Re: powernow not working on AMD Athlon XP Mobile
Hi Sepherosa,
> Well, I think I will need AMD 06h family BKDG to figure out how the
> ACPI VID and FID should be used. The code we have messed w/ so far is
> according to AMD 0fh family BKDG.
I'm doing some research and looking how other operating systems works
with my Athlon XP-M processor.
1) The DFly's powernow driver is almost identical to the NetBSD's
powernow_k8 driver (NetBSD has a powernow_k7 driver too) , probably
this is the reason why on DFly the powernow module returns an error
message when I try to load it, it simply does not support K7
architectures.
2) On FreeBSD: sys/x86/cpufreq/hwpstate.c
Looks like my CPU family (0x06) is not supported:
did = AMD_10H_11H_CUR_DID(msr);
fid = AMD_10H_11H_CUR_FID(msr);
switch(family) {
case 0x11:
/* fid/did to frequency */
hwpstate_set[i].freq = 100 * (fid + 0x08) / (1 << did);
break;
case 0x10:
/* fid/did to frequency */
hwpstate_set[i].freq = 100 * (fid + 0x10) / (1 << did);
break;
default:
HWPSTATE_DEBUG(dev, "get_info_from_msr: AMD family %d CPU's are not
implemented yet. sorry.\n", family);
return (ENXIO);
break;
}
---
But in their powernow driver my cpu family *is* supported:
sys/x86/cpufreq/powernow.c
--
switch (cpu_id & 0xf00) {
case 0x600:
sfid = PN7_STA_SFID(status);
mfid = PN7_STA_MFID(status);
cfid = PN7_STA_CFID(status);
sc->pn_type = PN7_TYPE;
sc->fsb = rate / 10 / pn7_fid_to_mult[cfid];
/*
* If start FID is different to max FID, then it is a
* mobile processor. If not, it is a low powered desktop
* processor.
*/
if (PN7_STA_SFID(status) != PN7_STA_MFID(status)) {
sc->vid_to_volts = pn7_mobile_vid_to_volts;
device_set_desc(dev, "PowerNow! K7");
} else {
sc->vid_to_volts = pn7_desktop_vid_to_volts;
device_set_desc(dev, "Cool`n'Quiet K7");
}
break;
---
Cheers,
--
Maurizio Lombardi
Re: powernow not working on AMD Athlon XP Mobile
On Thu, Sep 15, 2011 at 4:13 AM, Maurizio Lombardi wrote: >> You could try to set hw.acpi.cpu.pstate.strigent_check=0 in >> /boot/loader.conf. My K8 box needs that for it's messed up FID/VID. >> > > Thanks, > > I tried but the kernel freezes at boot. > > It prints a lot of error like: > > "cpu0: Invalid FID" > "cpu0: Invalid VID..." > > The last two lines it prints are a bit different: > > "cpu0: min VID mismatch real 9, lim 11" > "cpu0: min FID mismatch real 50, lim 6" Well, I think I will need AMD 06h family BKDG to figure out how the ACPI VID and FID should be used. The code we have messed w/ so far is according to AMD 0fh family BKDG. Best Regards, sephe -- Tomorrow Will Never Die
Re: powernow not working on AMD Athlon XP Mobile
> You could try to set hw.acpi.cpu.pstate.strigent_check=0 in > /boot/loader.conf. My K8 box needs that for it's messed up FID/VID. > Thanks, I tried but the kernel freezes at boot. It prints a lot of error like: "cpu0: Invalid FID" "cpu0: Invalid VID..." The last two lines it prints are a bit different: "cpu0: min VID mismatch real 9, lim 11" "cpu0: min FID mismatch real 50, lim 6" Cheers, -- Maurizio Lombardi
Re: powernow not working on AMD Athlon XP Mobile
On 14 sep 2011, at 20.45, Maurizio Lombardi wrote >> Ok. Remove the old patch and try this one: >> http://leaf.dragonflybsd.org/~sephe/pstate.c.diff1 > > Thanks for your support Sepherosa, but unfortunately it still does not work. > > The last patch triggered others errors like the following: > cpu0: Invalid FID 0x20, out [0xe, 0x16] > cpu0: Invalid VID 0x6, in [0xd, 0xb] You could try to set hw.acpi.cpu.pstate.strigent_check=0 in /boot/loader.conf. My K8 box needs that for it's messed up FID/VID. See http://bugs.dragonflybsd.org/issue2025 cheers, Max
