You are not logged in.
@Daniel_F. You provided a link with a patched udev ( http://www.electron2k.com/udev165arch/u … pkg.tar.xz ). Could you also post the file required to regenerate the package (PKGBUILD, original source, patch, etc...). I usually do not install precompiled packages if I cannot have the files to regenerate it (providing the source is also required by the GPL).
Read the whole topic ! (Basic reading skils are required to use the forum !) All the information you need to do it yourself is posted. I just used ABS and hand-patched the offending file, since it's just one line !
.
I don't work for you ! The compiled package was posted as a courtesy for those that can't / don't want to do it themselves. So don't come here demanding stuff. You are welcome not to install it !
.
About GPL compliance : we all know how it works !! You don't need to teach us ! ABS got the sources for me automatically, but you may not trust that source enough. So, here, just for you: http://www.kernel.org/pub/linux/utils/k … 165.tar.gz (can you trust that source ???)
The funny thing is that you are probably one of those people that downloaded dozens of those "oh so pretty" Windows screensavers. (from very reputable sorces, of couse !!)
Last edited by Daniel_F (2011-01-11 18:47:34)
Offline
Hi! I'm using Daniel_F's udev patched version since yesterday without any kernel panics, everything works fine
thank you!
Offline
Thanks go to the following users for the feedback:
Deadguy1
bones
rabid_works
nnoell
Mektub
Perre
eded
acoples
And for the other 67 people that downloaded it so far: how is it working for you ??
Offline
Read the whole topic ! (Basic reading skils are required to use the forum !) All the information you need to do it yourself is posted. I just used ABS and hand-patched the offending file, since it's just one line !
.I don't work for you ! The compiled package was posted as a courtesy for those that can't / don't want to do it themselves. So don't come here demanding stuff. You are welcome not to install it !
.About GPL compliance : we all know how it works !! You don't need to teach us ! ABS got the sources for me automatically, but you may not trust that source enough. So, here, just for you: http://www.kernel.org/pub/linux/utils/k … 165.tar.gz (can you trust that source ???)
It is bad practice to post a binary package without providing the full source, I mean the full source with the configure options, etc... abs do not provide the source of the patch, which I am interested in. You just say in post 33 that you have patched it according to a thread referenced by another forumer. Following the links of the previous posts, it is question of this patch: https://bugs.archlinux.org/task/22343?getfile=6224. Is it this really patch you are talking about? In this case, I provided the required buildscript myself since it seems that you are unwilling/unable to do it: http://sharesend.com/ftv9w
The funny thing is that you are probably one of those people that downloaded dozens of those "oh so pretty" Windows screensavers. (from very reputable sorces, of couse !!)
I just ask what I think is normal to ask: the full source of a provided binary package. Your only answer is that I have nothing to "demand" (while I have just asked) and that you know you have to provide it according to the GPL ! There is a little contradiction in your answer.
You imagination about "funny windows screensaver" is completely off topic. The real funny thing is that you probably spend more time to write your answer than what would have been necessary to provide the required buildscript (or at least a clear link to the patch).
Last edited by olive (2011-01-11 20:54:21)
Offline
--mod edit: tone it down if you want to be around longer on these forums
@olive: From post #33:
I've compiled a patched version of udev 165 according to post #36 of the thread referred to by sm4tik.
which, since you can't figure out on your own, means this: http://www.linuxquestions.org/questions … ost4212362
I even explained that on post #41
There is no need for build scripts or patch files or configure options. It's all vanilla ! Just use ABS, comment out one line of one file (it's easier to do it by hand !) and generate your package. No black magic, voodoo or anything like it... It can't get any simpler. That the beauty of Arch !
Now if you can't read or don't know how to use ABS that's not my fault.... (BTW: enjoy your kernel panics !)
Last edited by litemotiv (2011-01-11 21:55:28)
Offline
Daniel_F, the derogatory tone and remarks you are displaying in this thread are unacceptable. I ask you to stop with it immediately and react normally like everyone else, consider this you final warning.
ᶘ ᵒᴥᵒᶅ
Offline
Sorry for the trouble.
I'm removing the patched udev. Don't want to be liable for GPL violation, you know.
Won't be posting here again !!
Offline
Daniel F. Could you stop to feel insulted just because I have just asked a clear link to the patch. I you reread my post I did not accuse you of anything, just asked.
There was question of a lot of several patches and that was not clear. You at least provided what I want: you have builded the package according to the abs source and you have finally give a clear link to the patch. Was that so difficult?
Offline
I'm eagerly following this thread in hopes someone would make a PKGBUILD and a patch file. In the meantime my desktop and laptop keep getting panics.....
This is based on the linuxquestions post #36 http://www.linuxquestions.org/questions … ost4212362
This is my first attempt at creating and posting a patch and a PKGBUILD. It works for me, I hope it does for you....
ata_id.patch
--- ata_id.c 2010-11-09 19:30:53.000000000 -0500
+++ /home/cvandesande/abs/udev/src/udev-165/extras/ata_id/ata_id.c 2011-01-11 19:07:34.756572578 -0500
@@ -267,7 +267,7 @@
io_hdr.dxfer_direction = SG_DXFER_FROM_DEV;
io_hdr.timeout = COMMAND_TIMEOUT_MSEC;
- ret = ioctl(fd, SG_IO, &io_hdr);
+/* ret = ioctl(fd, SG_IO, &io_hdr);*/
if (ret != 0)
goto out;
} else {
New PKGBUILD
# $Id: PKGBUILD 104538 2011-01-03 22:13:06Z tpowa $
# Maintainer: Aaron Griffin <aaron@archlinux.org>
# Maintainer: Tobias Powalowski <tpowa@archlinux.org>
# Maintainer: Thomas Bächler <thomas@archlinux.org>
pkgbase="udev"
pkgname=('udev' 'udev-compat')
pkgver=165
pkgrel=1
arch=(i686 x86_64)
url="http://www.kernel.org/pub/linux/utils/kernel/hotplug/udev.html"
license=('GPL')
groups=('base')
# older initscripts versions required start_udev
options=(!makeflags !libtool)
makedepends=('glibc' 'coreutils' 'util-linux' 'pciutils' 'libusb-compat' 'glib2' 'kernel26' 'gperf' 'libxslt' 'gobject-introspection')
source=(http://www.kernel.org/pub/linux/utils/kernel/hotplug/$pkgname-$pkgver.tar.bz2
80-drivers.rules
81-arch.rules
load-modules.sh
cdsymlinks.sh
ata_id.patch)
md5sums=('74c62aa82969278960b7ee81e94b6d8f'
'4427855146513a4703ab5c7eb8a0156e'
'f17c99779173ddc79f866cf3a5de5e68'
'f91fddc67609b45b244a624977c4247b'
'2e808ee78d237c478b57af2a68d43769'
'9e858ca62bbb67e62306cd75db36c689')
build() {
cd $srcdir/$pkgname-$pkgver
patch extras/ata_id/ata_id.c < ${startdir}/ata_id.patch
./configure --sysconfdir=/etc --with-rootlibdir=/lib --libexecdir=/lib/udev\
--sbindir=/sbin --with-systemdsystemunitdir=/lib/systemd/system
make
}
package_udev() {
pkgdesc="The userspace dev tools (udev)"
depends=('glibc' 'coreutils' 'util-linux' 'libusb-compat' 'glib2'
'module-init-tools>=3.11' 'pciutils')
install=udev.install
backup=(etc/udev/udev.conf
etc/modprobe.d/framebuffer_blacklist.conf)
conflicts=('pcmcia-cs' 'hotplug' 'initscripts<2009.07')
replaces=('devfsd')
