This is the mail archive of the gcc-regression@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]

Re: 2 GCC regressions, 0 new, with your patch.

To: rth at cygnus dot com
Subject: Re: 2 GCC regressions, 0 new, with your patch.
From: Geoff Keating <geoffk at cygnus dot com>
Date: Mon, 31 Jul 2000 15:30:51 -0700
CC: gcc-regression at gcc dot gnu dot org
References: <200007311914.MAA01751@sloth.cygnus.com> <200007312020.NAA10279@localhost.cygnus.com> <20000731151111.C2351@cygnus.com>

> Date: Mon, 31 Jul 2000 15:11:11 -0700
> From: Richard Henderson <rth@cygnus.com>
> Cc: gcc-regression@gcc.gnu.org
> 
> On Mon, Jul 31, 2000 at 01:20:31PM -0700, Geoff Keating wrote:
> > It's not clear from the description of the relevant data structures
> > what it means to have NULL in the global_live_at_start variable.  Is
> > the global_live_at_start stuff valid at all at this point?
> 
> The global_live_at_start stuff is valid during SSA, and then
> not again until after flow1.  But you say this is happening
> in the cse2 invocation?

Yes.  Perhaps clear_log_links should also clear 
global_live_at_{start,end}?

-- 
- Geoffrey Keating <geoffk@cygnus.com>

Follow-Ups:
- Re: 2 GCC regressions, 0 new, with your patch.
  - From: Richard Henderson

References:
- 2 GCC regressions, 0 new, with your patch.
  - From: GCC regression checker
- Re: 2 GCC regressions, 0 new, with your patch.
  - From: Geoff Keating
- Re: 2 GCC regressions, 0 new, with your patch.
  - From: Richard Henderson

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]