[PATCH v2] drm/radeon/kms: remove benchmarks shorter than one page

[Alex Deucher]

On Mon, Jan 30, 2012 at 11:10 PM, Ilija Hadzic
<ihadzic at research.bell-labs.com> wrote:
> copy_blit operation works only on integral number of pages
> so benchmarks shorter than one page size (4K) do not make sense
>
> v2: use RADEON_GPU_PAGE_SIZE instead of "magic" 1024 number and
>    sweep sizes between 1x<page_size> to 16x<page_size> doubling
>    the size in each iteration; we get the same coverage, as
>    in the original benchmark, but guarantee integer multiples
>    of page size
>
> Signed-off-by: Ilija Hadzic <ihadzic at research.bell-labs.com>

Reviewed-by: Alex Deucher <alexander.deucher at amd.com>

> ---
>  drivers/gpu/drm/radeon/radeon_benchmark.c |   12 ++++++------
>  1 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c b/drivers/gpu/drm/radeon/radeon_benchmark.c
> index 815f234..d3c0002 100644
> --- a/drivers/gpu/drm/radeon/radeon_benchmark.c
> +++ b/drivers/gpu/drm/radeon/radeon_benchmark.c
> @@ -208,22 +208,22 @@ void radeon_benchmark(struct radeon_device *rdev, int test_number)
>                break;
>        case 3:
>                /* GTT to VRAM, buffer size sweep, powers of 2 */
> -               for (i = 1; i <= 65536; i <<= 1)
> -                       radeon_benchmark_move(rdev, i*1024,
> +               for (i = 1; i <= 16384; i <<= 1)
> +                       radeon_benchmark_move(rdev, i*RADEON_GPU_PAGE_SIZE,
>                                              RADEON_GEM_DOMAIN_GTT,
>                                              RADEON_GEM_DOMAIN_VRAM);
>                break;
>        case 4:
>                /* VRAM to GTT, buffer size sweep, powers of 2 */
> -               for (i = 1; i <= 65536; i <<= 1)
> -                       radeon_benchmark_move(rdev, i*1024,
> +               for (i = 1; i <= 16384; i <<= 1)
> +                       radeon_benchmark_move(rdev, i*RADEON_GPU_PAGE_SIZE,
>                                              RADEON_GEM_DOMAIN_VRAM,
>                                              RADEON_GEM_DOMAIN_GTT);
>                break;
>        case 5:
>                /* VRAM to VRAM, buffer size sweep, powers of 2 */
> -               for (i = 1; i <= 65536; i <<= 1)
> -                       radeon_benchmark_move(rdev, i*1024,
> +               for (i = 1; i <= 16384; i <<= 1)
> +                       radeon_benchmark_move(rdev, i*RADEON_GPU_PAGE_SIZE,
>                                              RADEON_GEM_DOMAIN_VRAM,
>                                              RADEON_GEM_DOMAIN_VRAM);
>                break;
> --
> 1.7.7
>
> _______________________________________________
> dri-devel mailing list
> dri-devel at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel