NForce2 + linux 2.6.3-rc2 + acpi,apic patches

NForce2 + linux 2.6.3-rc2 + acpi,apic patches

[Jesse Allen]

Hi,

I have noticed two patches in Andrew Morton's tree, 2.6.3-rc1-mm1, that may be 
helpful for me and other nforce 2 users.  The two patches are
nforce-irq-setup-fix.patch
8259-timer-ack-fix.patch

I have compiled 2.6.3-rc2 with those two patches included.  Here is 
/proc/interrupts after reboot:
  0:    2347129  local-APIC-edge  timer
  1:       6604    IO-APIC-edge  i8042
  2:          0          XT-PIC  cascade
  4:       4458    IO-APIC-edge  serial
  7:          0    IO-APIC-edge  parport0
  8:          1    IO-APIC-edge  rtc
  9:          0   IO-APIC-level  acpi
 14:       4456    IO-APIC-edge  ide0
 15:         17    IO-APIC-edge  ide1
 20:     239364   IO-APIC-level  eth0, ohci_hcd
 21:          0   IO-APIC-level  ehci_hcd, NVidia nForce2
 22:          0   IO-APIC-level  ohci_hcd


Timer is now local-APIC-edge instead of XT-PIC.  Seems to be better now.  I've 
checked to see if the problem with large clock time gain returned.  So far, it 
seems to be perfectly synced with my watch.

I don't know if local-APIC or IO-APIC are really any different with the timer.
If there is something I should know about the difference, someone please let
me know =)

Also, here's a snippet from dmesg:
enabled ExtINT on CPU#0
ESR value before enabling vector: 00000000
ESR value after enabling vector: 00000000
ENABLING IO-APIC IRQs
init IO_APIC IRQs
 IO-APIC (apicid-pin) 2-0, 2-16, 2-17, 2-18, 2-19, 2-20, 2-21, 2-22, 2-23 not connected.
..TIMER: vector=0x31 pin1=2 pin2=-1
..MP-BIOS bug: 8254 timer not connected to IO-APIC
...trying to set up timer (IRQ0) through the 8259A ...  failed.
...trying to set up timer as Virtual Wire IRQ... works.
Using local APIC timer interrupts.
calibrating APIC timer ...
..... CPU clock speed is 1912.0861 MHz.
..... host bus clock speed is 332.0671 MHz.


Thanks Ross and Maciej for the patches!

Jesse

Re: NForce2 + linux 2.6.3-rc2 + acpi,apic patches

[Mark Rutherford]

Jesse,

I saw this thread on the mailing list about your sucess with the nforce2...
I havent had much sucess with it, and was wondering if you have tried -rc3
Also, do you know if Andrew Morton's patches are cumulative?
would he have these patches in rc3-mm2 ?

I have been using my nforce2 machine in XT-PIC ... and its pretty bad.
I have some hardware that doesnt want to share the irq so it gets dicey.
I would like to get it fixed, I have posted, but I didnt get a response.

Thanks!


On Tuesday 10 February 2004 08:29 pm, Jesse Allen wrote:
> Hi,
>
> I have noticed two patches in Andrew Morton's tree, 2.6.3-rc1-mm1, that may
> be helpful for me and other nforce 2 users.  The two patches are
> nforce-irq-setup-fix.patch
> 8259-timer-ack-fix.patch
>
> I have compiled 2.6.3-rc2 with those two patches included.  Here is
> /proc/interrupts after reboot:
>   0:    2347129  local-APIC-edge  timer
>   1:       6604    IO-APIC-edge  i8042
>   2:          0          XT-PIC  cascade
>   4:       4458    IO-APIC-edge  serial
>   7:          0    IO-APIC-edge  parport0
>   8:          1    IO-APIC-edge  rtc
>   9:          0   IO-APIC-level  acpi
>  14:       4456    IO-APIC-edge  ide0
>  15:         17    IO-APIC-edge  ide1
>  20:     239364   IO-APIC-level  eth0, ohci_hcd
>  21:          0   IO-APIC-level  ehci_hcd, NVidia nForce2
>  22:          0   IO-APIC-level  ohci_hcd
>
>
> Timer is now local-APIC-edge instead of XT-PIC.  Seems to be better now. 
> I've checked to see if the problem with large clock time gain returned.  So
> far, it seems to be perfectly synced with my watch.
>
> I don't know if local-APIC or IO-APIC are really any different with the
> timer. If there is something I should know about the difference, someone
> please let me know =)
>
> Also, here's a snippet from dmesg:
> enabled ExtINT on CPU#0
> ESR value before enabling vector: 00000000
> ESR value after enabling vector: 00000000
> ENABLING IO-APIC IRQs
> init IO_APIC IRQs
>  IO-APIC (apicid-pin) 2-0, 2-16, 2-17, 2-18, 2-19, 2-20, 2-21, 2-22, 2-23
> not connected. ..TIMER: vector=0x31 pin1=2 pin2=-1
> ..MP-BIOS bug: 8254 timer not connected to IO-APIC
> ...trying to set up timer (IRQ0) through the 8259A ...  failed.
> ...trying to set up timer as Virtual Wire IRQ... works.
> Using local APIC timer interrupts.
> calibrating APIC timer ...
> ..... CPU clock speed is 1912.0861 MHz.
> ..... host bus clock speed is 332.0671 MHz.
>
>
> Thanks Ross and Maciej for the patches!
>
> Jesse
> -
> 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: NForce2 + linux 2.6.3-rc2 + acpi,apic patches

[Jesse Allen]

On Sun, Feb 22, 2004 at 12:42:47PM -0500, Mark Rutherford wrote:
> Jesse,
> 
> I saw this thread on the mailing list about your sucess with the nforce2...
> I havent had much sucess with it, and was wondering if you have tried -rc3

I have tried -rc3.  But I am now using vanilla 2.6.3.

> Also, do you know if Andrew Morton's patches are cumulative?
> would he have these patches in rc3-mm2 ?

probably.  Unless it is decided a patch is bad. Generally, they stay until 
something is determined so people will test it more.

> 
> I have been using my nforce2 machine in XT-PIC ... and its pretty bad.
> I have some hardware that doesnt want to share the irq so it gets dicey.
> I would like to get it fixed, I have posted, but I didnt get a response.
> 

What happens in APIC?

Jesse