udev-059 and default.hotplug script

udev-059 and default.hotplug script

[Phil K]

Gentoo here, udev-059 causes the
/etc/hotplug.d/default/default.hotplug script to
complain bitterly at boot on all references to the
/dev/null device.  I am using the hotplug 2004-09-23
script set.

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Greg KH]

On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> Gentoo here, udev-059 causes the
> /etc/hotplug.d/default/default.hotplug script to
> complain bitterly at boot on all references to the
> /dev/null device.  I am using the hotplug 2004-09-23
> script set.

This is a gentoo specific issue, stick with bugs.gentoo.org for now.

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 465 bytes --]

On Jul 02, Greg KH <greg@kroah.com> wrote:

> On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > Gentoo here, udev-059 causes the
> > /etc/hotplug.d/default/default.hotplug script to
> > complain bitterly at boot on all references to the
> > /dev/null device.  I am using the hotplug 2004-09-23
> > script set.
> This is a gentoo specific issue, stick with bugs.gentoo.org for now.
Actually it happens on Debian systems too.

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Greg KH]

On Fri, Jul 01, 2005 at 07:46:48PM -0700, Greg KH wrote:
> On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > Gentoo here, udev-059 causes the
> > /etc/hotplug.d/default/default.hotplug script to
> > complain bitterly at boot on all references to the
> > /dev/null device.  I am using the hotplug 2004-09-23
> > script set.
> 
> This is a gentoo specific issue, stick with bugs.gentoo.org for now.

Hm, no, this might be a udev specific issue.  I'm also now seeing this,
and in fact, can't get 2.6.12 to work at all with the latest udev (059).
Will debug more tonight...

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Greg KH]

On Sat, Jul 02, 2005 at 10:22:15AM +0200, Marco d'Itri wrote:
> On Jul 02, Greg KH <greg@kroah.com> wrote:
> 
> > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > Gentoo here, udev-059 causes the
> > > /etc/hotplug.d/default/default.hotplug script to
> > > complain bitterly at boot on all references to the
> > > /dev/null device.  I am using the hotplug 2004-09-23
> > > script set.
> > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> Actually it happens on Debian systems too.

Yeah, I don't know how my testing didn't uncover this :(

Does 2.6.12 work for you with the 059 udev?

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

On Jul 02, Greg KH <greg@kroah.com> wrote:

> Does 2.6.12 work for you with the 059 udev?
Yes. Should I upload 059 to the Debian archive or wait for a patch?

Actually I had big problems with the cxacru driver too, but since it's
still an out of tree driver I suppose that it needs to be updated in
some way. The symptoms are that if I let hotplug autoload it then the
system will hang. I do not understand exactly how this happens because
I'm not able to run ps even if I have another shell open at the time,
but sysrq-t shows that there are many udev_run_hotplugd processes
running.
If I blacklist and then manually load it with modprobe then everything
works.

Another issue is that some sysfs keys are apparently missing:

  looking at class device '/sys/devices/pci0000:00/0000:00:10.1/usb2/2-2':
    SUBSYSTEM="unknown"
    SYSFS{bConfigurationValue}="1"
    SYSFS{bDeviceClass}="ff"
    SYSFS{bDeviceProtocol}="00"
    SYSFS{bDeviceSubClass}="00"
    SYSFS{bMaxPower}="500mA"
    SYSFS{bNumConfigurations}="1"
    SYSFS{bNumInterfaces}=" 1"
    SYSFS{bcdDevice}="0001"
    SYSFS{bmAttributes}="80"
    SYSFS{configuration}=""
    SYSFS{devnum}="5"
    SYSFS{idProduct}="0020"
    SYSFS{idVendor}="0659"
    SYSFS{manufacturer}="-"
    SYSFS{maxchild}="0"
    SYSFS{product}="ADSL USB MODEM"
    SYSFS{serial}="83154158"
    SYSFS{speed}="12"
    SYSFS{version}=" 1.10"

-- 
ciao,
Marco


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Stefan Schweizer]

[-- Attachment #1: Type: text/plain, Size: 1188 bytes --]

Am Samstag 02 Juli 2005 17:58 schrieb Greg KH:
> On Sat, Jul 02, 2005 at 10:22:15AM +0200, Marco d'Itri wrote:
> > On Jul 02, Greg KH <greg@kroah.com> wrote:
> > > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > > Gentoo here, udev-059 causes the
> > > > /etc/hotplug.d/default/default.hotplug script to
> > > > complain bitterly at boot on all references to the
> > > > /dev/null device.  I am using the hotplug 2004-09-23
> > > > script set.
> > >
> > > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> >
> > Actually it happens on Debian systems too.
>
> Yeah, I don't know how my testing didn't uncover this :(
>
> Does 2.6.12 work for you with the 059 udev?

