From: Ming Lei <ming.lei@redhat.com>
To: Uday Shankar <ushankar@purestorage.com>
Cc: Jens Axboe <axboe@kernel.dk>,
Caleb Sander Mateos <csander@purestorage.com>,
Andrew Morton <akpm@linux-foundation.org>,
Shuah Khan <shuah@kernel.org>, Jonathan Corbet <corbet@lwn.net>,
linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-kselftest@vger.kernel.org, linux-doc@vger.kernel.org
Subject: Re: [PATCH v8 8/9] selftests: ublk: add stress test for per io daemons
Date: Fri, 30 May 2025 10:13:00 +0800 [thread overview]
Message-ID: <aDkULJG9zwqCWqTk@fedora> (raw)
In-Reply-To: <20250529-ublk_task_per_io-v8-8-e9d3b119336a@purestorage.com>
On Thu, May 29, 2025 at 05:47:17PM -0600, Uday Shankar wrote:
> Add a new test_stress_06 for the per io daemons feature. This is just a
> copy of test_stress_01 with the per_io_tasks flag added, with varying
> amounts of nthreads. This test is able to reproduce a panic which was
> caught manually during development [1]; in the current version of this
> patch set, it passes.
>
> Note that this commit also makes all stress tests using the
> run_io_and_remove helper more stressful by additionally exercising the
> batch submit (queue_rqs) path.
>
> [1] https://lore.kernel.org/linux-block/aDgwGoGCEpwd1mFY@fedora/
>
> Suggested-by: Ming Lei <ming.lei@redhat.com>
> Signed-off-by: Uday Shankar <ushankar@purestorage.com>
> ---
> tools/testing/selftests/ublk/Makefile | 1 +
> tools/testing/selftests/ublk/test_common.sh | 5 ++++
> tools/testing/selftests/ublk/test_stress_06.sh | 36 ++++++++++++++++++++++++++
> 3 files changed, 42 insertions(+)
>
> diff --git a/tools/testing/selftests/ublk/Makefile b/tools/testing/selftests/ublk/Makefile
> index 5d7f4ecfb81612f919a89eb442f948d6bfafe225..1fb1a95d452c2e9a7ed78cb8b12be2b759074e11 100644
> --- a/tools/testing/selftests/ublk/Makefile
> +++ b/tools/testing/selftests/ublk/Makefile
> @@ -38,6 +38,7 @@ TEST_PROGS += test_stress_02.sh
> TEST_PROGS += test_stress_03.sh
> TEST_PROGS += test_stress_04.sh
> TEST_PROGS += test_stress_05.sh
> +TEST_PROGS += test_stress_06.sh
>
> TEST_GEN_PROGS_EXTENDED = kublk
>
> diff --git a/tools/testing/selftests/ublk/test_common.sh b/tools/testing/selftests/ublk/test_common.sh
> index 0145569ee7e9a45b41898c2c789842b4c8380f18..8a4dbd09feb0a885ec7539f1a285ed6f437fe3ab 100755
> --- a/tools/testing/selftests/ublk/test_common.sh
> +++ b/tools/testing/selftests/ublk/test_common.sh
> @@ -278,6 +278,11 @@ __run_io_and_remove()
> fio --name=job1 --filename=/dev/ublkb"${dev_id}" --ioengine=libaio \
> --rw=randrw --norandommap --iodepth=256 --size="${size}" --numjobs="$(nproc)" \
> --runtime=20 --time_based > /dev/null 2>&1 &
> + fio --name=batchjob --filename=/dev/ublkb"${dev_id}" --ioengine=io_uring \
> + --rw=randrw --norandommap --iodepth=256 --size="${size}" \
> + --numjobs="$(nproc)" --runtime=20 --time_based \
> + --iodepth_batch_submit=32 --iodepth_batch_complete_min=32 \
> + --force_async=7 > /dev/null 2>&1 &
I think we can replace job1 with the batchjob simply.
> sleep 2
> if [ "${kill_server}" = "yes" ]; then
> local state
> diff --git a/tools/testing/selftests/ublk/test_stress_06.sh b/tools/testing/selftests/ublk/test_stress_06.sh
> new file mode 100755
> index 0000000000000000000000000000000000000000..3aee8521032e3962b8b070cda8eb295b01e7b124
> --- /dev/null
> +++ b/tools/testing/selftests/ublk/test_stress_06.sh
> @@ -0,0 +1,36 @@
> +#!/bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +
> +. "$(cd "$(dirname "$0")" && pwd)"/test_common.sh
> +TID="stress_06"
> +ERR_CODE=0
> +
> +ublk_io_and_remove()
> +{
> + run_io_and_remove "$@"
> + ERR_CODE=$?
> + if [ ${ERR_CODE} -ne 0 ]; then
> + echo "$TID failure: $*"
> + _show_result $TID $ERR_CODE
> + fi
> +}
> +
> +if ! _have_program fio; then
> + exit "$UBLK_SKIP_CODE"
> +fi
> +
> +_prep_test "stress" "run IO and remove device with per_io_tasks"
> +
> +_create_backfile 0 256M
> +_create_backfile 1 128M
> +_create_backfile 2 128M
> +
> +ublk_io_and_remove 8G -t null -q 4 --nthreads 5 --per_io_tasks &
> +ublk_io_and_remove 256M -t loop -q 4 --nthreads 3 --per_io_tasks \
> + "${UBLK_BACKFILES[0]}" &
> +ublk_io_and_remove 256M -t stripe -q 4 --nthreads 4 --per_io_tasks \
> + "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
> +wait
We plan to reuse the test for blktest or liburing test, so feature check
is needed.
