patches from the debian package

patches from the debian package

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 320 bytes --]

cdsymlinks_numeric: trivial fix, the variable is initialized to '', not 0

scsi_id_tmp: when udevstart is started, /tmp is not writeable.

I updated the rules files to the new style, and added support for devfs
names for hardware RAID controllers.

Also, please remove etc/udev/debian/udev.permissions.

-- 
ciao,
Marco

[-- Attachment #2: cdsymlinks_numeric --]
[-- Type: text/plain, Size: 635 bytes --]

diff -ruNp udev-050.orig/extras/cdsymlinks.sh udev-050/extras/cdsymlinks.sh
--- udev-050.orig/extras/cdsymlinks.sh	2004-12-18 06:53:07.000000000 +0100
+++ udev-050/extras/cdsymlinks.sh	2005-01-18 18:44:07.000000000 +0100
@@ -159,7 +159,7 @@ output () {
 	COUNT=$(($COUNT+1))
       done
       # If the target dev ($1) is the current dev ($i), we output its name.
-      if test $(($NUMBERED_LINKS)) -ne 0 || test "$COUNT" -eq 0; then
+      if test $(($NUMBERED_LINKS)) -ne 0 || test "$COUNT" = ''; then
 	test "$i" = "$1" && echo -n " $2$COUNT"
       fi
       # If the link isn't in our "existing links" list, add it and increment

[-- Attachment #3: scsi_id_tmp --]
[-- Type: text/plain, Size: 484 bytes --]

diff -ruNp udev-050.orig/extras/scsi_id/scsi_id.c udev-050/extras/scsi_id/scsi_id.c
--- udev-050.orig/extras/scsi_id/scsi_id.c	2004-12-18 06:53:07.000000000 +0100
+++ udev-050/extras/scsi_id/scsi_id.c	2005-01-01 22:19:15.000000000 +0100
@@ -44,8 +44,8 @@
 /*
  * temporary names for mknod.
  */
-#define TMP_DIR	"/tmp"
-#define TMP_PREFIX "scsi"
+#define TMP_DIR	"/dev"
+#define TMP_PREFIX "tmp-scsi"
 
 /*
  * XXX Note the 'e' (send output to stderr in all cases), and 'c' (callout)

[-- Attachment #4: udev.rules --]
[-- Type: text/plain, Size: 4679 bytes --]

# There are a number of modifiers that are allowed to be used in some
# of the different fields. They provide the following subsitutions:
#
# %n the "kernel number" of the device.
#    For example, 'sda3' has a "kernel number" of '3'
# %e the smallest number for that name which does not matches an existing node
# %k the kernel name for the device.
# %M the kernel major number for the device
# %m the kernel minor number for the device
# %b the bus id for the device
# %c the string returned by the PROGRAM
# %s{filename} the content of a sysfs attribute.
# %% the '%' char itself.
#

# default permissions for block devices
SUBSYSTEM="block",			GROUP="disk"
SUBSYSTEM="block", SYSFS{removable}="1", GROUP="floppy"

# workaround for devices which do not report media changes
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/model", RESULT="IOMEGA ZIP *", \
  NAME{all_partitions}="%k%n", GROUP="floppy"

# IDE devices
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/media", RESULT="cdrom*", \
  GROUP="cdrom"
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/media", RESULT="floppy*", \
  GROUP="floppy"
BUS="ide", KERNEL="ht[0-9]*",	GROUP="tape"
BUS="ide", KERNEL="nht[0-9]*",	GROUP="tape"

# SCSI devices
BUS="scsi", KERNEL="sr[0-9]*", SYSFS{type}="5", \
  NAME="scd%n", SYMLINK="sr%n",			GROUP="cdrom"
BUS="scsi", KERNEL="sg[0-9]*", SYSFS{type}="5",	GROUP="cdrom"
BUS="scsi", KERNEL="st[0-9]*",			GROUP="tape"
BUS="scsi", KERNEL="nst[0-9]*",			GROUP="tape"
BUS="scsi", KERNEL="osst[0-9]*",		GROUP="tape"
BUS="scsi", KERNEL="nosst[0-9]*",		GROUP="tape"

# USB devices
BUS="usb", KERNEL="hiddev*",	NAME="usb/%k"
BUS="usb", KERNEL="auer[0-9]*",	NAME="usb/%k"
BUS="usb", KERNEL="legousbtower*", NAME="usb/%k", MODE="0666"
BUS="usb", KERNEL="dabusb*",	NAME="usb/%k"
BUS="usb", KERNEL="cpad[0-9]*",	NAME="usb/%k"
BUS="usb", KERNEL="ttyUSB*", SYSFS{product}="Palm Handheld*", SYMLINK="pilot"

# serial devices
SUBSYSTEM="tty",	GROUP="dialout"
KERNEL="capi",		NAME="capi20", SYMLINK="isdn/capi20"
KERNEL="capi[0-9]*",	NAME="capi/%n"

# vc devices
KERNEL="ptmx",		MODE="0666"
KERNEL="tty",		MODE="0666"

# video devices
SUBSYSTEM="video4linux", GROUP="video"
KERNEL="dvb*",		PROGRAM="/etc/udev/scripts/dvb.sh %k", \
			NAME="%c", GROUP="video"
KERNEL="fb[0-9]*",	GROUP="video"
KERNEL="nvidia*",	GROUP="video"
KERNEL="card[0-9]*",	NAME="dri/%k", GROUP="video"

# misc devices
KERNEL="random",        MODE="0666"
KERNEL="urandom",       MODE="0444"
KERNEL="mem",           MODE="0640", GROUP="kmem"
KERNEL="kmem",          MODE="0640", GROUP="kmem"
KERNEL="port",          MODE="0640", GROUP="kmem"
KERNEL="full",          MODE="0666"
KERNEL="null",          MODE="0666"
KERNEL="zero",          MODE="0666"
KERNEL="inotify",       MODE="0666"
KERNEL="sgi_fetchop",   MODE="0666"
KERNEL="sonypi",        MODE="0666"
KERNEL="agpgart",       GROUP="video"
KERNEL="nvram",         GROUP="nvram"
KERNEL="rtc",           MODE="0664", GROUP="audio"
KERNEL="hw_random",	NAME="hwrng"

# printers and parallel devices
SUBSYSTEM="printer",	GROUP="lp"
BUS="usb", KERNEL="lp[0-9]*", NAME="usb/%k"
KERNEL="pt[0-9]*",	GROUP="tape"
KERNEL="pht[0-9]*",	GROUP="tape"

KERNEL="fd[0-9]*",	GROUP="floppy"

KERNEL="cdemu[0-9]*",	NAME="cdemu/%n", GROUP="cdrom"
KERNEL="pktcdvd[0-9]*",	NAME="pktcdvd/%n", GROUP="cdrom"
KERNEL="pktcdvd",	NAME="pktcdvd/control", MODE="0644"

KERNEL="cpu[0-9]*",	NAME="cpu/%n/cpuid"
KERNEL="msr[0-9]*",	NAME="cpu/%n/msr"
KERNEL="microcode",	NAME="cpu/microcode"

KERNEL="umad*",		NAME="infiniband/%k"
KERNEL="issm*",		NAME="infiniband/%k"

KERNEL="tap[0-9]*",	NAME="net/%k"
KERNEL="tun",		NAME="net/%k"

# ALSA devices
SUBSYSTEM="sound",	GROUP="audio"
KERNEL="controlC[0-9]*", NAME="snd/%k"
KERNEL="hwC[D0-9]*",	NAME="snd/%k"
KERNEL="pcmC[D0-9cp]*",	NAME="snd/%k"
KERNEL="midiC[D0-9]*",	NAME="snd/%k"
KERNEL="timer",		NAME="snd/%k"
KERNEL="seq",		NAME="snd/%k"

# input devices
KERNEL="mice",		NAME="input/%k"
KERNEL="mouse[0-9]*",	NAME="input/%k"
KERNEL="event[0-9]*",	NAME="input/%k", \
  PROGRAM="/etc/udev/scripts/inputdev.sh %k", RESULT="inputdev", \
  MODE="0664", GROUP="video"
KERNEL="event[0-9]*",	NAME="input/%k"
KERNEL="js[0-9]*",	NAME="input/%k", MODE="0664"
KERNEL="ts[0-9]*",	NAME="input/%k"
KERNEL="uinput",	NAME="input/%k"

