From: Joerg Roedel <joro@8bytes.org>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: Paul Brook <paul@codesourcery.com>,
qemu-devel@nongnu.org, avi@redhat.com, kvm@vger.kernel.org,
Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Subject: Re: [Qemu-devel] Re: [RFC PATCH 4/7] ide: IOMMU support
Date: Thu, 15 Jul 2010 16:45:34 +0200 [thread overview]
Message-ID: <20100715144534.GF23755@8bytes.org> (raw)
In-Reply-To: <4C3F02D2.2090006@codemonkey.ws>
On Thu, Jul 15, 2010 at 07:45:06AM -0500, Anthony Liguori wrote:
> On 07/15/2010 04:10 AM, Joerg Roedel wrote:
>> If this means a seperate interface for device dma accesses and not fold
>> that functionality into the cpu_physical_memory* interface I agree too :-)
>>
> No. PCI devices should never call cpu_physical_memory*.
Fully agreed.
> PCI devices should call pci_memory*.
>
> ISA devices should call isa_memory*.
This is a seperate interface. I like the idea and as you stated below it
has clear advantages, so lets go this way.
> All device memory accesses should go through their respective buses.
> There can be multiple IOMMUs at different levels of the device
> hierarchy. If you don't provide bus-level memory access functions that
> chain through the hierarchy, it's extremely difficult to implement all
> the necessary hooks to perform the translations at different places.
Joerg
WARNING: multiple messages have this Message-ID (diff)
From: Joerg Roedel <joro@8bytes.org>
To: Anthony Liguori <anthony@codemonkey.ws>
Cc: avi@redhat.com,
Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>,
Paul Brook <paul@codesourcery.com>,
kvm@vger.kernel.org, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] Re: [RFC PATCH 4/7] ide: IOMMU support
Date: Thu, 15 Jul 2010 16:45:34 +0200 [thread overview]
Message-ID: <20100715144534.GF23755@8bytes.org> (raw)
In-Reply-To: <4C3F02D2.2090006@codemonkey.ws>
On Thu, Jul 15, 2010 at 07:45:06AM -0500, Anthony Liguori wrote:
> On 07/15/2010 04:10 AM, Joerg Roedel wrote:
>> If this means a seperate interface for device dma accesses and not fold
>> that functionality into the cpu_physical_memory* interface I agree too :-)
>>
> No. PCI devices should never call cpu_physical_memory*.
Fully agreed.
> PCI devices should call pci_memory*.
>
> ISA devices should call isa_memory*.
This is a seperate interface. I like the idea and as you stated below it
has clear advantages, so lets go this way.
> All device memory accesses should go through their respective buses.
> There can be multiple IOMMUs at different levels of the device
> hierarchy. If you don't provide bus-level memory access functions that
> chain through the hierarchy, it's extremely difficult to implement all
> the necessary hooks to perform the translations at different places.
Joerg
next prev parent reply other threads:[~2010-07-15 14:45 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-14 5:45 [RFC PATCH 0/7] AMD IOMMU emulation patchset Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [RFC PATCH 1/7] Generic IOMMU layer Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 6:07 ` malc
2010-07-14 6:07 ` [Qemu-devel] " malc
2010-07-14 22:47 ` Eduard - Gabriel Munteanu
2010-07-14 22:47 ` Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [RFC PATCH 2/7] AMD IOMMU emulation Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 20:16 ` Paul Brook
2010-07-14 20:16 ` Paul Brook
2010-07-14 5:45 ` [RFC PATCH 3/7] pci: call IOMMU hooks Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 7:37 ` Isaku Yamahata
2010-07-14 7:37 ` Isaku Yamahata
2010-07-14 22:50 ` Eduard - Gabriel Munteanu
2010-07-14 22:50 ` Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [RFC PATCH 4/7] ide: IOMMU support Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 13:53 ` Paul Brook
2010-07-14 13:53 ` [Qemu-devel] " Paul Brook
2010-07-14 18:33 ` Joerg Roedel
2010-07-14 18:33 ` [Qemu-devel] " Joerg Roedel
2010-07-14 20:13 ` Paul Brook
2010-07-14 21:29 ` Anthony Liguori
2010-07-14 21:29 ` Anthony Liguori
2010-07-14 22:24 ` Chris Wright
2010-07-14 22:24 ` Chris Wright
2010-07-15 10:28 ` Paul Brook
2010-07-15 10:28 ` Paul Brook
2010-07-15 16:52 ` Chris Wright
2010-07-15 16:52 ` Chris Wright
2010-07-15 17:02 ` Avi Kivity
2010-07-15 17:02 ` Avi Kivity
2010-07-15 17:17 ` Chris Wright
2010-07-15 17:17 ` Chris Wright
2010-07-15 17:22 ` Avi Kivity
2010-07-15 17:22 ` Avi Kivity
2010-07-15 17:25 ` Chris Wright
2010-07-15 17:25 ` Chris Wright
2010-07-15 17:27 ` Eduard - Gabriel Munteanu
2010-07-15 17:27 ` Eduard - Gabriel Munteanu
2010-07-15 17:22 ` Joerg Roedel
2010-07-15 17:22 ` Joerg Roedel
2010-07-15 17:14 ` Chris Wright
2010-07-15 17:14 ` Chris Wright
2010-07-15 9:10 ` Joerg Roedel
2010-07-15 9:10 ` Joerg Roedel
2010-07-15 12:45 ` Anthony Liguori
2010-07-15 12:45 ` Anthony Liguori
2010-07-15 14:45 ` Joerg Roedel [this message]
2010-07-15 14:45 ` Joerg Roedel
2010-07-15 16:45 ` Eduard - Gabriel Munteanu
2010-07-15 16:45 ` Eduard - Gabriel Munteanu
2010-07-15 17:42 ` Anthony Liguori
2010-07-15 17:42 ` Anthony Liguori
2010-07-15 10:33 ` Paul Brook
2010-07-15 10:33 ` Paul Brook
2010-07-15 12:42 ` Anthony Liguori
2010-07-15 12:42 ` Anthony Liguori
2010-07-15 14:02 ` Paul Brook
2010-07-15 14:02 ` Paul Brook
2010-07-14 23:39 ` Eduard - Gabriel Munteanu
2010-07-14 23:39 ` Eduard - Gabriel Munteanu
2010-07-15 9:22 ` Joerg Roedel
2010-07-15 9:22 ` Joerg Roedel
2010-07-15 10:49 ` Paul Brook
2010-07-15 10:49 ` Paul Brook
2010-07-15 14:59 ` Joerg Roedel
2010-07-15 14:59 ` Joerg Roedel
2010-07-14 23:11 ` Eduard - Gabriel Munteanu
2010-07-14 23:11 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-15 10:58 ` Paul Brook
2010-07-15 10:58 ` [Qemu-devel] " Paul Brook
2010-07-14 5:45 ` [RFC PATCH 5/7] rtl8139: " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [RFC PATCH 6/7] eepro100: " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [RFC PATCH 7/7] ac97: " Eduard - Gabriel Munteanu
2010-07-14 5:45 ` [Qemu-devel] " Eduard - Gabriel Munteanu
2010-07-14 6:09 ` malc
2010-07-14 6:09 ` [Qemu-devel] " malc
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=20100715144534.GF23755@8bytes.org \
--to=joro@8bytes.org \
--cc=anthony@codemonkey.ws \
--cc=avi@redhat.com \
--cc=eduard.munteanu@linux360.ro \
--cc=kvm@vger.kernel.org \
--cc=paul@codesourcery.com \
--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 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.