[arch-dev-public] [signoff] udev-132-1

Tobias Powalowski

9 Nov 2008 9 Nov '08

9:31 a.m.

Hi update to latest version, please signoff for both arches. greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa@archlinux.org

Attachments:

signature.asc (application/pgp-signature — 197 bytes)

Show replies by date

Thomas Bächler

9 Nov 9 Nov

11:10 a.m.

Tobias Powalowski schrieb:

...

Hi update to latest version, please signoff for both arches.

greetings tpowa

signoff x86_64

Ronald van Haren

11:21 p.m.

On Sun, Nov 9, 2008 at 12:10 PM, Thomas Bächler wrote:

...

Tobias Powalowski schrieb:

...
Hi update to latest version, please signoff for both arches.

greetings tpowa

signoff x86_64

another x86_64 signoff Ronald

Dan McGee

11:23 p.m.

On Sun, Nov 9, 2008 at 5:21 PM, Ronald van Haren wrote:

...

On Sun, Nov 9, 2008 at 12:10 PM, Thomas Bächler wrote:

...
Tobias Powalowski schrieb:

...
Hi update to latest version, please signoff for both arches.

greetings tpowa

signoff x86_64

another x86_64 signoff

Booted fine on i686. However, I thought this was something that should be fixed by udev loading my modules, and it shows up in my dmesg output: Warning! ehci_hcd should always be loaded before uhci_hcd and ohci_hcd, not after dmesg has output from usbcore, ohci_hcd, and ehci_hcd in that order, so it does not appear to be loading in the correct order as the message indicates. -Dan

Tobias Powalowski

10 Nov 10 Nov

6:05 a.m.

...

Booted fine on i686. However, I thought this was something that should be fixed by udev loading my modules, and it shows up in my dmesg output:

Warning! ehci_hcd should always be loaded before uhci_hcd and ohci_hcd, not after

dmesg has output from usbcore, ohci_hcd, and ehci_hcd in that order, so it does not appear to be loading in the correct order as the message indicates.

-Dan Hi, http://bugs.archlinux.org/task/12009 I talked to brain0 yesterday, it's pretty difficult to do this on udev level, because how you want to trigger if a system has a usb2 controller inside.

Workaround for this warning is to load it in MODULES= in initcpio and rc.conf manually. greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa@archlinux.org

Thomas Bächler

12 Nov 12 Nov

8:01 a.m.

Dan McGee schrieb:

...

Booted fine on i686. However, I thought this was something that should be fixed by udev loading my modules, and it shows up in my dmesg output:

Warning! ehci_hcd should always be loaded before uhci_hcd and ohci_hcd, not after

dmesg has output from usbcore, ohci_hcd, and ehci_hcd in that order, so it does not appear to be loading in the correct order as the message indicates.

Udev has no ordering in loading modules, it is all random! Thus, the kernel devs wanting to ensure a particular order on loading modules completely contradicts the general usage of udev. IMO, this is a kernel bug, the order mustn't matter.

Aaron Griffin

3:39 p.m.

On Wed, Nov 12, 2008 at 2:01 AM, Thomas Bächler wrote:

...

Dan McGee schrieb:

...
Booted fine on i686. However, I thought this was something that should be fixed by udev loading my modules, and it shows up in my dmesg output:

Warning! ehci_hcd should always be loaded before uhci_hcd and ohci_hcd, not after

dmesg has output from usbcore, ohci_hcd, and ehci_hcd in that order, so it does not appear to be loading in the correct order as the message indicates.

Udev has no ordering in loading modules, it is all random! Thus, the kernel devs wanting to ensure a particular order on loading modules completely contradicts the general usage of udev. IMO, this is a kernel bug, the order mustn't matter.

You can handle this with modprobe. It's an age-old trick to load another module before the requested module is loaded. Perhaps we should ship a file that has that in there for uhci?

Eric Bélanger

8 Dec 8 Dec

8:12 p.m.

