Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
AMD64 system slow/unresponsive during disk access...
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... , 36, 37, 38  Next  
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo on AMD64
View previous topic :: View next topic  
Author Message
wazoo42
Apprentice
Apprentice


Joined: 13 Apr 2004
Posts: 165

PostPosted: Tue Aug 18, 2009 8:12 pm    Post subject: Reply with quote

Jens Axboe's patch (post 397 in the link MageSlayer gave above) worked wonders for my amd64 install (dual core opteron) with vanilla 2.6.30.4. I'll work on getting some concrete numbers, hopefully they'll backup the improved responsiveness I see in KDE4.
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Tue Aug 18, 2009 8:27 pm    Post subject: Reply with quote

wazoo42 wrote:
Jens Axboe's patch (post 397 in the link MageSlayer gave above) worked wonders for my amd64 install (dual core opteron) with vanilla 2.6.30.4. I'll work on getting some concrete numbers, hopefully they'll backup the improved responsiveness I see in KDE4.
That's interesting! Are you using the patch with NCQ or without. Can you post the output of:
Code:
cat /sys/block/*/queue/nr_requests
Back to top
View user's profile Send private message
MageSlayer
Apprentice
Apprentice


Joined: 26 Jul 2007
Posts: 228
Location: Ukraine

PostPosted: Tue Aug 18, 2009 8:27 pm    Post subject: Reply with quote

Yes. It would be great if you could provide some interbench numbers.
Back to top
View user's profile Send private message
wazoo42
Apprentice
Apprentice


Joined: 13 Apr 2004
Posts: 165

PostPosted: Wed Aug 19, 2009 12:32 am    Post subject: Reply with quote

It appears I'm using NCQ and the CFQ i/o scheduler.

Code:
benjfitz linux # cat /sys/block/*/queue/nr_requests
128
128
128
128
128
128
128
128

benjfitz linux # cat /sys/block/sda/queue/scheduler
noop anticipatory deadline [cfq]
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Wed Aug 19, 2009 12:55 am    Post subject: Reply with quote

wazoo42 wrote:
It appears I'm using NCQ and the CFQ i/o scheduler.

Code:
benjfitz linux # cat /sys/block/*/queue/nr_requests
128
128
128
128
128
128
128
128

benjfitz linux # cat /sys/block/sda/queue/scheduler
noop anticipatory deadline [cfq]
That's in contrast with what other people are seeing. hmmm...
Back to top
View user's profile Send private message
3Towers
n00b
n00b


Joined: 27 Mar 2004
Posts: 8

PostPosted: Sun Aug 23, 2009 3:34 pm    Post subject: Reply with quote

I have some more for you to test. For quite some time I had the same problems (unresponsive UI during IO operations and an almost completely freezed system when I was trying to rip a DVD with read errors). I tried several options already discussed in this thread (including the kernel ncq patch) with no success. Today I built a completely new kernel with the following options changed:
- disabled SMT (HyperThreading support) as my Q6700 doen't has HT
- enabled preemptile RCU
- enabled low latency desktop (forced preemption)
- disabled the IDE layer (only compiled libata SATA/PATA drivers)
additionally I swtched SATA mode in my BIOS to AHCI (IDE mode before) and set the RTC to 64bit.

With these settings all my problems are gone now! Only drawback: I'm not able to boot up Windowx XP with these BIOS settings if I need it again. *g*
When I have some more time I will try to figure out which option is responsible for this success and post it here.

Hope this can help some of you! :roll:
Back to top
View user's profile Send private message
energyman76b
Advocate
Advocate


Joined: 26 Mar 2003
Posts: 2031
Location: Germany

PostPosted: Sun Aug 23, 2009 5:28 pm    Post subject: Reply with quote

disable smt and ahci probably ;)

I don't have forced preemption nor rcu preemption and I don't suffer from long lags anymore. For a long time.
_________________
AidanJT wrote:

Libertardian denial of reality is wholly unimpressive and unconvincing, and simply serves to demonstrate what a bunch of delusional fools they all are.

Satan's got perfectly toned abs and rocks a c-cup.
Back to top
View user's profile Send private message
kernelOfTruth
Watchman
Watchman


Joined: 20 Dec 2005
Posts: 5720
Location: Vienna, Austria; Germany; hello world :)

PostPosted: Sun Aug 23, 2009 7:30 pm    Post subject: Reply with quote

energyman76b wrote:
disable smt and ahci probably ;)

I don't have forced preemption nor rcu preemption and I don't suffer from long lags anymore. For a long time.


