[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <bug-86471-13602-ApLc0iLpJM@https.bugzilla.kernel.org/>
Date: Thu, 23 Oct 2014 16:36:01 +0000
From: bugzilla-daemon@...zilla.kernel.org
To: linux-ext4@...r.kernel.org
Subject: [Bug 86471] drastic reduction of write performance after the
termination of mysqld
https://bugzilla.kernel.org/show_bug.cgi?id=86471
--- Comment #6 from Markus Königshaus <m.koenigshaus@....de> ---
I think I have found the error. The kernel calls apparently for the same page
dec_zone_page_state on (page, NR_FILE_DIRTY) twice, which means that the dirty
page counter underflows (is getting negative):
cat /proc/meminfo
MemTotal: 511492 kB
MemFree: 406972 kB
MemAvailable: 450536 kB
Buffers: 1840 kB
Cached: 52140 kB
SwapCached: 0 kB
Active: 54348 kB
Inactive: 34456 kB
Active(anon): 34904 kB
Inactive(anon): 5440 kB
Active(file): 19444 kB
Inactive(file): 29016 kB
Unevictable: 0 kB
Mlocked: 0 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 511492 kB
LowFree: 406972 kB
SwapTotal: 0 kB
SwapFree: 0 kB
>> Dirty: 4294967148 kB
Writeback: 0 kB
AnonPages: 34844 kB
Mapped: 13992 kB
Shmem: 5520 kB
Slab: 9196 kB
SReclaimable: 4976 kB
SUnreclaim: 4220 kB
KernelStack: 696 kB
PageTables: 1024 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
WritebackTmp: 0 kB
CommitLimit: 255744 kB
Committed_AS: 365452 kB
VmallocTotal: 499712 kB
VmallocUsed: 3512 kB
VmallocChunk: 495964 kB
The writeback - mechanism thinks that too much pages are dirty and disables the
writeback - cache.
Who can I send this bug?
Markus
--
You are receiving this mail because:
You are watching the assignee of the bug.--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists