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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
|
From 0d3a43711fc55bba4355010ec4c165bafad46c69 Mon Sep 17 00:00:00 2001
From: Kenneth Feng <kenneth.feng@amd.com>
Date: Fri, 11 Oct 2019 17:51:34 +0800
Subject: [PATCH 4133/4736] drm/amd/powerplay: bug fix for pcie parameters
override
Bug fix for pcie paramerers override on swsmu.
Below is a scenario to have this problem.
pptable definition on pcie dpm:
0 -> pcie gen speed:1, pcie lanes: *16
1 -> pcie gen speed:4, pcie lanes: *16
Then if we have a system only have the capbility:
pcie gen speed: 3, pcie lanes: *8,
we will override dpm 1 to pcie gen speed 3, pcie lanes *8.
But the code skips the dpm 0 configuration.
So the real pcie dpm parameters are:
0 -> pcie gen speed:1, pcie lanes: *16
1 -> pcie gen speed:3, pcie lanes: *8
Then the wrong pcie lanes will be toggled.
Signed-off-by: Kenneth Feng <kenneth.feng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Evan Quan <evan.quan@amd.com>
---
drivers/gpu/drm/amd/powerplay/amdgpu_smu.c | 44 -------------------
.../gpu/drm/amd/powerplay/inc/amdgpu_smu.h | 8 ++++
drivers/gpu/drm/amd/powerplay/navi10_ppt.c | 23 ++++++++++
drivers/gpu/drm/amd/powerplay/smu_v11_0.c | 44 +++++++++++++++++++
drivers/gpu/drm/amd/powerplay/vega20_ppt.c | 23 ++++++++++
5 files changed, 98 insertions(+), 44 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index a37a1b1d8abd..26cacc899dfe 100644
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -945,50 +945,6 @@ static int smu_fini_fb_allocations(struct smu_context *smu)
return 0;
}
-static int smu_override_pcie_parameters(struct smu_context *smu)
-{
- struct amdgpu_device *adev = smu->adev;
- uint32_t pcie_gen = 0, pcie_width = 0, smu_pcie_arg;
- int ret;
-
- if (adev->flags & AMD_IS_APU)
- return 0;
-
- if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
- pcie_gen = 3;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
- pcie_gen = 2;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
- pcie_gen = 1;
- else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
- pcie_gen = 0;
-
- /* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
- * Bit 15:8: PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
- * Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32
- */
- if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
- pcie_width = 6;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
- pcie_width = 5;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
- pcie_width = 4;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
- pcie_width = 3;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
- pcie_width = 2;
- else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
- pcie_width = 1;
-
- smu_pcie_arg = (1 << 16) | (pcie_gen << 8) | pcie_width;
- ret = smu_send_smc_msg_with_param(smu,
- SMU_MSG_OverridePcieParameters,
- smu_pcie_arg);
- if (ret)
- pr_err("[%s] Attempt to override pcie params failed!\n", __func__);
- return ret;
-}
-
static int smu_smc_table_hw_init(struct smu_context *smu,
bool initialize)
{
diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
index 401affdee49d..cdb845f5f23e 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
@@ -469,6 +469,7 @@ struct pptable_funcs {
int (*get_dpm_clk_limited)(struct smu_context *smu, enum smu_clk_type clk_type,
uint32_t dpm_level, uint32_t *freq);
int (*set_df_cstate)(struct smu_context *smu, enum pp_df_cstate state);
+ int (*update_pcie_parameters)(struct smu_context *smu, uint32_t pcie_gen_cap, uint32_t pcie_width_cap);
};
struct smu_funcs
@@ -551,6 +552,7 @@ struct smu_funcs
int (*mode2_reset)(struct smu_context *smu);
int (*get_dpm_ultimate_freq)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max);
int (*set_soft_freq_limited_range)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max);
+ int (*override_pcie_parameters)(struct smu_context *smu);
};
#define smu_init_microcode(smu) \
@@ -783,6 +785,12 @@ struct smu_funcs
#define smu_set_soft_freq_limited_range(smu, clk_type, min, max) \
((smu)->funcs->set_soft_freq_limited_range ? (smu)->funcs->set_soft_freq_limited_range((smu), (clk_type), (min), (max)) : -EINVAL)
+#define smu_override_pcie_parameters(smu) \
+ ((smu)->funcs->override_pcie_parameters ? (smu)->funcs->override_pcie_parameters((smu)) : 0)
+
+#define smu_update_pcie_parameters(smu, pcie_gen_cap, pcie_width_cap) \
+ ((smu)->ppt_funcs->update_pcie_parameters ? (smu)->ppt_funcs->update_pcie_parameters((smu), (pcie_gen_cap), (pcie_width_cap)) : 0)
+
extern int smu_get_atom_data_table(struct smu_context *smu, uint32_t table,
uint16_t *size, uint8_t *frev, uint8_t *crev,
uint8_t **addr);
diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
index e8e5c889cc95..b88aae9bb242 100644
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
@@ -1628,6 +1628,28 @@ static int navi10_get_power_limit(struct smu_context *smu,
return 0;
}
+static int navi10_update_pcie_parameters(struct smu_context *smu,
+ uint32_t pcie_gen_cap,
+ uint32_t pcie_width_cap)
+{
+ PPTable_t *pptable = smu->smu_table.driver_pptable;
+ int ret, i;
+ uint32_t smu_pcie_arg;
+
+ for (i = 0; i < NUM_LINK_LEVELS; i++) {
+ smu_pcie_arg = (i << 16) |
+ ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) :
+ (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ?
