This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/19196] New: Confusing identifiers in diagnostics
- From: "gdr at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 30 Dec 2004 02:05:35 -0000
- Subject: [Bug tree-optimization/19196] New: Confusing identifiers in diagnostics
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
Variables fabricated by tree-ssa passes get used in diagnostics instead
of those written by users in their codes. That can lead to really confusing
diagnostics. I noticed this wen compiling my own applications that I cannot
reduce to manageabe size. The snippet below reproduces the basic problem,
but is too simplistic. I hope a bugmaster would make it more complicated,
still manageable.
Consider
% cat sra-name.C
struct A {
int i;
A() { }
};
int main()
{
A a;
return a.i;
}
Compiling the above with current mainline gives
% g++ -O2 -Wall -c sra-name.C
sra-name.C: In function 'int main()':
sra-name.C:8: warning: 'a$i' is used uninitialized in this function
Notice the "a$i" which should have been "a.i".
In my application, I get weird names like 'SR.20649', which looks definitely
dubious and confusing.
--
Summary: Confusing identifiers in diagnostics
Product: gcc
Version: 4.0.0
Status: UNCONFIRMED
Keywords: diagnostic
Severity: normal
Priority: P2
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: gdr at gcc dot gnu dot org
CC: gcc-bugs at gcc dot gnu dot org
GCC host triplet: native, target-independent
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19196