This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patch] for PR 17531
- From: David Edelsohn <dje at watson dot ibm dot com>
- To: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>, Richard Henderson <rth at redhat dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Mon, 27 Sep 2004 21:53:40 -0400
- Subject: Re: [patch] for PR 17531
- References: <20040928012923.GB8959@atrey.karlin.mff.cuni.cz>
>>>>> Zdenek writes:
> my point is that regardless of what you found out in your debugging,
> my patch also fixes a bug in expand_binop and as such it should be
> considered as well. I am also fairly sure that it is the right
> fix for the PR in question and that your fix is just unnecessarily
> complicated, but I won't argue about it at this moment at night
> (I need more time to study your argumentation about it).
Your patch fixes a bug in expand_binop (and expand_twoval_binop
and prepare_cmp_insn, as mentioned by Paolo), if the those functions are
suppose to accept and be able to handle arguments that need conversion. I
believe that RTH is assuming different semantics for those functions and
is trying to ensure that those functions always receive arguments with the
correct mode.
I would suggest clarifying the assumed semantics to help make the
conversation more productive. If the semantics are not clear, the
functions probably require additional comments. This appears to be an
area of GCC that is ambiguous, or at least leaves too much open to
interpretation.
David