A quick survey in #gentoo-bugs showed that it is really slow (>1minute) on 
startup for kernels:
	2.6.12-ck3
	2.6.12-mm1 (my own)

Another user with 2.6.12-gentoo-r3 reportet that he has no problems at all

Furthermore people seem to have problems when they did not upgrade yet to 
linux-headers-2.6.11.

Can you please add linux-headers-2.6.11 to the udev DEPEND for the gentoo 
Ebuild and maybe package.mask it, when it is so unstable?

Regards,
Stefan

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Greg KH]

On Sat, Jul 02, 2005 at 08:55:42AM -0700, Greg KH wrote:
> On Fri, Jul 01, 2005 at 07:46:48PM -0700, Greg KH wrote:
> > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > Gentoo here, udev-059 causes the
> > > /etc/hotplug.d/default/default.hotplug script to
> > > complain bitterly at boot on all references to the
> > > /dev/null device.  I am using the hotplug 2004-09-23
> > > script set.
> > 
> > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> 
> Hm, no, this might be a udev specific issue.  I'm also now seeing this,
> and in fact, can't get 2.6.12 to work at all with the latest udev (059).
> Will debug more tonight...

Kay, this seems to be caused by udevstart running the
/etc/hotplug.d/defalt/default.hotplug link (points to the "load a
module" hotplug stuff) before we've created anything in /dev.  This is
because of the udev_run_hotplugd program rule.  But for udevstart, we
don't want to run any hotplug stuff :(

Hm, any hints on how to fix this loop?

gotta run, family stuff this afternoon...

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Sat, Jul 02, 2005 at 12:06:46PM -0700, Greg KH wrote:
> On Sat, Jul 02, 2005 at 08:55:42AM -0700, Greg KH wrote:
> > On Fri, Jul 01, 2005 at 07:46:48PM -0700, Greg KH wrote:
> > > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > > Gentoo here, udev-059 causes the
> > > > /etc/hotplug.d/default/default.hotplug script to
> > > > complain bitterly at boot on all references to the
> > > > /dev/null device.  I am using the hotplug 2004-09-23
> > > > script set.
> > > 
> > > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> > 
> > Hm, no, this might be a udev specific issue.  I'm also now seeing this,
> > and in fact, can't get 2.6.12 to work at all with the latest udev (059).
> > Will debug more tonight...
> 
> Kay, this seems to be caused by udevstart running the
> /etc/hotplug.d/defalt/default.hotplug link (points to the "load a
> module" hotplug stuff) before we've created anything in /dev.  This is
> because of the udev_run_hotplugd program rule.  But for udevstart, we
> don't want to run any hotplug stuff :(

Sure, that's wrong.

> Hm, any hints on how to fix this loop?

I hope so:
  ENV{UDEVD_EVENT}="1", RUN+="/sbin/udev_run_hotplugd"

should prevent the hotplug stuff run, if it's not from the kernel.

Cheers,
Kay


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Greg KH]

