Re: re: toolchain/45673: gcc on i386 current generates bad code for assembler code in lock.h

[David Laight <david%l8s.co.uk@localhost>]

The following reply was made to PR toolchain/45673; it has been noted by GNATS.

From: David Laight <david%l8s.co.uk@localhost>
To: Kai-Uwe Eckhardt <kuehro%gmx.de@localhost>
Cc: gnats-bugs%NetBSD.org@localhost
Subject: Re: re: toolchain/45673: gcc on i386 current generates bad code for 
assembler code in lock.h
Date: Tue, 13 Mar 2012 19:03:31 +0000

 On Tue, Mar 13, 2012 at 04:44:45PM +0100, Kai-Uwe Eckhardt wrote:
 >   
 > >  fortunately, the real bug seems to be the register constraints.
 > 
 > I have tested building lang/gcc46 with qQ in lock.h on 6.0_BETA i386
 > as suggested by J?rg Sonnenberger and it built fine, with no regressions
 > in the gcc testsuite compared to using q alone. 
 
 That just avoid the bug - it really shouldn't happen.
 
        David
 
 -- 
 David Laight: david%l8s.co.uk@localhost