# Zaptel
KERNEL="zapctl",	NAME="zap/ctl"
KERNEL="zaptimer",	NAME="zap/timer"
KERNEL="zapchannel",	NAME="zap/channel"
KERNEL="zappseudo",	NAME="zap/pseudo"
KERNEL="zap[0-9]*",	NAME="zap/%n"

# device mapper creates its own device nodes, so ignore these
KERNEL="dm-[0-9]*",	NAME=""
KERNEL="device-mapper",	NAME="mapper/control"


[-- Attachment #5: devfs.rules --]
[-- Type: text/plain, Size: 6950 bytes --]

# There are a number of modifiers that are allowed to be used in some
# of the different fields. They provide the following subsitutions:
#
# %n the "kernel number" of the device.
#    For example, 'sda3' has a "kernel number" of '3'
# %e the smallest number for that name which does not matches an existing node
# %k the kernel name for the device.
# %M the kernel major number for the device
# %m the kernel minor number for the device
# %b the bus id for the device
# %c the string returned by the PROGRAM
# %s{filename} the content of a sysfs attribute.
# %% the '%' char itself.
#

# default permissions for block devices
SUBSYSTEM="block",			GROUP="disk"
SUBSYSTEM="block", SYSFS{removable}="1", GROUP="floppy"

# workaround for devices which do not report media changes
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/model", RESULT="IOMEGA ZIP *", \
  NAME{all_partitions}="%k%n", GROUP="floppy"

# IDE devices
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/media", RESULT="cdrom*", \
  GROUP="cdrom"
BUS="ide", KERNEL="hd[a-z]", SYSFS{removable}="1", \
  PROGRAM="/bin/cat /proc/ide/%k/media", RESULT="floppy*", \
  GROUP="floppy"
BUS="ide", KERNEL="hd[a-z]*", \
  PROGRAM="/etc/udev/scripts/ide-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="ide", KERNEL="ht[0-9]*",	GROUP="tape"
BUS="ide", KERNEL="nht[0-9]*",	GROUP="tape"

# SCSI devices
BUS="scsi", KERNEL="sd[a-z]", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="scsi", KERNEL="sd[a-z][0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="scsi", KERNEL="sd[a-i][a-z]", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="scsi", KERNEL="sd[a-i][a-z][0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="scsi", KERNEL="s[gr][0-9]*", SYSFS{type}="5", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}", GROUP="cdrom"
BUS="scsi", KERNEL="sr[0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}", GROUP="cdrom"
BUS="scsi", KERNEL="sg[0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}"
BUS="scsi", KERNEL="st[0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}", GROUP="tape"
BUS="scsi", KERNEL="nst[0-9]*", \
  PROGRAM="/etc/udev/scripts/scsi-devfs.sh %k %b %n", \
  NAME="%c{1}", SYMLINK="%k %c{2}", GROUP="tape"

# RAID controllers
KERNEL="cciss!*",	PROGRAM="/etc/udev/scripts/raid-devfs.sh %k", \
			NAME="%c{1}", SYMLINK="%k"
KERNEL="ida!*",		PROGRAM="/etc/udev/scripts/raid-devfs.sh %k", \
			NAME="%c{1}", SYMLINK="%k"
KERNEL="rd!*",		PROGRAM="/etc/udev/scripts/raid-devfs.sh %k", \
			NAME="%c{1}", SYMLINK="%k"

# other block devices
KERNEL="md[0-9]*",	NAME="md/%n",
KERNEL="loop[0-9]*",	NAME="loop/%n"
KERNEL="ram[0-9]*",	NAME="rd/%n"

# USB devices
BUS="usb", KERNEL="hiddev*",	NAME="usb/%k"
BUS="usb", KERNEL="auer[0-9]*",	NAME="usb/%k"
BUS="usb", KERNEL="legousbtower*", NAME="usb/%k", MODE="0666"
BUS="usb", KERNEL="dabusb*",	NAME="usb/%k"
BUS="usb", KERNEL="cpad[0-9]*",	NAME="usb/%k"
BUS="usb", KERNEL="ttyUSB*", SYSFS{product}="Palm Handheld*", SYMLINK="pilot"

