mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: rwhron@earthlink.net
To: linux-kernel@vger.kernel.org, ltp-list@lists.sourceforge.net
Subject: VM tests on 2.4.14 2.4.15-pre[12] and Athlon opt
Date: Fri, 9 Nov 2001 20:46:22 -0500	[thread overview]
Message-ID: <20011109204622.B5791@earthlink.net> (raw)


Summary:	Athlon optimized kernel uses higher percentage of
		CPU (36% vs ~ 25%) and takes a little longer on memory 
		fill test.  Athlon kernel a little faster on mmap test.

Test:	Fill and write 80% of virtual memory with "mtest01 -w -p80"
	mmap, touch, msync, and munmap 500000 pages "mmap001 -m 500000"
	Listen to mp3 sampled at 128k. (usual test).
	Test with/without Athlon processor configuration.

There were no mp3blaster skips on the mmap001 tests.

config differences for athlon:

diff p5 k7
< CONFIG_M586=y
> CONFIG_MK7=y

< CONFIG_X86_L1_CACHE_SHIFT=5
< CONFIG_X86_USE_STRING_486=y
< CONFIG_X86_ALIGNMENT_16=y

> CONFIG_X86_L1_CACHE_SHIFT=6
> CONFIG_X86_TSC=y
> CONFIG_X86_GOOD_APIC=y
> CONFIG_X86_USE_3DNOW=y
> CONFIG_X86_PGE=y
> CONFIG_X86_USE_PPRO_CHECKSUM=y


2.4.14

mp3 played 248 of 301 seconds (82%)

Averages for 10 mtest01 runs
bytes allocated:                    1236900249
User time (seconds):                2.014
System time (seconds):              7.045
Elapsed (wall clock) time:          30.853
Percent of CPU this job got:        28.90
Major (requiring I/O) page faults:  106.5
Minor (reclaiming a frame) faults:  302773.1

Average for 5 mmap001 runs
bytes allocated:                    2048000000
User time (seconds):                19.518
System time (seconds):              18.276
Elapsed (wall clock seconds) time:  186.31
Percent of CPU this job got:        19.80
Major (requiring I/O) page faults:  500163.4
Minor (reclaiming a frame) faults:  42.6


2.4.15-pre1

mp3 played 257 of 303 seconds (84%)

Averages for 10 mtest01 runs
bytes allocated:                    1233649664
User time (seconds):                2.060
System time (seconds):              5.280
Elapsed (wall clock) time:          30.314
Percent of CPU this job got:        23.70
Major (requiring I/O) page faults:  106.3
Minor (reclaiming a frame) faults:  301976.7

Average for 5 mmap001 runs
bytes allocated:                    2048000000
User time (seconds):                19.360
System time (seconds):              17.950
Elapsed (wall clock seconds) time:  188.15
Percent of CPU this job got:        19.40
Major (requiring I/O) page faults:  500165.2
Minor (reclaiming a frame) faults:  41.0


A note from Alan Cox on lkml said 2.4.14 should handle buggy VIA 
chipsets.  My KT133A now runs without oops compiled as Athlon.  
(kernels <= 2.4.9 didn't on my IWill KK266 mobo).

The next two tests are athlon optimized.  To me, the biggest
revelation is the higher "Percent of CPU this job got" on the 
mtest01 runs.


2.4.15-pre1 - athlon optimized

mp3 played 275 of 317 seconds (86%)

Averages for 10 mtest01 runs
bytes allocated:                    1238577971
User time (seconds):                5.475
System time (seconds):              6.032
Elapsed (wall clock) time:          31.683
Percent of CPU this job got:        36.00
Major (requiring I/O) page faults:  106.4
Minor (reclaiming a frame) faults:  303181.0

Average for 5 mmap001 runs
bytes allocated:                    2048000000
User time (seconds):                19.218
System time (seconds):              18.214
Elapsed (wall clock seconds) time:  182.07
Percent of CPU this job got:        20.00
Major (requiring I/O) page faults:  500163.4
Minor (reclaiming a frame) faults:  42.6


2.4.15-pre2 - athlon optimized

mp3 played 268 of 314 seconds (85%)

Averages for 10 mtest01 runs
bytes allocated:                    1235851673
User time (seconds):                5.447
System time (seconds):              6.099
Elapsed (wall clock) time:          31.400
Percent of CPU this job got:        36.40
Major (requiring I/O) page faults:  106.8
Minor (reclaiming a frame) faults:  302515.9

Average for 5 mmap001 runs
bytes allocated:                    2048000000
User time (seconds):                19.566
System time (seconds):              18.472
Elapsed (wall clock seconds) time:  183.45
Percent of CPU this job got:        20.20
Major (requiring I/O) page faults:  500164.4
Minor (reclaiming a frame) faults:  41.8

-- 
Randy Hron


                 reply	other threads:[~2001-11-10  1:44 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20011109204622.B5791@earthlink.net \
    --to=rwhron@earthlink.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ltp-list@lists.sourceforge.net \
    /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

all inboxes | Powered by JetHome®