[llvm-commits] [patch] Drop DwarfRegNum from the D registers

[Akira Hatanaka]

Could you tell me what I need to add/change in
MipsFrameLowering::emitPrologue in order to emit the correct cfi_offset
directives when double precision callee-saved registers are saved?
Currently, the following directive gets emitted when double register $f20 is
saved,
.cfi_offset -1, -8

instead of something like this:
.cfi_offset 53, -4
.cfi_offset 52, -8

Should I just push back two MachineMoves (one for each of the paired single
precision registers)?

Thank you.

On Fri, May 27, 2011 at 8:44 AM, Rafael Avila de Espindola <
rafael.espindola at gmail.com> wrote:

> If I understand it correctly, the proper way to encode a D register in
> DWARF is with something like
>
> DW_OP_reg 32 DW_OP_piece 4 DW_OP_reg 33
>
> The attached patch drops the DwarfRegNum from the D registers. Is it OK?
>
> Cheers,
> Rafael
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20110531/f5d733ab/attachment.html>