Date:	Wed, 27 Aug 2008 22:58:11 +0100
From:	"Daniel J Blueman" <daniel.blueman@...il.com>
To:	"Stephen Hemminger" <shemminger@...tta.com>
Cc:	"Krzysztof Oledzki" <ole@....pl>,
	"Linux Netdev" <netdev@...r.kernel.org>,
	"Linux Networking" <linux-net@...r.kernel.org>
Subject: Re: [sky2, solved] transmit timeouts and firmware update...

On Wed, Aug 27, 2008 at 10:48 PM, Daniel J Blueman
<daniel.blueman@...il.com> wrote:
> On Wed, Aug 27, 2008 at 10:42 PM, Stephen Hemminger
> <shemminger@...tta.com> wrote:
>> On Wed, 27 Aug 2008 22:37:51 +0100
>> "Daniel J Blueman" <daniel.blueman@...il.com> wrote:
>>
>>> On Wed, Aug 27, 2008 at 9:27 PM, Stephen Hemminger
>>> <shemminger@...tta.com> wrote:
>>> > On Mon, 25 Aug 2008 15:22:06 -0400
>>> > Stephen Hemminger <shemminger@...tta.com> wrote:
>>> >
>>> >> On Mon, 25 Aug 2008 21:03:25 +0200 (CEST)
>>> >> Krzysztof Oledzki <ole@....pl> wrote:
>>> >>
>>> >> >
>>> >> >
>>> >> > On Mon, 25 Aug 2008, Daniel J Blueman wrote:
>>> >> >
>>> >> > > I (and a lot of other users) have been experiencing the frequent sky2
>>> >> > > transmit timeout problem [1] (on 88E8053/Yukon2 EC gig hardware); this
>>> >> > > is a result of the embedded NIC controller locking up, and I've found
>>> >> > > that updating the firmware addresses this issue. I'm still seeing a
>>> >> > > previous and different issue [2] from time to time though (silicon
>>> >> > > bug?).
>>> >> >
>>> >> > Thanks for the info. I've been dealing with this problem for some long
>>> >> > time, hopefully sky driver from the latest kernels is able to recover
>>> >> > from such hangs, so it is not that critical problem now.
>>> >> >
>>> >> > > Marvell shipping broken firmware is completely unpublicised or
>>> >> > > acknowledged, however updated firmware is available through your
>>> >> > > motherboard vendor, so all hope it not lost after all...
>>> >> >
>>> >> > Unfortunately not through all vendors. :( Or maybe not *yet*?
>>> >> >
>>> >> > > My 8053/EC is using firmware 2.2 (previously 1.9) - you can check in
>>> >> > > DOS with 'yukondg.exe' from
>>> >> > > http://www.marvell.com/drivers/files/yukondg_v6.53.4.3.zip .
>>> >> >
>>> >> > Stephen, is it possible for sky2 driver to check and print the
>>> >> > firmware version?
>>> >>
>>> >> It is possible to dump and program eeprom from linux, and the version
>>> >> maybe buried in the VPD, but some hardware has non-functional VPD.
>>> >
>>> > The revision number in the VPD matches the revision number (from PCI)
>>> > that is already printed by the driver. Since I don't have enough documentation
>>> > on the firmware to know the format, there is no way to print anything
>>> > really useful.
>>>
>>> $ lspci -d11ab:
>>> 02:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8053
>>> PCI-E Gigabit Ethernet Controller (rev 22)
>>> 07:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8036
>>> PCI-E Fast Ethernet Controller (rev 15)
>>>
>>> $ dmesg | grep sky2.*rev
>>> [   33.998133] sky2 0000:02:00.0: v1.20 addr 0xd6000000 irq 16
>>> Yukon-EC (0xb6) rev 2
>>> [   33.999085] sky2 0000:07:00.0: v1.20 addr 0xd8000000 irq 18
>>> Yukon-FE (0xb7) rev 1
>>>
>>> The 'rev' reported by the driver is from:
>>>
>>>         hw->chip_rev = (sky2_read8(hw, B2_MAC_CFG) & CFG_CHIP_R_MSK) >> 4;
>>>
>>> which is the chip/silicon revision, whereas the rev reported by the
>>> PCI device is the firmware revision, as you've found.
>>>
>>> How about:
>>>
>>> 02:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8053
>>> PCI-E Gigabit Ethernet Controller (silicon rev 2, firmware/PCI rev 22)
>>>
>>> This introduces perhaps something even more useful - we know there are
>>> really major issues with (eg) 88E8053 before firmware 2.2; is it worth
>>> adding a warning, such as:
>>>
>>> sky2: firmware earlier than rev 2.2 detected - if you experience
>>> transmit timeouts, ask vendor for 88E8053 firmware update
>>>
>>> There are a lot of bug reports on the web pointing the finger at the
>>> sky2 driver, and waiting for a 'fixed' version, but the issue is
>>> elsewhere...part of the problem, if there is (almost) no
>>> documentation/evidence of this problem posted anywhere.
>>>
>>> Many thanks,
>>>   Daniel
>>
>> lspci reports PCI rev.  driver reports chip rev.
>
> But the PCI revision is also the firmware revision - I didn't know
> this until you mentioned it, so it would be handy to make it explicit
> in the driver's output; also we don't always have the 'lspci' output
> in bug reports. Let me know if you'd like a patch...

Here [1] it is.

Signed-off-by: Daniel J Blueman <daniel.blueman@...il.com>
Description: Differentiate and print PCI and firmware revision against
silicon/chip revision

--- [1]

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index e24b25c..676a22f 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -4206,6 +4206,7 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
 	struct sky2_hw *hw;
 	int err, using_dac = 0, wol_default;
 	char buf1[16];
+	unsigned char pci_rev;

 	err = pci_enable_device(pdev);
 	if (err) {
@@ -4276,10 +4277,11 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
 	if (err)
 		goto err_out_iounmap;

-	dev_info(&pdev->dev, "v%s addr 0x%llx irq %d Yukon-2 %s rev %d\n",
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &pci_rev);
+	dev_info(&pdev->dev, "v%s addr 0x%llx irq %d Yukon-2 %s silicon rev
%d, firmware/PCI rev %d\n",
 		 DRV_VERSION, (unsigned long long)pci_resource_start(pdev, 0),
 		 pdev->irq, sky2_name(hw->chip_id, buf1, sizeof(buf1)),
-		 hw->chip_rev);
+		 hw->chip_rev, pci_rev);

 	sky2_reset(hw);

-- 
Daniel J Blueman
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists