On 13-09-07 9:01 AM, Richard Sandiford wrote:
Jakub Jelinek writes:
On Sat, Sep 07, 2013 at 12:37:14PM +0100, Richard Sandiford wrote:
Steven Bosscher writes:
Can you please add a test case?
What kind of test would you suggest? Do we have a harness for testing
that -O2 and -O2 -g .text out
Jakub Jelinek writes:
> On Sat, Sep 07, 2013 at 12:37:14PM +0100, Richard Sandiford wrote:
>> Steven Bosscher writes:
>> > Can you please add a test case?
>>
>> What kind of test would you suggest? Do we have a harness for testing
>> that -O2 and -O2 -g .text output is identical?
>
> No, but we
Jakub Jelinek writes:
> On Sat, Sep 07, 2013 at 12:37:14PM +0100, Richard Sandiford wrote:
>> Steven Bosscher writes:
>> > Can you please add a test case?
>>
>> What kind of test would you suggest? Do we have a harness for testing
>> that -O2 and -O2 -g .text output is identical?
>
> No, but we
On Sat, Sep 07, 2013 at 12:37:14PM +0100, Richard Sandiford wrote:
> Steven Bosscher writes:
> > Can you please add a test case?
>
> What kind of test would you suggest? Do we have a harness for testing
> that -O2 and -O2 -g .text output is identical?
No, but we have -fcompare-debug, which is e
On Sat, Sep 7, 2013 at 1:37 PM, Richard Sandiford wrote:
> Steven Bosscher writes:
>> Can you please add a test case?
>
> What kind of test would you suggest? Do we have a harness for testing
> that -O2 and -O2 -g .text output is identical?
Not .text, but the assembly output or the final RTL du
Steven Bosscher writes:
> Can you please add a test case?
What kind of test would you suggest? Do we have a harness for testing
that -O2 and -O2 -g .text output is identical?
I suppose the main testcase will be bootstrap if wide-int ever does get
accepted.
Thanks,
Richard
On Sat, Sep 7, 2013 at 11:14 AM, Richard Sandiford wrote:
> The problem seems to be split across IRA and LRA. In IRA we have:
>
> FOR_EACH_BB (bb)
> FOR_BB_INSNS (bb, insn)
> {
> if (! INSN_P (insn))
> continue;
> for_each_rtx (&insn, set_paradoxical_subreg, (
While doing some work on wide-int, I hit a bootstrap comparison failure that
was caused by the RA output depending on debug insns. There was a register R
that was used normally by all "real" insns, but which was used in a paradoxical
subreg by one of the debug insns. There was also a natural equi
