diff options
| author | Hawking Zhang <Hawking.Zhang@amd.com> | 2021-04-02 14:39:36 +0800 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2021-04-09 16:51:04 -0400 |
| commit | 6e36f23193cc870856a41e87281f62fb2b04bd1f (patch) | |
| tree | cf35baff233629f14cbce1e3b48f9fe316ffdfbd /drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | |
| parent | 87da0cc101e723833446e52971fac5fa7358dec5 (diff) | |
| download | linux-stable-6e36f23193cc870856a41e87281f62fb2b04bd1f.tar.gz linux-stable-6e36f23193cc870856a41e87281f62fb2b04bd1f.tar.bz2 linux-stable-6e36f23193cc870856a41e87281f62fb2b04bd1f.zip | |
drm/amdgpu: split nbio callbacks into ras and non-ras ones
nbio ras is not managed by gpu driver when gpu is
connected to cpu through xgmi. split nbio callbacks
into ras and non-ras ones so gpu driver only
initializes nbio ras callbacks when it manages
nbio ras.
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: John Clements <John.Clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 30 |
1 files changed, 24 insertions, 6 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 1708045e2a0d..ac3f4c3266bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -804,8 +804,9 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev, adev->mmhub.funcs->query_ras_error_status(adev); break; case AMDGPU_RAS_BLOCK__PCIE_BIF: - if (adev->nbio.funcs->query_ras_error_count) - adev->nbio.funcs->query_ras_error_count(adev, &err_data); + if (adev->nbio.ras_funcs && + adev->nbio.ras_funcs->query_ras_error_count) + adev->nbio.ras_funcs->query_ras_error_count(adev, &err_data); break; case AMDGPU_RAS_BLOCK__XGMI_WAFL: amdgpu_xgmi_query_ras_error_count(adev, &err_data); @@ -2030,14 +2031,31 @@ int amdgpu_ras_init(struct amdgpu_device *adev) /* Might need get this flag from vbios. */ con->flags = RAS_DEFAULT_FLAGS; - if (adev->nbio.funcs->init_ras_controller_interrupt) { - r = adev->nbio.funcs->init_ras_controller_interrupt(adev); + /* initialize nbio ras function ahead of any other + * ras functions so hardware fatal error interrupt + * can be enabled as early as possible */ + switch (adev->asic_type) { + case CHIP_VEGA20: + case CHIP_ARCTURUS: + case CHIP_ALDEBARAN: + if (!adev->gmc.xgmi.connected_to_cpu) + adev->nbio.ras_funcs = &nbio_v7_4_ras_funcs; + break; + default: + /* nbio ras is not available */ + break; + } + + if (adev->nbio.ras_funcs && + adev->nbio.ras_funcs->init_ras_controller_interrupt) { + r = adev->nbio.ras_funcs->init_ras_controller_interrupt(adev); if (r) goto release_con; } - if (adev->nbio.funcs->init_ras_err_event_athub_interrupt) { - r = adev->nbio.funcs->init_ras_err_event_athub_interrupt(adev); + if (adev->nbio.ras_funcs && + adev->nbio.ras_funcs->init_ras_err_event_athub_interrupt) { + r = adev->nbio.ras_funcs->init_ras_err_event_athub_interrupt(adev); if (r) goto release_con; } |