On 09/05/2023 03:50, Tejun Heo wrote:
> BACKGROUND
> ==========
>
> 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").
That's the wrong patch description, it should be:
4c16bd327c74 ("workqueue: implement NUMA affinity for unbound workqueues")
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.
>
> WHAT TO LOOK FOR
> ================
>
> The conversions are from
>
> alloc_workqueue(WQ_UNBOUND | flags, 1, args..)
>
> to
>
> alloc_ordered_workqueue(flags, args...)
>
> which don't cause any functional changes. If you know that fully ordered
> execution is not ncessary, please let me know. I'll drop the conversion and
ncessary -> necessary
> instead add a comment noting the fact to reduce confusion while conversion
> is in progress.
>
> If you aren't fully sure, it's completely fine to let the conversion
> through. The behavior will stay exactly the same and we can always
> reconsider later.
>
> As there are follow-up workqueue core changes, I'd really appreciate if the
> patch can be routed through the workqueue tree w/ your acks. Thanks.
Note that the max line length of 75 is exceeded in this commit message.
>
> Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
> Cc: Ming Qian <ming.qian@xxxxxxx>
> Cc: Shijie Qin <shijie.qin@xxxxxxx>
> Cc: Zhou Peng <eagle.zhou@xxxxxxx>
> Cc: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx>
> Cc: linux-media@xxxxxxxxxxxxxxx
> ---
> drivers/media/platform/amphion/vpu_core.c | 2 +-
> drivers/media/platform/amphion/vpu_v4l2.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/media/platform/amphion/vpu_core.c b/drivers/media/platform/amphion/vpu_core.c
> index de23627a119a..43d85a54268b 100644
> --- a/drivers/media/platform/amphion/vpu_core.c
> +++ b/drivers/media/platform/amphion/vpu_core.c
> @@ -254,7 +254,7 @@ static int vpu_core_register(struct device *dev, struct vpu_core *core)
> if (vpu_core_is_exist(vpu, core))
> return 0;
>
> - core->workqueue = alloc_workqueue("vpu", WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
> + core->workqueue = alloc_ordered_workqueue("vpu", WQ_MEM_RECLAIM);
> if (!core->workqueue) {
> dev_err(core->dev, "fail to alloc workqueue\n");
> return -ENOMEM;
> diff --git a/drivers/media/platform/amphion/vpu_v4l2.c b/drivers/media/platform/amphion/vpu_v4l2.c
> index 6773b885597c..a48edb445eea 100644
> --- a/drivers/media/platform/amphion/vpu_v4l2.c
> +++ b/drivers/media/platform/amphion/vpu_v4l2.c
> @@ -740,7 +740,7 @@ int vpu_v4l2_open(struct file *file, struct vpu_inst *inst)
> inst->fh.ctrl_handler = &inst->ctrl_handler;
> file->private_data = &inst->fh;
> inst->state = VPU_CODEC_STATE_DEINIT;
> - inst->workqueue = alloc_workqueue("vpu_inst", WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
> + inst->workqueue = alloc_ordered_workqueue("vpu_inst", WQ_MEM_RECLAIM);
> if (inst->workqueue) {
> INIT_WORK(&inst->msg_work, vpu_inst_run_work);
> ret = kfifo_init(&inst->msg_fifo,
Acked-by: Hans Verkuil <hverkuil-cisco@xxxxxxxxx>
Feel free to take this with the commit log changes mentioned above.
Regards,
Hans
