Athlon Bug Stomper Success Reports for 2.4.14

Athlon Bug Stomper Success Reports for 2.4.14

[Maxwell Spangler]

My Athlon Thunderbird has been happily compiling kernels for over 24 hours
straight using 2.4.14 with an Athlon-optimized kernel.  It could not do this
in many past kernels and I believe the code snippet offered in September is
working nicely.  There was a long thread about this back then, but I haven't
seen any Athlon problem reports between that time and now.  For the lurkers on
the list, and the benefit of archives, could we please record a few successes?

If you are using an Athlon with 2.4.14 and you previously had problems but now
have stability with an Athlon optimized, kernel, please respond.  (Just a few
people, please.)

Oh, hardware is Tyan Trinity KT-A (S2390B) with Athlon Thunderbird 1.2Ghz,
512M Crucial PC133 Cas2 RAM, many drives..etc.

My sincere appreciation and thanks to all that were involved in tracking this
down and making the fix possible.

/*
 * Nobody seems to know what this does. Damn.
 *
 * But it does seem to fix some unspecified problem
 * with 'movntq' copies on Athlons.
 *
 * VIA 8363 chipset:
 *  - bit 7 at offset 0x55: Debug (RW)
 */
static void __init pci_fixup_via_athlon_bug(struct pci_dev *d)
{
  u8 v;
  pci_read_config_byte(d, 0x55, &v);
  if (v & 0x80) {
    printk("Trying to stomp on Athlon bug...\n");
    v &= 0x7f; /* clear bit 55.7 */
    pci_write_config_byte(d, 0x55, v);
  }
}

-------------------------------------------------------------------------------
Maxwell Spangler
Program Writer
Greenbelt, Maryland, U.S.A.
Washington D.C. Metropolitan Area

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Dan Hollis]

On Wed, 7 Nov 2001, Maxwell Spangler wrote:
>  * Nobody seems to know what this does. Damn.
>  * But it does seem to fix some unspecified problem
>  * with 'movntq' copies on Athlons.
>  * VIA 8363 chipset:
>  *  - bit 7 at offset 0x55: Debug (RW)

I've given alan a full description of what this does.

-Dan
-- 
[-] Omae no subete no kichi wa ore no mono da. [-]

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Andreas Boman]

I have a abit KT7A board (via kt133a chipset). This board has an "Enhance Chip Performence" option in BIOS, previously I have been able to boot athlon optimized kernels with that option off, and anything but athlon optimized kernels with it on. Now with 2.4.14 I can boot an Athlon optimized kernel with it on. (I wonder what that option _really_ does -guess I'll have to find some benchmark to run now...)


	Andreas

On Wed, 7 Nov 2001 00:54:06 -0500 (EST)
Maxwell Spangler <maxwax@mindspring.com> wrote:

> 
> My Athlon Thunderbird has been happily compiling kernels for over 24 hours
> straight using 2.4.14 with an Athlon-optimized kernel.  It could not do this
> in many past kernels and I believe the code snippet offered in September is
> working nicely.  There was a long thread about this back then, but I haven't
> seen any Athlon problem reports between that time and now.  For the lurkers on
> the list, and the benefit of archives, could we please record a few successes?
> 
> If you are using an Athlon with 2.4.14 and you previously had problems but now
> have stability with an Athlon optimized, kernel, please respond.  (Just a few
> people, please.)
> 
> Oh, hardware is Tyan Trinity KT-A (S2390B) with Athlon Thunderbird 1.2Ghz,
> 512M Crucial PC133 Cas2 RAM, many drives..etc.
> 
> My sincere appreciation and thanks to all that were involved in tracking this
> down and making the fix possible.
> 
> /*
>  * Nobody seems to know what this does. Damn.
>  *
>  * But it does seem to fix some unspecified problem
>  * with 'movntq' copies on Athlons.
>  *
>  * VIA 8363 chipset:
>  *  - bit 7 at offset 0x55: Debug (RW)
>  */
> static void __init pci_fixup_via_athlon_bug(struct pci_dev *d)
> {
>   u8 v;
>   pci_read_config_byte(d, 0x55, &v);
>   if (v & 0x80) {
>     printk("Trying to stomp on Athlon bug...\n");
>     v &= 0x7f; /* clear bit 55.7 */
>     pci_write_config_byte(d, 0x55, v);
>   }
> }
> 
> -------------------------------------------------------------------------------
> Maxwell Spangler
> Program Writer
> Greenbelt, Maryland, U.S.A.
> Washington D.C. Metropolitan Area
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Eric Ortega]

On Wed, 7 Nov 2001, Maxwell Spangler wrote:

> If you are using an Athlon with 2.4.14 and you previously had problems but now
> have stability with an Athlon optimized, kernel, please respond.  (Just a few
> people, please.)

