This is a non regression test about interoperablity. The test used is conectathon 2004. Hardware configurations are:
- 2x x86_64 machine (Intel Xeon), 1GByte ram
- 2x IA32 machine, 2GBytes ram
Connectathon test results
| Machines | IA32 Client | x86_64 Client |
| IA32 Server | PASSED | PASSED |
| x86_64 Server | Failed | Failed |
Failure details
While executing Connectathon locking tests, there's a failure on the server. Here is the trace:
ep 13 16:23:40 nfs4 kernel: general protection fault: 0000 [1] SMP
Sep 13 16:23:40 nfs4 kernel: CPU 3
Sep 13 16:23:40 nfs4 kernel: Modules linked in: radeon drm parport_pc lp parport autofs4 binfmt_misc video button battery ac i2c_i801 i2c_core ide_cd
Sep 13 16:23:40 nfs4 kernel: Pid: 3451, comm: nfsd Not tainted 2.6.13-CITI_NFS4_ALL-1
Sep 13 16:23:40 nfs4 kernel: RIP: 0010:[] {kref_get+4}
Sep 13 16:23:40 nfs4 kernel: RSP: 0018:ffff81007775fb48 EFLAGS: 00010202
Sep 13 16:23:40 nfs4 kernel: RAX: 0000000000000000 RBX: 6666666666666666 RCX: ffff8100748355c0
Sep 13 16:23:40 nfs4 kernel: RDX: ffff8100749adc28 RSI: ffff8100748355c0 RDI: 6666666666666666
Sep 13 16:23:40 nfs4 kernel: RBP: ffff81007734d240 R08: ffff810074835588 R09: ffff81007734d228
Sep 13 16:23:40 nfs4 kernel: R10: 0000000000000000 R11: ffff810076c2cab0 R12: ffff81007734d1f0
Sep 13 16:23:40 nfs4 kernel: R13: ffff810074835588 R14: ffff810074835620 R15: ffff81007734d230
Sep 13 16:23:40 nfs4 kernel: FS: 00002aaaaaad03a0(0000) GS:ffffffff80668980(0000) knlGS:0000000000000000
Sep 13 16:23:40 nfs4 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
Sep 13 16:23:40 nfs4 kernel: CR2: 00007fffff864b0c CR3: 0000000077e5f000 CR4: 00000000000006e0
Sep 13 16:23:40 nfs4 kernel: Process nfsd (pid: 3451, threadinfo ffff81007775e000, task ffff81007ec951f0)
Sep 13 16:23:40 nfs4 kernel: Stack: 6666666666666666 ffffffff802335e6 ffff81007734d1f0 ffff81007cc72b00
Sep 13 16:23:40 nfs4 kernel: ffff81007775fe10 ffff810074835cc0 ffff8100751bb7a0 ffff8100748355d0
Sep 13 16:23:40 nfs4 kernel: 0000000000000000 0000000000000286
Sep 13 16:23:40 nfs4 kernel: Call Trace:{nfsd4_lock+519} {export_decode_fh+76}
Sep 13 16:23:40 nfs4 kernel: {fh_verify+1346} {nfsd4_proc_compound+1955}
Sep 13 16:23:40 nfs4 kernel: {svc_tcp_recvfrom+2084} {thread_return+0}
Sep 13 16:23:40 nfs4 kernel: {thread_return+82} {svc_sendto+276}
Sep 13 16:23:40 nfs4 kernel: {buffered_rmqueue+493} {lock_timer_base+27}
Sep 13 16:23:40 nfs4 kernel: {try_to_del_timer_sync+86} {ip_map_lookup+508}
Sep 13 16:23:40 nfs4 kernel: {nfsd4_decode_compound+3865} {nfsd_dispatch+226}
Sep 13 16:23:40 nfs4 kernel: {svc_process+1033} {__down_read+50}
Sep 13 16:23:40 nfs4 kernel: {nfsd+456} {schedule_tail+57}
Sep 13 16:23:40 nfs4 kernel: {child_rip+8} {nfsd+0}
Sep 13 16:23:40 nfs4 kernel: {child_rip+0}
Sep 13 16:23:40 nfs4 kernel:
Sep 13 16:23:40 nfs4 kernel: Code: 8b 07 85 c0 75 24 b9 20 00 00 00 48 c7 c2 8b c1 46 80 48 c7
Sep 13 16:23:40 nfs4 kernel: RIP {kref_get+4} RSP
The bug occurs randomly, and is not easy to reproduce. Most of the locking tests attemps are OK. It may depend of the memory state
Software configuration
- For both client and server: Linux 2.6.13-CITI_NFS4_ALL-1
