scratch this one, I've just thought of a better one (coming in a few minutes)
On Mon, 30 Jan 2012, Ilija Hadzic wrote: > copy_blit operation works only on integral number of pages > so benchmarks shorter than one page size (4K) do not make sense > > Signed-off-by: Ilija Hadzic <ihadzic at research.bell-labs.com> > --- > drivers/gpu/drm/radeon/radeon_benchmark.c | 6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c > b/drivers/gpu/drm/radeon/radeon_benchmark.c > index 815f234..a500493 100644 > --- a/drivers/gpu/drm/radeon/radeon_benchmark.c > +++ b/drivers/gpu/drm/radeon/radeon_benchmark.c > @@ -208,21 +208,21 @@ 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) > + for (i = 4; i <= 65536; i <<= 1) > radeon_benchmark_move(rdev, i*1024, > 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) > + for (i = 4; i <= 65536; i <<= 1) > radeon_benchmark_move(rdev, i*1024, > 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) > + for (i = 4; i <= 65536; i <<= 1) > radeon_benchmark_move(rdev, i*1024, > RADEON_GEM_DOMAIN_VRAM, > RADEON_GEM_DOMAIN_VRAM); > -- > 1.7.7 > >
