sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Thomas Glanzmann]

Hello,
my sky2 network card in my intel mac mini just stopped working again on
me. After a reboot it worked again. This time there is no dmesg output
related to the problem. :-( Am I the only one who sees that?

        Thomas

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 809 bytes --]

Am Freitag, den 01.09.2006, 20:41 +0200 schrieb Thomas Glanzmann:
> Hello,
> my sky2 network card in my intel mac mini just stopped working again on
> me. After a reboot it worked again. This time there is no dmesg output
> related to the problem. :-( Am I the only one who sees that?

Nope, same here on an Asus P5W DH Deluxe mainboard. The sky2 NIC just
silently dies after some time. Rmmod + modprobe sky2 used to re-enable
the NIC IIRC. Since this bug makes the driver practically unusable I
have since switched to a PCI NIC (which is a shame considering the 2
gigabit sky2 NICs on the mainboard...).


-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sat, 2 Sep 2006, Matthias Hentges wrote:

> Am Freitag, den 01.09.2006, 20:41 +0200 schrieb Thomas Glanzmann:
> > Hello,
> > my sky2 network card in my intel mac mini just stopped working again on
> > me. After a reboot it worked again. This time there is no dmesg output
> > related to the problem. :-( Am I the only one who sees that?
>
> Nope, same here on an Asus P5W DH Deluxe mainboard. The sky2 NIC just
> silently dies after some time. Rmmod + modprobe sky2 used to re-enable
> the NIC IIRC. Since this bug makes the driver practically unusable I
> have since switched to a PCI NIC (which is a shame considering the 2
> gigabit sky2 NICs on the mainboard...).

Has this not been fixed in the 2.6.18 git?

>
>
> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: H 0

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 1066 bytes --]

Am Freitag, den 01.09.2006, 20:57 -0400 schrieb shogunx:
> On Sat, 2 Sep 2006, Matthias Hentges wrote:
> 
> > Am Freitag, den 01.09.2006, 20:41 +0200 schrieb Thomas Glanzmann:
> > > Hello,
> > > my sky2 network card in my intel mac mini just stopped working again on
> > > me. After a reboot it worked again. This time there is no dmesg output
> > > related to the problem. :-( Am I the only one who sees that?
> >
> > Nope, same here on an Asus P5W DH Deluxe mainboard. The sky2 NIC just
> > silently dies after some time. Rmmod + modprobe sky2 used to re-enable
> > the NIC IIRC. Since this bug makes the driver practically unusable I
> > have since switched to a PCI NIC (which is a shame considering the 2
> > gigabit sky2 NICs on the mainboard...).
> 
> Has this not been fixed in the 2.6.18 git?

Good question. I'll try 2.6.18-rc4-mm3 and report back.
-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

> >
> > Has this not been fixed in the 2.6.18 git?
>
> Good question. I'll try 2.6.18-rc4-mm3 and report back.

I am having no problems with 2.6.18-rc5, which I just built and tested.

> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: H 0

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 536 bytes --]

Am Freitag, den 01.09.2006, 22:41 -0400 schrieb shogunx:
> > >
> > > Has this not been fixed in the 2.6.18 git?
> >
> > Good question. I'll try 2.6.18-rc4-mm3 and report back.
> 
> I am having no problems with 2.6.18-rc5, which I just built and tested.

The NIC is up and running for about 9hrs now w/ -rc4-mm3, thanks for the
heads up!
-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Stephen Hemminger]

Matthias Hentges wrote:
> Am Freitag, den 01.09.2006, 22:41 -0400 schrieb shogunx:
>   
>>>> Has this not been fixed in the 2.6.18 git?
>>>>         
>>> Good question. I'll try 2.6.18-rc4-mm3 and report back.
>>>       
>> I am having no problems with 2.6.18-rc5, which I just built and tested.
>>     
>
> The NIC is up and running for about 9hrs now w/ -rc4-mm3, thanks for the
> heads up!
>   

My theory still unproven, is that there is a problem with transmit flow 
control and alignment. I have
no direct relation to Marvell, and they tell me nothing about the 
hardware bugs. But it took two
weeks to find a problem where receive flow control was busted when the 
receive buffer was not
aligned on 8 byte boundary. The receiver would stop and not resume. To 
workaround, the driver
ensures alignment of receive buffers. The only clue was that the receive 
DMA FIFO always
had a partial count left in it.

