Lock test failures + BUG on server-cluster-locking-api branch
Marc Eshel
eshel at almaden.ibm.com
Tue Jan 30 23:07:36 EST 2007
Bruce, this looks like another case of uninitialized file_lock. The cfl in
the following code is not setting fl_ops and fl_lmops to NULL.
Marc.
int fcntl_getlk(struct file *filp, struct flock __user *l)
{
struct file_lock *fl, cfl, file_lock;
struct flock flock;
int error;
error = -EFAULT;
if (copy_from_user(&flock, l, sizeof(flock)))
goto out;
error = -EINVAL;
if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
goto out;
error = flock_to_posix_lock(filp, &file_lock, &flock);
if (error)
goto out;
error = vfs_test_lock(filp, &file_lock, &cfl);
"Bryce Harrington" <bryceharrington at gmail.com>
Sent by: nfsv4-bounces at linux-nfs.org
01/30/2007 03:55 PM
To
NFSv4 at linux-nfs.org
cc
Subject
Lock test failures + BUG on server-cluster-locking-api branch
Hi,
Starting on Jan 19th the server-cluster-locking-api branch started
failing the cthon testsuite on the locking tests:
http://crucible.osdl.org/runs/3704/test_output/cthon.sys.log
Starting LOCKING tests: test directory /mnt/nfs02/nfs03.test (arg: -t)
Testing native post-LFS locking
Creating parent/child synchronization pipes.
Test #1 - Test regions of an unlocked file.
runtests: line 44: 8392 Segmentation fault $i $TESTARGS $NFSTESTDIR
lock tests failed
Tests failed, leaving /mnt/nfs02 mounted
In addition, there's a kernel BUG:
http://crucible.osdl.org/runs/3704/sysinfo/nfs03.console
** Run 3704: Running './server -q -a -t nfs02 -o hard,intr -p /' ***
BUG: unable to handle kernel NULL pointer dereference at virtual
address 0000000e
printing eip:
c016b526
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU: 0
EIP: 0060:[<c016b526>] Not tainted VLI
EFLAGS: 00010202 (2.6.20-rc5-g6698b7f-server-cluster-locking-api #1)
EIP is at locks_release_private+0x26/0x3b
eax: 00000002 ebx: f5933f00 ecx: f5933f00 edx: f5933e90
esi: f5933e90 edi: f77a6618 ebp: f5933e4c esp: f5933e48
ds: 007b es: 007b ss: 0068
Process tlocklfs (pid: 8392, ti=f5933000 task=f7706550 task.ti=f5933000)
Stack: f5933f00 f5933e5c c016b69d 00000000 f5933f00 f5933e70 c016ce8d
00000000
f5933f00 f5933e90 f5933f7c c016cf3f bfc5ce10 f77a6618 00000001 00000000
00000001 08048ee8 000200d2 00000044 f5a588a0 00001155 00000000 f709c678
Call Trace:
[<c010388f>] show_trace_log_lvl+0x19/0x2e
[<c0103959>] show_stack_log_lvl+0xa1/0xa9
[<c0103b40>] show_registers+0x19c/0x2e3
[<c0103e0b>] die+0x114/0x1d9
[<c011636c>] do_page_fault+0x45a/0x53c
[<c047346c>] error_code+0x7c/0x84
[<c016b69d>] locks_copy_lock+0xe/0x36
[<c016ce8d>] vfs_test_lock+0x22/0x59
[<c016cf3f>] fcntl_getlk+0x7b/0x15a
[<c0169466>] do_fcntl+0xe3/0x140
[<c016955a>] sys_fcntl64+0x5a/0x6e
[<c0102b4c>] syscall_call+0x7/0xb
=======================
Code: 1a ff ff 5d c3 55 89 e5 53 89 c3 8b 40 58 85 c0 74 12 8b 50 0c
85 d2 74 04 89 d8 ff d2 c7 43 58 00 00 00 00 8b 43 5c 85 c0 74 12 <8b>
50 0c 85 d2 74 04 89 d8 ff d2 c7 43 5c 00 00 00 00 5b 5d c3
EIP: [<c016b526>] locks_release_private+0x26/0x3b SS:ESP 0068:f5933e48
_______________________________________________
NFSv4 mailing list
NFSv4 at linux-nfs.org
http://linux-nfs.org/cgi-bin/mailman/listinfo/nfsv4
More information about the NFSv4
mailing list