[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20250910093051.1693097-1-david@redhat.com>
Date: Wed, 10 Sep 2025 11:30:51 +0200
From: David Hildenbrand <david@...hat.com>
To: linux-kernel@...r.kernel.org
Cc: linux-mm@...ck.org,
linux-kselftest@...r.kernel.org,
David Hildenbrand <david@...hat.com>,
Andrew Morton <akpm@...ux-foundation.org>,
Jason Gunthorpe <jgg@...pe.ca>,
John Hubbard <jhubbard@...dia.com>,
Peter Xu <peterx@...hat.com>,
Lorenzo Stoakes <lorenzo.stoakes@...cle.com>,
"Liam R. Howlett" <Liam.Howlett@...cle.com>,
Vlastimil Babka <vbabka@...e.cz>,
Mike Rapoport <rppt@...nel.org>,
Suren Baghdasaryan <surenb@...gle.com>,
Michal Hocko <mhocko@...e.com>,
Shuah Khan <shuah@...nel.org>
Subject: [PATCH v1] selftests/mm: gup_tests: option to GUP all pages in a single call
We recently missed detecting an issue during early testing because
the default (!all) tests would not trigger it and even when running
"all" tests it only would happen sometimes because of races.
So let's allow for an easy way to specify "GUP all pages in a single
call", extend the test matrix and extend our default (!all) tests.
By GUP'ing all pages in a single call, with the default size of 128MiB
we'll cover multiple leaf page tables / PMDs on architectures with sane
THP sizes.
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Jason Gunthorpe <jgg@...pe.ca>
Cc: John Hubbard <jhubbard@...dia.com>
Cc: Peter Xu <peterx@...hat.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
Cc: "Liam R. Howlett" <Liam.Howlett@...cle.com>
Cc: Vlastimil Babka <vbabka@...e.cz>
Cc: Mike Rapoport <rppt@...nel.org>
Cc: Suren Baghdasaryan <surenb@...gle.com>
Cc: Michal Hocko <mhocko@...e.com>
Cc: Shuah Khan <shuah@...nel.org>
Signed-off-by: David Hildenbrand <david@...hat.com>
---
tools/testing/selftests/mm/gup_test.c | 2 ++
tools/testing/selftests/mm/run_vmtests.sh | 8 +++++---
2 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/tools/testing/selftests/mm/gup_test.c b/tools/testing/selftests/mm/gup_test.c
index bdeaac67ff9aa..8900b840c17a7 100644
--- a/tools/testing/selftests/mm/gup_test.c
+++ b/tools/testing/selftests/mm/gup_test.c
@@ -139,6 +139,8 @@ int main(int argc, char **argv)
break;
case 'n':
nr_pages = atoi(optarg);
+ if (nr_pages < 0)
+ nr_pages = size / psize();
break;
case 't':
thp = 1;
diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh
index 9e88cc25b9df2..6240e579b3ba5 100755
--- a/tools/testing/selftests/mm/run_vmtests.sh
+++ b/tools/testing/selftests/mm/run_vmtests.sh
@@ -138,7 +138,7 @@ run_gup_matrix() {
# -n: How many pages to fetch together? 512 is special
# because it's default thp size (or 2M on x86), 123 to
# just test partial gup when hit a huge in whatever form
- for num in "-n 1" "-n 512" "-n 123"; do
+ for num in "-n 1" "-n 512" "-n 123" "-n -1"; do
CATEGORY="gup_test" run_test ./gup_test \
$huge $test_cmd $write $share $num
done
@@ -313,9 +313,11 @@ if $RUN_ALL; then
run_gup_matrix
else
# get_user_pages_fast() benchmark
- CATEGORY="gup_test" run_test ./gup_test -u
+ CATEGORY="gup_test" run_test ./gup_test -u -n 1
+ CATEGORY="gup_test" run_test ./gup_test -u -n -1
# pin_user_pages_fast() benchmark
- CATEGORY="gup_test" run_test ./gup_test -a
+ CATEGORY="gup_test" run_test ./gup_test -a -n 1
+ CATEGORY="gup_test" run_test ./gup_test -a -n -1
fi
# Dump pages 0, 19, and 4096, using pin_user_pages:
CATEGORY="gup_test" run_test ./gup_test -ct -F 0x1 0 19 0x1000
--
2.50.1
Powered by blists - more mailing lists