Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
On Wed, May 17, 2023 at 01:55:00PM +0200, Frederik Harwath wrote: > Thanks for the explanation. But actually doing this would require a > complete rewrite which would almost certainly imply that mainline GCC > would not support the loop transformations for a long time. I don't think it needs complete rewrite, the change to use OMP_UNROLL/OMP_TILE should actually simplify stuff when you already have some other extra construct to handle the clauses if it isn't nested into something else, so I wouldn't expect it needs more than 2-3 hours of work. It is true that doing the transformation on trees rather than high gimple is something different, but again it doesn't require everything to be rewritten and we have code to do code copying both on trees and high and low gimple in tree-inline.cc, so the unrolling can just use different APIs to perform it. I'd still prefer to do it like that, I think it will pay back in maintainance costs. If you don't get to this within say 2 weeks, I'll try to do the conversion myself. Jakub
Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
Hi Jakub,
On 16.05.23 13:00, Jakub Jelinek wrote:
On Tue, May 16, 2023 at 11:45:16AM +0200, Frederik Harwath wrote:
The place where different compilers implement the loop transformations
was discussed in an OpenMP loop transformation meeting last year. Two
compilers (another one and GCC with this patch series) transformed
the loops
in the middle end after the handling of data sharing, one planned to
do so.
Yet another vendor had not yet decided where it will be implemented.
Clang
currently does everything in the front end, but it was mentioned that
this
might change in the future e.g. for code sharing with Flang. Implementing
the loop transformations late could potentially
complicate the implementation of transformations which require
adjustments
of the data sharing clauses, but this is known and consequentially,
no such
When already in the FE we determine how many canonical loops a particular
loop transformation creates, I think the primary changes I'd like to
see is
really have OMP_UNROLL/OMP_TILE GENERIC statements (see below) and
consider
where is the best spot to lower it. I believe for data sharing it is best
done during gimplification before the containing loops are handled, it is
already shared code among all the FEs, I think will make it easier to
handle
data sharing right and gimplification is also where doacross processing is
done. While there is restriction that ordered clause is incompatible with
generated loops from tile construct, there isn't one for unroll (unless
"The ordered clause must not appear on a worksharing-loop directive if
the associated loops
include the generated loops of a tile directive."
means unroll partial implicitly because partial unroll tiles the loop, but
it doesn't say it acts as if it was a tile construct), so we'd have to
handle
#pragma omp for ordered(2)
for (int i = 0; i < 64; i++)
#pragma omp unroll partial(4)
for (int j = 0; j < 64; j++)
{
#pragma omp ordered depend (sink: i - 1, j - 2)
#pragma omp ordered depend (source)
}
and I think handling it after gimplification is going to be increasingly
harder. Of course another possibility is ask lang committee to clarify
unless it has been clarified already in 6.0 (but in TR11 it is not).
I do not really expect that we will have to handle this. Questions
concerning
the correctness of code after applying loop transformations came up several
times since I have been following the design meetings and the result was
always either that nothing will be changed, because the loop transformations
are not expected to ensure the correctness of enclosing directives, or that
the use of the problematic construct in conjunction with loop
transformations
will be forbidden. Concerning the use of "ordered" on transformed loops, the
latter approach was suggested for all transformations, cf. issue #3494
in the
private OpenMP spec repository. I see that you have already asked for
clarification
on unroll. I suppose this could also be fixed after gimplification with
reasonable effort. But let's just wait for the result of that discussion
before we
continue worrying about this.
Also, I think creating temporaries is easier to be done during
gimplification than later.
This has not caused problems with the current approach.
Another option is as you implemented a separate pre-omp-lowering pass,
and another one would be do it in the omplower pass, which has actually
several subpasses internally, do it in the scan phase. Disadvantage of
a completely separate pass is that we have to walk the whole IL again,
while doing it in the scan phase means we avoid that cost. We already
do there similar transformations, scan_omp_simd transforms simd constructs
into if (...) simd else simt and then we process it with normal
scan_omp_for
on what we've created. So, if you insist doing it after gimplification
perhaps for compatibility with other non-LLVM compilers, I'd prefer to
do it there rather than in a completely separate pass.
I see. This would be possible. My current approach is indeed rather
wasteful because the pass is not restricted to functions that actually
use loop transformations. I could add an attribute to such functions
that could be used to avoid the execution of the pass and hence
the gimple walk on functions that do not use transformations.
This is necessary to represent the loop nest that is affected by the
loop transformations by a single OMP_FOR to meet the expectations
of all later OpenMP code transformations. This is also the major
reason why the loop transformations are represented by clauses
instead of representing them as "OMP_UNROLL/OMP_TILE as
GENERIC constructs like OMP_FOR" as you suggest below. Since the
I really don't see why. We try to represent what we see in the source
as OpenMP constructs as those constructs. We already have a precedent
with composite loop constructs, where for the combined constructs which
aren't innermost we temporarily use NULL
OMP_FOR_{INIT,COND,INCR,ORIG_DECL
Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
On Tue, May 16, 2023 at 11:45:16AM +0200, Frederik Harwath wrote:
> The place where different compilers implement the loop transformations
> was discussed in an OpenMP loop transformation meeting last year. Two
> compilers (another one and GCC with this patch series) transformed the loops
> in the middle end after the handling of data sharing, one planned to do so.
> Yet another vendor had not yet decided where it will be implemented. Clang
> currently does everything in the front end, but it was mentioned that this
> might change in the future e.g. for code sharing with Flang. Implementing
> the loop transformations late could potentially
> complicate the implementation of transformations which require adjustments
> of the data sharing clauses, but this is known and consequentially, no such
When already in the FE we determine how many canonical loops a particular
loop transformation creates, I think the primary changes I'd like to see is
really have OMP_UNROLL/OMP_TILE GENERIC statements (see below) and consider
where is the best spot to lower it. I believe for data sharing it is best
done during gimplification before the containing loops are handled, it is
already shared code among all the FEs, I think will make it easier to handle
data sharing right and gimplification is also where doacross processing is
done. While there is restriction that ordered clause is incompatible with
generated loops from tile construct, there isn't one for unroll (unless
"The ordered clause must not appear on a worksharing-loop directive if the
associated loops
include the generated loops of a tile directive."
means unroll partial implicitly because partial unroll tiles the loop, but
it doesn't say it acts as if it was a tile construct), so we'd have to handle
#pragma omp for ordered(2)
for (int i = 0; i < 64; i++)
#pragma omp unroll partial(4)
for (int j = 0; j < 64; j++)
{
#pragma omp ordered depend (sink: i - 1, j - 2)
#pragma omp ordered depend (source)
}
and I think handling it after gimplification is going to be increasingly
harder. Of course another possibility is ask lang committee to clarify
unless it has been clarified already in 6.0 (but in TR11 it is not).
Also, I think creating temporaries is easier to be done during
gimplification than later.
Another option is as you implemented a separate pre-omp-lowering pass,
and another one would be do it in the omplower pass, which has actually
several subpasses internally, do it in the scan phase. Disadvantage of
a completely separate pass is that we have to walk the whole IL again,
while doing it in the scan phase means we avoid that cost. We already
do there similar transformations, scan_omp_simd transforms simd constructs
into if (...) simd else simt and then we process it with normal scan_omp_for
on what we've created. So, if you insist doing it after gimplification
perhaps for compatibility with other non-LLVM compilers, I'd prefer to
do it there rather than in a completely separate pass.
> transformations are planned for OpenMP 6.0. In particular, the "apply"
> clause therefore only permits loop-transforming constructs to be applied to
> the loops generated from other loop
> transformations in TR11.
>
> > The normal loop constructs (OMP_FOR, OMP_SIMD, OMP_DISTRIBUTE, OMP_LOOP)
> > already need to know given their collapse/ordered how many loops they are
> > actually associated with and the loop transformation constructs can change
> > that.
> > So, I think we need to do the loop transformations in the FEs, that doesn't
> > mean we need to write everything 3 times, once for each frontend.
> > Already now, e.g. various stuff is shared between C and C++ FEs in c-family,
> > though how much can be shared between c-family and Fortran is to be
> > discovered.
> > Or at least partially, to the extent that we compute how many canonical
> > loops the loop transformations result in, what artificial iterators they
> > will use etc., so that during gimplification we can take all that into
> > account and then can do the actual transformations later.
>
> The patches in this patch series already do compute how many canonical
> loop nests result from the loop transformations in the front end.
Good.
> This is necessary to represent the loop nest that is affected by the
> loop transformations by a single OMP_FOR to meet the expectations
> of all later OpenMP code transformations. This is also the major
> reason why the loop transformations are represented by clauses
> instead of representing them as "OMP_UNROLL/OMP_TILE as
> GENERIC constructs like OMP_FOR" as you suggest below. Since the
I really don't see why. We try to represent what we see in the source
as OpenMP constructs as those constructs. We already have a precedent
with composite loop constructs, where for the combined constructs which
aren't innermost we temporarily use NULL OMP_FOR_{INIT,COND,INCR,ORIG_DECLS}
vectors to stand for this will be some loop, but the details for it aren
Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
Hi Jakub,
On 15.05.23 12:19, Jakub Jelinek wrote:
On Fri, Mar 24, 2023 at 04:30:38PM +0100, Frederik Harwath wrote:
this patch series implements the OpenMP 5.1 "unroll" and "tile"
constructs. It includes changes to the C,C++, and Fortran front end
for parsing the new constructs and a new middle-end
"omp_transform_loops" pass which implements the transformations in a
source language agnostic way.
I'm afraid we can't do it this way, at least not completely.
The OpenMP requirements and what is being discussed for further loop
transformations pretty much requires parts of it to be done as soon as possible.
My understanding is that that is where other implementations implement that
too and would also prefer GCC not to be the only implementation that takes
significantly different decision in that case from other implementations
The place where different compilers implement the loop transformations
was discussed in an OpenMP loop transformation meeting last year. Two
compilers (another one and GCC with this patch series) transformed the
loops in the middle end after the handling of data sharing, one planned
to do so. Yet another vendor had not yet decided where it will be
implemented. Clang currently does everything in the front end, but it
was mentioned that this might change in the future e.g. for code sharing
with Flang. Implementing the loop transformations late could potentially
complicate the implementation of transformations which require
adjustments of the data sharing clauses, but this is known and
consequentially, no such transformations are planned for OpenMP 6.0. In
particular, the "apply" clause therefore only permits loop-transforming
constructs to be applied to the loops generated from other loop
transformations in TR11.
The normal loop constructs (OMP_FOR, OMP_SIMD, OMP_DISTRIBUTE, OMP_LOOP)
already need to know given their collapse/ordered how many loops they are
actually associated with and the loop transformation constructs can change
that.
So, I think we need to do the loop transformations in the FEs, that doesn't
mean we need to write everything 3 times, once for each frontend.
Already now, e.g. various stuff is shared between C and C++ FEs in c-family,
though how much can be shared between c-family and Fortran is to be
discovered.
Or at least partially, to the extent that we compute how many canonical
loops the loop transformations result in, what artificial iterators they
will use etc., so that during gimplification we can take all that into
account and then can do the actual transformations later.
The patches in this patch series already do compute how many canonical
loop nests result from the loop transformations in the front end.