We're running an Athlon 1.4 GHz with an IWill KK266 board which finally 
booted with Athlon optimizations (2.4.14). 

We've also been running processes on it all night without issues.

We're very grateful and happy to have this fairly well resolved, expecially
since we have 11 more of these machines.  :)

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Gerhard Mack]

On Tue, 6 Nov 2001, Dan Hollis wrote:

> On Wed, 7 Nov 2001, Maxwell Spangler wrote:
> >  * Nobody seems to know what this does. Damn.
> >  * But it does seem to fix some unspecified problem
> >  * with 'movntq' copies on Athlons.
> >  * VIA 8363 chipset:
> >  *  - bit 7 at offset 0x55: Debug (RW)
> 
> I've given alan a full description of what this does.

Is there a place the rest of us can find out? I'm very curious to find out
what it was.

	Gerhard

--
Gerhard Mack

gmack@innerfire.net

<>< As a computer I find your faith in technology amusing.

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Lost Logic]

Now by athlon optimizations, you just mean setting "Processor Type" to 
Athlon/Duron/K7, right?  if that is the case, my athlon 1.0ghz worked as 
of kernel 2.4.9 on an Epox 8kta+.

--Brandon

Maxwell Spangler wrote:

>My Athlon Thunderbird has been happily compiling kernels for over 24 hours
>straight using 2.4.14 with an Athlon-optimized kernel.  It could not do this
>in many past kernels and I believe the code snippet offered in September is
>working nicely.  There was a long thread about this back then, but I haven't
>seen any Athlon problem reports between that time and now.  For the lurkers on
>the list, and the benefit of archives, could we please record a few successes?
>
>If you are using an Athlon with 2.4.14 and you previously had problems but now
>have stability with an Athlon optimized, kernel, please respond.  (Just a few
>people, please.)
>
>Oh, hardware is Tyan Trinity KT-A (S2390B) with Athlon Thunderbird 1.2Ghz,
>512M Crucial PC133 Cas2 RAM, many drives..etc.
>
>My sincere appreciation and thanks to all that were involved in tracking this
>down and making the fix possible.
>
>/*
> * Nobody seems to know what this does. Damn.
> *
> * But it does seem to fix some unspecified problem
> * with 'movntq' copies on Athlons.
> *
> * VIA 8363 chipset:
> *  - bit 7 at offset 0x55: Debug (RW)
> */
>static void __init pci_fixup_via_athlon_bug(struct pci_dev *d)
>{
>  u8 v;
>  pci_read_config_byte(d, 0x55, &v);
>  if (v & 0x80) {
>    printk("Trying to stomp on Athlon bug...\n");
>    v &= 0x7f; /* clear bit 55.7 */
>    pci_write_config_byte(d, 0x55, v);
>  }
>}
>
>-------------------------------------------------------------------------------
>Maxwell Spangler
>Program Writer
>Greenbelt, Maryland, U.S.A.
>Washington D.C. Metropolitan Area
>
>-
>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>Please read the FAQ at  http://www.tux.org/lkml/
>
>

Re: Athlon Bug Stomper Success Reports for 2.4.14

[Mikael Johansson]

Hello Maxwell and All!

On Wed, 7 Nov 2001 00:54:06 -0500 (EST) Maxwell Spangler
<maxwax@mindspring.com> wrote:

> If you are using an Athlon with 2.4.14 and you previously had problems but now
> have stability with an Athlon optimized, kernel, please respond.  (Just a few
> people, please.)

Two success stories here, on 2.4.13-ac8 (up to 2.4.10 no success):

1.) Abit KT7A
2.) MSI K7T Turbo

Both machines otherwise identical:
3x512 MB SDRAM
2x40 GB IBM Deskstar, etc.

So thank you very much for the patch!

Have a nice day,
    Mikael J.