[mythtv-users] 0.14 + VIA M1k + UDMA6 + full duplex =prebufferingpause - FIXED

[Scott Rowe]

Aha! Mikhail, that was indeed the issue. So it seems that both the 693 and
695 have this bug. 

Anyone using any other version of the Nehemiah processor? If I can confirm
this, I'll pass it along to the various epia wiki / HOWTO maintainers.
(Assuming it's not widely known / already fixed.)

691 = fine
693 = broke
695 = broke 

-Scott Rowe

> -----Original Message-----
> From: mythtv-users-bounces at mythtv.org 
> [mailto:mythtv-users-bounces at mythtv.org] On Behalf Of Scott Rowe
> Sent: Friday, March 26, 2004 8:51 PM
> To: 'Discussion about mythtv'
> Subject: RE: [mythtv-users] 0.14 + VIA M1k + UDMA6 + full 
> duplex =prebufferingpause
> 
>  
> > -----Original Message-----
> > From: mythtv-users-bounces at mythtv.org 
> > [mailto:mythtv-users-bounces at mythtv.org] On Behalf Of Mikhail D'Vane
> > Sent: Thursday, March 25, 2004 8:34 PM
> > To: mythtv-users at mythtv.org
> > Subject: RE: [mythtv-users] 0.14 + VIA M1k + UDMA6 + full duplex 
> > =prebufferingpause
> > 
> > You probably have a stepping 3 of the Nehemiah processor.  
> > This processor (or perhaps the kernel support for it) does 
> not report 
> > mmx capabilities correctly.
> > 
> > One of the directories in /proc will tell you your CPU stepping.  
> > (Sorry, not in front of any linux machine right now, so I 
> can remember 
> > exact directory - it's name is pretty obvious.)
> > 
> > If you have a stepping 3, you need to hard code the edx 
> return value 
> > in the cputest.c file and recompile mythtv (or maybe just 
> > mythfrontend).  You can find the exact information on what 
> to add to 
> > the cputest.c file if you search the archive for posts by me with 
> > "cputest.c" in them.
> > 
> > (cputest.c is in the myth source code.  If you go to the top level 
> > directory where you unpacked mythtv, just do "find -name 
> cputest.c" to 
> > get the path.)
> > 
> > Hope this is your problem - I puzzled over it for *months* before 
> > stumbling onto the solution (Google is your friend - try "edx 
> > cputest.c epia" and you'll find the info I did.)
> > 
> > Mik
>  
>  
>  
>  Actually, cpuid is 695 (stepping = 5).. So it may not just be
>  693 that has this problem (forgot to mention that I'd looked 
>  at this, but since it was 695, I ignored it.) 
>  
>  [root at mythbox root]# cat /proc/cpuinfo
>  processor       : 0
>  vendor_id       : CentaurHauls
>  cpu family      : 6
>  model           : 9
>  model name      : VIA Nehemiah
>  stepping        : 5
>  cpu MHz         : 999.838
>  fdiv_bug        : no
>  hlt_bug         : no
>  f00f_bug        : no
>  coma_bug        : no
>  fpu             : yes
>  fpu_exception   : yes
>  cpuid level     : 1
>  wp              : yes
>  flags           : fpu de pse tsc msr mtrr pge cmov mmx fxsr sse
>  bogomips        : 1992.29 
>  
>  I'll recompile anyway with the fix in place, and if it does 
> turn out to be the issue, I guess the stepping 5 version also 
> has the same issue as the 3. 
>  
>  Live and learn!
>  
>  Thanks,
>  -Scott 
>  
> 
>