On Wed, 12 Nov 2008, Aaron Griffin wrote:

...

On Wed, Nov 12, 2008 at 2:01 AM, Thomas Bächler wrote:

...
Dan McGee schrieb:

...
Booted fine on i686. However, I thought this was something that should be fixed by udev loading my modules, and it shows up in my dmesg output:

Warning! ehci_hcd should always be loaded before uhci_hcd and ohci_hcd, not after

dmesg has output from usbcore, ohci_hcd, and ehci_hcd in that order, so it does not appear to be loading in the correct order as the message indicates.

Udev has no ordering in loading modules, it is all random! Thus, the kernel devs wanting to ensure a particular order on loading modules completely contradicts the general usage of udev. IMO, this is a kernel bug, the order mustn't matter.

You can handle this with modprobe. It's an age-old trick to load another module before the requested module is loaded. Perhaps we should ship a file that has that in there for uhci?

This has been in testing for a month. I don't know if there are pending bug reports but I haven't noticed any problems. So signoff for both arches. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

Thayer Williams

9:21 p.m.

On Mon, Dec 8, 2008 at 12:12 PM, Eric Bélanger wrote:

...

This has been in testing for a month. I don't know if there are pending bug reports but I haven't noticed any problems. So signoff for both arches.

Agreed, I second the signoff for i686.

Tobias Powalowski

9:32 p.m.

Am Montag 08 Dezember 2008 schrieb Thayer Williams:

...

On Mon, Dec 8, 2008 at 12:12 PM, Eric Bélanger

wrote:

...
This has been in testing for a month. I don't know if there are pending bug reports but I haven't noticed any problems. So signoff for both arches.

Agreed, I second the signoff for i686. I think there was an issue, with symlinks, also a new version is out since quite some time i would like to skip this package movement and just bump it to latest version again. and resignoff.

greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa@archlinux.org

Thayer Williams

9:39 p.m.

On Mon, Dec 8, 2008 at 1:32 PM, Tobias Powalowski wrote:

...

Am Montag 08 Dezember 2008 schrieb Thayer Williams:

...
On Mon, Dec 8, 2008 at 12:12 PM, Eric Bélanger

wrote:

...
This has been in testing for a month. I don't know if there are pending bug reports but I haven't noticed any problems. So signoff for both arches.

Agreed, I second the signoff for i686. I think there was an issue, with symlinks, also a new version is out since quite some time i would like to skip this package movement and just bump it to latest version again. and resignoff.

That's cool, bump it and I'll have a go. Not sure what it was about the symlinks, but I do make use of them in my automount rules so I'll check that out too.

Aaron Griffin

9:46 p.m.

On Mon, Dec 8, 2008 at 3:39 PM, Thayer Williams wrote:

...

On Mon, Dec 8, 2008 at 1:32 PM, Tobias Powalowski wrote:

...
Am Montag 08 Dezember 2008 schrieb Thayer Williams:

...
On Mon, Dec 8, 2008 at 12:12 PM, Eric Bélanger

wrote:

...
This has been in testing for a month. I don't know if there are pending bug reports but I haven't noticed any problems. So signoff for both arches.

Agreed, I second the signoff for i686. I think there was an issue, with symlinks, also a new version is out since quite some time i would like to skip this package movement and just bump it to latest version again. and resignoff.

That's cool, bump it and I'll have a go. Not sure what it was about the symlinks, but I do make use of them in my automount rules so I'll check that out too.

It'd cd and dvd related. Something about udev trying to symlink say /dev/dvd0 to /dev/cd/dvd-4:0:0:0 but only /dev/cd/dvd- exists (missing the bus info). I actually noticed it myself, but it wasn't a huge deal as the device was still there, symlinks were just messed

5633

Age (days ago)

5662

Last active (days ago)

List overview

Download

11 comments

7 participants

participants (7)

Aaron Griffin
Dan McGee
Eric Bélanger
Ronald van Haren
Thayer Williams
Thomas Bächler
Tobias Powalowski