udev: move notify rule
PATCH 1/1 udev: move notify rule
Move notify rule into the first rules file. The RUN commands is executed after everything is finished, so it makes no difference if it is set early or late.
udev/10-dm.rules.in +6 -0
@@ -26,6 +26,12 @@ TEST!="$env{DM_SBIN_PATH}/dmsetup", GOTO="dm_end" # kernels >= 2.6.31 only. ENV{DM_COOKIE}=="?*", IMPORT{program}="$env{DM_SBIN_PATH}/dmsetup udevflags $env{DM_COOKIE}" +# These rules are responsible for sending a notification to a process +# waiting for completion of udev rules. The process is identified by +# a cookie value sent within "change" and "remove" events (the cookie +# value is set before by that process for every action requested). +ACTION=="change|remove", ENV{DM_COOKIE}=="?*", RUN+="$env{DM_SBIN_PATH}/dmsetup udevcomplete $env{DM_COOKIE}" + # Normally, we would test for DM_UDEV_DISABLE_DM_RULES_FLAG here and skip # the rules if set. However, we need to set DM_* environment variables # for now to properly filter out inappropriate events. This dependency
udev/95-dm-notify.rules deleted +0 -18
@@ -1,18 +0,0 @@ -# Copyright (C) 2009 Red Hat, Inc. All rights reserved. -# -# This file is part of LVM2. - -# Udev rules for device-mapper devices. -# -# These rules are responsible for sending a notification to a process -# waiting for completion of udev rules. The process is identified by -# a cookie value sent within "change" and "remove" events (the cookie -# value is set before by that process for every action requested). - -SUBSYSTEM!="block", GOTO="dm_end" -KERNEL!="dm-[0-9]*", GOTO="dm_end" -ACTION!="change|remove", GOTO="dm_end" - -ENV{DM_COOKIE}=="?*", RUN+="$env{DM_SBIN_PATH}/dmsetup udevcomplete $env{DM_COOKIE}" - -LABEL="dm_end"
udev/Makefile.in +1 -2
@@ -16,7 +16,7 @@ top_srcdir = @top_srcdir@ top_builddir = @top_builddir@ VPATH = @srcdir@ -DM_RULES=10-dm.rules 13-dm-disk.rules 95-dm-notify.rules +DM_RULES=10-dm.rules 13-dm-disk.rules LVM_RULES=11-dm-lvm.rules DM_DIR=$(shell grep "\#define DM_DIR" $(top_srcdir)/libdm/misc/dm-ioctl.h | awk '{print $$3}') --
One does not thank logic.
-- Sarek, "Journey to Babel", stardate 3842.4
PATCH 1/1 udev: move notify rule
Hi,
Quoted 4 lines · Bastian Blank
Move notify rule into the first rules file. The RUN commands is executed after everything is finished, so it makes no difference if it is set early or late.
yes, this is true with respect to all the other udev processing. But the point of having it as 95-dm.rules is that there could be some other RUN+="..." line in the other rules. These are processed in sequence of their order of appearance then and we really would like to send the notification after all this processing is done (considering those "RUNs" as well)...
Peter
PATCH 1/1 udev: move notify rule
Quoted 6 lines · Peter Rajnoha
Quoted 3 lines · Bastian Blank
Move notify rule into the first rules file. The RUN commands is executed after everything is finished, so it makes no difference if it is set early or late.
yes, this is true with respect to all the other udev processing. But the point of having it as 95-dm.rules is that there could be some other RUN+="..." line in the other rules.
And there are no constraints about the order of this executions. If you know more then the udev docu, fix it there.
Bastian
Signature · 2 lines
Where there's no emotion, there's no motive for violence. -- Spock, "Dagger of the Mind", stardate 2715.1
PATCH 1/1 udev: move notify rule
Quoted 10 lines · Bastian Blank
Quoted 6 lines · Peter Rajnoha
Quoted 3 lines · Bastian Blank
Move notify rule into the first rules file. The RUN commands is executed after everything is finished, so it makes no difference if it is set early or late.
yes, this is true with respect to all the other udev processing. But the point of having it as 95-dm.rules is that there could be some other RUN+="..." line in the other rules.
And there are no constraints about the order of this executions. If you know more then the udev docu, fix it there.
OK, maybe I should assure myself with udev guys so we're 100% sure (it's not so essential now, but it could be in the future - there are more plans for udev usage with respect to LVM, so I need to be sure about this ordering, that's true...).
But this is the behaviour I observed when I tested this. But you're right, I'll make sure...
Peter