Home » Mailing list archives » linux-kernel » 2008 » March » 23

Re: Performance degradation measurement [was Re: [PATCH 06/37] Security: Separate task security context from task_struct [ver #34]]

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
[view in full thread]
From: David Howells <dhowells@...>
To: <Valdis.Kletnieks@...>
Cc: <dhowells@...>, <akpm@...>, <torvalds@...>, <Trond.Myklebust@...>, <chuck.lever@...>, <nfsv4@...>, <linux-kernel@...>, <linux-fsdevel@...>, <selinux@...>, <linux-security-module@...>
Subject: Re: Performance degradation measurement [was Re: [PATCH 06/37] Security: Separate task security context from task_struct [ver #34]]
Date: Sunday, March 23, 2008 - 8:40 am

Valdis.Kletnieks@vt.edu wrote:


That's a good point.  I missed that, probably because I *knew* it would be
slower with SELinux enabled, and so just assumed that it was.

That's really odd. There should be no disk accesses happening (the
pagecache/buffer cache is preloaded and noatime is turned on), the CPUs are
running at top speed at all times, and there's lots of free RAM available.
Network activity should also be minimal, so I'm not sure why there's so much
variance.

I'll re-run my tests from a kernel running a single bash and nothing else, see
if I can get more consistent data.

David
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
Messages in current thread:
[PATCH 00/37] Permit filesystem local caching [ver #34], David Howells, (Thu Feb 28, 8:43 pm)
[PATCH 36/37] NFS: Display local caching state [ver #34], David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 26/37] NFS: Define and create superblock-level object..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 24/37] NFS: Register NFS for caching and retrieve the..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 20/37] CacheFiles: A cache that backs onto a mounted ..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 34/37] NFS: Read pages from FS-Cache into an NFS inod..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 28/37] NFS: Use local disk inode cache [ver #34], David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 30/37] NFS: Add some new I/O event counters for FS-Ca..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 22/37] NFS: Add FS-Cache option bit and debug bit [ve..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 23/37] NFS: Permit local filesystem caching to be ena..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 21/37] NFS: Add comment banners to some NFS functions..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 31/37] NFS: FS-Cache page management [ver #34], David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 29/37] NFS: Invalidate FsCache page flags when cache ..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 27/37] NFS: Define and create inode-level cache objec..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 37/37] NFS: Add mount options to enable local caching..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 18/37] CacheFiles: Permit the page lock state to be m..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 35/37] NFS: Store pages from an NFS inode into a loca..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 33/37] NFS: nfs_readpage_async() needs to be accessib..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 25/37] NFS: Define and create server-level objects [v..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 32/37] NFS: Add read context retention for FS-Cache t..., David Howells, (Thu Feb 28, 8:46 pm)
[PATCH 13/37] FS-Cache: Provide an add_wait_queue_tail() fun..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 19/37] CacheFiles: Export things for CacheFiles [ver ..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 17/37] CacheFiles: Add a hook to write a single page ..., David Howells, (Thu Feb 28, 8:45 pm)
[PATCH 14/37] FS-Cache: Generic filesystem caching facility ..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 15/37] CacheFiles: Add missing copy_page export for i..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 16/37] CacheFiles: Be consistent about the use of map..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 05/37] Security: Change current-&gt;fs[ug]id to curre..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 12/37] FS-Cache: Recruit a couple of page flags for c..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 11/37] FS-Cache: Release page-&gt;private after faile..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 03/37] KEYS: Allow the callout data to be passed as a..., David Howells, (Thu Feb 28, 8:43 pm)
[PATCH 10/37] Security: Make NFSD work with detached securit..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 07/37] Security: De-embed task security record from t..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 09/37] Security: Allow kernel services to override LS..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 08/37] Security: Add a kernel_service object class to..., David Howells, (Thu Feb 28, 8:44 pm)
[PATCH 06/37] Security: Separate task security context from ..., David Howells, (Thu Feb 28, 8:44 pm)
Performance degradation measurement [was Re: [PATCH 06/37] S..., David Howells, (Wed Mar 19, 1:12 pm)
Re: Performance degradation measurement [was Re: [PATCH 06/3..., , (Sun Mar 23, 1:17 am)
Re: Performance degradation measurement [was Re: [PATCH 06/3..., David Howells, (Sun Mar 23, 8:40 am)
[PATCH 04/37] KEYS: Add keyctl function to get a security la..., David Howells, (Thu Feb 28, 8:43 pm)
[PATCH 02/37] KEYS: Check starting keyring as part of search..., David Howells, (Thu Feb 28, 8:43 pm)
[PATCH 01/37] KEYS: Increase the payload size when instantia..., David Howells, (Thu Feb 28, 8:43 pm)