* ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time
@ 2006-11-24 21:39 gary.czek
2006-11-27 3:19 ` Tejun Heo
0 siblings, 1 reply; 5+ messages in thread
From: gary.czek @ 2006-11-24 21:39 UTC (permalink / raw)
To: linux-kernel
Hi, I have problem with my notebook Fujitsu-Siemens V8010. It has Intel
ICH6M chipset with SATA Controller. And SATA II disk Fujitsu MHT2040BH
with NCQ. If there is request on disk, iowait time of CPU gets to 100%
and whole system gets totally unresponsible. For example apt upgrade (of
average 10 packages totaling 30MB in .debs) gets 30 minutes. CPU iowait
time gets about 95% for whole 30 minutes.
My notebook details:
CPU: Intel Celeron M 1,4GHz
MEM: 256MB 333MHz
HDD: Fujitsu MHT2040BH SATA II, NCQ, 5400rpm, 8MB buffer
SWP: 512MB swap partition
Chipset: ICH6M 82801FBM
GPU: Intel i915GM integrated
kernel: 2.6.19-rc5
SATA Controller/disk driver: ata_piix and ahci tested, but results of
both were almost the same.
dmesg:
[ 12.746000] ata_piix 0000:00:1f.1: version 2.00ac6
[ 12.746000] ACPI: PCI Interrupt 0000:00:1f.1[A] -> GSI 18 (level, low) -> IRQ 18
[ 12.746000] PCI: Setting latency timer of device 0000:00:1f.1 to 64
[ 12.746000] ata1: PATA max UDMA/100 cmd 0x1F0 ctl 0x3F6 bmdma 0x1810 irq 14
[ 12.746000] ata2: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0x1818 irq 15
[ 12.746000] scsi0 : ata_piix
[ 13.049000] ata1.00: ATAPI, max UDMA/33
[ 13.201000] ata1.00: configured for UDMA/33
[ 13.201000] scsi1 : ata_piix
[ 13.201000] ata2: port disabled. ignoring.
[ 13.201000] ATA: abnormal status 0xFF on port 0x177
[ 13.204000] scsi 0:0:0:0: CD-ROM HL-DT-ST RW/DVD GCC-4243N 1.03 PQ: 0 ANSI: 5
[ 13.206000] sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray
[ 13.206000] Uniform CD-ROM driver Revision: 3.20
[ 13.206000] sr 0:0:0:0: Attached scsi CD-ROM sr0
[ 13.206000] ata_piix 0000:00:1f.2: MAP [ P0 P2 XX XX ]
[ 13.206000] ata_piix 0000:00:1f.2: invalid MAP value 128
[ 13.206000] ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 19 (level, low) -> IRQ 19
[ 13.206000] PCI: Setting latency timer of device 0000:00:1f.2 to 64
[ 13.206000] ata3: SATA max UDMA/133 cmd 0x2088 ctl 0x2082 bmdma 0x18B0 irq 19
[ 13.206000] ata4: SATA max UDMA/133 cmd 0x18A8 ctl 0x180E bmdma 0x18B8 irq 19
[ 13.206000] scsi2 : ata_piix
[ 13.359000] ata3.00: ATA-7, max UDMA/100, 78140160 sectors: LBA48 NCQ (depth 0/32)
[ 13.359000] ata3.00: ata3: dev 0 multi count 16
[ 13.362000] ata3.00: configured for UDMA/100
[ 13.362000] scsi3 : ata_piix
[ 13.513000] ATA: abnormal status 0x7F on port 0x18AF
[ 13.514000] scsi 2:0:0:0: Direct-Access ATA FUJITSU MHT2040B 0000 PQ: 0 ANSI: 5
[ 13.514000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 13.514000] sda: Write Protect is off
[ 13.514000] sda: Mode Sense: 00 3a 00 00
[ 13.514000] SCSI device sda: drive cache: write back
[ 13.514000] SCSI device sda: 78140160 512-byte hdwr sectors (40008 MB)
[ 13.514000] sda: Write Protect is off
[ 13.514000] sda: Mode Sense: 00 3a 00 00
[ 13.514000] SCSI device sda: drive cache: write back
[ 13.514000] sda: sda1 sda2 sda3
[ 13.574000] sd 2:0:0:0: Attached scsi disk sda
[ 13.880000] Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
[ 13.880000] ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
[ 13.880000] ide0: I/O resource 0x1F0-0x1F7 not free.
