Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
my harddrive performance is sucking
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
kyefun
n00b
n00b


Joined: 21 Jun 2002
Posts: 7

PostPosted: Fri Jun 21, 2002 3:02 am    Post subject: Reply with quote

I have a Maxtor 20 GB 7200 RPM harrddrive
Just wondering if these results meet the Standards

Settings:
/dev/ide/host0/bus1/target0/lun0/disc:
setting 32-bit I/O support flag to 1
setting using_dma to 1 (on)
setting xfermode to 68 (UltraDMA mode4)
I/O support = 1 (32-bit)
using_dma = 1 (on)

Results:
/dev/ide/host0/bus1/target0/lun0/disc:
Timing buffer-cache reads: 128 MB in 0.96 seconds =133.33 MB/sec
Timing buffered disk reads: 64 MB in 2.47 seconds = 25.91 MB/sec

The Results Vary by Around 10MB
(highest was 140.83 for cached and 28 for non-cache)

Is this Normal?
_________________
Little White Asian
Back to top
View user's profile Send private message
delta407
Bodhisattva
Bodhisattva


Joined: 23 Apr 2002
Posts: 2876
Location: Chicago, IL

PostPosted: Fri Jun 21, 2002 3:26 am    Post subject: Reply with quote

133.33 MB/sec... you're maxing out your IDE bus. That's a Good Thing. :D
Back to top
View user's profile Send private message
kormoc
Apprentice
Apprentice


Joined: 17 Jun 2002
Posts: 272
Location: Seattle, WA

PostPosted: Wed Jun 26, 2002 9:38 am    Post subject: What the Reply with quote

What in the world??? My laptop can use udma10 mode!!! 300 meg /sec burst!!!
From hdparm -I /dev/hda
[code]
Capabilities:
LBA, IORDY(can be disabled)
ECC bytes: 46 Queue depth: 1
Standby timer values: spec'd by Vendor, with device specific minimum
r/w multiple sector transfer: Max = 16 Current = 16
Advanced power management level: 128
DMA: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4 udma10
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=120ns IORDY flow control=120ns
[/code]
Back to top
View user's profile Send private message
AutoBot
l33t
l33t


Joined: 22 Apr 2002
Posts: 968
Location: Usually Out

PostPosted: Wed Jun 26, 2002 4:06 pm    Post subject: Reply with quote

Are you sure it can actually use it though?
_________________
This message self destructed a long time ago.
Back to top
View user's profile Send private message
abhishek
Retired Dev
Retired Dev


Joined: 28 Jun 2002
Posts: 393
Location: Los Angeles, CA

PostPosted: Sun Jun 30, 2002 10:18 am    Post subject: Reply with quote

Ae there sime tests I can run that would show some effcts of dma? Here's waht I started with:
Code:
Timing buffer-cache reads:   128 MB in  0.80 seconds =160.60 MB/sec
 Timing buffered disk reads:  64 MB in 10.40 seconds =  6.15 MB/sec
Then I tryed hdparm -c 1 -d 1 /dev/hda and got
Code:
Timing buffer-cache reads:   128 MB in  0.81 seconds =157.44 MB/sec
 Timing buffered disk reads:  64 MB in  1.67 seconds = 38.39 MB/sec
Then setting udma mode 5 didnt change much. After setting that i got
Code:
Timing buffer-cache reads:   128 MB in  0.82 seconds =156.29 MB/sec
 Timing buffered disk reads:  64 MB in  1.70 seconds = 37.62 MB/sec
which is virtually no diff. but its cool that i got a 6x increase with that. Seagate 7200 80gig. btw, maybe this should be moved to tweaks, even though it started as hardware?
Back to top
View user's profile Send private message
abhishek
Retired Dev
Retired Dev


Joined: 28 Jun 2002
Posts: 393
Location: Los Angeles, CA

PostPosted: Sun Jun 30, 2002 10:23 am    Post subject: Reply with quote

c_kuzmanic wrote:
This is for those of you with a kt266/82XX chipset:

Contrary to what you might think this chipset was not fully supported until recently, when code submitted by Vojtech Pavlik was incorporated into the 2.4.19-pre9.patch . None of the kernels that you can emerge through portage have that patch, so if you want full support for that chipset you have to build your own patched kernel. Keep in mind that this patch is a work in progress.

So what you need to do is emerge vanilla sources, then head over to ftp.kernel.org get 2.4.19-pre9.patch and apply that patch to the vanilla sources.

Optionally you can also get the pre-emptive kernel patch and apply that AFTER you applied the patch mentioned above.

