Re: tux+php

[Alex Antropoff <alex tirastel md>]

On Mon, Mar 04, 2002 at 11:43:29PM +0200, Valkanas Nikos wrote:
> Sasha is correct. The code is right, the patch is unnecessary. Alex, I have not been able to replicate your problem using php 4.1.1 and tux 2.2.3 using O4 optimization on both.
Ok, I have RH 7.2
gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-98)
glibc-2.2.4-19.3
tux-2.2.3
CFLAGS="-march=pentium -pipe -O"
I really have the problem. So ? :-) 
> 
> Nikos
> 
> > -----Original Message-----
> > From: Sascha Schumann [mailto:sascha@schumann.cx]
> > Sent: Monday, March 04, 2002 2:45 PM
> > To: tux-list@redhat.com
> > Subject: Re: tux+php
> > 
> > 
> > > I spend some time to make clean test standalone program,
> > > and found what problem is in incorrect(or correct :-) work
> > > with struct *io_vec array, I add
> > > zeroing of vec array at line 108 of php_tux.c( simple memset)
> > > after allocating of vec, that solved my problem.
> > > Now php_tux.c can be compiled with any optimization options and
> > > works perfectly without any crashes.
> > 
> >     That's interesting, because it apparently demonstrates a bug
> >     in those compilers.  The code, at least, is correct: It
> >     allocates the array, populates all slots (iov_base and
> >     iov_len) and passes it off to writev() at a later point in
> >     time.
> > 
> >     Can you send me the standalone program? Thanks.
> > 
> >     - Sascha                                     Experience IRCG
> >       http://schumann.cx/                http://schumann.cx/ircg
> > 
> > 
> > 
> > _______________________________________________
> > tux-list mailing list
> > tux-list@redhat.com
> > https://listman.redhat.com/mailman/listinfo/tux-list
> > 
> 
> 
> 
> _______________________________________________
> tux-list mailing list
> tux-list@redhat.com
> https://listman.redhat.com/mailman/listinfo/tux-list

-- 
Regards,
Alex Antropoff