[Intel-gfx] [PATCH] drm/i915/gt: Poison GTT scratch pages

[Matthew Auld]

On Thu, 23 Jan 2020 at 08:51, Chris Wilson <chris at chris-wilson.co.uk> wrote:
>
> Using a clear page for scratch means that we have relatively benign
> errors in case it is accidentally used, but that can be rather too
> benign for debugging. If we poison the scratch, ideally it quickly
> results in an obvious error.
>
> Suggested-by: Mika Kuoppala <mika.kuoppala at linux.intel.com>
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>

[snip]

>
> +static void poison_scratch_page(struct page *page, unsigned long size)
> +{
> +       if (!IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))
> +               return;
> +
> +       kunmap_atomic(memset(kmap_atomic(page), POISON_FREE, size));

kmap works for !order-zero?

> +}
> +
>  int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
>  {
>         unsigned long size;
> @@ -331,6 +339,17 @@ int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp)
>                 if (unlikely(!page))
>                         goto skip;
>
> +               /*
> +                * Use a non-zero scratch page for debugging.
> +                *
> +                * We want a value that should be reasonably obvious
> +                * to spot in the error state, while also causing a GPU hang
> +                * if executed. We prefer using a clear page in production, so
> +                * should it ever be accidentally used, the effect should be
> +                * fairly benign.
> +                */
> +               poison_scratch_page(page, size);
> +
>                 addr = dma_map_page_attrs(vm->dma,
>                                           page, 0, size,
>                                           PCI_DMA_BIDIRECTIONAL,
> --
> 2.25.0
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx