From: viresh.kumar@linaro.org (Viresh Kumar)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 6/9] PM / OPP: dt-bindings: Add opp-interconnect-bw
Date: Wed, 10 Oct 2018 15:29:51 +0530 [thread overview]
Message-ID: <20181010095951.hgeiy7yxsqi6behd@vireshk-i7> (raw)
In-Reply-To: <0998a374-6cb0-9218-d2e3-92f8ee9861ed@linaro.org>
On 27-09-18, 11:23, Georgi Djakov wrote:
> On 08/27/2018 06:11 PM, Jordan Crouse wrote:
> > Add the "opp-interconnect-bw" property to specify the
> > average and peak bandwidth for an interconnect path for
> > a specific operating power point. A separate bandwidth
> > pair can be specified for each of the interconnects
> > defined for the device by appending the interconnect
> > name to the property.
> >
> > Signed-off-by: Jordan Crouse <jcrouse@codeaurora.org>
> > ---
> > Documentation/devicetree/bindings/opp/opp.txt | 36 +++++++++++++++++++
> > 1 file changed, 36 insertions(+)
> >
> > diff --git a/Documentation/devicetree/bindings/opp/opp.txt b/Documentation/devicetree/bindings/opp/opp.txt
> > index c396c4c0af92..d714c084f36d 100644
> > --- a/Documentation/devicetree/bindings/opp/opp.txt
> > +++ b/Documentation/devicetree/bindings/opp/opp.txt
> > @@ -170,6 +170,11 @@ Optional properties:
> > functioning of the current device at the current OPP (where this property is
> > present).
> >
> > +- opp-interconnect-bw-<name>: This is an array of pairs specifying the average
> > + and peak bandwidth in bytes per second for the interconnect path known by
> > + 'name'. This should match the name(s) specified by interconnect-names in the
> > + device definition.
> > +
> > Example 1: Single cluster Dual-core ARM cortex A9, switch DVFS states together.
> >
> > / {
> > @@ -543,3 +548,34 @@ Example 6: opp-microvolt-<name>, opp-microamp-<name>:
> > };
> > };
> > };
> > +
> > +Example 7: opp-interconnect-bw:
> > +(example: leaf device with frequency and BW quotas)
> > +
> > +/ {
> > + soc {
> > + gpu at 5000000 {
> > + ...
> > + interconnects = <&qnoc 26 &qnoc 512>;
> > + interconnect-names = "port0";
> > + ...
> > + operating-points-v2 = <&gpu_opp_table>;
> > + };
> > + };
> > +
> > + gpu_opp_table: opp_table0 {
> > + compatible = "operating-points-v2";
> > +
> > + opp-710000000 {
> > + op-hz = /bits/ 64 <710000000>;
> > + /* Set peak bandwidth @ 7216000 KB/s */
> > + opp-interconnect-bw-port0 = /bits/ 64 <0 7216000000>;
>
> This seems a bit long. I would suggest the following instead.
> If there is only one path:
> /* average bandwidth = 0 KB/s, peak bandwidth = 7216000 KB/s */
> opp-bw-KBps = <0 7216000>;
> or
> opp-bw-MBps = <0 7216>;
>
> If there are multiple paths:
> opp-bw-KBps-port0 = <0 7216000>;
> opp-bw-KBps-port1 = <0 1000000>;
>
> The above follows a convention similar to opp-microvolt, where at
> runtime the platform can pick a <name> and a matching opp-bw-KBps-<name>
> property. If the platform doesn't pick a specific <name> or <name> does
> not match with any of the opp-bw-KBps-<name> properties, then
> opp-bw-KBps shall be used if present.
> For now supporting only KBps values seems enough to cover all present
> platforms, so we can start with this and based on the requirements of
> future platforms we can add MBps etc.
+1
And yes I am fine with such bindings getting introduced to the OPP
core.
I am not sure if this would fit well, but have a look at
"required-opps" property in OPP bindings and see if that can be used
instead of adding new bindings here. Again, I am not sure if that
should be done :)
--
viresh
next prev parent reply other threads:[~2018-10-10 9:59 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-27 15:11 [PATCH 0/9] Add interconnect support + bindings for A630 GPU Jordan Crouse
2018-08-27 15:11 ` [PATCH 1/9] drm/msm/a6xx: rnndb updates for a6xx Jordan Crouse
2018-08-27 15:11 ` [PATCH 2/9] drm/msm/a6xx: Fix PDC register overlap Jordan Crouse
2018-08-27 15:11 ` [PATCH 3/9] drm/msm/a6xx: Rename gmu phandle to qcom,gmu Jordan Crouse
2018-08-27 15:11 ` [PATCH 4/9] dt-bindings: Document qcom,adreno-gmu Jordan Crouse
2018-08-27 15:11 ` [PATCH 5/9] arm64: dts: sdm845: Add gpu and gmu device nodes Jordan Crouse
2018-08-28 10:30 ` Vivek Gautam
2018-10-10 9:46 ` Viresh Kumar
2018-10-10 14:29 ` Jordan Crouse
2018-10-10 14:31 ` Viresh Kumar
2018-10-10 14:48 ` [Freedreno] " Jordan Crouse
2018-10-10 14:51 ` Viresh Kumar
2018-10-10 15:10 ` Jordan Crouse
2018-10-11 5:02 ` Viresh Kumar
2018-10-11 14:54 ` Jordan Crouse
2018-10-15 10:03 ` Viresh Kumar
2018-10-15 14:34 ` Jordan Crouse
2018-10-22 10:38 ` Viresh Kumar
2018-10-22 13:20 ` Niklas Cassel
2018-10-22 14:37 ` Jordan Crouse
2018-10-22 14:34 ` Jordan Crouse
2018-10-17 18:28 ` Doug Anderson
2018-08-27 15:11 ` [PATCH 6/9] PM / OPP: dt-bindings: Add opp-interconnect-bw Jordan Crouse
2018-09-27 8:23 ` Georgi Djakov
2018-10-10 9:59 ` Viresh Kumar [this message]
2018-10-10 14:27 ` Jordan Crouse
2018-10-10 14:29 ` Viresh Kumar
2018-10-15 14:34 ` Rob Herring
2018-10-15 15:12 ` Jordan Crouse
2018-08-27 15:11 ` [PATCH 7/9] OPP: Add dev_pm_opp_get_interconnect_bw() Jordan Crouse
2018-10-05 6:36 ` Sharat Masetty
2018-10-05 17:13 ` Jordan Crouse
2018-08-27 15:11 ` [PATCH 8/9] drm/msm/a6xx: Add support for an interconnect path Jordan Crouse
2018-08-28 1:23 ` kbuild test robot
2018-08-27 15:11 ` [PATCH 9/9] arm64: dts: Add interconnect for the GPU on SDM845 Jordan Crouse
2018-08-28 18:16 ` [Freedreno] " Jordan Crouse
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=20181010095951.hgeiy7yxsqi6behd@vireshk-i7 \
--to=viresh.kumar@linaro.org \
--cc=linux-arm-kernel@lists.infradead.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