This is necessary to represent the loop nest that is affected by the
loop transformations by a single OMP_FOR to meet the expectations
of all later OpenMP code transformations. This is also the major
reason why the loop transformations are represented by clauses
instead of representing them as "OMP_UNROLL/OMP_TILE as
GENERIC constructs like OMP_FOR" as you suggest below. Since the
loop transformations may also appear on inner loops of a collapsed
loop nest (i.e. within the collapsed depth), representing the
transformation by OMP_FOR-like constructs would imply that a collapsed
loop nest would have to be broken apart into single loops. Perhaps this
could be handled somehow, but the collapsed loop nest would have to be
re-assembled to meet the expectations of e.g. gimplification.
The clause representation is also much better suited for the upcoming
OpenMP "apply" clause where the transformations will not appear
as directives in front of actual loops but inside of other clauses.
In fact, the loop transformation clauses in the implementation already
specify the level of a loop nest to which they apply and it could
be possible to re-use this handling for "apply".
My initial reaction also was to implement the loop transformations
as OMP_FOR-like constructs and the patch actually introduces an
OMP_LOOP_TRANS construct which is used to represent loops that
are not going to be associated with another OpenMP directive after
the transformation, e.g.
void foo () {
#pragma omp tile sizes (4, 8, 16)
for (int i = 0; i < 64; ++i)
{
...
}
}
You suggest to implement the loop transformations during gimplification.
I am not sure if gimplification is actually well-suited to implement the
depth-first evaluation of the loop transformations. I also believe that
gimplification already handles too many things which conceptually are
not related to the translation to GIMPLE. Having a separate pass seems
to be the right move to achieve a better separation of concerns. I think
this will be even more important in the future as the size of the loop
transformation implementation keeps growing. As you mention below,
several new constructs are already planned.
For C, I think the lowering of loop transformation constructs or at least
determining w
Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
On Mon, May 15, 2023 at 12:19:00PM +0200, Jakub Jelinek via Gcc-patches wrote:
> For C++ in templates we obviously need to defer that until instantiations,
> the constants in the clauses etc. could be template parameters etc.
Even in C++ the how many canonical loop nest form loops does this
transformation generate can be probably answered during parsing at least
for the 5.1/5.2 loop transformations.
I think we don't really allow
template
void foo ()
{
#pragma omp for collapse(2)
#pragma omp tile sizes(args...)
for (int i = 0; i < 64; i++)
for (int j = 0; j < 64; j++)
for (int k = 0; k < 64; k++)
;
}
there how many arguments sizes clause has would be determined only
after instantiation. Of course, we don't know the exact values...
Jakub
Re: [PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
On Fri, Mar 24, 2023 at 04:30:38PM +0100, Frederik Harwath wrote:
> this patch series implements the OpenMP 5.1 "unroll" and "tile"
> constructs. It includes changes to the C,C++, and Fortran front end
> for parsing the new constructs and a new middle-end
> "omp_transform_loops" pass which implements the transformations in a
> source language agnostic way.
I'm afraid we can't do it this way, at least not completely.
The OpenMP requirements and what is being discussed for further loop
transformations pretty much requires parts of it to be done as soon as possible.
My understanding is that that is where other implementations implement that
too and would also prefer GCC not to be the only implementation that takes
significantly different decision in that case from other implementations
like e.g. in the offloading case (where all other implementations
preprocess/parse etc. source multiple times compared to GCC splitting stuff
only at IPA time; this affects what can be done with metadirectives,
declare variant etc.).
Now, e.g. data sharing is done almost exclusively during gimplification,
the proposed pass is later than that; it needs to be done before the data
sharing. Ditto doacross handling.
The normal loop constructs (OMP_FOR, OMP_SIMD, OMP_DISTRIBUTE, OMP_LOOP)
already need to know given their collapse/ordered how many loops they are
actually associated with and the loop transformation constructs can change
that.
So, I think we need to do the loop transformations in the FEs, that doesn't
mean we need to write everything 3 times, once for each frontend.