# serial devices
SUBSYSTEM="tty",	GROUP="dialout"
KERNEL="ttyS[0-9]*",	NAME="tts/%n"
KERNEL="ttyUSB[0-9]*",	NAME="tts/USB%n"
KERNEL="rfcomm[0-9]*",	NAME="bluetooth/rfcomm/%n",
KERNEL="capi",		NAME="capi20", SYMLINK="isdn/capi20"
KERNEL="capi[0-9]*",	NAME="capi/%n"

# vc devices
KERNEL="ptmx",		MODE="0666"
KERNEL="tty",		MODE="0666"
KERNEL="tty[0-9]*",	NAME="vc/%n"
KERNEL="vcs",		NAME="vcc/0"
KERNEL="vcs[0-9]*",	NAME="vcc/%n"
KERNEL="vcsa",		NAME="vcc/a0"
KERNEL="vcsa[0-9]*",	NAME="vcc/a%n"

# video devices
SUBSYSTEM="video4linux", GROUP="video"
KERNEL="video[0-9]*",	NAME="v4l/%k"
KERNEL="radio[0-9]*",	NAME="v4l/%k"
KERNEL="vbi[0-9]*",	NAME="v4l/%k"
KERNEL="vtx[0-9]*",	NAME="v4l/%k"
KERNEL="dvb*",		PROGRAM="/etc/udev/scripts/dvb.sh %k", \
			NAME="%c", GROUP="video"
KERNEL="fb[0-9]*",	NAME="fb/%n", GROUP="video"
KERNEL="nvidia*",	GROUP="video"
KERNEL="card[0-9]*",	NAME="dri/%k", GROUP="video"

# misc devices
KERNEL="random",	MODE="0666"
KERNEL="urandom",	MODE="0444"
KERNEL="mem",		MODE="0640", GROUP="kmem"
KERNEL="kmem",		MODE="0640", GROUP="kmem"
KERNEL="port",		MODE="0640", GROUP="kmem"
KERNEL="full",		MODE="0666"
KERNEL="null",		MODE="0666"
KERNEL="zero",		MODE="0666"
KERNEL="inotify",	MODE="0666"
KERNEL="sgi_fetchop",	MODE="0666"
KERNEL="sonypi",	MODE="0666"
KERNEL="apm_bios",	NAME="misc/%k",
KERNEL="agpgart",	NAME="misc/%k", GROUP="video"
KERNEL="nvram",		NAME="misc/%k", GROUP="nvram"
KERNEL="psaux",		NAME="misc/%k"
KERNEL="rtc",		NAME="misc/%k", MODE="0644", GROUP="audio"
KERNEL="hw_random",	NAME="hwrng"

# printers and parallel devices
SUBSYSTEM="printer",	GROUP="lp"
BUS="usb", KERNEL="lp[0-9]*", NAME="usb/%k"
KERNEL="lp[0-9]*",	NAME="printers/%n"
KERNEL="pt[0-9]*",	GROUP="tape"
KERNEL="pht[0-9]*",	GROUP="tape"

KERNEL="i2c-[0-9]*",	NAME="i2c/%n"
KERNEL="fd[0-9]*",	NAME="floppy/%n", GROUP="floppy"

KERNEL="cdemu[0-9]*",	NAME="cdemu/%n", GROUP="cdrom"
KERNEL="pktcdvd[0-9]*",	NAME="pktcdvd/%n", GROUP="cdrom"
KERNEL="pktcdvd",	NAME="pktcdvd/control", MODE="0644"

KERNEL="cpu[0-9]*",	NAME="cpu/%n/cpuid"
KERNEL="msr[0-9]*",	NAME="cpu/%n/msr"
KERNEL="microcode",	NAME="cpu/microcode"

KERNEL="umad*",		NAME="infiniband/%k"
KERNEL="issm*",		NAME="infiniband/%k"

KERNEL="tap[0-9]*",	NAME="net/%k"
KERNEL="tun",		NAME="net/%k"

# sound devices
SUBSYSTEM="sound",	GROUP="audio"
KERNEL="controlC[0-9]*", NAME="snd/%k"
KERNEL="hwC[D0-9]*",	NAME="snd/%k"
KERNEL="pcmC[D0-9cp]*",	NAME="snd/%k"
KERNEL="midiC[D0-9]*",	NAME="snd/%k"
KERNEL="timer",		NAME="snd/%k"
KERNEL="seq",		NAME="snd/%k"
KERNEL="audio*",	NAME="sound/%k"
KERNEL="dsp*",		NAME="sound/%k"
KERNEL="adsp*",		NAME="sound/%k"
KERNEL="midi*",		NAME="sound/%k"
KERNEL="mixer*",	NAME="sound/%k"
KERNEL="sequencer*",	NAME="sound/%k"
KERNEL="amidi*",	NAME="sound/%k"
KERNEL="dmmidi*",	NAME="sound/%k"
KERNEL="admmidi*",	NAME="sound/%k"

# input devices
KERNEL="mice",		NAME="input/%k"
KERNEL="mouse[0-9]*",	NAME="input/%k"
KERNEL="event[0-9]*",   NAME="input/%k", \
  PROGRAM="/etc/udev/scripts/inputdev.sh %k", RESULT="inputdev", \
  MODE="0664", GROUP="video"
KERNEL="event[0-9]*",	NAME="input/%k"
KERNEL="js[0-9]*",	NAME="input/%k", MODE="0664"
KERNEL="ts[0-9]*",	NAME="input/%k"
KERNEL="uinput",	NAME="input/%k"

# Zaptel
KERNEL="zapctl",	NAME="zap/ctl"
KERNEL="zaptimer",	NAME="zap/timer"
KERNEL="zapchannel",	NAME="zap/channel"
KERNEL="zappseudo",	NAME="zap/pseudo"
KERNEL="zap[0-9]*",	NAME="zap/%n"

# device mapper creates its own device nodes, so ignore these
KERNEL="dm-[0-9]*",	NAME=""
KERNEL="device-mapper",	NAME="mapper/control"


[-- Attachment #6: raid-devfs.sh --]
[-- Type: application/x-sh, Size: 1218 bytes --]

Re: patches from the debian package

[Kay Sievers]

On Sat, 2005-02-05 at 16:50 +0100, Marco d'Itri wrote:
> cdsymlinks_numeric: trivial fix, the variable is initialized to '', not 0
> 
> scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> 
> I updated the rules files to the new style, and added support for devfs
> names for hardware RAID controllers.
> 
> Also, please remove etc/udev/debian/udev.permissions.

Everything applied. You may check it here:
  http://vrfy.bkbits.net:8080/udev/ChangeSet@-2d?nav=index.html

Thanks,
Kay



-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
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: patches from the debian package

[Patrick Mansfield]

On Sat, Feb 05, 2005 at 04:50:31PM +0100, Marco d'Itri wrote:

> 
> scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> 

What if /dev does not exist (udev_root modified) or perhaps not writable?

Does udev put temp files under /dev?

-- Patrick Mansfield


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
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: patches from the debian package

[Marco d'Itri]

[-- Attachment #1: Type: text/plain, Size: 406 bytes --]

On Feb 06, Patrick Mansfield <patmans@us.ibm.com> wrote:

> > scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> What if /dev does not exist (udev_root modified) or perhaps not writable?
Good point, but I believe that this is a much more uncommon situation.

> Does udev put temp files under /dev?
Yes, the location for the udev database is defined in udev.conf.

-- 
ciao,
Marco

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

Re: patches from the debian package

[Kay Sievers]

On Sun, 2005-02-06 at 11:05 -0800, Patrick Mansfield wrote:
> On Sat, Feb 05, 2005 at 04:50:31PM +0100, Marco d'Itri wrote:
> 
> > 
> > scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> > 
> 
> What if /dev does not exist (udev_root modified) or perhaps not writable?
> 
> Does udev put temp files under /dev?

Hmm, what about letting udev create the temporary node, instead of doing
it in the *_id programs itself? That may be the best solution.

Something like:
  PROGRAM="/sbin/scsi_id -d %N", ...

What do you think?

Thanks,
Kay



-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
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: patches from the debian package

[Patrick Mansfield]

On Mon, Feb 07, 2005 at 12:13:09AM +0100, Kay Sievers wrote:
> On Sun, 2005-02-06 at 11:05 -0800, Patrick Mansfield wrote:
> > On Sat, Feb 05, 2005 at 04:50:31PM +0100, Marco d'Itri wrote:
> > 
> > > 
> > > scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> > > 
> > 
> > What if /dev does not exist (udev_root modified) or perhaps not writable?
> > 
> > Does udev put temp files under /dev?
> 
> Hmm, what about letting udev create the temporary node, instead of doing
> it in the *_id programs itself? That may be the best solution.
> 
> Something like:
>   PROGRAM="/sbin/scsi_id -d %N", ...
> 
> What do you think?

Yes, that would be nice.

scsi_id pre-dates the addition of any of the % code, and relies on the
DEVPATH environment variable.

So it also needs code changes to use the DEVPATH plus any command line
option, and/or a new % value for the sysfs path, like:

	PROGRAM="scsi_id -d %N -s %p", ...

It should still figure out whether or not to print a trailing "\n" for
normal command line usage. DEVPATH set could be the trigger, though that
is not clean (using -s <devpath> for sysfs path, but then using DEVPATH to
figure out if \n is printed).

-- Patrick Mansfield


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Kay Sievers]

On Mon, 2005-02-07 at 10:39 -0800, Patrick Mansfield wrote:
> On Mon, Feb 07, 2005 at 12:13:09AM +0100, Kay Sievers wrote:
> > On Sun, 2005-02-06 at 11:05 -0800, Patrick Mansfield wrote:
> > > On Sat, Feb 05, 2005 at 04:50:31PM +0100, Marco d'Itri wrote:
> > > 
> > > > 
> > > > scsi_id_tmp: when udevstart is started, /tmp is not writeable.
> > > > 
> > > 
> > > What if /dev does not exist (udev_root modified) or perhaps not writable?
> > > 
> > > Does udev put temp files under /dev?
> > 
> > Hmm, what about letting udev create the temporary node, instead of doing
> > it in the *_id programs itself? That may be the best solution.
> > 
> > Something like:
> >   PROGRAM="/sbin/scsi_id -d %N", ...
> > 
> > What do you think?
> 
> Yes, that would be nice.
> 
> scsi_id pre-dates the addition of any of the % code, and relies on the
> DEVPATH environment variable.
> 
> So it also needs code changes to use the DEVPATH plus any command line
> option, and/or a new % value for the sysfs path, like:
> 
> 	PROGRAM="scsi_id -d %N -s %p", ...

Ok, I did that. We have %N and %p now. 

> It should still figure out whether or not to print a trailing "\n" for
> normal command line usage. DEVPATH set could be the trigger, though that
> is not clean (using -s <devpath> for sysfs path, but then using DEVPATH to
> figure out if \n is printed).

RESULT does not care about a trailing \n.

Thanks,
Kay



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Kay Sievers]

[offlist]

On Tue, 2005-02-08 at 20:50 +0100, Kay Sievers wrote:
> On Mon, 2005-02-07 at 10:39 -0800, Patrick Mansfield wrote:
> > On Mon, Feb 07, 2005 at 12:13:09AM +0100, Kay Sievers wrote:
> > > On Sun, 2005-02-06 at 11:05 -0800, Patrick Mansfield wrote:

> > > Hmm, what about letting udev create the temporary node, instead of doing
> > > it in the *_id programs itself? That may be the best solution.
> > > 
> > > Something like:
> > >   PROGRAM="/sbin/scsi_id -d %N", ...
> > > 
> > > What do you think?
> > 
> > Yes, that would be nice.
> > 
> > scsi_id pre-dates the addition of any of the % code, and relies on the
> > DEVPATH environment variable.
> > 
> > So it also needs code changes to use the DEVPATH plus any command line
> > option, and/or a new % value for the sysfs path, like:
> > 
> > 	PROGRAM="scsi_id -d %N -s %p", ...
> 
> Ok, I did that. We have %N and %p now. 

