
- return KSFT_SKIP;
+ ksft_test_result_skip("Skipping test: Transparent Hugepages not available\n");
+ return;
}
size_t hpage_len = read_pmd_pagesize();
--
2.50.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH v2] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled
2025-06-21 0:48 [PATCH] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled Li Wang
2025-06-23 9:01 ` Lorenzo Stoakes
@ 2025-06-24 3:27 ` Li Wang
2025-06-24 8:25 ` David Hildenbrand
1 sibling, 1 reply; 4+ messages in thread
From: Li Wang @ 2025-06-24 3:27 UTC (permalink / raw)
To: akpm, lorenzo.stoakes, linux-kselftest, linux-mm, linux-kernel
Cc: Aruna Ramakrishna, Bagas Sanjaya, Catalin Marinas, Dave Hansen,
David Hildenbrand, Joey Gouly, Johannes Weiner, Keith Lucas,
Ryan Roberts, Shuah Khan
Some systems (e.g. minimal or real-time kernels) may not enable
Transparent Hugepages (THP), causing MADV_HUGEPAGE to return EINVAL.
This patch introduces a runtime check using the existing THP sysfs
interface and skips the hugepage merging test (`-H`) when THP is
not available.
To avoid those failures:
# -----------------------------
# running ./ksm_tests -H -s 100
# -----------------------------
# ksm_tests: MADV_HUGEPAGE: Invalid argument
# [FAIL]
not ok 1 ksm_tests -H -s 100 # exit=2
# --------------------
# running ./khugepaged
# --------------------
# Reading PMD pagesize failed# [FAIL]
not ok 1 khugepaged # exit=1
# --------------------
# running ./soft-dirty
# --------------------
# TAP version 13
# 1..15
# ok 1 Test test_simple
# ok 2 Test test_vma_reuse dirty bit of allocated page
# ok 3 Test test_vma_reuse dirty bit of reused address page
# Bail out! Reading PMD pagesize failed# Planned tests != run tests (15 != 3)
# # Totals: pass:3 fail:0 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 1 soft-dirty # exit=1
# SUMMARY: PASS=0 SKIP=0 FAIL=1
# -------------------
# running ./migration
# -------------------
# TAP version 13
# 1..3
# # Starting 3 tests from 1 test cases.
# # RUN migration.private_anon ...
# # OK migration.private_anon
# ok 1 migration.private_anon
# # RUN migration.shared_anon ...
# # OK migration.shared_anon
# ok 2 migration.shared_anon
# # RUN migration.private_anon_thp ...
# # migration.c:196:private_anon_thp:Expected madvise(ptr, TWOMEG, MADV_HUGEPAGE) (-1) == 0 (0)
# # private_anon_thp: Test terminated by assertion
# # FAIL migration.private_anon_thp
# not ok 3 migration.private_anon_thp
# # FAILED: 2 / 3 tests passed.
# # Totals: pass:2 fail:1 xfail:0 xpass:0 skip:0 error:0
# [FAIL]
not ok 1 migration # exit=1
Signed-off-by: Li Wang <liwang@redhat.com>
Cc: Aruna Ramakrishna <aruna.ramakrishna@oracle.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Keith Lucas <keith.lucas@oracle.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
---
Notes:
v1 --> v2:
* replaces printf() + return KSFT_SKIP with a standardized
ksft_test_result_skip() call and return in soft-dirty.c
to avoid test build issue.
tools/testing/selftests/mm/khugepaged.c | 5 +++++
tools/testing/selftests/mm/ksm_tests.c | 6 ++++++
tools/testing/selftests/mm/migration.c | 8 ++++++++
tools/testing/selftests/mm/soft-dirty.c | 9 ++++++++-
tools/testing/selftests/mm/thp_settings.c | 11 +++++++++++
tools/testing/selftests/mm/thp_settings.h | 2 ++
6 files changed, 40 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/mm/khugepaged.c b/tools/testing/selftests/mm/khugepaged.c
index 8a4d34cce36b..6822eb7ea58e 100644
--- a/tools/testing/selftests/mm/khugepaged.c
+++ b/tools/testing/selftests/mm/khugepaged.c
@@ -1190,6 +1190,11 @@ int main(int argc, char **argv)
.read_ahead_kb = 0,
};
+ if (!thp_is_enabled()) {
+ printf("Transparent Hugepages not available\n");
+ return KSFT_SKIP;
+ }
+
parse_test_type(argc, argv);
setbuf(stdout, NULL);
diff --git a/tools/testing/selftests/mm/ksm_tests.c b/tools/testing/selftests/mm/ksm_tests.c
index e80deac1436b..b77462b5c240 100644
--- a/tools/testing/selftests/mm/ksm_tests.c
+++ b/tools/testing/selftests/mm/ksm_tests.c
@@ -15,6 +15,7 @@
#include "../kselftest.h"
#include <include/vdso/time64.h>
#include "vm_util.h"
+#include "thp_settings.h"
#define KSM_SYSFS_PATH "/sys/kernel/mm/ksm/"
#define KSM_FP(s) (KSM_SYSFS_PATH s)
@@ -527,6 +528,11 @@ static int ksm_merge_hugepages_time(int merge_type, int mapping, int prot,
unsigned long scan_time_ns;
int pagemap_fd, n_normal_pages, n_huge_pages;
+ if (!thp_is_enabled()) {
+ printf("Transparent Hugepages not available\n");
+ return KSFT_SKIP;
+ }
+
map_size *= MB;
size_t len = map_size;
diff --git a/tools/testing/selftests/mm/migration.c b/tools/testing/selftests/mm/migration.c
index 1e3a595fbf01..a306f8bab087 100644
--- a/tools/testing/selftests/mm/migration.c
+++ b/tools/testing/selftests/mm/migration.c
@@ -5,6 +5,8 @@
*/
#include "../kselftest_harness.h"
+#include "thp_settings.h"
+
#include <strings.h>
#include <pthread.h>
#include <numa.h>
@@ -185,6 +187,9 @@ TEST_F_TIMEOUT(migration, private_anon_thp, 2*RUNTIME)
uint64_t *ptr;
int i;
+ if (!thp_is_enabled())
+ SKIP(return, "Transparent Hugepages not available");
+
if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0)
SKIP(return, "Not enough threads or NUMA nodes available");
@@ -214,6 +219,9 @@ TEST_F_TIMEOUT(migration, shared_anon_thp, 2*RUNTIME)
uint64_t *ptr;
int i;
+ if (!thp_is_enabled())
+ SKIP(return, "Transparent Hugepages not available");
+
if (self->nthreads < 2 || self->n1 < 0 || self->n2 < 0)
SKIP(return, "Not enough threads or NUMA nodes available");
diff --git a/tools/testing/selftests/mm/soft-dirty.c b/tools/testing/selftests/mm/soft-dirty.c
index 8e1462ce0532..8a3f2b4b2186 100644
--- a/tools/testing/selftests/mm/soft-dirty.c
+++ b/tools/testing/selftests/mm/soft-dirty.c
@@ -6,8 +6,10 @@
#include <stdint.h>
#include <malloc.h>
#include <sys/mman.h>
+
#include "../kselftest.h"
#include "vm_util.h"
+#include "thp_settings.h"
#define PAGEMAP_FILE_PATH "/proc/self/pagemap"
#define TEST_ITERATIONS 10000
@@ -78,8 +80,13 @@ static void test_hugepage(int pagemap_fd, int pagesize)
{
char *map;
int i, ret;
- size_t hpage_len = read_pmd_pagesize();
+ if (!thp_is_enabled()) {
+ ksft_test_result_skip("Transparent Hugepages not available\n");
+ return;
+ }
+
+ size_t hpage_len = read_pmd_pagesize();
if (!hpage_len)
ksft_exit_fail_msg("Reading PMD pagesize failed");
diff --git a/tools/testing/selftests/mm/thp_settings.c b/tools/testing/selftests/mm/thp_settings.c
index ad872af1c81a..bad60ac52874 100644
--- a/tools/testing/selftests/mm/thp_settings.c
+++ b/tools/testing/selftests/mm/thp_settings.c
@@ -381,3 +381,14 @@ unsigned long thp_shmem_supported_orders(void)
{
return __thp_supported_orders(true);
}
+
+bool thp_is_enabled(void)
+{
+ if (access(THP_SYSFS, F_OK) != 0)
+ return false;
+
+ int mode = thp_read_string("enabled", thp_enabled_strings);
+
+ /* THP is considered enabled if it's either "always" or "madvise" */
+ return mode == 1 || mode == 3;
+}
diff --git a/tools/testing/selftests/mm/thp_settings.h b/tools/testing/selftests/mm/thp_settings.h
index fc131d23d593..6c07f70beee9 100644
--- a/tools/testing/selftests/mm/thp_settings.h
+++ b/tools/testing/selftests/mm/thp_settings.h
@@ -84,4 +84,6 @@ void thp_set_read_ahead_path(char *path);
unsigned long thp_supported_orders(void);
unsigned long thp_shmem_supported_orders(void);
+bool thp_is_enabled(void);
+
#endif /* __THP_SETTINGS_H__ */
--
2.49.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v2] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled
2025-06-24 3:27 ` [PATCH v2] " Li Wang
@ 2025-06-24 8:25 ` David Hildenbrand
0 siblings, 0 replies; 4+ messages in thread
From: David Hildenbrand @ 2025-06-24 8:25 UTC (permalink / raw)
To: Li Wang, akpm, lorenzo.stoakes, linux-kselftest, linux-mm,
linux-kernel
Cc: Aruna Ramakrishna, Bagas Sanjaya, Catalin Marinas, Dave Hansen,
Joey Gouly, Johannes Weiner, Keith Lucas, Ryan Roberts,
Shuah Khan
On 24.06.25 05:27, Li Wang wrote:
> Some systems (e.g. minimal or real-time kernels) may not enable
> Transparent Hugepages (THP), causing MADV_HUGEPAGE to return EINVAL.
> This patch introduces a runtime check using the existing THP sysfs
> interface and skips the hugepage merging test (`-H`) when THP is
> not available.
>
> To avoid those failures:
But we deliberately have in tools/testing/selftests/mm/config:
CONFIG_TRANSPARENT_HUGEPAGE=y
So isn't this rather a test setup issue? Meaning, the environment is
not well prepared.
--
Cheers,
David / dhildenb
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2025-06-24 8:25 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-06-21 0:48 [PATCH] ksm_tests: Skip hugepage test when Transparent Hugepages are disabled Li Wang
2025-06-23 9:01 ` Lorenzo Stoakes
2025-06-24 3:27 ` [PATCH v2] " Li Wang
2025-06-24 8:25 ` David Hildenbrand
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).