I did this and my system (Soyo Dragon +, kt266a/VT8233CE) runs smoother than ever:)

Really? my soyo dragon runs fine with 2.4.19-gentoo-r7. My drives work decently.
Back to top
View user's profile Send private message
felysium
n00b
n00b


Joined: 27 May 2002
Posts: 43
Location: Netherlands

PostPosted: Sun Jun 30, 2002 4:58 pm    Post subject: Reply with quote

data_the_android wrote:
c_kuzmanic wrote:

quote

Really? my soyo dragon runs fine with 2.4.19-gentoo-r7. My drives work decently.

The patch is already in gentoo-sources.

You only have to compile support for VIA chipset in the kernel.
I forgot to do that first and my system overall speed was really slow, although I switched dma on to the hdd...
Back to top
View user's profile Send private message
abhishek
Retired Dev
Retired Dev


Joined: 28 Jun 2002
Posts: 393
Location: Los Angeles, CA

PostPosted: Mon Jul 01, 2002 1:22 am    Post subject: Reply with quote

felysium wrote:

The patch is already in gentoo-sources.

You only have to compile support for VIA chipset in the kernel.
I forgot to do that first and my system overall speed was really slow, although I switched dma on to the hdd...

Cool. Go gentoo!
Back to top
View user's profile Send private message
cyc
Tux's lil' helper
Tux's lil' helper


Joined: 28 Jun 2002
Posts: 89

PostPosted: Wed Jul 03, 2002 11:59 am    Post subject: Reply with quote

damn, since i use xfs i cant use the gentoo-sources. or does that work
_________________
http://www.gentoo-de.org
Gentoo userpage
Back to top
View user's profile Send private message
cyc
Tux's lil' helper
Tux's lil' helper


Joined: 28 Jun 2002
Posts: 89

PostPosted: Wed Jul 03, 2002 6:02 pm    Post subject: Reply with quote

omg, is this normal:
hdparm -tT /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.50 seconds =256.00 MB/sec
Timing buffered disk reads: 64 MB in 1.59 seconds = 40.25 MB/sec

I have a 40 GB Seagate, 7200 rpm, using xfs how said before. seems to be great performance
_________________
http://www.gentoo-de.org
Gentoo userpage
Back to top
View user's profile Send private message
AutoBot
l33t
l33t


Joined: 22 Apr 2002
Posts: 968
Location: Usually Out

PostPosted: Wed Jul 03, 2002 10:27 pm    Post subject: Reply with quote

cyc wrote:
omg, is this normal:
hdparm -tT /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.50 seconds =256.00 MB/sec
Timing buffered disk reads: 64 MB in 1.59 seconds = 40.25 MB/sec

I have a 40 GB Seagate, 7200 rpm, using xfs how said before. seems to be great performance


That does appear to be great performance.
_________________
This message self destructed a long time ago.
Back to top
View user's profile Send private message
ee99ee2
Guru
Guru


Joined: 18 Jun 2002
Posts: 307
Location: Murfreesboro, TN, USA

PostPosted: Fri Jul 05, 2002 7:21 pm    Post subject: Reply with quote

WOAH

Here's a before
Code:
Timing buffer-cache reads: 128 MB in 0.42 seconds =159 MB/sec
Timing buffered disk reads: 64 MB in 2.76 seconds = 3.23 MB/sec


And amazeingly, here's after "hdparm -c3 -m16 -X69 -d1 -u1" on /dev/hde and /dev/hdg:

Code:
Timing buffer-cache reads: 128 MB in 0.42 seconds =160 MB/sec
Timing buffered disk reads: 64 MB in 2.76 seconds = 40.25 MB/sec


Both of those above are an average of about 4 tests on both hard drives.... none of them were off more then 1 MB/sec between every other one.

Question:
What's my bottlekneck that's causeing the Timing buffer-cache read to top out at 160 MB/sec? I've been people with over 200. This ATA/100 UDMA5 on 2 7200RPM Maxtor 60GB HD's w/ 2MB Cache.

Another question that's kinda scary:
I booted into single user mode to try this the first time, and when I rebooted back into normal mode, none of the settings were saved. In fact, I noticed that every time I changed it, it wouldn't be saved after a reboot. So I just aded the command to my /etc/conf.d/local.start file and it works fine now. BUT, one major thing... when I set it to UDMA, even though my mobo is suppost to support it, I get the following error in /var/log/messages:

Code:
ide2: Speed warnings UDMA 3/4/5 is not functional.
ide3: Speed warnings UDMA 3/4/5 is not functional.


