drm/i915: Write GPU relocs harder with gen3

Under moderate amounts of GPU stress, we can observe on Bearlake and
Pineview (later gen3 models) that we execute the following batch buffer
before the write into the batch is coherent. Adding extra (tested with
upto 32x) MI_FLUSH to either the invalidation, flush or both phases does
not solve the incoherency issue with the relocations, but emitting the
MI_STORE_DWORD_IMM twice does. So be it.

Fixes: 7dd4f6729f ("drm/i915: Async GPU relocation processing")
Testcase: igt/gem_tiled_fence_blits # blb/pnv
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20181119154153.15327-1-chris@chris-wilson.co.uk
(cherry picked from commit 7fa28e1469)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
This commit is contained in:
Chris Wilson 2018-11-19 15:41:53 +00:00 committed by Joonas Lahtinen
parent 8830f26bcd
commit f8577fb3c2
1 changed files with 6 additions and 1 deletions

View File

@ -1268,7 +1268,7 @@ relocate_entry(struct i915_vma *vma,
else if (gen >= 4)
len = 4;
else
len = 3;
len = 6;
batch = reloc_gpu(eb, vma, len);
if (IS_ERR(batch))
@ -1309,6 +1309,11 @@ relocate_entry(struct i915_vma *vma,
*batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL;
*batch++ = addr;
*batch++ = target_offset;
/* And again for good measure (blb/pnv) */
*batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL;
*batch++ = addr;
*batch++ = target_offset;
}
goto out;