printk/sched: Introduce special printk_sched() for those awkward moments

There's a few awkward printk()s inside of scheduler guts that people
prefer to keep but really are rather deadlock prone. Fudge around it
by storing the text in a per-cpu buffer and poll it using the existing
printk_tick() handler.

This will drop output when its more frequent than once a tick, however
only the affinity thing could possible go that fast and for that just
one should suffice to notify the admin he's done something silly..

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/n/tip-wua3lmkt3dg8nfts66o6brne@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>

4 files changed, 55 insertions, 5 deletions

diff --git a/include/linux/printk.h b/include/linux/printk.h
index f0e22f75143f..1f77a4174ee0 100644
--- a/include/linux/printk.h
+++ b/include/linux/printk.h
@@ -101,6 +101,11 @@ asmlinkage __printf(1, 2) __cold
 int printk(const char *fmt, ...);
 
 /*
+ * Special printk facility for scheduler use only, _DO_NOT_USE_ !
+ */
+__printf(1, 2) __cold int printk_sched(const char *fmt, ...);
+
+/*
  * Please don't use printk_ratelimit(), because it shares ratelimiting state
  * with all other unrelated printk_ratelimit() callsites.  Instead use
  * printk_ratelimited() or plain old __ratelimit().
@@ -127,6 +132,11 @@ int printk(const char *s, ...)
 {
         return 0;
 }
+static inline __printf(1, 2) __cold
+int printk_sched(const char *s, ...)
+{
+        return 0;
+}
 static inline int printk_ratelimit(void)
 {
         return 0;
diff --git a/kernel/printk.c b/kernel/printk.c
index 13c0a1143f49..7ca7ba591e21 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1208,13 +1208,47 @@ int is_console_locked(void)
         return console_locked;
 }
 
+/*
+ * Delayed printk facility, for scheduler-internal messages:
+ */
+#define PRINTK_BUF_SIZE         512
+
+#define PRINTK_PENDING_WAKEUP   0x01
+#define PRINTK_PENDING_SCHED    0x02
+
 static DEFINE_PER_CPU(int, printk_pending);
+static DEFINE_PER_CPU(char [PRINTK_BUF_SIZE], printk_sched_buf);
+
+int printk_sched(const char *fmt, ...)
+{
+        unsigned long flags;
+        va_list args;
+        char *buf;
+        int r;
+
+        local_irq_save(flags);
+        buf = __get_cpu_var(printk_sched_buf);
+
+        va_start(args, fmt);
+        r = vsnprintf(buf, PRINTK_BUF_SIZE, fmt, args);
+        va_end(args);
+
+        __this_cpu_or(printk_pending, PRINTK_PENDING_SCHED);
+        local_irq_restore(flags);
+
+        return r;
+}
 
 void printk_tick(void)
 {
         if (__this_cpu_read(printk_pending)) {
-                __this_cpu_write(printk_pending, 0);
-                wake_up_interruptible(&log_wait);
+                int pending = __this_cpu_xchg(printk_pending, 0);
+                if (pending & PRINTK_PENDING_SCHED) {
+                        char *buf = __get_cpu_var(printk_sched_buf);
+                        printk(KERN_WARNING "[sched_delayed] %s", buf);
+                }
+                if (pending & PRINTK_PENDING_WAKEUP)
+                        wake_up_interruptible(&log_wait);
         }
 }
 
@@ -1228,7 +1262,7 @@ int printk_needs_cpu(int cpu)
 void wake_up_klogd(void)
 {
         if (waitqueue_active(&log_wait))
-                this_cpu_write(printk_pending, 1);
+                this_cpu_or(printk_pending, PRINTK_PENDING_WAKEUP);
 }
 
 /**
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index b1ccce819ce2..8781cec7c3e6 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1284,7 +1284,7 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
          * leave kernel.
          */
         if (p->mm && printk_ratelimit()) {
-                printk(KERN_INFO "process %d (%s) no longer affine to cpu%d\n",
+                printk_sched("process %d (%s) no longer affine to cpu%d\n",
                                 task_pid_nr(p), p->comm, cpu);
         }
 
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 7f7e7cdcb472..b60dad720173 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -864,8 +864,14 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
                  * but accrue some time due to boosting.
                  */
                 if (likely(rt_b->rt_runtime)) {
+                        static bool once = false;
+
                         rt_rq->rt_throttled = 1;
-                        printk_once(KERN_WARNING "sched: RT throttling activated\n");
+
+                        if (!once) {
+                                once = true;
+                                printk_sched("sched: RT throttling activated\n");
+                        }
                 } else {
                         /*
                          * In case we did anyway, make it go away,