Displaying 3 results from an estimated 3 matches for "add37".
Did you mean:
add32
2017 Feb 08
3
Problem ScheduleDAG on PowerPC, X86 works fine.
...ticular test case on x86, because the dag ends up looking like (uaddo X,
(adde Y, 0, Carry)), which the transform doesn't match. That is not really
relevant to the issue here, because the x86 backend does work even if the
transform gets triggered (replacing the final store with "store i64 %add37,
i64* %r, align 8" will make this happen)
On Tue, Feb 7, 2017 at 5:44 PM, Amaury SECHET <deadalnix at gmail.com> wrote:
> Making this a value instead of a glue looks like a good longer term
> solution, but it doesn't quite cut it as a short term one. It would require
> to...
2017 Feb 09
2
Problem ScheduleDAG on PowerPC, X86 works fine.
...ecause the dag ends up looking like (uaddo X,
>> (adde Y, 0, Carry)), which the transform doesn't match. That is not really
>> relevant to the issue here, because the x86 backend does work even if the
>> transform gets triggered (replacing the final store with "store i64 %add37,
>> i64* %r, align 8" will make this happen)
>>
>>
>> On Tue, Feb 7, 2017 at 5:44 PM, Amaury SECHET <deadalnix at gmail.com>
>> wrote:
>>
>>> Making this a value instead of a glue looks like a good longer term
>>> solution, but it doe...
2017 Feb 07
2
Problem ScheduleDAG on PowerPC, X86 works fine.
Would it not make sense to refactor the code so those don't use glue rather
than emitting them with glue and then getting rid of it. There are times
when we would like to emit these in separate blocks but can't (presumably
because of the glue).
On Tue, Feb 7, 2017 at 9:15 PM, James Y Knight via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
> That's seems really odd that