I got to test that with smt, thanks !

you don't need ahci anyways with your SCSI disks ;)
_________________
https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.3.0-r2
2.6.37.2_plus_v1: BFS, CFS,THP,compaction, zcache or TOI
Hardcore Linux user since 2004 :D
Back to top
View user's profile Send private message
energyman76b
Advocate
Advocate


Joined: 26 Mar 2003
Posts: 2031
Location: Germany

PostPosted: Sun Aug 23, 2009 8:03 pm    Post subject: Reply with quote

kernelOfTruth wrote:
energyman76b wrote:
disable smt and ahci probably ;)

I don't have forced preemption nor rcu preemption and I don't suffer from long lags anymore. For a long time.


I got to test that with smt, thanks !

you don't need ahci anyways with your SCSI disks ;)


my superfast scsi disks are in my other box ;)
_________________
AidanJT wrote:

Libertardian denial of reality is wholly unimpressive and unconvincing, and simply serves to demonstrate what a bunch of delusional fools they all are.

Satan's got perfectly toned abs and rocks a c-cup.
Back to top
View user's profile Send private message
luispa
Guru
Guru


Joined: 17 Mar 2006
Posts: 343
Location: España

PostPosted: Wed Aug 26, 2009 7:20 pm    Post subject: Reply with quote

wrc1944 wrote:
For those having this problem, especially if they have SATA drives, it would probably be worth a shot to try the deadline scheduler instead of cfq.

Everything I've read over the last year or so seems to indicate there is still an I/O problem with cfq on some systems, and also that generally with SATA drives deadline is often a better scheduler that cfq. Kernel >=2.6.30-rc4 seemed to improve it somewhat (as mentioned), but I'm still sticking with deadline myself until I'm convinced this is really fixed with cfq.

You need to enable support in your kernel (probably already has it, but check your .config file). If not, you'll need to recompile your kernel and enable deadline, but if it does already have it, just append your grub kernel line with
Code:
 elevator=deadline

and reboot.
If it makes a difference great, but if not, just remove the append.


wrc1944, thanks for the update. I spend some time with .30-rc4 but still suffering the issue I mentioned...

luispa wrote:
@fangorn
Thanks for the information, as I said here is the result with 2.6.29: no problem, back to normal behaviour. I'm not suffering problems with I/O now. Obviously I cant add any value here, but my experience. 2.6.28: Ok, 2.6.30: I/O issue, 2.6.29: Ok.
Luis


I'm going to try with your suggestion (deadline) and report back.

Luis
Back to top
View user's profile Send private message
luispa
Guru
Guru


Joined: 17 Mar 2006
Posts: 343
Location: España

PostPosted: Thu Aug 27, 2009 10:44 am    Post subject: Reply with quote

luispa wrote:


I'm going to try with your suggestion (deadline) and report back.

Luis


My report:

I've noticed now with .30 & deadline the system is more responsive but still seeing that disk writtes (syncing) are somehow delayed meaning sporadic reponsiveness, so in general I feel that the system is more reliable with .29.

With interbench didn't notice the difference. I've run it under runlevel 3 though, maybe I should run it under runlevel 1 as recommended.

Here is the output:
Code:

bolica interbench-0.30 # cat 2.6.29-gentoo-r5.log

Using 805133 loops per ms, running every load for 30 seconds
Benchmarking kernel 2.6.29-gentoo-r5 at datestamp 200908270935

--- Benchmarking simulated cpu of Audio in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.013 +/- 0.088       1.93       100           100
Video     0.006 +/- 0.00695    0.015       100           100
X      0.01 +/- 0.061       1.47       100           100
Burn     0.022 +/- 0.261       5.79       100           100
Write      0.01 +/- 0.0375     0.733       100           100
Read     0.012 +/- 0.0677      1.64       100           100
Compile     0.111 +/- 0.983       16.6       100           100
Memload     0.021 +/- 0.088       1.52       100           100

--- Benchmarking simulated cpu of Video in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.007 +/- 0.0205     0.809       100           100
X     0.015 +/- 0.138       3.45       100           100
Burn     0.016 +/- 0.152       4.17       100           100
Write     0.007 +/- 0.0153     0.481       100           100
Read     0.008 +/- 0.0144     0.406       100           100
Compile     0.023 +/- 0.357       10.9       100           100
Memload     0.017 +/- 0.027      0.668       100           100

--- Benchmarking simulated cpu of X in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None         0 +/- 0.00111    0.013       100           100
Video         0 +/- 0.000812    0.01       100           100
Burn     0.533 +/- 2.57          18      86.2          84.4
Write         0 +/- 0.00153    0.017       100           100
Read         0 +/- 0.0014     0.018       100           100
Compile     0.446 +/- 2.16          14      92.6          89.7
Memload         0 +/- 0.00165    0.018       100           100

--- Benchmarking simulated cpu of Gaming in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU
None         0 +/- 0              0       100
Video         0 +/- 0              0       100
X         0 +/- 0              0       100
Burn      1.35 +/- 4.43        19.7      98.7
Write         0 +/- 0              0       100
Read         0 +/- 0              0       100
Compile      4.89 +/- 8.49        34.1      95.3
Memload         0 +/- 0              0       100


Code:
bolica interbench-0.30 # cat 2.6.30-gentoo-r4.log

Using 805133 loops per ms, running every load for 30 seconds
Benchmarking kernel 2.6.30-gentoo-r4 at datestamp 200908271023

--- Benchmarking simulated cpu of Audio in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.007 +/- 0.00727    0.015       100           100
Video     0.005 +/- 0.00515    0.011       100           100
X     0.006 +/- 0.00629    0.012       100           100
Burn     0.015 +/- 0.107       1.93       100           100
Write     0.008 +/- 0.00859    0.017       100           100
Read     0.008 +/- 0.00939    0.098       100           100
Compile     0.005 +/- 0.0054     0.014       100           100
Memload     0.021 +/- 0.122       2.45       100           100

--- Benchmarking simulated cpu of Video in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.005 +/- 0.00553    0.015       100           100
X     0.005 +/- 0.00572    0.019       100           100
Burn     0.021 +/- 0.227       5.66       100           100
Write     0.006 +/- 0.00943     0.24       100           100
Read     0.006 +/- 0.00627    0.017       100           100
Compile     0.277 +/- 2.62        35.3      99.8            99
Memload     0.016 +/- 0.0319     0.719       100           100

--- Benchmarking simulated cpu of X in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None         0 +/- 0.000983   0.012       100           100
Video         0 +/- 0.000592   0.008       100           100
Burn     0.903 +/- 3.25          14        89          84.6
Write         0 +/- 0.00117    0.012       100           100
Read         0 +/- 0.00119    0.017       100           100
Compile      1.04 +/- 4             28      85.6            81
Memload         0 +/- 0.00156    0.021       100           100

--- Benchmarking simulated cpu of Gaming in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU
None         0 +/- 0              0       100
Video         0 +/- 0              0       100
X         0 +/- 0              0       100
Burn      1.38 +/- 4.24        14.3      98.6
Write         0 +/- 0              0       100
Read         0 +/- 0              0       100
Compile      3.35 +/- 7.49        34.4      96.8
Memload         0 +/- 0              0       100



Thanks,
Luis
Back to top
View user's profile Send private message
sidamos
Apprentice
Apprentice


Joined: 16 Dec 2007
Posts: 164

PostPosted: Fri Sep 04, 2009 9:10 am    Post subject: Reply with quote

I have the same problem with AMD64x2 5400+ (running Gentoo 32 bit) and PATA disks. Also, they started with 2.6.30, AFAIK. I have low latency desktop on. I have tried cfq and anticipatory scheduler. However, IMHO this should not be relevant if the heavy IO is on another disk (in my case), see http://forums.gentoo.org/viewtopic-p-5958024.html
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Sat Sep 05, 2009 7:34 pm    Post subject: Reply with quote

something brewing in the 2.6.31 release, which may be good news. anyone running 2.6.31 RCs here?

http://www.techworld.com.au/article/317416/kernel_2_6_31_speed_up_linux_desktop
Back to top
View user's profile Send private message
sidamos
Apprentice
Apprentice


Joined: 16 Dec 2007
Posts: 164

PostPosted: Sat Sep 05, 2009 8:32 pm    Post subject: Reply with quote

devsk wrote:
something brewing in the 2.6.31 release, which may be good news. anyone running 2.6.31 RCs here?

http://www.techworld.com.au/article/317416/kernel_2_6_31_speed_up_linux_desktop


This link does not work.
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Sun Sep 06, 2009 3:03 am    Post subject: Reply with quote

sidamos wrote:
devsk wrote:
something brewing in the 2.6.31 release, which may be good news. anyone running 2.6.31 RCs here?

http://www.techworld.com.au/article/317416/kernel_2_6_31_speed_up_linux_desktop


This link does not work.
I just clicked on in your post and it did work! anyway, here is the relevant part:
Quote:


The kernel developers have been working on improvements to desktop interactivity, particularly when it's under memory pressure since the last release, version 2.6.30, in June.

Desktop applications can experience long and noticeable pauses when the application's code path jumps to a part of the code that is not cached in memory and needs to be read from the disk, which is slower.

However, recent kernel memory management scalability work can result in a desktop environment with poor interactivity as applications become unresponsive too easily.

In version 2.6.31, some heuristics have been used to make it much harder to move the “mapped executable pages” out of the list of active pages, according to Kernelnewbies.org.

“The result is an improved desktop experience; benchmarks on memory tight desktops show clock time and major faults reduced by 50 per cent, and pswpin numbers (memory reads from disk) are reduced to about one-third. That means X desktop responsiveness is doubled under high memory pressure.”

Furthermore, memory flushing benchmarks in a file server shows the number of major faults going from 50 to 3 during 10 per cent cache hot reads.
Back to top
View user's profile Send private message
keenblade
Veteran
Veteran


Joined: 03 Oct 2004
Posts: 1085

PostPosted: Sun Sep 06, 2009 7:42 am    Post subject: Reply with quote

devsk wrote:
something brewing in the 2.6.31 release, which may be good news. anyone running 2.6.31 RCs here?

I have 2.6.31-rc5 here. Here the interbench resulsts:
Code:

Using 979314 loops per ms, running every load for 30 seconds
Benchmarking kernel 2.6.31-rc5 at datestamp 200909060950

--- Benchmarking simulated cpu of Audio in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.085 +/- 0.379       6.64       100           100
Video     0.122 +/- 0.652       11.8       100           100
X     0.208 +/- 1.29        16.3       100           100
Burn     0.035 +/- 0.211       4.15       100           100
Write     0.255 +/- 1.36        21.5       100           100
Read     0.117 +/- 0.875       13.6       100           100
Compile      0.04 +/- 0.383       8.73       100           100
Memload     0.024 +/- 0.101       2.29       100           100

--- Benchmarking simulated cpu of Video in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None     0.101 +/- 1.2         31.9      99.9          99.7
X     0.275 +/- 2.32        35.8      99.4          99.2
Burn     0.072 +/- 1.97        81.8      99.9          99.9
Write     0.242 +/- 2.6         67.3      99.7          99.4
Read     0.093 +/- 0.926       32.4       100          99.9
Compile     0.076 +/- 0.881       27.1       100          99.9
Memload     0.074 +/- 0.486       16.7       100          99.9

--- Benchmarking simulated cpu of X in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU  % Deadlines Met
None         0 +/- 0.00419    0.063       100           100
Video         0 +/- 0.00283    0.037       100           100
Burn      6.15 +/- 36.5         434      52.5          51.1
Write         0 +/- 0.00193    0.023       100           100
Read         0 +/- 0.00198    0.025       100           100
Compile      9.32 +/- 39.4         408      33.5          29.8
Memload         0 +/- 0.00567    0.068       100           100

--- Benchmarking simulated cpu of Gaming in the presence of simulated ---
Load   Latency +/- SD (ms)  Max Latency   % Desired CPU
None     0.016 +/- 0.395       9.57       100
Video     0.428 +/- 5.17         103      99.6
X     0.355 +/- 5.11         103      99.6
Burn      59.1 +/- 140          952      62.9
Write     0.132 +/- 1.33        22.9      99.9
Read     0.097 +/- 1.77        38.2      99.9
Compile      49.3 +/- 89.1         911        67
Memload         0 +/- 0              0       100

Note : I don't know how to interpret these results, but it seemed slow to me. Kernel is from linus git. No patch used and I have ccache, if it matters.
_________________
Anyway it's all the same at the end...
Need help to get it working: "x-fi surround 5.1"
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Sun Sep 06, 2009 4:18 pm    Post subject: Reply with quote

I think its the relative interpretation which may be useful. If you have 2.6.30 (or any other older) kernel, then boot into it and re-run interbench and let's see what numbers it throws up.
Back to top
View user's profile Send private message
Elv13
Guru
Guru


Joined: 13 Nov 2005
Posts: 382
Location: Socialist land of North America

PostPosted: Sun Sep 06, 2009 7:15 pm    Post subject: Reply with quote

In slashdot today:
http://linux.slashdot.org/story/09/09/05/161230/Kernel-2631-To-Speed-Up-Linux-Desktop?art_pos=20
Back to top
View user's profile Send private message
d-fens
Tux's lil' helper
Tux's lil' helper


Joined: 09 Jan 2004
Posts: 88

PostPosted: Tue Sep 08, 2009 1:03 pm    Post subject: Reply with quote

sounds promising, a pity that i can't try it this week - but one question remains for me:

does the optimized (reduced) disk access just work around the root problem, the slow disk access starting from 2.18?
Back to top
View user's profile Send private message
BitJam
Advocate
Advocate


Joined: 12 Aug 2003
Posts: 2454
Location: Silver City, NM

PostPosted: Thu Sep 10, 2009 1:30 am    Post subject: Reply with quote

I've been running amd64 for a couple of years and here is what I've noticed. As long as I had only one hard drive on the system, I never had any problem. Then as soon as I got a 2nd hard drive, transfer of large files from one drive to the other would be dog slow.

I've been recently playing with LiveUSBs and I've noticed something very strange. Transfer of large files to the usb drive is fairly fast if the usb is formated with fat32 but when I format the usb with ext2, the transfers become dog slow, such as 0.5 megabytes per second.

I've seen this on two out of two usb drives I've tested.
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Thu Sep 10, 2009 4:40 am    Post subject: Reply with quote

2.6.31 goes stable. Will try it later tomorrow. Anybody else up for some benchmarking and comparisons with respect to this bug?
Back to top
View user's profile Send private message
mamunata
Apprentice
Apprentice


Joined: 30 Nov 2004
Posts: 169

PostPosted: Thu Sep 10, 2009 4:02 pm    Post subject: Reply with quote

devsk wrote:
2.6.31 goes stable. Will try it later tomorrow. Anybody else up for some benchmarking and comparisons with respect to this bug?

I've compiled 2.6.31 but almost no difference:
-hdparm shows even slower speed - about 20MB/s on my laptop (wit 2.6.30 was 25MB/s)
-on heavy disk load desktop environment responds slowly and system load is 2-3
-swap is used rarely
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Thu Sep 10, 2009 6:55 pm    Post subject: Reply with quote

mamunata wrote:
devsk wrote:
2.6.31 goes stable. Will try it later tomorrow. Anybody else up for some benchmarking and comparisons with respect to this bug?

I've compiled 2.6.31 but almost no difference:
-hdparm shows even slower speed - about 20MB/s on my laptop (wit 2.6.30 was 25MB/s)
-on heavy disk load desktop environment responds slowly and system load is 2-3
-swap is used rarely
Thanks for replying because I was gonna do it today. I think I am gonna just pass the 2.6.31 release then.
Back to top
View user's profile Send private message
Elv13
Guru
Guru


Joined: 13 Nov 2005
Posts: 382
Location: Socialist land of North America

PostPosted: Sat Sep 12, 2009 2:42 am    Post subject: Reply with quote

So far so good for me! Kernel 2.6.30 was a huge improvement over 2.6.2[0-9] and 2.6.31 seem to ba as stable and fast. And I finally removed debugging support from my kernel, I hope to never have it enable again :p. I think this solve the issue in my case, 64bit Linux is now exploiting the whole potential of my computer!

EDIT: Ignore my comment, skip this kernel, it suck, I am reverting to 2.6.30 right now.
Back to top
View user's profile Send private message
devsk
Advocate
Advocate


Joined: 24 Oct 2003
Posts: 2734
Location: Bay Area, CA

PostPosted: Sat Sep 12, 2009 3:10 am    Post subject: Reply with quote

Elv13 wrote:
So far so good for me! Kernel 2.6.30 was a huge improvement over 2.6.2[0-9] and 2.6.31 seem to ba as stable and fast. And I finally removed debugging support from my kernel, I hope to never have it enable again :p. I think this solve the issue in my case, 64bit Linux is now exploiting the whole potential of my computer!

EDIT: Ignore my comment, skip this kernel, it suck, I am reverting to 2.6.30 right now.
what happened? elaborate please.
Back to top
View user's profile Send private message
Display posts from previous:   
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Gentoo on AMD64 All times are GMT
Goto page Previous  1, 2, 3 ... , 36, 37, 38  Next
Page 37 of 38

 
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