From: Greg Law <glaw@solarflare.com>
To: Caitlin Bestler <caitlinb@broadcom.com>
Cc: xen@solarflare.com, xen-devel@lists.xensource.com
Subject: Re: RFC -- virtualisable NIC support in Xen
Date: Wed, 27 Sep 2006 13:53:28 +0100 [thread overview]
Message-ID: <451A7448.4000103@solarflare.com> (raw)
In-Reply-To: <54AD0F12E08D1541B826BE97C98F99F1A31057@NT-SJCA-0751.brcm.ad.broadcom.com>
Caitlin Bestler wrote:
>
>
>
>>-----Original Message-----
>>From: Greg Law [mailto:glaw@solarflare.com]
>>Sent: Friday, September 22, 2006 4:05 AM
>>To: Caitlin Bestler
>>Cc: xen@solarflare.com; xen-devel@lists.xensource.com
>>Subject: Re: [Xen-devel] RFC -- virtualisable NIC support in Xen
>>
>>Caitlin Bestler wrote:
>>
>>>When the App/DomU posts a buffer (for TX or RX), what
>>
>>address does it
>>
>>>use, and how does the hardware understand it?
>>
>>It uses a "buffer address". This is essentially an index
>>into a "buffer-table" which is stored on the NIC. This table
>>is used to lookup physical addresses that the NIC will DMA to/from.
>>
>>The buffer-table is programmed from Dom0.
>>
>>The buffer-table is effectively our IOMMU/page-tables.
>>
>
>
> Over what scope is the buffer handle unique?
The buffer table is per NIC. The handle into that buffer table is tied to a
particular Virtual Interface, meaning a guest driver can only DMA to/from
buffers associated with the virtual interface to which it has access (the VIs
are all mapped page aligned so that the usual paging can be used to prevent one
guest from accessing the VI of another).
Can I ask what your interest is here? Do you think Broadcom might be interested
in the kind of model we're proposing?
Cheers,
Greg
next prev parent reply other threads:[~2006-09-27 12:53 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <54AD0F12E08D1541B826BE97C98F99F1A30F2F@NT-SJCA-0751.brcm.ad.broadcom.com>
2006-09-22 11:04 ` RFC -- virtualisable NIC support in Xen Greg Law
2006-09-22 18:13 ` Caitlin Bestler
2006-09-27 12:53 ` Greg Law [this message]
2006-09-20 16:21 Greg Law
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=451A7448.4000103@solarflare.com \
--to=glaw@solarflare.com \
--cc=caitlinb@broadcom.com \
--cc=xen-devel@lists.xensource.com \
--cc=xen@solarflare.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.