Already now, e.g. various stuff is shared between C and C++ FEs in c-family,
though how much can be shared between c-family and Fortran is to be
discovered.
Or at least partially, to the extent that we compute how many canonical
loops the loop transformations result in, what artificial iterators they
will use etc., so that during gimplification we can take all that into
account and then can do the actual transformations later.
For C, I think the lowering of loop transformation constructs or at least
determining what it means can be done right after we actually parse it and
before we finalize the OMP_FOR eetc. that wraps it if any. As discussed last
week at F2F, I think we want to remember in OMP_FOR_ORIG_DECLS the user
iterators on the loop transformation constructs and take it into account
for data sharing purposes.
For C++ in templates we obviously need to defer that until instantiations,
the constants in the clauses etc. could be template parameters etc.
For Fortran during resolving.
> The "unroll" and "tile" directives are
> internally implemented as clauses. This fits the representation of
So perhaps just use OMP_UNROLL/OMP_TILE as GENERIC constructs like
OMP_FOR etc. but with some argument where from the early loop
transformation analysis you can remember the important stuff, whether
does the loop transformation result in a canonical loop nest or not
and in the former case with how many nested loops.
And then handle the actual transformation IMHO best at gimplification
time, find them in the OMP_FOR etc. body if they are nested in there,
let the transformation happen on GENERIC before the containing OMP_FOR
etc. if any is actually finalized and from the transformation remember
the original user decls and what should happen with them for data sharing
(e.g. lastprivate/lastprivate conditional).
>From the slides I saw last week, a lot of other transformations are in the
planning, like loop reversal etc.
And, I think even in OpenMP 5.1 nothing prevents e.g.
#pragma omp for collapse(3) // etc.
#pragma omp tile sizes (4, 2, 2)
#pragma omp tile sizes (4, 8, 16)
for (int i = 0; i < 64; ++i)
for (int j = 0; j < 64; ++j)
for (int k = 0; k < 64; ++k)
body;
where the inner tile takes the i and j loops and makes
for (int i1 = 0; i1 < 64; i1 += 4)
for (int j1 = 0; j1 < 64; j1 += 8)
for (int k1 = 0; k1 < 64; k1 += 16)
for (int i2 = 0; i2 < 4; i2++)
{
int i = i1 + i2;
for (int j2 = 0; j2 < 8; j2++)
{
int j = j1 + j2;
for (int k2 = 0; k2 < 16; k2++)
{
int k = k1 + k2;
body;
}
}
}
out of it with 3 outer loops which have canonical loop form (the rest
doesn't). And then the second tile takes the outermost 3 of those generated
loops and tiles them again, making it into again 3 canonical loop form
loops plus stuff inside of it.
Or one can replace the
#pragma omp for collapse(3) // etc.
with
#pragma omp for
#pragma omp unroll partial(2)
which furthermore unrolls the outermost generated loop from the outer tile
turning it into 1 canonical loop form loop plus stuff in it.
