ACPI S3 fails to re-init NIC on Asus A7V

ACPI S3 fails to re-init NIC on Asus A7V

[Carsten Aulbert]

Hi all,

Please consider the following problem:

Home Server (daisy):

Asus A7V board
Duron 650 MHz
Debian sarge (not 100% up2date right now - from March 22nd)
SiS900 (also tried 3Com 3c509-TX-M without a change)

Kernels tried:
2.4.22 (IIRC), 2.6.4, 2.6.4 with acpi-patch 20040311, 2.6.6

Suspending to S3 works fine, resume also (except with the 
onboard-Promise chip, but that's not a big issue), however, trying to 
use the network after resume gives
NETDEV WATCHDOG: eth0: transmit timed out

This also did not help:
/etc/init.d/networking stop
modprobe -r sis900
echo 3 > /proc/acpi/sleep

[resume]

modprobe sis900
-> module is loaded successfully
/etc/init.d/networking start
-> sets the static routes correctly
pinging a machine on the network yields the same error as before.

(Also tried 3Com 3c509-TXM to no avail)

To spare you about ~1000 lines of output, I've put the output of dmesg, 
lspci -v, acpidmp and an excerpt from /var/log/syslog with enabled 
sis900-debugging here
http://carsten.welcomes-you.com/misc-tmp/acpi-s3/

Does this problem ring any bell, do you need more input from my side? 
I'm currently running out of options and ideas.

TIA for any help

Carsten

PS: I'm not subscribed but monitor linux.kernel, but feel free to Cc me.

Re: ACPI S3 fails to re-init NIC on Asus A7V

[Arjan van de Ven]

On Thu, May 27, 2004 at 10:14:50PM +0200, Carsten Aulbert wrote:
> Hi all,
> 
> Please consider the following problem:
> 
> Home Server (daisy):
> 
> Asus A7V board
> Duron 650 MHz
> Debian sarge (not 100% up2date right now - from March 22nd)
> SiS900 (also tried 3Com 3c509-TX-M without a change)
> 
> Kernels tried:
> 2.4.22 (IIRC), 2.6.4, 2.6.4 with acpi-patch 20040311, 2.6.6
> 
> Suspending to S3 works fine, resume also (except with the 
> onboard-Promise chip, but that's not a big issue), however, trying to 
> use the network after resume gives
> NETDEV WATCHDOG: eth0: transmit timed out



please try this patch:


diff -urNp linux-1100/drivers/pci/pci.c linux-1110/drivers/pci/pci.c
--- linux-1100/drivers/pci/pci.c
+++ linux-1110/drivers/pci/pci.c
@@ -385,6 +385,7 @@ pci_enable_device_bars(struct pci_dev *d
 int
 pci_enable_device(struct pci_dev *dev)
 {
+	dev->is_enabled = 1;
 	return pci_enable_device_bars(dev, (1 << PCI_NUM_RESOURCES) - 1);
 }
 
@@ -399,6 +400,9 @@ void
 pci_disable_device(struct pci_dev *dev)
 {
 	u16 pci_command;
+	
+	dev->is_enabled = 0;
+	dev->is_busmaster = 0;
 
 	pci_read_config_word(dev, PCI_COMMAND, &pci_command);
 	if (pci_command & PCI_COMMAND_MASTER) {
@@ -601,6 +605,7 @@ pci_set_master(struct pci_dev *dev)
 		cmd |= PCI_COMMAND_MASTER;
 		pci_write_config_word(dev, PCI_COMMAND, cmd);
 	}
+	dev->is_busmaster = 1;
 	pcibios_set_master(dev);
 }
 
diff -urNp linux-1100/drivers/pci/pci-driver.c linux-1110/drivers/pci/pci-driver.c
--- linux-1100/drivers/pci/pci-driver.c
+++ linux-1110/drivers/pci/pci-driver.c
@@ -299,10 +299,30 @@ static int pci_device_suspend(struct dev
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
 	struct pci_driver * drv = pci_dev->driver;
+	int i = 0;
 
 	if (drv && drv->suspend)
-		return drv->suspend(pci_dev,state);
-	return 0;
+		i = drv->suspend(pci_dev,state);
+		
+	pci_save_state(pci_dev, pci_dev->saved_config_space);
+	return i;
+}
+
+
+/* 
+ * Default resume method for devices that have no driver provided resume,
+ * or not even a driver at all.
+ */
+static void pci_default_resume(struct pci_dev *pci_dev)
+{
+	/* restore the PCI config space */
+	pci_restore_state(pci_dev, pci_dev->saved_config_space);
+	/* if the device was enabled before suspend, reenable */
+	if (pci_dev->is_enabled)
+		pci_enable_device(pci_dev);
+	/* if the device was busmaster before the suspend, make it busmaster again */
+	if (pci_dev->is_busmaster)
+		pci_set_master(pci_dev);
 }
 