There may well be a similar problem on transmit; since driver has no 
control over transmit buffer
alignment, fixing it would require copying most transmit data, or a hack 
all the way up in protocols.
Maybe if the driver lied about the hard header length, it could fool the 
protocols.

Probably better just to always negotiate no transmit flow control.

-- 
VGER BF report: H 0.232987

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sat, 2 Sep 2006, Matthias Hentges wrote:

> Am Freitag, den 01.09.2006, 22:41 -0400 schrieb shogunx:
> > > >
> > > > Has this not been fixed in the 2.6.18 git?
> > >
> > > Good question. I'll try 2.6.18-rc4-mm3 and report back.
> >
> > I am having no problems with 2.6.18-rc5, which I just built and tested.
>
> The NIC is up and running for about 9hrs now w/ -rc4-mm3, thanks for the
> heads up!

Hey, no worries.  I have a friend who has has that problem for some time,
and I just got one of those cards myself, albeint in an ExpressCard
format.

Glad its working.

Scott


> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: H 1.37565e-05

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 1113 bytes --]

Am Samstag, den 02.09.2006, 15:41 -0400 schrieb shogunx:
> On Sat, 2 Sep 2006, Matthias Hentges wrote:
> 
> > Am Freitag, den 01.09.2006, 22:41 -0400 schrieb shogunx:
> > > > >
> > > > > Has this not been fixed in the 2.6.18 git?
> > > >
> > > > Good question. I'll try 2.6.18-rc4-mm3 and report back.
> > >
> > > I am having no problems with 2.6.18-rc5, which I just built and tested.
> >
> > The NIC is up and running for about 9hrs now w/ -rc4-mm3, thanks for the
> > heads up!
> 
> Hey, no worries.  I have a friend who has has that problem for some time,
> and I just got one of those cards myself, albeint in an ExpressCard
> format.
> 
> Glad its working.

Well, it just crapped out on me again :(

Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
out
Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing

Only a rmmod / modprobe cycle helps at this point.
-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sat, 2 Sep 2006, Matthias Hentges wrote:

> Am Samstag, den 02.09.2006, 15:41 -0400 schrieb shogunx:
> > On Sat, 2 Sep 2006, Matthias Hentges wrote:
> >
> > > Am Freitag, den 01.09.2006, 22:41 -0400 schrieb shogunx:
> > > > > >
> > > > > > Has this not been fixed in the 2.6.18 git?
> > > > >
> > > > > Good question. I'll try 2.6.18-rc4-mm3 and report back.
> > > >
> > > > I am having no problems with 2.6.18-rc5, which I just built and tested.
> > >
> > > The NIC is up and running for about 9hrs now w/ -rc4-mm3, thanks for the
> > > heads up!
> >
> > Hey, no worries.  I have a friend who has has that problem for some time,
> > and I just got one of those cards myself, albeint in an ExpressCard
> > format.
> >
> > Glad its working.
>
> Well, it just crapped out on me again :(
>
> Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> out
> Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
>
> Only a rmmod / modprobe cycle helps at this point.

Really?  What is the error condition causing it?  On my friends lap, which
has an integrated sky2, his drops out with a full sustained TX...
uploading to another box for example, at about 4-8MB of transfer.  The
fix in his case is ifdown eth0 && ifup eth0.  I have
yet to see the error occur at all on my ExpressCard device, either with
2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
cannot get it to fail under any conditions.


> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: U 0.5

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 1254 bytes --]

Am Samstag, den 02.09.2006, 19:11 -0400 schrieb shogunx:
> On Sat, 2 Sep 2006, Matthias Hentges wrote:

> > Well, it just crapped out on me again :(
> >
> > Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> > out
> > Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
> >
> > Only a rmmod / modprobe cycle helps at this point.
> 
> Really?  What is the error condition causing it?  On my friends lap, which
> has an integrated sky2, his drops out with a full sustained TX...
> uploading to another box for example, at about 4-8MB of transfer.  The
> fix in his case is ifdown eth0 && ifup eth0.  I have
> yet to see the error occur at all on my ExpressCard device, either with
> 2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
> cannot get it to fail under any conditions.
> 

I have yet to find a reproduceable way to trigger the bug but I'll try a
few things tomorrow.
Currently it appears to be completely ranom. I've loaded the driver w/
debug=10, maybe it'll give some clues.
-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sun, 3 Sep 2006, Matthias Hentges wrote:

