On Sat, Sep 23, 2023 at 2:54 PM Hanke Zhang via Gcc wrote:
>
> Hi, I have been very interested in loop fusion recently. I found that
> both LLVM and icc have implemented this optimization. I also noticed
> that gcc does not seem to implement it.
>
> I would like to ask if gcc
Hi, I have been very interested in loop fusion recently. I found that
both LLVM and icc have implemented this optimization. I also noticed
that gcc does not seem to implement it.
I would like to ask if gcc have any plans to implement this
optimization? In addition, I also found that there is a
On 04/24/2018 09:18 AM, Richard Biener wrote:
On Mon, Apr 23, 2018 at 8:35 PM, Toon Moene wrote:
On 04/23/2018 01:00 PM, Richard Biener wrote:
Note that while it looks "obvious" in the above source fragment the IL
that is presented to optimizers may make things a lot less "low-hanging".
>>> compiler):
>>>
>>> 'And multiple consecutive array statements with the same shape are
>>> “fused”
>>> exactly so that the compiler can generate good cache use. This sort of
>>> optimization is pretty low hanging fruit.'
>>&g
tion from one of the members of the committee (about "their"
compiler):
'And multiple consecutive array statements with the same shape are “fused”
exactly so that the compiler can generate good cache use. This sort of
optimization is pretty low hanging fruit.'
As far as I can see l
LEVG) = PGFL(1:NPROMA,1:NFLEVG,YS%MP)
>> >>
>> >> The reaction from one of the members of the committee (about "their"
>> >> compiler):
>> >>
>> >> 'And multiple consecutive array statements with the same shape are
>> >
ot;
> >> compiler):
> >>
> >> 'And multiple consecutive array statements with the same shape are
> “fused”
> >> exactly so that the compiler can generate good cache use. This sort of
> >> optimization is pretty low hanging fruit.'
> >&
y so that the compiler can generate good cache use. This sort of
>> optimization is pretty low hanging fruit.'
>>
>> As far as I can see loop fusion as a stand-alone optimization is not
>> supported as yet, although some mention is made in the context of graphite.
>>
rom one of the members of the committee (about "their"
> compiler):
>
> 'And multiple consecutive array statements with the same shape are “fused”
> exactly so that the compiler can generate good cache use. This sort of
> optimization is pretty low hanging fruit.'
rom one of the members of the committee (about "their"
> compiler):
>
> 'And multiple consecutive array statements with the same shape are “fused”
> exactly so that the compiler can generate good cache use. This sort of
> optimization is pretty low hanging fruit.'
nsecutive array statements with the same shape are
“fused” exactly so that the compiler can generate good cache use. This
sort of optimization is pretty low hanging fruit.'
As far as I can see loop fusion as a stand-alone optimization is not
supported as yet, although some mention is ma
On August 16, 2015 1:21:22 PM GMT+02:00, Ajit Kumar Agarwal
wrote:
>All:
>
>Loop fusion is an important optimizations that fuses the set of Loops
>if the following condition is valid.
GCC does not implement loop fusion at all.
Richard.
>1) Loops are conformant ( i.e. they have
All:
Loop fusion is an important optimizations that fuses the set of Loops if the
following condition is valid.
1) Loops are conformant ( i.e. they have same iteration count).
2. Loops are control equivalent. The control equivalence of the loops can be
identified with the dominator and post
On Wed, Apr 22, 2015 at 10:05 PM, Toon Moene wrote:
> On 04/22/2015 09:10 PM, Steven Bosscher wrote:
>
>> On Wed, Apr 22, 2015 at 6:59 PM, Toon Moene wrote:
>
>
>>> Why is loop fusion important, especially in Fortran 90 and later programs
>>> ?
>>>
&
On 04/22/2015 09:10 PM, Steven Bosscher wrote:
On Wed, Apr 22, 2015 at 6:59 PM, Toon Moene wrote:
Why is loop fusion important, especially in Fortran 90 and later programs ?
Because without it, every array assignment is a single loop nest, isolated
from related, same-shape assignments
On Wed, Apr 22, 2015 at 6:59 PM, Toon Moene wrote:
> Why is loop fusion important, especially in Fortran 90 and later programs ?
>
> Because without it, every array assignment is a single loop nest, isolated
> from related, same-shape assignments.
Why is this a bad thing? When yo
/computing/our-facilities).
In the context of GCC this is relevant, because one of the three
compilers is gfortran (version 4.9.2).
One of his slides discussed the differences in optimizations that the
three compilers offer; I was surprised to learn that GCC/gfortran
doesn't do loop f
Sebastian Pop wrote:
Hi Toon,
On Mon, Nov 30, 2009 at 13:51, Toon Moene wrote:
Sebastian,
Obviously, this could benefit from loop fusion, by combining the four
assignments in one loop.
Is that on the horizon ?
Yes, that's for gcc-4.6.
Thanks !
--
Toon Moene - e-ma
Hi Toon,
On Mon, Nov 30, 2009 at 13:51, Toon Moene wrote:
> Sebastian,
>
> I saw you updated the Graphite Wiki page, and I wondered if there are any
> concrete plans on loop fusion.
>
> The reason I ask is because in Fortran (as of Fortran 90) one often sees
> assig
)
where the four assignments all have the structure of loops like:
DO I = 1, N
DO J = 1, N
X(J,I) = OP(A(J,I), B(J,I))
ENDDO
ENDDO
Obviously, this could benefit from loop fusion, by combining the four
assignments in one loop.
Provided that it were still possible to vectorize suitable
Sebastian,
I saw you updated the Graphite Wiki page, and I wondered if there are
any concrete plans on loop fusion.
The reason I ask is because in Fortran (as of Fortran 90) one often sees
assignments like:
REAL, ALLOCATABLE :: A(:,:), B(:,:), C(:,:), D(:,:), E(:,:), F(:,:)
! ... READ IN
21 matches
Mail list logo
