Skip to contentSkip to conversation

Move udev cache creation after checkroot

RFC PATCH In review Koen Kooi 4 participants 7 messages Jan 12, 2010 – Jan 23, 2010

lore
Share & tools
lore

Koen Kooi

author
#0 lore raw

RFC PATCH 1/1 Move udev cache creation after checkroot

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

regards,

Koen
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFLTH8oMkyGM64RGpERAqk0AJ4/5rUWj/Ciyfy0lfcDnM1SgqUN9gCeOr9f QDA6nnADhKgOt6LEQZK+AUA=
=EtI0
-----END PGP SIGNATURE-----

From 1a77c8e64b2ba25e8f08b6b577c59312ed87a9e6 Mon Sep 17 00:00:00 2001
From: Koen Kooi <[email protected]>
Date: Tue, 12 Jan 2010 14:48:59 +0100
Subject: [PATCH] udev 141: move creation of /etc/dev.tar to after checkroot

* / isn't always mounted rw when udev runs, so lets create /etc/dev.tar after checkroot has run

---
 recipes/udev/udev-141/cache |   18 ++++++++++++++++++
 recipes/udev/udev-141/init  |    8 --------
 recipes/udev/udev_141.bb    |   10 ++++++++--
 3 files changed, 26 insertions(+), 10 deletions(-)
 create mode 100644 recipes/udev/udev-141/cache
recipes/udev/udev-141/cache new file +18 -0
new file mode 100644
index 0000000..4db1466
@@ -0,0 +1,18 @@
+#!/bin/sh -e
+
+export TZ=/etc/localtime
+
+[ -f /etc/default/udev ] && . /etc/default/udev
+
+echo "Caching udev devnodes"
+
+			if [ "$DEVCACHE" != "" ]; then
+				echo -n "Populating dev cache"
+				(cd /; tar cf $DEVCACHE dev)
+				mv /tmp/uname /etc/udev/saved.uname
+				mv /tmp/cmdline /etc/udev/saved.cmdline
+				mv /tmp/atags /etc/udev/saved.atags
+				echo
+			fi
+
+exit 0
recipes/udev/udev-141/init +0 -8
index a0bdf40..24ca9d8 100644
@@ -68,14 +68,6 @@ kill_udevd > "/dev/null" 2>&1
 		else
 			/sbin/udevadm trigger
 			/sbin/udevadm settle
-			if [ "$DEVCACHE" != "" ]; then
-				echo -n "Populating dev cache"
-				(cd /; tar cf $DEVCACHE dev)
-				mv /tmp/uname /etc/udev/saved.uname
-				mv /tmp/cmdline /etc/udev/saved.cmdline
-				mv /tmp/atags /etc/udev/saved.atags
-				echo
-			fi
 		fi
 
 exit 0
recipes/udev/udev_141.bb +8 -2
index 00ba971..df5e346 100644
@@ -6,7 +6,7 @@ LICENSE = "GPL"
 # Untested
 DEFAULT_PREFERENCE = "-1"
 
-PR = "r19"
+PR = "r20"
 
 # needed for init.d script
 RDEPENDS_${PN} += "udev-utils"
@@ -24,7 +24,9 @@ SRC_URI += " \
        file://network.sh \
        file://local.rules \
        file://default \
-       file://init"
+       file://init \
+       file://cache \
+"
 
 SRC_URI_append_h2200 = " file://50-hostap_cs.rules "
 PACKAGE_ARCH_h2200 = "h2200"
@@ -70,6 +72,7 @@ do_install () {
 	oe_runmake 'DESTDIR=${D}' INSTALL=install install
 	install -d ${D}${sysconfdir}/init.d
 	install -m 0755 ${WORKDIR}/init ${D}${sysconfdir}/init.d/udev
+	install -m 0755 ${WORKDIR}/cache ${D}${sysconfdir}/init.d/udev-cache
 
 	install -d ${D}${sysconfdir}/default
 	install -m 0755 ${WORKDIR}/default ${D}${sysconfdir}/default/udev
@@ -112,7 +115,10 @@ do_install_append_bug() {
 	install -m 0644 ${WORKDIR}/bmi_eventpipe.sh ${D}${sysconfdir}/udev/scripts/bmi_eventpipe.sh
 }
 
+# Create the cache after checkroot has run
 pkg_postinst_${PN}_append() {
+update-rc.d $OPT udev-cache start 12 S .
+
 if [ -d $D/lib/udev/rules.d ] ; then
 	echo "$D/lib/udev/rules.d is not a symlink, fixing that"
 	mv $D/lib/udev/rules.d/* $D${sysconfdir}/udev/rules.d/
Signature · 1 line
1.6.5
Acked-by on v1 1/1
#1 lore raw
Quoted 12 lines · Koen Kooi

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

That patch seems to match what I have been saying but not had time to code.

author
#2 lore raw
In reply to Koen Kooi

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

ping

Quoted 6 lines · Koen Kooi

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Quoted 6 lines

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

regards,

Koen

_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFLTZerMkyGM64RGpERAg9CAKCNetCgK7Xs4b2UFLsn3pOoHRrSpgCgulfl H2qZxaChQx0kSuc/8Jguhmg=
=4lQ5
-----END PGP SIGNATURE-----

Acked-by on v1 1/1
#3 lore raw
In reply to Koen Kooi

Dnia wtorek, 12 stycznia 2010 o 14:54:48 Koen Kooi napisał(a):

Quoted 5 lines

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Regards,

Signature · 3 lines
JID:      [email protected]
Website:  http://marcin.juszkiewicz.com.pl/
LinkedIn: http://www.linkedin.com/in/marcinjuszkiewicz
author
#4 lore raw
In reply to Koen Kooi

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

It received 2 Acks, no NAKs, works well in the field: it's applied now

Quoted 6 lines · Koen Kooi

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Quoted 6 lines

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

regards,

Koen

_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFLTzN0MkyGM64RGpERAj0SAJ9zMcapZJRhN5WqngZ5Xqse9q+jJwCeMTR8 UUtNtGO21HZL99oyBYKQ8ug=
=LmQG
-----END PGP SIGNATURE-----

#5 lore raw
Quoted 18 lines · Koen Kooi

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

It received 2 Acks, no NAKs, works well in the field: it's applied now

Quoted 12 lines · Koen Kooi

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

regards,

Koen

I tested it today(sorry for testing late)
And boot still hang with udev + ro
maybe it fixes a different issue then

Denis.

author
#6 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 16 lines · gnutoo
Quoted 15 lines · Koen Kooi

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

It received 2 Acks, no NAKs, works well in the field: it's applied now

Quoted 9 lines · Koen Kooi

Hi,

What's your opinion on attached (untested) patch? If there's some consensus that this is a step in the right direction I'll create a working & tested patch to commit.

Nota bene: I'm not going to discuss the merits of udev caching, only the impact of this patch to the current situation.

Quoted 6 lines
Quoted 3 lines
Quoted 3 lines

regards,

Koen

I tested it today(sorry for testing late)
And boot still hang with udev + ro
maybe it fixes a different issue then

This wasn't about hangs, this was about trying to write to a RO fs :)

regards,

Koen
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFLWz6GMkyGM64RGpERApVIAJ9GAr80qKR5ZlXfmOTK7byZEcxDiwCfYDi1 MIGa4OyxBXh5XZ9eXfLvNnQ=
=kIp/
-----END PGP SIGNATURE-----