> Am Samstag, den 02.09.2006, 19:11 -0400 schrieb shogunx:
> > On Sat, 2 Sep 2006, Matthias Hentges wrote:
>
> > > Well, it just crapped out on me again :(
> > >
> > > Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> > > out
> > > Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
> > >
> > > Only a rmmod / modprobe cycle helps at this point.
> >
> > Really?  What is the error condition causing it?  On my friends lap, which
> > has an integrated sky2, his drops out with a full sustained TX...
> > uploading to another box for example, at about 4-8MB of transfer.  The
> > fix in his case is ifdown eth0 && ifup eth0.  I have
> > yet to see the error occur at all on my ExpressCard device, either with
> > 2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
> > cannot get it to fail under any conditions.
> >
>
> I have yet to find a reproduceable way to trigger the bug but I'll try a
> few things tomorrow.
> Currently it appears to be completely ranom. I've loaded the driver w/
> debug=10, maybe it'll give some clues.

Ack.  Awaiting more info.  I pushed it pretty hard last night with both
kernel revisions, scp'ing cd iso images and kernel tarballs back and forth
across the interface, and could not get it to lock.  I am using a 88E8053
chipset.  I'll ask my friend what chipset his is.  Perhaps its a
different bug that is hitting you now...

> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: U 0.5

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sat, 2 Sep 2006, shogunx wrote:

> On Sun, 3 Sep 2006, Matthias Hentges wrote:
>
> > Am Samstag, den 02.09.2006, 19:11 -0400 schrieb shogunx:
> > > On Sat, 2 Sep 2006, Matthias Hentges wrote:
> >
> > > > Well, it just crapped out on me again :(
> > > >
> > > > Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> > > > out
> > > > Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
> > > >
> > > > Only a rmmod / modprobe cycle helps at this point.
> > >
> > > Really?  What is the error condition causing it?  On my friends lap, which
> > > has an integrated sky2, his drops out with a full sustained TX...
> > > uploading to another box for example, at about 4-8MB of transfer.  The
> > > fix in his case is ifdown eth0 && ifup eth0.  I have
> > > yet to see the error occur at all on my ExpressCard device, either with
> > > 2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
> > > cannot get it to fail under any conditions.
> > >
> >
> > I have yet to find a reproduceable way to trigger the bug but I'll try a
> > few things tomorrow.
> > Currently it appears to be completely ranom. I've loaded the driver w/
> > debug=10, maybe it'll give some clues.
>
> Ack.  Awaiting more info.  I pushed it pretty hard last night with both
> kernel revisions, scp'ing cd iso images and kernel tarballs back and forth
> across the interface, and could not get it to lock.  I am using a 88E8053
> chipset.  I'll ask my friend what chipset his is.

He is using a 88E8036, JFYI.


Perhaps its a
> different bug that is hitting you now...
>
> > --
> > Matthias 'CoreDump' Hentges
> >
> > Webmaster of hentges.net and OpenZaurus developer.
> > You can reach me in #openzaurus on Freenode.
> >
> > My OS: Debian SID. Geek by Nature, Linux by Choice
> >
>
> sleekfreak pirate broadcast
> http://sleekfreak.ath.cx:81/
>
>
> --
> VGER BF report: U 0.5
> -
> 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/
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: U 0.500006

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Denis Vlasenko]

On Sunday 03 September 2006 01:49, shogunx wrote:
> > > > Well, it just crapped out on me again :(
> > > >
> > > > Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> > > > out
> > > > Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
> > > >
> > > > Only a rmmod / modprobe cycle helps at this point.
> > >
> > > Really?  What is the error condition causing it?  On my friends lap, which
> > > has an integrated sky2, his drops out with a full sustained TX...
> > > uploading to another box for example, at about 4-8MB of transfer.  The
> > > fix in his case is ifdown eth0 && ifup eth0.  I have
> > > yet to see the error occur at all on my ExpressCard device, either with
> > > 2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
> > > cannot get it to fail under any conditions.
> > >
> >
> > I have yet to find a reproduceable way to trigger the bug but I'll try a
> > few things tomorrow.
> > Currently it appears to be completely ranom. I've loaded the driver w/
> > debug=10, maybe it'll give some clues.
> 
> Ack.  Awaiting more info.  I pushed it pretty hard last night with both
> kernel revisions, scp'ing cd iso images and kernel tarballs back and forth
> across the interface, and could not get it to lock.  I am using a 88E8053
> chipset.  I'll ask my friend what chipset his is.  Perhaps its a
> different bug that is hitting you now...

scp isn't "pushing very hard". It takes some time to do ssh crypto
and even if your CPU is fast enough for that to be not an issue,
scp is using TCP, which is _designed_ to not saturate links to 100.00%.

