[Intel-gfx] [PATCH 04/18] drm/i915: Only emit the 'send bug report' once for a GPU hang

[Mika Kuoppala]

Chris Wilson <chris at chris-wilson.co.uk> writes:

> Use a locked xchg to ensure that the global log message giving
> instructions on how to send a bug report is emitted precisely once.
>

Well you must have run into this as this patch appeared.
Arguably better than some read/write once trickery.

Reviewed-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>

> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> ---
>  drivers/gpu/drm/i915/i915_gpu_error.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
> index b0ae055b6c82..e284bd76fa86 100644
> --- a/drivers/gpu/drm/i915/i915_gpu_error.c
> +++ b/drivers/gpu/drm/i915/i915_gpu_error.c
> @@ -1765,7 +1765,7 @@ void i915_capture_error_state(struct drm_i915_private *i915,
>  		return;
>  	}
>  
> -	if (!warned &&
> +	if (!xchg(&warned, true) &&
>  	    ktime_get_real_seconds() - DRIVER_TIMESTAMP < DAY_AS_SECONDS(180)) {
>  		pr_info("GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.\n");
>  		pr_info("Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel\n");
> @@ -1773,7 +1773,6 @@ void i915_capture_error_state(struct drm_i915_private *i915,
>  		pr_info("The GPU crash dump is required to analyze GPU hangs, so please always attach it.\n");
>  		pr_info("GPU crash dump saved to /sys/class/drm/card%d/error\n",
>  			i915->drm.primary->index);
> -		warned = true;
>  	}
>  }
>  
> -- 
> 2.23.0.rc1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx