This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug ada/35050] [4.x regression] renames entities not in symbol table
- From: "tero dot koskinen at iki dot fi" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 29 Feb 2008 15:50:17 -0000
- Subject: [Bug ada/35050] [4.x regression] renames entities not in symbol table
- References: <bug-35050-9333@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #1 from tero dot koskinen at iki dot fi 2008-02-29 15:50 -------
I think the problem is in Xref_lib package (xref_lib.adb). It doesn't parse the
current .ali file format properly.
The patch (against gcc-trunk r131693) at
http://iki.fi/tero.koskinen/gcc/gcc-ada-xref_lib_adb.diff fixes parsing for me.
Some details:
The code expects .ali file format for renamings to be:
4V12*My_plus{integer}=4:58 2|11r20
But if I understood the code correctly, the format is:
4V12*My_plus=4:58{integer} 2|11r20
The patch moves =NUMBER:NUMBER handling before {integer}&co. handling.
gnatxref output after the patch:
Add function
Decl: pb_renaming.ads 5:12
Ref: pb_renaming.adb 11:8
My_plus function
Decl: pb_renaming.ads 4:12
Ref: pb_renaming.adb 11:20
I haven't tested very complex cases, but for random gtkada code the patch
seemed to work.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35050