Your message dated Sun, 28 Apr 2013 21:50:22 +0100 with message-id <1367182222.29152.13.camel@jacala.jungle.funky-badger.org> and subject line Re: Bug#705650: unblock: kmod/9-3 has caused the Debian Bug report #705650, regarding unblock: kmod/9-3 to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what this message is talking about, this may indicate a serious mail system misconfiguration somewhere. Please contact owner@bugs.debian.org immediately.) -- 705650: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=705650 Debian Bug Tracking System Contact owner@bugs.debian.org with problems
--- Begin Message ---
- To: Debian Bug Tracking System <submit@bugs.debian.org>
- Subject: unblock: kmod/9-3
- From: Marco d'Itri <md@linux.it>
- Date: Thu, 18 Apr 2013 03:19:18 +0200
- Message-id: <[🔎] 20130418011918.GA19182@bongo.bofh.it>
Package: release.debian.org Severity: normal User: release.debian.org@packages.debian.org Usertags: unblock Please unblock package kmod The first two bugs are important enough that they have to be fixed in a stable release. kmod (9-3) unstable; urgency=low * Added backported patch dot_kcmdline: correctly parse kcmdline parameters containing a dot. (Closes: #689872) * Added backported patch bad_alias_assertion: stop modprobe from aborting from an assertion because of some invalid but common configuration directives. (Closes: #674110) * Removed the nfs4 alias from aliases.conf as requested by the kernel team. (Closes: #683972) * Removed the unnecessary build-dependency on perl. (Closes: #697750) -- Marco d'Itri <md@linux.it> Sun, 07 Apr 2013 18:19:01 +0200 diff -Nru kmod-9/debian/changelog kmod-9/debian/changelog --- kmod-9/debian/changelog 2012-08-27 00:40:32.000000000 +0200 +++ kmod-9/debian/changelog 2013-04-07 18:35:22.000000000 +0200 @@ -1,3 +1,16 @@ +kmod (9-3) unstable; urgency=low + + * Added backported patch dot_kcmdline: correctly parse kcmdline parameters + containing a dot. (Closes: #689872) + * Added backported patch bad_alias_assertion: stop modprobe from aborting + from an assertion because of some invalid but common configuration + directives. (Closes: #674110) + * Removed the nfs4 alias from aliases.conf as requested by the kernel + team. (Closes: #683972) + * Removed the unnecessary build-dependency on perl. (Closes: #697750) + + -- Marco d'Itri <md@linux.it> Sun, 07 Apr 2013 18:19:01 +0200 + kmod (9-2) unstable; urgency=low * Backported upstream fixes to make modprobe work like in module-init-tools diff -Nru kmod-9/debian/control kmod-9/debian/control --- kmod-9/debian/control 2012-06-25 03:03:52.000000000 +0200 +++ kmod-9/debian/control 2013-04-07 18:35:36.000000000 +0200 @@ -2,7 +2,7 @@ Section: admin Priority: important Maintainer: Marco d'Itri <md@linux.it> -Build-Depends: debhelper (>> 8.1.3), perl +Build-Depends: debhelper (>> 8.1.3) Standards-Version: 3.9.3.1 Package: kmod diff -Nru kmod-9/debian/patches/aliases_conf kmod-9/debian/patches/aliases_conf --- kmod-9/debian/patches/aliases_conf 2011-12-17 00:04:04.000000000 +0100 +++ kmod-9/debian/patches/aliases_conf 2013-04-07 18:36:39.000000000 +0200 @@ -160,8 +160,6 @@ + +alias usbdevfs usbcore + -+alias nfs4 nfs -+ +# work around other kernel issues ############################################ +# The EHCI driver should be loaded before the ones for low speed controllers +# or some devices may be confused when they are disconnected and reconnected. diff -Nru kmod-9/debian/patches/bad_alias_assertion kmod-9/debian/patches/bad_alias_assertion --- kmod-9/debian/patches/bad_alias_assertion 1970-01-01 01:00:00.000000000 +0100 +++ kmod-9/debian/patches/bad_alias_assertion 2013-04-07 18:28:51.000000000 +0200 @@ -0,0 +1,40 @@ +Description: fix assertion on --show-depends with bogus config file + Some common but invalid modprobe.d configuration directives caused + modprobe --show-depends (which is used by update-initramfs) to crash. +Origin: upstream, commit:ace742fa9aee1aec5931d5ee8a51fa9b8f0d94e0 +Applied-Upstream: commit:ace742fa9aee1aec5931d5ee8a51fa9b8f0d94e0 +Bug-Debian: http://bugs.debian.org/674110 +--- + +commit