I've got a headless system. It's a testing x86. It has got a "Silicon Integrated Systems [SiS] SiS900 PCI Fast Ethernet (rev 90)" ethernet controller. It's always worked fine with 2.6.27 (latest gentoo revision). I've updated to 2.6.28 .. (make silentoldconfig) and now networking doesn't work anymore. dhcpcd waits and then assigns "an old ip from some assumed cache file" (that has nothing to do with my 192.168.* LAN but is something else like 167 or something) Also manual configuration doesn't work anymore. On the other hand the same static config works if I load the 2.6.27 kernel. net.eth0 is recognized also by 2.6.28, though. And but it is fired up at bootup, also. Indeed the full-duplex link appears in the 2.6.28 kernel log (as it appeared in 2.6.27).
please check out http://bugzilla.kernel.org/show_bug.cgi?id=11649 can you report back your mac address for both kernel versions? thanks
Is your card recognised from your system? Can you see your card if you ran ifconfig -a ? Could you post a full dmesg log *after* you try to assign an IP to your card with dhcp?
There were no code changes to sis900.c or sis900.h between 2.6.27 and 2.6.28, so this is most likely a problem unrelated to the driver.. need the info requested above to proceed further.
(In reply to comment #1) > please check out http://bugzilla.kernel.org/show_bug.cgi?id=11649 > can you report back your mac address for both kernel versions? > thanks > MAC is recognized ok, now there's also a valid IP (thanks to static configuration). But servers don't work. I've caught a watchdog sis900 error in dmesg. (In reply to comment #2) > Is your card recognised from your system? Can you see your card if you ran > ifconfig -a ? > > Could you post a full dmesg log *after* you try to assign an IP to your card > with dhcp? > After dhcpcd eth0: eth0: Media Link On 100mbps full-duplex ------------[ cut here ]------------ WARNING: at net/sched/sch_generic.c:226 dev_watchdog+0xf2/0x143() NETDEV WATCHDOG: eth0 (sis900): transmit timed out Pid: 0, comm: swapper Tainted: G W 2.6.28-gentoo #1 Call Trace: [<c0116491>] warn_slowpath+0x55/0x73 [<c01bd4d6>] __prop_inc_percpu_max+0x22/0x67 [<c0147649>] virt_to_head_page+0x2a/0x35 [<c015f991>] end_bio_bh_io_sync+0x0/0x34 [<c0106c1f>] read_tsc+0x6/0x22 [<c0106c1f>] read_tsc+0x6/0x22 [<c0126cd5>] getnstimeofday+0x32/0xaf [<c010d3af>] lapic_next_event+0x10/0x13 [<c01292d4>] clockevents_program_event+0xbb/0xc9 [<c01bebe4>] strlcpy+0x11/0x3d [<c0228f70>] dev_watchdog+0xf2/0x143 [<c0106c1f>] read_tsc+0x6/0x22 [<c0126cd5>] getnstimeofday+0x32/0xaf [<c010d3af>] lapic_next_event+0x10/0x13 [<c01292d4>] clockevents_program_event+0xbb/0xc9 [<c0228e7e>] dev_watchdog+0x0/0x143 [<c011c628>] run_timer_softirq+0x101/0x145 [<c011991e>] __do_softirq+0x4a/0xbc [<c01199b2>] do_softirq+0x22/0x26 [<c0119a68>] irq_exit+0x25/0x55 [<c010da5f>] smp_apic_timer_interrupt+0x5c/0x64 [<c01036a8>] apic_timer_interrupt+0x28/0x30 [<c0107713>] default_idle+0x28/0x3b [<c0101e10>] cpu_idle+0x41/0x65 ---[ end trace 4eaa2a86a8e2da22 ]--- eth0: Transmit timeout, status 00000004 00000249 eth0: Transmit timeout, status 00000004 00000249 eth0: Transmit timeout, status 00000004 00000249 eth0: Transmit timeout, status 00000004 00000249 eth0: Transmit timeout, status 00000004 00000240 eth0: Transmit timeout, status 00000004 00000240 eth0: Transmit timeout, status 00000004 00000240 eth0: Transmit timeout, status 00000004 00000240 eth0: Transmit timeout, status 00000004 00000240
OK, lets get this filed upstream. Please try to reproduce this on vanilla-sources-2.6.28.
(In reply to comment #5) > OK, lets get this filed upstream. Please try to reproduce this on > vanilla-sources-2.6.28. > Can you tell me how? I'll install vanilla-sources and copy over the .config Will it work even without patches? I'll use grub fallback just in case. Then what should I do? Post dmesg?
Yes that's right
Created attachment 178101 [details] 2.6.28 vanilla kernel dmesg after trying to assign a ip though dhcpcd As said, the IP had already been manually set, but I can't reach the machine. ifconfig (same as ifconfig -a) eth0 Link encap:Ethernet HWaddr OK inet addr:192.168.OK Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B) Interrupt:22 Base address:0xa000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Great, thanks for testing. Please now file an upstream networking bug for this at http://bugzilla.kernel.org. State that it is a 2.6.28 regression (2.6.27 worked) and attach your last dmesg output, mentioning that the error is included inside. Please post the new bug URL here when done, thanks!
(In reply to comment #9) > Great, thanks for testing. Please now file an upstream networking bug for this > at http://bugzilla.kernel.org. State that it is a 2.6.28 regression (2.6.27 > worked) and attach your last dmesg output, mentioning that the error is > included inside. Please post the new bug URL here when done, thanks! > It's http://bugzilla.kernel.org/show_bug.cgi?id=12432 Thanks for your advices!
Thanks, will track that
invalid, was a user configuration error (disabled PCI quirks)
