public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Arnd Bergmann <arnd@arndb.de>
To: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Paul Mackerras <paulus@samba.org>,
	linuxppc64-dev@ozlabs.org, linux-kernel@vger.kernel.org,
	arjan@infradead.org, viro@ftp.linux.org.uk
Subject: Re: [PATCH 02/14] spufs: fix local store page refcounting
Date: Tue, 6 Dec 2005 19:49:30 +0100	[thread overview]
Message-ID: <200512061949.33482.arnd@arndb.de> (raw)
In-Reply-To: <1133869108.7968.1.camel@localhost>

On Dinsdag 06 Dezember 2005 12:38, Pekka Enberg wrote:
> On Dinsdag 06 Dezember 2005 01:51, Paul Mackerras wrote:
> > > Remind me again why spufs is under arch/powerpc/ rather than fs/ ?
> 
> On Tue, 2005-12-06 at 11:18 +0100, Arnd Bergmann wrote:
> > We had a discussion about this in August, after the patch
> > at http://patchwork.ozlabs.org/linuxppc64/patch?id=2140
> > 
> > Nobody had voiced any objections against the arch/powerpc location,
> > and Pekka had good reasons against fs/, so I changed it.
> 
> It had arch specific hooks which IMHO do not belong into fs/.

Since the discussion came up again in irc, I looked up the existing file
systems.

outside of fs/, we have the following file systems.

find -name \*.c | grep -v ^./fs | xargs grep struct.file_system_type.*=
./arch/ia64/kernel/perfmon.c:static struct file_system_type pfm_fs_type = {
./drivers/infiniband/core/uverbs_main.c:static struct file_system_type uverbs_event_fs = {
./drivers/isdn/capi/capifs.c:static struct file_system_type capifs_fs_type = {
./drivers/misc/ibmasm/ibmasmfs.c:static struct file_system_type ibmasmfs_type = {
./drivers/oprofile/oprofilefs.c:static struct file_system_type oprofilefs_type = {
./drivers/usb/core/inode.c:static struct file_system_type usb_fs_type = {
./drivers/usb/gadget/inode.c:static struct file_system_type gadgetfs_type = {
./ipc/mqueue.c:static struct file_system_type mqueue_fs_type = {
./kernel/cpuset.c:static struct file_system_type cpuset_fs_type = {
./kernel/futex.c:static struct file_system_type futex_fs_type = {
./mm/shmem.c:static struct file_system_type tmpfs_fs_type = {
./mm/tiny-shmem.c:static struct file_system_type tmpfs_fs_type = {
./net/socket.c:static struct file_system_type sock_fs_type = {
./net/sunrpc/rpc_pipe.c:static struct file_system_type rpc_pipe_fs_type = {
./security/inode.c:static struct file_system_type fs_type = {
./security/selinux/selinuxfs.c:static struct file_system_type sel_fs_type = {

In fs/, most code deals with actual files stored on a disk or similar,
with the exception of:

./fs/binfmt_misc.c:static struct file_system_type bm_fs_type = {
./fs/block_dev.c:static struct file_system_type bd_type = {
./fs/debugfs/inode.c:static struct file_system_type debug_fs_type = {
./fs/devfs/base.c:static struct file_system_type devfs_fs_type = {
./fs/devpts/inode.c:static struct file_system_type devpts_fs_type = {
./fs/eventpoll.c:static struct file_system_type eventpoll_fs_type = {
./fs/hugetlbfs/inode.c:static struct file_system_type hugetlbfs_fs_type = {
./fs/inotify.c:static struct file_system_type inotify_fs_type = {
./fs/openpromfs/inode.c:static struct file_system_type openprom_fs_type = {
./fs/pipe.c:static struct file_system_type pipe_fs_type = {
./fs/proc/root.c:static struct file_system_type proc_fs_type = {
./fs/relayfs/inode.c:static struct file_system_type relayfs_fs_type = {
./fs/sysfs/mount.c:static struct file_system_type sysfs_fs_type = {

I guess there is no strict rule where these file systems go to, e.g.
hugetlbs could just as well live near mm/shmem.c or any of those outside
of fs/ could be moved in there.

I don't really care where I put spufs, but I would prefer to move
the files only one more time at most.
Initially, they were in fs/spufs, and I moved them to
arch/powerpc/platforms/cell/spufs at Pekkas suggestion.

	Arnd <><

       reply	other threads:[~2005-12-06 18:47 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20051206035220.097737000@localhost>
     [not found] ` <200512061118.19633.arnd@arndb.de>
     [not found]   ` <1133869108.7968.1.camel@localhost>
2005-12-06 18:49     ` Arnd Bergmann [this message]
2005-12-06 19:05       ` [PATCH 02/14] spufs: fix local store page refcounting Pekka Enberg
2005-12-06 21:10         ` Paul Mackerras
2005-12-06 21:41           ` Pekka Enberg
2005-12-06 22:19             ` Paul Mackerras
2005-12-06 22:27               ` Arnd Bergmann
2005-12-07  2:26               ` Al Viro
2005-12-07  3:15                 ` Paul Mackerras
2005-12-07  8:21                   ` Pekka Enberg
2005-12-07 10:17                   ` Al Viro
2005-12-06 22:14           ` Nathan Lynch

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200512061949.33482.arnd@arndb.de \
    --to=arnd@arndb.de \
    --cc=arjan@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc64-dev@ozlabs.org \
    --cc=paulus@samba.org \
    --cc=penberg@cs.helsinki.fi \
    --cc=viro@ftp.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox