From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
To: Matthew Brost <matthew.brost@intel.com>,
<intel-gfx@lists.freedesktop.org>,
<dri-devel@lists.freedesktop.org>
Cc: <daniel.vetter@ffwll.ch>
Subject: Re: [Intel-gfx] [PATCH 22/27] drm/i915/guc: Drop pin count check trick between sched_disable and re-pin
Date: Wed, 25 Aug 2021 17:50:24 -0700 [thread overview]
Message-ID: <884337fb-191e-ee9c-9f65-3b601eb3c97d@intel.com> (raw)
In-Reply-To: <20210819061639.21051-23-matthew.brost@intel.com>
On 8/18/2021 11:16 PM, Matthew Brost wrote:
> Drop pin count check trick between a sched_disable and re-pin, now rely
> on the lock and counter of the number of committed requests to determine
> if scheduling should be disabled on the context.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Daniele
> ---
> drivers/gpu/drm/i915/gt/intel_context_types.h | 2 +
> .../gpu/drm/i915/gt/uc/intel_guc_submission.c | 49 ++++++++++++-------
> 2 files changed, 34 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
> index d5d643b04d54..524a35a78bf4 100644
> --- a/drivers/gpu/drm/i915/gt/intel_context_types.h
> +++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
> @@ -169,6 +169,8 @@ struct intel_context {
> struct list_head fences;
> /* GuC context blocked fence */
> struct i915_sw_fence blocked_fence;
> + /* GuC committed requests */
> + int number_committed_requests;
> } guc_state;
>
> struct {
> diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
> index 5f77f25322ca..3e90985b0c1b 100644
> --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
> +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
> @@ -248,6 +248,25 @@ static inline void decr_context_blocked(struct intel_context *ce)
> ce->guc_state.sched_state -= SCHED_STATE_BLOCKED;
> }
>
> +static inline bool context_has_committed_requests(struct intel_context *ce)
> +{
> + return !!ce->guc_state.number_committed_requests;
> +}
> +
> +static inline void incr_context_committed_requests(struct intel_context *ce)
> +{
> + lockdep_assert_held(&ce->guc_state.lock);
> + ++ce->guc_state.number_committed_requests;
> + GEM_BUG_ON(ce->guc_state.number_committed_requests < 0);
> +}
> +
> +static inline void decr_context_committed_requests(struct intel_context *ce)
> +{
> + lockdep_assert_held(&ce->guc_state.lock);
> + --ce->guc_state.number_committed_requests;
> + GEM_BUG_ON(ce->guc_state.number_committed_requests < 0);
> +}
> +
> static inline bool context_guc_id_invalid(struct intel_context *ce)
> {
> return ce->guc_id == GUC_INVALID_LRC_ID;
> @@ -1751,14 +1770,11 @@ static void guc_context_sched_disable(struct intel_context *ce)
> spin_lock_irqsave(&ce->guc_state.lock, flags);
>
> /*
> - * We have to check if the context has been disabled by another thread.
> - * We also have to check if the context has been pinned again as another
> - * pin operation is allowed to pass this function. Checking the pin
> - * count, within ce->guc_state.lock, synchronizes this function with
> - * guc_request_alloc ensuring a request doesn't slip through the
> - * 'context_pending_disable' fence. Checking within the spin lock (can't
> - * sleep) ensures another process doesn't pin this context and generate
> - * a request before we set the 'context_pending_disable' flag here.
> + * We have to check if the context has been disabled by another thread,
> + * check if submssion has been disabled to seal a race with reset and
> + * finally check if any more requests have been committed to the
> + * context ensursing that a request doesn't slip through the
> + * 'context_pending_disable' fence.
> */
> enabled = context_enabled(ce);
> if (unlikely(!enabled || submission_disabled(guc))) {
> @@ -1767,7 +1783,8 @@ static void guc_context_sched_disable(struct intel_context *ce)
> spin_unlock_irqrestore(&ce->guc_state.lock, flags);
> goto unpin;
> }
> - if (unlikely(atomic_add_unless(&ce->pin_count, -2, 2))) {
> + if (unlikely(context_has_committed_requests(ce))) {
> + intel_context_sched_disable_unpin(ce);
> spin_unlock_irqrestore(&ce->guc_state.lock, flags);
> return;
> }
> @@ -1800,6 +1817,7 @@ static void __guc_context_destroy(struct intel_context *ce)
> ce->guc_prio_count[GUC_CLIENT_PRIORITY_HIGH] ||
> ce->guc_prio_count[GUC_CLIENT_PRIORITY_KMD_NORMAL] ||
> ce->guc_prio_count[GUC_CLIENT_PRIORITY_NORMAL]);
> + GEM_BUG_ON(ce->guc_state.number_committed_requests);
>
> lrc_fini(ce);
> intel_context_fini(ce);
> @@ -2030,6 +2048,10 @@ static void remove_from_context(struct i915_request *rq)
>
> spin_unlock_irq(&ce->guc_active.lock);
>
> + spin_lock_irq(&ce->guc_state.lock);
> + decr_context_committed_requests(ce);
> + spin_unlock_irq(&ce->guc_state.lock);
> +
> atomic_dec(&ce->guc_id_ref);
> i915_request_notify_execute_cb_imm(rq);
> }
> @@ -2177,15 +2199,7 @@ static int guc_request_alloc(struct i915_request *rq)
> * schedule enable or context registration if either G2H is pending
> * respectfully. Once a G2H returns, the fence is released that is
> * blocking these requests (see guc_signal_context_fence).
> - *
> - * We can safely check the below fields outside of the lock as it isn't
> - * possible for these fields to transition from being clear to set but
> - * converse is possible, hence the need for the check within the lock.
> */
> - if (likely(!context_wait_for_deregister_to_register(ce) &&
> - !context_pending_disable(ce)))
> - return 0;
> -
> spin_lock_irqsave(&ce->guc_state.lock, flags);
> if (context_wait_for_deregister_to_register(ce) ||
> context_pending_disable(ce)) {
> @@ -2194,6 +2208,7 @@ static int guc_request_alloc(struct i915_request *rq)
>
> list_add_tail(&rq->guc_fence_link, &ce->guc_state.fences);
> }
> + incr_context_committed_requests(ce);
> spin_unlock_irqrestore(&ce->guc_state.lock, flags);
>
> return 0;
next prev parent reply other threads:[~2021-08-26 0:50 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-19 6:16 [Intel-gfx] [PATCH 00/27] Clean up GuC CI failures, simplify locking, and kernel DOC Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 01/27] drm/i915/guc: Fix blocked context accounting Matthew Brost
2021-08-24 23:24 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 02/27] drm/i915/guc: Fix outstanding G2H accounting Matthew Brost
2021-08-19 21:31 ` Daniele Ceraolo Spurio
2021-08-19 21:30 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 03/27] drm/i915/guc: Unwind context requests in reverse order Matthew Brost
2021-08-19 23:54 ` Daniele Ceraolo Spurio
2021-08-19 23:53 ` Matthew Brost
2021-08-20 0:03 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 04/27] drm/i915/guc: Don't drop ce->guc_active.lock when unwinding context Matthew Brost
2021-08-20 0:01 ` Daniele Ceraolo Spurio
2021-08-19 23:58 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 05/27] drm/i915/guc: Process all G2H message at once in work queue Matthew Brost
2021-08-20 0:06 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 06/27] drm/i915/guc: Workaround reset G2H is received after schedule done G2H Matthew Brost
2021-08-24 23:31 ` Daniele Ceraolo Spurio
2021-08-25 4:05 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 07/27] Revert "drm/i915/gt: Propagate change in error status to children on unhold" Matthew Brost
2021-08-20 19:47 ` Jason Ekstrand
2021-08-19 6:16 ` [Intel-gfx] [PATCH 08/27] drm/i915/selftests: Add a cancel request selftest that triggers a reset Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 09/27] drm/i915/guc: Kick tasklet after queuing a request Matthew Brost
2021-08-20 18:31 ` Daniele Ceraolo Spurio
