From: Orion Poplawski <orion@cora.nwra.com>
To: linux-kernel@vger.kernel.org
Subject: Slow IDE disk write performance with 2.6.6-bk1 on tyan S2882 dual proc opteron
Date: Thu, 13 May 2004 16:29:43 -0600 [thread overview]
Message-ID: <c80ssq$cn7$1@sea.gmane.org> (raw)
I'm testing out a brand new dual processor Opteron 244 system with 4GB
RAM and Tyan S2882 motherboard and finding slow performance writing to
the IDE disk. Install is 64-bit Fedora Core 2 Test 3 with the latest
development updates. I've seen similar performance with the standard
Fedora kernel and the 2.6.6 kernel, as well as the 2.6.6-bk1 kernel
which these reported results are from. I've also tried "acpi=off"
without much effect, though I didn't disable acpi in the bios.
The following script:
opcontrol --start
dd if=/dev/zero of=/dev/hda8 bs=1M count=5000
sync
opcontrol --stop
takes the following time:
real 10m59.788s
user 0m0.077s
sys 0m22.934s
or about 7.6 MB/s which seems quite slow.
bonnie++ to an ext3 filesystem reports 16590 KB/s writing and 43606 KB/s
reading.
hdparm -t /dev/hda reports about 40MB/s reading, so it seems to be a
write issue.
/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 256 (on)
geometry = 16383/255/63, sectors = 156301488, start = 0
oprofile of the test scripts gives (I hope this is useful, I've never
run oprofile before):
CPU: AMD64 processors, speed 1792.39 MHz (estimated)
Counted RETIRED_INSNS events (Retired instructions (includes exceptions,
interrupts, re-syncs)) with a unit mask of 0x00 (No unit mask) count 100000
RETIRED_INSNS:...|
samples| %|
------------------
19896771 99.6125 vmlinux-2.6.6-bk1
65018 0.3255 oprofiled
11850 0.0593 oprofile
235 0.0012 libc-2.3.3.so
115 5.8e-04 bash
54 2.7e-04 ld-2.3.3.so
47 2.4e-04 ext3
36 1.8e-04 jbd
samples % symbol name
19654156 98.7806 poll_idle
12981 0.0652 free_block
10297 0.0518 mempool_alloc
9860 0.0496 kmem_cache_alloc
9439 0.0474 __make_request
8722 0.0438 rb_next
8498 0.0427 ll_back_merge_fn
8380 0.0421 as_merge
7716 0.0388 .text.lock.char_dev
7578 0.0381 __block_prepare_write
7370 0.0370 .text.lock.ll_rw_blk
7209 0.0362 __block_write_full_page
6908 0.0347 kmem_cache_free
6270 0.0315 end_buffer_async_write
5973 0.0300 blk_rq_map_sg
5614 0.0282 blk_recount_segments
4866 0.0245 generic_make_request
4865 0.0245 cache_alloc_refill
4762 0.0239 mempool_free
4649 0.0234 submit_bio
4131 0.0208 as_merged_request
3835 0.0193 bio_alloc
3084 0.0155 as_latter_request
2903 0.0146 drop_buffers
2659 0.0134 radix_tree_delete
2532 0.0127 init_buffer_head
2335 0.0117 wake_up_buffer
1976 0.0099 __delay
1940 0.0098 buffered_rmqueue
1858 0.0093 __block_commit_write
1816 0.0091 radix_tree_tag_set
1813 0.0091 radix_tree_tag_clear
1759 0.0088 free_pages_bulk
1742 0.0088 end_bio_bh_io_sync
1736 0.0087 __rmqueue
1731 0.0087 drive_stat_acct
1620 0.0081 __set_page_dirty_nobuffers
1535 0.0077 create_buffers
1517 0.0076 radix_tree_insert
1447 0.0073 as_add_arq_hash
1360 0.0068 shrink_list
1335 0.0067 generic_file_aio_write_nolock
1329 0.0067 elv_latter_request
1305 0.0066 test_set_page_writeback
1271 0.0064 free_hot_cold_page
1269 0.0064 shrink_cache
1257 0.0063 cache_init_objs
1247 0.0063 page_waitqueue
1239 0.0062 release_pages
1228 0.0062 submit_bh
1210 0.0061 bio_endio
1168 0.0059 __end_that_request_first
1163 0.0058 mempool_free_slab
1153 0.0058 blkdev_get_block
1091 0.0055 elv_try_last_merge
1065 0.0054 unlock_page
System info:
00:07.1 IDE interface: Advanced Micro Devices [AMD] AMD-8111 IDE (rev
03) (prog-if 8a [Master SecP PriP])
Subsystem: Advanced Micro Devices [AMD] AMD-8111 IDE
Flags: bus master, medium devsel, latency 32
I/O ports at ffa0 [size=16]
Disk is a WDC WD800JB-00ETA0
----------AMD BusMastering IDE Configuration----------------
Driver Version: 2.13
South Bridge: 0000:00:07.1
Revision: IDE 0x3
Highest DMA rate: UDMA133
BM-DMA base: 0xffa0
PCI clock: 33.3MHz
-----------------------Primary IDE-------Secondary IDE------
Prefetch Buffer: yes yes
Post Write Buffer: yes yes
Enabled: yes yes
Simplex only: no no
Cable Type: 80w 40w
-------------------drive0----drive1----drive2----drive3-----
Transfer Mode: UDMA PIO PIO PIO
Address Setup: 30ns 120ns 120ns 120ns
Cmd Active: 90ns 90ns 480ns 480ns
Cmd Recovery: 30ns 30ns 480ns 480ns
Data Active: 90ns 330ns 330ns 330ns
Data Recovery: 30ns 270ns 270ns 270ns
Cycle Time: 20ns 600ns 600ns 600ns
Transfer Rate: 99.9MB/s 3.3MB/s 3.3MB/s 3.3MB/s
Any help would be greatly appreciated.
- Orion
reply other threads:[~2004-05-13 22:30 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='c80ssq$cn7$1@sea.gmane.org' \
--to=orion@cora.nwra.com \
--cc=linux-kernel@vger.kernel.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