Nothing really serious here, mainly just a matter of nit-picking :-/
From: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Subject: sched: check for SD_SERIALIZE atomically in rebalance_domains()
For CONFIG_SCHED_DEBUG && CONFIG_SYSCT configs, sd->flags can be altered
while being manipulated in rebalance_domains(). Let's do an atomic check.
We rely here on the atomicity of read/write accesses for aligned words.
Signed-off-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>
---
diff --git a/kernel/sched.c b/kernel/sched.c
index 4a378bc..e106e61 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4153,6 +4153,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
/* Earliest time when we have to do rebalance again */
unsigned long next_balance = jiffies + 60*HZ;
int update_next_balance = 0;
+ int need_serialize;
cpumask_t tmp;
for_each_domain(cpu, sd) {
@@ -4170,8 +4171,9 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
if (interval > HZ*NR_CPUS/10)
interval = HZ*NR_CPUS/10;
+ need_serialize = sd->flags & SD_SERIALIZE;
- if (sd->flags & SD_SERIALIZE) {
+ if (need_serialize) {
if (!spin_trylock(&balancing))
goto out;
}
@@ -4187,7 +4189,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
}
sd->last_balance = jiffies;
}
- if (sd->flags & SD_SERIALIZE)
+ if (need_serialize)
spin_unlock(&balancing);
out:
if (time_after(next_balance, sd->last_balance + interval)) {
--
| Greg KH | Re: PCIE |
| Rafael J. Wysocki | [Bug #11237] corrupt PMD after resume |
| Greg Kroah-Hartman | [PATCH 005/196] Chinese: add translation of SubmittingDrivers |
| Chuck Ebbert | Why do so many machines need "noapic"? |
git: | |
| Thomas Glanzmann | fatal: serious inflate inconsistency |
| Yossi Leybovich | corrupt object on git-gc |
| Johannes Schindelin | Re: What's in git.git (stable) |
| Alex Riesen | [PATCH] Allow git-diff exit with codes similar to diff(1) |
| Bertram Scharpf | First install: Grub doesn't find partitions |
| Renaud Allard | very weak bridge performance |
| Richard Stallman | Real men don't attack straw men |
| Wijnand Wiersma | Almost success: OpenBSD on Xen |
| Thor Lancelot Simon | Re: sysctl knob to let sugid processes dump core (pr 15994) |
| Nathan J. Williams | Re: Call for testing: ACPI standby/suspend support |
| matthew green | re: @booted_kernel magic symlink? |
| Greg A. Woods | Re: Fork bomb protection patch |
