Francesco Gringoli wrote: > On Jul 23, 2009, at 4:05 AM, Larry Finger wrote: > >> Francesco Gringoli wrote: >>> Larry, I think this could be one of the causes of the malfunctioning you >>> reported before. If you have some time (and indeed if you feel like >>> doing it :-) ) please test this firmware, it will be great. >> >> Francesco, >> >> The system ran about 30 minutes, then crashed. I missed the first >> oops, but caught a kernel panic with formal traceback on my i386 system: >> >> b43_dma_handle_txstatus + 0x1ee/0x2fa >> b43_handle_txstatus + 0x45/0x52 >> >> The call in b43_dma_handle_status is at line 1405: >> >> unmap_descbuffer(ring, meta->dmaaddr, meta->skb->len, >> 1); >> >> The oops was in drivers/net/wireless/b43/xmit.h:171 in the call to >> b43_is_old_txhdr_format(). It appears that dev->fw.rev causes the oops. >> >> As usual, I was running an infinite loop of tcpperf in one console and >> a flood ping in a second. >> >> Larry > Hi Larry, > > Many many thanks. > > As I focussed almost only on the firmware side during these last months, > I forgot to upgrade the kernel :-) . Which version have you used to do > the test? Are you still using the PCcard32 (rev11/a)? I will try this > summer to replicate precisely your setup. I also think it is time to use > the support provided by Michael for telling the kernel which features > are present.
The dump traceback was for 2.6.31-rc1-wl. The card is a Cardbus version of the BCM4318 on i386 architecture. I have now upgraded the kernel for that machine to 2.6.31-rc3-wl, but as there have been no recent changes in b43, I expect the same result. I will be adding some diagnostics to the interrupt routine to detect bad values and get better information for your debugging. Just knowint that there was a "fatal exception in interrupt" crash isn't very helpful. I will also test with my PCIe version of a BCM4311, which also uses rev 5 firmware. That will be on x86_64 architecture. Larry _______________________________________________ Bcm43xx-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/bcm43xx-dev