On Sat, Jul 02, 2005 at 09:37:40PM +0200, Kay Sievers wrote:
> On Sat, Jul 02, 2005 at 12:06:46PM -0700, Greg KH wrote:
> > On Sat, Jul 02, 2005 at 08:55:42AM -0700, Greg KH wrote:
> > > On Fri, Jul 01, 2005 at 07:46:48PM -0700, Greg KH wrote:
> > > > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > > > Gentoo here, udev-059 causes the
> > > > > /etc/hotplug.d/default/default.hotplug script to
> > > > > complain bitterly at boot on all references to the
> > > > > /dev/null device.  I am using the hotplug 2004-09-23
> > > > > script set.
> > > > 
> > > > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> > > 
> > > Hm, no, this might be a udev specific issue.  I'm also now seeing this,
> > > and in fact, can't get 2.6.12 to work at all with the latest udev (059).
> > > Will debug more tonight...
> > 
> > Kay, this seems to be caused by udevstart running the
> > /etc/hotplug.d/defalt/default.hotplug link (points to the "load a
> > module" hotplug stuff) before we've created anything in /dev.  This is
> > because of the udev_run_hotplugd program rule.  But for udevstart, we
> > don't want to run any hotplug stuff :(
> 
> Sure, that's wrong.
> 
> > Hm, any hints on how to fix this loop?
> 
> I hope so:
>   ENV{UDEVD_EVENT}="1", RUN+="/sbin/udev_run_hotplugd"
> 
> should prevent the hotplug stuff run, if it's not from the kernel.

Much better, thanks.  I've released 060 with this fix (for the gentoo
rules) in it.

Sorry for my shoddy testing on the 059 gentoo release, that was my
fault.

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Greg KH]

On Sat, Jul 02, 2005 at 08:59:13PM +0200, Stefan Schweizer wrote:
> Am Samstag 02 Juli 2005 17:58 schrieb Greg KH:
> > On Sat, Jul 02, 2005 at 10:22:15AM +0200, Marco d'Itri wrote:
> > > On Jul 02, Greg KH <greg@kroah.com> wrote:
> > > > On Fri, Jul 01, 2005 at 06:28:16PM -0700, Phil K wrote:
> > > > > Gentoo here, udev-059 causes the
> > > > > /etc/hotplug.d/default/default.hotplug script to
> > > > > complain bitterly at boot on all references to the
> > > > > /dev/null device.  I am using the hotplug 2004-09-23
> > > > > script set.
> > > >
> > > > This is a gentoo specific issue, stick with bugs.gentoo.org for now.
> > >
> > > Actually it happens on Debian systems too.
> >
> > Yeah, I don't know how my testing didn't uncover this :(
> >
> > Does 2.6.12 work for you with the 059 udev?
> 
> A quick survey in #gentoo-bugs showed that it is really slow (>1minute) on 
> startup for kernels:
> 	2.6.12-ck3
> 	2.6.12-mm1 (my own)

Yes, until I convert all gentoo packages that put stuff in /etc/dev.d/
to use a udev RUN= rule, it will be a bit slower, but in the end, it
will be a few orders of magnitude faster, give us a few weeks :)

> Another user with 2.6.12-gentoo-r3 reportet that he has no problems at all

They were lucky...

> Furthermore people seem to have problems when they did not upgrade yet to 
> linux-headers-2.6.11.

Yes, known bug, see the gentoo bugzilla for details.

> Can you please add linux-headers-2.6.11 to the udev DEPEND for the gentoo 
> Ebuild and maybe package.mask it, when it is so unstable?

No, I fixed udev (well, Marco did:
http://www.kernel.org/git/?p=linux/hotplug/udev.git;a=commit;h¡05fe495bdc95a24eaf2db557104534f6e1db34
)

060 is now out that fixes all of the above issues on gentoo, hopefully
debian will also be happier with it.  Sleep time now...

thanks,

greg k-h


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 294 bytes --]

On Jul 02, Kay Sievers <kay.sievers@vrfy.org> wrote:

> I hope so:
>   ENV{UDEVD_EVENT}=="1", RUN+="/sbin/udev_run_hotplugd"
> 
> should prevent the hotplug stuff run, if it's not from the kernel.
What about other RUN events? Should each one check for $UDEVD_EVENT?

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Sun, Jul 03, 2005 at 12:22:42PM +0200, Marco d'Itri wrote:
> On Jul 02, Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > I hope so:
> >   ENV{UDEVD_EVENT}="1", RUN+="/sbin/udev_run_hotplugd"
> > 
> > should prevent the hotplug stuff run, if it's not from the kernel.
> What about other RUN events? Should each one check for $UDEVD_EVENT?

There is no general rule. It depends on your setup and how you organize
things on bootup. SUSE does kernel event-replay from initramfs. udevstart
and coldplug are not used anymore.

udevstart did dev.d/ also with the old udev versions, so only the
hotplug.d/ replacement RUN rules you may disable with matching
$UDEVD_EVENT.

