From: Tejun Heo <tj@kernel.org>
To: jiangshanlai@gmail.com
Cc: linux-kernel@vger.kernel.org, kernel-team@meta.com
Subject: [PATCHSET v2 wq/for-6.5-cleanup-ordered] workqueue: Ordered workqueue creation cleanup
Date: Mon, 8 May 2023 15:50:19 -1000 [thread overview]
Message-ID: <20230509015032.3768622-1-tj@kernel.org> (raw)
Hello,
v2: Acked patches are applied to wq/for-6.5-cleanup-ordered. Some conversion
patches were dropped (e.g. because they were using WQ_SYSFS and thus
can't be ordered) and fixed. The final patch to remove implicit ordered
promotion logic was broken and could trigger WARN spuriously. Fixed.
When multiple work items are queued to a workqueue, their execution order
doesn't match the queueing order. They may get executed in any order and
simultaneously. When fully serialized execution - one by one in the queueing
order - is needed, an ordered workqueue should be used which can be created
with alloc_ordered_workqueue().
However, alloc_ordered_workqueue() was a later addition. Before it, an
ordered workqueue could be obtained by creating an UNBOUND workqueue with
@max_active==1. This originally was an implementation side-effect which was
broken by 4c16bd327c74 ("workqueue: restore WQ_UNBOUND/max_active==1 to be
ordered"). Because there were users that depended on the ordered execution,
5c0338c68706 ("workqueue: restore WQ_UNBOUND/max_active==1 to be ordered")
made workqueue allocation path to implicitly promote UNBOUND workqueues w/
@max_active==1 to ordered workqueues.
While this has worked okay, overloading the UNBOUND allocation interface
this way creates other issues. It's difficult to tell whether a given
workqueue actually needs to be ordered and users that legitimately want a
min concurrency level wq unexpectedly gets an ordered one instead. With
planned UNBOUND workqueue updates to improve execution locality and more
prevalence of chiplet designs which can benefit from such improvements, this
isn't a state we wanna be in forever.
This patch series audits all callsites that create an UNBOUND workqueue w/
@max_active==1 and converts them to alloc_ordered_workqueue() as necessary
and contains the following 13 patches on top of wq/for-6.5-cleanup-ordered
branch.
0001-scsi-ncr53c8xx-Use-default-max_active-for-hostdata-w.patch
0002-wifi-mwifiex-Use-default-max_active-for-workqueues.patch
0003-dm-integrity-Use-alloc_ordered_workqueue-to-create-o.patch
0004-media-amphion-Use-alloc_ordered_workqueue-to-create-.patch
0005-wifi-ath10-11-12k-Use-alloc_ordered_workqueue-to-cre.patch
0006-net-wwan-t7xx-Use-alloc_ordered_workqueue-to-create-.patch
0007-soc-qcom-qmi-Use-alloc_ordered_workqueue-to-create-o.patch
0008-btrfs-Use-alloc_ordered_workqueue-to-create-ordered-.patch
0009-net-qrtr-Use-alloc_ordered_workqueue-to-create-order.patch
0010-rxrpc-Use-alloc_ordered_workqueue-to-create-ordered-.patch
0011-crypto-octeontx2-Use-alloc_ordered_workqueue-to-crea.patch
0012-media-coda-Use-alloc_ordered_workqueue-to-create-ord.patch
0013-workqueue-Don-t-implicitly-make-UNBOUND-workqueues-w.patch
0001-0012 convert the existing users and 0013 drops the implicit ordered
promotion logic from alloc_workqueue(). I'll keep an eye out for a while
after merging 0022. Thankfully, these are pretty easy to grep for. The
patches can also be found in the following git branch.
git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq.git cleanup-ordered-v2
diffstat follows. Thanks.
drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c | 12 ++++++------
drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c | 6 +++---
drivers/md/dm-integrity.c | 4 ++--
drivers/md/dm.c | 2 +-
drivers/media/platform/amphion/vpu_core.c | 2 +-
drivers/media/platform/amphion/vpu_v4l2.c | 2 +-
drivers/media/platform/chips-media/coda-common.c | 2 +-
drivers/net/wireless/ath/ath10k/qmi.c | 3 +--
drivers/net/wireless/ath/ath11k/qmi.c | 3 +--
drivers/net/wireless/ath/ath12k/qmi.c | 3 +--
drivers/net/wireless/marvell/mwifiex/cfg80211.c | 4 ++--
drivers/net/wireless/marvell/mwifiex/main.c | 8 ++++----
drivers/net/wwan/t7xx/t7xx_hif_cldma.c | 13 +++++++------
drivers/net/wwan/t7xx/t7xx_hif_dpmaif_tx.c | 5 +++--
drivers/scsi/NCR5380.c | 2 +-
drivers/soc/qcom/qmi_interface.c | 2 +-
fs/btrfs/disk-io.c | 2 +-
fs/btrfs/scrub.c | 6 ++++--
include/linux/workqueue.h | 4 +---
kernel/workqueue.c | 23 ++++-------------------
net/qrtr/ns.c | 2 +-
net/rxrpc/af_rxrpc.c | 2 +-
22 files changed, 48 insertions(+), 64 deletions(-)
next reply other threads:[~2023-05-09 1:50 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-09 1:50 Tejun Heo [this message]
2023-05-09 1:50 ` [PATCH 01/13] scsi: ncr53c8xx: Use default @max_active for hostdata->work_q Tejun Heo
2023-05-21 2:48 ` [PATCH RESEND " Tejun Heo
2023-05-21 6:42 ` Finn Thain
2023-05-22 22:06 ` Martin K. Petersen
2023-05-23 0:58 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 02/13] wifi: mwifiex: Use default @max_active for workqueues Tejun Heo
2023-05-10 8:45 ` Kalle Valo
2023-05-10 18:09 ` Brian Norris
2023-05-10 18:16 ` Tejun Heo
2023-05-10 18:57 ` Brian Norris
2023-05-10 19:19 ` Tejun Heo
2023-05-10 19:50 ` Brian Norris
2023-05-19 0:36 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 03/13] dm integrity: Use alloc_ordered_workqueue() to create ordered workqueues Tejun Heo
2023-05-25 22:10 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 04/13] media: amphion: " Tejun Heo
2023-05-23 12:19 ` Hans Verkuil
2023-05-23 21:25 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 05/13] wifi: ath10/11/12k: " Tejun Heo
2023-05-10 8:41 ` Kalle Valo
2023-05-19 0:41 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 06/13] net: wwan: t7xx: " Tejun Heo
2023-05-25 22:11 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 07/13] soc: qcom: qmi: " Tejun Heo
2023-05-09 1:50 ` [PATCH 08/13] btrfs: " Tejun Heo
2023-05-09 14:53 ` David Sterba
2023-05-09 15:57 ` Tejun Heo
2023-05-09 23:36 ` David Sterba
2023-05-09 23:47 ` Tejun Heo
2023-05-25 23:33 ` Tejun Heo
2023-05-26 12:52 ` David Sterba
2023-05-09 1:50 ` [PATCH 09/13] net: qrtr: " Tejun Heo
2023-05-25 22:14 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 10/13] rxrpc: " Tejun Heo
2023-05-25 22:14 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 11/13] crypto: octeontx2: " Tejun Heo
2023-05-09 2:18 ` Herbert Xu
2023-05-19 0:42 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 12/13] media: coda: " Tejun Heo
2023-05-09 10:46 ` Philipp Zabel
2023-05-19 0:50 ` Tejun Heo
2023-05-09 1:50 ` [PATCH 13/13] workqueue: Don't implicitly make UNBOUND workqueues w/ @max_active==1 ordered Tejun Heo
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=20230509015032.3768622-1-tj@kernel.org \
--to=tj@kernel.org \
--cc=jiangshanlai@gmail.com \
--cc=kernel-team@meta.com \
--cc=linux-kernel@vger.kernel.org \
/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