On Wed, 2004-04-28 at 12:00, Thomas Quinot wrote: > * Stefan Ehmann, 2004-04-14 : > > > Since latest cvsup kernel doesn't boot about half of the time. Without > > ATAPICAM it just boots fine every time. > > Can you show a kernel backtrace at the point where the kernel hangs? > > > (probe0:ata0:0:0:0): error 22 > > (probe1:ata0:0:1:0): error 22 > > Do you hava anything connected to ata0? First of all, my first analysis turned out to be pretty wrong: The problem also happens with older kernels and also without atapicam. Though with atapicam disabled it seems to be much less likely. atacontrol list gives this: ATA channel 0: Master: ad0 <ExcelStor Technology J340/V22OA63A> ATA/ATAPI revision 6 Slave: ad1 <ST340823A/3.39> ATA/ATAPI revision 5 ATA channel 1: Master: acd0 <CR-4804TE/3.0D> ATA/ATAPI revision 0 Slave: acd1 <JLMS XJ-HD166S/DS1A> ATA/ATAPI revision 5 I changed master/slave of the cdrom devices which seemed to fix it at first but turned out to just improve chance of successfull booting. trace in ddb shows this after manual break to debugger scgetc() sckbdevent() atkdb_intr() atkbd_isa_intr() ithread_loop() fork_exit() fork_trampoline() A photo with a bit more detail can be found here (to save me frome typing all those numbers): http://stud4.tuwien.ac.at/~e0125637/fbsd/hang.jpgReceived on Wed Apr 28 2004 - 21:28:33 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:52 UTC