| Message ID | 20230509015032.3768622-12-tj@kernel.org (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | Herbert Xu |
| Headers | show |
| Series | None | expand |
On Mon, May 08, 2023 at 03:50:30PM -1000, 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"). 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 > 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. > > Signed-off-by: Tejun Heo <tj@kernel.org> > Cc: Boris Brezillon <bbrezillon@kernel.org> > Cc: Arnaud Ebalard <arno@natisbad.org> > Cc: Srujana Challa <schalla@marvell.com> > Cc: Herbert Xu <herbert@gondor.apana.org.au> > Cc: "David S. Miller" <davem@davemloft.net> > Cc: Shijith Thotton <sthotton@marvell.com> > Cc: Vladis Dronov <vdronov@redhat.com> > Cc: Vincent Mailhol <mailhol.vincent@wanadoo.fr> > Cc: Wolfram Sang <wsa+renesas@sang-engineering.com> > Cc: Alexander Lobakin <alobakin@pm.me> > Cc: Minghao Chi <chi.minghao@zte.com.cn> > Cc: ye xingchen <ye.xingchen@zte.com.cn> > Cc: linux-crypto@vger.kernel.org > --- > drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c | 12 ++++++------ > drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c | 6 +++--- > 2 files changed, 9 insertions(+), 9 deletions(-) Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Applied to wq/for-6.5-cleanup-ordered. Thanks.
diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c b/drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c index ddf6e913c1c4..30e6acfc93d9 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c @@ -357,9 +357,9 @@ static int cptpf_vfpf_mbox_init(struct otx2_cptpf_dev *cptpf, int num_vfs) u64 vfpf_mbox_base; int err, i; - cptpf->vfpf_mbox_wq = alloc_workqueue("cpt_vfpf_mailbox", - WQ_UNBOUND | WQ_HIGHPRI | - WQ_MEM_RECLAIM, 1); + cptpf->vfpf_mbox_wq = + alloc_ordered_workqueue("cpt_vfpf_mailbox", + WQ_HIGHPRI | WQ_MEM_RECLAIM); if (!cptpf->vfpf_mbox_wq) return -ENOMEM; @@ -453,9 +453,9 @@ static int cptpf_afpf_mbox_init(struct otx2_cptpf_dev *cptpf) resource_size_t offset; int err; - cptpf->afpf_mbox_wq = alloc_workqueue("cpt_afpf_mailbox", - WQ_UNBOUND | WQ_HIGHPRI | - WQ_MEM_RECLAIM, 1); + cptpf->afpf_mbox_wq = + alloc_ordered_workqueue("cpt_afpf_mailbox", + WQ_HIGHPRI | WQ_MEM_RECLAIM); if (!cptpf->afpf_mbox_wq) return -ENOMEM; diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c b/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c index 392e9fee05e8..6023a7adb70c 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c @@ -75,9 +75,9 @@ static int cptvf_pfvf_mbox_init(struct otx2_cptvf_dev *cptvf) resource_size_t offset, size; int ret; - cptvf->pfvf_mbox_wq = alloc_workqueue("cpt_pfvf_mailbox", - WQ_UNBOUND | WQ_HIGHPRI | - WQ_MEM_RECLAIM, 1); + cptvf->pfvf_mbox_wq = + alloc_ordered_workqueue("cpt_pfvf_mailbox", + WQ_HIGHPRI | WQ_MEM_RECLAIM); if (!cptvf->pfvf_mbox_wq) return -ENOMEM;
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"). 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 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. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Boris Brezillon <bbrezillon@kernel.org> Cc: Arnaud Ebalard <arno@natisbad.org> Cc: Srujana Challa <schalla@marvell.com> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: "David S. Miller" <davem@davemloft.net> Cc: Shijith Thotton <sthotton@marvell.com> Cc: Vladis Dronov <vdronov@redhat.com> Cc: Vincent Mailhol <mailhol.vincent@wanadoo.fr> Cc: Wolfram Sang <wsa+renesas@sang-engineering.com> Cc: Alexander Lobakin <alobakin@pm.me> Cc: Minghao Chi <chi.minghao@zte.com.cn> Cc: ye xingchen <ye.xingchen@zte.com.cn> Cc: linux-crypto@vger.kernel.org --- drivers/crypto/marvell/octeontx2/otx2_cptpf_main.c | 12 ++++++------ drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c | 6 +++--- 2 files changed, 9 insertions(+), 9 deletions(-)