From: Andrzej Hajda <andrzej.hajda@intel.com>
To: Jani Nikula <jani.nikula@linux.intel.com>,
Joonas Lahtinen <joonas.lahtinen@linux.intel.com>,
Rodrigo Vivi <rodrigo.vivi@intel.com>,
Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>,
David Airlie <airlied@gmail.com>, Daniel Vetter <daniel@ffwll.ch>
Cc: netdev@vger.kernel.org, intel-gfx@lists.freedesktop.org,
linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
Chris Wilson <chris@chris-wilson.co.uk>,
Eric Dumazet <edumazet@google.com>,
Chris Wilson <chris.p.wilson@intel.com>,
Andrzej Hajda <andrzej.hajda@intel.com>,
Jakub Kicinski <kuba@kernel.org>,
Dmitry Vyukov <dvyukov@google.com>
Subject: [PATCH v4 10/10] drm/i915/gt: Hold a wakeref for the active VM
Date: Mon, 06 Mar 2023 17:32:06 +0100 [thread overview]
Message-ID: <20230224-track_gt-v4-10-464e8ab4c9ab@intel.com> (raw)
In-Reply-To: <20230224-track_gt-v4-0-464e8ab4c9ab@intel.com>
From: Chris Wilson <chris@chris-wilson.co.uk>
There may be a disconnect between the GT used by the engine and the GT
used for the VM, requiring us to hold a wakeref on both while the GPU is
active with this request.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
---
drivers/gpu/drm/i915/gt/intel_context.h | 15 +++++++++++----
drivers/gpu/drm/i915/gt/intel_context_types.h | 2 ++
drivers/gpu/drm/i915/gt/intel_engine_pm.c | 4 ++++
3 files changed, 17 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h
index 0a8d553da3f439..582faa21181e58 100644
--- a/drivers/gpu/drm/i915/gt/intel_context.h
+++ b/drivers/gpu/drm/i915/gt/intel_context.h
@@ -14,6 +14,7 @@
#include "i915_drv.h"
#include "intel_context_types.h"
#include "intel_engine_types.h"
+#include "intel_gt_pm.h"
#include "intel_ring_types.h"
#include "intel_timeline_types.h"
#include "i915_trace.h"
@@ -207,8 +208,11 @@ void intel_context_exit_engine(struct intel_context *ce);
static inline void intel_context_enter(struct intel_context *ce)
{
lockdep_assert_held(&ce->timeline->mutex);
- if (!ce->active_count++)
- ce->ops->enter(ce);
+ if (ce->active_count++)
+ return;
+
+ ce->ops->enter(ce);
+ ce->wakeref = intel_gt_pm_get(ce->vm->gt);
}
static inline void intel_context_mark_active(struct intel_context *ce)
@@ -222,8 +226,11 @@ static inline void intel_context_exit(struct intel_context *ce)
{
lockdep_assert_held(&ce->timeline->mutex);
GEM_BUG_ON(!ce->active_count);
- if (!--ce->active_count)
- ce->ops->exit(ce);
+ if (--ce->active_count)
+ return;
+
+ intel_gt_pm_put_async(ce->vm->gt, ce->wakeref);
+ ce->ops->exit(ce);
}
static inline struct intel_context *intel_context_get(struct intel_context *ce)
diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
index e36670f2e6260b..5dc39a9d7a501c 100644
--- a/drivers/gpu/drm/i915/gt/intel_context_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
@@ -17,6 +17,7 @@
#include "i915_utils.h"
#include "intel_engine_types.h"
#include "intel_sseu.h"
+#include "intel_wakeref.h"
#include "uc/intel_guc_fwif.h"
@@ -110,6 +111,7 @@ struct intel_context {
u32 ring_size;
struct intel_ring *ring;
struct intel_timeline *timeline;
+ intel_wakeref_t wakeref;
unsigned long flags;
#define CONTEXT_BARRIER_BIT 0
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
index 7063dea2112943..c2d17c97bfe989 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
@@ -114,6 +114,10 @@ __queue_and_release_pm(struct i915_request *rq,
ENGINE_TRACE(engine, "parking\n");
+ GEM_BUG_ON(rq->context->active_count != 1);
+ __intel_gt_pm_get(engine->gt);
+ rq->context->wakeref = intel_wakeref_track(&engine->gt->wakeref);
+
/*
* We have to serialise all potential retirement paths with our
* submission, as we don't want to underflow either the
--
2.34.1
next prev parent reply other threads:[~2023-03-06 16:33 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-06 16:31 [PATCH v4 00/10] drm/i915: use ref_tracker library for tracking wakerefs Andrzej Hajda
2023-03-06 16:31 ` [PATCH v4 01/10] lib/ref_tracker: add unlocked leak print helper Andrzej Hajda
2023-03-19 22:59 ` [Intel-gfx] " Andi Shyti
2023-03-20 15:30 ` Andrzej Hajda
2023-03-06 16:31 ` [PATCH v4 02/10] lib/ref_tracker: __ref_tracker_dir_print improve printing Andrzej Hajda
2023-03-19 23:11 ` [Intel-gfx] " Andi Shyti
2023-03-20 15:39 ` Andrzej Hajda
2023-03-06 16:31 ` [PATCH v4 03/10] lib/ref_tracker: add printing to memory buffer Andrzej Hajda
2023-03-19 23:18 ` [Intel-gfx] " Andi Shyti
2023-03-20 18:24 ` Andrzej Hajda
2023-03-06 16:32 ` [PATCH v4 04/10] lib/ref_tracker: remove warnings in case of allocation failure Andrzej Hajda
2023-03-19 23:19 ` [Intel-gfx] " Andi Shyti
2023-03-06 16:32 ` [PATCH v4 05/10] drm/i915: Separate wakeref tracking types from rpm Andrzej Hajda
2023-03-06 16:32 ` [PATCH v4 06/10] drm/i915: Separate wakeref tracking Andrzej Hajda
2023-03-19 23:57 ` [Intel-gfx] " Andi Shyti
2023-03-20 18:39 ` Andrzej Hajda
2023-03-06 16:32 ` [PATCH v4 07/10] drm/i915: Track leaked gt->wakerefs Andrzej Hajda
2023-03-06 16:32 ` [PATCH v4 08/10] drm/i915: Correct type of wakeref variable Andrzej Hajda
2023-03-19 23:58 ` [Intel-gfx] " Andi Shyti
2023-03-06 16:32 ` [PATCH v4 09/10] drm/i915: replace Intel internal tracker with kernel core ref_tracker Andrzej Hajda
2023-03-06 16:32 ` Andrzej Hajda [this message]
2023-03-17 11:58 ` [Intel-gfx] [PATCH v4 00/10] drm/i915: use ref_tracker library for tracking wakerefs Andrzej Hajda
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=20230224-track_gt-v4-10-464e8ab4c9ab@intel.com \
--to=andrzej.hajda@intel.com \
--cc=airlied@gmail.com \
--cc=chris.p.wilson@intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=dvyukov@google.com \
--cc=edumazet@google.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jani.nikula@linux.intel.com \
--cc=joonas.lahtinen@linux.intel.com \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=rodrigo.vivi@intel.com \
--cc=tvrtko.ursulin@linux.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).