drm/amdgpu: update ras capability's query based on mem ecc configuration
RAS support capability needs to be updated on top of different
memeory ECC enablement, and remove redundant memory ecc check
in gmc module for vega20 and arcturus.
v2: check HBM ECC enablement and set ras mask accordingly.
v3: avoid to invoke atomfirmware interface to query twice.
Suggested-by: Hawking Zhang <[email protected]>
Signed-off-by: Guchun Chen <[email protected]>
Reviewed-by: Hawking Zhang <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index ce8548d..c3dca4ca 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1765,18 +1765,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
*hw_supported = 0;
*supported = 0;
- if (amdgpu_sriov_vf(adev) ||
+ if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw ||
(adev->asic_type != CHIP_VEGA20 &&
adev->asic_type != CHIP_ARCTURUS))
return;
- if (adev->is_atom_fw &&
- (amdgpu_atomfirmware_mem_ecc_supported(adev) ||
- amdgpu_atomfirmware_sram_ecc_supported(adev)))
- *hw_supported = AMDGPU_RAS_BLOCK_MASK;
+ if (amdgpu_atomfirmware_mem_ecc_supported(adev)) {
+ DRM_INFO("HBM ECC is active.\n");
+ *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC |
+ 1 << AMDGPU_RAS_BLOCK__DF);
+ } else
+ DRM_INFO("HBM ECC is not presented.\n");
+
+ if (amdgpu_atomfirmware_sram_ecc_supported(adev)) {
+ DRM_INFO("SRAM ECC is active.\n");
+ *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC |
+ 1 << AMDGPU_RAS_BLOCK__DF);
+ } else
+ DRM_INFO("SRAM ECC is not presented.\n");
+
+ /* hw_supported needs to be aligned with RAS block mask. */
+ *hw_supported &= AMDGPU_RAS_BLOCK_MASK;
*supported = amdgpu_ras_enable == 0 ?
- 0 : *hw_supported & amdgpu_ras_mask;
+ 0 : *hw_supported & amdgpu_ras_mask;
}
int amdgpu_ras_init(struct amdgpu_device *adev)
