Re: msk watchdog timeout

On Mon, Oct 01, 2007 at 03:39:47PM +0800, Kudo Chien wrote:
 > Hi,
 > 
 > Once again, I had the msk watchdog timeout problem under moderate load. I've
 > surveyed msk related posts before.
 > There are many reports and solutions, but they seems not help for me.
 > 
 > Whenever I try to get a FreeBSD iso file through ftp (transfer rate is about
 > 3 MB/s), msk would be down. Some messages from dmesg are like:
 > msk0: watchdog timeout
 > msk0: link state changed to DOWN
 > msk0: link state changed to UP
 > mskc0: Tx descriptor error
 > mskc0: Tx descriptor error
 > msk0: watchdog timeout
 > msk0: link state changed to DOWN
 > msk0: link state changed to UP
 > 
 > I used FreeBSD 6.2-STABLE before, since there is some related patch for msk
 > at FreeBSD-Current (if_msk.c rev. 1.18). I upgrade to Current, but the
 > problem seems not solved. Certainly, I tried to add hw.msk.legacy_intr="1"
 > in /boot/loader.conf. Whether disabling  MSI or not, the problem is as
 > before. In order to avoid shared interrupt, I've also disabled USB device in
 > BIOS, but not help.
 > 
 > Could you please tell me how to solve this problem.
 > Thanks a lot
 > 
 > ---
 > My host is Intel Core 2 Duo E6300 + Onboard Marvel 8056 Gigabit LAN
 > Controller (My MB is GA-965G-DS3 of Gigabyte,
 > http://www.gigabyte.com.tw/Products/Motherboard/Products_Spec.aspx?ClassValue=Motherboard&ProductID=2417&ProductName=GA-965G-DS3
 > )
 > 
 > uname:
 > 7.0-CURRENT FreeBSD 7.0-CURRENT #2: Thu Sep 27 00:11:01 CST 2007
 > root_at_localhost:/usr/obj/usr/src/sys/SMP  i386
 > 
 > ifconfig:
 > msk0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
 >         options=9a<TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM>
 >         ether 00:16:e6:de:bf:ed
 >         inet x.x.x.x netmask 0xffffff00 broadcast x.x.x.x
 >         media: Ethernet autoselect (1000baseTX <full-duplex,flag2>)
 >         status: active
 > 
 > boot message:
 > msk0: <Marvell Technology Group Ltd. Yukon EC Ultra Id 0xb4 Rev 0x02> on
 > mskc0
 > msk0: Ethernet address: 00:16:e6:de:bf:ed
 > miibus0: <MII bus> on msk0
 > e1000phy0: <Marvell 88E1149 Gigabit PHY> PHY 0 on miibus0
 > e1000phy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX-FDX,
 > auto
 > mskc0: [ITHREAD]
 > 
 > vmstat -i:
 > interrupt                          total       rate
 > irq1: atkbd0                           6          0
 > irq6: fdc0                             1          0
 > irq14: ata0                        97074          0
 > irq15: ata1                          629          0
 > irq16: mskc0                      288008          1
 > irq19: atapci0+                     1522          0
 > cpu0: timer                    421587304       1922
 > cpu1: timer                    421586922       1922
 > Total                          843561466       3846
 > 
 > 
 > pciconf -lv:
 > mskc0_at_pci3:0:0: class=0x020000 card=0xe0001458 chip=0x436411ab rev=0x14
 > hdr=0x00
 >     vendor     = 'Marvell Semiconductor (Was: Galileo Technology Ltd)'
 >     class      = network
 >     subclass   = ethernet
 > 

It seems that you have Yukon EC Ultra and 88E1149 PHY. Due to lack of
hardwares Yukon EC Ultra was not tested by me and it may have bugs.
It's also known that 88E1149 PHY has some issues which were not solved
yet. I have a WIP version that may fix Yukon EC Ultra issues but I'm
not sure whether it would work or not. Would you try attached patch?

-- 
Regards,
Pyun YongHyeon