From 0df12b142ce917fd8395d4837f15bd473e60872f Mon Sep 17 00:00:00 2001 From: johann <johann> Date: Tue, 7 Oct 2008 15:07:06 +0000 Subject: [PATCH] Branch b1_6 b=17026 i=shadow i=oleg Description: (ptllnd_peer.c:557:kptllnd_peer_check_sends()) ASSERTION(!in_interrupt()) failed Details: fix stack overflow in the distributed lock manager by defering export eviction after a failed ast to the elt thread instead of handling it in the dlm interpret routine. --- lustre/ChangeLog | 7 +++++++ lustre/ldlm/ldlm_lockd.c | 7 +++++++ lustre/mds/mds_reint.c | 3 ++- 3 files changed, 16 insertions(+), 1 deletion(-) diff --git a/lustre/ChangeLog b/lustre/ChangeLog index 7e06d24f65..a0bd3d0231 100644 --- a/lustre/ChangeLog +++ b/lustre/ChangeLog @@ -144,6 +144,13 @@ Bugzilla : 14095 Description: Add lustre_start utility to start or stop multiple Lustre servers from a CSV file. +Severity : normal +Bugzilla : 17026 +Description: (ptllnd_peer.c:557:kptllnd_peer_check_sends()) ASSERTION(!in_interrupt()) failed +Details : fix stack overflow in the distributed lock manager by defering export + eviction after a failed ast to the elt thread instead of handling + it in the dlm interpret routine. + -------------------------------------------------------------------------- 2008-08-31 Sun Microsystems, Inc. diff --git a/lustre/ldlm/ldlm_lockd.c b/lustre/ldlm/ldlm_lockd.c index 7dfbbd1a2c..a6cc1a2309 100644 --- a/lustre/ldlm/ldlm_lockd.c +++ b/lustre/ldlm/ldlm_lockd.c @@ -465,7 +465,14 @@ static void ldlm_failed_ast(struct ldlm_lock *lock, int rc, if (obd_dump_on_timeout) libcfs_debug_dumplog(); +#ifdef __KERNEL__ + spin_lock_bh(&waiting_locks_spinlock); + list_add(&lock->l_pending_chain, &expired_lock_thread.elt_expired_locks); + cfs_waitq_signal(&expired_lock_thread.elt_waitq); + spin_unlock_bh(&waiting_locks_spinlock); +#else class_fail_export(lock->l_export); +#endif } static int ldlm_handle_ast_error(struct ldlm_lock *lock, diff --git a/lustre/mds/mds_reint.c b/lustre/mds/mds_reint.c index 2dc03816b9..ab8330d899 100644 --- a/lustre/mds/mds_reint.c +++ b/lustre/mds/mds_reint.c @@ -2393,7 +2393,8 @@ no_unlink: GOTO(cleanup, rc); cleanup: - rc = mds_finish_transno(mds, de_tgtdir ? de_tgtdir->d_inode : NULL, + rc = mds_finish_transno(mds, de_tgtdir && !IS_ERR(de_tgtdir) ? + de_tgtdir->d_inode : NULL, handle, req, rc, 0, 0); switch (cleanup_phase) { -- GitLab