Harddisk to slow on HP 6715s

[PC-Ente]

1 Installed Ubuntu Feisty on my Notebook HP 6715s.

The 32bit-Version works to slow with the Hard Disk. At least only 1 Mbit write/read speed. This Bug is allsow known in German Forum with oder Users
http://forum.ubuntuusers.de/topic/104391/?highlight=6715s

The Only possibility to was to USE 64bit Ubuntu Feisty. The Harddisk works with normal Speed Here

lspci:

1 jan@laptop:~$ sudo lspci
2 Password:
3 00:00.0 Host bridge: ATI Technologies Inc Unknown device 7910
4 00:01.0 PCI bridge: ATI Technologies Inc Unknown device 7912
5 00:04.0 PCI bridge: ATI Technologies Inc Unknown device 7914
6 00:05.0 PCI bridge: ATI Technologies Inc Unknown device 7915
7 00:06.0 PCI bridge: ATI Technologies Inc Unknown device 7916
8 00:12.0 SATA controller: ATI Technologies Inc SB600 Non-Raid-5 SATA
9 00:13.0 USB Controller: ATI Technologies Inc SB600 USB (OHCI0)
10 00:13.1 USB Controller: ATI Technologies Inc SB600 USB (OHCI1)
11 00:13.2 USB Controller: ATI Technologies Inc SB600 USB (OHCI2)
12 00:13.3 USB Controller: ATI Technologies Inc SB600 USB (OHCI3)
13 00:13.4 USB Controller: ATI Technologies Inc SB600 USB (OHCI4)
14 00:13.5 USB Controller: ATI Technologies Inc SB600 USB Controller (EHCI)
15 00:14.0 SMBus: ATI Technologies Inc SB600 SMBus (rev 14)
16 00:14.1 IDE interface: ATI Technologies Inc SB600 IDE
17 00:14.2 Audio device: ATI Technologies Inc SB600 Azalia
18 00:14.3 ISA bridge: ATI Technologies Inc SB600 PCI to LPC Bridge
19 00:14.4 PCI bridge: ATI Technologies Inc SB600 PCI to PCI Bridge
20 00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] HyperTransport Technology Configuration
21 00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Address Map
22 00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] DRAM Controller
23 00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Miscellaneous Control
24 01:05.0 VGA compatible controller: ATI Technologies Inc Unknown device 791f
25 02:04.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev b6)
26 10:00.0 Ethernet controller: Broadcom Corporation Unknown device 1713 (rev 02)
27 30:00.0 Network controller: Broadcom Corporation Dell Wireless 1390 WLAN Mini-PCI Card (rev 02)
28 jan@laptop:~$
29

Detailed scan from windows (not my scan an other user with the same notebook made it)
http://rafb.net/p/3GHWPB94.html

[Mindphaser (mindo83)]

I own the same Laptop (difference: CPU AMD Semperon 3600+ 32bit only) and can confirm this problem !

hdparm -tT /dev/sda says this (no c/p because i am on my other PC now) :

cached readings: 1206 MB in 2 sec. = 602 MB/s
buffered readings: 8 MB in 1,4 sec. = 1,8 MB/s (<--- bad :( )

I can add that its not only the HardDisk that is slow, the CD/DVD-Drive and USB is horribly slow too.
If I connect a USB-Mouse and the HDD (or CD/DVD) is working, the mouse-cursor responds very slow while it moves smooth using the touchpad.

[Mindphaser (mindo83)]

sorry, but I forgot to mention that this problem is present on gutsy too

[Unkraut (unkraut2)]

Same problem in Hardy (up to now)

[Mindphaser (mindo83)]

Its fixed upstreamt, running 2.6.25-rc7 on my notebook and it works very well (if booting with noapic).
Bad is that they ruined it somewhere between rc7 and now.
I could try to find the patch which solved this problem, but I doubt any dev would care, because it seems it was reverted because it caused problems on other systems :(

[Jean-Baptiste Lallement (jibel)]

Could someone with this hardware test with the latest kernel in Intrepid and report here. Thanks.

[Mindphaser (mindo83)]

Hello, I've tested the latest Intrepid kernel (2.6.26-5) but it didn't helped :( The "noapic" parameter didn't helped, like in a vanilla 2.6.26

I'm currently running 2.6.26 on my Notebook with a patched processor_idle.c, as mentioned above I found my Hardware working fine during the 2.6.25 developement cycle but later the problem re-appeared... actually caused by this change:

http://www.kernel.org/diff/diffview.cgi?file=%2Fpub%2Flinux%2Fkernel%2Fv2.6%2Fsnapshots%2Fold%2Fpatch-2.6.25-rc7-git4.bz2;z=47

So I was happy until 2.6.25.3 came out, but after reverting this:

http://www.kernel.org/diff/diffview.cgi?file=%2Fpub%2Flinux%2Fkernel%2Fv2.6%2Fpatch-2.6.25.3.bz2;z=15

It was working again... dont ask me what these patches actually do because I'm no kernel-dev or hardware-guru...if you know more I am interested to know what exactly is happening there :)

Reverting the changes also works fine for 2.6.26

greets, Matze

[Jean-Baptiste Lallement (jibel)]

Can you try to boot the latest kernel and pass the boot param: processor.max_cstate=1 ?

Please also include the following additional information, as required by the Ubuntu Kernel Team:
1. Please include the output of the command "uname -a" in your next response.
2. Please run the command "dmesg > dmesg.log" after a fresh boot and attach the resulting file "dmesg.log" to this bug report.
3. Please run the command "sudo lspci -vvnn > lspci-vvnn.log" and attach the resulting file "lspci-vvnn.log" to this bug report.

Thanks in advance

[Mindphaser (mindo83)]

The suggested "processor.max_cstate=1" parameter seems to be unknown, but here is the dmesg.log

[Mindphaser (mindo83)]

okay, I figured out what J-B was up to and made this small patch... compiled my custom 2.6.26.2 with it and.... it works like a charm *HAPPY*

[Jean-Baptiste Lallement (jibel)]

Thanks for finding this. I'm marking it as confirmed. Thanks for taking the time to make Ubuntu better!

[Leann Ogasawara (leannogasawara)]

The Ubuntu Kernel Team is planning to move to the 2.6.27 kernel for the upcoming Intrepid Ibex 8.10 release. As a result, the kernel team would appreciate it if you could please test this newer 2.6.27 Ubuntu kernel. There are one of two ways you should be able to test:

1) If you are comfortable installing packages on your own, the linux-image-2.6.27-* package is currently available for you to install and test.

--or--

2) The upcoming Alpha5 for Intrepid Ibex 8.10 will contain this newer 2.6.27 Ubuntu kernel. Alpha5 is set to be released Thursday Sept 4. Please watch http://www.ubuntu.com/testing for Alpha5 to be announced. You should then be able to test via a LiveCD.

Please let us know immediately if this newer 2.6.27 kernel resolves the bug reported here or if the issue remains. More importantly, please open a new bug report for each new bug/regression introduced by the 2.6.27 kernel and tag the bug report with 'linux-2.6.27'. Also, please specifically note if the issue does or does not appear in the 2.6.26 kernel. Thanks again, we really appreicate your help and feedback.

[Mindphaser (mindo83)]

Hello again
Ive tested the 2.6.27-2 and the issue remains. Without the "noapic" kernel parameter the HDD seems to have normal speed, but the fan is constantly running on fastest (and loudest) spped and system hangs up while "Loading Hardware Drivers...".
With the noapic parameter the system boots fine with the fan keeping quiet, but HDD is terribly slow again. I attach a dmesg.log using this configuration in hope it could help you to identify the problem.

I tested 2.6.27-rc5 from kernel.org yesterday with the same results, but after applying the processor_idle.patch, HDD and System runs fine, at least when using "noapic".

2.6.26 + processor_idle.patch is the best working solution for this problem now, there it runs fine with and without the noapic parameter.

greets, Matze

[yakuizhao (yakui-zhao)]

Will you please try the boot option of "processor.max_cstate=2"?
It is noted that the processor driver (in drivers/acpi/processor) should be compiled as built-in kernel modules.
Thanks.

[Mindphaser (mindo83)]

See my answer here: http://bugzilla.kernel.org/show_bug.cgi?id=11416#c3
In short: max_cstate=2 works fine on the first look but re-introduce random system freezes which were avoided by booting with "noapic".
Maybe SOMEONE ELSE THAN ME could find out more or even different results ;-)

greets, Matze

[Unkraut (unkraut2)]

On my computer passing nohz=off to the Kernel solves all issues! This will disable the Tickless-feature.

I filed a related bug. See: http://bugzilla.kernel.org/show_bug.cgi?id=10778

Best regards,
Simon Waid

[yakuizhao (yakui-zhao)]

Thanks for the info.
Does the system work well if adding the boot option of "nolapic_timer"?
Thanks

[Mindphaser (mindo83)]

Greetings,
adding the "nolapic_timer" parameter leads to system freeze/blank screen during boot before getting to init :-(

[Łukasz Góralczyk (lukasz-goralczyk)]

After upgrade from Ubuntu 8.04 to 8.10 I encountered the same problem - slow disk access, it takes long for applications to start. Passing "nohz=off" option to the kernel fixes the problem (I didn't tried other solutions).

More info on my hardware: Benq P52 notebook, amd64 turion x2 tl-52, 2 GB RAM, Seagate 320 GB HD, ATI based mainboard (SB400), ATI X1600 graphics card.

uname -a:
Linux sagittarius 2.6.27-4-generic #1 SMP Mon Sep 22 04:40:15 UTC 2008 x86_64 GNU/Linux

System:
Ubuntu Intrepid 8.10

[Łukasz Góralczyk (lukasz-goralczyk)]

I no longer suffer from this "bug". After playing with menu.lst I forgot to add "nohz=off" but I didn't notice the slow disk problem.

[PC-Ente (pcente)]

So I Installed the 64bit Beta version of Ubuntu 8.10 Intrepid, The Setup crasht the first time with the typical "Black Screen" Frezze, after booting with acpi=off everything worked fine. So after Booting and updateing with acpi=off I started without any Kernel flags to See what happens. The Disk is Slow as it was on Feisty 32bit i Rember.
A Boot with nohz=off works for me, but bug still is alive in 2.6.27

[PC-Ente (pcente)]

So i just Cecked everything on my Laptop and have to tell you, nohz or nohz=off insn't fixing anything at my maschine. HDD is still slow, its only working with acpi=off. I'm going to search for a resolution in other bug reports the next days, but until now I left here my dmseg and hdparm logs with differnt kernel flags. maybe someone will check/need it and it helps ...

[Mindphaser (mindo83)]

Hi Jan, which Laptop Model you have ?
Seems that CPU/Chipset is identical... if youre familiar with compiling your own Kernel, you could try the processor_idle.patch uploaded by me, but if you dont have a Compaq 6715s you need to apply the DMI_* strings in the patch before applying.
For 2.6.27 you also need to take a look at this :

http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.27.y.git;a=commitdiff;h=e84956f92a846246b09b34f2a728329c386d250f

If you have a Laptop of the types mention there you don't need to do anything.
Other you should look at the dmi_system_table struct in arch/x86/kernel/acpi/boot.c and do whats needed that these quirks work for your system.
If I didnt change the quirk for "Compaq 6715b" there to "Compaq 6715s" my cooling-fan is running on highest speed all the time, and performance slows down constantly

then, everything should maybe work fine on your system in 32bit mode.
Just a little bit of work ^^-.-

greets, Matze

[PC-Ente (pcente)]

I own the Model HP6715s with AMD Sempron 3400+, I'll ceck what u wrote and post here again. thx

[knuutsen (peter-vornberg)]

Hey all,

i've the same problem that Jan explained. Also have 6715s with amd turion x2 60. I'm interested in 8.10 64bit but have the same probs as described here. Mindphaser, is this mess fixed till the end of october? Or do i have to make the idle patch to get the messy SB600 work properly with the final ubuntu 8.10?

thx in advance

knuutsen

[Unkraut (unkraut2)]

Hey knuutsen,

the current version of 8.10 works with Kernel 2.6.25 (however I don't know if support won't be dropped). This kernel supports all Hardware-Components of the 6715s, including WLAN. I uploaded two ready-made .deb Packages, with work well on my machine (6715s).

1. Install Intrepid
2. Get .deb them from:
http://www.gigasize.com/get.php?d=nnpwrfzycfc
http://www.gigasize.com/get.php?d=dyy3xv4j3lb

(If you like, source from: ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.2.tar.bz2)

3. Install & Test 2.6.25.2
4. If it works, remove all 2.6.27 kernel packages, including "linux-image, linux-headers-generic, linux-generic" to prevent the system from installing kernel-updates.

Best regards,
Simon

[Mindphaser (mindo83)]

AFAIK systems which are capable of running in 64bit (and actually running in 64bit mode) make no problems, but for 32bit you need the idle patch AND (for 2.6.27 kernel) do the tweak in arch/x86/kernel/acpi/boot.c I've mentioned abve.

But then, everything works just fine, except ATI graphics but thats another painfull story ....

greets

[knuutsen (peter-vornberg)]

Hey Simon,
thanks for this value tip.
i will try it after installing the final version of 8.10. . This makes
me feel good in hoping that the wlan prob is solved as well.

thx much knuutsen

SimonW schrieb:
> Hey knuutsen,
>
> the current version of 8.10 works with Kernel 2.6.25 (however I don't
> know if support won't be dropped). This kernel supports all Hardware-
> Components of the 6715s, including WLAN. I uploaded two ready-made .deb
> Packages, with work well on my machine (6715s).
>
> 1. Install Intrepid
> 2. Get .deb them from:
> http://www.gigasize.com/get.php?d=nnpwrfzycfc
> http://www.gigasize.com/get.php?d=dyy3xv4j3lb
>
> (If you like, source from:
> ftp://ftp.kernel.org/pub/linux/kernel/v2.6/linux-2.6.25.2.tar.bz2)
>
> 3. Install & Test 2.6.25.2
> 4. If it works, remove all 2.6.27 kernel packages, including "linux-image, linux-headers-generic, linux-generic" to prevent the system from installing kernel-updates.
>
> Best regards,
> Simon
>

[knuutsen (peter-vornberg)]

Hey Mindphaser,
i will go on 64 bit cause its much faster on 6715s as 32 bit.
Thx for replying
So i will try Simons debs...
best regards knuutsen

Mindphaser schrieb:
> AFAIK systems which are capable of running in 64bit (and actually
> running in 64bit mode) make no problems, but for 32bit you need the idle
> patch AND (for 2.6.27 kernel) do the tweak in
> arch/x86/kernel/acpi/boot.c I've mentioned abve.
>
> But then, everything works just fine, except ATI graphics but thats
> another painfull story ....
>
> greets
>

[knuutsen (peter-vornberg)]

Hey Simon,
i did what you suggested..but i wasn't successfull. Nevertheless i will now play hardy back on live at my laptop, after that, i will install the debs again you suggested and after that i will update online to ibex.... will give my results here in 4 - 5 hours i guess.
Any suggestions would be appreciated.... knuutsen

[Unkraut (unkraut2)]

Hey knuutsen!

What is the exact problem?

I tell you what I did: I compiled the kernel with 8.04 and updated to 8.10 afterwards. However it could be, that the packets I have installed differ from what is the standard.

good luck & best regards,

Simon
PS: I forgot: The latest version of fglrx (8.10 i think) did not work with the kernel I posted. You can move over to radeonhd; except for performance it has only advantages.

[knuutsen (peter-vornberg)]

Hey Simon,
so far your packages did'nt work well for me. I couldn't start with your
linux-image and - header intrepid ibex. Maybe i have different hardware
in my 6715s. You can see my configuration at my blog:
knuutsen.wordpress.com .

Anyway, i reinstalled Hardy with the newest kernel in there and i was
suprised, that wlan works well now; thanks to spa driver and in
combination with the newest wicd. So far i decided to wait with the
upgrade to ibex. The disadvantages are too much for my special 6715s. So
i like to take the newest fglrx in hardy and in ibex it is not possible.
On the other hand ibex is interesting for dvbt i read.

Besides i will wait till i can boot intrepid ibex without any flags.
Hopefully it comes soon.

Thx for your advise....cya knuutsen

SimonW schrieb:
> Hey knuutsen!
>
> What is the exact problem?
>
> I tell you what I did: I compiled the kernel with 8.04 and updated to
> 8.10 afterwards. However it could be, that the packets I have installed
> differ from what is the standard.
>
> good luck & best regards,
>
> Simon
> PS: I forgot: The latest version of fglrx (8.10 i think) did not work with the kernel I posted. You can move over to radeonhd; except for performance it has only advantages.
>

[knuutsen (peter-vornberg)]

Hey Simon,

early that morning i had the last freeze with hardy + 2.6.24.21 (released 20.10.). After that, the decision was clear to update with "sudo update-manager -d" to intrepid. During updating it freezes again - crap - . At 7 am taking the normal 64 bit intrepid live cd and i installed the intrepid again. To make sure that this works before pressing the installation button i was pressing F6 and added the follows; nohz=off and noapic. I did that only to be sure having a quick install. It went well after the first boot i still got the fan blowing me away. I set the flag "noapic" in the menu.lst. This fixed the prob. Now it works very well and i'm hopefully that my 6715s HP will not freeze again under intrepid. I also installed fglrx from envy; type in terminal " envyng -t" and it works fine yet.

But now i also silence with the fan i never had before...hmm.. lmsensors ... says that the temperature is ok ... and acpi from ubuntu is working... it sounds all to be good now my question at you:

What does "noapic" exactly do? Is there some danger to my laptop to run this flag for ever? :-)
greets knuutsen

[Launchpad Janitor (janitor)]

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.

[Oibaf (oibaf)]

Apparently, this is now fixed in jaunty:

$ sudo hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 902 MB in 2.00 seconds = 450.42 MB/sec
 Timing buffered disk reads: 126 MB in 3.00 seconds = 41.98 MB/sec