powernow-k8 patch

powernow-k8 patch

[mark.langsdorf]

[-- Attachment #1: Type: text/plain, Size: 488 bytes --]

AMD recently errata'd the definition of the PSB/PST
for recent Athlon 64 and Opteron parts.  The errata
allows for a second, identical PST for those parts.
The current powernow-k8 driver will not work in 
PST/PSB mode on those parts because it requires 
there be 1 and only 1 PST.

This patch adjusts the powernow-k8 driver to 
work with that errata.

Mark Langsdorf
Advanced Micro Devices        mark.langsdorf@amd.com
5204 E. Ben White Blvd        Phone: 512.602.3756
Austin, TX 78704


[-- Attachment #2: powernow-k8-new-cpus-2.6.8.1.patch --]
[-- Type: application/octet-stream, Size: 1209 bytes --]

--- arch/i386/kernel/cpu/cpufreq/powernow-k8.c.orig	2004-08-14 05:55:33.000000000 -0500
+++ arch/i386/kernel/cpu/cpufreq/powernow-k8.c	2004-08-18 14:33:56.953872328 -0500
@@ -39,7 +39,7 @@
 
 #define PFX "powernow-k8: "
 #define BFX PFX "BIOS error: "
-#define VERSION "version 1.00.09b"
+#define VERSION "version 1.00.09e"
 #include "powernow-k8.h"
 
 /* serialize freq changes  */
@@ -596,6 +596,8 @@ static int find_psb_table(struct powerno
 	unsigned int i;
 	u32 mvs;
 	u8 maxvid;
+	u32 cpst = 0;
+	u32 thiscpuid;
 
 	for (i = 0xc0000; i < 0xffff0; i += 0x10) {
 		/* Scan BIOS looking for the signature. */
@@ -634,7 +636,14 @@ static int find_psb_table(struct powerno
 		dprintk(KERN_INFO PFX "maximum voltage step: %d - 0x%x\n", mvs, data->vidmvs);
 
 		dprintk(KERN_DEBUG PFX "numpst: 0x%x\n", psb->numpst);
-		if (psb->numpst != 1) {
+		cpst = psb->numpst;
+		if ((psb->cpuid == 0x00000fc0) || (psb->cpuid == 0x00000fe0) ){
+			thiscpuid = cpuid_eax(CPUID_PROCESSOR_SIGNATURE);
+			if ((thiscpuid == 0x00000fc0) || (thiscpuid == 0x00000fe0) ) {
+				cpst = 1;
+			}
+		}
+		if (cpst != 1) {
 			printk(KERN_ERR BFX "numpst must be 1\n");
 			return -ENODEV;
 		}

[-- Attachment #3: Type: text/plain, Size: 143 bytes --]

_______________________________________________
Cpufreq mailing list
Cpufreq@www.linux.org.uk
http://www.linux.org.uk/mailman/listinfo/cpufreq

Re: powernow-k8 patch

[Ari Pollak]

Does that happen to fix the bug I reported earlier in powernow-k8?

mark.langsdorf@amd.com wrote:
> AMD recently errata'd the definition of the PSB/PST
> for recent Athlon 64 and Opteron parts.  The errata
> allows for a second, identical PST for those parts.
> The current powernow-k8 driver will not work in 
> PST/PSB mode on those parts because it requires 
> there be 1 and only 1 PST.
> 
> This patch adjusts the powernow-k8 driver to 
> work with that errata.
> 
> Mark Langsdorf
> Advanced Micro Devices        mark.langsdorf@amd.com
> 5204 E. Ben White Blvd        Phone: 512.602.3756
> Austin, TX 78704

RE: powernow-k8 patch

[mark.langsdorf]

Ari Pollak wrote:
> mark.langsdorf@amd.com wrote:
> > AMD recently errata'd the definition of the PSB/PST
> > for recent Athlon 64 and Opteron parts.  The errata
> > allows for a second, identical PST for those parts.
> > The current powernow-k8 driver will not work in
> > PST/PSB mode on those parts because it requires 
> > there be 1 and only 1 PST.
> > 
> > This patch adjusts the powernow-k8 driver to
> > work with that errata.

> Does that happen to fix the bug I reported earlier in powernow-k8?

No.  Your problem is caused by a malformed PST table
that has too many "low" frequency entries and by a 
PSB header that isn't the correct revision (1.4).  
Complain to your BIOS vendor.  Tell them that their 
BIOS does not support PN! on Win2K.

I'm curious as to why it doesn't work under ACPI.  
Can you instantiate the ACPI version and give me the
debug messages?

-Mark Langsdorf
AMD, Inc.

Re: powernow-k8 patch

[Ari Pollak]

 > No.  Your problem is caused by a malformed PST table
 > that has too many "low" frequency entries and by a
 > PSB header that isn't the correct revision (1.4).
 > Complain to your BIOS vendor.  Tell them that their
 > BIOS does not support PN! on Win2K.
That's interesting, since IIRC AMD distributed a similar MSI motherboard 
as a demo/reference. Maybe there's some obscure BIOS setting I'm missing.

 > I'm curious as to why it doesn't work under ACPI.
 > Can you instantiate the ACPI version and give me the
 > debug messages?
I'm also wondering why powernow-k8 doesn't fallback to retrieving 
information from ACPI, which is what the documentation says it's 
supposed to do.

Not sure if this is what you want, but here's the ACPI debugging stuff 
from dmesg:
ACPI: RSDP (v000 AMI                                       ) @ 0x000fa3f0
ACPI: RSDT (v001 AMIINT VIA_K8   0x00000010 MSFT 0x00000097) @ 0x1fff0000
ACPI: FADT (v001 AMIINT VIA_K8   0x00000011 MSFT 0x00000097) @ 0x1fff0030
ACPI: MADT (v001 AMIINT VIA_K8   0x00000009 MSFT 0x00000097) @ 0x1fff00c0
ACPI: DSDT (v001    VIA   VIA_K8 0x00001000 MSFT 0x0100000d) @ 0x00000000
ACPI: Local APIC address 0xfee00000
ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled)
ACPI: Subsystem revision 20040326
  tbxface-0117 [03] acpi_load_tables      : ACPI Tables successfully 
acquired
Parsing all Control 
Methods:.....................................................................................................................................................
Table [DSDT](id F004) - 520 Objects with 48 Devices 149 Methods 26 Regions
ACPI Namespace successfully loaded at root c03c0c7c
ACPI: IRQ9 SCI: Edge set to Level Trigger.
evxfevnt-0093 [04] acpi_enable           : Transition to ACPI mode 
successful
evgpeblk-0867 [06] ev_create_gpe_block   : GPE 00 to 15 [_GPE] 2 regs at 
0000000000000820 on int 9
evgpeblk-0925 [06] ev_create_gpe_block   : Found 0 Wake, Enabled 6 
Runtime GPEs in this block
Completing Region/Field/Buffer/Package 
initialization:.......................................................................................................
Initialized 26/26 Regions 9/9 Fields 44/44 Buffers 24/24 Packages (529 
nodes)
Executing all Device _STA and_INI 
methods:..................................................
50 Devices found containing: 50 _STA, 1 _INI methods
ACPI: Interpreter enabled
ACPI: Using PIC for interrupt routing
ACPI: PCI Root Bridge [PCI0] (00:00)
PCI: Probing PCI hardware (bus 00)
ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
ACPI: Power Resource [URP1] (off)
ACPI: Power Resource [URP2] (off)
ACPI: Power Resource [FDDP] (off)
ACPI: Power Resource [LPTP] (off)
ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 6 7 10 *11 12 14 15)
ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 6 7 *10 11 12 14 15)
ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 *5 6 7 10 11 12 14 15)
ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 6 7 10 11 *12 14 15)
ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 6 7 10 11 12 14 15) *0, 
disabled.
ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 6 7 10 11 12 14 15) *0, 
disabled.
ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 6 7 10 11 12 14 15) *0, 
disabled.
ACPI: PCI Interrupt Link [LNKH] (IRQs 3 4 5 6 7 10 11 12 14 15) *0, 
disabled.
PCI: Using ACPI for IRQ routing
ACPI: PCI Interrupt Link [LNKB] enabled at IRQ 10
ACPI: PCI interrupt 0000:00:06.0[A] -> GSI 10 (level, low) -> IRQ 10
ACPI: PCI Interrupt Link [LNKA] enabled at IRQ 11
ACPI: PCI interrupt 0000:00:0b.0[A] -> GSI 11 (level, low) -> IRQ 11
ACPI: PCI interrupt 0000:00:0f.0[B] -> GSI 10 (level, low) -> IRQ 10
ACPI: PCI interrupt 0000:00:0f.1[A] -> GSI 11 (level, low) -> IRQ 11
ACPI: PCI interrupt 0000:00:10.0[A] -> GSI 11 (level, low) -> IRQ 11
ACPI: PCI interrupt 0000:00:10.1[A] -> GSI 11 (level, low) -> IRQ 11
ACPI: PCI interrupt 0000:00:10.2[B] -> GSI 10 (level, low) -> IRQ 10
ACPI: PCI interrupt 0000:00:10.3[B] -> GSI 10 (level, low) -> IRQ 10
ACPI: PCI Interrupt Link [LNKC] enabled at IRQ 5
ACPI: PCI interrupt 0000:00:10.4[C] -> GSI 5 (level, low) -> IRQ 5
ACPI: PCI interrupt 0000:00:11.5[C] -> GSI 5 (level, low) -> IRQ 5
ACPI: PCI interrupt 0000:01:00.0[A] -> GSI 11 (level, low) -> IRQ 11

