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
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
|
From 2d2a5a052ea8e4ada3cfabafd23a9e1b896a23ee Mon Sep 17 00:00:00 2001
From: Le Ma <le.ma@amd.com>
Date: Tue, 26 Nov 2019 22:12:31 +0800
Subject: [PATCH 4662/4736] drm/amdgpu: add concurrent baco reset support for
XGMI
Currently each XGMI node reset wq does not run in parrallel if bound to same
cpu. Make change to bound the xgmi_reset_work item to different cpus.
XGMI requires all nodes enter into baco within very close proximity before
any node exit baco. So schedule the xgmi_reset_work wq twice for enter/exit
baco respectively.
To use baco for XGMI, PMFW supported for baco on XGMI needs to be involved.
The case that PSP reset and baco reset coexist within an XGMI hive never exist
and is not in the consideration.
v2: define use_baco flag to simplify the code for xgmi baco sequence
Change-Id: I9c08cf90134f940b42e20d2129ff87fba761c532
Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 +
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 82 ++++++++++++++++++----
2 files changed, 72 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 4eddee90553b..566ae8bf2ba7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1040,6 +1040,8 @@ struct amdgpu_device {
bool pm_sysfs_en;
bool ucode_sysfs_en;
+
+ bool in_baco;
};
static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 863590e169ac..2ca9d556c084 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2663,7 +2663,13 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
struct amdgpu_device *adev =
container_of(__work, struct amdgpu_device, xgmi_reset_work);
- adev->asic_reset_res = amdgpu_asic_reset(adev);
+ if (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO)
+ adev->asic_reset_res = (adev->in_baco == false) ?
+ amdgpu_device_baco_enter(adev->ddev) :
+ amdgpu_device_baco_exit(adev->ddev);
+ else
+ adev->asic_reset_res = amdgpu_asic_reset(adev);
+
if (adev->asic_reset_res)
DRM_WARN("ASIC reset failed with error, %d for drm dev, %s",
adev->asic_reset_res, adev->ddev->unique);
@@ -3795,13 +3801,18 @@ static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
return r;
}
-static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
+static int amdgpu_do_asic_reset(struct amdgpu_device *adev,
+ struct amdgpu_hive_info *hive,
struct list_head *device_list_handle,
bool *need_full_reset_arg)
{
struct amdgpu_device *tmp_adev = NULL;
bool need_full_reset = *need_full_reset_arg, vram_lost = false;
int r = 0;
+ int cpu = smp_processor_id();
+ bool use_baco =
+ (amdgpu_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) ?
+ true : false;
/*
* ASIC reset has to be done on all HGMI hive nodes ASAP
@@ -3809,21 +3820,24 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
*/
if (need_full_reset) {
list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
- /* For XGMI run all resets in parallel to speed up the process */
+ /*
+ * For XGMI run all resets in parallel to speed up the
+ * process by scheduling the highpri wq on different
+ * cpus. For XGMI with baco reset, all nodes must enter
+ * baco within close proximity before anyone exit.
+ */
if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
- if (!queue_work(system_highpri_wq, &tmp_adev->xgmi_reset_work))
+ if (!queue_work_on(cpu, system_highpri_wq,
+ &tmp_adev->xgmi_reset_work))
r = -EALREADY;
+ cpu = cpumask_next(cpu, cpu_online_mask);
} else
r = amdgpu_asic_reset(tmp_adev);
-
- if (r) {
- DRM_ERROR("ASIC reset failed with error, %d for drm dev, %s",
- r, tmp_adev->ddev->unique);
+ if (r)
break;
- }
}
- /* For XGMI wait for all PSP resets to complete before proceed */
+ /* For XGMI wait for all work to complete before proceed */
if (!r) {
list_for_each_entry(tmp_adev, device_list_handle,
gmc.xgmi.head) {
@@ -3832,11 +3846,54 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive,
r = tmp_adev->asic_reset_res;
if (r)
break;
+ if (use_baco)
+ tmp_adev->in_baco = true;
}
}
}
- }
+ /*
+ * For XGMI with baco reset, need exit baco phase by scheduling
+ * xgmi_reset_work one more time. PSP reset and sGPU skips this
+ * phase. Not assume the situation that PSP reset and baco reset
+ * coexist within an XGMI hive.
+ */
+
+ if (!r && use_baco) {
+ cpu = smp_processor_id();
+ list_for_each_entry(tmp_adev, device_list_handle,
+ gmc.xgmi.head) {
+ if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
+ if (!queue_work_on(cpu,
+ system_highpri_wq,
+ &tmp_adev->xgmi_reset_work))
+ r = -EALREADY;
+ if (r)
+ break;
+ cpu = cpumask_next(cpu, cpu_online_mask);
+ }
+ }
+ }
+
+ if (!r && use_baco) {
+ list_for_each_entry(tmp_adev, device_list_handle,
+ gmc.xgmi.head) {
+ if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) {
+ flush_work(&tmp_adev->xgmi_reset_work);
+ r = tmp_adev->asic_reset_res;
+ if (r)
+ break;
+ tmp_adev->in_baco = false;
+ }
+ }
+ }
+
+ if (r) {
+ DRM_ERROR("ASIC reset failed with error, %d for drm dev, %s",
+ r, tmp_adev->ddev->unique);
+ goto end;
+ }
+ }
list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) {
if (need_full_reset) {
@@ -4121,7 +4178,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
if (r)
adev->asic_reset_res = r;
} else {
- r = amdgpu_do_asic_reset(hive, device_list_handle, &need_full_reset);
+ r = amdgpu_do_asic_reset(adev, hive, device_list_handle,
+ &need_full_reset);
if (r && r == -EAGAIN)
goto retry;
}
--
2.17.1
|
