On Tue, Nov 18, 2014 at 02:06:37PM +0000, Chris Wilson wrote:
> On Tue, Nov 18, 2014 at 02:40:28PM +0100, Daniel Vetter wrote:
> > The only thing the kernel can do is pin the buffers, which essentially
> > means no swapped tiled objects.
> >
> > Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxxx>
> > ---
> > tests/gem_tiled_swapping.c | 22 ++++++++++++++++++++++
> > 1 file changed, 22 insertions(+)
> >
> > diff --git a/tests/gem_tiled_swapping.c b/tests/gem_tiled_swapping.c
> > index 69d1cfae94d2..0839df00151e 100644
> > --- a/tests/gem_tiled_swapping.c
> > +++ b/tests/gem_tiled_swapping.c
> > @@ -61,6 +61,7 @@
> > #include "drmtest.h"
> > #include "intel_io.h"
> > #include "igt_aux.h"
> > +#include "igt_debugfs.h"
> >
> > #define WIDTH 512
> > #define HEIGHT 512
> > @@ -143,6 +144,25 @@ static void thread_fini(struct thread *t)
> > free(t->idx_arr);
> > }
> >
> > +static void check_memory_layout(void)
> > +{
> > + FILE *tiling_debugfs_file;
> > + char *line = NULL;
> > + size_t sz = 0;
> > +
> > + tiling_debugfs_file = igt_debugfs_fopen("i915_swizzle_info", "r");
> > + igt_assert(tiling_debugfs_file);
> > +
> > + while (getline(&line, &sz, tiling_debugfs_file) > 0) {
> > + if (strstr(line, "L-shaped") != 0)
> > + continue;
> > +
> > + igt_skip("L-shaped memory configuration detected\n");
> > + }
> > +
> > + igt_debug("normal memory configuration detected, continuing\n");
> > +}
>
> igt_check_memory():
> if (flags & CHECK_SWAP && !check_memory_layout())
> skip()
>
> That should then cover us for all tests that want to use swap. (And if
> not, we are missing some more igt_check_memory/igt_require_memory.)
Lots of our memory thrashin tests use linear objects, and I don't really
want to disable all these tests. I know somewhat futile since the only big
things are tiled objects, still.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
