All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <jens.axboe@oracle.com>
To: Christoph Hellwig <hch@infradead.org>,
	linux-kernel@vger.kernel.org, cotte@de.ibm.com, hugh@veritas.com,
	neilb@suse.de, zanussi@us.ibm.com,
	Linus Torvalds <torvalds@osdl.org>
Subject: Re: [PATCH] sendfile removal
Date: Thu, 31 May 2007 13:05:50 +0200	[thread overview]
Message-ID: <20070531110550.GR32105@kernel.dk> (raw)
In-Reply-To: <20070531105543.GA25676@infradead.org>

On Thu, May 31 2007, Christoph Hellwig wrote:
> On Thu, May 31, 2007 at 12:33:16PM +0200, Jens Axboe wrote:
> > - nfds: The ->rq_sendfile_ok optimization is gone for now. I can't
> >   determine the value of it, but I'm assuming it's there for a reason.
> >   Any chance this can be converted to splice, or use something else than
> >   ->sendfile()? CC'ed Neil.
> 
> sendfile useage in nfsd avoids a data copy and allows to use checksum
> offloading.  it's quite important for nfs server workloads.

OK, I hope Neil can provide some input on how to convert it. Of course
I'm just fishing for Neil to actually do that work :-)

> > Apart from that, it was mostly straight forward. Almost everybody uses
> > generic_file_sendfile(), which makes the conversion easy. I changed loop
> > to use do_generic_file_read() instead of sendfile, it works for me...
> 
> > diff --git a/drivers/block/loop.c b/drivers/block/loop.c
> > index 5526ead..92bac14 100644
> > --- a/drivers/block/loop.c
> > +++ b/drivers/block/loop.c
> > @@ -435,16 +435,24 @@ do_lo_receive(struct loop_device *lo,
> >  {
> >  	struct lo_read_data cookie;
> >  	struct file *file;
> > -	int retval;
> > +	read_descriptor_t desc;
> > +
> > +	desc.written = 0;
> > +	desc.count = bvec->bv_len;
> > +	desc.arg.data = &cookie;
> > +	desc.error = 0;
> >  
> >  	cookie.lo = lo;
> >  	cookie.page = bvec->bv_page;
> >  	cookie.offset = bvec->bv_offset;
> >  	cookie.bsize = bsize;
> >  	file = lo->lo_backing_file;
> > -	retval = file->f_op->sendfile(file, &pos, bvec->bv_len,
> > -			lo_read_actor, &cookie);
> > -	return (retval < 0)? retval: 0;
> > +
> > +	do_generic_file_read(file, &pos, &desc, lo_read_actor);
> 
> This change is wrong.  loop or any existing user of ->sendfile absolutely
> needs to go through a file operations vector so that file-system specific
> actions such as locking are performed.  This is required at least for the
> clustered filesystems and XFS.   The right way to implement this is
> via do_splice_direct or something similar.
> 
> do_generic_file_read is only a library function for filesystem use
> and should never be called directly.

I'll convert it to do_splice_direct(), thanks.

-- 
Jens Axboe


  reply	other threads:[~2007-05-31 11:06 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-31 10:33 [PATCH] sendfile removal Jens Axboe
2007-05-31 10:47 ` Jens Axboe
2007-05-31 10:47 ` Eric Dumazet
2007-05-31 10:53   ` Jens Axboe
2007-06-01  4:09     ` H. Peter Anvin
2007-06-01  5:41       ` Jens Axboe
2007-06-01  5:50         ` H. Peter Anvin
2007-06-01  7:22           ` Eric Dumazet
2007-06-01 15:52             ` H. Peter Anvin
2007-06-01 16:18               ` Linus Torvalds
2007-06-01 16:47                 ` Eric Dumazet
2007-06-01 16:53                 ` H. Peter Anvin
2007-06-02 15:02                   ` Jens Axboe
2007-06-02 15:01                 ` Jens Axboe
2007-06-02 15:40                   ` Linus Torvalds
2007-06-02 16:35                     ` Jens Axboe
2007-06-03 13:05                     ` Fengguang Wu
2007-06-03 13:05                       ` Fengguang Wu
2007-06-03 14:29                       ` Fengguang Wu
2007-06-03 14:29                         ` Fengguang Wu
2007-06-04  0:46                         ` Fengguang Wu
2007-06-04  0:46                           ` Fengguang Wu
2007-06-04  8:05                           ` Jens Axboe
2007-06-04 11:22                             ` Fengguang Wu
2007-06-04 11:22                               ` Fengguang Wu
2007-06-01 16:22               ` Pádraig Brady
2007-05-31 10:55 ` Christoph Hellwig
2007-05-31 11:05   ` Jens Axboe [this message]
2007-05-31 12:26     ` Neil Brown
2007-05-31 12:27       ` Jens Axboe
2007-06-01  2:44         ` [PATCH] sendfile removal (nfsd update) Neil Brown
2007-06-01  5:44           ` Jens Axboe
2007-06-01  8:01             ` Jens Axboe
2007-06-01  8:15     ` [PATCH] sendfile removal Jens Axboe
2007-05-31 11:04 ` Carsten Otte
2007-05-31 11:06   ` Jens Axboe
2007-05-31 15:33 ` Tom Zanussi
2007-05-31 19:01   ` Jens Axboe
2007-05-31 17:06 ` Hugh Dickins
2007-05-31 17:31   ` Christoph Hellwig
2007-05-31 19:03   ` Jens Axboe

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=20070531110550.GR32105@kernel.dk \
    --to=jens.axboe@oracle.com \
    --cc=cotte@de.ibm.com \
    --cc=hch@infradead.org \
    --cc=hugh@veritas.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=neilb@suse.de \
    --cc=torvalds@osdl.org \
    --cc=zanussi@us.ibm.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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.