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
|
From a01add62c600eb697dc1cb18d2dc07fd5969d98f Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Wed, 31 Jul 2019 20:23:01 +0800
Subject: [PATCH 3200/4256] drm/amdgpu: add umc v6_1 query error count support
Implement umc query_ras_error_count function to support querry
both correctable and uncorrectable error
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Dennis Li <dennis.li@amd.com>
---
drivers/gpu/drm/amd/amdgpu/Makefile | 4 +
drivers/gpu/drm/amd/amdgpu/umc_v6_1.c | 162 ++++++++++++++++++++++++++
drivers/gpu/drm/amd/amdgpu/umc_v6_1.h | 39 +++++++
3 files changed, 205 insertions(+)
create mode 100644 drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
create mode 100644 drivers/gpu/drm/amd/amdgpu/umc_v6_1.h
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index bb2e00bbdee1..797f8a7e4f72 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -81,6 +81,10 @@ amdgpu-y += \
gfxhub_v1_0.o mmhub_v1_0.o gmc_v9_0.o gfxhub_v1_1.o mmhub_v9_4.o \
gfxhub_v2_0.o mmhub_v2_0.o gmc_v10_0.o
+# add UMC block
+amdgpu-y += \
+ umc_v6_1.o
+
# add IH block
amdgpu-y += \
amdgpu_irq.o \
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
new file mode 100644
index 000000000000..1ca5ae642946
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c
@@ -0,0 +1,162 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "umc_v6_1.h"
+#include "amdgpu_ras.h"
+#include "amdgpu.h"
+
+#include "rsmu/rsmu_0_0_2_offset.h"
+#include "rsmu/rsmu_0_0_2_sh_mask.h"
+#include "umc/umc_6_1_1_offset.h"
+#include "umc/umc_6_1_1_sh_mask.h"
+
+static void umc_v6_1_enable_umc_index_mode(struct amdgpu_device *adev,
+ uint32_t umc_instance)
+{
+ uint32_t rsmu_umc_index;
+
+ rsmu_umc_index = RREG32_SOC15(RSMU, 0,
+ mmRSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU);
+ rsmu_umc_index = REG_SET_FIELD(rsmu_umc_index,
+ RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU,
+ RSMU_UMC_INDEX_MODE_EN, 1);
+ rsmu_umc_index = REG_SET_FIELD(rsmu_umc_index,
+ RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU,
+ RSMU_UMC_INDEX_INSTANCE, umc_instance);
+ rsmu_umc_index = REG_SET_FIELD(rsmu_umc_index,
+ RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU,
+ RSMU_UMC_INDEX_WREN, 1 << umc_instance);
+ WREG32_SOC15(RSMU, 0, mmRSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU,
+ rsmu_umc_index);
+}
+
+static void umc_v6_1_disable_umc_index_mode(struct amdgpu_device *adev)
+{
+ WREG32_FIELD15(RSMU, 0, RSMU_UMC_INDEX_REGISTER_NBIF_VG20_GPU,
+ RSMU_UMC_INDEX_MODE_EN, 0);
+}
+
+static void umc_v6_1_query_correctable_error_count(struct amdgpu_device *adev,
+ uint32_t umc_reg_offset,
+ unsigned long *error_count)
+{
+ uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr;
+ uint32_t ecc_err_cnt, ecc_err_cnt_addr;
+ uint64_t mc_umc_status;
+ uint32_t mc_umc_status_addr;
+
+ ecc_err_cnt_sel_addr =
+ SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_EccErrCntSel);
+ ecc_err_cnt_addr =
+ SOC15_REG_OFFSET(UMC, 0, mmUMCCH0_0_EccErrCnt);
+ mc_umc_status_addr =
+ SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_STATUST0);
+
+ /* select the lower chip and check the error count */
+ ecc_err_cnt_sel = RREG32(ecc_err_cnt_sel_addr + umc_reg_offset);
+ ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_EccErrCntSel,
+ EccErrCntCsSel, 0);
+ WREG32(ecc_err_cnt_sel_addr + umc_reg_offset, ecc_err_cnt_sel);
+ ecc_err_cnt = RREG32(ecc_err_cnt_addr + umc_reg_offset);
+ *error_count +=
+ REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_EccErrCnt, EccErrCnt);
+ /* clear the lower chip err count */
+ WREG32(ecc_err_cnt_addr + umc_reg_offset, 0);
