| View previous topic :: View next topic |
| Author |
Message |
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
 Posted: Sun Aug 22, 2004 2:20 pm Post subject: 2.6.x sis5513 100%cpu@file-i/o [->bugzilla.kernel.org] |
 |
|
Im not sure if this is the right place to post such a question, but what is going wrong here:
After booting to the new 2.6.8-gentoo-r1 kernel i discover bad hd operation performance.
| Code: |
anfortas hefe # hdparm /dev/hda
/dev/hda:
multcount = 0 (off)
IO_support = 0 (default 16-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 256 (on)
geometry = 65535/16/63, sectors = 78140160, start = 0
anfortas hefe # hdparm -T /dev/hda
/dev/hda:
Timing buffer-cache reads: 1448 MB in 2.00 seconds = 722.30 MB/sec
anfortas hefe # hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 72 MB in 3.03 seconds = 23.74 MB/sec
|
I think the throuput is ok, but when timing with parameter -t the cpu usage (mainly sys) goes up to 100%. In comparison to the former 2.4.26-gentoo-r9 kernel where the cpu usage is very low when timing with parameter -t (not recognizable) but the throuput is the same! (same hdparm configuration only -d1 enabled).
For short:
2.6.8 kernel: hdparm -T .. -> nice throuput, 100% cpu.
2.6.8 kernel: hdparm -t .. -> nice throuput, 100% cpu.
2.4.26 kernel: hdparm -T .. -> nice (slightly better) throuput, 100% cpu.
2.4.26 kernel: hdparm -t .. -> nice throutput, 0% cpu.
To make you a bit familiar with my system:
| Code: |
anfortas root # lspci
0000:00:00.0 Host bridge: Silicon Integrated Systems [SiS] SiS645DX Host & Memory & AGP Controller
0000:00:01.0 PCI bridge: Silicon Integrated Systems [SiS] Virtual PCI-to-PCI bridge (AGP)
0000:00:02.0 ISA bridge: Silicon Integrated Systems [SiS] SiS962 [MuTIOL Media IO] (rev 04)
0000:00:02.1 SMBus: Silicon Integrated Systems [SiS] SiS961/2 SMBus Controller
0000:00:02.3 FireWire (IEEE 1394): Silicon Integrated Systems [SiS] FireWire Controller
0000:00:02.5 IDE interface: Silicon Integrated Systems [SiS] 5513 [IDE]
0000:00:02.6 Modem: Silicon Integrated Systems [SiS] AC'97 Modem Controller (rev a0)
0000:00:02.7 Multimedia audio controller: Silicon Integrated Systems [SiS] Sound Controller (rev a0)
0000:00:03.0 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
0000:00:03.1 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
0000:00:03.2 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 0f)
0000:00:03.3 USB Controller: Silicon Integrated Systems [SiS] USB 2.0 Controller
0000:00:04.0 Ethernet controller: Silicon Integrated Systems [SiS] SiS900 PCI Fast Ethernet (rev 91)
0000:00:0a.0 CardBus bridge: ENE Technology Inc CB1410 Cardbus Controller
0000:01:00.0 VGA compatible controller: ATI Technologies Inc Radeon R250 Lf [Radeon Mobility 9000 M9] (rev 01)
anfortas root # cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Pentium(R) 4 CPU 2.40GHz
stepping : 7
cpu MHz : 2390.792
cache size : 512 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe cid
bogomips : 4718.59
|
And this is my ide config in make menuconfig:
| Code: |
│ │ <*> ATA/ATAPI/MFM/RLL support │ │
│ │ <*> Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support │ │
│ │ --- Please see Documentation/ide.txt for help/info on IDE drives │ │
│ │ [ ] Support for SATA (deprecated; conflicts with libata SATA driver) │ │
│ │ [ ] Use old disk-only driver on primary interface │ │
│ │ <*> Include IDE/ATA-2 DISK support │ │
│ │ [ ] Use multi-mode by default │ │
│ │ < > PCMCIA IDE support │ │
│ │ < > Include IDE/ATAPI CDROM support │ │
│ │ < > Include IDE/ATAPI TAPE support (EXPERIMENTAL) │ │
│ │ < > Include IDE/ATAPI FLOPPY support │ │
│ │ <*> SCSI emulation support │ │
│ │ [ ] IDE Taskfile Access │ │
│ │ [ ] IDE Taskfile IO (EXPERIMENTAL) │ │
│ │ --- IDE chipset support/bugfixes │ │
│ │ < > generic/default IDE chipset support │ │
│ │ [ ] CMD640 chipset bugfix/support │ │
│ │ [ ] PNP EIDE support │ │
│ │ [*] PCI IDE chipset support │ │
│ │ [ ] Sharing PCI IDE interrupts support │ │
│ │ [ ] Boot off-board chipsets first support │ │
│ │ < > Generic PCI IDE Chipset Support │ │
