public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Tobias Burnus <burnus@net-b.de>
Cc: gcc patches <gcc-patches@gcc.gnu.org>, gfortran <fortran@gcc.gnu.org>
Subject: [Patch, Fortran] PR57553 - fix two STORAGE_SIZE bugs
Date: Fri, 07 Jun 2013 16:11:00 -0000	[thread overview]
Message-ID: <51B20641.4040509@net-b.de> (raw)

[-- Attachment #1: Type: text/plain, Size: 356 bytes --]

This patch fixes two issues:
*  storage_size('aa')  was rejected as constant expression - as 
ts.u.cl->length == 0.
* In trans*.c, there was a fold_convert missing (-> ICE). Additionally, 
I have replaced the detour to generate a tree containing the value "8" 
via a fortran expression.

Build and regtested on x86-64-gnu-linux.
OK for the trunk?

Tobias


[-- Attachment #2: storage-size-simplify.diff --]
[-- Type: text/x-patch, Size: 2695 bytes --]

2013-06-07  Tobias Burnus  <burnus@net-b.de>

	PR fortran/57553
	* simplify.c (gfc_simplify_storage_size): Handle literal
	strings.
	* trans-intrinsic.c (gfc_conv_intrinsic_storage_size):
	Add missing fold_convert.

diff --git a/gcc/fortran/simplify.c b/gcc/fortran/simplify.c
index 815043b..683d58b 100644
--- a/gcc/fortran/simplify.c
+++ b/gcc/fortran/simplify.c
@@ -5717,7 +5717,7 @@ gfc_simplify_storage_size (gfc_expr *x,
   if (x->ts.type == BT_CLASS || x->ts.deferred)
     return NULL;
 
-  if (x->ts.type == BT_CHARACTER
+  if (x->ts.type == BT_CHARACTER && x->expr_type != EXPR_CONSTANT
       && (!x->ts.u.cl || !x->ts.u.cl->length
 	  || x->ts.u.cl->length->expr_type != EXPR_CONSTANT))
     return NULL;
diff --git a/gcc/fortran/trans-intrinsic.c b/gcc/fortran/trans-intrinsic.c
index eca907e..3fbf193 100644
--- a/gcc/fortran/trans-intrinsic.c
+++ b/gcc/fortran/trans-intrinsic.c
@@ -5249,12 +5249,10 @@ static void
 gfc_conv_intrinsic_storage_size (gfc_se *se, gfc_expr *expr)
 {
   gfc_expr *arg;
-  gfc_se argse,eight;
+  gfc_se argse;
   tree type, result_type, tmp;
 
   arg = expr->value.function.actual->expr;
-  gfc_init_se (&eight, NULL);
-  gfc_conv_expr (&eight, gfc_get_int_expr (expr->ts.kind, NULL, 8));
   
   gfc_init_se (&argse, NULL);
   result_type = gfc_get_int_type (expr->ts.kind);
@@ -5285,11 +5283,12 @@ gfc_conv_intrinsic_storage_size (gfc_se *se, gfc_expr *expr)
   if (arg->ts.type == BT_CHARACTER)
     tmp = size_of_string_in_bytes (arg->ts.kind, argse.string_length);
   else
-    tmp = fold_convert (result_type, size_in_bytes (type)); 
+    tmp = size_in_bytes (type); 
+  tmp = fold_convert (result_type, tmp);
 
 done:
   se->expr = fold_build2_loc (input_location, MULT_EXPR, result_type, tmp,
-			      eight.expr);
+			      build_int_cst (result_type, BITS_PER_UNIT));
   gfc_add_block_to_block (&se->pre, &argse.pre);
 }
 
--- /dev/null	2013-06-07 09:13:23.024185858 +0200
+++ gcc/gcc/testsuite/gfortran.dg/storage_size_4.f90	2013-06-07 17:34:12.719284544 +0200
@@ -0,0 +1,23 @@
+! { dg-do compile }
+! { dg-options "-fdump-tree-original" }
+!
+! PR fortran/57553
+!
+! Ensure that there is no ICE and that compile-time simplication works.
+!
+  use iso_fortran_env
+  implicit none
+  integer, parameter :: ESize = storage_size('a')
+  integer, parameter :: ESize2 = storage_size('aa')
+  if ( ESize/CHARACTER_STORAGE_SIZE /= 1) call abort()
+  if ( ESize2/CHARACTER_STORAGE_SIZE /= 2) call abort()
+end
+
+subroutine S ( A )
+  character(len=*), intent(in) :: A
+  integer :: ESize = 4
+  esize = ( storage_size(a) + 7 ) / 8
+end
+
+! { dg-final { scan-tree-dump-not "abort" "original" } }
+! { dg-final { cleanup-tree-dump "original" } }


             reply	other threads:[~2013-06-07 16:11 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-07 16:11 Tobias Burnus [this message]
2013-06-08 10:56 ` Mikael Morin

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=51B20641.4040509@net-b.de \
    --to=burnus@net-b.de \
    --cc=fortran@gcc.gnu.org \
    --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).