on 29/07/2009 14:21 Andriy Gapon said the following: > Your panics are superficially similar but seem to be different. > But it is hard to tell as function argument values are not available in your > backtraces for the interesting calls. > One difference that I see is that your panics happen one level below _sx_xlock, in > sx_xlock_hard and sx argument value appears to be far from NULL > (0xffffff0043557d50) - in the panic that started this thread it was near NULL. > Another difference is that you panics do not involve zfs_znode_dmu_fini and > mu_buf_update_user, in your case sx_xlock is called directly from > zfs_freebsd_reclaim. So it must a problem with a different lock. BTW, have you tried to reproduce the problem with INVARIANTS enabled? Do you have crashdumps with debugging symbols? -- Andriy GaponReceived on Wed Jul 29 2009 - 09:42:21 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:52 UTC