You are not logged in.
Hi everyone,
So, I took my first plunge into Arch linux two days ago. I am pleasantly surprised at the ease and cleanliness of installing and setting up Arch64 without any major hiccups. Thanks to the nice, clean documentation available online, I quickly corrected any small problems I ran into (see wireless). However, I can't help but notice how my usb mouse (MS wireless mouse 5000), when attached to an AC powered Kensington 7-port usb2 hub, hangs intermittently for one or two seconds. I can see the hub's lights flickering when this hang occurs. This problem vanishes when I attach the same mouse to my laptop's own usb port. I never seen this problem when using Fedora 10 or Kubuntu 08.10 and 09.04 (kernels <<2.6.32).
There is a large number of repeating events in kernel.log that reminds me of the time when my intel wireless card caused a series of kernel panics in 2.6.27.
Feb 18 11:54:10 o-toro kernel: Uniform Multi-Platform E-IDE driver
Feb 18 11:54:10 o-toro kernel: Probing IDE interface ide0...
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:54:10 o-toro kernel: ide0 at 0x6100-0x6107,0x610e on irq 19
Feb 18 11:54:10 o-toro kernel: Probing IDE interface ide0...
Feb 18 11:54:10 o-toro kernel: Probing IDE interface ide0...
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:54:10 o-toro kernel: Probing IDE interface ide0...
Feb 18 11:54:10 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
....
Feb 18 11:56:53 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:54 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:54 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:55 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:55 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:55 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:57 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:57 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:57 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:58 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:58 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:56:59 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
Feb 18 11:57:00 o-toro kernel: usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
here's a partial output of dmesg
usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
usb 2-1.1: reset full speed USB device using ehci_hcd and address 4
usb usb3: root hub lost power or was reset
usb usb4: root hub lost power or was reset
usb usb5: root hub lost power or was reset
usb usb6: root hub lost power or was reset
usb usb7: root hub lost power or was reset
mkinitcpio.conf (this problem still occured with or without the usb hook added in)
HOOKS="base udev autodetect pata scsi sata filesystems usb usbinput"
rc.conf:
DAEMONS=(@syslog-ng !network !networkmanager netfs @crond alsa hal wicd kdm)
Has anyone encountered this problem before? Any suggestions?
Last edited by ssri (2010-06-26 17:58:00)
Offline
I'm beginning to believe the problem lies with the recognition of the mouse in question.
I stuck another MS wireless mouse in the hub and it worked perfectly
~$lsusb | grep -i mouse
Bus 002 Device 009: ID 045e:00e1 Microsoft Corp. Wireless Laser Mouse 6000 Reciever
When I placed the mouse I use nowadays back in, the problem returns
Bus 002 Device 010: ID 045e:0745 Microsoft Corp.
Any hope in getting it properly recognized by the kernel?
Offline
well, modprobe -r ehci_hcd does the job, but I'm concerned that I will not have any usb2 available to my other usb devices...
Offline
Any suggestions? Looks like a 2.6.32 regression
http://lkml.org/lkml/2009/12/3/418
Last edited by ssri (2010-02-20 17:59:24)
Offline
Loaded up lucid alpha 2 livecd, running 2.6.32-10, to find out that the bug is not there. I did a lsmod and found out that usbhid is loaded but ehci_hcd is nowhere to be found. hmmm...
Offline
I have a similar problem, as both my keyboard and mouse sometimes stops responding when they are connected via my usb hub. The problem has been far worse earlier, now it only happens a couple of times a day.
It is weird though, they both still receive power (the light is on on both) but there is just no response on keypresses or mouse moves. The problem corrects itself if I reconnect the hub, but that is really not a very good solution.
Edit:
Forgot to say that 'lsusb' doesn't give any indication that I have either plugged in. 'lsmod' doesn't show 'ehci_hcd' either, only as a sub-module (or whatever it is) for 'usbcore'
$ lsmod | grep usb
usbhid 38311 0
hid 75845 1 usbhid
usb_storage 41702 0
lirc_mceusb 12697 0
usbcore 140485 6 usbhid,usb_storage,ohci_hcd,ehci_hcd,lirc_mceusb
lirc_dev 10325 1 lirc_mceusb
scsi_mod 92676 5 usb_storage,sg,sr_mod,sd_mod,libata
Last edited by Sharpeee (2010-03-11 23:14:53)
Offline
the plot thickens...
http://bugs.archlinux.org/task/18866
https://bugzilla.kernel.org/show_bug.cgi?id=15649
the default usb 2.0 scheduler is dumb.
Offline
Nicely found ssri! That means that there actually IS a solution.
Please dev, enable 'CONFIG_USB_EHCI_TT_NEWSCHED' in the default kernel.
Offline
Nicely found ssri! That means that there actually IS a solution.
Please dev, enable 'CONFIG_USB_EHCI_TT_NEWSCHED' in the default kernel.
I just want to report that my issue went away after enabling "CONFIG_USB_EHCI_TT_NEWSCHED" and compiling the latest 2.6.32 kernel from kernel.org. Thanks bjacob for the extremely helpful bugreport. Of course, it would be most helpful that the devs enabled this option from the start. Even though compiling the kernel is pretty easy thanks to arch's wiki and the build scripts found within, I do not think that I'll enjoy spending the time compiling every single update. Call me lazy if you will. Marking this thread as solved.
Last edited by ssri (2010-04-04 09:55:59)
Offline
http://kernel.xc.net/html/linux-2.6.19/ … T_NEWSCHED
'Linux kernel 2.6.19 on i386.
CONFIG_USB_EHCI_TT_NEWSCHED:
This changes the periodic scheduling code to fill more of the low
and full speed bandwidth available from the Transaction Translator
(TT) in USB 2.0 hubs. Without this, only one transfer will be
issued in each microframe, significantly reducing the number of
periodic low/fullspeed transfers possible.
If you have multiple periodic low/fullspeed devices connected to a
highspeed USB hub which is connected to a highspeed USB Host
Controller, and some of those devices will not work correctly
(possibly due to "ENOSPC" or "-28" errors), say Y.
If unsure, say N.'
Offline
Update:
Enabling 'CONFIG_USB_EHCI_TT_NEWSCHED' has fixed this problem everytime since 2.6.32. Just sync abs, modify kernel26's PKGBUILD and uncomment make menuconfig, enabaled the new EHCI scheduler under devices and USB, makepkg -c -i and I'm done.
Last edited by ssri (2010-06-26 17:57:11)
Offline
Hi,
I have the same mouse and mine won't respond unless I do modprobe -r ehci_hcd. I am experienced in several other distros (mainly Ubuntu) but not Arch. I've got as far as the fact that I can recompile the kernel and solve the problem but in the last post I don't understand where to "modify kernel26's PKGBUILD and uncomment make menuconfig".
Any help would be greatly appreciated!
Offline
Hi,
I have the same mouse and mine won't respond unless I do modprobe -r ehci_hcd. I am experienced in several other distros (mainly Ubuntu) but not Arch. I've got as far as the fact that I can recompile the kernel and solve the problem but in the last post I don't understand where to "modify kernel26's PKGBUILD and uncomment make menuconfig".
Any help would be greatly appreciated!
Sync ABS (http://wiki.archlinux.org/index.php/Arch_Build_System).
$ cp /var/abs/core/kernel26/* /whatever/your/directory/you're/building/your/kernel/.
Then you use your favorite editor to open up PKGBUILD and remove the "#" that is in front of "make menuconfig", which is on line 47 (as of today). Then follow the rest of my procedure outlined in my last post.
Also, to save compile time, while you have PKGBUILD opened in your editor you can remove 'kernel26-manpages' 'kernel26-docs' from pkgname and delete the lines "package_kernel26-manpages() {" and "package_kernel26-docs() {" on down (delete lines 223-247). You can install both of those packages from pacman.
Last edited by ssri (2010-08-21 07:59:47)
Offline
This worked quite well, except for a few false starts on my part. Thank you very much!
Offline