[ 13.880000] ide0: ports already in use, skipping probe
[ 13.880000] ide1: I/O resource 0x170-0x177 not free.
[ 13.880000] ide1: ports already in use, skipping probe
hdparm:
gary@ntb:~$ sudo hdparm -tT -d1 -v /dev/sda
/dev/sda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Inappropriate ioctl for device
IO_support = 0 (default 16-bit)
readonly = 0 (off)
readahead = 256 (on)
geometry = 4864/255/63, sectors = 78140160, start = 0
Timing cached reads: 1920 MB in 2.00 seconds = 960.43 MB/sec
Timing buffered disk reads: 88 MB in 3.02 seconds = 29.11 MB/sec
my measures has
average buffered disk: 22 MB/sec
maximal buffered disk: 34 MB/sec
On cnet is said that this hdd has maximum internal data transfer rate
(how fast it can actually pull data off the disk surface) of 53.9 MBps.
lshw:
gary@ntb:~$ sudo lshw -class ide -class storage -class disk
*-storage
description: Mass storage controller
product: PCIxx21 Integrated FlashMedia Controller
vendor: Texas Instruments
physical id: 9.3
bus info: pci@06:09.3
version: 00
width: 32 bits
clock: 33MHz
capabilities: storage bus_master cap_list
configuration: driver=tifm_7xx1
resources: iomemory:b0106000-b0107fff irq:17
*-ide
description: IDE interface
product: 82801FB/FBM/FR/FW/FRW (ICH6 Family) IDE Controller
vendor: Intel Corporation
physical id: 1f.1
bus info: pci@00:1f.1
logical name: scsi0
version: 04
width: 32 bits
clock: 33MHz
capabilities: ide bus_master emulated
configuration: driver=ata_piix
resources: ioport:1810-181f irq:18
*-cdrom UNCLAIMED
description: SCSI CD-ROM
product: RW/DVD GCC-4243N
vendor: HL-DT-ST
physical id: 0.0.0
bus info: scsi@0:0.0.0
version: 1.03
capabilities: removable
configuration: ansiversion=5
*-storage
description: SATA controller
product: 82801FBM (ICH6M) SATA Controller
vendor: Intel Corporation
physical id: 1f.2
bus info: pci@00:1f.2
logical name: scsi2
version: 04
width: 32 bits
clock: 66MHz
capabilities: storage ahci_1.0 bus_master cap_list emulated
configuration: driver=ata_piix
resources: ioport:2088-208f ioport:2080-2083 ioport:18a8-18af ioport:180c-180f ioport:18b0-18bf iomemory:b0040c00-b0040fff irq:19
*-disk
description: SCSI Disk
product: FUJITSU MHT2040B
vendor: ATA
physical id: 0.0.0
bus info: scsi@2:0.0.0
logical name: /dev/sda
version: 0000
serial: NR29T5A26VMV
size: 37GB
capabilities: partitioned partitioned:dos
configuration: ansiversion=5
*-volume:0
description: Linux filesystem partition
physical id: 1
bus info: scsi@2:0.0.0,1
logical name: /dev/sda1
capacity: 9538MB
capabilities: primary bootable
*-volume:1
description: Linux swap / Solaris partition
physical id: 2
bus info: scsi@2:0.0.0,2
logical name: /dev/sda2
capacity: 517MB
capabilities: primary nofs
*-volume:2
description: Linux filesystem partition
physical id: 3
bus info: scsi@2:0.0.0,3
logical name: /dev/sda3
capacity: 27GB
capabilities: primary
I have no other idea how to solve this problem. I'm not sure whether
disk uses correct driver. Friend told me that it may be solved by
increasing RAM (i plan buy additional 1024MB RAM for christmas).
I'm nearly hopeless. I'm trying to find the solution for more than
4months.
Thanks in advance for every hint.