 static int pci_device_resume(struct device * dev)
@@ -312,6 +332,8 @@ static int pci_device_resume(struct devi
 
 	if (drv && drv->resume)
 		drv->resume(pci_dev);
+	else
+		pci_default_resume(pci_dev);
 	return 0;
 }
 
diff -urNp linux-1100/include/linux/pci.h linux-1110/include/linux/pci.h
--- linux-1100/include/linux/pci.h
+++ linux-1110/include/linux/pci.h
@@ -488,6 +488,11 @@ struct pci_dev {
 	/* These fields are used by common fixups */
 	unsigned int	transparent:1;	/* Transparent PCI bridge */
 	unsigned int	multifunction:1;/* Part of multi-function device */
+	/* keep track of device state */
+	unsigned int	is_enabled:1;	/* pci_enable_device has been called */
+	unsigned int	is_busmaster:1; /* device is busmaster */
+	
+	unsigned int 	saved_config_space[16]; /* config space saved at suspend time */
 #ifdef CONFIG_PCI_NAMES
 #define PCI_NAME_SIZE	96
 #define PCI_NAME_HALF	__stringify(43)	/* less than half to handle slop */

Re: ACPI S3 fails to re-init NIC on Asus A7V

[Carsten Aulbert]

Arjan van de Ven wrote:

> 
> please try this patch:

[...]

Thanks a lot for the patch, applied it to vanilla 2.6.6, but that did 
not help (at all :( ).

in both cases (with and without unloading the sis900 module before 
suspending) I get:

/var/log/syslog
[...]
May 28 07:54:03 daisy kernel: PM: Preparing system for suspend
May 28 07:54:03 daisy kernel: Stopping tasks: 
==========================================|
May 28 07:54:03 daisy kernel: PM: Entering state.
May 28 07:54:03 daisy kernel:  hwsleep-0304 [12] acpi_enter_sleep_state: 
Entering sleep state [S3]
May 28 07:54:03 daisy kernel: Back to C!
May 28 07:54:03 daisy kernel: spurious 8259A interrupt: IRQ7.
May 28 07:54:03 daisy kernel: PM: Finishing up.
May 28 07:54:03 daisy kernel: blk: queue d7de0c00, I/O limit 4095Mb 
(mask 0xffffffff)
May 28 07:54:03 daisy kernel: blk: queue d7de0400, I/O limit 4095Mb 
(mask 0xffffffff)
May 28 07:54:03 daisy sleepd[584]: 34 sec sleep; resetting timer
May 28 07:54:03 daisy kernel: Restarting tasks... done
May 28 07:54:04 daisy kernel: MCE: The hardware reports a non fatal, 
correctable incident occurred on CPU
0.
May 28 07:54:04 daisy kernel: Bank 1: ddd1ffdd22063591
May 28 07:54:04 daisy kernel: MCE: The hardware reports a non fatal, 
correctable incident occurred on CPU
0.
May 28 07:54:04 daisy kernel: Bank 2: a00060000002042e

[ modprobe sis900 debug=4 ]

May 28 07:54:26 daisy kernel: sis900.c: v1.08.07 11/02/2003
May 28 07:54:26 daisy kernel: eth0: SiS 900 Internal MII PHY transceiver 
found at address 1.
May 28 07:54:26 daisy kernel: eth0: Using transceiver found at address 1 
as default
May 28 07:54:28 daisy kernel: eth0: SiS 900 PCI Fast Ethernet at 0xa400, 
IRQ 11, 00:30:ab:01:7b:6e.
May 28 07:54:34 daisy kernel: eth0: Receive Filter Addrss[0]=3000
May 28 07:54:34 daisy kernel: eth0: Receive Filter Addrss[1]=1ab
May 28 07:54:34 daisy kernel: eth0: Receive Filter Addrss[2]=6e7b
May 28 07:54:34 daisy kernel: eth0: TX descriptor register loaded with: 
177d6000
May 28 07:54:34 daisy kernel: eth0: RX descriptor register loaded with: 
177d1000
May 28 07:54:36 daisy kernel: eth0: Media Link On 100mbps full-duplex
May 28 07:54:48 daisy kernel: eth0: Queued Tx packet at d73f0b96 size 42 
to slot 0.
May 28 07:54:54 daisy kernel: NETDEV WATCHDOG: eth0: transmit timed out
May 28 07:54:54 daisy kernel: eth0: exiting interrupt, interrupt status 
= 0x0x00000000.
May 28 07:54:54 daisy kernel: eth0: Transmit timeout, status 00000004 
00000241
May 28 07:54:54 daisy kernel: eth0: Queued Tx packet at d73f0996 size 42 
to slot 0.

[...]

That's all info I can currently squeeze out of the box (due to my 
limited knowledge I suppose), anyone with any idea, how to proceed?


Cheers

Carsten