From: Josh Triplett <josh@joshtriplett.org>
To: Pieter Smith <pieter@boesman.nl>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
open list <linux-kernel@vger.kernel.org>,
Arnd Bergmann <arnd@arndb.de>
Subject: Re: [PATCH 51/56] drivers/char/mem: support compiling out splice
Date: Sun, 16 Nov 2014 10:20:07 -0800 [thread overview]
Message-ID: <20141116182007.GA5032@thin> (raw)
In-Reply-To: <CAPho-_LCj7bxJ2EvuZtBZXD1buH1V+nKEiobTJUATYmGYVRWcA@mail.gmail.com>
[Please don't top-post.]
On Sun, Nov 16, 2014 at 01:18:27PM +0100, Pieter Smith wrote:
> I had a look at the numbers (see below): For a tinyconfig kernel, savings
> outside fs/splice.c are negligible. With the addition of networking
> however, net/core and net/ipv4 can be squeezed for usable savings.
>
> Should I leave networking (and the rest) until we can come up with a
> non-icky approach outside fs/splice.c?
>
> *THE NUMBERS*
>
> Given a tinyconfig, fs/splice.c can free up 8.2K (3.5K + 4.7K) with
> CONFIG_SYSCALL_SPLICE. The rest are not worth the effort:
> fs/splice:
> syscalls only: add/remove: 0/16 grow/shrink: 2/5 up/down: 114/-3693
> (-3579)
> the remainder: add/remove: 0/24 grow/shrink: 0/4 up/down: 0/-4824
> (-4824)
Nice!
Go ahead and submit the patches for that portion, and the rest can wait
until we get compiler support for omitting structure fields.
- Josh Triplett
next prev parent reply other threads:[~2014-11-16 18:20 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1415913813-362-1-git-send-email-pieter@boesman.nl>
2014-11-13 21:22 ` [PATCH 01/56] moved sendfile syscall to splice translation unit Pieter Smith
2014-11-13 21:22 ` [PATCH 02/56] moved kernel_write out of " Pieter Smith
2014-11-13 21:22 ` [PATCH 03/56] fs: Support compiling out splice-family syscalls Pieter Smith
2014-11-13 21:22 ` [PATCH 04/56] fs: Macros to define splice file_operations Pieter Smith
2014-11-13 21:49 ` Richard Weinberger
2014-11-13 22:24 ` josh
2014-11-13 21:51 ` Al Viro
2014-11-13 21:22 ` [PATCH 05/56] fs/lustre: support compiling out splice Pieter Smith
2014-11-13 22:09 ` Greg Kroah-Hartman
2014-11-13 21:22 ` [PATCH 06/56] fs/adfs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 07/56] fs/affs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 08/56] fs/afs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 09/56] fs/bad_inode: " Pieter Smith
2014-11-13 21:22 ` [PATCH 10/56] fs/block_dev: " Pieter Smith
2014-11-13 21:22 ` [PATCH 11/56] fs/bfs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 12/56] fs/btrfs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 13/56] fs/ceph: " Pieter Smith
2014-11-13 21:22 ` [PATCH 14/56] fs/cifs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 15/56] fs/coda: " Pieter Smith
2014-11-13 21:22 ` [PATCH 16/56] fs/encryptfs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 17/56] fs/exofs: " Pieter Smith
2014-11-16 9:02 ` [osd-dev] " Boaz Harrosh
2014-11-13 21:22 ` [PATCH 18/56] fs/ext2: " Pieter Smith
2014-11-13 21:22 ` [PATCH 19/56] fs/ext3: " Pieter Smith
2014-11-13 21:22 ` [PATCH 20/56] fs/ext4: " Pieter Smith
2014-11-14 0:28 ` josh
2014-11-13 21:22 ` [PATCH 21/56] fs/f2fs: " Pieter Smith