I got to reading about it on Google, and I found that UDMA has been disabled in my IDE chipset b/c it's not stable? How can I get around this, and what would happen if I forced it to run at UDMA5 all the time, which I can do with the local.start script.

Is there somewhere I can put the same thing that's in my local.start script to get hdparm to change the HD mode before everything starts loading so my system will boot up faster?

-ee99ee2
Back to top
View user's profile Send private message
Sp4rky
n00b
n00b


Joined: 10 Jul 2002
Posts: 15

PostPosted: Wed Jul 10, 2002 7:19 pm    Post subject: Reply with quote

I'm having the same problem.

My mb (abit VT6X4) support ATA66 (that would be udma4).

My HD (Maxtor 91531U3) supports ATA66:

Code:
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 *udma4


But I get the following warning when I do hdparm -X 68:

Code:
ide0: Speed warnings UDMA 3/4/5 is not functional.


The performance i'm getting so far:

Code:

Timing buffer-cache reads:   128 MB in  0.89 seconds =143.82 MB/sec
 Timing buffered disk reads:  64 MB in  2.66 seconds = 24.06 MB/sec


When I read this thread, that isn't a very bad performance, but I wonder if it would be any better if I was able to set -X 68.

Could the CDRom drive (also on ide0) prevent this?

On ide1 there's a cd burner (hp9100+), so I think that is why they put (didn't built this machine myself) on ide0.

Someone an idea on how to force udma4 performance?

Ruud

[EDIT]
so oke i was a bit hasty with the keyboard. Turns out I didn't have VIA support compiled in the kernel. Did that and now -X68 doesn't give the error/warning. However, performance hasn't improved a bit[/EDIT]
_________________
| Asus A7M266-D | dual XP2400+ t-bred | 1024Mb PC2100 REG ECC | Geforce4 Ti4600 |
| 3Ware 7006-2 | 2x WD800JB | LiteOn LTR-48125w | HEC-425VD-PT power |
Back to top
View user's profile Send private message
whansard
n00b
n00b


Joined: 19 Jul 2002
Posts: 4

PostPosted: Thu Jul 25, 2002 2:31 pm    Post subject: Maxium transfer rate of drive. Reply with quote

Don't get all caught up in the numbers here. The point of this all isn't to get
the highest numbers you can, its really to get the the controller and chipset
set up to transfer data as fast as the drive can. As soon as you get the
controller set up to run as fast as the drive can, you will not see any more
significant performance increases. I have an old 7200 rpm udma 66 drive
that will only transfer data at 16 mb's a sec when configured correctly. Its
that the drive was one of the first to come out at that speed, and thats as
fast as the heads can read the data from the platters. I have a new one
thats udma 133 that reads at 46 mb's a sec.
You ought to test your computer at these different settings by running
something like
time dd if=/dev/zero of=1000mbfile bs=1000k count=1000
That will time your machine writing a 1 gig file. Be sure to delete it.
thats one way you can see the real performance differences.
By the way, I've found that dd transfers much faster with the larger
block sizes, so i changed the defaults in the code on my machine and
recompiled it. Call me crazy.
Back to top
View user's profile Send private message
swat
Tux's lil' helper
Tux's lil' helper


Joined: 21 Jul 2002
Posts: 144

PostPosted: Sat Aug 03, 2002 12:46 pm    Post subject: Reply with quote

Hmm..

Ok - I have tried applying optimizations:

/dev/hde:

Model=MAXTOR 6L060J3, FwRev=A93.0500, SerialNo=663201217072
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=DualPortCache, BuffSize=1819kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117266688
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=no WriteCache=enabled
Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1: 1 2 3 4 5


sudo /sbin/hdparm -u 1 -m 16 -c 1 -d 1 /dev/hde

Yet i still only get:

/dev/hde:
Timing buffer-cache reads: 128 MB in 1.02 seconds =125.00 MB/sec
Timing buffered disk reads: 64 MB in 1.67 seconds = 38.25 MB/sec

Any ideas why it isn't any better ? :(

swat
Back to top
View user's profile Send private message
dioxmat
Bodhisattva
Bodhisattva


Joined: 04 May 2002
Posts: 709
Location: /home/mat

PostPosted: Tue Aug 06, 2002 9:00 am    Post subject: Reply with quote

Quote:

/dev/hde:
Timing buffer-cache reads: 128 MB in 1.02 seconds =125.00 MB/sec
Timing buffered disk reads: 64 MB in 1.67 seconds = 38.25 MB/sec

