Re: [Bug] invalid mac address after rebooting (2.6.12-rc2-mm2)

[Daniel Ritz <daniel.ritz@gmx.ch>]

On Friday 15 April 2005 08:43, Peter Baumann wrote:
> On Thu, Apr 14, 2005 at 07:40:52PM +0200, Daniel Ritz wrote:
> > 
> > could you apply this debuggin patch instead and send me the dmsg output
> > plus output from lspci, lspci -vvvn. also please send me a hexdump from
> > /proc/bus/pci/00/0b.0
> > 
> > i think i'm having some 3coms around so maybe i can reprocude :)
> > 
> > rgds
> > -daniel
> > 
> > --- 1.81/drivers/pci/pci.c	2005-03-03 08:17:57 +01:00
> > +++ edited/drivers/pci/pci.c	2005-04-05 00:37:13 +02:00
> > @@ -268,7 +268,7 @@
> >  		return -EIO; 
> >  
> >  	pci_read_config_word(dev,pm + PCI_PM_PMC,&pmc);
> > -	if ((pmc & PCI_PM_CAP_VER_MASK) > 2) {
> > +	if ((pmc & PCI_PM_CAP_VER_MASK) > 3) {
> >  		printk(KERN_DEBUG
> >  		       "PCI: %s has unsupported PM cap regs version (%u)\n",
> >  		       pci_name(dev), pmc & PCI_PM_CAP_VER_MASK);
> > @@ -287,15 +287,19 @@
> >  	 * This doesn't affect PME_Status, disables PME_En, and
> >  	 * sets PowerState to 0.
> >  	 */
> > -	if (dev->current_state >= PCI_D3hot)
> > +	printk("PCI: %s pmc: %04x, current_state, pmcsr: %08x", pci_name(dev), pmc, dev->current_state);

for the record: i messed up that printk, but anyway...

> > +	if (dev->current_state >= PCI_D3hot) {
> >  		pmcsr = 0;
> > -	else {
> > +		printk("0");
> > +	} else {
> >  		pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr);
> > +		printk("%04x", pmcsr);
> >  		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
> >  		pmcsr |= state;
> >  	}
> >  
> >  	/* enter specified state */
> > +	printk(", new: %04x\n", pmcsr);
> >  	pci_write_config_word(dev, pm + PCI_PM_CTRL, pmcsr);
> >  
> >  	/* Mandatory power management transition delays */
> > 
> 
> I applied this one (just for the record, the offsets weren't correct for 
> 2.6.12-rc2-mm3, but it applied cleanly)

(it was against mainline, not -mm)

> 
> Attached is the info you asked for.
> 
> ...1.txt  = first boot (working)
> ...2.txt  = second boot (not working)
> 
> -Peter
> 

from your dmesg:
PCI: 0000:00:0b.0 pmc: 7601, current_state, pmcsr: 000000040, new: 0000
ACPI: PCI Interrupt 0000:00:0b.0[A] -> GSI 19 (level, low) -> IRQ 19
3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c905B Cyclone 100baseTx at 0xac00. Vers LK1.1.19
PCI: 0000:00:0b.0 pmc: 7601, current_state, pmcsr: 000000000000, new: 0003

so the device is sent to D3 right after it is up. nice.

and second boot:
PCI: 0000:00:0b.0 pmc: 7601, current_state, pmcsr: 000000040, new: 0000
PCI: Enabling device 0000:00:0b.0 (0000 -> 0003)
ACPI: PCI Interrupt 0000:00:0b.0[A] -> GSI 19 (level, low) -> IRQ 19
3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c905B Cyclone 100baseTx at 0x1000. Vers LK1.1.19
PCI: Setting latency timer of device 0000:00:0b.0 to 64
*** EEPROM MAC address is invalid.
3c59x: vortex_probe1 fails.  Returns -22
3c59x: probe of 0000:00:0b.0 failed with error -22

to me it looks like during the second boot the device is in D3 and has
troubles coming out of it. unfortunatley that is not made visible by the
debugging patch. a hexdump of the device in a second boot without
loading the 3c59x driver would tell us...

the diff in the hexdumps of the config space proves that the device
is programmed wrong:
--- hexdump-2.6.12-rc2-mm3-withdebugpatch_1.txt 2005-04-17 21:27:28.000000000 +0200
+++ hexdump-2.6.12-rc2-mm3-withdebugpatch_2.txt 2005-04-17 21:27:32.000000000 +0200
@@ -1,7 +1,7 @@
-0000000 10b7 9055 0007 0210 0024 0200 2008 0000
-0000010 ac01 0000 2000 e800 0000 0000 0000 0000
+0000000 10b7 9055 0003 0210 0024 0200 4000 0000
+0000010 0001 0000 0000 0000 0000 0000 0000 0000

so the I/O ports and the iomem are not set!

 0000020 0000 0000 0000 0000 0000 0000 10b7 9055
-0000030 0000 0000 00dc 0000 0000 0000 010a 0a0a
+0000030 0000 0000 00dc 0000 0000 0000 0100 0a0a
 0000040 0000 0000 0000 0000 0000 0000 0000 0000
 0000050 0000 0000 0040 0000 0000 0000 0000 0000
 0000060 0000 0000 0000 0000 0000 0000 0000 0000

but to make it short: i think it's a driver bug. the device doesn't seem to be
correctly reprogrammed when it is in D3 initially. and then it shouldn't be
put into D3 during the first boot. so please try with the attached patch.

rgds
-daniel

PS: does somebody have a datasheet of a 3com chip around? i could need one.

-----

[PATCH] 3c59x: only put the device into D3 when we're actually using WOL

Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>

--- 1.77/drivers/net/3c59x.c	2005-03-03 06:00:42 +01:00
+++ edited/drivers/net/3c59x.c	2005-04-17 22:17:19 +02:00
@@ -1581,7 +1581,8 @@
 
 	if (VORTEX_PCI(vp)) {
 		pci_set_power_state(VORTEX_PCI(vp), PCI_D0);	/* Go active */
-		pci_restore_state(VORTEX_PCI(vp));
+		if (vp->pm_state_valid)
+			pci_restore_state(VORTEX_PCI(vp));
 		pci_enable_device(VORTEX_PCI(vp));
 	}
 
@@ -2741,6 +2742,7 @@
 		outl(0, ioaddr + DownListPtr);
 
 	if (final_down && VORTEX_PCI(vp)) {
+		vp->pm_state_valid = 1;
 		pci_save_state(VORTEX_PCI(vp));
 		acpi_set_WOL(dev);
 	}
@@ -3243,9 +3245,10 @@
 		outw(RxEnable, ioaddr + EL3_CMD);
 
 		pci_enable_wake(VORTEX_PCI(vp), 0, 1);
+
+		/* Change the power state to D3; RxEnable doesn't take effect. */
+		pci_set_power_state(VORTEX_PCI(vp), PCI_D3hot);
 	}
-	/* Change the power state to D3; RxEnable doesn't take effect. */
-	pci_set_power_state(VORTEX_PCI(vp), PCI_D3hot);
 }
 
 


-
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/