From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:36227) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eqHKC-0003Km-0W for guix-patches@gnu.org; Mon, 26 Feb 2018 06:52:08 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eqHK6-0003u5-SK for guix-patches@gnu.org; Mon, 26 Feb 2018 06:52:08 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:53162) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1eqHK6-0003tf-PN for guix-patches@gnu.org; Mon, 26 Feb 2018 06:52:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1eqHK6-00075e-Eo for guix-patches@gnu.org; Mon, 26 Feb 2018 06:52:02 -0500 Subject: [bug#30604] [PATCH 3/4] linux-initrd: Add kmod. Resent-Message-ID: References: <20180225114557.816-1-dannym@scratchpost.org> <20180225114816.869-1-dannym@scratchpost.org> <20180225114816.869-3-dannym@scratchpost.org> <87vaelf9ln.fsf@gmail.com> <20180225160750.46bf5b50@scratchpost.org> From: Mathieu Othacehe In-reply-to: <20180225160750.46bf5b50@scratchpost.org> Date: Mon, 26 Feb 2018 12:51:51 +0100 Message-ID: <87h8q4j7e0.fsf@gmail.com> MIME-Version: 1.0 Content-Type: text/plain List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: Danny Milosavljevic Cc: 30604@debbugs.gnu.org Hey Danny, > Because it's an option for specifying the location of "Module.symvers" - and > I don't know whether guix uses it (probably not). If one doesn't specify an > option, depmod will default to the running kernel - which is not what we want. > > I should elaborate in the comment that, if we start using Module.symvers, we > MUST pass "-E" there. Maybe better to even just check for the file existence > and add it right now, otherwise we might forget later. What do you think? Reading 'depmod' manpage, I understand that -E and -F are mutually exclusive. Both options seem to have an interest only if -e is supplied to "reports any symbols which a module needs which are not supplied by other modules or the kernel". So maybe something like "-F (string-append #$linux "/System.map") -e" would make more sense ? Mathieu