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time
2006-11-24 21:39 ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time gary.czek
@ 2006-11-27 3:19 ` Tejun Heo
2006-11-30 17:06 ` gary.czek
0 siblings, 1 reply; 5+ messages in thread
From: Tejun Heo @ 2006-11-27 3:19 UTC (permalink / raw)
To: gary.czek; +Cc: linux-kernel
gary.czek wrote:
> Hi, I have problem with my notebook Fujitsu-Siemens V8010. It has Intel
> ICH6M chipset with SATA Controller. And SATA II disk Fujitsu MHT2040BH
> with NCQ. If there is request on disk, iowait time of CPU gets to 100%
> and whole system gets totally unresponsible. For example apt upgrade (of
> average 10 packages totaling 30MB in .debs) gets 30 minutes. CPU iowait
> time gets about 95% for whole 30 minutes.
>
> My notebook details:
> CPU: Intel Celeron M 1,4GHz
> MEM: 256MB 333MHz
> HDD: Fujitsu MHT2040BH SATA II, NCQ, 5400rpm, 8MB buffer
> SWP: 512MB swap partition
> Chipset: ICH6M 82801FBM
> GPU: Intel i915GM integrated
>
> kernel: 2.6.19-rc5
> SATA Controller/disk driver: ata_piix and ahci tested, but results of
> both were almost the same.
1. does 'mount -o remount,barrier=0 /' change anything?
2. 256MB is really small if you're running modern desktop environment.
Please post the result of 'vmstat 5' while the machine is really slow.
--
tejun
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time
2006-11-27 3:19 ` Tejun Heo
@ 2006-11-30 17:06 ` gary.czek
2006-12-01 0:50 ` Tejun Heo
0 siblings, 1 reply; 5+ messages in thread
From: gary.czek @ 2006-11-30 17:06 UTC (permalink / raw)
To: Tejun Heo; +Cc: linux-kernel
Tejun Heo <htejun@gmail.com> wrote:
> gary.czek wrote:
> > Hi, I have problem with my notebook Fujitsu-Siemens V8010. It has
> > Intel ICH6M chipset with SATA Controller. And SATA II disk Fujitsu
> > MHT2040BH with NCQ. If there is request on disk, iowait time of CPU
> > gets to 100% and whole system gets totally unresponsible. For
> > example apt upgrade (of average 10 packages totaling 30MB in .debs)
> > gets 30 minutes. CPU iowait time gets about 95% for whole 30
> > minutes.
> >
> > My notebook details:
> > CPU: Intel Celeron M 1,4GHz
> > MEM: 256MB 333MHz
> > HDD: Fujitsu MHT2040BH SATA II, NCQ, 5400rpm, 8MB buffer
> > SWP: 512MB swap partition
> > Chipset: ICH6M 82801FBM
> > GPU: Intel i915GM integrated
> >
> > kernel: 2.6.19-rc5
> > SATA Controller/disk driver: ata_piix and ahci tested, but results
> > of both were almost the same.
>
> 1. does 'mount -o remount,barrier=0 /' change anything?
>
> 2. 256MB is really small if you're running modern desktop
> environment. Please post the result of 'vmstat 5' while the machine
> is really slow.
>
First of all. I've reinstalled Ubuntu and installed Xfce instead of
Gnome. It seemed to me that it was much faster after reinstall. But as
I installed as much programs as I had installed before reinstall it
seems to me that it goes back to problems. Maybe it could by because of
I have running MySQL and PostgreSQL dbs. But they are accessed
minimally-not big queries and not so frequently (Only for use on
localhost). I use for some work postgres and for other mysql.
1. no, that changes nothing. Maybe I just didn't recognised the
difference.
2. if I coldn't find the real problem, I'll buy 1Gig RAM immediatelly.
