Linux-Fsdevel Archive on lore.kernel.org
 help / color / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Nate Karstens <nate.karstens@garmin.com>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Jeff Layton <jlayton@kernel.org>,
	"J. Bruce Fields" <bfields@fieldses.org>,
	Arnd Bergmann <arnd@arndb.de>,
	Richard Henderson <rth@twiddle.net>,
	Ivan Kokshaysky <ink@jurassic.park.msu.ru>,
	Matt Turner <mattst88@gmail.com>,
	"James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>,
	Helge Deller <deller@gmx.de>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	linux-fsdevel@vger.kernel.org, linux-arch@vger.kernel.org,
	linux-alpha@vger.kernel.org, linux-parisc@vger.kernel.org,
	sparclinux@vger.kernel.org, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org
Cc: Changli Gao <xiaosuo@gmail.com>
Subject: Re: [PATCH 1/4] fs: Implement close-on-fork
Date: Mon, 20 Apr 2020 03:25:49 -0700
Message-ID: <36dce9b4-a0bf-0015-f6bc-1006938545b1@gmail.com> (raw)
In-Reply-To: <20200420071548.62112-2-nate.karstens@garmin.com>



On 4/20/20 12:15 AM, Nate Karstens wrote:
> The close-on-fork flag causes the file descriptor to be closed
> atomically in the child process before the child process returns
> from fork(). Implement this feature and provide a method to
> get/set the close-on-fork flag using fcntl(2).
> 
> This functionality was approved by the Austin Common Standards
> Revision Group for inclusion in the next revision of the POSIX
> standard (see issue 1318 in the Austin Group Defect Tracker).

Oh well... yet another feature slowing down a critical path.

> 
> Co-developed-by: Changli Gao <xiaosuo@gmail.com>
> Signed-off-by: Changli Gao <xiaosuo@gmail.com>
> Signed-off-by: Nate Karstens <nate.karstens@garmin.com>
> ---
>  fs/fcntl.c                             |  2 ++
>  fs/file.c                              | 50 +++++++++++++++++++++++++-
>  include/linux/fdtable.h                |  7 ++++
>  include/linux/file.h                   |  2 ++
>  include/uapi/asm-generic/fcntl.h       |  5 +--
>  tools/include/uapi/asm-generic/fcntl.h |  5 +--
>  6 files changed, 66 insertions(+), 5 deletions(-)
> 
> diff --git a/fs/fcntl.c b/fs/fcntl.c
> index 2e4c0fa2074b..23964abf4a1a 100644
> --- a/fs/fcntl.c
> +++ b/fs/fcntl.c
> @@ -335,10 +335,12 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
>  		break;
>  	case F_GETFD:
>  		err = get_close_on_exec(fd) ? FD_CLOEXEC : 0;
> +		err |= get_close_on_fork(fd) ? FD_CLOFORK : 0;
>  		break;
>  	case F_SETFD:
>  		err = 0;
>  		set_close_on_exec(fd, arg & FD_CLOEXEC);
> +		set_close_on_fork(fd, arg & FD_CLOFORK);
>  		break;
>  	case F_GETFL:
>  		err = filp->f_flags;
> diff --git a/fs/file.c b/fs/file.c
> index c8a4e4c86e55..de7260ba718d 100644
> --- a/fs/file.c
> +++ b/fs/file.c
> @@ -57,6 +57,8 @@ static void copy_fd_bitmaps(struct fdtable *nfdt, struct fdtable *ofdt,
>  	memset((char *)nfdt->open_fds + cpy, 0, set);
>  	memcpy(nfdt->close_on_exec, ofdt->close_on_exec, cpy);
>  	memset((char *)nfdt->close_on_exec + cpy, 0, set);
> +	memcpy(nfdt->close_on_fork, ofdt->close_on_fork, cpy);
> +	memset((char *)nfdt->close_on_fork + cpy, 0, set);
>  

I suggest we group the two bits of a file (close_on_exec, close_on_fork) together,
so that we do not have to dirty two separate cache lines.

Otherwise we will add yet another cache line miss at every file opening/closing for processes
with big file tables.

Ie having a _single_ bitmap array, even bit for close_on_exec, odd bit for close_on_fork

static inline void __set_close_on_exec(unsigned int fd, struct fdtable *fdt)
{
	__set_bit(fd * 2, fdt->close_on_fork_exec);
}

static inline void __set_close_on_fork(unsigned int fd, struct fdtable *fdt)
{
	__set_bit(fd * 2 + 1, fdt->close_on_fork_exec);
}

Also the F_GETFD/F_SETFD implementation must use a single function call,
to not acquire the spinlock twice.



  reply index

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-20  7:15 Nate Karstens
2020-04-20  7:15 ` [PATCH 1/4] fs: " Nate Karstens
2020-04-20 10:25   ` Eric Dumazet [this message]
2020-04-22  3:38     ` Changli Gao
2020-04-22  3:41       ` Changli Gao
2020-04-22  8:35     ` David Laight
2020-05-01 14:45     ` Karstens, Nate
2020-05-01 15:23       ` Matthew Wilcox
2020-05-03 13:52       ` David Laight
2020-04-22 15:36   ` Karstens, Nate
2020-04-22 15:43     ` Matthew Wilcox
2020-04-22 16:02       ` Karstens, Nate
2020-04-22 16:31         ` Bernd Petrovitsch
2020-04-22 16:55           ` David Laight
2020-04-23 12:34             ` Bernd Petrovitsch
2020-04-20  7:15 ` [PATCH 2/4] fs: Add O_CLOFORK flag for open(2) and dup3(2) Nate Karstens
2020-04-20  7:15 ` [PATCH 3/4] fs: Add F_DUPFD_CLOFORK to fcntl(2) Nate Karstens
2020-04-20  7:15 ` [PATCH 4/4] net: Add SOCK_CLOFORK Nate Karstens
2020-04-22 14:32 ` Implement close-on-fork James Bottomley
2020-04-22 15:01 ` Al Viro
2020-04-22 15:18   ` Matthew Wilcox
2020-04-22 15:34     ` James Bottomley
2020-04-22 16:00     ` Al Viro
2020-04-22 16:13       ` Al Viro
2020-05-04 13:46       ` Karstens, Nate

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=36dce9b4-a0bf-0015-f6bc-1006938545b1@gmail.com \
    --to=eric.dumazet@gmail.com \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=arnd@arndb.de \
    --cc=bfields@fieldses.org \
    --cc=davem@davemloft.net \
    --cc=deller@gmx.de \
    --cc=ink@jurassic.park.msu.ru \
    --cc=jlayton@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-alpha@vger.kernel.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-parisc@vger.kernel.org \
    --cc=mattst88@gmail.com \
    --cc=nate.karstens@garmin.com \
    --cc=netdev@vger.kernel.org \
    --cc=rth@twiddle.net \
    --cc=sparclinux@vger.kernel.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=xiaosuo@gmail.com \
    /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

Linux-Fsdevel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-fsdevel/0 linux-fsdevel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-fsdevel linux-fsdevel/ https://lore.kernel.org/linux-fsdevel \
		linux-fsdevel@vger.kernel.org
	public-inbox-index linux-fsdevel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-fsdevel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git