But it may also be possible to modify udevstart to replace coldplug.
Don't know, you may have a better view, if this is possible?

Btw:
The next major udev version will depend on kernel 2.6.12 and will support
fast node creation in the udev daemon itself. _Simple_ rules will be able to
handled by the daemon itself. That will hopefully make event-replay fast
and easy and be a better approach than udevstart and coldplug.

Thanks,
Kay



-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 2052 bytes --]

On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:

> There is no general rule. It depends on your setup and how you organize
> things on bootup. SUSE does kernel event-replay from initramfs. udevstart
> and coldplug are not used anymore.
> 
> udevstart did dev.d/ also with the old udev versions, so only the
> hotplug.d/ replacement RUN rules you may disable with matching
> $UDEVD_EVENT.
> 
> But it may also be possible to modify udevstart to replace coldplug.
> Don't know, you may have a better view, if this is possible?
This same tought occurred to me before, coldplugging could come for
free since udevstart already walks sysfs and generates events for the
PCI and USB devices, but I am not sure about all the possible
consequences.

My first tought about this is that it's too early to mandate using
initramfs, so udevstart should continue to be fully supported until
other reasons will make using initramfs mandatory for everybody.
My second tought is that I do not understand exactly which events are
generated and when in these cases, both when drivers are modular and
statically built in the kernel:
- initramfs with events replaying
- udevstart
- coldplugging

so I would welcome some documentation about this.
(The "when" part is very important, because it's crucial to know at
which point of the boot process the RUN scripts will be started.)

Did you look yet at my hotplug-light scripts?
http://www.bofh.it/~md/debian/

> The next major udev version will depend on kernel 2.6.12 and will support
> fast node creation in the udev daemon itself. _Simple_ rules will be able to
> handled by the daemon itself. That will hopefully make event-replay fast
> and easy and be a better approach than udevstart and coldplug.
Why? How was udev determined to be slower?
I am a bit concerned about increasing complexity and code size.

(And I am very concerned about the dependency on a kernel newer than the
2.6.8 shipped in Debian 3.1, which I fear will make upgrades painful.)

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Sun, Jul 03, 2005 at 02:15:47PM +0200, Marco d'Itri wrote:
> On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > There is no general rule. It depends on your setup and how you organize
> > things on bootup. SUSE does kernel event-replay from initramfs. udevstart
> > and coldplug are not used anymore.
> > 
> > udevstart did dev.d/ also with the old udev versions, so only the
> > hotplug.d/ replacement RUN rules you may disable with matching
> > $UDEVD_EVENT.
> > 
> > But it may also be possible to modify udevstart to replace coldplug.
> > Don't know, you may have a better view, if this is possible?
> This same tought occurred to me before, coldplugging could come for
> free since udevstart already walks sysfs and generates events for the
> PCI and USB devices, but I am not sure about all the possible
> consequences.

But we need to walk through /sys/devices to catch devices there
for modules-load, ..., right? That is not handled by udevstart
currently.

> My first tought about this is that it's too early to mandate using
> initramfs, so udevstart should continue to be fully supported until
> other reasons will make using initramfs mandatory for everybody.

Yes, righ. It is not mandatory to use replay. But it works very nicely
and I expect we all will go that route sooner or later. Time will tell... :)

> My second tought is that I do not understand exactly which events are
> generated and when in these cases, both when drivers are modular and
> statically built in the kernel:
> - initramfs with events replaying

All events are catched and can be replayed. No logic needed, it's all
possible with the normal udev rules without any coldplug-scan or
whatever.

> - udevstart

Handles only devices with a node.

> - coldplugging

Scans for devices to load modules too.

Maybe we can combine udevstart and coldplugging, but someone that will
actually use it will need to help here. I'm expected to concentrate on the
event-replay stuff.

> so I would welcome some documentation about this.
> (The "when" part is very important, because it's crucial to know at
> which point of the boot process the RUN scripts will be started.)

We have an initramfs that stores every event with udeveventrecorder in
the filesystem and when the real root is mounted udevd is started with
the exec_queue blocked. After feeding the events to udevd, the queue
handling is switched on and udevd starts with the very fist events the
kernel has emitted and seamlessly receives and handles all new events
that arrive in the meantime in the queue.

> Did you look yet at my hotplug-light scripts?
> http://www.bofh.it/~md/debian/

Only a short look and it looks great compared to the current ones. But SUSE
does not use the hotplug package, it has it's own version which is highly
integrated into the configuration system.
We are going to revamp major parts and replace a huge part of it.

> > The next major udev version will depend on kernel 2.6.12 and will support
> > fast node creation in the udev daemon itself. _Simple_ rules will be able to
> > handled by the daemon itself. That will hopefully make event-replay fast
> > and easy and be a better approach than udevstart and coldplug.
> Why? How was udev determined to be slower?

On event replay you fire ~800 processes in a few seconds, while most of
these events are simple devices without any further processing. Tests showed
and my coworkers needed to convince me that doing this in-process is much
faster and you save some seconds bootup-time. (using udevstart does not have
this problem, cause it already does everything with only one process)

> I am a bit concerned about increasing complexity and code size.

It's not getting bigger, it will be lot smaller I expect, cause I will
remove a lot of old code and half of wait_for_sysfs, cause this is solved
with kernel changes. It will also be able to work without sysfs for simple
rules, which will make some people very happy.

> (And I am very concerned about the dependency on a kernel newer than the
> 2.6.8 shipped in Debian 3.1, which I fear will make upgrades painful.)

Debian will need to have shorter release cycles than 3 years. :)

The next version will definitely not work with kernels earlier than 2.6.12
and a next HAL version will require that udev version to be functional.
The recent HAL already depends on 2.6.11 anyway.
But I see no problem for Debian to stick with that version, or make some
tricky boot script to decide what udev version to run.
Also I don't see a problem to have official maintence releases on top of
the version before the cut if that is useful.

We have no other option to make progress with our integration work and need
to get rid of the old stuff during the next months to reach that goal.  We
will see big changes in the dynamic hardware handling and all that stuff will
only be possible with recent kernels, udev and HAL versions.

The commercial distributions have definitely no ressources to support so
many options and will just require specific kernel versions to work with
the rest of the system. They just make sure that the systems will run with
recent custom kernels, therefore udevstart will also be supported as long
as it is needed. But support for older kernels will need to be dropped
to keep the packages maintainable if the difference is that big like the
netlink-uevent, the modalias-support, no need to wait_for_sysfs for most
of the events and the $MAJOR/$MINOR in the event env.

Thanks,
Kay


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 806 bytes --]

On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:

> But we need to walk through /sys/devices to catch devices there
> for modules-load, ..., right? That is not handled by udevstart
> currently.
udevstart should walk /sys/bus/*/devices/ to support $MODALIAS (look at
my modalias.rc).

> On event replay you fire ~800 processes in a few seconds, while most of
> these events are simple devices without any further processing. Tests showed
> and my coworkers needed to convince me that doing this in-process is much
> faster and you save some seconds bootup-time. (using udevstart does not have
> this problem, cause it already does everything with only one process)
Why not move the whole udev in udevd then, and make it fork and continue
processing only when needed?

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Sun, Jul 03, 2005 at 06:26:29PM +0200, Marco d'Itri wrote:
> On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > But we need to walk through /sys/devices to catch devices there
> > for modules-load, ..., right? That is not handled by udevstart
> > currently.
> udevstart should walk /sys/bus/*/devices/ to support $MODALIAS (look at
> my modalias.rc).

Is there anything else than adding modalias support to udevstart to
replace what coldplug is doing today? But hey, this also only works
with 2.6.12 too. :)

> > On event replay you fire ~800 processes in a few seconds, while most of
> > these events are simple devices without any further processing. Tests showed
> > and my coworkers needed to convince me that doing this in-process is much
> > faster and you save some seconds bootup-time. (using udevstart does not have
> > this problem, cause it already does everything with only one process)
> Why not move the whole udev in udevd then, and make it fork and continue
> processing only when needed?

Sure, that's the plan and already works that way, but with kernels older
2.6.12, udevd would need to read the "dev" file in sysfs and the sysfs
files may arrive at any time later cause of bad sysfs-timing. So we
would need wait_for_sysfs in the daemon with stat() loops for concurrent
events in parallel compared to just use $MAJOR/$MINOR from the netlink event.
I don't say that this is impossible, but it definitely is a total mess.