and after I try to load the acpi module:
ACPI: Processor [CPU1] (supports C1)
acpi_processor_perf-0301 [30] acpi_cpufreq_cpu_init : Unsupported 
address space [127, 127]

RE: powernow-k8 patch

[mark.langsdorf]

>  > No.  Your problem is caused by a malformed PST table
>  > that has too many "low" frequency entries and by a
>  > PSB header that isn't the correct revision (1.4).
>  > Complain to your BIOS vendor.  Tell them that their
>  > BIOS does not support PN! on Win2K.
>
> That's interesting, since IIRC AMD distributed a similar MSI 
> motherboard as a demo/reference. Maybe there's some obscure
> BIOS setting I'm missing.

No comment.
 
>  > I'm curious as to why it doesn't work under ACPI.
>  > Can you instantiate the ACPI version and give me the
>  > debug messages?

> I'm also wondering why powernow-k8 doesn't fallback to retrieving 
> information from ACPI, which is what the documentation says it's 
> supposed to do.
 
> Not sure if this is what you want, but here's the ACPI 
> debugging stuff 

Actually, I meant the powernow-k8 information.  If it failed
to load, it should give some reason why.

> and after I try to load the acpi module:
> ACPI: Processor [CPU1] (supports C1)
> acpi_processor_perf-0301 [30] acpi_cpufreq_cpu_init : Unsupported 
> address space [127, 127]

It looks like there's something wrong with the ACPI module,
but I don't know what that means.

-Mark Langsdorf
AMD, Inc.

Re: powernow-k8 patch

[Ari Pollak]

> Actually, I meant the powernow-k8 information.  If it failed
> to load, it should give some reason why.

The only messages that got output were the 3 I put in my original e-mail.

Re: powernow-k8 patch

[Ari Pollak]

> Actually, I meant the powernow-k8 information.  If it failed
> to load, it should give some reason why.

The only messages that got output were the 3 I put in my original e-mail.

Re: powernow-k8 patch

[Bruno Ducrot]

On Thu, Aug 19, 2004 at 02:26:18PM -0500, mark.langsdorf@amd.com wrote:
> >  > No.  Your problem is caused by a malformed PST table
> >  > that has too many "low" frequency entries and by a
> >  > PSB header that isn't the correct revision (1.4).
> >  > Complain to your BIOS vendor.  Tell them that their
> >  > BIOS does not support PN! on Win2K.
> >
> > That's interesting, since IIRC AMD distributed a similar MSI 
> > motherboard as a demo/reference. Maybe there's some obscure
> > BIOS setting I'm missing.
> 
> No comment.

Maybe you have to check a powernow option on BIOS?

> >  > I'm curious as to why it doesn't work under ACPI.
> >  > Can you instantiate the ACPI version and give me the
> >  > debug messages?
> 
> > I'm also wondering why powernow-k8 doesn't fallback to retrieving 
> > information from ACPI, which is what the documentation says it's 
> > supposed to do.
>  
> > Not sure if this is what you want, but here's the ACPI 
> > debugging stuff 
> 
> Actually, I meant the powernow-k8 information.  If it failed
> to load, it should give some reason why.
> 
> > and after I try to load the acpi module:
> > ACPI: Processor [CPU1] (supports C1)
> > acpi_processor_perf-0301 [30] acpi_cpufreq_cpu_init : Unsupported 
> > address space [127, 127]
> 
> It looks like there's something wrong with the ACPI module,
> but I don't know what that means.

The last 2 lines are from the cpufreq ACPI driver (the IO version).

Cheers,
-- 
Bruno Ducrot

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.

Re: powernow-k8 patch

[Ari Pollak]

Bruno Ducrot <ducrot <at> poupinou.org> writes:
> Maybe you have to check a powernow option on BIOS?
I've gone through every setting, and the only related option I could find is for
Cool n' Quiet, which I have enabled.

> The last 2 lines are from the cpufreq ACPI driver (the IO version).
Right, that's the one I loaded.