+
+ /* select the higher chip and check the err counter */
+ ecc_err_cnt_sel = REG_SET_FIELD(ecc_err_cnt_sel, UMCCH0_0_EccErrCntSel,
+ EccErrCntCsSel, 1);
+ WREG32(ecc_err_cnt_sel_addr + umc_reg_offset, ecc_err_cnt_sel);
+ ecc_err_cnt = RREG32(ecc_err_cnt_addr + umc_reg_offset);
+ *error_count +=
+ REG_GET_FIELD(ecc_err_cnt, UMCCH0_0_EccErrCnt, EccErrCnt);
+ /* clear the higher chip err count */
+ WREG32(ecc_err_cnt_addr + umc_reg_offset, 0);
+
+ /* check for SRAM correctable error
+ MCUMC_STATUS is a 64 bit register */
+ mc_umc_status =
+ RREG32(mc_umc_status_addr + umc_reg_offset);
+ mc_umc_status |=
+ (uint64_t)RREG32(mc_umc_status_addr + umc_reg_offset + 1) << 32;
+ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, ErrorCodeExt) == 6 &&
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)
+ *error_count += 1;
+
+ /* clear the MCUMC_STATUS */
+ WREG32(mc_umc_status_addr + umc_reg_offset, 0);
+ WREG32(mc_umc_status_addr + umc_reg_offset + 1, 0);
+}
+
+static void umc_v6_1_querry_uncorrectable_error_count(struct amdgpu_device *adev,
+ uint32_t umc_reg_offset,
+ unsigned long *error_count)
+{
+ uint64_t mc_umc_status;
+ uint32_t mc_umc_status_addr;
+
+ mc_umc_status_addr =
+ SOC15_REG_OFFSET(UMC, 0, mmMCA_UMC_UMC0_MCUMC_STATUST0);
+
+ /* check the MCUMC_STATUS */
+ mc_umc_status = RREG32(mc_umc_status_addr + umc_reg_offset);
+ mc_umc_status |=
+ (uint64_t)RREG32(mc_umc_status_addr + umc_reg_offset + 1) << 32;
+
+ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, ErrorCodeExt) == 6 &&
+ (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 ||
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 ||
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 ||
+ REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1))
+ *error_count += 1;
+
+ /* clear the MCUMC_STATUS */
+ WREG32(mc_umc_status_addr + umc_reg_offset, 0);
+ WREG32(mc_umc_status_addr + umc_reg_offset + 1, 0);
+}
+
+static void umc_v6_1_query_ras_error_count(struct amdgpu_device *adev,
+ void *ras_error_status)
+{
+ struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
+ uint32_t umc_inst, channel_inst, umc_reg_offset;
+
+ for (umc_inst = 0; umc_inst < UMC_V6_1_UMC_INSTANCE_NUM; umc_inst++) {
+ /* enable the index mode to query eror count per channel */
+ umc_v6_1_enable_umc_index_mode(adev, umc_inst);
+ for (channel_inst = 0; channel_inst < UMC_V6_1_CHANNEL_INSTANCE_NUM; channel_inst++) {
+ /* calc the register offset according to channel instance */
+ umc_reg_offset = UMC_V6_1_PER_CHANNEL_OFFSET * channel_inst;
+ umc_v6_1_query_correctable_error_count(adev, umc_reg_offset,
+ &(err_data->ce_count));
+ umc_v6_1_querry_uncorrectable_error_count(adev, umc_reg_offset,
+ &(err_data->ue_count));
+ }
+ }
+ umc_v6_1_disable_umc_index_mode(adev);
+}
+
+const struct amdgpu_umc_funcs umc_v6_1_funcs = {
+ .query_ras_error_count = umc_v6_1_query_ras_error_count,
+};
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h
new file mode 100644
index 000000000000..d25ae414f4d8
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2019 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#ifndef __UMC_V6_1_H__
+#define __UMC_V6_1_H__
+
+#include "soc15_common.h"
+
+/* HBM Memory Channel Width */
+#define UMC_V6_1_HBM_MEMORY_CHANNEL_WIDTH 128
+/* number of umc channel instance with memory map register access */
+#define UMC_V6_1_CHANNEL_INSTANCE_NUM 4
+/* number of umc instance with memory map register access */
+#define UMC_V6_1_UMC_INSTANCE_NUM 8
+/* UMC regiser per channel offset */
+#define UMC_V6_1_PER_CHANNEL_OFFSET 0x800
+
+extern const struct amdgpu_umc_funcs umc_v6_1_funcs;
+
+#endif
--
2.17.1
|
