From: Christoph Hellwig <hch@infradead.org>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: "Steinar H. Gunderson" <sesse@google.com>,
Christoph Hellwig <hch@infradead.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org
Subject: Re: [PATCH] Add support for usbfs zerocopy.
Date: Fri, 8 Jan 2016 01:45:35 -0800 [thread overview]
Message-ID: <20160108094535.GA17286@infradead.org> (raw)
In-Reply-To: <Pine.LNX.4.44L0.1601061032000.1579-100000@iolanthe.rowland.org>
On Wed, Jan 06, 2016 at 10:35:05AM -0500, Alan Stern wrote:
> Indeed, the I/O operations we are using with mmap here are not reads or
> writes; they are ioctls. As far as I know, the kernel doesn't have any
> defined interface for zerocopy ioctls.
IF it was using mmap for I/O it would read in through the page fault
handler an then mark the page dirty for writeback by the VM. Thats
clearly not the case.
Instead it's using mmap on a file as a pecial purpose anonymous
memory allocator, bypassing the VM and VM policies, including
allowing to pin kernel memory that way.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@infradead.org>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: "Steinar H. Gunderson" <sesse@google.com>,
Christoph Hellwig <hch@infradead.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org
Subject: Re: [PATCH] Add support for usbfs zerocopy.
Date: Fri, 8 Jan 2016 01:45:35 -0800 [thread overview]
Message-ID: <20160108094535.GA17286@infradead.org> (raw)
In-Reply-To: <Pine.LNX.4.44L0.1601061032000.1579-100000@iolanthe.rowland.org>
On Wed, Jan 06, 2016 at 10:35:05AM -0500, Alan Stern wrote:
> Indeed, the I/O operations we are using with mmap here are not reads or
> writes; they are ioctls. As far as I know, the kernel doesn't have any
> defined interface for zerocopy ioctls.
IF it was using mmap for I/O it would read in through the page fault
handler an then mark the page dirty for writeback by the VM. Thats
clearly not the case.
Instead it's using mmap on a file as a pecial purpose anonymous
memory allocator, bypassing the VM and VM policies, including
allowing to pin kernel memory that way.
next prev parent reply other threads:[~2016-01-08 9:45 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <Pine.LNX.4.44L0.1601051024110.1666-100000@iolanthe.rowland.org>
2016-01-05 16:27 ` Does vm_operations_struct require a .owner field? Alan Stern
2016-01-05 16:27 ` Alan Stern
2016-01-05 20:58 ` Kirill A. Shutemov
2016-01-05 20:58 ` Kirill A. Shutemov
2016-01-05 21:31 ` Alan Stern
2016-01-05 21:31 ` Alan Stern
2015-11-26 0:19 ` [PATCH] Add support for usbfs zerocopy Steinar H. Gunderson
2016-01-06 0:11 ` Greg Kroah-Hartman
2015-11-26 0:19 ` [PATCH v2] " Steinar H. Gunderson
2016-01-07 2:01 ` Lingzhu Xiang
2016-01-07 15:40 ` Alan Stern
2016-01-11 11:08 ` Oliver Neukum
[not found] ` <20160111161504.GG10641@char.us.oracle.com>
2016-01-12 10:00 ` Oliver Neukum
2016-01-09 4:00 ` Lingzhu Xiang
2016-01-09 10:57 ` Steinar H. Gunderson
2016-01-24 21:12 ` Greg Kroah-Hartman
2016-01-25 8:03 ` Steinar H. Gunderson
2016-02-02 10:34 ` Steinar H. Gunderson
2016-02-03 21:23 ` Greg Kroah-Hartman
2016-02-03 21:58 ` [PATCH v4] " Steinar H. Gunderson
2016-02-03 22:09 ` [PATCH v2] " Steinar H. Gunderson
2016-02-03 22:15 ` Felipe Balbi
2016-02-03 23:40 ` Steinar H. Gunderson
2016-02-04 10:17 ` Bjørn Mork
2016-02-04 10:26 ` Steinar H. Gunderson
2016-02-12 20:55 ` Steinar H. Gunderson
2016-01-06 0:24 ` [PATCH] " Steinar H. Gunderson
2016-01-06 6:49 ` Christoph Hellwig
2016-01-06 14:45 ` Steinar H. Gunderson
2016-01-06 15:22 ` Peter Stuge
2016-01-06 15:36 ` Steinar H. Gunderson
2016-01-06 15:39 ` Alan Stern
2016-01-06 15:35 ` Alan Stern
2016-01-08 9:45 ` Christoph Hellwig [this message]
2016-01-08 9:45 ` Christoph Hellwig
2016-01-08 10:22 ` David Laight
2016-01-08 10:22 ` David Laight
2016-01-08 16:02 ` Oliver Neukum
2016-01-08 16:02 ` Oliver Neukum
2016-01-12 21:26 ` Steinar H. Gunderson
2016-01-12 21:26 ` Steinar H. Gunderson
2016-01-12 22:05 ` Alan Stern
2016-01-12 22:05 ` Alan Stern
2016-01-13 6:59 ` Lingzhu Xiang
2016-01-13 17:21 ` Greg Kroah-Hartman
2016-01-08 9:43 ` Christoph Hellwig
2016-02-24 19:30 ` Sasha Levin
2016-02-24 19:33 ` Steinar H. Gunderson
2016-01-05 23:54 ` Does vm_operations_struct require a .owner field? Steinar H. Gunderson
2016-01-05 23:54 ` Steinar H. Gunderson
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=20160108094535.GA17286@infradead.org \
--to=hch@infradead.org \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-usb@vger.kernel.org \
--cc=sesse@google.com \
--cc=stern@rowland.harvard.edu \
/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.