You can pull that into your tree:
  bk pull bk://vrfy.bkbits.net/udev

if you are faster than Greg. :)

Cheers,
Kay



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Patrick Mansfield]

On Tue, Feb 08, 2005 at 08:50:10PM +0100, Kay Sievers wrote:
> On Mon, 2005-02-07 at 10:39 -0800, Patrick Mansfield wrote:
> > So it also needs code changes to use the DEVPATH plus any command line
> > option, and/or a new % value for the sysfs path, like:
> > 
> > 	PROGRAM="scsi_id -d %N -s %p", ...
> 
> Ok, I did that. We will have %N and %p now. 

> > It should still figure out whether or not to print a trailing "\n" for
> > normal command line usage. DEVPATH set could be the trigger, though that
> > is not clean (using -s <devpath> for sysfs path, but then using DEVPATH to
> > figure out if \n is printed).
> 
> RESULT does not care about a trailing \n.

Okay. Then there is no change needed in scsi_id, and that code should be
removed from it. I'm not sure if scsi_id DEVPATH environment variable
usage should be removed (for backwards compatibility).

Given this change, can you revert the tmp directory change (if Marco's OK
with that)? Mainly for non-udev command line usage.

Thanks ...

-- Patrick Mansfield


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Greg KH]

On Tue, Feb 08, 2005 at 08:52:29PM +0100, Kay Sievers wrote:
> [offlist]

Hm, looks like it was on the list to me :)

> On Tue, 2005-02-08 at 20:50 +0100, Kay Sievers wrote:
> > On Mon, 2005-02-07 at 10:39 -0800, Patrick Mansfield wrote:
> > > On Mon, Feb 07, 2005 at 12:13:09AM +0100, Kay Sievers wrote:
> > > > On Sun, 2005-02-06 at 11:05 -0800, Patrick Mansfield wrote:
> 
> > > > Hmm, what about letting udev create the temporary node, instead of doing
> > > > it in the *_id programs itself? That may be the best solution.
> > > > 
> > > > Something like:
> > > >   PROGRAM="/sbin/scsi_id -d %N", ...
> > > > 
> > > > What do you think?
> > > 
> > > Yes, that would be nice.
> > > 
> > > scsi_id pre-dates the addition of any of the % code, and relies on the
> > > DEVPATH environment variable.
> > > 
> > > So it also needs code changes to use the DEVPATH plus any command line
> > > option, and/or a new % value for the sysfs path, like:
> > > 
> > > 	PROGRAM="scsi_id -d %N -s %p", ...
> > 
> > Ok, I did that. We have %N and %p now. 
> 
> You can pull that into your tree:
>   bk pull bk://vrfy.bkbits.net/udev
> 
> if you are faster than Greg. :)

I've pulled this into my tree and pushed it out now.

thanks,

greg k-h


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Kay Sievers]

On Tue, 2005-02-08 at 12:15 -0800, Greg KH wrote:
> On Tue, Feb 08, 2005 at 08:52:29PM +0100, Kay Sievers wrote:
> > [offlist]
> 
> Hm, looks like it was on the list to me :)

Yeah, bad. But it was harmless, puh. :)

> > if you are faster than Greg. :)
> 
> I've pulled this into my tree and pushed it out now.

Wow, you were really fast. I will try it again sometimes, if it works
to send private mails to the list for speedup :)

Kay 



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Kay Sievers]

On Tue, 2005-02-08 at 12:11 -0800, Patrick Mansfield wrote:
> On Tue, Feb 08, 2005 at 08:50:10PM +0100, Kay Sievers wrote:
> > On Mon, 2005-02-07 at 10:39 -0800, Patrick Mansfield wrote:
> > > So it also needs code changes to use the DEVPATH plus any command line
> > > option, and/or a new % value for the sysfs path, like:
> > > 
> > > 	PROGRAM="scsi_id -d %N -s %p", ...
> > 
> > Ok, I did that. We will have %N and %p now. 
> 
> > > It should still figure out whether or not to print a trailing "\n" for
> > > normal command line usage. DEVPATH set could be the trigger, though that
> > > is not clean (using -s <devpath> for sysfs path, but then using DEVPATH to
> > > figure out if \n is printed).
> > 
> > RESULT does not care about a trailing \n.
> 
> Okay. Then there is no change needed in scsi_id, and that code should be
> removed from it. I'm not sure if scsi_id DEVPATH environment variable
> usage should be removed (for backwards compatibility).

