From: Nicolin Chen <nicolinc@nvidia•com>
To: Pranjal Shrivastava <praan@google•com>
Cc: <iommu@lists•linux.dev>, Will Deacon <will@kernel•org>,
Joerg Roedel <joro@8bytes•org>,
Robin Murphy <robin.murphy@arm•com>,
Jason Gunthorpe <jgg@ziepe•ca>,
Mostafa Saleh <smostafa@google•com>,
Daniel Mentz <danielmentz@google•com>,
Ashish Mhetre <amhetre@nvidia•com>,
<linux-arm-kernel@lists•infradead.org>
Subject: Re: [PATCH v7 10/11] iommu/arm-smmu-v3: Invoke pm_runtime before hw access
Date: Thu, 28 May 2026 15:01:13 -0700 [thread overview]
Message-ID: <ahi7KWov1+GC3xVx@Asurada-Nvidia> (raw)
In-Reply-To: <ahi3uY9YZZtR1Lq5@google.com>
On Thu, May 28, 2026 at 09:46:33PM +0000, Pranjal Shrivastava wrote:
> On Thu, May 28, 2026 at 01:28:15PM -0700, Nicolin Chen wrote:
> > On Wed, May 27, 2026 at 10:14:06PM +0000, Pranjal Shrivastava wrote:
> > > TLB and CFG invalidations are
> > > elided if the SMMU is suspended by observing the CMDQ_PROD_STOP_FLAG via
> > > the arm_smmu_can_elide() helper.
> >
> > All the arm_smmu_can_elide() call sites here would eventually elide
> > the commands in arm_smmu_cmdq_issue_cmdlist() that is already gated
> > by CMDQ_PROD_STOP_FLAG? It doesn't seem necessary to gate again?
>
> While issue_cmdlist() would eventually elide these commands, the
> can_elide() check is necessary to return early during suspension.
>
> This avoids unnecessary stack allocation, cmd building, and spinlock
> contention on the cmdq->lock for threads that are anyway about to be
> elided.
We aren't in the perf sensitive path.. most of those aren't going
to be that bad.
arm_smmu_cmdq_shared_lock() on the other hand is taken at step 2,
and the STOP flag in the same function is gated at step 1?
> By dropping these requests immediately, we significantly reduce cacheline
> bouncing and contention during unmap storms.
How significantly, so as to justify invading every command issue()
call site, which would be difficult to maintain? If we really need
an early return, it would be nicer to have a common place at least.
> Furthermore, the early check
> also allows us to specifically trigger the WARN_ON_ONCE() for broken
> devlinks.
This doesn't sound very appealing to me.. And even if there is a
very plausible reason, it should be noted in the commit message.
Nicolin
next prev parent reply other threads:[~2026-05-28 22:02 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-27 22:13 [PATCH v7 00/11] iommu/arm-smmu-v3: Implement Runtime/System Sleep ops Pranjal Shrivastava
2026-05-27 22:13 ` [PATCH v7 01/11] iommu/arm-smmu-v3: Refactor arm_smmu_setup_irqs Pranjal Shrivastava
2026-05-27 22:13 ` [PATCH v7 02/11] iommu/arm-smmu-v3: Add a helper to drain cmd queues Pranjal Shrivastava
2026-05-28 1:35 ` Nicolin Chen
2026-05-28 10:34 ` Pranjal Shrivastava
2026-05-28 22:09 ` Nicolin Chen
2026-05-29 14:32 ` Pranjal Shrivastava
2026-05-27 22:13 ` [PATCH v7 03/11] iommu/tegra241-cmdqv: Add a helper to drain VCMDQs Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 04/11] iommu/tegra241-cmdqv: Restore PROD and CONS after resume Pranjal Shrivastava
2026-05-28 18:14 ` Nicolin Chen
2026-05-27 22:14 ` [PATCH v7 05/11] iommu/arm-smmu-v3: Cache and restore MSI config Pranjal Shrivastava
2026-05-28 18:36 ` Nicolin Chen
2026-05-28 21:57 ` Pranjal Shrivastava
2026-05-28 22:03 ` Nicolin Chen
2026-05-27 22:14 ` [PATCH v7 06/11] iommu/arm-smmu-v3: Handle gerror during suspend Pranjal Shrivastava
2026-05-28 18:53 ` Nicolin Chen
2026-05-28 21:59 ` Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 07/11] iommu/arm-smmu-v3: Add CMDQ_PROD_STOP_FLAG to gate CMDQ submissions Pranjal Shrivastava
2026-05-28 19:41 ` Nicolin Chen
2026-05-28 21:57 ` Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 08/11] iommu/arm-smmu-v3: Implement pm_runtime & system sleep ops Pranjal Shrivastava
2026-05-28 19:39 ` Nicolin Chen
2026-05-28 21:21 ` Pranjal Shrivastava
2026-05-28 22:13 ` Nicolin Chen
2026-05-28 23:30 ` Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 09/11] iommu/arm-smmu-v3: Enable pm_runtime and setup devlinks Pranjal Shrivastava
2026-05-28 20:13 ` Nicolin Chen
2026-05-28 21:36 ` Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 10/11] iommu/arm-smmu-v3: Invoke pm_runtime before hw access Pranjal Shrivastava
2026-05-28 20:28 ` Nicolin Chen
2026-05-28 21:46 ` Pranjal Shrivastava
2026-05-28 22:01 ` Nicolin Chen [this message]
2026-05-28 22:25 ` Pranjal Shrivastava
2026-05-28 23:18 ` Nicolin Chen
2026-05-29 14:48 ` Pranjal Shrivastava
2026-06-03 22:18 ` Daniel Mentz
2026-06-04 7:18 ` Pranjal Shrivastava
2026-05-27 22:14 ` [PATCH v7 11/11] iommu/arm-smmu-v3: Add KUnit unit tests for Runtime PM Pranjal Shrivastava
2026-05-28 21:43 ` Nicolin Chen
2026-05-28 23:10 ` Pranjal Shrivastava
2026-05-28 23:21 ` Nicolin Chen
2026-05-28 23:33 ` Pranjal Shrivastava
2026-05-28 18:05 ` [PATCH v7 00/11] iommu/arm-smmu-v3: Implement Runtime/System Sleep ops Nicolin Chen
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=ahi7KWov1+GC3xVx@Asurada-Nvidia \
--to=nicolinc@nvidia$(echo .)com \
--cc=amhetre@nvidia$(echo .)com \
--cc=danielmentz@google$(echo .)com \
--cc=iommu@lists$(echo .)linux.dev \
--cc=jgg@ziepe$(echo .)ca \
--cc=joro@8bytes$(echo .)org \
--cc=linux-arm-kernel@lists$(echo .)infradead.org \
--cc=praan@google$(echo .)com \
--cc=robin.murphy@arm$(echo .)com \
--cc=smostafa@google$(echo .)com \
--cc=will@kernel$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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