public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jakub Jelinek <jakub@redhat.com>
To: gcc-patches@gcc.gnu.org
Subject: [committed] c++, openmp: Handle some OMP_*/OACC_* constructs during constant expression evaluation [PR108607]
Date: Wed, 1 Feb 2023 10:44:54 +0100	[thread overview]
Message-ID: <Y9o0lpdPLeh4JF6U@tucnak> (raw)

Hi!

While potential_constant_expression_1 handled most of OMP_* codes (by saying that
they aren't potential constant expressions), OMP_SCOPE was missing in that list.
I've also added OMP_SCAN, though that is less important (similarly to OMP_SECTION
it ought to appear solely inside of OMP_{FOR,SIMD} resp. OMP_SECTIONS).
As the testcase shows, it isn't enough, potential_constant_expression_1
can catch only some cases, as soon as one uses switch or ifs where at least
one of the possible paths could be constant expression, we can run into the
same codes during cxx_eval_constant_expression, so this patch handles those
there as well.

Bootstrapped/regtested on x86_64-linux and i686-linux, committed to trunk
so far.

2023-02-01  Jakub Jelinek  <jakub@redhat.com>

	PR c++/108607
	* constexpr.cc (cxx_eval_constant_expression): Handle OMP_*
	and OACC_* constructs as non-constant.
	(potential_constant_expression_1): Handle OMP_SCAN and OMP_SCOPE.

	* g++.dg/gomp/pr108607.C: New test.

--- gcc/cp/constexpr.cc.jj	2023-01-16 11:52:16.000000000 +0100
+++ gcc/cp/constexpr.cc	2023-01-31 14:44:36.686618639 +0100
@@ -8001,6 +8001,51 @@ cxx_eval_constant_expression (const cons
       r = cxx_eval_bit_cast (ctx, t, non_constant_p, overflow_p);
       break;
 
+    case OMP_PARALLEL:
+    case OMP_TASK:
+    case OMP_FOR:
+    case OMP_SIMD:
+    case OMP_DISTRIBUTE:
+    case OMP_TASKLOOP:
+    case OMP_LOOP:
+    case OMP_TEAMS:
+    case OMP_TARGET_DATA:
+    case OMP_TARGET:
+    case OMP_SECTIONS:
+    case OMP_ORDERED:
+    case OMP_CRITICAL:
+    case OMP_SINGLE:
+    case OMP_SCAN:
+    case OMP_SCOPE:
+    case OMP_SECTION:
+    case OMP_MASTER:
+    case OMP_MASKED:
+    case OMP_TASKGROUP:
+    case OMP_TARGET_UPDATE:
+    case OMP_TARGET_ENTER_DATA:
+    case OMP_TARGET_EXIT_DATA:
+    case OMP_ATOMIC:
+    case OMP_ATOMIC_READ:
+    case OMP_ATOMIC_CAPTURE_OLD:
+    case OMP_ATOMIC_CAPTURE_NEW:
+    case OMP_DEPOBJ:
+    case OACC_PARALLEL:
+    case OACC_KERNELS:
+    case OACC_SERIAL:
+    case OACC_DATA:
+    case OACC_HOST_DATA:
+    case OACC_LOOP:
+    case OACC_CACHE:
+    case OACC_DECLARE:
+    case OACC_ENTER_DATA:
+    case OACC_EXIT_DATA:
+    case OACC_UPDATE:
+      if (!ctx->quiet)
+	error_at (EXPR_LOCATION (t),
+		  "statement is not a constant expression");
+      *non_constant_p = true;
+      break;
+
     default:
       if (STATEMENT_CODE_P (TREE_CODE (t)))
 	{
@@ -9471,6 +9516,8 @@ potential_constant_expression_1 (tree t,
     case OMP_ORDERED:
     case OMP_CRITICAL:
     case OMP_SINGLE:
+    case OMP_SCAN:
+    case OMP_SCOPE:
     case OMP_SECTION:
     case OMP_MASTER:
     case OMP_MASKED:
--- gcc/testsuite/g++.dg/gomp/pr108607.C.jj	2023-01-31 14:52:11.914936116 +0100
+++ gcc/testsuite/g++.dg/gomp/pr108607.C	2023-01-31 14:52:36.501575191 +0100
@@ -0,0 +1,47 @@
+// PR c++/108607
+// { dg-do compile { target c++14 } }
+// { dg-options "-fopenmp" }
+
+constexpr int
+bar (int x)
+{
+  return x;
+}
+
+constexpr int
+foo (int x)			// { dg-message "declared here" "" { target c++20_down } }
+{				// { dg-message "is not usable as a 'constexpr' function because" "" { target c++23 } .-1 }
+  #pragma omp scope		// { dg-warning "is not a constant expression" "" { target c++20_down } }
+  x = bar (x);			// { dg-error "is not a constant expression" "" { target c++23 } .-1 }
+  return x;
+}
+
+constexpr int
+baz (int x)
+{
+  switch (x)
+    {
+    case 42:
+      return 0;
+    case 2:
+      #pragma omp scope		// { dg-error "statement is not a constant expression" }
+      x = bar (x);
+      return x;
+    case 3:
+      #pragma omp parallel	// { dg-error "statement is not a constant expression" }
+      x = bar (x);
+      return x;
+    case 4:
+      #pragma omp task		// { dg-error "statement is not a constant expression" }
+      x = bar (x);
+      return x;
+    default:
+      return -1;
+    }
+}
+
+constexpr int a = foo (1);	// { dg-error "called in a constant expression" }
+constexpr int b = baz (42);
+constexpr int c = baz (2);
+constexpr int d = baz (3);
+constexpr int e = baz (4);

	Jakub


                 reply	other threads:[~2023-02-01  9:44 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Y9o0lpdPLeh4JF6U@tucnak \
    --to=jakub@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).