1 /*
2  * Copyright 2021 Advanced Micro Devices, Inc.
3  *
4  * Permission is hereby granted, free of charge, to any person obtaining a
5  * copy of this software and associated documentation files (the "Software"),
6  * to deal in the Software without restriction, including without limitation
7  * the rights to use, copy, modify, merge, publish, distribute, sublicense,
8  * and/or sell copies of the Software, and to permit persons to whom the
9  * Software is furnished to do so, subject to the following conditions:
10  *
11  * The above copyright notice and this permission notice shall be included in
12  * all copies or substantial portions of the Software.
13  *
14  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
15  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
16  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
17  * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
18  * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
19  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
20  * OTHER DEALINGS IN THE SOFTWARE.
21  *
22  */
23 #include "amdgpu_ras.h"
24 #include "amdgpu.h"
25 #include "amdgpu_mca.h"
26 
27 #include "umc/umc_6_7_0_offset.h"
28 #include "umc/umc_6_7_0_sh_mask.h"
29 
amdgpu_mca_query_correctable_error_count(struct amdgpu_device * adev,uint64_t mc_status_addr,unsigned long * error_count)30 void amdgpu_mca_query_correctable_error_count(struct amdgpu_device *adev,
31 					      uint64_t mc_status_addr,
32 					      unsigned long *error_count)
33 {
34 	uint64_t mc_status = RREG64_PCIE(mc_status_addr);
35 
36 	if (REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
37 	    REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)
38 		*error_count += 1;
39 }
40 
amdgpu_mca_query_uncorrectable_error_count(struct amdgpu_device * adev,uint64_t mc_status_addr,unsigned long * error_count)41 void amdgpu_mca_query_uncorrectable_error_count(struct amdgpu_device *adev,
42 						uint64_t mc_status_addr,
43 						unsigned long *error_count)
44 {
45 	uint64_t mc_status = RREG64_PCIE(mc_status_addr);
46 
47 	if ((REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) &&
48 	    (REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1 ||
49 	    REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 ||
50 	    REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 ||
51 	    REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 ||
52 	    REG_GET_FIELD(mc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1))
53 		*error_count += 1;
54 }
55 
amdgpu_mca_reset_error_count(struct amdgpu_device * adev,uint64_t mc_status_addr)56 void amdgpu_mca_reset_error_count(struct amdgpu_device *adev,
57 				  uint64_t mc_status_addr)
58 {
59 	WREG64_PCIE(mc_status_addr, 0x0ULL);
60 }
61 
amdgpu_mca_query_ras_error_count(struct amdgpu_device * adev,uint64_t mc_status_addr,void * ras_error_status)62 void amdgpu_mca_query_ras_error_count(struct amdgpu_device *adev,
63 				      uint64_t mc_status_addr,
64 				      void *ras_error_status)
65 {
66 	struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status;
67 
68 	amdgpu_mca_query_correctable_error_count(adev, mc_status_addr, &(err_data->ce_count));
69 	amdgpu_mca_query_uncorrectable_error_count(adev, mc_status_addr, &(err_data->ue_count));
70 
71 	amdgpu_mca_reset_error_count(adev, mc_status_addr);
72 }
73 
amdgpu_mca_ras_late_init(struct amdgpu_device * adev,struct amdgpu_mca_ras * mca_dev)74 int amdgpu_mca_ras_late_init(struct amdgpu_device *adev,
75 			     struct amdgpu_mca_ras *mca_dev)
76 {
77 	int r;
78 	struct ras_ih_if ih_info = {
79 		.cb = NULL,
80 	};
81 	struct ras_fs_if fs_info = {
82 		.sysfs_name = mca_dev->ras_funcs->sysfs_name,
83 	};
84 
85 	if (!mca_dev->ras_if) {
86 		mca_dev->ras_if = kmalloc(sizeof(struct ras_common_if), GFP_KERNEL);
87 		if (!mca_dev->ras_if)
88 			return -ENOMEM;
89 		mca_dev->ras_if->block = mca_dev->ras_funcs->ras_block;
90 		mca_dev->ras_if->sub_block_index = mca_dev->ras_funcs->ras_sub_block;
91 		mca_dev->ras_if->type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
92 	}
93 	ih_info.head = fs_info.head = *mca_dev->ras_if;
94 	r = amdgpu_ras_late_init(adev, mca_dev->ras_if,
95 				 &fs_info, &ih_info);
96 	if (r || !amdgpu_ras_is_supported(adev, mca_dev->ras_if->block)) {
97 		kfree(mca_dev->ras_if);
98 		mca_dev->ras_if = NULL;
99 	}
100 
101 	return r;
102 }
103 
amdgpu_mca_ras_fini(struct amdgpu_device * adev,struct amdgpu_mca_ras * mca_dev)104 void amdgpu_mca_ras_fini(struct amdgpu_device *adev,
105 			 struct amdgpu_mca_ras *mca_dev)
106 {
107 	struct ras_ih_if ih_info = {
108 		.cb = NULL,
109 	};
110 
111 	if (!mca_dev->ras_if)
112 		return;
113 
114 	amdgpu_ras_late_fini(adev, mca_dev->ras_if, &ih_info);
115 	kfree(mca_dev->ras_if);
116 	mca_dev->ras_if = NULL;
117 }