This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [patch]: Patch for wince
On Thursday 01 January 2009 18:44:12, Kai Tietz wrote:
> >>>>2009-01-01 Kai Tietz <kai.tietz@onevision.com>
> >>>>
> >>>> * windres.c (set_endianess): Cut name between '-' for search.
> >>>>
> >>Hmm, well alloca would be possible too. But isn't the scope of
> >>validity of stack just within the current frame, and not necessarily
> >>out side of it? This was the reason why I used xmalloc/free here.
> >
> > AFAIK, The stack frame of the caller would be the duration of the
> > function call.
> >
> > I probably missed this in a previous message but can you show the
> > target triple that this is supposed to be adapted to?
> E.g. "pe-arm-wince-little"
>
> I attached corrected patch.
Hmm, are you extracting these from cegcc's svn repository?
The version there does use alloca, but it also adjusted windres.exp
so it actually runs on arm*-*-mingw32ce*:
--- ChangeLog.ce (revisão 1129)
+++ ChangeLog.ce (revisão 1130)
@@ -1,3 +1,11 @@
+2008-01-10 Pedro Alves <pedro_alves@portugalmail.pt>
+
+ binutils/
+ * testsuite/binutils-all/windres/windres.c: Enable testing on
+ *-*-mingw*.
+ * windres.c (set_endianess): Better extract the target's
+ architecture from the bfd target name.
...
Index: binutils/testsuite/binutils-all/windres/windres.exp
===================================================================
--- binutils/testsuite/binutils-all/windres/windres.exp (revisão 1129)
+++ binutils/testsuite/binutils-all/windres/windres.exp (revisão 1130)
@@ -19,7 +19,7 @@
# Written by DJ Delorie <dj@redhat.com>
-if {![istarget "i*86-*-*"] && ![istarget "x86_64-*-mingw*"] } {
+if {![istarget "i*86-*-*"] && ![istarget "*-*-mingw*"] } {
return
}
..
--
Pedro Alves