From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Xiaoguang Chen <chenxg@marvell.com>,
Viresh Kumar <viresh.kumar@linaro.org>,
"Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
Krzysztof Kozlowski <k.kozlowski@samsung.com>
Subject: [PATCH 3.10 40/41] cpufreq: Fix governor start/stop race condition
Date: Fri, 11 Apr 2014 09:10:11 -0700 [thread overview]
Message-ID: <20140411160935.906769115@linuxfoundation.org> (raw)
In-Reply-To: <20140411160932.865173041@linuxfoundation.org>
3.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xiaoguang Chen <chenxg@marvell.com>
commit 95731ebb114c5f0c028459388560fc2a72fe5049 upstream.
Cpufreq governors' stop and start operations should be carried out
in sequence. Otherwise, there will be unexpected behavior, like in
the example below.
Suppose there are 4 CPUs and policy->cpu=CPU0, CPU1/2/3 are linked
to CPU0. The normal sequence is:
1) Current governor is userspace. An application tries to set the
governor to ondemand. It will call __cpufreq_set_policy() in
which it will stop the userspace governor and then start the
ondemand governor.
2) Current governor is userspace. The online of CPU3 runs on CPU0.
It will call cpufreq_add_policy_cpu() in which it will first
stop the userspace governor, and then start it again.
If the sequence of the above two cases interleaves, it becomes:
1) Application stops userspace governor
2) Hotplug stops userspace governor
which is a problem, because the governor shouldn't be stopped twice
in a row. What happens next is:
3) Application starts ondemand governor
4) Hotplug starts a governor
In step 4, the hotplug is supposed to start the userspace governor,
but now the governor has been changed by the application to ondemand,
so the ondemand governor is started once again, which is incorrect.
The solution is to prevent policy governors from being stopped
multiple times in a row. A governor should only be stopped once for
one policy. After it has been stopped, no more governor stop
operations should be executed.
Also add a mutex to serialize governor operations.
[rjw: Changelog. And you owe me a beverage of my choice.]
Signed-off-by: Xiaoguang Chen <chenxg@marvell.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/cpufreq/cpufreq.c | 24 ++++++++++++++++++++++++
include/linux/cpufreq.h | 1 +
2 files changed, 25 insertions(+)
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -46,6 +46,7 @@ static DEFINE_PER_CPU(struct cpufreq_pol
static DEFINE_PER_CPU(char[CPUFREQ_NAME_LEN], cpufreq_cpu_governor);
#endif
static DEFINE_RWLOCK(cpufreq_driver_lock);
+static DEFINE_MUTEX(cpufreq_governor_lock);
/*
* cpu_policy_rwsem is a per CPU reader-writer semaphore designed to cure
@@ -1563,6 +1564,21 @@ static int __cpufreq_governor(struct cpu
pr_debug("__cpufreq_governor for CPU %u, event %u\n",
policy->cpu, event);
+
+ mutex_lock(&cpufreq_governor_lock);
+ if ((!policy->governor_enabled && (event == CPUFREQ_GOV_STOP)) ||
+ (policy->governor_enabled && (event == CPUFREQ_GOV_START))) {
+ mutex_unlock(&cpufreq_governor_lock);
+ return -EBUSY;
+ }
+
+ if (event == CPUFREQ_GOV_STOP)
+ policy->governor_enabled = false;
+ else if (event == CPUFREQ_GOV_START)
+ policy->governor_enabled = true;
+
+ mutex_unlock(&cpufreq_governor_lock);
+
ret = policy->governor->governor(policy, event);
if (!ret) {
@@ -1570,6 +1586,14 @@ static int __cpufreq_governor(struct cpu
policy->governor->initialized++;
else if (event == CPUFREQ_GOV_POLICY_EXIT)
policy->governor->initialized--;
+ } else {
+ /* Restore original values */
+ mutex_lock(&cpufreq_governor_lock);
+ if (event == CPUFREQ_GOV_STOP)
+ policy->governor_enabled = true;
+ else if (event == CPUFREQ_GOV_START)
+ policy->governor_enabled = false;
+ mutex_unlock(&cpufreq_governor_lock);
}
/* we keep one module reference alive for
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -107,6 +107,7 @@ struct cpufreq_policy {
unsigned int policy; /* see above */
struct cpufreq_governor *governor; /* see below */
void *governor_data;
+ bool governor_enabled; /* governor start/stop flag */
struct work_struct update; /* if update_policy() needs to be
* called, but you're in IRQ context */
