From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: <intel-xe@lists.freedesktop.org>
Cc: "Rodrigo Vivi" <rodrigo.vivi@intel.com>,
"Matthew Brost" <matthew.brost@intel.com>,
"Francois Dugast" <francois.dugast@intel.com>,
"Thomas Hellström" <thomas.hellstrom@linux.intel.com>
Subject: [PATCH 3/7] drm/xe: Relax runtime pm protection during execution
Date: Wed, 8 May 2024 16:07:03 -0400 [thread overview]
Message-ID: <20240508200707.375414-3-rodrigo.vivi@intel.com> (raw)
In-Reply-To: <20240508200707.375414-1-rodrigo.vivi@intel.com>
Limit the protection only during moments of actual job execution,
and introduce protection for guc submit fini, which is currently
unprotected due to the absence of exec_queue life protection.
In the regular use case scenario, user space will create an
exec queue, and keep it alive to reuse that until it is done
with that kind of workload.
For the regular desktop cases, it means that the exec_queue
is alive even on idle scenarios where display goes off. This
is unacceptable since this would entirely block runtime PM
indefinitely, blocking deeper Package-C state. This would be
a waste drainage of power.
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
drivers/gpu/drm/xe/xe_exec_queue.c | 14 --------------
drivers/gpu/drm/xe/xe_guc_submit.c | 3 +++
drivers/gpu/drm/xe/xe_sched_job.c | 10 +++-------
3 files changed, 6 insertions(+), 21 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 395de93579fa..33b03605a1d1 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -106,7 +106,6 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe,
static int __xe_exec_queue_init(struct xe_exec_queue *q)
{
- struct xe_device *xe = gt_to_xe(q->gt);
int i, err;
for (i = 0; i < q->width; ++i) {
@@ -119,17 +118,6 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q)
if (err)
goto err_lrc;
- /*
- * Normally the user vm holds an rpm ref to keep the device
- * awake, and the context holds a ref for the vm, however for
- * some engines we use the kernels migrate vm underneath which offers no
- * such rpm ref, or we lack a vm. Make sure we keep a ref here, so we
- * can perform GuC CT actions when needed. Caller is expected to have
- * already grabbed the rpm ref outside any sensitive locks.
- */
- if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) && (q->flags & EXEC_QUEUE_FLAG_VM || !q->vm))
- xe_pm_runtime_get_noresume(xe);
-
return 0;
err_lrc:
@@ -216,8 +204,6 @@ void xe_exec_queue_fini(struct xe_exec_queue *q)
for (i = 0; i < q->width; ++i)
xe_lrc_finish(q->lrc + i);
- if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) && (q->flags & EXEC_QUEUE_FLAG_VM || !q->vm))
- xe_pm_runtime_put(gt_to_xe(q->gt));
__xe_exec_queue_free(q);
}
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 2c0aa3443cd9..8b1afbe3058c 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -35,6 +35,7 @@
#include "xe_macros.h"
#include "xe_map.h"
#include "xe_mocs.h"
+#include "xe_pm.h"
#include "xe_ring_ops_types.h"
#include "xe_sched_job.h"
#include "xe_trace.h"
@@ -1073,6 +1074,7 @@ static void __guc_exec_queue_fini_async(struct work_struct *w)
struct xe_exec_queue *q = ge->q;
struct xe_guc *guc = exec_queue_to_guc(q);
+ xe_pm_runtime_get(guc_to_xe(guc));
trace_xe_exec_queue_destroy(q);
if (xe_exec_queue_is_lr(q))
@@ -1083,6 +1085,7 @@ static void __guc_exec_queue_fini_async(struct work_struct *w)
kfree(ge);
xe_exec_queue_fini(q);
+ xe_pm_runtime_put(guc_to_xe(guc));
}
static void guc_exec_queue_fini_async(struct xe_exec_queue *q)
diff --git a/drivers/gpu/drm/xe/xe_sched_job.c b/drivers/gpu/drm/xe/xe_sched_job.c
index cd8a2fba5438..a4e030f5e019 100644
--- a/drivers/gpu/drm/xe/xe_sched_job.c
+++ b/drivers/gpu/drm/xe/xe_sched_job.c
@@ -158,11 +158,7 @@ struct xe_sched_job *xe_sched_job_create(struct xe_exec_queue *q,
for (i = 0; i < width; ++i)
job->batch_addr[i] = batch_addr[i];
- /* All other jobs require a VM to be open which has a ref */
- if (unlikely(q->flags & EXEC_QUEUE_FLAG_KERNEL))
- xe_pm_runtime_get_noresume(job_to_xe(job));
- xe_device_assert_mem_access(job_to_xe(job));
-
+ xe_pm_runtime_get_noresume(job_to_xe(job));
trace_xe_sched_job_create(job);
return job;
@@ -191,13 +187,13 @@ void xe_sched_job_destroy(struct kref *ref)
{
struct xe_sched_job *job =
container_of(ref, struct xe_sched_job, refcount);
+ struct xe_device *xe = job_to_xe(job);
- if (unlikely(job->q->flags & EXEC_QUEUE_FLAG_KERNEL))
- xe_pm_runtime_put(job_to_xe(job));
xe_exec_queue_put(job->q);
dma_fence_put(job->fence);
drm_sched_job_cleanup(&job->drm);
job_free(job);
+ xe_pm_runtime_put(xe);
}
void xe_sched_job_set_error(struct xe_sched_job *job, int error)
--
2.44.0
next prev parent reply other threads:[~2024-05-08 20:07 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-08 20:07 [PATCH 1/7] drm/xe: Fix xe_pm_runtime_get_if_active return Rodrigo Vivi
2024-05-08 20:07 ` [PATCH 2/7] drm/xe: Fix xe_pm_runtime_get_if_in_use documentation Rodrigo Vivi
2024-05-08 20:07 ` Rodrigo Vivi [this message]
2024-05-08 20:07 ` [PATCH 4/7] drm/xe: Relax runtime pm protection around VM Rodrigo Vivi
2024-05-09 15:48 ` Matthew Brost
2024-05-13 13:16 ` Thomas Hellström
2024-05-08 20:07 ` [PATCH 5/7] drm/xe: Prepare display for D3Cold Rodrigo Vivi
2024-05-09 11:36 ` Gupta, Anshuman
2024-05-08 20:07 ` [PATCH 6/7] drm/xe: Stop checking for power_lost on D3Cold Rodrigo Vivi
2024-05-08 20:07 ` [PATCH 7/7] drm/xe: Enable D3Cold on 'low' VRAM utilization Rodrigo Vivi
2024-05-09 15:37 ` Matthew Brost
2024-05-08 20:13 ` ✓ CI.Patch_applied: success for series starting with [1/7] drm/xe: Fix xe_pm_runtime_get_if_active return Patchwork
2024-05-08 20:14 ` ✓ CI.checkpatch: " Patchwork
2024-05-08 20:15 ` ✓ CI.KUnit: " Patchwork
2024-05-08 20:26 ` ✓ CI.Build: " Patchwork
2024-05-08 20:29 ` ✓ CI.Hooks: " Patchwork
2024-05-08 20:30 ` ✓ CI.checksparse: " Patchwork
2024-05-08 21:03 ` ✓ CI.BAT: " Patchwork
2024-05-09 8:20 ` ✗ CI.FULL: failure " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2024-05-09 19:16 [PATCH 1/7] " Rodrigo Vivi
2024-05-09 19:16 ` [PATCH 3/7] drm/xe: Relax runtime pm protection during execution Rodrigo Vivi
2024-05-03 19:13 [PATCH 0/7] Unlock deeper package-C states (PC-10) and D3Cold Rodrigo Vivi
2024-05-03 19:13 ` [PATCH 3/7] drm/xe: Relax runtime pm protection during execution Rodrigo Vivi
2024-05-06 11:57 ` Thomas Hellström
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=20240508200707.375414-3-rodrigo.vivi@intel.com \
--to=rodrigo.vivi@intel.com \
--cc=francois.dugast@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.brost@intel.com \
--cc=thomas.hellstrom@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 read-only IMAP folder(s) and NNTP newsgroup(s).