linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Slaby <jslaby@suse.cz>
To: stable@vger.kernel.org
Cc: linux-kernel@vger.kernel.org,
	Bodo Stroesser <bstroesser@ts.fujitsu.com>,
	Corey Minyard <cminyard@mvista.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jiri Slaby <jslaby@suse.cz>
Subject: [PATCH 3.12 20/50] ipmi: Fix a race restarting the timer
Date: Mon,  5 May 2014 14:43:02 +0200	[thread overview]
Message-ID: <10083c00587485c79d86b5c425464418cfc3c586.1399292849.git.jslaby@suse.cz> (raw)
In-Reply-To: <7d4f4737432af6216e86975e587331b9d8b08063.1399292849.git.jslaby@suse.cz>
In-Reply-To: <cover.1399292849.git.jslaby@suse.cz>

From: Bodo Stroesser <bstroesser@ts.fujitsu.com>

3.12-stable review patch.  If anyone has any objections, please let me know.

===============

commit 48e8ac2979920ffa39117e2d725afa3a749bfe8d upstream.

With recent changes it is possible for the timer handler to detect an
idle interface and not start the timer, but the thread to start an
operation at the same time.  The thread will not start the timer in that
instance, resulting in the timer not running.

Instead, move all timer operations under the lock and start the timer in
the thread if it detect non-idle and the timer is not already running.
Moving under locks allows the last timeout to be set in both the thread
and the timer.  'Timer is not running' means that the timer is not
pending and smi_timeout() is not running.  So we need a flag to detect
this correctly.

Also fix a few other timeout bugs: setting the last timeout when the
interrupt has to be disabled and the timer started, and setting the last
timeout in check_start_timer_thread possibly racing with the timer

Signed-off-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Bodo Stroesser <bstroesser@ts.fujitsu.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
---
 drivers/char/ipmi/ipmi_si_intf.c | 46 ++++++++++++++++++++++++----------------
 1 file changed, 28 insertions(+), 18 deletions(-)

diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c
index 15e4a6031934..e5bdd1a2f541 100644
--- a/drivers/char/ipmi/ipmi_si_intf.c
+++ b/drivers/char/ipmi/ipmi_si_intf.c
@@ -249,6 +249,9 @@ struct smi_info {
 	/* The timer for this si. */
 	struct timer_list   si_timer;
 
+	/* This flag is set, if the timer is running (timer_pending() isn't enough) */
+	bool		    timer_running;
+
 	/* The time (in jiffies) the last timeout occurred at. */
 	unsigned long       last_timeout_jiffies;
 
@@ -435,6 +438,13 @@ static void start_clear_flags(struct smi_info *smi_info)
 	smi_info->si_state = SI_CLEARING_FLAGS;
 }
 
+static void smi_mod_timer(struct smi_info *smi_info, unsigned long new_val)
+{
+	smi_info->last_timeout_jiffies = jiffies;
+	mod_timer(&smi_info->si_timer, new_val);
+	smi_info->timer_running = true;
+}
+
 /*
  * When we have a situtaion where we run out of memory and cannot
  * allocate messages, we just leave them in the BMC and run the system
@@ -447,8 +457,7 @@ static inline void disable_si_irq(struct smi_info *smi_info)
 		start_disable_irq(smi_info);
 		smi_info->interrupt_disabled = 1;
 		if (!atomic_read(&smi_info->stop_operation))
-			mod_timer(&smi_info->si_timer,
-				  jiffies + SI_TIMEOUT_JIFFIES);
+			smi_mod_timer(smi_info, jiffies + SI_TIMEOUT_JIFFIES);
 	}
 }
 
@@ -908,15 +917,7 @@ static void sender(void                *send_info,
 		list_add_tail(&msg->link, &smi_info->xmit_msgs);
 
 	if (smi_info->si_state == SI_NORMAL && smi_info->curr_msg == NULL) {
-		/*
-		 * last_timeout_jiffies is updated here to avoid
-		 * smi_timeout() handler passing very large time_diff
-		 * value to smi_event_handler() that causes
-		 * the send command to abort.
-		 */
-		smi_info->last_timeout_jiffies = jiffies;
-
-		mod_timer(&smi_info->si_timer, jiffies + SI_TIMEOUT_JIFFIES);
+		smi_mod_timer(smi_info, jiffies + SI_TIMEOUT_JIFFIES);
 
 		if (smi_info->thread)
 			wake_up_process(smi_info->thread);
@@ -1005,6 +1006,17 @@ static int ipmi_thread(void *data)
 
 		spin_lock_irqsave(&(smi_info->si_lock), flags);
 		smi_result = smi_event_handler(smi_info, 0);
+
+		/*
+		 * If the driver is doing something, there is a possible
+		 * race with the timer.  If the timer handler see idle,
+		 * and the thread here sees something else, the timer
+		 * handler won't restart the timer even though it is
+		 * required.  So start it here if necessary.
+		 */
+		if (smi_result != SI_SM_IDLE && !smi_info->timer_running)
+			smi_mod_timer(smi_info, jiffies + SI_TIMEOUT_JIFFIES);
+
 		spin_unlock_irqrestore(&(smi_info->si_lock), flags);
 		busy_wait = ipmi_thread_busy_wait(smi_result, smi_info,
 						  &busy_until);
@@ -1074,10 +1086,6 @@ static void smi_timeout(unsigned long data)
 		     * SI_USEC_PER_JIFFY);
 	smi_result = smi_event_handler(smi_info, time_diff);
 
-	spin_unlock_irqrestore(&(smi_info->si_lock), flags);
-
-	smi_info->last_timeout_jiffies = jiffies_now;
-
 	if ((smi_info->irq) && (!smi_info->interrupt_disabled)) {
 		/* Running with interrupts, only do long timeouts. */
 		timeout = jiffies + SI_TIMEOUT_JIFFIES;
@@ -1099,7 +1107,10 @@ static void smi_timeout(unsigned long data)
 
  do_mod_timer:
 	if (smi_result != SI_SM_IDLE)
-		mod_timer(&(smi_info->si_timer), timeout);
+		smi_mod_timer(smi_info, timeout);
+	else
+		smi_info->timer_running = false;
+	spin_unlock_irqrestore(&(smi_info->si_lock), flags);
 }
 
 static irqreturn_t si_irq_handler(int irq, void *data)
@@ -1147,8 +1158,7 @@ static int smi_start_processing(void       *send_info,
 
 	/* Set up the timer that drives the interface. */
 	setup_timer(&new_smi->si_timer, smi_timeout, (long)new_smi);
-	new_smi->last_timeout_jiffies = jiffies;
-	mod_timer(&new_smi->si_timer, jiffies + SI_TIMEOUT_JIFFIES);
+	smi_mod_timer(new_smi, jiffies + SI_TIMEOUT_JIFFIES);
 
 	/*
 	 * Check if the user forcefully enabled the daemon.
-- 
1.9.2


  parent reply	other threads:[~2014-05-05 12:50 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-05 12:28 [PATCH 3.12 00/50] 3.12.19-stable review Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 01/50] openvswitch: fix vport-netdev unregister Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 02/50] brcmsmac: fix deadlock on missing firmware Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 03/50] /dev/mem: handle out-of-bounds read/write Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 04/50] drivers/net: tulip_remove_one needs to call pci_disable_device() Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 05/50] Bluetooth: Add support for Intel Bluetooth device [8087:0a2a] Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 06/50] iommu/amd: Fix PASID format in INVALIDATE_IOTLB_PAGES command Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 07/50] usbatm: Fix dynamic_debug / ratelimited atm_dbg and atm_rldbg macros Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 08/50] printk: pr_debug_ratelimited: check state first to reduce "callbacks suppressed" messages Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 09/50] dcache: restore error on restart in prepend_path Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 10/50] __dentry_path() fixes Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 11/50] i2c: i801: enable Intel BayTrail SMBUS Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 12/50] e1000e: Fix no connectivity when driver loaded with cable out Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 13/50] ACPI / EC: Process rather than discard events in acpi_ec_clear Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 14/50] ARM: 7840/1: LPAE: don't reject mapping /dev/mem above 4GB Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 15/50] x86/quirks: Add workaround for AMD F16h Erratum792 Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 16/50] amd64_edac: Fix logic to determine channel for F15 M30h processors Jiri Slaby
2014-05-05 12:42 ` [PATCH 3.12 17/50] backing_dev: fix hung task on sync Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 18/50] bdi: avoid oops on device removal Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 19/50] virtio_balloon: don't softlockup on huge balloon changes Jiri Slaby
2014-05-05 12:43 ` Jiri Slaby [this message]
2014-05-05 12:43 ` [PATCH 3.12 21/50] KVM: ioapic: fix assignment of ioapic->rtc_status.pending_eoi (CVE-2014-0155) Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 22/50] net: ipv4: current group_info should be put after using Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 24/50] powerpc/8xx: mfspr SPRN_TBRx in lieu of mftb/mftbu is not supported Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 25/50] ACPI / sleep: remove panic in case hardware has changed after S4 Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 26/50] user namespace: fix incorrect memory barriers Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 27/50] x86: Adjust irq remapping quirk for older revisions of 5500/5520 chipsets Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 28/50] PCI: designware: Fix RC BAR to be single 64-bit non-prefetchable memory BAR Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 29/50] PCI: designware: Fix iATU programming for cfg1, io and mem viewport Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 30/50] ACPI / button: Add ACPI Button event via netlink routine Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 31/50] staging: comedi: 8255_pci: initialize MITE data window Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 32/50] tty: Set correct tty name in 'active' sysfs attribute Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 33/50] tty: Fix low_latency BUG Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 35/50] Bluetooth: Fix removing Long Term Key Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 36/50] xfs: fix directory hash ordering bug Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 37/50] Btrfs: skip submitting barrier for missing device Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 38/50] Btrfs: fix deadlock with nested trans handles Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 39/50] ext4: fix error return from ext4_ext_handle_uninitialized_extents() Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 40/50] ext4: fix partial cluster handling for bigalloc file systems Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 41/50] ext4: fix premature freeing of partial clusters split across leaf blocks Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 42/50] jffs2: Fix segmentation fault found in stress test Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 43/50] jffs2: Fix crash due to truncation of csize Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 44/50] jffs2: avoid soft-lockup in jffs2_reserve_space_gc() Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 45/50] jffs2: remove from wait queue after schedule() Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 46/50] sparc32: fix build failure for arch_jump_label_transform Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 47/50] sparc64: don't treat 64-bit syscall return codes as 32-bit Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 48/50] sparc64: Make sure %pil interrupts are enabled during hypervisor yield Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 49/50] wait: fix reparent_leader() vs EXIT_DEAD->EXIT_ZOMBIE race Jiri Slaby
2014-05-05 12:43 ` [PATCH 3.12 50/50] exit: call disassociate_ctty() before exit_task_namespaces() Jiri Slaby
2014-05-05 15:45 ` [PATCH 3.12 00/50] 3.12.19-stable review Guenter Roeck
2014-05-06 14:57 ` Shuah Khan
2014-05-09  8:32   ` Jiri Slaby

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=10083c00587485c79d86b5c425464418cfc3c586.1399292849.git.jslaby@suse.cz \
    --to=jslaby@suse.cz \
    --cc=bstroesser@ts.fujitsu.com \
    --cc=cminyard@mvista.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.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).