cd $srcdir/$pkgname-$pkgver
make DESTDIR=${pkgdir} install
# Replace original 80-drivers.rules with custom one.
install -D -m644 $srcdir/80-drivers.rules $pkgdir/lib/udev/rules.d/80-drivers.rules
# Install our rule for permissions and symlinks
install -D -m644 $srcdir/81-arch.rules $pkgdir/lib/udev/rules.d/81-arch.rules
# install our module loading subsystem
install -D -m755 $srcdir/load-modules.sh $pkgdir/lib/udev/load-modules.sh
# install cdsymlinks.sh
install -D -m755 $srcdir/cdsymlinks.sh $pkgdir/lib/udev/cdsymlinks.sh
# disable error logging to prevent startup failures printed to vc on boot
sed -i -e 's|udev_log="err"|udev_log="0"|g' $pkgdir/etc/udev/udev.conf
# disable persistent cdromsymlinks and network by default
# and move it to /etc/udev/rules.d
mv $pkgdir/lib/udev/rules.d/75-persistent-net-generator.rules \
$pkgdir/etc/udev/rules.d/75-persistent-net-generator.rules.optional
mv $pkgdir/lib/udev/rules.d/75-cd-aliases-generator.rules \
$pkgdir/etc/udev/rules.d/75-cd-aliases-generator.rules.optional
# create framebuffer blacklist
mkdir -p $pkgdir/etc/modprobe.d/
for mod in $(find /lib/modules/*/kernel/drivers/video -name '*fb.ko' -exec basename {} .ko \;); do
echo "blacklist $mod" >> $pkgdir/etc/modprobe.d/framebuffer_blacklist.conf
done
# create static devices in /lib/udev/devices/
mkdir ${pkgdir}/lib/udev/devices/pts
mkdir ${pkgdir}/lib/udev/devices/shm
mknod -m 0600 ${pkgdir}/lib/udev/devices/console c 5 1
mknod -m 0666 ${pkgdir}/lib/udev/devices/null c 1 3
mknod -m 0660 ${pkgdir}/lib/udev/devices/zero c 1 5
mknod -m 0666 ${pkgdir}/lib/udev/devices/kmsg c 1 11
ln -snf /proc/self/fd ${pkgdir}/lib/udev/devices/fd
ln -snf /proc/self/fd/0 ${pkgdir}/lib/udev/devices/stdin
ln -snf /proc/self/fd/1 ${pkgdir}/lib/udev/devices/stdout
ln -snf /proc/self/fd/2 ${pkgdir}/lib/udev/devices/stderr
ln -snf /proc/kcore ${pkgdir}/lib/udev/devices/core
# these static devices are created for convenience, to autoload the modules if necessary
# /dev/loop0
mknod -m 0660 ${pkgdir}/lib/udev/devices/loop0 b 7 0
chgrp disk ${pkgdir}/lib/udev/devices/loop0
# /dev/net/tun
mkdir ${pkgdir}/lib/udev/devices/net
mknod -m 0666 ${pkgdir}/lib/udev/devices/net/tun c 10 200
# /dev/fuse
mknod -m 0666 ${pkgdir}/lib/udev/devices/fuse c 10 229
# /dev/ppp
mknod -m 0600 ${pkgdir}/lib/udev/devices/ppp c 108 0
# Replace dialout/tape/cdrom group in rules with uucp/storage/optical group
for i in $pkgdir/lib/udev/rules.d/*.rules; do
sed -i -e 's#GROUP="dialout"#GROUP="uucp"#g;
s#GROUP="tape"#GROUP="storage"#g;
s#GROUP="cdrom"#GROUP="optical"#g' $i
done
}
package_udev-compat() {
pkgdesc="The userspace dev tools (udev) - additional rules for older kernels"
depends=('udev')
groups=('')
cd $srcdir/$pkgname-$pkgver
install -d -m755 ${pkgdir}/lib/${pkgname}/rules.d
install -D -m644 ${srcdir}/${pkgname}-${pkgver}/rules/misc/30-kernel-compat.rules ${pkgdir}/lib/udev/rules.d/30-kernel-compat.rules
}
Last edited by monzer4819 (2011-01-12 00:41:48)
Offline
Its happening here too; although I didn't notice it until Jan 09 when I rebooted into the 2.6.37-ck kernel, guess I'm just lucky to usually have it work on the 1st boot?
Confirmed on all but fallback image on this old P4 laptop
=== START OF INFORMATION SECTION ===
Model Family: SAMSUNG SpinPoint M series
Device Model: SAMSUNG MP0603H
Serial Number: S03ZJ10Y628227
Firmware Version: UD100-14
User Capacity: 60,060,155,904 bytes
Device is: In smartctl database [for details use: -P show]
ATA Version is: 7
ATA Standard is: ATA/ATAPI-7 T13 1532D revision 0
Local Time is: Tue Jan 11 23:42:47 2011 EST
SMART support is: Available - device has SMART capability.
SMART support is: Enabled
=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED
Unfortunately no log of the panic; guess you need a camera
Any society that would give up a little liberty to gain a little security will deserve neither and lose both.
-Benjamin Franklin
The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man.
-George Bernard Shaw
Offline
i just met the same problem.
Offline
Booted correctly again with 2.6.37... A few times without a lockup.
Offline
I justed wanted to say that I'm running a self compiled version of 2.6.37 (very similar to arch defaults) and still experienced the panic, as did some people running Slackware Current. See http://www.linuxquestions.org/questions … ost4216527
Although it was intermittent, the patch I posted here yesterday appears to have corrected the issue. If I experience another panic, I will post it here.
Offline
Yeah, I just had the same kernel panic with .37 kernel, with the stock 135 udev, and also with 134. The fallback image seems being unaffected.
Last time it happened while copying files between two external USB2 drives.
Will try the above 135 patch, and report back.
Microshaft delenda est
Offline
Me too, just had a full lockup when a dvd was burnt and the tray opened. Downgraded to 164-3 until we get the official update...
Offline
Daniel_F 's patch (via olive 's buildscript) fixed it for me so far..
~10 boots with no freeze. Thanks both.
Offline
...
Last edited by Mr. Alex (2011-01-12 16:26:26)
Offline
Just noticed the following "glitch" in /var/log/messages after downgrading to 164:
Jan 12 17:15:00 localhost kernel: udev[38]: starting version 165
And 40 lines below that:
Jan 12 17:15:00 localhost kernel: udev[787]: starting version 164
I downgraded udev by using pacman -U old_package. I even removed udev completely and installed only version 164 but /var/log/messages keeps reporting that it starts 165 before 164. We'll see if that gives problems.
Offline
did you recreate initrd?
Offline
Ah yes of course, that did the trick!
I have to backup about 50-100 DVD's so I'll see if downgrading udev fixes my systems freezes.
Offline
same issue here. downgraded udev to 164 lets see if it helps.
Offline
.
Last edited by silent (2013-08-21 08:49:33)
Offline
We patched our udev-165-2 like this:
#
# Chakra Packages for Chakra, part of chakra-project.org
#
# maintainer (i686): Phil Miller <philm[at]chakra-project[dog]org>
# maintainer (x86_64): Manuel Tortosa <manutortosa[at]chakra-project[dot]org>
pkgbase="udev"
pkgname=('udev' 'udev-compat')
pkgver=165
pkgrel=2
arch=(i686 x86_64)
url="http://www.kernel.org/pub/linux/utils/kernel/hotplug/udev.html"
license=('GPL')
groups=('base')
# older initscripts versions required start_udev
options=(!makeflags !libtool)
makedepends=('glibc' 'coreutils' 'util-linux' 'libusb' 'pciutils' 'glib2' 'kernel26' 'gperf' 'libxslt' 'gobject-introspection>=0.10.0')
source=(http://www.kernel.org/pub/linux/utils/kernel/hotplug/$pkgname-$pkgver.tar.bz2
http://chakra-project.org/sources/udev/udev-165-ata-pass-through-memalign.patch
80-drivers.rules
81-chakra.rules
load-modules.sh
cdsymlinks.sh)
md5sums=('74c62aa82969278960b7ee81e94b6d8f'
'6e9b693ec3d2c3dadff240094d1cb09a'
'4427855146513a4703ab5c7eb8a0156e'
'a2bf84747f44915ec9c5e6e88b077845'
'f91fddc67609b45b244a624977c4247b'
'2e808ee78d237c478b57af2a68d43769')
build() {
cd $srcdir/$pkgname-$pkgver
patch -p1 < $srcdir/udev-165-ata-pass-through-memalign.patch || exit 1
./configure --sysconfdir=/etc --with-rootlibdir=/lib --libexecdir=/lib/udev\
--sbindir=/sbin --with-systemdsystemunitdir=/lib/systemd/system
make
}
package_udev() {
pkgdesc="The userspace dev tools (udev)"
depends=('glibc' 'coreutils' 'util-linux' 'libusb' 'pciutils' 'glib2'
'module-init-tools>=3.11')
install=udev.install
backup=(etc/udev/udev.conf
etc/modprobe.d/framebuffer_blacklist.conf)
conflicts=('pcmcia-cs' 'hotplug' 'initscripts<2009.07')
replaces=('devfsd')
cd $srcdir/$pkgname-$pkgver
make DESTDIR=${pkgdir} install
# Replace original 80-drivers.rules with custom one.
install -D -m644 $srcdir/80-drivers.rules $pkgdir/lib/udev/rules.d/80-drivers.rules
# Install our rule for permissions and symlinks
install -D -m644 $srcdir/81-chakra.rules $pkgdir/lib/udev/rules.d/81-chakra.rules
# install our module loading subsystem
install -D -m755 $srcdir/load-modules.sh $pkgdir/lib/udev/load-modules.sh
# install cdsymlinks.sh
install -D -m755 $srcdir/cdsymlinks.sh $pkgdir/lib/udev/cdsymlinks.sh
# disable error logging to prevent startup failures printed to vc on boot
sed -i -e 's|udev_log="err"|udev_log="0"|g' $pkgdir/etc/udev/udev.conf
# disable persistent cdromsymlinks and network by default
# and move it to /etc/udev/rules.d
mv $pkgdir/lib/udev/rules.d/75-persistent-net-generator.rules \
$pkgdir/etc/udev/rules.d/75-persistent-net-generator.rules.optional
mv $pkgdir/lib/udev/rules.d/75-cd-aliases-generator.rules \
$pkgdir/etc/udev/rules.d/75-cd-aliases-generator.rules.optional
# create framebuffer blacklist
mkdir -p $pkgdir/etc/modprobe.d/
for mod in $(find /lib/modules/*/kernel/drivers/video -name '*fb.ko' -exec basename {} .ko \;); do
echo "blacklist $mod" >> $pkgdir/etc/modprobe.d/framebuffer_blacklist.conf
done
# create static devices in /lib/udev/devices/
mkdir ${pkgdir}/lib/udev/devices/pts
mkdir ${pkgdir}/lib/udev/devices/shm
mknod -m 0600 ${pkgdir}/lib/udev/devices/console c 5 1
mknod -m 0666 ${pkgdir}/lib/udev/devices/null c 1 3
mknod -m 0660 ${pkgdir}/lib/udev/devices/zero c 1 5
mknod -m 0666 ${pkgdir}/lib/udev/devices/kmsg c 1 11
ln -snf /proc/self/fd ${pkgdir}/lib/udev/devices/fd
ln -snf /proc/self/fd/0 ${pkgdir}/lib/udev/devices/stdin
ln -snf /proc/self/fd/1 ${pkgdir}/lib/udev/devices/stdout
ln -snf /proc/self/fd/2 ${pkgdir}/lib/udev/devices/stderr
ln -snf /proc/kcore ${pkgdir}/lib/udev/devices/core
# these static devices are created for convenience, to autoload the modules if necessary
# /dev/loop0
mknod -m 0660 ${pkgdir}/lib/udev/devices/loop0 b 7 0
chgrp disk ${pkgdir}/lib/udev/devices/loop0
# /dev/net/tun
mkdir ${pkgdir}/lib/udev/devices/net
mknod -m 0666 ${pkgdir}/lib/udev/devices/net/tun c 10 200
# /dev/fuse
mknod -m 0666 ${pkgdir}/lib/udev/devices/fuse c 10 229
# /dev/ppp
mknod -m 0600 ${pkgdir}/lib/udev/devices/ppp c 108 0
# Replace dialout/tape/cdrom group in rules with uucp/storage/optical group
for i in $pkgdir/lib/udev/rules.d/*.rules; do
sed -i -e 's#GROUP="dialout"#GROUP="uucp"#g;
s#GROUP="tape"#GROUP="storage"#g;
s#GROUP="cdrom"#GROUP="optical"#g' $i
done
}
package_udev-compat() {
pkgdesc="The userspace dev tools (udev) - additional rules for older kernels"
depends=('udev')
groups=('')
cd $srcdir/$pkgname-$pkgver
install -d -m755 ${pkgdir}/lib/${pkgname}/rules.d
install -D -m644 ${srcdir}/${pkgname}-${pkgver}/rules/misc/30-kernel-compat.rules ${pkgdir}/lib/udev/rules.d/30-kernel-compat.rules
}
visit chakra-project.org to get latest Chakra GNU/Linux
Offline
I've been using udev 165 patched with the ata-pass-through-memalign patch (as per Amnon82's post above) for a couple of days now, no problems yet.
Offline
Amnon82, the memalign patch is working for you? I've been using it some, and haven't had any problems with it. Just curious, how many people do you have using it (eg how much testing is it getting)?
Edited to add:
I see that Slackware has also patched udev165. From the -current changelog:
a/udev-165-x86_64-2.txz: Rebuilt.
Revert patch to use ata_id, not scsi_id, on ATAPI devices.
Using ata_id on ATAPI devices can crash the kernel when udev starts at boot.
Thanks to EdGr on LQ for tracking down the responsible commit.
Last edited by beerad (2011-01-12 21:13:21)
Offline