Any ideas why it isn't any better


hey, this is quite good :)
you might want udma6, though I dont know if you will actually notice any difference...

btw, I see everyone here is using -m 16 ...
Always been using -m 32 (yes my disks do suppor that :)

btw, a quote from the hdparm manual :

Quote:

Western Digital recommends lower settings of 4 to 8 on many of their drives, due tiny (32kB) drive buffers and non-optimized buffering algorithms.
The -i flag can be used to find the maximum setting supported by an installed drive (look for MaxMultSect in the output).


this is probably not true any more with modern disks, but you never know :)
_________________
mat
Back to top
View user's profile Send private message
squareroot
n00b
n00b


Joined: 09 Aug 2002
Posts: 22
Location: Orlando, Florida

PostPosted: Sat Aug 10, 2002 3:43 pm    Post subject: Warning - file corruption Reply with quote

Sp4rky wrote:
I'm having the same problem.

My mb (abit VT6X4) support ATA66 (that would be udma4).

My HD (Maxtor 91531U3) supports ATA66:

Code:
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 *udma4


But I get the following warning when I do hdparm -X 68:

Code:
ide0: Speed warnings UDMA 3/4/5 is not functional.


The performance i'm getting so far:

Code:

Timing buffer-cache reads:   128 MB in  0.89 seconds =143.82 MB/sec
 Timing buffered disk reads:  64 MB in  2.66 seconds = 24.06 MB/sec


When I read this thread, that isn't a very bad performance, but I wonder if it would be any better if I was able to set -X 68.

Could the CDRom drive (also on ide0) prevent this?

On ide1 there's a cd burner (hp9100+), so I think that is why they put (didn't built this machine myself) on ide0.

Someone an idea on how to force udma4 performance?

Ruud

[EDIT]
so oke i was a bit hasty with the keyboard. Turns out I didn't have VIA support compiled in the kernel. Did that and now -X68 doesn't give the error/warning. However, performance hasn't improved a bit[/EDIT]



Sp4rky, you were lucky you didn't have faster drives. The reason for the limitation, and kernel option, is that a lot of machines with buggy Via chipsets can have "massive file corruption" 8O if UDMA modes > 2 (33MB/s) are enabled.

I found out the hard way over a year ago. Everything tested good until I transferred over 100 MB. at once at >32 MB/s. UDMA 2 can never sustain this rate. It is barely possible with UDMA 4, fairly common with UDMA 5. I did a good bit of testing settings before I used them for real work.

Until I had a disk where data came off the head fast enough to sustain that rate it wasn't even possible. (Forget the big numbers on the outside of the box.) If I hadn't had >100 MB. of RAM there wouldn't have been such a transfer anyway. Other people had no trouble until they installed a new sound card. It isn't worthwhile to find out what you can get away with.

Gentoo kernels have the Via option set by default. People, don't turn it off if you ever plan to run it on a machine with a Via 686a or 686b chip. (Some may still be sold AFAIK. ) The easy option for software is to permanently limit performance. Gentoo is different.
Back to top
View user's profile Send private message
Parabola
n00b
n00b


Joined: 29 Aug 2002
Posts: 54

PostPosted: Fri Sep 06, 2002 3:27 pm    Post subject: Reply with quote

Nice! My new numbers:

Code:

/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.54 seconds =237.04 MB/sec
 Timing buffered disk reads:  64 MB in  1.59 seconds = 40.25 MB/sec


I am also interested in if there is anything like this for the SCSI side. Currently my ATA is outperforming my SCSI. :(
_________________
Josh
Back to top
View user's profile Send private message
squanto
Guru
Guru


Joined: 20 Apr 2002
Posts: 524
Location: Rochester, NY, USA

PostPosted: Fri Sep 06, 2002 8:23 pm    Post subject: Reply with quote

Code:
Maxtor 60 gig 5400 drive:
/dev/hdb:
 Timing buffered disk reads:  64 MB in  2.25 seconds = 28.43 MB/sec
/dev/hdb:
 Timing buffer-cache reads:   128 MB in  0.50 seconds =254.47 MB/sec


WD Special Edition 80 gig 7200 drive with 8MB of cache:
/dev/hda:
 Timing buffered disk reads:  64 MB in  1.39 seconds = 45.98 MB/sec
/dev/hda:
 Timing buffer-cache reads:   128 MB in  0.48 seconds =267.22 MB/sec


I think I am saturating my bus :lol: pretty well.
I like my WD Special Edition :D

