3.5.7.19 -stable review patch. If anyone has any objections, please let me know. ------------------ From: Egbert Eich <eich@xxxxxxxx> commit ecaac1c866bcda4780a963b3d18cd310d971aea3 upstream. When a BO gets pinned the placement may get changed. If the memory is mapped into user space and user space has already accessed the mapped range the page tables are set up but now point to the wrong memory. Set bo.mdev->dev_mapping in mgag200_bo_create() to make sure that ttm_bo_unmap_virtual() called from ttm_bo_handle_move_mem() will take care of this. v2: Don't call ttm_bo_unmap_virtual() in mgag200_bo_pin(), fix comment. Signed-off-by: Egbert Eich <eich@xxxxxxxx> Signed-off-by: Dave Airlie <airlied@xxxxxxxxxx> Signed-off-by: Luis Henriques <luis.henriques@xxxxxxxxxxxxx> --- drivers/gpu/drm/mgag200/mgag200_ttm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/mgag200/mgag200_ttm.c b/drivers/gpu/drm/mgag200/mgag200_ttm.c index a707394..95b5377 100644 --- a/drivers/gpu/drm/mgag200/mgag200_ttm.c +++ b/drivers/gpu/drm/mgag200/mgag200_ttm.c @@ -347,6 +347,7 @@ int mgag200_bo_create(struct drm_device *dev, int size, int align, mgabo->gem.driver_private = NULL; mgabo->bo.bdev = &mdev->ttm.bdev; + mgabo->bo.bdev->dev_mapping = dev->dev_mapping; mgag200_ttm_placement(mgabo, TTM_PL_FLAG_VRAM | TTM_PL_FLAG_SYSTEM); -- 1.8.3.2 -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html