qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Scott Wood <scottwood@freescale.com>
To: Stuart Yoder <b08248@gmail.com>
Cc: Wood Scott-B07421 <B07421@freescale.com>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"agraf@suse.de" <agraf@suse.de>,
	"iommu@lists.linux-foundation.org"
	<iommu@lists.linux-foundation.org>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	Yoder Stuart-B08248 <B08248@freescale.com>,
	Alex Williamson <alex.williamson@redhat.com>,
	Bhushan Bharat-R65777 <R65777@freescale.com>
Subject: Re: [Qemu-devel] RFC: vfio API changes needed for powerpc
Date: Wed, 3 Apr 2013 14:26:04 -0500	[thread overview]
Message-ID: <1365017164.25627.7@snotra> (raw)
In-Reply-To: <CALRxmdDHYp1+svjwDrhMc9w2p7H8t87PphEZ1t+97C-EHDR+3g@mail.gmail.com> (from b08248@gmail.com on Wed Apr  3 14:09:45 2013)

On 04/03/2013 02:09:45 PM, Stuart Yoder wrote:
> > Would is be possible for userspace to simply leave room for MSI bank
> > mapping (how much room could be determined by something like
> > VFIO_IOMMU_GET_MSI_BANK_COUNT) then document the API that userspace  
> can
> > DMA_MAP starting at the 0x0 address of the aperture, growing up, and
> > VFIO will map banks on demand at the top of the aperture, growing  
> down?
> > Wouldn't that avoid a lot of issues with userspace needing to know
> > anything about MSI banks (other than count) and coordinating irq  
> numbers
> > and enabling handlers?
> 
> This is basically option #A in the original proposals sent.   I like
> this approach, in that it
> is simpler and keeps user space mostly out of this...which is
> consistent with how things are done
> on x86.  User space just needs to know how many windows to leave at
> the top of the aperture.
> The kernel then has the flexibility to use those windows how it wants.
> 
> But one question, is when should the kernel actually map (and unmap)
> the MSI banks.   One thing we need to do is enable the aperture...and  
> current
> thinking is that is done on the first DMA_MAP.   Similarly when the  
> last mapping
> is unmapped we could also umap the MSI banks.
> 
> Sequence would be something like:
> 
>         VFIO_GROUP_SET_CONTAINER     // add groups to the container
> 
>         VFIO_SET_IOMMU(VFIO_FSL_PAMU)    // set iommu model
> 
>         cnt = VFIO_IOMMU_GET_MSI_BANK_COUNT    // returns max # of  
> MSI banks
> 
>         VFIO_IOMMU_SET_ATTR(ATTR_GEOMETRY) // set overall aperture
> 
>         VFIO_IOMMU_SET_ATTR(ATTR_WINDOWS) // set # of windows,
> including MSI banks
> 
>         VFIO_IOMMU_MAP_DMA    // map the guest's memory
>                ---> kernel enables aperture and maps needed MSI banks  
> here

Maps them where?

What if there is more than one explicit DMA mapping?  What if DMA  
mappings are changed during operation?

-Scott

  parent reply	other threads:[~2013-04-03 19:26 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-02 17:32 [Qemu-devel] RFC: vfio API changes needed for powerpc Yoder Stuart-B08248
2013-04-02 19:39 ` Scott Wood
2013-04-02 20:38   ` Stuart Yoder
2013-04-02 20:47     ` Scott Wood
2013-04-02 20:58       ` Stuart Yoder
2013-04-02 20:32 ` Alex Williamson
2013-04-02 20:54   ` Stuart Yoder
2013-04-02 21:16     ` Alex Williamson
2013-04-02 22:13       ` Scott Wood
2013-04-03  2:54         ` Alex Williamson
2013-04-02 20:57   ` Scott Wood
2013-04-02 21:08     ` Stuart Yoder
2013-04-02 21:38       ` Alex Williamson
2013-04-02 22:50         ` Scott Wood
2013-04-03  3:37           ` Alex Williamson
2013-04-03 19:09             ` Stuart Yoder
2013-04-03 19:18               ` Scott Wood
2013-04-03 19:43                 ` Stuart Yoder
2013-04-03 20:00                   ` Scott Wood
2013-04-03 19:23               ` Alex Williamson
2013-04-03 19:26               ` Scott Wood [this message]
2013-04-03 21:19             ` Scott Wood
2013-04-03 18:32           ` Stuart Yoder
2013-04-03 18:39             ` Scott Wood
2013-04-02 21:55       ` Scott Wood
2013-04-02 21:32     ` Alex Williamson
2013-04-02 22:44       ` Scott Wood
2013-04-03  3:12         ` Alex Williamson
2013-04-03 18:25           ` Stuart Yoder
2013-04-03 21:25           ` Scott Wood

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=1365017164.25627.7@snotra \
    --to=scottwood@freescale.com \
    --cc=B07421@freescale.com \
    --cc=B08248@freescale.com \
    --cc=R65777@freescale.com \
    --cc=agraf@suse.de \
    --cc=alex.williamson@redhat.com \
    --cc=b08248@gmail.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=kvm@vger.kernel.org \
    --cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).