│ │ < > OPTi 82C621 chipset enhanced support (EXPERIMENTAL) │ │
│ │ < > RZ1000 chipset bugfix/support │ │
│ │ [*] Generic PCI bus-master DMA support │ │
│ │ [ ] Force enable legacy 2.0.X HOSTS to use DMA │ │
│ │ [*] Use PCI DMA by default when available │ │
│ │ [ ] Enable DMA only for disks │ │
│ │ < > AEC62XX chipset support │ │
<some lines removed, but they are all disabled>
│ │ <*> SiS5513 chipset support │ │
│ │ < > SLC90E66 chipset support │ │
│ │ < > Tekram TRM290 chipset support │ │
│ │ < > VIA82CXXX chipset support │ │
│ │ [ ] Other IDE chipset support │ │
│ │ [ ] IGNORE word93 Validation BITS
|
Some bootmessages:
| Code: |
Aug 22 14:56:15 anfortas kernel: Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
Aug 22 14:56:15 anfortas kernel: ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
Aug 22 14:56:15 anfortas kernel: SIS5513: IDE controller at PCI slot 0000:00:02.5
Aug 22 14:56:15 anfortas kernel: ACPI: PCI interrupt 0000:00:02.5[A] -> GSI 16 (level, low) -> IRQ 16
Aug 22 14:56:15 anfortas kernel: SIS5513: chipset revision 0
Aug 22 14:56:15 anfortas kernel: SIS5513: not 100%% native mode: will probe irqs later
Aug 22 14:56:15 anfortas kernel: SIS5513: SiS 962/963 MuTIOL IDE UDMA133 controller
Aug 22 14:56:15 anfortas kernel: ide0: BM-DMA at 0x1000-0x1007, BIOS settings: hda:DMA, hdb:pio
Aug 22 14:56:15 anfortas kernel: ide1: BM-DMA at 0x1008-0x100f, BIOS settings: hdc:DMA, hdd:pio
Aug 22 14:56:15 anfortas kernel: hda: TOSHIBA MK4021GAS, ATA DISK drive
Aug 22 14:56:15 anfortas kernel: Using anticipatory io scheduler
Aug 22 14:56:15 anfortas kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Aug 22 14:56:15 anfortas kernel: hdc: QSI CD-RW/DVD-ROM SBW-241, ATAPI CD/DVD-ROM drive
Aug 22 14:56:15 anfortas kernel: ide1 at 0x170-0x177,0x376 on irq 15
Aug 22 14:56:15 anfortas kernel: hda: max request size: 128KiB
Aug 22 14:56:15 anfortas kernel: hda: 78140160 sectors (40007 MB), CHS=65535/16/63, UDMA(33)
Aug 22 14:56:15 anfortas kernel: /dev/ide/host0/bus0/target0/lun0: p1 p2 p3 < p5 p6 p7 p8 >
Aug 22 14:56:15 anfortas kernel: ide-scsi is deprecated for cd burning! Use ide-cd and give dev=/dev/hdX as device
Aug 22 14:56:15 anfortas kernel: scsi0 : SCSI host adapter emulation for IDE ATAPI devices
Aug 22 14:56:15 anfortas kernel: Vendor: QSI Model: CDRW/DVD SBW-241 Rev: VK02
Aug 22 14:56:15 anfortas kernel: Type: CD-ROM ANSI SCSI revision: 02
Aug 22 14:56:15 anfortas kernel: sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray
Aug 22 14:56:15 anfortas kernel: Uniform CD-ROM driver Revision: 3.20
|
Bunch of questions:
1. Is someone able to reproduce this issue?
2. Where do i get more information about kernel options and their (side)-effects? (The kernel-how-to on kernel.org seems to be very old.)
3. May this problem be submitted to the sis5513.c author?
4. I read the top of sis5513.c but did not understand what is said. May someone give me a hint?
5. There are a few other strange kernel messages, that appear in both 2.4.26 and 2.6.8 kernels, could they be informative in this issue? Here is an example (look at the timestamps):
| Code: |
Aug 22 15:23:57 anfortas kernel: APIC error on CPU0: 40(40)
Aug 22 15:24:28 anfortas last message repeated 53 times
|
I compiled and tried various configurations regarding smp and acpi, but the message still appears. I dont get any ooops or something.... may i ignore this message?
6. Is there someone familiar with phoenix-bioses? I got another question, but all the manuals and faqs dont say a word about it. Its a pain: The scrennshot in the manual shows certain options (these show up in my bios as well), but they only comment a few in the text.
7. I activeted pci hotplug in the kernel (no module). Could this be the problem?
And to make things clear, i am keen on switching to kernel 2.6 because i want to use Intel Pentium 4 clock modulation.
If it is needed i will submit my linux-2.6.8-gentoo-r1/.config aswell. Dont hesitate to ask.
Have a nice sunday folks, i like you to say anything that comes to your minds.
Last edited by hefe on Wed Aug 25, 2004 1:51 am; edited 2 times in total |
|
| Back to top |
|
 |
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
| Posted: Sun Aug 22, 2004 2:51 pm Post subject: Re: 2.6.8-gentoo-r1: sis5513 100% cpu hdparm -t /dev/hda |
|
|
| hefe wrote: |
7. I activeted pci hotplug in the kernel (no module). Could this be the problem?
|
Just tried it without pci hotplug, no difference -> 7. solved 
Now its your turn! |
|
| Back to top |
|
|
DaMouse
Apprentice
Joined: 18 Jul 2003
Posts: 233
|
| Posted: Sun Aug 22, 2004 3:39 pm Post subject: |
|
|
I just tried this on my SiS system (same IDE chipset driver) and hdparm -t hits 5% CPU max but -T goes up to 65% which isn't 100% and i'm unsure why it would be, lemme just try my sda disk  . Same kinda results with sda which is a SiS964 so perhaps yours is a bitta bodged, hope this proved slightly useful.
Gouranga!
-DaMouse
_________________
I am the dark lord DaMouse I eat souls and wear the dressing gown of evil. |
|
| Back to top |
|
|
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
| Posted: Sun Aug 22, 2004 4:42 pm Post subject: |
|
|
| DaMouse wrote: |
...yours is a bitta bodged...
|
What do you mean with that? (me german, no slang plugin )
Do you have a gericom laptop too?
Could you post your exact kernel version?
| DaMouse wrote: |
Same kinda results with sda which is a SiS964
|
Same like mine results or same like you previous (and imho better results as mine). Do you have scsi emulation enabled with that ide disk (im a bit confused, sorry)? could you give me more hints on how your config is? thanks alot.... i hope we'r getting closer to that thingy. |
|
| Back to top |
|
|
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
| Posted: Sun Aug 22, 2004 8:55 pm Post subject: 2.6.8-gentoo-r1: sis5513 100% cpu hdparm -t /dev/hda [solved |
|
|
Hi there!
Step 1: remove the /usr/src/linux -> <to old version> -link
Step 2: create a new symlink (/usr/src/linux that points to the correct source)
Step 3: emerge -C hdparm
Step 4: emerge hdparm
Step 5: fix the config stuff that emerge complains about
Now timing with hdparm -t parameter does not utilize the cpu (only a bit) -> great as i want it.
But: is there someone who can help me about the unsolved questions in my first post?
Thanks for help. |
|
| Back to top |
|
|
DaMouse
Apprentice
Joined: 18 Jul 2003
Posts: 233
|
| Posted: Sun Aug 22, 2004 9:14 pm Post subject: |
|
|
I'm not on a laptop, I was running a version of the kernel with staircase 7.I atop 2.6.8.1-mm3 and the sda disk was a true SATA controller. to get help on kernel options hitting "?" in the menu brings up help if available.
Gouranga!
-DaMouse
_________________
I am the dark lord DaMouse I eat souls and wear the dressing gown of evil. |
|
| Back to top |
|
|
DaMouse
Apprentice
Joined: 18 Jul 2003
Posts: 233
|
| Posted: Sun Aug 22, 2004 9:19 pm Post subject: |
|
|
Oh and the results for sda were similar to my previous results
Gouranga!
-DaMouse
_________________
I am the dark lord DaMouse I eat souls and wear the dressing gown of evil. |
|
| Back to top |
|
|
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
| Posted: Mon Aug 23, 2004 10:42 am Post subject: Re: 2.6.8-gentoo-r1: sis5513 100% cpu hdparm -t /dev/hda [so |
|
|
| hefe wrote: | Hi there!
Step 1: remove the /usr/src/linux -> <to old version> -link
Step 2: create a new symlink (/usr/src/linux that points to the correct source)
Step 3: emerge -C hdparm
Step 4: emerge hdparm
Step 5: fix the config stuff that emerge complains about
Now timing with hdparm -t parameter does not utilize the cpu (only a bit) -> great as i want it.
But: is there someone who can help me about the unsolved questions in my first post?
Thanks for help. |
This is bullshit! I was sure i ran on kernel 2.6.8 but i wasnt, and so this fix does not work. I thought i was a bit insider of computers, but actually i am not! This thing screws me up.
Status: not solved!
| DaMouse wrote: |
I'm not on a laptop, I was running a version of the kernel with staircase 7.I atop 2.6.8.1-mm3 and the sda disk was a true SATA controller. to get help on kernel options hitting "?" in the menu brings up help if available.
|
Maybe i have a sata controller too and the autodetection think its an 40c or 80c ide controller?
Is there a way to emerge a 2.6.8.1-mm3 kernel? or am i supposed to rip it by my own, please give me some assistance with it, i would like to try 2.6.8.1-mm3. Actually i will try everything possible to get this to work (on a 2.6.x kernel). (dma disk access should never utilize the cpu like that, arg... *sniff*)
regarding "?" in menuconfig: this help rocks, but sometimes it says "no help to that option" like that.
Thank you so far, i feel as if i am the only one with a gericom "masterpiece xl 2440" laptop. |
|
| Back to top |
|
|
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
|
| Back to top |
|
|
hefe
n00b
Joined: 18 Aug 2004
Posts: 58
Location: Germany
|
|
| Back to top |
|
|
|
Kernel & Hardware