With netlink, modalias and $MAJOR/$MINOR, only available after 2.6.11/2.6.12,
there is a good reason to drop old kernel-versions.

Thanks,
Kay


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 725 bytes --]

On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:

> > > But we need to walk through /sys/devices to catch devices there
> > > for modules-load, ..., right? That is not handled by udevstart
> > > currently.
> > udevstart should walk /sys/bus/*/devices/ to support $MODALIAS (look at
> > my modalias.rc).
> Is there anything else than adding modalias support to udevstart to
> replace what coldplug is doing today? But hey, this also only works
Look at hotplug-light. pnp can trivially get $MODALIAS support and serio
can get it with some effort, but I'm not sure about ide.rc and scsi.rc.
But I see nothing wrong with keeping a few shell coldplug scripts, they
are not a boot bottleneck.

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Sun, Jul 03, 2005 at 10:15:34PM +0200, Marco d'Itri wrote:
> On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > > > But we need to walk through /sys/devices to catch devices there
> > > > for modules-load, ..., right? That is not handled by udevstart
> > > > currently.
> > > udevstart should walk /sys/bus/*/devices/ to support $MODALIAS (look at
> > > my modalias.rc).
> > Is there anything else than adding modalias support to udevstart to
> > replace what coldplug is doing today? But hey, this also only works
> Look at hotplug-light. pnp can trivially get $MODALIAS support and serio
> can get it with some effort, but I'm not sure about ide.rc and scsi.rc.
> But I see nothing wrong with keeping a few shell coldplug scripts, they
> are not a boot bottleneck.

I agree, it's fine to do it that way. An alternative to running the
scan-scripts and invoke something itself, may be to synthesize events
by reading sysfs and pass them as an event to the daemon which will
handle it.
That way it would be very similar to event-replay and would use the same
rules as the normal event. Also events would run in parallel managed by
udevd.
That may be a nice fallback if the events from initramfs are not
available, but similar in operation. What do you think?

Kay


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 540 bytes --]

On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:

> > (And I am very concerned about the dependency on a kernel newer than the
> > 2.6.8 shipped in Debian 3.1, which I fear will make upgrades painful.)
> Debian will need to have shorter release cycles than 3 years. :)
This is not actually relevant, there would be the same risks of
breakage while upgrading even if it had a 2.6.10 or 2.6.11 kernel
because users need to upgrade to 2.6.12 and reboot with the old udev 059
before they can install udev 060.

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: udev-059 and default.hotplug script

[Kay Sievers]

On Tue, Jul 05, 2005 at 12:28:43PM +0200, Marco d'Itri wrote:
> On Jul 03, Kay Sievers <kay.sievers@vrfy.org> wrote:
> 
> > > (And I am very concerned about the dependency on a kernel newer than the
> > > 2.6.8 shipped in Debian 3.1, which I fear will make upgrades painful.)
> > Debian will need to have shorter release cycles than 3 years. :)
> This is not actually relevant, there would be the same risks of
> breakage while upgrading even if it had a 2.6.10 or 2.6.11 kernel
> because users need to upgrade to 2.6.12 and reboot with the old udev 059
> before they can install udev 060.

The point is that you should require a certain kernel version for a
actual release. Easy as that, and no risk. But I see the problem in your
case, that sometimes you don't have a _actual_ release. :)

Kay


-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id\x16492&op=click
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: udev-059 and default.hotplug script

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 704 bytes --]

On Jul 05, Kay Sievers <kay.sievers@vrfy.org> wrote:

> > This is not actually relevant, there would be the same risks of
> > breakage while upgrading even if it had a 2.6.10 or 2.6.11 kernel
> > because users need to upgrade to 2.6.12 and reboot with the old udev 059
> > before they can install udev 060.
> The point is that you should require a certain kernel version for a
> actual release. Easy as that, and no risk. But I see the problem in your
And my point is that it's not possible to upgrade them at the same time,
at least for Debian (because the udev package needs to check the version
of the *running* kernel and abort the upgrade if it's not recent enough).

-- 
ciao,
Marco

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]