From: Leonard Crestez <leonard.crestez@nxp.com>
To: "Artur Świgoń" <a.swigon@partner.samsung.com>,
"Georgi Djakov" <georgi.djakov@linaro.org>
Cc: "devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-samsung-soc@vger.kernel.org"
<linux-samsung-soc@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
"dri-devel@lists.freedesktop.org"
<dri-devel@lists.freedesktop.org>,
"krzk@kernel.org" <krzk@kernel.org>,
"cw00.choi@samsung.com" <cw00.choi@samsung.com>,
"myungjoo.ham@samsung.com" <myungjoo.ham@samsung.com>,
"inki.dae@samsung.com" <inki.dae@samsung.com>,
"sw0312.kim@samsung.com" <sw0312.kim@samsung.com>,
"georgi.djakov@linaro.org" <georgi.djakov@linaro.org>,
"m.szyprowski@samsung.com" <m.szyprowski@samsung.com>
Subject: Re: [RFC PATCH 09/11] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Tue, 6 Aug 2019 13:41:31 +0000 [thread overview]
Message-ID: <VI1PR04MB5055BED59960B4F4147479AEEED50@VI1PR04MB5055.eurprd04.prod.outlook.com> (raw)
In-Reply-To: 20190723122016.30279-10-a.swigon@partner.samsung.com
On 23.07.2019 15:21, Artur Świgoń wrote:
> +static int exynos_bus_icc_aggregate(struct icc_node *node, u32 avg_bw,
> + u32 peak_bw, u32 *agg_avg, u32 *agg_peak)
> +{
> + *agg_peak = *agg_avg = peak_bw;
> +
> + return 0;
> +}
The only current provider aggregates "avg" with "sum" and "peak" with
"max", any particular reason to do something different? This function
doesn't even really do aggregation, if there is a second request for "0"
then the first request is lost.
I didn't find any docs but my interpretation of avg/peak is that "avg"
is for constant traffic like a display or VPU pushing pixels and "peak"
is for variable traffic like networking. I assume devices which make
"peak" requests are aggregated with max because they're not expected to
all max-out together.
In PATCH 11 you're making a bandwidth request based on resolution, that
traffic is constant and guaranteed to happend while the display is on so
it would make sense to request it as an "avg" and aggregate it with "sum".
--
Regards,
Leonard
WARNING: multiple messages have this Message-ID (diff)
From: Leonard Crestez <leonard.crestez@nxp.com>
To: "Artur Świgoń" <a.swigon@partner.samsung.com>
Cc: "devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-samsung-soc@vger.kernel.org"
<linux-samsung-soc@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
"dri-devel@lists.freedesktop.org"
<dri-devel@lists.freedesktop.org>,
"krzk@kernel.org" <krzk@kernel.org>,
"cw00.choi@samsung.com" <cw00.choi@samsung.com>,
"myungjoo.ham@samsung.com" <myungjoo.ham@samsung.com>,
"inki.dae@samsung.com" <inki.dae@samsung.com>,
"sw0312.kim@samsung.com" <sw0312.kim@samsung.com>,
"georgi.djakov@linaro.org" <georgi.djakov@linaro.org>,
"m.szyprowski@samsung.com" <m.szyprowski@samsung.com>
Subject: Re: [RFC PATCH 09/11] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Tue, 6 Aug 2019 13:41:31 +0000 [thread overview]
Message-ID: <VI1PR04MB5055BED59960B4F4147479AEEED50@VI1PR04MB5055.eurprd04.prod.outlook.com> (raw)
In-Reply-To: 20190723122016.30279-10-a.swigon@partner.samsung.com
On 23.07.2019 15:21, Artur Świgoń wrote:
> +static int exynos_bus_icc_aggregate(struct icc_node *node, u32 avg_bw,
> + u32 peak_bw, u32 *agg_avg, u32 *agg_peak)
> +{
> + *agg_peak = *agg_avg = peak_bw;
> +
> + return 0;
> +}
The only current provider aggregates "avg" with "sum" and "peak" with
"max", any particular reason to do something different? This function
doesn't even really do aggregation, if there is a second request for "0"
then the first request is lost.
I didn't find any docs but my interpretation of avg/peak is that "avg"
is for constant traffic like a display or VPU pushing pixels and "peak"
is for variable traffic like networking. I assume devices which make
"peak" requests are aggregated with max because they're not expected to
all max-out together.
In PATCH 11 you're making a bandwidth request based on resolution, that
traffic is constant and guaranteed to happend while the display is on so
it would make sense to request it as an "avg" and aggregate it with "sum".