"vmstat 5" when system is really slow shows following:
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
0 9 285904 3044 832 42088 75 46 247 98 484 746 5 2 80 13
0 4 285560 3628 692 43724 686 71 913 418 952 1651 18 3 0 79
0 3 285000 3616 688 43100 822 15 1431 766 939 1761 16 4 0 80
0 3 284372 3612 700 42344 703 0 1234 43 941 1697 18 3 7 72
1 2 284012 3176 580 42800 406 0 1054 74 834 1602 15 4 36 45
0 3 283820 3888 824 40044 413 23 1102 95 865 1767 17 3 0 80
0 1 286364 3904 900 41508 63 540 1023 629 928 1646 15 3 0 82
0 2 287500 4884 1120 44840 388 338 1606 431 902 1670 13 4 0 83
0 10 291148 3104 1296 46088 396 827 923 877 917 1521 10 3 0 87
0 3 292952 3728 1492 47524 259 474 730 606 889 1754 15 2 0 84
0 2 294960 3540 1588 47488 633 632 970 678 863 1630 12 2 0 86
0 2 297444 3828 1676 46496 648 674 774 735 886 1906 14 2 0 84
0 3 298184 3616 1684 46420 593 290 759 331 844 1792 12 3 0 85
0 5 297836 3852 1508 43576 632 168 990 229 814 1605 10 1 0 88
0 3 297996 4084 1240 41552 876 315 981 345 866 1466 7 2 0 91
0 5 296632 3272 1100 38324 1084 19 1593 59 878 1756 8 2 0 90
0 9 295768 3840 1128 38332 801 14 1244 170 870 1385 9 2 0 89
0 5 294748 3812 1068 37924 705 0 1053 61 844 1547 12 2 0 86
0 6 293664 3588 1100 37844 832 0 1206 23 848 1765 16 1 0 82
0 2 293040 4308 1152 37388 638 5 1070 101 680 1524 13 1 0 86
2 4 294140 3908 1248 37772 30 235 750 425 791 1539 14 3 0 83
0 2 295608 3876 1452 38568 185 356 676 464 883 1658 12 3 0 85
0 4 297092 3552 1500 38188 497 424 842 532 871 1433 9 2 0 89
0 4 298484 4004 1488 38136 612 416 806 563 866 1371 9 2 0 89
0 5 300004 4376 1476 38524 468 433 913 490 773 1475 12 1 0 86
0 3 303592 3948 1508 39556 286 844 993 1022 459 1457 16 1 0 82
0 2 305068 3644 1428 39000 399 385 868 482 871 1579 12 3 0 84
0 2 304912 3460 1464 40136 317 52 1200 139 710 1573 12 2 0 87
0 3 305848 3924 1808 39856 462 326 794 470 659 1491 13 3 0 84
0 9 312892 3260 1200 48304 354 1590 1477 5350 846 3194 8 6 0 87
0 12 314740 3356 1252 48252 747 478 1306 1416 643 1370 2 1 0 96
0 9 315192 3800 1380 46372 727 260 1390 1706 430 1484 2 1 0 97
0 5 313996 3716 1540 43844 826 49 1387 181 457 1311 0 1 0 99
0 9 313088 2928 1528 40200 1021 25 1742 67 435 1275 1 1 0 98
0 10 312608 3664 1484 38452 714 0 1438 436 391 1139 1 1 0 98
0 10 311692 3520 1564 37928 986 0 1876 87 431 1133 1 1 0 98
0 8 297540 3436 1532 38008 1000 0 1813 111 763 1360 3 2 0 95
0 10 296264 3636 1292 39056 796 0 2186 196 853 1456 1 2 0 97
0 10 295476 3948 984 40616 440 25 2082 162 783 1380 1 1 0 97
0 9 293580 25044 1088 44084 346 0 1915 159 373 1339 3 1 0 96
0 21 292148 16892 1220 45908 1228 0 1618 286 414 2652 1 2 0 97
0 20 290812 3420 1276 48468 1349 0 1862 14 575 1823 3 3 0 94
0 25 290764 3172 1036 45176 831 231 1337 289 824 3387 1 2 0 96
0 20 289884 4144 772 42756 802 22 1173 34 794 1227 0 2 0 98
0 23 289976 3612 644 44300 722 186 1204 211 595 1203 0 2 0 98
0 22 289108 3088 720 45608 694 0 1087 28 379 1092 3 0 0 97
0 24 288328 3824 844 44304 699 13 1266 61 382 1210 2 1 0 97
0 18 287232 3792 816 42800 1013 0 1258 46 392 1209 5 1 0 94
0 19 286932 4348 764 44516 598 104 1489 141 379 1127 25 0 0 75
0 17 286136 3924 836 44672 678 6 1015 71 403 1122 3 1 0 96
0 20 286872 3116 688 45728 566 318 1508 345 445 1017 2 1 0 96
0 18 288712 4008 736 45680 403 446 1121 474 759 950 2 1 0 97
0 22 288700 3068 652 46824 722 163 1489 208 684 1135 1 1 0 98
0 30 289812 3204 720 46408 471 307 1002 317 454 1240 7 1 0 92
0 26 290196 2940 744 45684 726 228 1573 304 467 1198 3 1 0 96
