On Thu, 2019-02-21 at 21:41 +0330, Henri Verbeet wrote:
On Wed, 20 Feb 2019 at 21:11, Matteo Bruni mbruni@codeweavers.com wrote:
dlls/d3d9/d3d9_private.h | 2 ++ dlls/d3d9/device.c | 24 ++++++++++++++++++++++-- dlls/d3d9/tests/visual.c | 22 ++++++++++++++++++++++ dlls/d3d9/vertexdeclaration.c | 2 +- 4 files changed, 47 insertions(+), 3 deletions(-)
Why do we care? (I've read the bug report, but assume I haven't.) One of the reasons for being slightly sceptical about the "Wine-Bug:" tags is that we want that kind of information in the source itself instead of having to dig through bug reports.
I can add a comment in the code explaining that this isn't just an optimization, sure. I thought that the test would be enough but I guess I can see, some time from now, having to look up the patch introducing the code to figure out what's going on is not great.
I could add a comment to the new test too, although I don't think it could be more specific than something like "Applications in the wild do release vertex buffers without unbinding them, make sure things don't break in this case".