1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
|
From 083acdc7a3158786d53e46a0e43ff7cf55135f88 Mon Sep 17 00:00:00 2001
From: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Date: Wed, 30 May 2018 15:28:52 -0400
Subject: [PATCH 4593/5725] drm/amdgpu: move amdgpu_ctx_mgr_entity_fini to
f_ops flush hook (V4)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
With this we can now terminate jobs enqueue into SW queue the moment
the task is being killed instead of waiting for last user of
drm file to release it.
Also stop checking for kref_read(&ctx->refcount) == 1 when
calling drm_sched_entity_do_release since other task
might still hold a reference to this entity but we don't
care since KILL means terminate job submission regardless
of what other tasks are doing.
v2:
Use returned remaining timeout as parameter for the next call.
Rebase.
v3:
Switch to working with jiffies.
Streamline remainder TO usage.
Rebase.
v4:
Rebase.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Kalyan Alle <kalyan.alle@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 14 ++++++++------
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 12 ++++++++++++
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 1 -
include/drm/gpu_scheduler.h | 6 ++++--
4 files changed, 24 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 58795d4..6f6fdab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -465,26 +465,28 @@ void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr)
struct amdgpu_ctx *ctx;
struct idr *idp;
uint32_t id, i;
+ long max_wait = MAX_WAIT_SCHED_ENTITY_Q_EMPTY;
idp = &mgr->ctx_handles;
+ mutex_lock(&mgr->lock);
idr_for_each_entry(idp, ctx, id) {
- if (!ctx->adev)
+ if (!ctx->adev) {
+ mutex_unlock(&mgr->lock);
return;
+ }
for (i = 0; i < ctx->adev->num_rings; i++) {
if (ctx->adev->rings[i] == &ctx->adev->gfx.kiq.ring)
continue;
- if (kref_read(&ctx->refcount) == 1)
- drm_sched_entity_do_release(&ctx->adev->rings[i]->sched,
- &ctx->rings[i].entity);
- else
- DRM_ERROR("ctx %p is still alive\n", ctx);
+ max_wait = drm_sched_entity_do_release(&ctx->adev->rings[i]->sched,
+ &ctx->rings[i].entity, max_wait);
}
}
+ mutex_unlock(&mgr->lock);
}
void amdgpu_ctx_mgr_entity_cleanup(struct amdgpu_ctx_mgr *mgr)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 1c0cf9a..96694c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -861,9 +861,21 @@ static const struct dev_pm_ops amdgpu_pm_ops = {
.runtime_idle = amdgpu_pmops_runtime_idle,
};
+static int amdgpu_flush(struct file *f, fl_owner_t id)
+{
+ struct drm_file *file_priv = f->private_data;
+ struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
+
+ amdgpu_ctx_mgr_entity_fini(&fpriv->ctx_mgr);
+
+ return 0;
+}
+
+
static const struct file_operations amdgpu_driver_kms_fops = {
.owner = THIS_MODULE,
.open = drm_open,
+ .flush = amdgpu_flush,
.release = drm_release,
.unlocked_ioctl = amdgpu_drm_ioctl,
.mmap = amdgpu_mmap,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 86087c1..a39919c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -964,7 +964,6 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
return;
pm_runtime_get_sync(dev->dev);
- amdgpu_ctx_mgr_entity_fini(&fpriv->ctx_mgr);
if (adev->asic_type != CHIP_RAVEN) {
amdgpu_uvd_free_handles(adev, file_priv);
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index 86c7344..7ae23fb 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -27,6 +27,8 @@
#include <drm/spsc_queue.h>
#include <linux/dma-fence.h>
+#define MAX_WAIT_SCHED_ENTITY_Q_EMPTY msecs_to_jiffies(1000)
+
struct drm_gpu_scheduler;
struct drm_sched_rq;
@@ -151,8 +153,8 @@ int drm_sched_entity_init(struct drm_gpu_scheduler *sched,
struct drm_sched_entity *entity,
struct drm_sched_rq *rq,
atomic_t *guilty);
-void drm_sched_entity_do_release(struct drm_gpu_scheduler *sched,
- struct drm_sched_entity *entity);
+long drm_sched_entity_do_release(struct drm_gpu_scheduler *sched,
+ struct drm_sched_entity *entity, long timeout);
void drm_sched_entity_cleanup(struct drm_gpu_scheduler *sched,
struct drm_sched_entity *entity);
void drm_sched_entity_fini(struct drm_gpu_scheduler *sched,
--
2.7.4
|
