[Intel-gfx] [PATCH 1/1] drm/i915: Fixing error code for WOPCM initialization

Sujaritha sujaritha.sundaresan at intel.com
Wed Mar 6 00:11:38 UTC 2019


On 3/5/19 2:06 PM, Daniele Ceraolo Spurio wrote:
>
>
> On 3/5/19 2:05 PM, Daniele Ceraolo Spurio wrote:
>>
>>
>> On 3/4/19 4:55 PM, Sujaritha Sundaresan wrote:
>>> Replacing the -E2BIG error code return for WOPCM
>>> initialization with -ENODEV. This will prevent the pci from
>>> picking this up as a warning during fault injection testing.
>>
>> To clarify, we want to silence this:
>>
>
> forgot to paste the log...
>
> <4> [381.569479] i915: probe of 0000:00:02.0 failed with error -7
>
> Daniele
>
>>
>>
>>>
>>> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio at intel.com>
>>> Cc: Michal Wajdeczko <michal.wajdeczko at intel.com>
>>> Signed-off-by: Sujaritha Sundaresan <sujaritha.sundaresan at intel.com>
>>> ---
>>>   drivers/gpu/drm/i915/intel_wopcm.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/intel_wopcm.c 
>>> b/drivers/gpu/drm/i915/intel_wopcm.c
>>> index f82a415ea2ba..a651557e6e4e 100644
>>> --- a/drivers/gpu/drm/i915/intel_wopcm.c
>>> +++ b/drivers/gpu/drm/i915/intel_wopcm.c
>>> @@ -169,7 +169,7 @@ int intel_wopcm_init(struct intel_wopcm *wopcm)
>>>       GEM_BUG_ON(!wopcm->size);
>>>       if (i915_inject_load_failure())
>>> -        return -E2BIG;
>>> +        return -ENODEV;
>>
>> Would it be better to just change the final return code in 
>> i915_pci_probe() to ENODEV?
>>
>> i.e.
>>      err = i915_driver_load(pdev, ent);
>>      if (err)
>>          return i915_error_injected() ? -ENODEV : err;
>>
>> This way we can test that the possible error codes (E2BIG in this 
>> case) don't hit issues in dedicated cases during onion unwinding.
>>
>> Daniele
>>
>>>       if (guc_fw_size >= wopcm->size) {
>>>           DRM_ERROR("GuC FW (%uKiB) is too big to fit in WOPCM.",
>>>

Okay this approach seems to provide a wider fix. Will make this change. 
Thanks for the review.

Sujaritha



More information about the Intel-gfx mailing list