2014-11-13 21:22 ` [PATCH 22/56] fs/fat: " Pieter Smith
2014-11-13 21:23 ` [PATCH 23/56] fs/fuse: " Pieter Smith
2014-11-13 21:23 ` [PATCH 24/56] fs/gfs2: " Pieter Smith
2014-11-13 21:23 ` [PATCH 25/56] fs/hfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 26/56] fs/hfsplus: " Pieter Smith
2014-11-13 21:23 ` [PATCH 27/56] fs/hostfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 28/56] fs/hpfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 29/56] fs/jffs2: " Pieter Smith
2014-11-13 21:23 ` [PATCH 30/56] fs/jfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 31/56] fs/minix: " Pieter Smith
2014-11-13 21:23 ` [PATCH 32/56] fs/nfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 33/56] fs/nfsd: " Pieter Smith
2014-11-13 21:23 ` [PATCH 34/56] fs/nilfs2: " Pieter Smith
2014-11-13 21:23 ` [PATCH 35/56] fs/ntfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 36/56] fs/ocfs2: " Pieter Smith
2014-11-13 21:23 ` [PATCH 37/56] fs/omfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 38/56] fs/ramfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 39/56] fs/reiserfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 40/56] fs/romfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 41/56] fs/sysv: " Pieter Smith
2014-11-13 21:23 ` [PATCH 42/56] fs/ubifs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 43/56] fs/udf: " Pieter Smith
2014-11-13 21:23 ` [PATCH 44/56] fs/ufs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 45/56] fs/xfs: " Pieter Smith
2014-11-13 21:23 ` [PATCH 46/56] kernel/relay: " Pieter Smith
2014-11-13 21:23 ` [PATCH 47/56] kenel/trace: " Pieter Smith
2014-11-17 20:33 ` Steven Rostedt
2014-11-13 21:23 ` [PATCH 48/56] mm/shmem: " Pieter Smith
2014-11-13 21:23 ` [PATCH 49/56] net/socket: " Pieter Smith
2014-11-13 21:23 ` [PATCH 50/56] fs/read_write: " Pieter Smith
2014-11-13 21:23 ` [PATCH 51/56] drivers/char/mem: " Pieter Smith
2014-11-13 22:09 ` Greg Kroah-Hartman
[not found] ` <CAPho-_JJGy0cwBVfWKL1Gt9ZQZM+Odo7W05bKQ2JLO+TM-ABJA@mail.gmail.com>
2014-11-13 22:31 ` josh
2014-11-13 23:34 ` Greg Kroah-Hartman
2014-11-14 0:19 ` josh
2014-11-14 3:27 ` Greg Kroah-Hartman
[not found] ` <CAPho-_KUc-+c=X6xtLTD2F-o4qi+YpYdnw6F1cx4kniezfs7aw@mail.gmail.com>
2014-11-14 23:25 ` josh
[not found] ` <CAPho-_LCj7bxJ2EvuZtBZXD1buH1V+nKEiobTJUATYmGYVRWcA@mail.gmail.com>
2014-11-16 18:20 ` Josh Triplett [this message]
[not found] ` <CAPho-_+ZXHYGB9-d19NR9u5tcOpN1Ytg80NZgvScF=YQ-6SdNA@mail.gmail.com>
2014-11-18 22:42 ` josh
2014-11-13 21:23 ` [PATCH 52/56] drivers/char/virtio: " Pieter Smith
2014-11-13 22:09 ` Greg Kroah-Hartman
2014-11-13 21:23 ` [PATCH 53/56] net/ipv6: " Pieter Smith
2014-11-13 21:23 ` [PATCH 54/56] net/ipv4: " Pieter Smith
2014-11-13 21:23 ` [PATCH 55/56] net/core: " Pieter Smith
2014-11-13 21:23 ` [PATCH 56/56] fs/splice: full support for " Pieter Smith
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=20141116182007.GA5032@thin \
--to=josh@joshtriplett.org \
--cc=arnd@arndb.de \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pieter@boesman.nl \
/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