diff mbox series

[2/2] OpenMP: Duplicate checking for map clauses in Fortran (PR107214)

Message ID 20221207191355.2e43ea14@squid.athome
State New
Headers show
Series [1/2] OpenMP/Fortran: Combined directives with map/firstprivate of same symbol | expand

Commit Message

Julian Brown Dec. 7, 2022, 7:13 p.m. UTC
> Hi Julian,
> 
> I had a first quick lock at this patch, I should have a closer look
> later. However, I stumbled over the following:
> 
> On 20.10.22 18:14, Julian Brown wrote:
> > typedef struct gfc_symbol
> > {
> > ...
> >    struct gfc_symbol *old_symbol;
> >
> >    unsigned mark:1, comp_mark:1, data_mark:1, dev_mark:1,
> > gen_mark:1; unsigned reduc_mark:1, gfc_new:1;
> >
> >    struct gfc_symbol *tlink;
> >
> >    unsigned equiv_built:1;
> >    ...  
> I know that this was the case before, but can you move the mark:1 etc.
> after 'tlink'? In that case all bitfields are grouped together. If I
> have not miscounted, we have currently 7 bits before and 9 bits after
> 'tlink' and grouping them together reduced pointless padding.
> 
> * * *
> > +      else if (n->sym->mark)
> > +     gfc_error ("Symbol %qs present on both data and map clauses "
> > +                "at %L", n->sym->name, &n->where);  
> 
> I wonder whether that also rejects the following – which seems to be
> valid. The 'map' goes to 'target' and the 'firstprivate' to
> 'parallel', cf. OpenMP 5.2, "17.2 Clauses on Combined and Composite
> Constructs", [340:3-4 & 12-14]. (BTW: While some fixes went into 5.1
> regarding this section, a likewise wording is already in 5.0.)
> 
> (Testing showed: it give an ICE without the patch and an error with.)

...and this patch avoids the error for combined directives, and
reorders the gfc_symbol bitfields.

--

This patch adds duplicate checking for OpenMP "map" clauses, taking some
cues from the implementation for C in c-typeck.cc:c_finish_omp_clauses
(and similar for C++).

In addition to the existing use of the "mark" and "comp_mark" bitfields
in the gfc_symbol structure, the patch adds several new bits handling
duplicate checking within various categories of clause types.  If "mark"
is being used for map clauses, we need to use different bits for other
clauses for cases where "map" and some other clause can refer to the
same symbol (e.g. "map(n) shared(n)").

This version of the patch avoids flagging variables that are listed on
both map and firstprivate clauses when they are on a combined directive,
as they get moved to separate nested directives later (see previous
patch in series).

Tested with offloading to NVPTX alongside previous patch (and
dependencies).  OK?

2022-12-06  Julian Brown  <julian@codesourcery.com>

gcc/fortran/
        PR fortran/107214
        * gfortran.h (gfc_symbol): Add data_mark, dev_mark, gen_mark and
        reduc_mark bitfields.
        * openmp.cc (resolve_omp_clauses): Use above bitfields to improve
        duplicate clause detection.

gcc/testsuite/
        PR fortran/107214
        * gfortran.dg/gomp/pr107214.f90: New test.

Comments

Tobias Burnus Dec. 8, 2022, 12:04 p.m. UTC | #1
Hi Julian:

On 07.12.22 20:13, Julian Brown wrote:
>> I know that this was the case before, but can you move the mark:1 etc.
>> after 'tlink'? In that case all bitfields are grouped together.
Thanks for doing so.
>> I wonder whether that also rejects the following – which seems to be
>> valid. The 'map' goes to 'target' and the 'firstprivate' to
>> 'parallel', cf. OpenMP 5.2, "17.2 Clauses on Combined and Composite
>> Constructs", [340:3-4 & 12-14]. (BTW: While some fixes went into 5.1
>> regarding this section, a likewise wording is already in 5.0.)
>>
>> (Testing showed: it give an ICE without the patch and an error with.)
> ...and this patch avoids the error for combined directives, and
> reorders the gfc_symbol bitfields.

All in all, I am fine with the patch - but I spotted some issues.

First, I think you need to set for some error cases mark = 0 to avoid duplicated errors.
Namely:

   ! Outputs the error twice ('Symbol ‘y’ present on multiple clauses')
   !$omp target has_device_addr(y) firstprivate(y)
   block; end block

  * * *

Additionally, I think it would be good to have besides 'target' + map/firstprivate (→ error)
also a testcase for 'target simd' + map/firstprivate → error

And I think also gives-no-error checks all combined 'target ...' that take firstprivate
should be added, cf. your own patch - possibly with looking at the original dump (scan-tree-dump)
to see that the clause is properly attached correctly. Example for 'target teams':

   !$omp target teams map(x) firstprivate(x)
   block; end block

(Works but no testcase.)

