Re: [PATCH v5 2/2] drm/xe: Increase devcoredump timeout

[Rodrigo Vivi]

On Mon, Jun 10, 2024 at 09:11:33AM -0700, José Roberto de Souza wrote:
> 5 minutes is too short for a regular user to search and understand
> what he needs to do to report capture devcoredump and report a bug to
> us, so here increasing this timeout to 1 hour.
> 
> Cc: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>
> Cc: Jonathan Cavitt <jonathan.cavitt@xxxxxxxxx>
> Acked-by: Jonathan Cavitt <jonathan.cavitt@xxxxxxxxx>
> Signed-off-by: José Roberto de Souza <jose.souza@xxxxxxxxx>
> ---
>  drivers/gpu/drm/xe/xe_devcoredump.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c
> index d7f2d19a77c10..62c2b10fbf1d2 100644
> --- a/drivers/gpu/drm/xe/xe_devcoredump.c
> +++ b/drivers/gpu/drm/xe/xe_devcoredump.c
> @@ -53,6 +53,9 @@
>  
>  #ifdef CONFIG_DEV_COREDUMP
>  
> +/* 1 hour timeout */
> +#define XE_COREDUMP_TIMEOUT_JIFFIES (60 * 60 * HZ)

o.O! 1h?!
we should likely already add a config option for that.
but anyway, let's move with that and adjust as we go.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@xxxxxxxxx>

> +
>  static struct xe_device *coredump_to_xe(const struct xe_devcoredump *coredump)
>  {
>  	return container_of(coredump, struct xe_device, devcoredump);
> @@ -247,8 +250,9 @@ void xe_devcoredump(struct xe_sched_job *job)
>  	drm_info(&xe->drm, "Check your /sys/class/drm/card%d/device/devcoredump/data\n",
>  		 xe->drm.primary->index);
>  
> -	dev_coredumpm(xe->drm.dev, THIS_MODULE, coredump, 0, GFP_KERNEL,
> -		      xe_devcoredump_read, xe_devcoredump_free);
> +	dev_coredumpm_timeout(xe->drm.dev, THIS_MODULE, coredump, 0, GFP_KERNEL,
> +			      xe_devcoredump_read, xe_devcoredump_free,
> +			      XE_COREDUMP_TIMEOUT_JIFFIES);
>  }
>  
>  static void xe_driver_devcoredump_fini(void *arg)
> -- 
> 2.45.2
>