From: Tobias Burnus <burnus@net-b.de>
To: gfortran g <fortran@gcc.gnu.org>, gcc patches <gcc-patches@gcc.gnu.org>
Subject: [Patch,Fortran] PR 45756 Fix muliple-decl issue with PARAMETER
Date: Tue, 28 Sep 2010 15:52:00 -0000 [thread overview]
Message-ID: <4CA1E78C.1010802@net-b.de> (raw)
[-- Attachment #1: Type: text/plain, Size: 529 bytes --]
For PARAMETERs in MODULEs which are of derived type or array-valued
gfortran creates a static variable. This patch ensures that no extra
declaration is generated for use associated parameters.
If one runs the testcase (see PR) without patch, the result of
-fdump-tree-original-uid is:
iD.1554 = paraD.1556[0];
jD.1559 = paraD.1561[0];
with the patch, the result is:
iD.1513 = paraD.1510[0];
jD.1516 = paraD.1510[0];
Build on x86-64-linux and currently regtesting; if it succeeds:
OK for the trunk?
Tobias
[-- Attachment #2: gsym-parameter.diff --]
[-- Type: text/x-patch, Size: 1629 bytes --]
2010-09-28 Tobias Burnus <burnus@net-b.de>
PR fortran/45756
* trans-decl.c (gfc_get_symbol_decl): Use gsym for decl of
module parameters.
Index: gcc/fortran/trans-decl.c
===================================================================
--- gcc/fortran/trans-decl.c (revision 164682)
+++ gcc/fortran/trans-decl.c (working copy)
@@ -1133,11 +1133,18 @@ gfc_get_symbol_decl (gfc_symbol * sym)
if (sym->backend_decl)
return sym->backend_decl;
+ /* Special case for array-valued named constants from intrinsic
+ procedures; those are inlined. */
+ if (sym->attr.use_assoc && sym->from_intmod
+ && sym->attr.flavor == FL_PARAMETER)
+ intrinsic_array_parameter = true;
+
/* If use associated and whole file compilation, use the module
declaration. */
if (gfc_option.flag_whole_file
- && sym->attr.flavor == FL_VARIABLE
- && sym->attr.use_assoc
+ && (sym->attr.flavor == FL_VARIABLE
+ || sym->attr.flavor == FL_PARAMETER)
+ && sym->attr.use_assoc && !intrinsic_array_parameter
&& sym->module)
{
gfc_gsymbol *gsym;
@@ -1182,12 +1189,6 @@ gfc_get_symbol_decl (gfc_symbol * sym)
if (sym->attr.intrinsic)
internal_error ("intrinsic variable which isn't a procedure");
- /* Special case for array-valued named constants from intrinsic
- procedures; those are inlined. */
- if (sym->attr.use_assoc && sym->from_intmod && sym->attr.dimension
- && sym->attr.flavor == FL_PARAMETER)
- intrinsic_array_parameter = true;
-
/* Create string length decl first so that they can be used in the
type declaration. */
if (sym->ts.type == BT_CHARACTER)
next reply other threads:[~2010-09-28 13:03 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-28 15:52 Tobias Burnus [this message]
2010-09-28 16:30 ` Jerry DeLisle
2010-09-28 16:49 ` Paul Richard Thomas
2010-09-28 16:56 ` Tobias Burnus
2010-09-28 17:11 ` Paul Richard Thomas
2010-09-28 18:24 ` Tobias Burnus
2010-09-28 18:53 ` Paul Richard Thomas
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=4CA1E78C.1010802@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).