Hello Michael, A very good Morning to you.
On Wed, Nov 20, 2019 at 10:58 PM Michael Kruse wrote:
> Am Mi., 20. Nov. 2019 um 10:21 Uhr schrieb HAPPY Mahto
> :
> >> #pragma clang loop vectorize_assume_alignment(32)
> >> for(int i = 0;i < n; i++){
> >> a[i] = b[i] + i*i;
> >> }
> >
> > for this al
Am Mi., 20. Nov. 2019 um 10:21 Uhr schrieb HAPPY Mahto
:
>> #pragma clang loop vectorize_assume_alignment(32)
>> for(int i = 0;i < n; i++){
>> a[i] = b[i] + i*i;
>> }
>
> for this all-access inside the loop will be aligned to 32bit,
> ex IR
>>
>> for.cond:
Hello Michael,
Very sorry for the late reply, we had exams and assignments this week and I
had to read about _builtin_assume_aligned as I didn't come across this.
#pragma clang loop vectorize_assume_alignment(32)
> for(int i = 0;i < n; i++){
> a[i] = b[i] + i*i;
> }
>
for this all-access inside t
Meinersbur added a comment.
Could you elaborate why this is better than `__builtin_assume_aligned`?
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D69897/new/
https://reviews.llvm.org/D69897
___
cfe-comm
lebedev.ri added a comment.
To reword, if `vectorize_assume_alignment(32)` is *NOT* lowered
via `CodeGenFunction::EmitAlignmentAssumption()` in clang frontend,
but lowered into `LoopAttributes::VectorizeAssumeAlignment`,
then once the alignment that was specified does not match reality,
there wil
m-happy updated this revision to Diff 229190.
m-happy added a comment.
Updated the syntax as suggested by Michael Kruse, and added functionality to
specify the number used for alignment instead of just using 32bits.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://revi
