Skip to contentSkip to conversation

revert "Optimize sys_times for a single thread process"

PATCH In review Oleg Nesterov 2 participants 3 messages Feb 22, 2006

lore
Share & tools
lore
#0 lore raw

PATCH 2/3 revert "Optimize sys_times for a single thread process"

This patch reverts 'CONFIG_SMP && thread_group_empty()' optimization in sys_times(). The reason is that the next patch breaks memory ordering which is needed for that optimization.

tasklist_lock in sys_times() will be eliminated completely by further patch.

kernel/exit.c renamed from kernel/exit.c~2_REVERT +1 -5
@@ -137,11 +137,7 @@ repeat:
 	ptrace_unlink(p);
 	BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children));
 	__exit_signal(p);
-	/*
-	 * Note that the fastpath in sys_times depends on __exit_signal having
-	 * updated the counters before a task is removed from the tasklist of
-	 * the process by __unhash_process.
-	 */
+
 	__unhash_process(p);
 
 	/*
kernel/sys.c renamed from kernel/sys.c~2_REVERT +25 -59

94 lines not on this page · raw message

#1 lore raw
Quoted 2 lines · Oleg Nesterov

tasklist_lock in sys_times() will be eliminated completely by further patch.

Where is that patch? The patch would simply drop the locks?

#2 lore raw

Re: [PATCH 2/3] revert "Optimize sys_times for a single threadprocess"

Quoted 7 lines · Christoph Lameter
Quoted 2 lines · Oleg Nesterov

tasklist_lock in sys_times() will be eliminated completely by further patch.

Where is that patch? The patch would simply drop the locks?

Just sent it: "[PATCH 1/6] sys_times: don't take tasklist_lock"

It only drops tasklist_lock, sighand->siglock is still needed, of course.

Oleg.