Hello Robert Watson wrote: >On Mon, 16 Feb 2004, Jiri Mikulas wrote: > >>Hello, just only for info I got this panic on: FreeBSD 5.2-CURRENT #1: >>Mon Feb 16 00:53:39 CET 2004 >> >>I built few kernels from saturday morning to sunday evening and with all >>of them i got this panic .. (my last functional kernel is from FreeBSD >>5.2-CURRENT #0: Mon Feb 9 19:53:47 CET 2004) Jiri >> >> > >It looks like 'ifp' in ip_output.c is NULL here (the offset of if_hwassist >is 0x98). Are you running with DUMMYNET? I'm wondering if the recent >commit to clean up MT_TAG pseudo-mbufs has a loose end somewhere. Try >backing ouf the following commit locally: > >mlaier 2004/02/13 11:14:16 PST > > FreeBSD src repository > > Modified files: > sys/net bridge.c if_ethersubr.c > sys/netinet in_proto.c ip_divert.c ip_dummynet.c > ip_dummynet.h ip_fastfwd.c ip_fw.h > ip_fw2.c ip_input.c ip_output.c ip_var.h > tcp_debug.c tcp_input.c > sys/sys mbuf.h > Added files: > sys/netinet ip_divert.h > > > yes I'm runnig DUMMYNET >Also, it would be quite interesting to see the contents of 'inspect *m' >and 'inspect *ro' in stack frame #6. > > > ~~~cut~~~ (kgdb) up 6 #6 0xc05726c8 in ip_output (m=0xc10f7200, opt=0x0, ro=0xc2abab68, flags=1, imo=0x0, inp=0x0) at /usr/src/sys/netinet/ip_output.c:983 983 sw_csum = m->m_pkthdr.csum_flags & ~ifp->if_hwassist; (kgdb) inspect *m $1 = {m_hdr = {mh_next = 0x0, mh_nextpkt = 0x0, mh_data = 0xc10f7250 "E", mh_len = 48, mh_flags = 2, mh_type = 1}, M_dat = {MH = { MH_pkthdr = {rcvif = 0xc27b4000, len = 48, header = 0x14, csum_flags = 1, csum_data = 393990, tags = {slh_first = 0x0}}, MH_dat = { MH_ext = {ext_buf = 0x1020108---Can't read userspace from dump, or kernel process--- (kgdb) inspect *ro $2 = {ro_rt = 0x0, ro_dst = {sa_len = 0 '\0', sa_family = 0 '\0', sa_data = '\0' <repeats 13 times>}} ~~~cut~~~ is it all ? thanks JiriReceived on Sun Feb 15 2004 - 23:36:49 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:43 UTC