Andrew
Back to top
View user's profile Send private message
pilla
Administrator
Administrator


Joined: 07 Aug 2002
Posts: 7214
Location: Pelotas, BR

PostPosted: Fri Sep 06, 2002 10:24 pm    Post subject: Reply with quote

Don't forget that buffer-cache reads only use memory access (no access to the disk)

from man hdparm
Code:

-T     Perform  timings  of cache reads for benchmark and comparison pur-
              poses.  For meaningful results, this operation should be  repeated
              2-3  times  on  an otherwise inactive system (no other active pro-
              cesses) with at least a couple of megabytes of free memory.   This
              displays the speed of reading directly from the Linux buffer cache
              without disk access.  This measurement is essentially  an  indica-
              tion  of the throughput of the processor, cache, and memory of the
              system under test.  If the -t flag is also specified, then a  cor-
              rection  factor  based  on  the outcome of -T will be incorporated
              into the result reported for the -t operation.


This is an interesting link (I've posted in another thread also)
http://linux.oreillynet.com/pub/a/linux/2000/06/29/hdparm.html
Back to top
View user's profile Send private message
squanto
Guru
Guru


Joined: 20 Apr 2002
Posts: 524
Location: Rochester, NY, USA

PostPosted: Sat Sep 07, 2002 4:57 am    Post subject: Reply with quote

Bloody Bastard: thank you for enlightening me. I did not know that. Now it makes sense that my 2 drives would have similar ratings for buffered cache reads. :P

Andrew

Can anyone beat my 45.98MB / sec? just curious. I run ReiserFS with notail on my WD special edition, on Epox 8KHA+ board. :?:
Back to top
View user's profile Send private message
pilla
Administrator
Administrator


Joined: 07 Aug 2002
Posts: 7214
Location: Pelotas, BR

PostPosted: Sat Sep 07, 2002 1:24 pm    Post subject: Reply with quote

You're welcome 8)

BTW, you should run benchmarks more than one time to get relevant results. I would say at least 7 times (I would be more confortable with 20), but I'm not taking in account statisc rules -- just magic numbers.


squanto wrote:
Bloody Bastard: thank you for enlightening me. I did not know that. Now it makes sense that my 2 drives would have similar ratings for buffered cache reads. :P

Andrew

Can anyone beat my 45.98MB / sec? just curious. I run ReiserFS with notail on my WD special edition, on Epox 8KHA+ board. :?:
Back to top
View user's profile Send private message
squanto
Guru
Guru


Joined: 20 Apr 2002
Posts: 524
Location: Rochester, NY, USA

PostPosted: Sat Sep 07, 2002 3:32 pm    Post subject: Reply with quote

Bloody Bastard wrote:
You're welcome 8)

BTW, you should run benchmarks more than one time to get relevant results. I would say at least 7 times (I would be more confortable with 20), but I'm not taking in account statisc rules -- just magic numbers.


I ran it 8 times, and they were all within 1MB/sec of that number, but I can post all if you want :P , but anyways, not that it is a competition, I am just happy that my $107 didn't go to waste, but actually got me a high proformance drive for once.
Now if only I could afford an opteron.... :drool: :drool:

Code:
bash-2.05a$ su
Password:
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.40 seconds = 45.81 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.41 seconds = 45.26 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.39 seconds = 45.91 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.39 seconds = 46.04 MB/sec <-I beat myself
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.39 seconds = 45.98 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.40 seconds = 45.81 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.40 seconds = 45.88 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.39 seconds = 45.91 MB/sec
bash-2.05a# hdparm -t /dev/hda

/dev/hda:
 Timing buffered disk reads:  64 MB in  1.40 seconds = 45.85 MB/sec


bash-2.05a# uptime
 11:36am  up 2 days,  2:42,  1 user,  load average: 0.40, 0.83, 0.94
:D


Andrew
Back to top
View user's profile Send private message
TEB
n00b
n00b


Joined: 19 Sep 2002
Posts: 17

PostPosted: Thu Sep 19, 2002 2:14 pm    Post subject: Reply with quote

Hi, very interested thread. I have seen the same problems here.

I have 2 cases : 1 Compaq EVO N800c and a selfbuilt rack mounted server.

The Compaq EVO N800c uses an Intel 845 chipset on a vanilla RH7.3 kernel.
The selfbuilt server uses an Intel E7500 chipset on a tweaked 2.4.19 customkernel with 3ware 7500-12 controller support inbuilt.

Ok i used the following commands in HDparm for the ide-drive (maxtor 160gb 5400rpm):

[root@videoserver f1]# hdparm -X69 -d1 -u1 -m16 -c3 /dev/hda

/dev/hda:
setting 32-bit I/O support flag to 3
setting multcount to 16
setting unmaskirq to 1 (on)
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
setting xfermode to 69 (UltraDMA mode5)
multcount = 16 (on)
I/O support = 3 (32-bit w/sync)
unmaskirq = 1 (on)
using_dma = 0 (off)

The following benchmark on both /dev/hda through the ICH2 controller and /dev/sda through the 3ware controller:

[root@videoserver f1]# hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.28 seconds =457.14 MB/sec
Timing buffered disk reads: 64 MB in 14.51 seconds = 4.41 MB/sec

/dev/sda:
Timing buffer-cache reads: 128 MB in 0.29 seconds =441.38 MB/sec
Timing buffered disk reads: 64 MB in 0.48 seconds =133.33 MB/sec

I compiled the 2.4.19 kernel with support for the Intel Piix driver. And the 3ware card is on a 266MB max speed (64bitx33mhz). Could it be that there some switch to turn it from 32bit to 64bit ? so in fact its using 32bit mode now?

Anyway over to the portable. I was very dissapointet with the speed here. Its a 1.7ghz P4 with a 5400rpm toshiba drive. I hoped to get more than this:

[root@localhost root]# hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.35 seconds =365.71 MB/sec
Timing buffered disk reads: 64 MB in 11.33 seconds = 5.65 MB/sec

[root@localhost root]# hdparm -c1 -d1 -u1 -X69 /dev/hda

/dev/hda:
setting 32-bit I/O support flag to 1
setting unmaskirq to 1 (on)
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
setting xfermode to 69 (UltraDMA mode5)
I/O support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 0 (off)

Well i see the same problem here. Thou i use a stock kernel from RH.73 i would guess the 845 chipset was supported in DMA mode and not in Generic IDE mode.

When i do a hdparm -I /dev/hda i get:

[root@localhost root]# hdparm -I /dev/hda

/dev/hda:

non-removable ATA device, with non-removable media
Model Number: TOSHIBA MK4019GAX
Serial Number: 52IC0366T
Firmware Revision: FA001A
Standards:
Supported: 1 2 3 4 5
Likely used: 5
Configuration:
Logical max current
cylinders 16383 17475
heads 16 15
sectors/track 63 63
bytes/track: 0 (obsolete)
bytes/sector: 0 (obsolete)
current sector capacity: 16513875
LBA user addressable sectors = 78140160
Capabilities:
LBA, IORDY(can be disabled)
ECC bytes: 46 Queue depth: 1
Standby timer values: spec'd by standard, no device specific minimum
r/w multiple sector transfer: Max = 16 Current = 16
Advanced power management level: 128
DMA: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4 udma5
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=120ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* NOP cmd
* READ BUFFER cmd
* WRITE BUFFER cmd
* Host Protected Area feature set
* look-ahead
* write cache
* Power Management feature set
Security Mode feature set
* SMART feature set
SET MAX security extension
* Advanced Power Management feature set
Security:
Master password revision code = 65534
supported
not enabled
not locked
frozen
not expired: security count
not supported: enhanced erase
36min for SECURITY ERASE UNIT.
HW reset results:
CBLID- above Vih
Device num = 0 determined by CSEL
Checksum: correct

When i do a dmesg i get:

Linux version 2.4.18-3 (bhcompile@daffy.perf.redhat.com) (gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110)) #1 Thu Apr 18 07:37:53 EDT 2002
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 000000001ffd0000 (usable)
BIOS-e820: 000000001ffd0000 - 000000001fff0c00 (reserved)
BIOS-e820: 000000001fff0c00 - 000000001fffc000 (ACPI NVS)
BIOS-e820: 000000001fffc000 - 0000000020000000 (reserved)
On node 0 totalpages: 131024
zone(0): 4096 pages.
zone(1): 126928 pages.
zone(2): 0 pages.
Kernel command line: auto BOOT_IMAGE=linux ro root=303 BOOT_FILE=/boot/vmlinuz-2.4.18-3 hdc=ide-scsi
ide_setup: hdc=ide-scsi
Initializing CPU#0
Detected 1694.525 MHz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 3381.65 BogoMIPS
Memory: 513856k/524096k available (1119k kernel code, 9852k reserved, 775k data, 280k init, 0k highmem)
Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
Inode cache hash table entries: 32768 (order: 6, 262144 bytes)
Mount-cache hash table entries: 8192 (order: 4, 65536 bytes)
Buffer cache hash table entries: 32768 (order: 5, 131072 bytes)
Page-cache hash table entries: 131072 (order: 7, 524288 bytes)
CPU: Before vendor init, caps: 3febf9ff 00000000 00000000, vendor = 0
CPU: L1 I cache: 12K, L1 D cache: 8K
CPU: L2 cache: 512K
CPU: After vendor init, caps: 3febf9ff 00000000 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: After generic, caps: 3febf9ff 00000000 00000000 00000000
CPU: Common caps: 3febf9ff 00000000 00000000 00000000
CPU: Intel(R) Pentium(R) 4 Mobile CPU 1.70GHz stepping 04
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.40 (20010327) Richard Gooch (rgooch@atnf.csiro.au)
mtrr: detected mtrr type: Intel
PCI: PCI BIOS revision 2.10 entry at 0xf031f, last bus=3
PCI: Using configuration type 1
PCI: Probing PCI hardware
Unknown bridge resource 2: assuming transparent
PCI: Using IRQ router PIIX [8086/248c] at 00:1f.0
PCI: Found IRQ 10 for device 02:0e.1
PCI: Sharing IRQ 10 with 02:08.0
PCI: Sharing IRQ 10 with 02:0e.0
PCI: Sharing IRQ 10 with 02:0e.2
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
apm: BIOS version 1.2 Flags 0x03 (Driver version 1.16)
Starting kswapd
VFS: Diskquotas version dquot_6.5.0 initialized
pty: 2048 Unix98 ptys configured
Serial driver version 5.05c (2001-07-08) with MANY_PORTS MULTIPORT SHARE_IRQ SERIAL_PCI ISAPNP enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS02 at 0x03e8 (irq = 4) is a 16550A
Real Time Clock Driver v1.10e
block: 992 slots per queue, batch=248
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev f9
PCI: Device 00:1f.1 not available because of resource collisions
PIIX4: (ide_setup_pci_device:) Could not enable device.
hda: TOSHIBA MK4019GAX, ATA DISK drive
hdc: DW-28E, ATAPI CD/DVD-ROM drive
ide2: ports already in use, skipping probe
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 78140160 sectors (40008 MB), CHS=5168/240/63
ide-floppy driver 0.99.newide
Partition check:
hda: hda1 hda2 hda3
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
ide-floppy driver 0.99.newide
md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
md: Autodetecting RAID arrays.
md: autorun ...
md: ... autorun DONE.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
IP: routing cache hash table of 4096 buckets, 32Kbytes
TCP: Hash tables configured (established 32768 bind 32768)
Linux IP multicast router 0.06 plus PIM-SM
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
RAMDISK: Compressed image found at block 0
Freeing initrd memory: 121k freed
VFS: Mounted root (ext2 filesystem).
Journalled Block Device driver loaded
kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.
Freeing unused kernel memory: 280k freed
Adding Swap: 521632k swap-space (priority -1)
usb.c: registered new driver usbdevfs
usb.c: registered new driver hub
PCI: Found IRQ 10 for device 02:0e.0
PCI: Sharing IRQ 10 with 02:08.0
PCI: Sharing IRQ 10 with 02:0e.1
PCI: Sharing IRQ 10 with 02:0e.2
usb-ohci.c: USB OHCI at membase 0xe0888000, IRQ 10
usb-ohci.c: usb-02:0e.0, NEC Corporation USB
usb.c: new USB bus registered, assigned bus number 1
hub.c: USB hub found
hub.c: 3 ports detected
PCI: Found IRQ 10 for device 02:0e.1
PCI: Sharing IRQ 10 with 02:08.0
PCI: Sharing IRQ 10 with 02:0e.0
PCI: Sharing IRQ 10 with 02:0e.2
usb-ohci.c: USB OHCI at membase 0xe088a000, IRQ 10
usb-ohci.c: usb-02:0e.1, NEC Corporation USB (#2)
usb.c: new USB bus registered, assigned bus number 2
hub.c: USB hub found
hub.c: 2 ports detected
ehci-hcd.c: $Revision: 1.1 $ USB 2.0 'Enhanced' Host Controller (EHCI) Driver
ehci-hcd.c: block sizes: qh 96 qtd 96 itd 160 sitd 64
PCI: Found IRQ 10 for device 02:0e.2
PCI: Sharing IRQ 10 with 02:08.0
PCI: Sharing IRQ 10 with 02:0e.0
PCI: Sharing IRQ 10 with 02:0e.1
hcd.c: ehci-hcd @ 02:0e.2, PCI device 1033:00e0 (NEC Corporation)
hcd.c: PCI latency reduced to max 34
hcd.c: irq 10, pci mem e08b4000
usb.c: new USB bus registered, assigned bus number 3
ehci-dbg.c: ehci_start hcs_params 0x2395 dbg=0 cc=2 pcc=3 ports=5
ehci-dbg.c: 02:0e.2: ehci_start portroute 1 0 1 0 0
ehci-dbg.c: ehci_start hcc_params 0x0002 caching 0 uframes 256/512/1024
ehci-hcd.c: reset 80002 cmd (park)=0 ithresh=8 period=1024 Reset HALT
ehci-hcd.c: init 10000 cmd (park)=0 ithresh=1 period=1024 HALT
ehci-hcd.c: USB 2.0 support enabled, EHCI rev 0.95
hcd.c: 02:0e.2 root hub device address 1
hub.c: USB hub found
hub.c: 5 ports detected
EXT3 FS 2.4-0.9.17, 10 Jan 2002 on ide0(3,3), internal journal
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.17, 10 Jan 2002 on ide0(3,1), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
SCSI subsystem driver Revision: 1.00
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
Vendor: TEAC Model: DW-28E Rev: A.0B
Type: CD-ROM ANSI SCSI revision: 02
parport0: PC-style at 0x378 (0x778) [PCSPP,TRISTATE,EPP]
parport0: irq 7 detected
eepro100.c:v1.09j-t 9/29/99 Donald Becker http://www.scyld.com/network/eepro100.html
eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin <saw@saw.sw.com.sg> and others
PCI: Found IRQ 10 for device 02:08.0
PCI: Sharing IRQ 10 with 02:0e.0
PCI: Sharing IRQ 10 with 02:0e.1
PCI: Sharing IRQ 10 with 02:0e.2
eth0: OEM i82557/i82558 10/100 Ethernet, 00:08:02:64:9C:D7, IRQ 10.
Board assembly 000000-000, Physical connectors present: RJ45
Primary interface chip i82555 PHY #1.
General self-test: passed.
Serial sub-system self-test: passed.
Internal registers self-test: passed.
ROM checksum self-test: passed (0x04f4518b).
Linux Kernel Card Services 3.1.22
options: [pci] [cardbus] [pm]
PCI: Found IRQ 11 for device 02:06.0
Yenta IRQ list 0298, PCI irq11
Socket status: 30000006
cs: IO port probe 0x0c00-0x0cff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
Linux agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 439M
agpgart: Detected Intel i845 chipset
agpgart: AGP aperture is 256M @ 0x60000000
[drm] AGP 0.99 on Unknown @ 0x60000000 256MB
[drm] Initialized radeon 1.1.1 20010405 on minor 0
Intel 810 + AC97 Audio, version 0.21, 07:42:52 Apr 18 2002
PCI: Found IRQ 5 for device 00:1f.5
PCI: Sharing IRQ 5 with 02:04.0
PCI: Setting latency timer of device 00:1f.5 to 64
i810: Intel ICH3 found at IO 0x4400 and 0x4000, IRQ 5
i810_audio: Audio Controller supports 6 channels.
ac97_codec: AC97 Audio codec, id: 0x4144:0x5363 (Unknown)
i810_audio: AC'97 codec 0 Unable to map surround DAC's (or DAC's not present), total channels = 2
ide0: unexpected interrupt, status=0x58, count=1
ide0: unexpected interrupt, status=0x58, count=2
ide0: unexpected interrupt, status=0x58, count=3
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: Speed warnings UDMA 3/4/5 is not functional.
ide0: unexpected interrupt, status=0x58, count=4
ide0: Speed warnings UDMA 3/4/5 is not functional.


Anyone have any idea what i should do to add DMA support to the kernel for both machines? Even on the portable it locks up when i try to send much data on the pure-ftpd thats running on it. Well it doesnt lockup but it sorta stops for a few seq's even if TOP says pureftp uses 25% cpu and not 100%

best regards dragmore
Back to top
View user's profile Send private message
pilla
Administrator
Administrator


Joined: 07 Aug 2002
Posts: 7214
Location: Pelotas, BR

PostPosted: Thu Sep 19, 2002 3:22 pm    Post subject: Reply with quote

The problem may be in your network card. eepro100 is known by having some problems.... make a search in Gentoo forums and I 'm sure you'll find some valuable info about it.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6, 7, 8, 9  Next
Page 3 of 9

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum