linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Usama Arif <usamaarif642@gmail.com>
To: David Hildenbrand <david@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-mm@kvack.org
Cc: linux-fsdevel@vger.kernel.org, corbet@lwn.net, rppt@kernel.org,
	surenb@google.com, mhocko@suse.com, hannes@cmpxchg.org,
	baohua@kernel.org, shakeel.butt@linux.dev, riel@surriel.com,
	ziy@nvidia.com, laoar.shao@gmail.com, dev.jain@arm.com,
	baolin.wang@linux.alibaba.com, npache@redhat.com,
	lorenzo.stoakes@oracle.com, Liam.Howlett@oracle.com,
	ryan.roberts@arm.com, vbabka@suse.cz, jannh@google.com,
	Arnd Bergmann <arnd@arndb.de>,
	sj@kernel.org, linux-kernel@vger.kernel.org,
	linux-doc@vger.kernel.org, kernel-team@meta.com
Subject: Re: [PATCH v3 5/6] selftests: prctl: introduce tests for disabling THPs completely
Date: Tue, 5 Aug 2025 15:26:16 +0100	[thread overview]
Message-ID: <955ed533-dd7e-45e6-a2cb-f379a5878ef7@gmail.com> (raw)
In-Reply-To: <e608f766-8750-4781-bd23-8fa95b6d683a@redhat.com>



On 05/08/2025 13:55, David Hildenbrand wrote:
> On 05.08.25 14:46, Usama Arif wrote:
>>
>>
>> On 05/08/2025 13:39, David Hildenbrand wrote:
>>>> +FIXTURE_SETUP(prctl_thp_disable_completely)
>>>> +{
>>>> +    if (!thp_available())
>>>> +        SKIP(return, "Transparent Hugepages not available\n");
>>>> +
>>>> +    self->pmdsize = read_pmd_pagesize();
>>>> +    if (!self->pmdsize)
>>>> +        SKIP(return, "Unable to read PMD size\n");
>>>> +
>>>> +    thp_save_settings();
>>>> +    thp_read_settings(&self->settings);
>>>> +    self->settings.thp_enabled = variant->thp_policy;
>>>> +    self->settings.hugepages[sz2ord(self->pmdsize, getpagesize())].enabled = THP_INHERIT;
>>>
>>> Oh, one more thing: should we set all other sizes also to THP_INHERIT or (for simplicity) THP_NEVER?
>>>
>>
>> hmm do we need to? I am hoping that we should always get the PMD size THP no matter what the policy
>> for others is in the scenario we have?
> 
> Assuming 64K is set to "always", couldn't it mislead us in the "madvise"/"never" test cases in some scenarios?
> 

I tried it with 64K to always and seems to be ok. check_huge_anon checks AnonHugepages from smaps which
only indicates the pmd mapped THPs only. So I think should be ok?
Happy to set them to never and the change is probably something simple like below (untested), but
just trying to understand better the need for it.

diff --git a/tools/testing/selftests/mm/prctl_thp_disable.c b/tools/testing/selftests/mm/prctl_thp_disable.c
index 4b072d4ecc30..ee692c397835 100644
--- a/tools/testing/selftests/mm/prctl_thp_disable.c
+++ b/tools/testing/selftests/mm/prctl_thp_disable.c
@@ -142,6 +142,8 @@ FIXTURE_SETUP(prctl_thp_disable_completely)
        thp_save_settings();
        thp_read_settings(&self->settings);
        self->settings.thp_enabled = variant->thp_policy;
+       for (int i = 1; i < sz2ord(self->pmdsize, getpagesize()); i++)
+               self->settings.hugepages[i].enabled = THP_INHERIT;
        self->settings.hugepages[sz2ord(self->pmdsize, getpagesize())].enabled = THP_INHERIT;
        thp_write_settings(&self->settings);
 }
@@ -244,6 +246,8 @@ FIXTURE_SETUP(prctl_thp_disable_except_madvise)
        thp_save_settings();
        thp_read_settings(&self->settings);
        self->settings.thp_enabled = variant->thp_policy;
+       for (int i = 1; i < sz2ord(self->pmdsize, getpagesize()); i++)
+               self->settings.hugepages[i].enabled = THP_INHERIT;
        self->settings.hugepages[sz2ord(self->pmdsize, getpagesize())].enabled = THP_INHERIT;
        thp_write_settings(&self->settings);
 }
 


[root@vm7 vmuser]# echo always >  /sys/kernel/mm/transparent_hugepage/hugepages-64kB/enabled                                                                                                                                                                                                                        
[root@vm7 vmuser]# ./prctl_thp_disable                                                                                                                                                                                                                                                                              
TAP version 13                                                                                                                                                                                                                                                                                                      
1..12                                                                                                                                                                                                                                                                                                               
# Starting 12 tests from 6 test cases.                                                                                                                                                                                                                                                                              
#  RUN           prctl_thp_disable_completely.never.nofork ...
#            OK  prctl_thp_disable_completely.never.nofork
ok 1 prctl_thp_disable_completely.never.nofork
#  RUN           prctl_thp_disable_completely.never.fork ...
#            OK  prctl_thp_disable_completely.never.fork
ok 2 prctl_thp_disable_completely.never.fork
#  RUN           prctl_thp_disable_completely.madvise.nofork ...
#            OK  prctl_thp_disable_completely.madvise.nofork
ok 3 prctl_thp_disable_completely.madvise.nofork
#  RUN           prctl_thp_disable_completely.madvise.fork ...
#            OK  prctl_thp_disable_completely.madvise.fork
ok 4 prctl_thp_disable_completely.madvise.fork
#  RUN           prctl_thp_disable_completely.always.nofork ...
#            OK  prctl_thp_disable_completely.always.nofork
ok 5 prctl_thp_disable_completely.always.nofork
#  RUN           prctl_thp_disable_completely.always.fork ...
#            OK  prctl_thp_disable_completely.always.fork
ok 6 prctl_thp_disable_completely.always.fork
#  RUN           prctl_thp_disable_except_madvise.never.nofork ...
#            OK  prctl_thp_disable_except_madvise.never.nofork
ok 7 prctl_thp_disable_except_madvise.never.nofork
#  RUN           prctl_thp_disable_except_madvise.never.fork ...
#            OK  prctl_thp_disable_except_madvise.never.fork
ok 8 prctl_thp_disable_except_madvise.never.fork
#  RUN           prctl_thp_disable_except_madvise.madvise.nofork ...
#            OK  prctl_thp_disable_except_madvise.madvise.nofork
ok 9 prctl_thp_disable_except_madvise.madvise.nofork
#  RUN           prctl_thp_disable_except_madvise.madvise.fork ...
#            OK  prctl_thp_disable_except_madvise.madvise.fork
ok 10 prctl_thp_disable_except_madvise.madvise.fork
#  RUN           prctl_thp_disable_except_madvise.always.nofork ...
#            OK  prctl_thp_disable_except_madvise.always.nofork
ok 11 prctl_thp_disable_except_madvise.always.nofork
#  RUN           prctl_thp_disable_except_madvise.always.fork ...
#            OK  prctl_thp_disable_except_madvise.always.fork
ok 12 prctl_thp_disable_except_madvise.always.fork
# PASSED: 12 / 12 tests passed.
# Totals: pass:12 fail:0 xfail:0 xpass:0 skip:0 error:0

  reply	other threads:[~2025-08-05 14:26 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-04 15:40 [PATCH v3 0/6] prctl: extend PR_SET_THP_DISABLE to only provide THPs when advised Usama Arif
2025-08-04 15:40 ` [PATCH v3 1/6] prctl: extend PR_SET_THP_DISABLE to optionally exclude VM_HUGEPAGE Usama Arif
2025-08-04 15:40 ` [PATCH v3 2/6] mm/huge_memory: convert "tva_flags" to "enum tva_type" Usama Arif
2025-08-05  1:41   ` Baolin Wang
2025-08-05 14:35   ` Lorenzo Stoakes
2025-08-04 15:40 ` [PATCH v3 3/6] mm/huge_memory: respect MADV_COLLAPSE with PR_THP_DISABLE_EXCEPT_ADVISED Usama Arif
2025-08-05  1:49   ` Baolin Wang
2025-08-05 14:43   ` Lorenzo Stoakes
2025-08-05 15:08     ` Usama Arif
2025-08-04 15:40 ` [PATCH v3 4/6] docs: transhuge: document process level THP controls Usama Arif
2025-08-05 10:24   ` David Hildenbrand
2025-08-05 12:09     ` Usama Arif
2025-08-05 15:08   ` Lorenzo Stoakes
2025-08-04 15:40 ` [PATCH v3 5/6] selftests: prctl: introduce tests for disabling THPs completely Usama Arif
2025-08-05 10:30   ` David Hildenbrand
2025-08-05 12:13     ` Usama Arif
2025-08-05 12:16       ` David Hildenbrand
2025-08-05 12:39   ` David Hildenbrand
2025-08-05 12:46     ` Usama Arif
2025-08-05 12:55       ` David Hildenbrand
2025-08-05 14:26         ` Usama Arif [this message]
2025-08-05 14:32           ` David Hildenbrand
2025-08-04 15:40 ` [PATCH v3 6/6] selftests: prctl: introduce tests for disabling THPs except for madvise Usama Arif
2025-08-05 10:36   ` David Hildenbrand
2025-08-05 12:19     ` Usama Arif
2025-08-05 12:29       ` David Hildenbrand
2025-08-05 12:31         ` Usama Arif
2025-08-04 15:46 ` [PATCH v3 0/6] prctl: extend PR_SET_THP_DISABLE to only provide THPs when advised Usama Arif
2025-08-13  6:06 ` Lorenzo Stoakes
2025-08-13  8:07   ` Usama Arif

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=955ed533-dd7e-45e6-a2cb-f379a5878ef7@gmail.com \
    --to=usamaarif642@gmail.com \
    --cc=Liam.Howlett@oracle.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=baohua@kernel.org \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=corbet@lwn.net \
    --cc=david@redhat.com \
    --cc=dev.jain@arm.com \
    --cc=hannes@cmpxchg.org \
    --cc=jannh@google.com \
    --cc=kernel-team@meta.com \
    --cc=laoar.shao@gmail.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=mhocko@suse.com \
    --cc=npache@redhat.com \
    --cc=riel@surriel.com \
    --cc=rppt@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=shakeel.butt@linux.dev \
    --cc=sj@kernel.org \
    --cc=surenb@google.com \
    --cc=vbabka@suse.cz \
    --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;
as well as URLs for NNTP newsgroup(s).