This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] PR17913:[4.0 Regression] ICE jumping into statementexpression
- From: "Joseph S. Myers" <jsm at polyomino dot org dot uk>
- To: Paolo Bonzini <bonzini at gnu dot org>
- Cc: Steven Bosscher <stevenb at suse dot de>, gcc-patches at gcc dot gnu dot org
- Date: Thu, 28 Oct 2004 12:56:42 +0000 (UTC)
- Subject: Re: [PATCH] PR17913:[4.0 Regression] ICE jumping into statementexpression
- References: <2084962.1098880316640.SLOX.WebMail.firstname.lastname@example.org><4180DE14.email@example.com>
On Thu, 28 Oct 2004, Paolo Bonzini wrote:
> Ouch. Think about an extremely simplistic stack-based code generator, and
> feed it
With tree-ssa we should no longer have such problems arising from jumping
into pushing function arguments (etc.).
> From some experiments, gcc manages to cope with this even with very complex
> nested expressions, but making labels local would make a lot of sense. I
> second your opinion and suggest marking as ice-on-invalid, closing as WONTFIX,
> and opening another accepts-invalid PR.
An ICE is a bug, whether the code is valid or invalid. Marking unfixed
bugs that apply to current mainline WONTFIX is not appropriate. (WONTFIX
makes sense for bugs fixed in mainline for which a fix won't be backported
but a separate bug requests a backport, or for bugs in targets or options
that *have been* (not will be) removed. But if program X with options Y
gives an ICE, this is a genuine bug as long as options Y are accepted, and
should not be marked WONTFIX.
As I explained, making labels local leads to other problems and is not a
Joseph S. Myers http://www.srcf.ucam.org/~jsm28/gcc/
firstname.lastname@example.org (personal mail)
email@example.com (CodeSourcery mail)
firstname.lastname@example.org (Bugzilla assignments and CCs)