This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [tree-ssa] Bootstrap failure on i686-linux
- From: Diego Novillo <dnovillo at redhat dot com>
- To: Jeff Law <law at redhat dot com>
- Cc: Paul Brook <paul at nowt dot org>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>, Richard Henderson <rth at redhat dot com>, Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- Date: Tue, 02 Sep 2003 14:07:13 -0400
- Subject: Re: [tree-ssa] Bootstrap failure on i686-linux
- Organization: Red Hat Canada
- References: <200309021459.h82ExwIX020092@speedy.slc.redhat.com>
On Tue, 2003-09-02 at 10:59, law@redhat.com wrote:
> Attached. Compile with -O2 -pedantic. The problem is the structures
> that the block statement iterators depend on are mucked up for block #3
> after PRE has completed.
>
Cool. Thanks Jeff.
PRE seems to have exposed another bug in the edge insertion routines.
>From what I've seen so far, handle_switch_split is mishandling an
insertion at [ XXX ].
> switch (((re_opcode_t) * p++))
> {
> unconditional_jump:
> ;
> case jump:
[ XXX ]
> do
> {
> (mcnt) = *(p) & 0377;
> }
> while (0);
I should have a fix for this soon-ish.
Diego.