[vorbis-dev] Re: ogg123 1.0rc3 thread issue [FIXED] [#112]

[Christian Weisgerber]

Segher Boessenkool <segher at chello.nl> wrote:

> I checked a fix for this problem into xiph.org cvs.  Only tested
> on OpenBSD/x86 (and regression tested on Linux/ppc), so you might
> want to test it on FreeBSD.

Works fine.  I've updated the {Free,Open}BSD ports.

> On OpenBSD 3.0, we still get bitten by the bug in libc_r where
> signals arriving while a thread that doesn't handle them runs,

If somebody wants to run RC3 on OpenBSD 3.0 or earlier, building
with pth will probably do.  In fact, I'll give this a try on 2.9/i386.
The official port is for -current only, anyway.

> Btw, why do the BSD ports remove -O from the Makefiles?  It's
> actually needed (for libvorbis, at least).

Because it's ports policy.

Note that the patch doesn't remove optimization in general, it only
removes the entirely idiotic fixed argument "-O20", so the ports
will be built with the optimization level passed in at configuration
time.  (Defaults to -O on FreeBSD and -O2 on OpenBSD, FWIW.)

Optimization options are a system/admin preference.  Ports are to
honor these.  There are excellent reasons for this.  Of course the
Xiph crew is free to disagree, but I will continue to fix this *bug*
in the ports.

For instance, on FreeBSD/alpha, any attempt to use a optimization
level higher than 1 will give you this:

cc1: warning: 
***
***     The -O2 flag TRIGGERS KNOWN OPTIMIZER BUGS ON THIS PLATFORM
***

If I want to build really fast Ogg Vorbis code there, I won't use
GCC with "-O20" but rather Compaq's compiler with "-fast".


-- 
Christian "naddy" Weisgerber                          naddy at mips.inka.de

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-dev-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.