[PATCH 3.10 35/86] drm/vmwgfx: Make sure user-space cant DMA across buffer object boundaries v2

From: Greg Kroah-Hartman
Date: Thu May 29 2014 - 00:54:33 EST


3.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Hellstrom <thellstrom@xxxxxxxxxx>

commit cbd75e97a525e3819c02dc18bc2d67aa544c9e45 upstream.

We already check that the buffer object we're accessing is registered with
the file. Now also make sure that we can't DMA across buffer object boundaries.

v2: Code commenting update.

Signed-off-by: Thomas Hellstrom <thellstrom@xxxxxxxxxx>
Reviewed-by: Jakob Bornecrantz <jakob@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)

--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -834,14 +834,36 @@ static int vmw_cmd_dma(struct vmw_privat
SVGA3dCmdSurfaceDMA dma;
} *cmd;
int ret;
+ SVGA3dCmdSurfaceDMASuffix *suffix;
+ uint32_t bo_size;

cmd = container_of(header, struct vmw_dma_cmd, header);
+ suffix = (SVGA3dCmdSurfaceDMASuffix *)((unsigned long) &cmd->dma +
+ header->size - sizeof(*suffix));
+
+ /* Make sure device and verifier stays in sync. */
+ if (unlikely(suffix->suffixSize != sizeof(*suffix))) {
+ DRM_ERROR("Invalid DMA suffix size.\n");
+ return -EINVAL;
+ }
+
ret = vmw_translate_guest_ptr(dev_priv, sw_context,
&cmd->dma.guest.ptr,
&vmw_bo);
if (unlikely(ret != 0))
return ret;

+ /* Make sure DMA doesn't cross BO boundaries. */
+ bo_size = vmw_bo->base.num_pages * PAGE_SIZE;
+ if (unlikely(cmd->dma.guest.ptr.offset > bo_size)) {
+ DRM_ERROR("Invalid DMA offset.\n");
+ return -EINVAL;
+ }
+
+ bo_size -= cmd->dma.guest.ptr.offset;
+ if (unlikely(suffix->maximumOffset > bo_size))
+ suffix->maximumOffset = bo_size;
+
ret = vmw_cmd_res_check(dev_priv, sw_context, vmw_res_surface,
user_surface_converter, &cmd->dma.host.sid,
NULL);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/