linux-kernel - [PATCH -mm] lower migration thread/stop machine prio

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-Id: <20060809024637.231688000@mvista.com>
Date:	Tue, 08 Aug 2006 19:46:37 -0700
From:	dwalker@...sta.com
To:	akpm@...l.org
Cc:	linux-kernel@...r.kernel.org
Subject: [PATCH -mm] lower migration thread/stop machine prio

We should give some space at the top to allow things to schedule
above everything in a default system. I think this takes better
advantage of all the priorities we have. 

This moves the stop_machine and migration threads to SCHED_FIFO 
prio 80.

Signed-Off-By: Daniel Walker <dwalker@...sta.com>

---
 include/linux/sched.h        |    3 +++
 include/linux/stop_machine.h |    3 +++
 kernel/sched.c               |    3 +--
 kernel/stop_machine.c        |    2 +-
 4 files changed, 8 insertions(+), 3 deletions(-)

Index: linux-2.6.17/include/linux/sched.h
===================================================================
--- linux-2.6.17.orig/include/linux/sched.h
+++ linux-2.6.17/include/linux/sched.h
@@ -510,6 +510,9 @@ struct signal_struct {
 #define has_rt_policy(p) \
 	unlikely((p)->policy != SCHED_NORMAL && (p)->policy != SCHED_BATCH)
 
+/* Must be high prio: stop_machine expects to yield to it. */
+#define MIGRATION_THREAD_PRIO (MAX_RT_PRIO-20)
+
 /*
  * Some day this will be a full-fledged user tracking system..
  */
Index: linux-2.6.17/include/linux/stop_machine.h
===================================================================
--- linux-2.6.17.orig/include/linux/stop_machine.h
+++ linux-2.6.17/include/linux/stop_machine.h
@@ -8,6 +8,9 @@
 #include <asm/system.h>
 
 #if defined(CONFIG_STOP_MACHINE) && defined(CONFIG_SMP)
+
+#define STOP_MACHINE_PRIO (MAX_RT_PRIO-20)
+
 /**
  * stop_machine_run: freeze the machine on all CPUs and run this function
  * @fn: the function to run
Index: linux-2.6.17/kernel/sched.c
===================================================================
--- linux-2.6.17.orig/kernel/sched.c
+++ linux-2.6.17/kernel/sched.c
@@ -5289,9 +5289,8 @@ migration_call(struct notifier_block *nf
 			return NOTIFY_BAD;
 		p->flags |= PF_NOFREEZE;
 		kthread_bind(p, cpu);
-		/* Must be high prio: stop_machine expects to yield to it. */
 		rq = task_rq_lock(p, &flags);
-		__setscheduler(p, SCHED_FIFO, MAX_RT_PRIO-1);
+		__setscheduler(p, SCHED_FIFO, MIGRATION_THREAD_PRIO);
 		task_rq_unlock(rq, &flags);
 		cpu_rq(cpu)->migration_thread = p;
 		break;
Index: linux-2.6.17/kernel/stop_machine.c
===================================================================
--- linux-2.6.17.orig/kernel/stop_machine.c
+++ linux-2.6.17/kernel/stop_machine.c
@@ -86,7 +86,7 @@ static void stopmachine_set_state(enum s
 static int stop_machine(void)
 {
 	int i, ret = 0;
-	struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+	struct sched_param param = { .sched_priority = STOP_MACHINE_PRIO };
 
 	/* One high-prio thread per cpu.  We'll do this one. */
 	sched_setscheduler(current, SCHED_FIFO, &param);
--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/