  * * *

The following is not diagnosed and gives an ICE:

!$omp target in_reduction(+: x) private(x)
   block; end block
end

The C testcase properly has:
   error: ‘x’ appears more than once in data-sharing clauses

Note: Using 'firstprivate' instead of 'private' shows the proper error also in Fortran.


The following does not ICE but does not make sense (and is rejected in C):

     4 | #pragma omp target private(x) map(x)

vs.

   !$omp target map(x) private(x)
   block; end block

(The latter produces "#pragma omp target private(x.0) map(tofrom:*x.0)", ups!)

  * * *

I also note that 'simd' accepts private such that

#pragma omp target simd private(x) map(x)
  for (int i=0; i < 0; i++)
  ;

!$omp target simd map(x) private(x)
do i = 1, 0; end do

is valid. (It is accepted by gcc and gfortran, i.e. it just needs to be added as testcase.)

  * * *

I note that C rejects {map(x),firstprivate(x)} + {has_device_addr(x),is_device_ptr(x)}',
but gfortran + your patch accepts:

   !$omp target map(x) has_device_addr(x)
   !$omp target map(x) is_device_ptr(x)

while

   !$omp target firstprivate(x) has_device_addr(x)
   !$omp target firstprivate(x) is_device_ptr(x)

is rejected – showing the error message twice.

Expected: I think it should show an error in all four cases - but only once.

> 2022-12-06  Julian Brown  <julian@codesourcery.com>
>
> gcc/fortran/
>          PR fortran/107214
>          * gfortran.h (gfc_symbol): Add data_mark, dev_mark, gen_mark and
>          reduc_mark bitfields.
>          * openmp.cc (resolve_omp_clauses): Use above bitfields to improve
>          duplicate clause detection.
>
> gcc/testsuite/
>          PR fortran/107214
>          * gfortran.dg/gomp/pr107214.f90: New test.

Thanks,

Tobias

-----------------
Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht München, HRB 106955
Julian Brown Dec. 10, 2022, 12:10 p.m. UTC | #2
On Thu, 8 Dec 2022 13:04:20 +0100
Tobias Burnus <tobias@codesourcery.com> wrote:

> All in all, I am fine with the patch - but I spotted some issues.
> 
> First, I think you need to set for some error cases mark = 0 to avoid
> duplicated errors. Namely:
> 
>    ! Outputs the error twice ('Symbol ‘y’ present on multiple
> clauses') !$omp target has_device_addr(y) firstprivate(y)
>    block; end block
> 
>   * * *
> 
> Additionally, I think it would be good to have besides 'target' +
> map/firstprivate (→ error) also a testcase for 'target simd' +
> map/firstprivate → error
> 
> And I think also gives-no-error checks all combined 'target ...' that
> take firstprivate should be added, cf. your own patch - possibly with
> looking at the original dump (scan-tree-dump) to see that the clause
> is properly attached correctly. Example for 'target teams':
> 
>    !$omp target teams map(x) firstprivate(x)
>    block; end block
> 
> (Works but no testcase.)
> 
>   * * *
> 
> The following is not diagnosed and gives an ICE:
> 
> !$omp target in_reduction(+: x) private(x)
>    block; end block
> end
> 
> The C testcase properly has:
>    error: ‘x’ appears more than once in data-sharing clauses
> 
> Note: Using 'firstprivate' instead of 'private' shows the proper
> error also in Fortran.
> 
> 
> The following does not ICE but does not make sense (and is rejected
> in C):
> 
>      4 | #pragma omp target private(x) map(x)
> 
> vs.
> 
>    !$omp target map(x) private(x)
>    block; end block
> 
> (The latter produces "#pragma omp target private(x.0)
> map(tofrom:*x.0)", ups!)
> 
>   * * *
> 
> I also note that 'simd' accepts private such that
> 
> #pragma omp target simd private(x) map(x)
>   for (int i=0; i < 0; i++)
>   ;
> 
> !$omp target simd map(x) private(x)
> do i = 1, 0; end do
> 
> is valid. (It is accepted by gcc and gfortran, i.e. it just needs to
> be added as testcase.)
> 
>   * * *
> 
> I note that C rejects {map(x),firstprivate(x)} +
> {has_device_addr(x),is_device_ptr(x)}', but gfortran + your patch
> accepts:
> 
>    !$omp target map(x) has_device_addr(x)
>    !$omp target map(x) is_device_ptr(x)
> 
> while
> 
>    !$omp target firstprivate(x) has_device_addr(x)
>    !$omp target firstprivate(x) is_device_ptr(x)
> 
> is rejected – showing the error message twice.
> 
> Expected: I think it should show an error in all four cases - but
> only once.

