From: Mathieu Desnoyers Date: Wed, 9 Jan 2013 17:40:55 +0000 (-0500) Subject: Fix: statedump hang/too early completion due to logic error X-Git-Tag: v2.2.0-rc1~21 X-Git-Url: https://git.lttng.org./?a=commitdiff_plain;h=7a7128e0ad5de4e9c7b245dae38cc9b2f65339c9;p=lttng-modules.git Fix: statedump hang/too early completion due to logic error The previous "Fix: statedump hang due to incorrect wait/wakeup use" was not actually fixing the real problem. The issue is that we should pass the expected condition to wait_event() rather than its contrary. This bug has been sitting there for a while. I suspect that a recent change in the Linux scheduler behavior for newly spawned worker threads might have contributed to trigger the hang more reliably. The effects of this bugs are: - possible hang of the lttng-sessiond (within the kernel) at tracing start, - the statedump end event is traced before all worker threads have actually completed, which can confuse LTTng viewer state systems. Reported-by: Phil Wilshire Signed-off-by: Mathieu Desnoyers --- diff --git a/lttng-statedump-impl.c b/lttng-statedump-impl.c index 765fcea8..bbe0a156 100644 --- a/lttng-statedump-impl.c +++ b/lttng-statedump-impl.c @@ -397,7 +397,7 @@ int do_lttng_statedump(struct lttng_session *session) schedule_delayed_work_on(cpu, &cpu_work[cpu], 0); } /* Wait for all threads to run */ - wait_event(statedump_wq, (atomic_read(&kernel_threads_to_run) != 0)); + __wait_event(statedump_wq, (atomic_read(&kernel_threads_to_run) == 0)); put_online_cpus(); /* Our work is done */ printk(KERN_DEBUG "LTT state dump end\n");