--
Regards,
Leonard
WARNING: multiple messages have this Message-ID (diff)
From: Leonard Crestez <leonard.crestez@nxp.com>
To: "Artur Świgoń" <a.swigon@partner.samsung.com>,
"Georgi Djakov" <georgi.djakov@linaro.org>
Cc: "devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
"linux-samsung-soc@vger.kernel.org"
<linux-samsung-soc@vger.kernel.org>,
"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
"sw0312.kim@samsung.com" <sw0312.kim@samsung.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"dri-devel@lists.freedesktop.org"
<dri-devel@lists.freedesktop.org>,
"inki.dae@samsung.com" <inki.dae@samsung.com>,
"cw00.choi@samsung.com" <cw00.choi@samsung.com>,
"myungjoo.ham@samsung.com" <myungjoo.ham@samsung.com>,
"krzk@kernel.org" <krzk@kernel.org>,
"georgi.djakov@linaro.org" <georgi.djakov@linaro.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"m.szyprowski@samsung.com" <m.szyprowski@samsung.com>
Subject: Re: [RFC PATCH 09/11] devfreq: exynos-bus: Add interconnect functionality to exynos-bus
Date: Tue, 6 Aug 2019 13:41:31 +0000 [thread overview]
Message-ID: <VI1PR04MB5055BED59960B4F4147479AEEED50@VI1PR04MB5055.eurprd04.prod.outlook.com> (raw)
In-Reply-To: 20190723122016.30279-10-a.swigon@partner.samsung.com
On 23.07.2019 15:21, Artur Świgoń wrote:
> +static int exynos_bus_icc_aggregate(struct icc_node *node, u32 avg_bw,
> + u32 peak_bw, u32 *agg_avg, u32 *agg_peak)
> +{
> + *agg_peak = *agg_avg = peak_bw;
> +
> + return 0;
> +}
The only current provider aggregates "avg" with "sum" and "peak" with
"max", any particular reason to do something different? This function
doesn't even really do aggregation, if there is a second request for "0"
then the first request is lost.
I didn't find any docs but my interpretation of avg/peak is that "avg"
is for constant traffic like a display or VPU pushing pixels and "peak"
is for variable traffic like networking. I assume devices which make
"peak" requests are aggregated with max because they're not expected to
all max-out together.
In PATCH 11 you're making a bandwidth request based on resolution, that
traffic is constant and guaranteed to happend while the display is on so
it would make sense to request it as an "avg" and aggregate it with "sum".
--
Regards,
Leonard
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-08-06 13:41 UTC|newest]
Thread overview: 111+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20190723122022eucas1p2f568f74f981f9de9012eb693c3b446d5@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 00/11] Simple QoS for exynos-bus driver using interconnect Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
[not found] ` <CGME20190723122022eucas1p1266d90873d564894bd852c20140f8474@eucas1p1.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 01/11] devfreq: exynos-bus: Extract exynos_bus_profile_init() Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:07 ` Krzysztof Kozlowski
2019-07-24 19:07 ` Krzysztof Kozlowski
2019-07-31 13:00 ` Artur Świgoń
2019-07-31 13:00 ` Artur Świgoń
2019-08-05 9:56 ` Krzysztof Kozlowski
2019-08-05 9:56 ` Krzysztof Kozlowski
2019-07-25 12:43 ` Chanwoo Choi
2019-07-25 12:43 ` Chanwoo Choi
2019-07-26 10:42 ` Krzysztof Kozlowski
2019-07-26 10:42 ` Krzysztof Kozlowski
2019-07-26 10:42 ` Krzysztof Kozlowski
2019-07-26 10:52 ` Chanwoo Choi
2019-07-26 10:52 ` Chanwoo Choi
[not found] ` <CGME20190723122023eucas1p2ff56c00b60a676ed85d9fe159a1839f2@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 02/11] devfreq: exynos-bus: Extract exynos_bus_profile_init_passive() Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:08 ` Krzysztof Kozlowski
2019-07-24 19:08 ` Krzysztof Kozlowski
2019-07-24 19:08 ` Krzysztof Kozlowski
2019-07-25 12:45 ` Chanwoo Choi
2019-07-25 12:45 ` Chanwoo Choi
[not found] ` <CGME20190723122024eucas1p1ff060d072132bfbc8a8a1d10fa1f90f8@eucas1p1.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 03/11] devfreq: exynos-bus: Change goto-based logic to if-else logic Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:10 ` Krzysztof Kozlowski
2019-07-24 19:10 ` Krzysztof Kozlowski
2019-07-25 12:56 ` Chanwoo Choi
2019-07-25 12:56 ` Chanwoo Choi
2019-07-25 13:02 ` Chanwoo Choi
2019-07-25 13:02 ` Chanwoo Choi
[not found] ` <CGME20190723122024eucas1p25a480ccddaa69ee1d0f1a07960ca3f22@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 04/11] devfreq: exynos-bus: Clean up code Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:14 ` Krzysztof Kozlowski
