[Mesa-dev] [PATCH v2] texgetimage: Add check for the target argument to GetTextureSubImage
Ilia Mirkin
imirkin at alum.mit.edu
Thu Feb 2 20:24:04 UTC 2017
On Thu, Feb 2, 2017 at 2:50 PM, Eduardo Lima Mitev <elima at igalia.com> wrote:
> This is currently not being checked and e.g a multisample texture image is passed
> down to the driver hook. On i965, it is crashing the driver with an assertion:
>
> intel_mipmap_tree.c:3125: intel_miptree_map: Assertion `mt->num_samples <= 1' failed.
>
> Similar check exists for GetTextureSubImage.
>
> v2: (Ilia Mirkin) Use legal_getteximage_target() instead.
> ---
> src/mesa/main/texgetimage.c | 11 +++++++++++
> 1 file changed, 11 insertions(+)
>
> diff --git a/src/mesa/main/texgetimage.c b/src/mesa/main/texgetimage.c
> index d5cb1636605..c9a0ef0b8ac 100644
> --- a/src/mesa/main/texgetimage.c
> +++ b/src/mesa/main/texgetimage.c
> @@ -1465,6 +1465,17 @@ _mesa_GetTextureSubImage(GLuint texture, GLint level,
> return;
> }
>
> + if (!legal_getteximage_target(ctx, texObj->Target, true)) {
Sorry, perhaps I wasn't clear enough. Actually I don't think this
needs to be called here at all - it doesn't really add anything, since
a target isn't passed in here, and the extension checking isn't really
important since the texture object can't just appear out of nowhere
with an illegal target. [And the same argument applies to
GetTextureImage as well...] Just the if statement below without the
else clause. I believe the rest is covered by getteximage_error_check.
> + if (texObj->Target == GL_TEXTURE_BUFFER ||
> + texObj->Target == GL_TEXTURE_2D_MULTISAMPLE) {
GL_TEXTURE_2D_MULTISAMPLE_ARRAY as well presumably?
> + _mesa_error(ctx, GL_INVALID_OPERATION,
> + "%s(buffer/multisample texture)", caller);
> + } else {
> + _mesa_error(ctx, GL_INVALID_ENUM, "%s(target)", caller);
> + }
> + return;
> + }
> +
> if (getteximage_error_check(ctx, texObj, texObj->Target, level,
> xoffset, yoffset, zoffset, width, height, depth,
> format, type, bufSize, pixels, caller)) {
> --
> 2.11.0
>
More information about the mesa-dev
mailing list