> I've just started to work on fixing this in LRA. But I am not going to
> work on fixing this in reload. It would be just wasting enormous amount
> of time.
Yes, fixing it in LRA only would be sufficient for PR rtl-optimization/59477,
since it's an x86-specific issue. Other architectures aren
On 1/20/2014, 4:12 AM, Eric Botcazou wrote:
I think the problem is still either a missed feature in LRA/reload
(split live-ranges), a problem in how we represent calls & argument
setup (magic hard-reg uses), or a backend problem (should spill
itself if necessary, via a post-reload splitter or alw
> I think the problem is still either a missed feature in LRA/reload
> (split live-ranges), a problem in how we represent calls & argument
> setup (magic hard-reg uses), or a backend problem (should spill
> itself if necessary, via a post-reload splitter or always spill
> and undo via a peephole2).
On 16 January 2014 08:26, Jakub Jelinek wrote:
> I wonder if instead of not creating the LOG_LINKS we could just somehow taint
> them, add a flag to struct insn_link that we need to be careful about
> combining across that link, then in combine_instructions or the flags together
> from all the us
On Wed, 15 Jan 2014, Jakub Jelinek wrote:
> Hi!
>
> As discussed in the PR, when combine combines something across a setter
> of a likely spilled non-fixed hard register, we may get RA ICE, as the
> combined insn might need for reload a hard reg that is live at the point
> where combiner has comb
On Thu, Jan 16, 2014 at 03:41:57AM +, Joern Rennecke wrote:
> On 15 January 2014 21:55, Jakub Jelinek wrote:
> ...
> > The patch removes the likely_spilled_retval stuff Joern wrote a few years
> > ago because I believe this change should obsolete that. But, have tried to
> > reproduce the ICE
On 15 January 2014 21:55, Jakub Jelinek wrote:
...
> The patch removes the likely_spilled_retval stuff Joern wrote a few years
> ago because I believe this change should obsolete that. But, have tried to
> reproduce the ICE using sh-elf target and haven't managed to do so, 4.9 is
> quite a differ
Hi!
As discussed in the PR, when combine combines something across a setter
of a likely spilled non-fixed hard register, we may get RA ICE, as the
combined insn might need for reload a hard reg that is live at the point
where combiner has combined the insn.
The following patch attempts to fix tha