2021-08-20 18:36 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 10/27] drm/i915/guc: Don't enable scheduling on a banned context, guc_id invalid, not registered Matthew Brost
2021-08-20 18:42 ` Daniele Ceraolo Spurio
2021-08-20 18:42 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 11/27] drm/i915/selftests: Fix memory corruption in live_lrc_isolation Matthew Brost
2021-08-25 0:07 ` Daniele Ceraolo Spurio
2021-08-25 20:03 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 12/27] drm/i915/selftests: Add initial GuC selftest for scrubbing lost G2H Matthew Brost
2021-08-25 0:58 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 13/27] drm/i915/guc: Take context ref when cancelling request Matthew Brost
2021-08-21 0:07 ` Daniele Ceraolo Spurio
2021-08-24 15:42 ` Matthew Brost
2021-08-25 1:21 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 14/27] drm/i915/guc: Don't touch guc_state.sched_state without a lock Matthew Brost
2021-08-25 1:20 ` Daniele Ceraolo Spurio
2021-08-25 1:44 ` Matthew Brost
2021-08-25 1:51 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 15/27] drm/i915/guc: Reset LRC descriptor if register returns -ENODEV Matthew Brost
2021-08-21 0:14 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 16/27] drm/i915: Allocate error capture in nowait context Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 17/27] drm/i915/guc: Flush G2H work queue during reset Matthew Brost
2021-08-21 0:25 ` Daniele Ceraolo Spurio
2021-08-24 15:44 ` Matthew Brost
2021-08-25 1:22 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 18/27] drm/i915/guc: Release submit fence from an irq_work Matthew Brost
2021-08-25 1:44 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 19/27] drm/i915/guc: Move guc_blocked fence to struct guc_state Matthew Brost
2021-08-21 0:30 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 20/27] drm/i915/guc: Rework and simplify locking Matthew Brost
2021-08-25 16:52 ` Daniele Ceraolo Spurio
2021-08-25 19:22 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 21/27] drm/i915/guc: Proper xarray usage for contexts_lookup Matthew Brost
2021-08-26 0:44 ` Daniele Ceraolo Spurio
2021-08-26 0:41 ` Matthew Brost
2021-08-26 0:48 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 22/27] drm/i915/guc: Drop pin count check trick between sched_disable and re-pin Matthew Brost
2021-08-26 0:50 ` Daniele Ceraolo Spurio [this message]
2021-08-19 6:16 ` [Intel-gfx] [PATCH 23/27] drm/i915/guc: Move GuC priority fields in context under guc_active Matthew Brost
2021-08-25 21:51 ` Daniele Ceraolo Spurio
2021-08-25 22:53 ` Matthew Brost
2021-08-25 23:04 ` Matthew Brost
2021-08-19 6:16 ` [Intel-gfx] [PATCH 24/27] drm/i915/guc: Move fields protected by guc->contexts_lock into sub structure Matthew Brost
2021-08-25 2:00 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 25/27] drm/i915/guc: Drop guc_active move everything into guc_state Matthew Brost
2021-08-26 0:54 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 26/27] drm/i915/guc: Add GuC kernel doc Matthew Brost
2021-08-26 1:03 ` Daniele Ceraolo Spurio
2021-08-19 6:16 ` [Intel-gfx] [PATCH 27/27] drm/i915/guc: Drop static inline functions intel_guc_submission.c Matthew Brost
2021-08-19 7:18 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for Clean up GuC CI failures, simplify locking, and kernel DOC (rev3) Patchwork
2021-08-19 7:20 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-08-19 7:51 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-08-19 9:08 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
2021-08-26 3:23 [Intel-gfx] [PATCH 00/27] Clean up GuC CI failures, simplify locking, and kernel DOC Matthew Brost
2021-08-26 3:23 ` [Intel-gfx] [PATCH 22/27] drm/i915/guc: Drop pin count check trick between sched_disable and re-pin Matthew Brost
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=884337fb-191e-ee9c-9f65-3b601eb3c97d@intel.com \
--to=daniele.ceraolospurio@intel.com \
--cc=daniel.vetter@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=matthew.brost@intel.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).