mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Badari Pulavarty <badari@us.ibm.com>
To: cel@citi.umich.edu (Chuck Lever)
Cc: badari@us.ibm.com (Badari Pulavarty),
	linux-aio@kvack.org, linux-kernel@vger.kernel.org (lkml),
	akpm@digeo.com
Subject: Re: [RFC] generic_file_direct_IO() argument changes for AIO
Date: Tue, 29 Oct 2002 11:28:55 -0800 (PST)	[thread overview]
Message-ID: <200210291928.g9TJSuG07916@eng2.beaverton.ibm.com> (raw)
In-Reply-To: <Pine.BSO.4.33.0210291414120.11378-100000@citi.umich.edu> from "Chuck Lever" at Oct 29, 2002 01:15:56 PM PST

> 
> On Tue, 29 Oct 2002, Badari Pulavarty wrote:
> 
> > Inorder to support AIO for raw/O_DIRECT, I need to add "struct kiocb *"
> > to generic_file_direct_IO() and all direct_IO() ops.
> >
> > generic_file_direct_IO(int rw, struct file *file, const struct iovec *iov,
> >         loff_t offset, unsigned long nr_segs)
> >
> > Instead of adding a new argument, I propose replace
> >
> > 	"struct file *file" argument with "struct kiocb *iocb"
> >
> > One can get "filp" from iocb->ki_filp.
> >
> > Any objections ?
> 
> here's a naive objection.  is there any preallocated data structure
> associated with a kiocb other than the struct itself?  if there is, that
> will have some scalability impact for applications like Oracle that like
> to create a very large database by opening thousands of files.
> 
> 	- Chuck Lever

If the application is not using Async IO, "kiocb" structure comes
from stack. Other than that, there are no preallocated strcutures
associated with kiocb.

- Badari

      reply	other threads:[~2002-10-29 19:23 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-10-29 18:19 Badari Pulavarty
2002-10-29 18:58 ` Andrew Morton
2002-10-29 19:15 ` Chuck Lever
2002-10-29 19:28   ` Badari Pulavarty [this message]

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=200210291928.g9TJSuG07916@eng2.beaverton.ibm.com \
    --to=badari@us.ibm.com \
    --cc=akpm@digeo.com \
    --cc=cel@citi.umich.edu \
    --cc=linux-aio@kvack.org \
    --cc=linux-kernel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®