Re: powernow not working on AMD Athlon XP Mobile
> Ok. Remove the old patch and try this one: > http://leaf.dragonflybsd.org/~sephe/pstate.c.diff1 Thanks for your support Sepherosa, but unfortunately it still does not work. The last patch triggered others errors like the following: cpu0: Invalid FID 0x20, out [0xe, 0x16] cpu0: Invalid VID 0x6, in [0xd, 0xb] I modified your patch to force the kernel to ignore those errors: http://www.dotter.it/~maurizio/acpic5.diff But the kernel freezes at boot when it tries to initialize the cpu driver -- Maurizio Lombardi
Re: powernow not working on AMD Athlon XP Mobile
On Wed, Sep 14, 2011 at 5:25 AM, Maurizio Lombardi wrote: >> Try following patch: >> http://leaf.dragonflybsd.org/~sephe/pstate_machdep.c.diff >> >> See whether following sysctl appears: >> hw.acpi.cpu.px_dom0.select > > I applied your patch and rebuild the kernel. > > Unfortunately it does not work: > > #sysctl -a | grep px > # > > Now it does not print anything at all! > Any ideas? Ok. Remove the old patch and try this one: http://leaf.dragonflybsd.org/~sephe/pstate.c.diff1 > > However, full dmesg attached. > > -- > > Maurizio Lombardi > -- Tomorrow Will Never Die
Re: powernow not working on AMD Athlon XP Mobile
> Try following patch: > http://leaf.dragonflybsd.org/~sephe/pstate_machdep.c.diff > > See whether following sysctl appears: > hw.acpi.cpu.px_dom0.select I applied your patch and rebuild the kernel. Unfortunately it does not work: #sysctl -a | grep px # Now it does not print anything at all! Any ideas? However, full dmesg attached. -- Maurizio Lombardi Copyright (c) 2003-2011 The DragonFly Project. Copyright (c) 1992-2003 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. madt_probe: can't locate MADT DragonFly v2.11.0.836.g3ddc3-DEVELOPMENT #0: Tue Sep 13 22:48:33 UTC 2011 root@:/usr/obj/usr/src/sys/LMB TSC clock: 1855182810 Hz, i8254 clock: 1193179 Hz CPU: mobile AMD Athlon(tm) XP2500+ (1855.20-MHz 686-class CPU) Origin = "AuthenticAMD" Id = 0x6a0 Stepping = 0 Features=0x383f9ff AMD Features=0xc0480800 real memory = 526972928 (502 MB) avail memory = 494817280 (471 MB) LAPIC: Can't find LAPIC module_register: module pci/ahci already exists! Module pci/ahci failed to register: 17 FQ scheduler policy version 1.1 loaded kbd0 at kbdmux0 disk scheduler: set policy of md0 to noop md0: Malloc disk ACPI: RSDP 0xf7290 00014 (v0 PTLTD ) ACPI: RSDT 0x1f6f8b67 00030 (v1 PTLTDRSDT 0604 LTP ) ACPI: FACP 0x1f6fee2b 00074 (v1 ATIRaptor 0604 ATI 000F4240) ACPI: DSDT 0x1f6f8b97 06294 (v1ATI U1_M1535 0604 MSFT 010D) ACPI: FACS 0x1f6fffc0 00040 ACPI: BOOT 0x1f6fee9f 00028 (v1 PTLTD $SBFTBL$ 0604 LTP 0001) ACPI: SSDT 0x1f6feec7 00139 (v1 PTLTD POWERNOW 0604 LTP 0001) npx0: on motherboard npx0: INT 16 interface Using MMX optimized bcopy/copyin/copyout cryptosoft0: on motherboard acpi0: on motherboard ACPI FADT: SCI testing interrupt mode ... ACPI FADT: SCI select level/low objcache_reclaimlist objcache_reclaimlist objcache_reclaimlist objcache_reclaimlist acpi0: Power Button (fixed) Warning: ACPI is disabling APM's device. You can't run both [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" acpi_timer0: <32-bit timer at 3.579545MHz> port 0x8008-0x800b on acpi0 acpi_ec0: port 0x66,0x62 on acpi0 [ACPI Debug] String [0x1C] "THERM: _Q80 - Thermal Change" [ACPI Debug] String [0x16] "Notify _TZ.THRM , 0x80" pcib0: port 0xcf8-0xcff on acpi0 pci0: on pcib0 agp0: port 0x8090-0x8093 mem 