Hmm, I don't trust myself with gotos yet because I still don't have much experience writing them. So maybe I'll use them for a different patch later.
On Wed, Apr 1, 2015 at 11:56 AM, Ilia Mirkin <[email protected]> wrote: > Reviewed-by: Ilia Mirkin <[email protected]> > > An alternative, btw, is to do a goto to the existing unlock. This > would have the advantage of keeping a single unlock, and a single > return path in the function. Your call, you get the R-b either way. > > -ilia > > On Wed, Apr 1, 2015 at 2:23 PM, Laura Ekstrand <[email protected]> > wrote: > > Ilia Mirkin found that I had forgotten to free the mutex in the error > case. > > --- > > src/mesa/main/bufferobj.c | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c > > index b67e1c4..c7ef90b 100644 > > --- a/src/mesa/main/bufferobj.c > > +++ b/src/mesa/main/bufferobj.c > > @@ -1353,6 +1353,7 @@ create_buffers(GLsizei n, GLuint *buffers, bool > dsa) > > buf = ctx->Driver.NewBufferObject(ctx, buffers[i]); > > if (!buf) { > > _mesa_error(ctx, GL_OUT_OF_MEMORY, "%s", func); > > + mtx_unlock(&ctx->Shared->Mutex); > > return; > > } > > } > > -- > > 2.1.0 > > > > _______________________________________________ > > mesa-dev mailing list > > [email protected] > > http://lists.freedesktop.org/mailman/listinfo/mesa-dev >
_______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
