[Patch, fortran] PR38657 - [4.3/4.4 Regression] PUBLIC/PRIVATE Common blocks

Paul Richard Thomas paul.richard.thomas@gmail.com
Mon Jan 5 18:07:00 GMT 2009


Yet another regression squashed.....

This is verging on obvious.... once the problem is found:-)  The check
during module writing that ensured that a common block did not get
written twice was using the common block name rather than the symtree.
 This is rather akin to renamed use associated symbols, since use
associated common blocks have their names mangled.  The testcase is a
fair elaboration on the original to try to mix private and public
common block variables.

Bootstrapped and regtested on x86_i64/FC9 - OK for trunk and a week
later for 4.3?

Paul

2009-01-05  Paul Thomas  <pault@gcc.gnu.org>

	PR fortran/38657
	* module.c (write_common_0): Use the name of the symtree rather
	than the common block, to determine if the common has been
	written.

2009-01-05  Paul Thomas  <pault@gcc.gnu.org>

	PR fortran/38657
	* gfortran.dg/module_commons_3.f90: New test.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: submit.diff
Type: text/x-patch
Size: 2881 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-patches/attachments/20090105/088607b6/attachment.bin>


More information about the Gcc-patches mailing list