Or of course as you have in your testcases, some loop transformation
constructs could be used on more nested loops, not necessarily before
the outermost one. But still, in all cases you need to know quite early
how m
[PATCH 0/7] openmp: OpenMP 5.1 loop transformation directives
Hi, this patch series implements the OpenMP 5.1 "unroll" and "tile" constructs. It includes changes to the C,C++, and Fortran front end for parsing the new constructs and a new middle-end "omp_transform_loops" pass which implements the transformations in a source language agnostic way. The "unroll" and "tile" directives are internally implemented as clauses. This fits the representation of collapsed loop nests by a single internal gomp_for construct. Loop transformations can be applied to loops at the different levels of such a loop nest and this can be represented well with the clause representation. The transformations can also be applied to loops which are not going to be associated with any OpenMP directive after the transformation. This is represented by a new gomp_for kind. Loops of this kind are lowered in the transformation pass since they are not subject to any further OpenMP-specific processing. The patches are roughly presented in the order of their development: Each construct is implemented in the Fortran front end first including the middle-end additions/changes, followed by a patch that adds the C and C++ front end changes. This initial implementation supports the loop transformation constructs on the outermost loop of a loop nest only. The support for applying the transformations to inner loops is then added in two further patches. The patches have been bootstrapped and tested on x86_64-linux-gnu with both nvptx-none and amdgcn-amdhsa offloading. Best regards, Frederik Frederik Harwath (7): openmp: Add Fortran support for "omp unroll" directive openmp: Add C/C++ support for "omp unroll" directive openacc: Rename OMP_CLAUSE_TILE to OMP_CLAUSE_OACC_TILE openmp: Add Fortran support for "omp tile" openmp: Add C/C++ support for "omp tile" openmp: Add Fortran support for loop transformations on inner loops openmp: Add C/C++ support for loop transformations on inner loops gcc/Makefile.in |1 + gcc/c-family/c-gimplify.cc|1 + gcc/c-family/c-omp.cc | 12 +- gcc/c-family/c-pragma.cc |2 + gcc/c-family/c-pragma.h |7 +- gcc/c/c-parser.cc | 403 +++- gcc/c/c-typeck.cc | 10 +- gcc/cp/cp-gimplify.cc |3 + gcc/cp/parser.cc | 453 - gcc/cp/pt.cc | 15 +- gcc/cp/semantics.cc | 104 +- gcc/fortran/dump-parse-tree.cc| 30 + gcc/fortran/gfortran.h| 12 +- gcc/fortran/match.h |2 + gcc/fortran/openmp.cc | 460 - gcc/fortran/parse.cc | 52 +- gcc/fortran/resolve.cc|6 + gcc/fortran/st.cc |2 + gcc/fortran/trans-openmp.cc | 187 +- gcc/fortran/trans.cc |2 + gcc/gimple-pretty-print.cc|6 + gcc/gimple.h |1 + gcc/gimplify.cc | 79 +- gcc/omp-general.cc| 22 +- gcc/omp-general.h |1 + gcc/omp-low.cc|6 +- gcc/omp-transform-loops.cc| 1773 + gcc/params.opt|9 + gcc/passes.def|1 + .../loop-transforms/imperfect-loop-nest.c | 12 + .../gomp/loop-transforms/tile-1.c | 164 ++ .../gomp/loop-transforms/tile-2.c | 183 ++ .../gomp/loop-transforms/tile-3.c | 117 ++ .../gomp/loop-transforms/tile-4.c | 322 +++ .../gomp/loop-transforms/tile-5.c | 150 ++ .../gomp/loop-transforms/tile-6.c | 34 + .../gomp/loop-transforms/tile-7.c | 31 + .../gomp/loop-transforms/tile-8.c | 40 + .../gomp/loop-transforms/unroll-1.c | 133 ++ .../gomp/loop-transforms/unroll-2.c | 95 + .../gomp/loop-transforms/unroll-3.c | 18 + .../gomp/loop-transforms/unroll-4.c | 19 + .../gomp/loop-transforms/unroll-5.c | 19 + .../gomp/loop-transforms/unroll-6.c | 20 + .../gomp/loop-transforms/unroll-7.c | 144 ++ .../gomp/loop-transforms/unroll-inner-1.c | 15 + .../gomp/loop-transforms/unroll-inner-2.c | 31 + .../gomp/loop-transforms/unroll-non-rect-1.c | 37 + .../gomp/loop-transforms/unroll-non-rect-2.c | 22 + .../gomp/loop-transforms/unroll-simd-1.c | 84 + .../g++.dg/gomp/loop-transforms/tile-1.h | 27 + .../g++.dg/gomp/loop-transforms/tile-1a.C | 27 + .../g++.dg/gomp/loop-transforms/tile-1b.C | 27 + .../g++.dg/gomp/loop-transforms/unroll-1.C
