From: hdoyu@nvidia•com (Hiroshi Doyu)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH v2] devicetree: Add generic IOMMU device tree bindings
Date: Fri, 30 May 2014 22:29:13 +0300 [thread overview]
Message-ID: <87tx87rrp2.fsf@nvidia.com> (raw)
In-Reply-To: <4545972.cM7IP1qTXQ@wuerfel>
Arnd Bergmann <arnd@arndb•de> writes:
>> > +Multiple-master IOMMU:
>> > +----------------------
>> > +
>> > + iommu {
>> > + /* the specifier represents the ID of the master */
>> > + #address-cells = <1>;
>> > + #size-cells = <0>;
>> > + };
>> > +
>> > + master {
>> > + /* device has master ID 42 in the IOMMU */
>> > + iommus = <&/iommu 42>;
>> > + };
>>
>> Presumably the ID would be the streamID on ARM's SMMU. How would a
>> master with 8 streamIDs be described? This is what Calxeda midway has
>> for SATA and I would expect that to be somewhat common. Either you
>> need some ID masking or you'll have lots of duplication when you have
>> windows.
>
> I don't understand the problem. If you have stream IDs 0 through 7,
> you would have
>
> master at a {
> ...
> iommus = <&smmu 0>;
> };
>
> master at b {
> ...
> iommus = <&smmu 1;
> };
>
> ...
>
> master at 12 {
> ...
> iommus = <&smmu 7;
> };
>
> and you don't need a window at all. Why would you need a mask of
> some sort?
IIUC the original problem, "a master with 8 streamIDs" means something
like below, where some devices have multiple IDs but some have a
single. A sinle #address-cells cannot afford those 2 masters at once.
iommu {
/* the specifier represents the ID of the master */
#address-cells = <1>;
#size-cells = <0>;
};
master at a {
...
iommus = <&smmu 1 2 3>; # 3 IDs
};
master at b {
...
iommus = <&smmu 4>; # 1 ID
};
Tegra,SMMU has a similar problem and we have used a fixed size bitmap(64
bit) to afford 64 stream IDs so that a single device can hold multiple
IDs. If we apply the same bitmap to the above exmaple:
iommu {
/* the specifier represents the ID of the master */
#address-cells = <1>;
#size-cells = <0>;
};
master at a {
...
iommus = <&smmu (BIT(1) | BIT(2) | BIT(3))>; # IDs 1 2 3
};
master at b {
...
iommus = <&smmu BIT(4)>; # ID 4
};
The disadvantage of this is that this limits the max number of streamIDs
to support. If # of streamID is increased later more than 64, this
format cannot cover any more. You have to predict the max # of streamIDs
in advance if steamID is statically assigned.
next prev parent reply other threads:[~2014-05-30 19:29 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1400877218-4113-1-git-send-email-thierry.reding@gmail.com>
2014-05-30 13:16 ` [PATCH v2] devicetree: Add generic IOMMU device tree bindings Rob Herring
2014-05-30 19:06 ` Arnd Bergmann
2014-05-30 19:29 ` Hiroshi Doyu [this message]
2014-05-30 19:54 ` Arnd Bergmann
2014-06-01 9:55 ` Will Deacon
2014-06-04 13:39 ` Thierry Reding
2014-06-04 13:44 ` Thierry Reding
2014-06-04 13:53 ` Arnd Bergmann
2014-06-04 13:56 ` Will Deacon
2014-06-04 14:01 ` Arnd Bergmann
2014-06-04 16:39 ` Will Deacon
2014-05-30 19:31 ` Rob Herring
2014-05-30 19:49 ` Arnd Bergmann
2014-06-02 10:41 ` Dave Martin
2014-06-04 14:35 ` Thierry Reding
2014-06-04 16:41 ` Will Deacon
2014-06-04 21:00 ` Thierry Reding
2014-06-05 19:10 ` Varun Sethi
2014-06-16 15:27 ` Will Deacon
2014-06-16 16:56 ` Stuart Yoder
2014-06-16 17:04 ` Will Deacon
2014-06-16 17:30 ` Arnd Bergmann
2014-06-16 18:53 ` Stuart Yoder
2014-06-17 10:26 ` Varun Sethi
2014-06-17 10:43 ` Will Deacon
2014-06-17 11:21 ` Varun Sethi
2014-06-17 14:50 ` Stuart Yoder
2014-06-18 9:29 ` Will Deacon
2014-06-17 14:39 ` Stuart Yoder
2014-06-20 23:16 ` Olav Haugan
2014-06-24 9:18 ` Will Deacon
2014-06-24 17:57 ` Olav Haugan
2014-06-24 18:11 ` Will Deacon
2014-06-24 18:20 ` Arnd Bergmann
2014-06-25 9:17 ` Will Deacon
2014-06-25 9:27 ` Arnd Bergmann
2014-06-25 9:38 ` Will Deacon
2014-06-25 9:48 ` Arnd Bergmann
2014-06-25 9:57 ` Will Deacon
2014-06-25 10:12 ` Arnd Bergmann
2014-06-25 10:14 ` Will Deacon
2014-06-24 21:35 ` Olav Haugan
2014-06-25 9:18 ` Will Deacon
2014-06-27 22:23 ` Olav Haugan
2014-06-30 9:52 ` Will Deacon
2014-07-09 1:07 ` Olav Haugan
2014-07-09 10:54 ` Will Deacon
2014-07-10 22:32 ` Olav Haugan
2014-07-11 12:24 ` Will Deacon
[not found] <20140606224542.GA22188@mithrandir>
2014-06-07 13:22 ` Arnd Bergmann
2014-06-09 10:49 ` Thierry Reding
2014-05-23 20:36 Thierry Reding
2014-05-29 15:52 ` Stephen Warren
2014-05-30 7:30 ` Thierry Reding
2014-05-30 11:27 ` Dave Martin
2014-05-30 19:11 ` Arnd Bergmann
2014-06-02 10:56 ` Dave Martin
2014-06-04 21:12 ` Thierry Reding
2014-06-16 12:57 ` Will Deacon
2014-06-17 11:58 ` Thierry Reding
2014-06-17 12:18 ` Will Deacon
2014-06-17 23:37 ` Thierry Reding
2014-06-18 10:14 ` Will Deacon
2014-06-20 15:53 ` Arnd Bergmann
2014-06-20 17:50 ` Will Deacon
2014-06-20 18:55 ` Arnd Bergmann
2014-05-30 11:22 ` Dave Martin
2014-05-30 19:01 ` Arnd Bergmann
2014-06-02 11:44 ` Dave Martin
2014-06-04 21:32 ` Thierry Reding
2014-06-05 9:42 ` Arnd Bergmann
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=87tx87rrp2.fsf@nvidia.com \
--to=hdoyu@nvidia$(echo .)com \
--cc=linux-arm-kernel@lists$(echo .)infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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