Home » Mailing list archives » git-commits-head » 2008 » March » 15

[NET]: Fix tbench regression in 2.6.25-rc1

Previous thread: sched: simplify sched_slice() by Linux Kernel Mailing List on Friday, March 14, 2008 - 9:59 pm. (1 message)

Next thread: tifm_sd: DATA_CARRY is not boolean in tifm_sd_transfer_data() by Linux Kernel Mailing List on Saturday, March 15, 2008 - 12:59 pm. (1 message)
[view original message]
From: Linux Kernel Mailing List
Subject: [NET]: Fix tbench regression in 2.6.25-rc1
Date: Saturday, March 15, 2008 - 12:59 pm

Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=f1dd9c...
Commit:     f1dd9c379cac7d5a76259e7dffcd5f8edc697d17
Parent:     22626216c46f2ec86287e75ea86dd9ac3df54265
Author:     Zhang Yanmin <yanmin.zhang@intel.com>
AuthorDate: Wed Mar 12 22:52:37 2008 -0700
Committer:  David S. Miller <davem@davemloft.net>
CommitDate: Wed Mar 12 22:52:37 2008 -0700

    [NET]: Fix tbench regression in 2.6.25-rc1
    
    Comparing with kernel 2.6.24, tbench result has regression with
    2.6.25-rc1.
    
    1) On 2 quad-core processor stoakley: 4%.
    2) On 4 quad-core processor tigerton: more than 30%.
    
    bisect located below patch.
    
    b4ce92775c2e7ff9cf79cca4e0a19c8c5fd6287b is first bad commit
    commit b4ce92775c2e7ff9cf79cca4e0a19c8c5fd6287b
    Author: Herbert Xu <herbert@gondor.apana.org.au>
    Date:   Tue Nov 13 21:33:32 2007 -0800
    
        [IPV6]: Move nfheader_len into rt6_info
    
        The dst member nfheader_len is only used by IPv6.  It's also currently
        creating a rather ugly alignment hole in struct dst.  Therefore this patch
        moves it from there into struct rt6_info.
    
    Above patch changes the cache line alignment, especially member
    __refcnt. I did a testing by adding 2 unsigned long pading before
    lastuse, so the 3 members, lastuse/__refcnt/__use, are moved to next
    cache line. The performance is recovered.
    
    I created a patch to rearrange the members in struct dst_entry.
    
    With Eric and Valdis Kletnieks's suggestion, I made finer arrangement.
    
    1) Move tclassid under ops in case CONFIG_NET_CLS_ROUTE=y. So
       sizeof(dst_entry)=200 no matter if CONFIG_NET_CLS_ROUTE=y/n. I
       tested many patches on my 16-core tigerton by moving tclassid to
       different place. It looks like tclassid could also have impact on
       performance.  If moving tclassid before metrics, or just don't move
       tclassid, the ...
[ message continues ]
Previous thread: sched: simplify sched_slice() by Linux Kernel Mailing List on Friday, March 14, 2008 - 9:59 pm. (1 message)

Next thread: tifm_sd: DATA_CARRY is not boolean in tifm_sd_transfer_data() by Linux Kernel Mailing List on Saturday, March 15, 2008 - 12:59 pm. (1 message)