drm/amdgpu: add umc retire unit element

It records how many bad pages are retired in one uncorrectable error.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
index a6951160f13af64c37c9033cc6513034cb1cf35d..f2bf979af5883596169c12aba6db42ceaaa66f30 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h
@@ -74,6 +74,8 @@ struct amdgpu_umc {
 
        /* UMC regiser per channel offset */
        uint32_t channel_offs;
+       /* how many pages are retired in one UE */
+       uint32_t retire_unit;
        /* channel index table of interleaved memory */
        const uint32_t *channel_idx_tbl;
        struct ras_common_if *ras_if;

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 7db1f1a7e33c305bf414af1bbb51f5c4e3155d60..ab2556ca984e1551be6f078d9056866e9f7e8a16 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -692,6 +692,7 @@ static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev)
                adev->umc.channel_inst_num = UMC_V8_7_CHANNEL_INSTANCE_NUM;
                adev->umc.umc_inst_num = UMC_V8_7_UMC_INSTANCE_NUM;
                adev->umc.channel_offs = UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA;
+               adev->umc.retire_unit = 1;
                adev->umc.channel_idx_tbl = &umc_v8_7_channel_idx_tbl[0][0];
                adev->umc.ras = &umc_v8_7_ras;
                break;

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 0a31a341aa43b586c97e4ea9ea39522b6164e696..85e0afc3d4f7f336715629779c44170eb4ab7c8f 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -570,6 +570,7 @@ static void gmc_v11_0_set_umc_funcs(struct amdgpu_device *adev)
                adev->umc.node_inst_num = adev->gmc.num_umc;
                adev->umc.max_ras_err_cnt_per_query = UMC_V8_10_TOTAL_CHANNEL_NUM(adev);
                adev->umc.channel_offs = UMC_V8_10_PER_CHANNEL_OFFSET;
+               adev->umc.retire_unit = UMC_V8_10_NA_COL_2BITS_POWER_OF_2_NUM;
                if (adev->umc.node_inst_num == 4)
                        adev->umc.channel_idx_tbl = &umc_v8_10_channel_idx_tbl_ext0[0][0][0];
                else

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index d65c6cea344515453a95c4fe35a21d71482d55af..b06170c00dfca9e174306f363a44e4b30b584e07 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1288,6 +1288,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
                adev->umc.channel_inst_num = UMC_V6_1_CHANNEL_INSTANCE_NUM;
                adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM;
                adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_VG20;
+               adev->umc.retire_unit = 1;
                adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0];
                adev->umc.ras = &umc_v6_1_ras;
                break;
@@ -1296,6 +1297,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
                adev->umc.channel_inst_num = UMC_V6_1_CHANNEL_INSTANCE_NUM;
                adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM;
                adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_ARCT;
+               adev->umc.retire_unit = 1;
                adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0];
                adev->umc.ras = &umc_v6_1_ras;
                break;
@@ -1305,6 +1307,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
                adev->umc.channel_inst_num = UMC_V6_7_CHANNEL_INSTANCE_NUM;
                adev->umc.umc_inst_num = UMC_V6_7_UMC_INSTANCE_NUM;
                adev->umc.channel_offs = UMC_V6_7_PER_CHANNEL_OFFSET;
+               adev->umc.retire_unit = (UMC_V6_7_NA_MAP_PA_NUM * 2);
                if (!adev->gmc.xgmi.connected_to_cpu)
                        adev->umc.ras = &umc_v6_7_ras;
                if (1 & adev->smuio.funcs->get_die_id(adev))