rjmccall added a comment.
That's a fair point. I agree that separate allocas would make this a lot
cleaner, in both IR and frontend implementation. We could also set an inalloca
bit (+ field index? or maybe keep the arg index -> field index mapping on the
CGFunctionInfo) on each arg info *in
rnk added a comment.
In https://reviews.llvm.org/D43842#1022546, @rjmccall wrote:
> Okay. In that case, this seems correct, although it seems to me that perhaps
> `inalloca` is not actually orthogonal to anything else. In fact, it seems to
> me that maybe `inalloca` ought to just be a bit on
compnerd closed this revision.
compnerd added a comment.
SVN r326362
Repository:
rC Clang
https://reviews.llvm.org/D43842
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
rjmccall accepted this revision.
rjmccall added a comment.
This revision is now accepted and ready to land.
Alright, yeah, we can fix that later. LGTM.
Repository:
rC Clang
https://reviews.llvm.org/D43842
___
cfe-commits mailing list
cfe-commits
compnerd added a comment.
Yeah, this is still an indirect return. I can see your point about the
representation, nfortunately, I think that change is way out of scope for this.
That would be a pretty large and invasive change to wire that through.
Repository:
rC Clang
https://reviews.llvm
rjmccall added a comment.
Okay. In that case, this seems correct, although it seems to me that perhaps
`inalloca` is not actually orthogonal to anything else. In fact, it seems to
me that maybe `inalloca` ought to just be a bit on the CGFunctionInfo and the
individual ABIInfos should be left
smeenai added a comment.
In https://reviews.llvm.org/D43842#1022498, @rjmccall wrote:
> Ugh, I hate `inalloca` *so much*.
>
> It's still an indirect return, right? It's just that the return-slot pointer
> has to get stored to the `inalloca` allocation like any other argument?
Correct.
Repos
rjmccall added a comment.
Ugh, I hate `inalloca` *so much*.
It's still an indirect return, right? It's just that the return-slot pointer
has to get stored to the `inalloca` allocation like any other argument?
Repository:
rC Clang
https://reviews.llvm.org/D43842
_
compnerd created this revision.
compnerd added reviewers: rjmccall, rnk.
objc_msgSend_stret takes a hidden parameter for the returned structure's
address for the construction. When the function signature is rewritten
for the inalloca passing, the return type is no longer marked as
indirect but ra