I believe this patch covers all the above cases (hopefully
appropriately generalised), at least for Fortran. I haven't attempted
to fix any missing cases for C, for now.

Re-tested with offloading to NVPTX (with a few supporting patches, as
before).

Does this look OK now?

Thanks,

Julian
Tobias Burnus Dec. 10, 2022, 12:48 p.m. UTC | #3
Hi Julian,

On 10.12.22 13:10, Julian Brown wrote:
> On Thu, 8 Dec 2022 13:04:20 +0100
> Tobias Burnus <tobias@codesourcery.com> wrote:
>> All in all, I am fine with the patch - but I spotted some issues.
...
> I believe this patch covers all the above cases (hopefully
> appropriately generalised), at least for Fortran. I haven't attempted
> to fix any missing cases for C, for now.
>
> Re-tested with offloading to NVPTX (with a few supporting patches, as
> before).
>
> Does this look OK now?

Yes, LGTM.

Thanks!

Tobias

-----------------
Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht München, HRB 106955
diff mbox series

Patch

From fa6d1e273449aff61833064027fed3787c13121f Mon Sep 17 00:00:00 2001
Message-Id: <fa6d1e273449aff61833064027fed3787c13121f.1670438768.git.julian@codesourcery.com>
In-Reply-To: <c66db363066913ae4939f2aa706427338b109d71.1670438768.git.julian@codesourcery.com>
References: <c66db363066913ae4939f2aa706427338b109d71.1670438768.git.julian@codesourcery.com>
From: Julian Brown <julian@codesourcery.com>
Date: Tue, 6 Dec 2022 23:10:58 +0000
Subject: [PATCH 2/2] OpenMP: Duplicate checking for map clauses in Fortran
 (PR107214)

This patch adds duplicate checking for OpenMP "map" clauses, taking some
cues from the implementation for C in c-typeck.cc:c_finish_omp_clauses
(and similar for C++).

In addition to the existing use of the "mark" and "comp_mark" bitfields
in the gfc_symbol structure, the patch adds several new bits handling
duplicate checking within various categories of clause types.  If "mark"
is being used for map clauses, we need to use different bits for other
clauses for cases where "map" and some other clause can refer to the
same symbol (e.g. "map(n) shared(n)").

This version of the patch avoids flagging variables that are listed on
both map and firstprivate clauses when they are on a combined directive,
as they get moved to separate nested directives later (see previous
patch in series).

Tested with offloading to NVPTX alongside previous patch (and
dependencies).  OK?

2022-12-06  Julian Brown  <julian@codesourcery.com>

gcc/fortran/
	PR fortran/107214
	* gfortran.h (gfc_symbol): Add data_mark, dev_mark, gen_mark and
	reduc_mark bitfields.
	* openmp.cc (resolve_omp_clauses): Use above bitfields to improve
	duplicate clause detection.

gcc/testsuite/
	PR fortran/107214
	* gfortran.dg/gomp/pr107214.f90: New test.
---
 gcc/fortran/gfortran.h                      | 32 ++++++---
 gcc/fortran/openmp.cc                       | 73 +++++++++++++++++----
 gcc/testsuite/gfortran.dg/gomp/pr107214.f90 |  7 ++
 3 files changed, 90 insertions(+), 22 deletions(-)
 create mode 100644 gcc/testsuite/gfortran.dg/gomp/pr107214.f90

diff --git a/gcc/fortran/gfortran.h b/gcc/fortran/gfortran.h
index df90ed39bea7..47a7f5552385 100644
--- a/gcc/fortran/gfortran.h
+++ b/gcc/fortran/gfortran.h
@@ -1871,16 +1871,6 @@  typedef struct gfc_symbol
 
   gfc_namelist *namelist, *namelist_tail;
 
-  /* Change management fields.  Symbols that might be modified by the
-     current statement have the mark member nonzero.  Of these symbols,
-     symbols with old_symbol equal to NULL are symbols created within
-     the current statement.  Otherwise, old_symbol points to a copy of
-     the old symbol. gfc_new is used in symbol.cc to flag new symbols.
-     comp_mark is used to indicate variables which have component accesses
-     in OpenMP/OpenACC directive clauses.  */
-  struct gfc_symbol *old_symbol;
-  unsigned mark:1, comp_mark:1, gfc_new:1;
-
   /* The tlink field is used in the front end to carry the module
      declaration of separate module procedures so that the characteristics
      can be compared with the corresponding declaration in a submodule. In
@@ -1888,6 +1878,28 @@  typedef struct gfc_symbol
      deferred initialization.  */
   struct gfc_symbol *tlink;
 
+  /* Change management fields.  Symbols that might be modified by the
+     current statement have the mark member nonzero.  Of these symbols,
+     symbols with old_symbol equal to NULL are symbols created within
+     the current statement.  Otherwise, old_symbol points to a copy of
+     the old symbol. gfc_new is used in symbol.cc to flag new symbols.
+     comp_mark is used to indicate variables which have component accesses
+     in OpenMP/OpenACC directive clauses (cf. c-typeck.cc:c_finish_omp_clauses,
+     map_field_head).
+     data_mark is used to check duplicate mappings for OpenMP data-sharing
+     clauses (see firstprivate_head/lastprivate_head in the above function).
+     dev_mark is used to check duplicate mappings for OpenMP
+     is_device_ptr/has_device_addr clauses (see is_on_device_head in above
+     function).
+     gen_mark is used to check duplicate mappings for OpenMP
+     use_device_ptr/use_device_addr/private/shared clauses (see generic_head in
+     above functon).
+     reduc_mark is used to check duplicate mappings for OpenMP reduction
+     clauses.  */
+  struct gfc_symbol *old_symbol;
+  unsigned mark:1, comp_mark:1, data_mark:1, dev_mark:1, gen_mark:1;
+  unsigned reduc_mark:1, gfc_new:1;
+
   /* Nonzero if all equivalences associated with this symbol have been
      processed.  */
   unsigned equiv_built:1;
diff --git a/gcc/fortran/openmp.cc b/gcc/fortran/openmp.cc
index 653c43f79ffb..63a14daa6d7b 100644
--- a/gcc/fortran/openmp.cc
+++ b/gcc/fortran/openmp.cc
@@ -7135,6 +7135,10 @@  resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses,
 	  continue;
 	n->sym->mark = 0;
 	n->sym->comp_mark = 0;
+	n->sym->data_mark = 0;
+	n->sym->dev_mark = 0;
+	n->sym->gen_mark = 0;
+	n->sym->reduc_mark = 0;
 	if (n->sym->attr.flavor == FL_VARIABLE
 	    || n->sym->attr.proc_pointer
 	    || (!code && (!n->sym->attr.dummy || n->sym->ns != ns)))
@@ -7203,7 +7207,6 @@  resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses,
 	&& list != OMP_LIST_LASTPRIVATE
 	&& list != OMP_LIST_ALIGNED
 	&& list != OMP_LIST_DEPEND
-	&& (list != OMP_LIST_MAP || openacc)
 	&& list != OMP_LIST_FROM
 	&& list != OMP_LIST_TO
 	&& (list != OMP_LIST_REDUCTION || !openacc)
@@ -7222,10 +7225,43 @@  resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses,
 	    for (gfc_ref *ref = n->expr->ref; ref; ref = ref->next)
 	      if (ref->type == REF_COMPONENT)
 		component_ref_p = true;
-	  if ((!component_ref_p && n->sym->comp_mark)
-	      || (component_ref_p && n->sym->mark))
-	    gfc_error ("Symbol %qs has mixed component and non-component "
-		       "accesses at %L", n->sym->name, &n->where);
+	  if ((list == OMP_LIST_IS_DEVICE_PTR
+	       || list == OMP_LIST_HAS_DEVICE_ADDR)
+	      && !component_ref_p)
+	    {
+	      if (n->sym->gen_mark || n->sym->dev_mark || n->sym->reduc_mark)
+		gfc_error ("Symbol %qs present on multiple clauses at %L",
+			   n->sym->name, &n->where);
+	      else
+		n->sym->dev_mark = 1;
+	    }
+	  else if ((list == OMP_LIST_USE_DEVICE_PTR
+		    || list == OMP_LIST_USE_DEVICE_ADDR
+		    || list == OMP_LIST_PRIVATE
+		    || list == OMP_LIST_SHARED)
+		   && !component_ref_p)
+	    {
+	      if (n->sym->gen_mark || n->sym->dev_mark || n->sym->reduc_mark)
+		gfc_error ("Symbol %qs present on multiple clauses at %L",
+			   n->sym->name, &n->where);
+	      else
+		n->sym->gen_mark = 1;
+	    }
+	  else if (list == OMP_LIST_REDUCTION && !component_ref_p)
+	    {
+	      if (n->sym->gen_mark || n->sym->dev_mark || n->sym->reduc_mark)
+		gfc_error ("Symbol %qs present on multiple clauses at %L",
+			   n->sym->name, &n->where);
+	      else
+		n->sym->reduc_mark = 1;
+	    }
+	  else if ((!component_ref_p && n->sym->comp_mark)
+		   || (component_ref_p && n->sym->mark))
+	    {
+	      if (openacc)
+		gfc_error ("Symbol %qs has mixed component and non-component "
+			   "accesses at %L", n->sym->name, &n->where);
+	    }
 	  else if (n->sym->mark)
 	    gfc_error ("Symbol %qs present on multiple clauses at %L",
 		       n->sym->name, &n->where);
@@ -7241,31 +7277,44 @@  resolve_omp_clauses (gfc_code *code, gfc_omp_clauses *omp_clauses,
   gcc_assert (OMP_LIST_LASTPRIVATE == OMP_LIST_FIRSTPRIVATE + 1);
   for (list = OMP_LIST_FIRSTPRIVATE; list <= OMP_LIST_LASTPRIVATE; list++)
     for (n = omp_clauses->lists[list]; n; n = n->next)
-      if (n->sym->mark)
+      if (n->sym->data_mark || n->sym->gen_mark || n->sym->dev_mark)
 	{
 	  gfc_error ("Symbol %qs present on multiple clauses at %L",
 		     n->sym->name, &n->where);
-	  n->sym->mark = 0;
+	  n->sym->data_mark = 0;
 	}
+      else if (n->sym->mark
+	       && code->op != EXEC_OMP_TARGET_TEAMS
+	       && code->op != EXEC_OMP_TARGET_TEAMS_DISTRIBUTE
+	       && code->op != EXEC_OMP_TARGET_TEAMS_LOOP
+	       && code->op != EXEC_OMP_TARGET_TEAMS_DISTRIBUTE_SIMD
+	       && code->op != EXEC_OMP_TARGET_TEAMS_DISTRIBUTE_PARALLEL_DO
+	       && code->op != EXEC_OMP_TARGET_PARALLEL
+	       && code->op != EXEC_OMP_TARGET_PARALLEL_DO
+	       && code->op != EXEC_OMP_TARGET_PARALLEL_LOOP
+	       && code->op != EXEC_OMP_TARGET_PARALLEL_DO_SIMD
+	       && code->op != EXEC_OMP_TARGET_TEAMS_DISTRIBUTE_PARALLEL_DO_SIMD)
+	gfc_error ("Symbol %qs present on both data and map clauses "
+		   "at %L", n->sym->name, &n->where);
 
   for (n = omp_clauses->lists[OMP_LIST_FIRSTPRIVATE]; n; n = n->next)
     {
-      if (n->sym->mark)
+      if (n->sym->data_mark || n->sym->gen_mark || n->sym->dev_mark)
 	gfc_error ("Symbol %qs present on multiple clauses at %L",
 		   n->sym->name, &n->where);
       else
-	n->sym->mark = 1;
+	n->sym->data_mark = 1;
     }
   for (n = omp_clauses->lists[OMP_LIST_LASTPRIVATE]; n; n = n->next)
-    n->sym->mark = 0;
+    n->sym->data_mark = 0;
 
   for (n = omp_clauses->lists[OMP_LIST_LASTPRIVATE]; n; n = n->next)
     {
-      if (n->sym->mark)
+      if (n->sym->data_mark || n->sym->gen_mark || n->sym->dev_mark)
 	gfc_error ("Symbol %qs present on multiple clauses at %L",
 		   n->sym->name, &n->where);
       else
-	n->sym->mark = 1;
+	n->sym->data_mark = 1;
     }
 
   for (n = omp_clauses->lists[OMP_LIST_ALIGNED]; n; n = n->next)
diff --git a/gcc/testsuite/gfortran.dg/gomp/pr107214.f90 b/gcc/testsuite/gfortran.dg/gomp/pr107214.f90
new file mode 100644
index 000000000000..25949934e840
--- /dev/null
+++ b/gcc/testsuite/gfortran.dg/gomp/pr107214.f90
@@ -0,0 +1,7 @@ 
+! { dg-do compile }
+
+program p
+   integer, allocatable :: a
+   !$omp target map(tofrom: a, a) ! { dg-error "Symbol 'a' present on multiple clauses" }
+   !$omp end target
+end
-- 
2.29.2