[cfe-commits] [PATCH] Add PNaCl ABIInfo

[David Meyer]

Eli,

Why is it generating lower-quality code? What kind of resolution do
you think we could come to on this issue?

For PNaCl, we lower first-class struct passing in a target-specific
way (in llc) so that we can approximately (but not exactly) match the
platform ABIs. We'd prefer full ABI compatibility, but we're willing
to live with partial compatibility for now.

Attached is a new patch. Also note, I don't (yet?) have commit access.

Thanks,
- pdox


On Tue, Sep 20, 2011 at 2:03 PM, Eli Friedman <eli.friedman at gmail.com> wrote:
> On Tue, Sep 20, 2011 at 10:09 AM, David Meyer <pdox at google.com> wrote:
>> Ping! Anybody available to review?
>
> I'm not really sure the ABIArgInfo::Abstract stuff is really
> appropriate, in that it is likely to generate low-quality code as-is.
> I was sort of waiting for a resolution to that discussion on cfe-dev.
>
> I don't see any issues with the patch otherwise, though.
>
> +      if (!DestPtr) {
> +         DestPtr = CreateMemTemp(RetTy, "agg.tmp");
> +      }
>
> Nit: LLVM style is no braces here.
>
> -Eli
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pnacl_abi_v2.patch
Type: text/x-patch
Size: 8304 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20110920/7f678b1e/attachment.bin>