From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753614AbbCaHhG (ORCPT ); Tue, 31 Mar 2015 03:37:06 -0400 Received: from shonan.sfc.wide.ad.jp ([203.178.142.130]:56151 "EHLO mail.sfc.wide.ad.jp" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753497AbbCaHg7 (ORCPT ); Tue, 31 Mar 2015 03:36:59 -0400 Date: Tue, 31 Mar 2015 16:36:55 +0900 Message-ID: From: Hajime Tazaki To: richard@nod.at Cc: linux-arch@vger.kernel.org, arnd@arndb.de, corbet@lwn.net, cl@linux.com, penberg@kernel.org, rientjes@google.com, iamjoonsoo.kim@lge.com, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, netdev@vger.kernel.org, linux-mm@kvack.org, jdike@addtoit.com, rusty@rustcorp.com.au, mathieu.lacage@gmail.com, christoph.paasch@gmail.com Subject: Re: [RFC PATCH 08/11] lib: other kernel glue layer code In-Reply-To: <5518419E.8010007@nod.at> References: <1427202642-1716-1-git-send-email-tazaki@sfc.wide.ad.jp> <1427202642-1716-9-git-send-email-tazaki@sfc.wide.ad.jp> <5518419E.8010007@nod.at> User-Agent: Wanderlust/2.15.9 (Almost Unreal) Emacs/22.3 Mule/5.0 (SAKAKI) X-Face: "+?b:_s\r$dbBx'Ur*k#`5|~\>v~i`PzaxANTx_S?J>:mTQrtm:c7'f5b~W2eX~Fl[0Pw,0bow)8r8Z5,D&>]C/'ujqr:fbY>]/52T^Q~cX*y5\?!"i<^Vp=zCNguAeyPH$`ZTv{di25X8,%@! MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Richard, thanks again. At Sun, 29 Mar 2015 20:17:02 +0200, Richard Weinberger wrote: > > Am 24.03.2015 um 14:10 schrieb Hajime Tazaki: > > These files are used to provide the same function calls so that other > > network stack code keeps untouched. > > > > Signed-off-by: Hajime Tazaki > > Signed-off-by: Christoph Paasch > > --- > > arch/lib/cred.c | 16 +++ > > arch/lib/dcache.c | 93 +++++++++++++++ > > arch/lib/filemap.c | 27 +++++ > > arch/lib/fs.c | 287 ++++++++++++++++++++++++++++++++++++++++++++ > > arch/lib/glue.c | 336 ++++++++++++++++++++++++++++++++++++++++++++++++++++ > > arch/lib/inode.c | 146 +++++++++++++++++++++++ > > arch/lib/modules.c | 36 ++++++ > > arch/lib/pid.c | 29 +++++ > > arch/lib/print.c | 56 +++++++++ > > arch/lib/proc.c | 164 +++++++++++++++++++++++++ > > arch/lib/random.c | 53 +++++++++ > > arch/lib/security.c | 45 +++++++ > > arch/lib/seq.c | 122 +++++++++++++++++++ > > arch/lib/splice.c | 20 ++++ > > arch/lib/super.c | 210 ++++++++++++++++++++++++++++++++ > > arch/lib/sysfs.c | 83 +++++++++++++ > > arch/lib/vmscan.c | 26 ++++ > > 17 files changed, 1749 insertions(+) > > BTW: Why do you need these stub implementations at all? > If I read your code correctly it is because you're linking against the whole net/ directory. > Let's take register_filesystem() again as example. net/socket.c references it in sock_init(). > Maybe it would make sense to split socket.c into two files, net/socket.c and net/sockfs.c. > Such that you could link only net/socket.o. > Of course you'd have to convince networking folks first. :D this is something that we've been avoided: we weren't able to estimate how much such splits will expand. > By linking selectively objects files from net/ you could get rid of a lot unneeded stubs. now I'm trying to minimize those stubs by reusing the original codes (i.e., fs/filesystems.c for register_filesystem()). I will let you know once I've done (maybe v2 RFC if you think it's better ?) -- Hajime