Give it a real hard beating with uni- and bidirectional UDP netcat flood! :)
--
vda

-- 
VGER BF report: U 0.511791

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

On Sun, 3 Sep 2006, Denis Vlasenko wrote:

> On Sunday 03 September 2006 01:49, shogunx wrote:
> > > > > Well, it just crapped out on me again :(
> > > > >
> > > > > Sep  2 23:36:13 localhost kernel: NETDEV WATCHDOG: eth2: transmit timed
> > > > > out
> > > > > Sep  2 23:36:13 localhost kernel: sky2 hardware hung? flushing
> > > > >
> > > > > Only a rmmod / modprobe cycle helps at this point.
> > > >
> > > > Really?  What is the error condition causing it?  On my friends lap, which
> > > > has an integrated sky2, his drops out with a full sustained TX...
> > > > uploading to another box for example, at about 4-8MB of transfer.  The
> > > > fix in his case is ifdown eth0 && ifup eth0.  I have
> > > > yet to see the error occur at all on my ExpressCard device, either with
> > > > 2.6.18-rc5 or 2.6.17.5.  I built the rc5 as a preemptive measure, but I
> > > > cannot get it to fail under any conditions.
> > > >
> > >
> > > I have yet to find a reproduceable way to trigger the bug but I'll try a
> > > few things tomorrow.
> > > Currently it appears to be completely ranom. I've loaded the driver w/
> > > debug=10, maybe it'll give some clues.
> >
> > Ack.  Awaiting more info.  I pushed it pretty hard last night with both
> > kernel revisions, scp'ing cd iso images and kernel tarballs back and forth
> > across the interface, and could not get it to lock.  I am using a 88E8053
> > chipset.  I'll ask my friend what chipset his is.  Perhaps its a
> > different bug that is hitting you now...
>
> scp isn't "pushing very hard". It takes some time to do ssh crypto
> and even if your CPU is fast enough for that to be not an issue,
> scp is using TCP, which is _designed_ to not saturate links to 100.00%.

thats the condition that caused the error on my friends box... scp a large
file up over the interface in question, and watch the link dump.

>
> Give it a real hard beating with uni- and bidirectional UDP netcat flood! :)

well, there is that.  i'll give it that pounding and see what happens.

> --
> vda
>
> --
> VGER BF report: U 0.511791
> -
> 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/
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/


-- 
VGER BF report: U 0.482991

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[shogunx]

>
> I have yet to find a reproduceable way to trigger the bug but I'll try a
> few things tomorrow.
> Currently it appears to be completely ranom. I've loaded the driver w/
> debug=10, maybe it'll give some clues.

Seen that error again?  I've done everything I can think of to get this
interface to fail, and I just can't do it.

> --
> Matthias 'CoreDump' Hentges
>
> Webmaster of hentges.net and OpenZaurus developer.
> You can reach me in #openzaurus on Freenode.
>
> My OS: Debian SID. Geek by Nature, Linux by Choice
>

sleekfreak pirate broadcast
http://sleekfreak.ath.cx:81/

Re: sky2 hangs on me again: This time 200 kb/s IPv4 traffic, not easily reproducable

[Matthias Hentges]

[-- Attachment #1: Type: text/plain, Size: 726 bytes --]

Am Montag, den 04.09.2006, 13:24 -0400 schrieb shogunx:
> >
> > I have yet to find a reproduceable way to trigger the bug but I'll try a
> > few things tomorrow.
> > Currently it appears to be completely ranom. I've loaded the driver w/
> > debug=10, maybe it'll give some clues.
> 
> Seen that error again?  I've done everything I can think of to get this
> interface to fail, and I just can't do it.
> 

After running almost a full day w/o problems, it freaked out on me when
debugging was disabled after a reboot *sigh*.
-- 
Matthias 'CoreDump' Hentges 

Webmaster of hentges.net and OpenZaurus developer.
You can reach me in #openzaurus on Freenode.

My OS: Debian SID. Geek by Nature, Linux by Choice

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]