ace742fa9aee1aec5931d5ee8a51fa9b8f0d94e0 +Author: Lucas De Marchi <lucas.de.marchi@gmail.com> +Date: Thu Mar 21 02:33:25 2013 -0300 + + modprobe: Fix assertion on --show-depends with bogus config file + + Putting something like "alias psmouse deadbeef" is a hackish way to + blacklist a module. While I don't encourage doing so, let's not explode + if we fiund such config files. + + A small difference from the behavior of module-init-tools: we exit with + 0 instead of 1. + +diff --git a/tools/modprobe.c b/tools/modprobe.c +index 64674b0..1b8c96e 100644 +--- a/tools/modprobe.c ++++ b/tools/modprobe.c +@@ -495,8 +495,12 @@ static void print_action(struct kmod_module *m, bool install, + path = kmod_module_get_path(m); + + if (path == NULL) { +- assert(kmod_module_get_initstate(m) == KMOD_MODULE_BUILTIN); +- printf("builtin %s\n", kmod_module_get_name(m)); ++ /* ++ * Either a builtin module, or an alias, print only for ++ * builtin ++ */ ++ if (kmod_module_get_initstate(m) == KMOD_MODULE_BUILTIN) ++ printf("builtin %s\n", kmod_module_get_name(m)); + } else + printf("insmod %s %s\n", kmod_module_get_path(m), options); + } diff -Nru kmod-9/debian/patches/dot_kcmdline kmod-9/debian/patches/dot_kcmdline --- kmod-9/debian/patches/dot_kcmdline 1970-01-01 01:00:00.000000000 +0100 +++ kmod-9/debian/patches/dot_kcmdline 2013-04-07 18:27:41.000000000 +0200 @@ -0,0 +1,36 @@ +Description: correctly parse kcmdline parameters containing a dot + Kernel command line parameters of modules were not parsed correctly if + the value contained a dot. +Origin: upstream, commit:66f3228d17d66d7e2dd484427259290fbc82b2f0 +Applied-Upstream: commit:66f3228d17d66d7e2dd484427259290fbc82b2f0 +Bug-Debian: http://bugs.debian.org/689872 +--- + +commit 66f3228d17d66d7e2dd484427259290fbc82b2f0 +Author: Lucas De Marchi <lucas.demarchi@profusion.mobi> +Date: Mon Oct 8 19:04:16 2012 -0300 + + libkmod: Add support for '.' in module parameter on kcmdline + + Otherwise we fail to parse arguments in kernel command line like + testmodule.testparam=1.5G + + Suggested-by: Selim T. Erdogan <selim@alumni.cs.utexas.edu> + +diff --git a/libkmod/libkmod-config.c b/libkmod/libkmod-config.c +index 70044f0..398468e 100644 +--- a/libkmod/libkmod-config.c ++++ b/libkmod/libkmod-config.c +@@ -567,8 +567,10 @@ static int kmod_config_parse_kcmdline(struct kmod_config *config) + modname = p + 1; + break; + case '.': +- *p = '\0'; +- param = p + 1; ++ if (param == NULL) { ++ *p = '\0'; ++ param = p + 1; ++ } + break; + case '=': + if (param != NULL) diff -Nru kmod-9/debian/patches/series kmod-9/debian/patches/series --- kmod-9/debian/patches/series 2012-08-27 03:38:47.000000000 +0200 +++ kmod-9/debian/patches/series 2013-04-07 18:21:17.000000000 +0200 @@ -1,3 +1,5 @@ +dot_kcmdline +bad_alias_assertion blacklist_aliased # fixes to be pushed upstream unblock kmod/9-3 -- ciao, MarcoAttachment: signature.asc
Description: Digital signature
--- End Message ---
--- Begin Message ---
- To: Cyril Brulebois <kibi@debian.org>
- Cc: Marco d'Itri <md@linux.it>, 705650-done@bugs.debian.org
- Subject: Re: Bug#705650: unblock: kmod/9-3
- From: "Adam D. Barratt" <adam@adam-barratt.org.uk>
- Date: Sun, 28 Apr 2013 21:50:22 +0100
- Message-id: <1367182222.29152.13.camel@jacala.jungle.funky-badger.org>
- In-reply-to: <[🔎] 20130428204633.GG4613@mraw.org>
- References: <[🔎] 20130418011918.GA19182@bongo.bofh.it> <[🔎] 1366463102.23177.11.camel@jacala.jungle.funky-badger.org> <[🔎] 20130428204633.GG4613@mraw.org>
On Sun, 2013-04-28 at 22:46 +0200, Cyril Brulebois wrote: > Adam D. Barratt <adam@adam-barratt.org.uk> (20/04/2013): > > On Thu, 2013-04-18 at 03:19 +0200, Marco d'Itri wrote: > > > Please unblock package kmod > > > > > > The first two bugs are important enough that they have to be fixed > > > in a stable release. > > > > It is very late to be changing such things though. :-( > > > > In any case, it'll need a d-i ack. > > While it's very late indeed, I guess merging it now makes sense anyway > (better find possible regressions now than introduce them into r1). Okay then; unblock{,-udeb}bed. Regards, Adam
--- End Message ---