0xe440-0xe4400fff,0xe800-0xebff at device 0.0 on pci0 pcib1: at device 1.0 on pci0 pci1: on pcib1 vgapci0: port 0x9000-0x90ff mem 0xe410-0xe410,0xec00-0xedff irq 10 at device 5.0 on pci1 ohci0: mem 0xe400-0xe4000fff irq 9 at device 2.0 on pci0 usb0: OHCI version 1.0, legacy support usb0: on ohci0 usb0: USB revision 1.0 uhub0: on usb0 uhub0: 4 ports with 4 removable, self powered pci0: (vendor 0x10b9, dev 0x5451) at device 6.0 irq 5 isab0: at device 7.0 on pci0 isa0: on isab0 pci0: (vendor 0x10b9, dev 0x5457) at device 8.0 irq 3 cbb0: irq 11 at device 10.0 on pci0 cardbus0: on cbb0 pccard0: <16-bit PCCard bus> on cbb0 atapci0: port 0x8080-0x808f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 16.0 on pci0 atapci0: using PIO transfers above 137GB as workaround for 48bit DMA access bug, expect reduced performance ata0: on atapci0 disk scheduler: set policy of ad0 to noop [ACPI Debug] String [0x1C] "THERM: _Q80 - Thermal Change" [ACPI Debug] String [0x16] "Notify _TZ.THRM , 0x80" [ACPI Debug] String [0x1C] "THERM: _Q80 - Thermal Change" [ACPI Debug] String [0x16] "Notify _TZ.THRM , 0x80" ad0: 152627MB at ata0-master UDMA100 ata1: on atapci0 disk scheduler: set policy of acd0 to noop acd0: CDRW at ata1-master PIO4 pci0: (vendor 0x10b9, dev 0x7101) at device 17.0 sis0: port 0x8c00-0x8cff mem 0xe4003000-0xe4003fff irq 11 at device 18.0 on pci0 miibus0: on sis0 ukphy0: on miibus0 ukphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto sis0: MAC address: 00:0f:20:c7:7f:8d acpi_button0: on acpi0 acpi_lid0: on acpi0 acpi_acad0: on acpi0 [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String
Re: powernow not working on AMD Athlon XP Mobile
On Sun, Sep 11, 2011 at 6:46 PM, Maurizio Lombardi wrote: >> >> Under most of the cases, powernow kernel module is outdated, you >> should use acpi p-states if possible. >> It looks like acpi p-states do exist, but lacking CPU driver. Could >> you post the dmesg? > > dmesg attached, thanks for your quick answer. Try following patch: http://leaf.dragonflybsd.org/~sephe/pstate_machdep.c.diff See whether following sysctl appears: hw.acpi.cpu.px_dom0.select > > -- > > Maurizio Lombardi > -- Tomorrow Will Never Die
Re: powernow not working on AMD Athlon XP Mobile
> > Under most of the cases, powernow kernel module is outdated, you > should use acpi p-states if possible. > It looks like acpi p-states do exist, but lacking CPU driver. Could > you post the dmesg? dmesg attached, thanks for your quick answer. -- Maurizio Lombardi Copyright (c) 2003-2011 The DragonFly Project. Copyright (c) 1992-2003 The FreeBSD Project. Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 The Regents of the University of California. All rights reserved. DragonFly v2.10.1.1.gf7ba0-RELEASE #1: Mon Apr 25 19:48:10 UTC 2011 r...@pkgbox32.dragonflybsd.org:/usr/obj/usr/src/sys/GENERIC TSC clock: 1855145616 Hz, i8254 clock: 1193155 Hz CPU: mobile AMD Athlon(tm) XP2500+ (1753.31-MHz 686-class CPU) Origin = "AuthenticAMD" Id = 0x6a0 Stepping = 0 Features=0x383f9ff AMD Features=0xc0480800 real memory = 526972928 (502 MB) avail memory = 495161344 (472 MB) module_register: module pci/ahci already exists! Module pci/ahci failed to register: 17 FQ scheduler policy version 1.0 loaded disk scheduler: set policy of md0 to noop md0: Malloc disk kbd0 at kbdmux0 ACPI: RSDP 0xf7290 00014 (v0 PTLTD ) ACPI: RSDT 0x1f6f8b67 00030 (v1 PTLTDRSDT 0604 LTP ) ACPI: FACP 0x1f6fee2b 00074 (v1 ATIRaptor 0604 ATI 000F4240) ACPI: DSDT 0x1f6f8b97 06294 (v1ATI U1_M1535 0604 MSFT 010D) ACPI: FACS 0x1f6fffc0 