2019-07-24 19:14 ` Krzysztof Kozlowski
2019-07-25 12:50 ` Chanwoo Choi
2019-07-25 12:50 ` Chanwoo Choi
2019-07-26 10:45 ` Krzysztof Kozlowski
2019-07-26 10:45 ` Krzysztof Kozlowski
2019-07-26 11:04 ` Chanwoo Choi
2019-07-26 11:04 ` Chanwoo Choi
[not found] ` <CGME20190723122025eucas1p251df372451e0b27ad7f2e3c89df60b64@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 05/11] icc: Export of_icc_get_from_provider() Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:15 ` Krzysztof Kozlowski
2019-07-24 19:15 ` Krzysztof Kozlowski
2019-07-24 19:15 ` Krzysztof Kozlowski
[not found] ` <CGME20190723122026eucas1p2acf705de2a47ba54f383d916f5383144@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 06/11] icc: Relax requirement in of_icc_get_from_provider() Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:16 ` Krzysztof Kozlowski
2019-07-24 19:16 ` Krzysztof Kozlowski
[not found] ` <CGME20190723122027eucas1p124f44370a63b16dcb765585761d661a3@eucas1p1.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 07/11] icc: Relax condition in apply_constraints() Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
[not found] ` <CGME20190723122027eucas1p24b1d76e3139f7cc52614d7613ff9ba98@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 08/11] arm: dts: exynos: Add parents and #interconnect-cells to Exynos4412 Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 19:24 ` Krzysztof Kozlowski
2019-07-24 19:24 ` Krzysztof Kozlowski
2019-07-24 19:24 ` Krzysztof Kozlowski
2019-07-31 13:00 ` Artur Świgoń
2019-07-31 13:00 ` Artur Świgoń
2019-07-25 13:13 ` Chanwoo Choi
2019-07-25 13:13 ` Chanwoo Choi
2019-07-26 12:02 ` Marek Szyprowski
2019-07-26 12:02 ` Marek Szyprowski
2019-07-29 1:20 ` Chanwoo Choi
2019-07-29 1:20 ` Chanwoo Choi
[not found] ` <CGME20190723122028eucas1p2eb75f35b810e71d6c590370aaff0997b@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 09/11] devfreq: exynos-bus: Add interconnect functionality to exynos-bus Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 18:36 ` Krzysztof Kozlowski
2019-07-24 18:36 ` Krzysztof Kozlowski
2019-07-31 13:01 ` Artur Świgoń
2019-07-31 13:01 ` Artur Świgoń
2019-07-26 8:05 ` Georgi Djakov
2019-07-26 8:05 ` Georgi Djakov
2019-08-01 7:59 ` Artur Świgoń
2019-08-01 7:59 ` Artur Świgoń
2019-08-07 14:21 ` Georgi Djakov
2019-08-07 14:21 ` Georgi Djakov
2019-08-08 13:28 ` Artur Świgoń
2019-08-08 13:28 ` Artur Świgoń
2019-07-29 1:52 ` Chanwoo Choi
2019-07-29 1:52 ` Chanwoo Choi
2019-08-08 13:18 ` Artur Świgoń
2019-08-08 13:18 ` Artur Świgoń
2019-08-09 2:17 ` Chanwoo Choi
2019-08-09 2:17 ` Chanwoo Choi
2019-08-08 15:00 ` Leonard Crestez
2019-08-08 15:00 ` Leonard Crestez
2019-08-08 15:00 ` Leonard Crestez
2019-08-19 23:44 ` Chanwoo Choi
2019-08-19 23:44 ` Chanwoo Choi
2019-08-19 23:44 ` Chanwoo Choi
2019-08-06 13:41 ` Leonard Crestez [this message]
2019-08-06 13:41 ` Leonard Crestez
2019-08-06 13:41 ` Leonard Crestez
2019-08-08 13:19 ` Artur Świgoń
2019-08-08 13:19 ` Artur Świgoń
2019-08-08 13:19 ` Artur Świgoń
[not found] ` <CGME20190723122029eucas1p21e1a51e759f9b605d2c89daf659af7bb@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 10/11] arm: dts: exynos: Add interconnects to Exynos4412 mixer Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
[not found] ` <CGME20190723122029eucas1p2915f536d9ef43a7bd043a878a553439f@eucas1p2.samsung.com>
2019-07-23 12:20 ` [RFC PATCH 11/11] drm: exynos: mixer: Add interconnect support Artur Świgoń
2019-07-23 12:20 ` Artur Świgoń
2019-07-24 18:52 ` Krzysztof Kozlowski
2019-07-24 18:52 ` Krzysztof Kozlowski
2019-07-24 18:53 ` [RFC PATCH 00/11] Simple QoS for exynos-bus driver using interconnect Krzysztof Kozlowski
2019-07-24 18:53 ` Krzysztof Kozlowski
2019-07-24 18:53 ` Krzysztof Kozlowski
2019-08-13 6:17 ` Chanwoo Choi
2019-08-13 6:17 ` Chanwoo Choi
2019-08-13 6:19 ` Chanwoo Choi
2019-08-13 6:19 ` Chanwoo Choi
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=VI1PR04MB5055BED59960B4F4147479AEEED50@VI1PR04MB5055.eurprd04.prod.outlook.com \
--to=leonard.crestez@nxp.com \
--cc=a.swigon@partner.samsung.com \
--cc=cw00.choi@samsung.com \
--cc=devicetree@vger.kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=georgi.djakov@linaro.org \
--cc=inki.dae@samsung.com \
--cc=krzk@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=linux-samsung-soc@vger.kernel.org \
--cc=m.szyprowski@samsung.com \
--cc=myungjoo.ham@samsung.com \
--cc=sw0312.kim@samsung.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.