I think that functionality it's not really needed outside of udev.
But you still use it for vpd@page0 validation. You may compare that with
inquiry@page0 instead of using the vendor/model values from sysfs?

> Given this change, can you revert the tmp directory change (if Marco's OK
> with that)? Mainly for non-udev command line usage.

I will do that. Care to give it a try, if it really works with the new
code. Also with the \n RESULT matching.

Thanks,
Kay



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Patrick Mansfield]

On Tue, Feb 08, 2005 at 09:38:46PM +0100, Kay Sievers wrote:
> On Tue, 2005-02-08 at 12:11 -0800, Patrick Mansfield wrote:

> > Okay. Then there is no change needed in scsi_id, and that code should be
> > removed from it. I'm not sure if scsi_id DEVPATH environment variable
> > usage should be removed (for backwards compatibility).
> 
> I think that functionality it's not really needed outside of udev.
> But you still use it for vpd@page0 validation. You may compare that with
> inquiry@page0 instead of using the vendor/model values from sysfs?

I'm not sure what you mean by the above. Can you clarify? Or does the
following answer your question?

The DEVPATH usage and hotplug_mode are mainly for use with udev, the code
is also there for ease of use in any hotplug program (though a script
can easily run scsi_id -s ${DEVPATH}).

There are also some other oddities based on hotplug_mode being set (use
syslog, don't warn on what would be errors for command line usage).

I get the vendor/model via sysfs as there are some quirks related to use
of the INQUIRY and I did not want to duplicate that code or the devinfo
list (black/white list, see drivers/scsi/scsi_scan.c and the scsi_devinfo.c
table; especially the BLIST_INQUIRY_36 and 58). 

In short scsi_inquiry() is always called with evpd set to 1.

> > Given this change, can you revert the tmp directory change (if Marco's OK
> > with that)? Mainly for non-udev command line usage.
> 
> I will do that. Care to give it a try, if it really works with the new
> code. Also with the \n RESULT matching.

OK, I will test it out.

I still haven't tried building and running udev from source on my laptop
(with FC3, better make sure I have a backup boot device).

-- Patrick Mansfield


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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: patches from the debian package

[Kay Sievers]

On Tue, 2005-02-08 at 13:32 -0800, Patrick Mansfield wrote:
> On Tue, Feb 08, 2005 at 09:38:46PM +0100, Kay Sievers wrote:
> > On Tue, 2005-02-08 at 12:11 -0800, Patrick Mansfield wrote:
> 
> > > Okay. Then there is no change needed in scsi_id, and that code should be
> > > removed from it. I'm not sure if scsi_id DEVPATH environment variable
> > > usage should be removed (for backwards compatibility).
> > 
> > I think that functionality it's not really needed outside of udev.
> > But you still use it for vpd@page0 validation. You may compare that with
> > inquiry@page0 instead of using the vendor/model values from sysfs?
> 
> I'm not sure what you mean by the above. Can you clarify? Or does the
> following answer your question?
> 
> The DEVPATH usage and hotplug_mode are mainly for use with udev, the code
> is also there for ease of use in any hotplug program (though a script
> can easily run scsi_id -s ${DEVPATH}).
> 
> There are also some other oddities based on hotplug_mode being set (use
> syslog, don't warn on what would be errors for command line usage).
> 
> I get the vendor/model via sysfs as there are some quirks related to use
> of the INQUIRY and I did not want to duplicate that code or the devinfo
> list (black/white list, see drivers/scsi/scsi_scan.c and the scsi_devinfo.c
> table; especially the BLIST_INQUIRY_36 and 58). 

I was just asking if a compare between the results of page0 vpd=0 and
vpd=1 would be sufficient to make sure that the right page0 is returned.
And we wouldn't need the devpath at all.
I see the reason now. Well, I like broken hardware. :)

Thanks,
Kay




-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id\x14396&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