From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Kevin Hao <haokexin@gmail.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: [PATCH 4.4 11/43] dump_stack: avoid the livelock of the dump_lock
Date: Mon, 11 Nov 2019 19:28:25 +0100 [thread overview]
Message-ID: <20191111181300.050698565@linuxfoundation.org> (raw)
In-Reply-To: <20191111181246.772983347@linuxfoundation.org>
From: Kevin Hao <haokexin@gmail.com>
commit 5cbf2fff3bba8d3c6a4d47c1754de1cf57e2b01f upstream.
In the current code, we use the atomic_cmpxchg() to serialize the output
of the dump_stack(), but this implementation suffers the thundering herd
problem. We have observed such kind of livelock on a Marvell cn96xx
board(24 cpus) when heavily using the dump_stack() in a kprobe handler.
Actually we can let the competitors to wait for the releasing of the
lock before jumping to atomic_cmpxchg(). This will definitely mitigate
the thundering herd problem. Thanks Linus for the suggestion.
[akpm@linux-foundation.org: fix comment]
Link: http://lkml.kernel.org/r/20191030031637.6025-1-haokexin@gmail.com
Fixes: b58d977432c8 ("dump_stack: serialize the output from dump_stack()")
Signed-off-by: Kevin Hao <haokexin@gmail.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
lib/dump_stack.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
--- a/lib/dump_stack.c
+++ b/lib/dump_stack.c
@@ -44,7 +44,12 @@ retry:
was_locked = 1;
} else {
local_irq_restore(flags);
- cpu_relax();
+ /*
+ * Wait for the lock to release before jumping to
+ * atomic_cmpxchg() in order to mitigate the thundering herd
+ * problem.
+ */
+ do { cpu_relax(); } while (atomic_read(&dump_lock) != -1);
goto retry;
}
next prev parent reply other threads:[~2019-11-11 19:11 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-11 18:28 [PATCH 4.4 00/43] 4.4.201-stable review Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 01/43] CDC-NCM: handle incomplete transfer of MTU Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 02/43] net: fix data-race in neigh_event_send() Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 03/43] NFC: fdp: fix incorrect free object Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 04/43] NFC: st21nfca: fix double free Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 05/43] qede: fix NULL pointer deref in __qede_remove() Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 06/43] nfc: netlink: fix double device reference drop Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 07/43] ALSA: bebob: fix to detect configured source of sampling clock for Focusrite Saffire Pro i/o series Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 08/43] ALSA: hda/ca0132 - Fix possible workqueue stall Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 09/43] mm, meminit: recalculate pcpu batch and high limits after init completes Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 10/43] mm, vmstat: hide /proc/pagetypeinfo from normal users Greg Kroah-Hartman
2019-11-11 18:28 ` Greg Kroah-Hartman [this message]
2019-11-11 18:28 ` [PATCH 4.4 12/43] perf tools: Fix time sorting Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 13/43] drm/radeon: fix si_enable_smc_cac() failed issue Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 14/43] ceph: fix use-after-free in __ceph_remove_cap() Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 15/43] iio: imu: adis16480: make sure provided frequency is positive Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 16/43] netfilter: nf_tables: Align nft_expr private data to 64-bit Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 17/43] netfilter: ipset: Fix an error code in ip_set_sockfn_get() Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 18/43] can: usb_8dev: fix use-after-free on disconnect Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 19/43] can: c_can: c_can_poll(): only read status register after status IRQ Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 20/43] can: peak_usb: fix a potential out-of-sync while decoding packets Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 21/43] can: gs_usb: gs_can_open(): prevent memory leak Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 22/43] can: peak_usb: fix slab info leak Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 23/43] drivers: usb: usbip: Add missing break statement to switch Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 24/43] configfs: fix a deadlock in configfs_symlink() Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 25/43] PCI: tegra: Enable Relaxed Ordering only for Tegra20 & Tegra30 Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 26/43] scsi: qla2xxx: fixup incorrect usage of host_byte Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 27/43] scsi: lpfc: Honor module parameter lpfc_use_adisc Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 28/43] ipvs: move old_secure_tcp into struct netns_ipvs Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 29/43] bonding: fix unexpected IFF_BONDING bit unset Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 30/43] usb: fsl: Check memory resource before releasing it Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 31/43] usb: gadget: udc: atmel: Fix interrupt storm in FIFO mode Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 32/43] usb: gadget: composite: Fix possible double free memory bug Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 33/43] usb: gadget: configfs: fix concurrent issue between composite APIs Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 34/43] perf/x86/amd/ibs: Fix reading of the IBS OpData register and thus precise RIP validity Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 35/43] USB: Skip endpoints with 0 maxpacket length Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 36/43] scsi: qla2xxx: stop timer in shutdown path Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 37/43] net: hisilicon: Fix "Trying to free already-free IRQ" Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 38/43] NFSv4: Dont allow a cached open with a revoked delegation Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 39/43] igb: Fix constant media auto sense switching when no cable is connected Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 40/43] e1000: fix memory leaks Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 41/43] can: flexcan: disable completely the ECC mechanism Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 42/43] mm/filemap.c: dont initiate writeback if mapping has no dirty pages Greg Kroah-Hartman
2019-11-11 18:28 ` [PATCH 4.4 43/43] cgroup,writeback: dont switch wbs immediately on dead wbs if the memcg is dead Greg Kroah-Hartman
2019-11-11 23:35 ` [PATCH 4.4 00/43] 4.4.201-stable review Guenter Roeck
2019-11-12 5:16 ` Greg Kroah-Hartman
2019-11-12 1:39 ` kernelci.org bot
2019-11-12 5:27 ` Greg Kroah-Hartman
2019-11-12 12:00 ` Jon Hunter
2019-11-12 13:27 ` Naresh Kamboju
2019-11-12 18:18 ` Guenter Roeck
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=20191111181300.050698565@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=akpm@linux-foundation.org \
--cc=haokexin@gmail.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).