00040 ACPI: BOOT 0x1f6fee9f 00028 (v1 PTLTD $SBFTBL$ 0604 LTP 0001) ACPI: SSDT 0x1f6feec7 00139 (v1 PTLTD POWERNOW 0604 LTP 0001) npx0: on motherboard npx0: INT 16 interface Using MMX optimized bcopy/copyin/copyout cryptosoft0: on motherboard acpi0: on motherboard ACPI FADT: SCI testing interrupt mode ... ACPI FADT: SCI select level/low objcache_reclaimlist objcache_reclaimlist objcache_reclaimlist objcache_reclaimlist acpi0: Power Button (fixed) Warning: ACPI is disabling APM's device. You can't run both [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" acpi_timer0: <32-bit timer at 3.579545MHz> port 0x8008-0x800b on acpi0 acpi_ec0: port 0x66,0x62 on acpi0 pcib0: port 0xcf8-0xcff on acpi0 pci0: on pcib0 agp0: port 0x8090-0x8093 mem 0xe440-0xe4400fff,0xe800-0xebff at device 0.0 on pci0 pcib1: at device 1.0 on pci0 pci1: on pcib1 vgapci0: port 0x9000-0x90ff mem 0xe410-0xe410,0xec00-0xedff irq 10 at device 5.0 on pci1 ohci0: mem 0xe400-0xe4000fff irq 9 at device 2.0 on pci0 usb0: OHCI version 1.0, legacy support usb0: on ohci0 usb0: USB revision 1.0 uhub0: on usb0 uhub0: 4 ports with 4 removable, self powered pci0: (vendor 0x10b9, dev 0x5451) at device 6.0 irq 5 isab0: at device 7.0 on pci0 isa0: on isab0 pci0: (vendor 0x10b9, dev 0x5457) at device 8.0 irq 3 cbb0: irq 11 at device 10.0 on pci0 cardbus0: on cbb0 pccard0: <16-bit PCCard bus> on cbb0 atapci0: port 0x8080-0x808f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 16.0 on pci0 atapci0: using PIO transfers above 137GB as workaround for 48bit DMA access bug, expect reduced performance ata0: on atapci0 disk scheduler: set policy of ad0 to noop ad0: 152627MB at ata0-master UDMA100 ata1: on atapci0 disk scheduler: set policy of acd0 to noop acd0: CDRW at ata1-master PIO4 pci0: (vendor 0x10b9, dev 0x7101) at device 17.0 sis0: port 0x8c00-0x8cff mem 0xe4003000-0xe4003fff irq 11 at device 18.0 on pci0 miibus0: on sis0 ukphy0: on miibus0 ukphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto sis0: MAC address: 00:0f:20:c7:7f:8d acpi_button0: on acpi0 acpi_lid0: on acpi0 acpi_acad0: on acpi0 [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" battery0: on acpi0 [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" [ACPI Debug] String [0x12] "CMBatt - _STA.BAT1" acpi_tz0: on acpi0 atkbdc0: port 0x64,0x60 irq 1 on acpi0 atkbd0: flags 0x1 irq 1 on atkbdc0 kbd1 at atkbd0 [ACPI Debug] String [0x1C] "THERM: _Q80 - Thermal Change" [ACPI Debug] String [0x16] "Notify _TZ.THRM , 0x80" [ACPI Debug] String [0x25] "Current Temperature C is --- " [ACPI Debug] Integ
Re: powernow not working on AMD Athlon XP Mobile
On Sun, Sep 11, 2011 at 5:51 PM, Maurizio Lombardi wrote: > I'm looking at the DFly BSD powernow and the FreeBSD's powernow > drivers differences. > If I'm not wrong, the DFly powernow does not support the K7 > architecture, is it correct? Under most of the cases, powernow kernel module is outdated, you should use acpi p-states if possible. It looks like acpi p-states do exist, but lacking CPU driver. Could you post the dmesg? Or at least post the lines like following one (usually at the beginning of ): CPU: AMD Phenom(tm) 9550 Quad-Core Processor (2200.09-MHz K8-class CPU) Origin = "AuthenticAMD" Id = 0x100f23 Stepping = 3 Best Regards, sephe -- Tomorrow Will Never Die
Re: powernow not working on AMD Athlon XP Mobile
I'm looking at the DFly BSD powernow and the FreeBSD's powernow drivers differences. If I'm not wrong, the DFly powernow does not support the K7 architecture, is it correct? -- Maurizio Lombardi
