On Saturday 26 March 2005 06:09 pm, Robert Watson wrote: > On Sat, 26 Mar 2005, Robert Watson wrote: > > Got the following panic while doing a buildworld on a dual xeon (quad > > logical processor) system today, using 6-CURRENT source from today. > > Actually, it looks like this is a false positive -- a race introduced in > local code because the debugging code to check for multiple frees (etc) > was getting called after mutexes/etc were released, so raced against > real-world re-allocation. Could there be a race in the current code as well? I've seen several instances of this with the Mbuf zone recently and there are other reports on the list of similar panics with other zones. In my panics, it all seems to have to do with fifo's taking to the socket code and the mbuf's being recycled. > Robert N M Watson > > > Robert N M Watson > > > > Slab at 0xc5857200, freei 2 = 0. > > panic: Duplicate free of item 0xc5862800 from zone 0xc55f8580(NAMEI) > > > > cpuid = 1 > > KDB: enter: panic > > [thread pid 10534 tid 100207 ] > > Stopped at kdb_enter+0x2b: nop > > db> trace > > Tracing pid 10534 tid 100207 td 0xc5e27000 > > kdb_enter(c08317e5) at kdb_enter+0x2b > > panic(c084b549,c5862800,c55f8580,c0839cfb,c084b52d) at panic+0x127 > > uma_dbg_free(c55f8580,0,c5862800) at uma_dbg_free+0x110 > > uma_zfree_arg(c55f8580,c5862800,0) at uma_zfree_arg+0xf4 > > namei(ebb3fbdc,c0626abe,c5e27000,c5e27000,c5e27000) at namei+0x245 > > vn_open_cred(ebb3fbdc,ebb3fcdc,1a4,c5ddec80,3) at vn_open_cred+0x278 > > vn_open(ebb3fbdc,ebb3fcdc,1a4,3,c0626abe) at vn_open+0x1e > > kern_open(c5e27000,8418900,0,8001,1b6) at kern_open+0xb6 > > open(c5e27000,ebb3fd14,3,b,202) at open+0x1a > > syscall(2f,2f,bfbf002f,0,0) at syscall+0x213 > > Xint0x80_syscall() at Xint0x80_syscall+0x1f > > --- syscall (5, FreeBSD ELF32, open), eip = 0x82d0e27, esp = 0xbfbfe2fc, > > ebp = 0xbfbfe318 --- > > db> show lockedvnods > > Locked vnodes > > > > 0xc61819fc: tag ufs, type VREG > > usecount 1, writecount 0, refcount 4 mountedhere 0 > > flags () > > v_object 0xc6174738 ref 0 pages 5 > > lock type ufs: EXCL (count 1) by thread 0xc5e27a10 (pid 10541) > > ino 2028972, on dev ar0s1e > > > > 0xc6b5fd50: tag ufs, type VREG > > usecount 1, writecount 0, refcount 3 mountedhere 0 > > flags () > > v_object 0xc76d3528 ref 0 pages 1 > > lock type ufs: EXCL (count 1) by thread 0xc624aa10 (pid 10538) > > ino 2028317, on dev ar0s1e > > > > > > _______________________________________________ > > freebsd-current_at_freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-current > > To unsubscribe, send any mail to > > "freebsd-current-unsubscribe_at_freebsd.org" > > _______________________________________________ > freebsd-current_at_freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe_at_freebsd.org" -- John Baldwin <jhb_at_FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" = http://www.FreeBSD.orgReceived on Tue Apr 05 2005 - 12:18:33 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:31 UTC