+ pptable->PcieLaneCount[i] : pcie_width_cap);
+ ret = smu_send_smc_msg_with_param(smu,
+ SMU_MSG_OverridePcieParameters,
+ smu_pcie_arg);
+ }
+
+ return ret;
+}
+
+
static const struct pptable_funcs navi10_ppt_funcs = {
.tables_init = navi10_tables_init,
.alloc_dpm_context = navi10_allocate_dpm_context,
@@ -1666,6 +1688,7 @@ static const struct pptable_funcs navi10_ppt_funcs = {
.get_thermal_temperature_range = navi10_get_thermal_temperature_range,
.display_disable_memory_clock_switch = navi10_display_disable_memory_clock_switch,
.get_power_limit = navi10_get_power_limit,
+ .update_pcie_parameters = navi10_update_pcie_parameters,
};
void navi10_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
index 9883f0a4471a..df1f2b99fed7 100644
--- a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
+++ b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -32,6 +32,7 @@
#include "vega20_ppt.h"
#include "arcturus_ppt.h"
#include "navi10_ppt.h"
+#include "amd_pcie.h"
#include "asic_reg/thm/thm_11_0_2_offset.h"
#include "asic_reg/thm/thm_11_0_2_sh_mask.h"
@@ -1791,6 +1792,48 @@ static int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum s
return ret;
}
+static int smu_v11_0_override_pcie_parameters(struct smu_context *smu)
+{
+ struct amdgpu_device *adev = smu->adev;
+ uint32_t pcie_gen = 0, pcie_width = 0;
+ int ret;
+
+ if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4)
+ pcie_gen = 3;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)
+ pcie_gen = 2;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2)
+ pcie_gen = 1;
+ else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1)
+ pcie_gen = 0;
+
+ /* Bit 31:16: LCLK DPM level. 0 is DPM0, and 1 is DPM1
+ * Bit 15:8: PCIE GEN, 0 to 3 corresponds to GEN1 to GEN4
+ * Bit 7:0: PCIE lane width, 1 to 7 corresponds is x1 to x32
+ */
+ if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X16)
+ pcie_width = 6;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X12)
+ pcie_width = 5;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X8)
+ pcie_width = 4;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X4)
+ pcie_width = 3;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X2)
+ pcie_width = 2;
+ else if (adev->pm.pcie_mlw_mask & CAIL_PCIE_LINK_WIDTH_SUPPORT_X1)
+ pcie_width = 1;
+
+ ret = smu_update_pcie_parameters(smu, pcie_gen, pcie_width);
+
+ if (ret)
+ pr_err("[%s] Attempt to override pcie params failed!\n", __func__);
+
+ return ret;
+
+}
+
+
static const struct smu_funcs smu_v11_0_funcs = {
.init_microcode = smu_v11_0_init_microcode,
.load_microcode = smu_v11_0_load_microcode,
@@ -1843,6 +1886,7 @@ static const struct smu_funcs smu_v11_0_funcs = {
.baco_reset = smu_v11_0_baco_reset,
.get_dpm_ultimate_freq = smu_v11_0_get_dpm_ultimate_freq,
.set_soft_freq_limited_range = smu_v11_0_set_soft_freq_limited_range,
+ .override_pcie_parameters = smu_v11_0_override_pcie_parameters,
};
void smu_v11_0_set_smu_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
index 1050566cb69a..a76ffd58404e 100644
--- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
@@ -3157,6 +3157,28 @@ static int vega20_set_df_cstate(struct smu_context *smu,
return smu_send_smc_msg_with_param(smu, SMU_MSG_DFCstateControl, state);
}
+static int vega20_update_pcie_parameters(struct smu_context *smu,
+ uint32_t pcie_gen_cap,
+ uint32_t pcie_width_cap)
+{
+ PPTable_t *pptable = smu->smu_table.driver_pptable;
+ int ret, i;
+ uint32_t smu_pcie_arg;
+
+ for (i = 0; i < NUM_LINK_LEVELS; i++) {
+ smu_pcie_arg = (i << 16) |
+ ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) :
+ (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ?
+ pptable->PcieLaneCount[i] : pcie_width_cap);
+ ret = smu_send_smc_msg_with_param(smu,
+ SMU_MSG_OverridePcieParameters,
+ smu_pcie_arg);
+ }
+
+ return ret;
+}
+
+
static const struct pptable_funcs vega20_ppt_funcs = {
.tables_init = vega20_tables_init,
.alloc_dpm_context = vega20_allocate_dpm_context,
@@ -3201,6 +3223,7 @@ static const struct pptable_funcs vega20_ppt_funcs = {
.set_watermarks_table = vega20_set_watermarks_table,
.get_thermal_temperature_range = vega20_get_thermal_temperature_range,
.set_df_cstate = vega20_set_df_cstate,
+ .update_pcie_parameters = vega20_update_pcie_parameters
};
void vega20_set_ppt_funcs(struct smu_context *smu)
--
2.17.1
|
