[Aurora-sparc-user] Happy Meal failure or kernel bug?

[Rafal Maszkowski]

On Fri, Nov 25, 2005 at 10:23:43AM +0000, Christian Joensson wrote:
> On 11/24/05, Christian Joensson <christian.joensson at gmail.com> wrote:
> > On 11/24/05, Dominik 'Rathann' Mierzejewski <D.Mierzejewski at icm.edu.pl> wrote:
> > > Nov 24 13:37:41 szkuner kernel: happy meal: Fry guys.SABRE0: PCI SERR signal asserted.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Error interrupt for happy meal, status = ffffffff
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal BigMAC SQE test failed.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal transmitter FIFO underrun, DMA error.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal MAX Packet size error.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal out of receive descriptors, packet dropped.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal rx DMA errors [ GenericError ParityError RxTagBotch ]
> > > Nov 24 13:37:41 szkuner kernel: eth0: EOP not set in happy meal transmit descriptor!
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal MIF interrupt.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal tx DMA errors [ GenericError LateError ParityErro TagBotch ]
> > > Nov 24 13:37:41 szkuner kernel: eth0: Happy Meal register access SBUS slave (parity) error.
> > > Nov 24 13:37:41 szkuner kernel: eth0: Resetting...
> > > repeated over and over, making the box crawl to a stop. I can't even
> > > log in. Rebooting fixes the problem.
> > > Could this be a bug in 2.0beta kernel or is this a sign of decaying
> > > hardware? My coworker mentioned freezes after extended periods of uptime.
> > > It was running Corona at the time.
> > > Ah, the box is one of my U10s: 440MHz, 256MB RAM, 9GB HD and onboard
> > > ATI.
> > hmm, odd, I used to be able to hang my U1 system, with hme, during
> > heavy network activity, such as downloading an insallation iso over
> > the network... I don't seem to be able to do that anymore, using
> > 2.0b2...
> > any good news on this old issue?
> there, got it using svn update on the gcc tree :)
> Nov 25 18:08:12 u1 kernel: NETDEV WATCHDOG: eth0: transmit timed out
> Nov 25 18:08:12 u1 kernel: eth0: transmit timed out, resetting
> Nov 25 18:08:12 u1 kernel: eth0: Happy Status 03030000 TX[000003ff:00000301]
> Nov 25 18:08:15 u1 kernel: eth0: Link is up using internal transceiver
> at 100Mb/s, Full Duplex.
> Nov 25 18:08:35 u1 kernel: nfs: server fw not responding, still trying

Recently I use Ultra 1 with 2.6.16-1.2241sp1 kernel and get Ethernet lockups
with messages about timeout, similar to the ones cited above. Any hope for me?
I can try to use some SBUS SCSI-Ethernet card if it may help.

R.
-- 
Jeżeli upadnie cywilizacja zachodnia, to będzie szansa na odbudowę cywilizacji
      chrześcijańskiej. (dr Stanisław Krajski, Radio Maryja, 10 X 2001)