next prev parent reply other threads:[~2014-04-11 16:10 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-11 16:09 [PATCH 3.10 00/41] 3.10.37-stable review Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 01/41] selinux: correctly label /proc inodes in use before the policy is loaded Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 02/41] powernow-k6: disable cache when changing frequency Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 03/41] powernow-k6: correctly initialize default parameters Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 04/41] powernow-k6: reorder frequencies Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 05/41] kbuild: fix make headers_install when path is too long Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 06/41] cpuidle: Check the result of cpuidle_get_driver() against NULL Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 07/41] net: fix for a race condition in the inet frag code Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 08/41] net: sctp: fix skb leakage in COOKIE ECHO path of chunk->auth_chunk Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 09/41] bridge: multicast: add sanity check for query source addresses Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 10/41] inet: frag: make sure forced eviction removes all frags Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 11/41] net: unix: non blocking recvmsg() should not return -EINTR Greg Kroah-Hartman
2014-04-11 16:21 ` Rainer Weikusat
2014-04-11 16:09 ` [PATCH 3.10 12/41] ipv6: Fix exthdrs offload registration Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 13/41] ipv6: dont set DST_NOCOUNT for remotely added routes Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 14/41] vlan: Set correct source MAC address with TX VLAN offload enabled Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 15/41] tcp: tcp_release_cb() should release socket ownership Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 16/41] net: socket: error on a negative msg_namelen Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 17/41] ipv6: Avoid unnecessary temporary addresses being generated Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 18/41] ipv6: ip6_append_data_mtu do not handle the mtu of the second fragment properly Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 19/41] vxlan: fix potential NULL dereference in arp_reduce() Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 20/41] rtnetlink: fix fdb notification flags Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 21/41] ipmr: fix mfc " Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 22/41] ip6mr: " Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 23/41] netpoll: fix the skb check in pkt_is_ns Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 24/41] tg3: Do not include vlan acceleration features in vlan_features Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 25/41] usbnet: include wait queue head in device structure Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 26/41] vlan: Set hard_header_len according to available acceleration Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 27/41] vhost: fix total length when packets are too short Greg Kroah-Hartman
2014-04-11 16:09 ` [PATCH 3.10 28/41] vhost: validate vhost_get_vq_desc return value Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 29/41] xen-netback: remove pointless clause from if statement Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 30/41] ipv6: some ipv6 statistic counters failed to disable bh Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 31/41] netlink: dont compare the nul-termination in nla_strcmp Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 32/41] isdnloop: Validate NUL-terminated strings from user Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 33/41] isdnloop: several buffer overflows Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 34/41] rds: prevent dereference of a NULL device in rds_iw_laddr_check Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 35/41] ARC: [nsimosci] Change .dts to use generic 8250 UART Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 36/41] ARC: [nsimosci] Unbork console Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 37/41] futex: Allow architectures to skip futex_atomic_cmpxchg_inatomic() test Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 38/41] m68k: Skip " Greg Kroah-Hartman
2014-04-11 16:10 ` [PATCH 3.10 39/41] crypto: ghash-clmulni-intel - use C implementation for setkey() Greg Kroah-Hartman
2014-04-11 16:10 ` Greg Kroah-Hartman [this message]
2014-04-11 16:10 ` [PATCH 3.10 41/41] cpufreq: Fix timer/workqueue corruption due to double queueing Greg Kroah-Hartman
2014-04-11 21:44 ` [PATCH 3.10 00/41] 3.10.37-stable review Guenter Roeck
2014-04-11 23:45 ` Shuah Khan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140411160935.906769115@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=chenxg@marvell.com \
--cc=k.kozlowski@samsung.com \
--cc=linux-kernel@vger.kernel.org \
--cc=rafael.j.wysocki@intel.com \
--cc=stable@vger.kernel.org \
--cc=viresh.kumar@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).