2 18 289556 3776 752 42448 825 46 1425 126 457 1552 15 1 0 85
0 15 288872 3684 772 41244 696 5 1326 30 423 1384 21 1 0 78
0 16 288168 3916 688 41116 661 5 1601 50 430 1266 12 1 0 88
0 15 287252 3460 768 40552 791 0 1431 58 456 1361 6 1 0 94
0 12 286428 3952 764 40156 694 8 1329 50 438 1330 4 1 0 95
0 13 286400 3472 796 40304 636 202 1585 235 450 1293 4 1 0 95
0 13 287240 3444 756 41528 398 295 1398 323 453 1202 10 1 0 88
0 10 287868 3996 816 42396 378 216 1378 274 459 1208 6 0 0 93
0 16 289792 2976 884 42992 404 474 1381 477 448 1302 8 1 0 90
1 13 289424 4032 772 41468 707 69 1407 128 914 1462 15 2 0 83
0 10 290736 3380 664 42468 508 434 1635 461 934 1551 31 1 0 68
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time
2006-11-30 17:06 ` gary.czek
@ 2006-12-01 0:50 ` Tejun Heo
2006-12-01 1:08 ` gary.czek
0 siblings, 1 reply; 5+ messages in thread
From: Tejun Heo @ 2006-12-01 0:50 UTC (permalink / raw)
To: gary.czek; +Cc: linux-kernel
gary.czek wrote:
> "vmstat 5" when system is really slow shows following:
>
> procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
> r b swpd free buff cache si so bi bo in cs us sy id wa
> 0 9 285904 3044 832 42088 75 46 247 98 484 746 5 2 80 13
> 0 4 285560 3628 692 43724 686 71 913 418 952 1651 18 3 0 79
> 0 3 285000 3616 688 43100 822 15 1431 766 939 1761 16 4 0 80
> 0 3 284372 3612 700 42344 703 0 1234 43 941 1697 18 3 7 72
> 1 2 284012 3176 580 42800 406 0 1054 74 834 1602 15 4 36 45
> 0 3 283820 3888 824 40044 413 23 1102 95 865 1767 17 3 0 80
> 0 1 286364 3904 900 41508 63 540 1023 629 928 1646 15 3 0 82
> 0 2 287500 4884 1120 44840 388 338 1606 431 902 1670 13 4 0 83
> 0 10 291148 3104 1296 46088 396 827 923 877 917 1521 10 3 0 87
> 0 3 292952 3728 1492 47524 259 474 730 606 889 1754 15 2 0 84
> 0 2 294960 3540 1588 47488 633 632 970 678 863 1630 12 2 0 86
> 0 2 297444 3828 1676 46496 648 674 774 735 886 1906 14 2 0 84
> 0 3 298184 3616 1684 46420 593 290 759 331 844 1792 12 3 0 85
> 0 5 297836 3852 1508 43576 632 168 990 229 814 1605 10 1 0 88
Your machine is thrashing. Working set size is over the available
memory and pages are continuously getting dropped and then brought back.
Run top and press 'M' after the list showed up. It will show who are
consuming all the memory. Adding 1G should solve the problem but just
another 256M will make a big difference too.
--
tejun
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time
2006-12-01 0:50 ` Tejun Heo
@ 2006-12-01 1:08 ` gary.czek
0 siblings, 0 replies; 5+ messages in thread
From: gary.czek @ 2006-12-01 1:08 UTC (permalink / raw)
To: Tejun Heo; +Cc: linux-kernel
On Fri, 01 Dec 2006 09:50:49 +0900
Tejun Heo <htejun@gmail.com> wrote:
> Your machine is thrashing. Working set size is over the available
> memory and pages are continuously getting dropped and then brought
> back. Run top and press 'M' after the list showed up. It will show
> who are consuming all the memory. Adding 1G should solve the problem
> but just another 256M will make a big difference too.
Well... The problem is solved.
Thanks a lot for your time.
I look forward for additional 1Gig.
Thanks a lot twice more time.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2006-12-01 1:09 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-11-24 21:39 ICH6M SATA Controller, SATA2 NCQ disk and high iowait CPU time gary.czek
2006-11-27 3:19 ` Tejun Heo
2006-11-30 17:06 ` gary.czek
2006-12-01 0:50 ` Tejun Heo
2006-12-01 1:08 ` gary.czek
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox