diff options
| author | Dave Airlie <airlied@redhat.com> | 2011-03-14 09:47:24 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2011-03-14 12:51:04 +1000 |
| commit | 5359533801e3dd3abca5b7d3d985b0b33fd9fe8b (patch) | |
| tree | 5ab728c08b4cb5fae0116396911ed3757540b7b8 /drivers/gpu/drm/radeon/r600_blit_kms.c | |
| parent | c640e8ca172c6a5c45abe8e2e8353900a84427fa (diff) | |
drm/radeon: fix problem with changing active VRAM size. (v2)
So we used to use lpfn directly to restrict VRAM when we couldn't
access the unmappable area, however this was removed in
93225b0d7bc030f4a93165347a65893685822d70 as it also restricted
the gtt placements. However it was only later noticed that this
broke on some hw.
This removes the active_vram_size, and just explicitly sets it
when it changes, TTM/drm_mm will always use the real_vram_size,
and the active vram size will change the TTM size used for lpfn
setting.
We should re-work the fpfn/lpfn to per-placement at some point
I suspect, but that is too late for this kernel.
Hopefully this addresses:
https://bugs.freedesktop.org/show_bug.cgi?id=35254
v2: fix reported useful VRAM size to userspace to be correct.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Diffstat (limited to 'drivers/gpu/drm/radeon/r600_blit_kms.c')
| -rw-r--r-- | drivers/gpu/drm/radeon/r600_blit_kms.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c index 41f7aafc97c..df68d91e819 100644 --- a/drivers/gpu/drm/radeon/r600_blit_kms.c +++ b/drivers/gpu/drm/radeon/r600_blit_kms.c @@ -558,7 +558,7 @@ done: dev_err(rdev->dev, "(%d) pin blit object failed\n", r); return r; } - rdev->mc.active_vram_size = rdev->mc.real_vram_size; + radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size); return 0; } @@ -566,7 +566,7 @@ void r600_blit_fini(struct radeon_device *rdev) { int r; - rdev->mc.active_vram_size = rdev->mc.visible_vram_size; + radeon_ttm_set_active_vram_size(rdev, rdev->mc.visible_vram_size); if (rdev->r600_blit.shader_obj == NULL) return; /* If we can't reserve the bo, unref should be enough to destroy |