But this patch shouldn't be a blocker, and it can be one follow-up with
something like the following with above batchjob change, then generic
feature can be covered completely wrt. stress test.
diff --git a/tools/testing/selftests/ublk/test_stress_03.sh b/tools/testing/selftests/ublk/test_stress_03.sh
index 7d728ce50774..f5908c1357d2 100755
--- a/tools/testing/selftests/ublk/test_stress_03.sh
+++ b/tools/testing/selftests/ublk/test_stress_03.sh
@@ -41,5 +41,13 @@ if _have_feature "AUTO_BUF_REG"; then
fi
wait
+if _have_feature "PER_IO_DAEMON"; then
+ ublk_io_and_remove 8G -t null -q 4 --nthreads 8 --per_io_tasks &
+ ublk_io_and_remove 256M -t loop -q 4 --nthreads 8 --per_io_tasks "${UBLK_BACKFILES[0]}" &
+ ublk_io_and_remove 256M -t stripe -q 4 --nthreads 8 --per_io_tasks "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+ ublk_io_and_remove 8G -t null -q 4 -z --nthreads 8 --per_io_tasks &
+fi
+wait
+
_cleanup_test "stress"
_show_result $TID $ERR_CODE
diff --git a/tools/testing/selftests/ublk/test_stress_04.sh b/tools/testing/selftests/ublk/test_stress_04.sh
index 9bcfa64ea1f0..be3ecdbe9bb4 100755
--- a/tools/testing/selftests/ublk/test_stress_04.sh
+++ b/tools/testing/selftests/ublk/test_stress_04.sh
@@ -38,6 +38,13 @@ if _have_feature "AUTO_BUF_REG"; then
ublk_io_and_kill_daemon 256M -t stripe -q 4 --auto_zc "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
ublk_io_and_kill_daemon 8G -t null -q 4 -z --auto_zc --auto_zc_fallback &
fi
+
+if _have_feature "PER_IO_DAEMON"; then
+ ublk_io_and_kill_daemon 8G -t null -q 4 --nthreads 8 --per_io_tasks &
+ ublk_io_and_kill_daemon 256M -t loop -q 4 --nthreads 8 --per_io_tasks "${UBLK_BACKFILES[0]}" &
+ ublk_io_and_kill_daemon 256M -t stripe -q 4 --nthreads 8 --per_io_tasks "${UBLK_BACKFILES[1]}" "${UBLK_BACKFILES[2]}" &
+ ublk_io_and_kill_daemon 8G -t null -q 4 -z --nthreads 8 --per_io_tasks &
+fi
wait
_cleanup_test "stress"
diff --git a/tools/testing/selftests/ublk/test_stress_05.sh b/tools/testing/selftests/ublk/test_stress_05.sh
index bcfc904cefc6..5634b1db1ded 100755
--- a/tools/testing/selftests/ublk/test_stress_05.sh
+++ b/tools/testing/selftests/ublk/test_stress_05.sh
@@ -69,5 +69,12 @@ if _have_feature "AUTO_BUF_REG"; then
done
fi
+if _have_feature "PER_IO_DAEMON"; then
+ ublk_io_and_remove 8G -t null -q 4 --nthreads 8 --per_io_tasks -r 1 -i "$reissue" &
+ ublk_io_and_remove 256M -t loop -q 4 --nthreads 8 --per_io_tasks -r 1 -i "$reissue" "${UBLK_BACKFILES[0]}" &
+ ublk_io_and_remove 8G -t null -q 4 -z --nthreads 8 --per_io_tasks -r 1 -i "$reissue" &
+fi
+wait
+
_cleanup_test "stress"
_show_result $TID $ERR_CODE
Thanks,
Ming
next prev parent reply other threads:[~2025-05-30 2:13 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-29 23:47 [PATCH v8 0/9] ublk: decouple server threads from ublk_queues/hctxs Uday Shankar
2025-05-29 23:47 ` [PATCH v8 1/9] ublk: have a per-io daemon instead of a per-queue daemon Uday Shankar
2025-05-30 1:33 ` Jens Axboe
2025-05-30 2:05 ` Ming Lei
2025-05-29 23:47 ` [PATCH v8 2/9] selftests: ublk: kublk: plumb q_id in io_uring user_data Uday Shankar
2025-05-29 23:47 ` [PATCH v8 3/9] selftests: ublk: kublk: tie sqe allocation to io instead of queue Uday Shankar
2025-05-29 23:47 ` [PATCH v8 4/9] selftests: ublk: kublk: lift queue initialization out of thread Uday Shankar
2025-05-29 23:47 ` [PATCH v8 5/9] selftests: ublk: kublk: move per-thread data out of ublk_queue Uday Shankar
2025-05-29 23:47 ` [PATCH v8 6/9] selftests: ublk: kublk: decouple ublk_queues from ublk server threads Uday Shankar
2025-05-29 23:47 ` [PATCH v8 7/9] selftests: ublk: add functional test for per io daemons Uday Shankar
2025-05-29 23:47 ` [PATCH v8 8/9] selftests: ublk: add stress " Uday Shankar
2025-05-30 2:13 ` Ming Lei [this message]
2025-05-29 23:47 ` [PATCH v8 9/9] Documentation: ublk: document UBLK_F_PER_IO_DAEMON Uday Shankar
2025-05-30 13:07 ` [PATCH v8 0/9] ublk: decouple server threads from ublk_queues/hctxs Jens Axboe
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=aDkULJG9zwqCWqTk@fedora \
--to=ming.lei@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=axboe@kernel.dk \
--cc=corbet@lwn.net \
--cc=csander@purestorage.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=shuah@kernel.org \
--cc=ushankar@purestorage.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).