[PATCH] d3dx9_36: Implement ID3DXSkinInfoImpl_UpdateSkinnedMesh.

Zebediah Figura z.figura12 at gmail.com
Wed Apr 17 12:20:14 CDT 2019


On 04/10/2019 12:40 PM, Vijay Kiran Kamuju wrote:
> It seems I have to do a deeper analysis of the code.
> I checked only whether the tests are successful or not.

For what it's worth (and since the concern was brought up to me offline
by someone else), I feel like I should point out that Staging patches
are not necessarily correct, either functionally or stylistically.
That's the basic reason (even if it's not the only reason) that those
patches are in Staging. Hence anyone who's interested in upstreaming
Staging patches should treat the patch as their own responsibility, and
should ensure that the patch is correct, and ideally have also tested it
themself. That's part of what the signoff means. It's also the reason
why Alistair and I are slow to upstream patches from Staging. There were
a few hundred we tried to push upstream at an early point when we took
over the repository, but most of those were the easy ones. We're still
stuck with 800 patches which are nontrivial or take a fair amount of
background knowledge to understand (like everything involved with ACLs)
or, worse, which have no bug report or test application attached to them
and thus no way to ensure that they are actually doing the right thing
and fixing the problem (or, even more critically, will still be doing
the right thing if we make some modifications.)

Bottom line is, inasmuch as you should be expected to understand and
write your own patches and ensure they're correct, the same rules apply
to patches someone else wrote. You can't implicitly trust that they're
correct, even according to that person's standards, and you certainly
can't trust that they're correct just because they're in Staging.



More information about the wine-devel mailing list