All of lore.kernel.org
 help / color / mirror / Atom feed
From: Viresh Kumar <viresh.kumar@linaro.org>
To: Steven Price <steven.price@arm.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>,
	Chanwoo Choi <cw00.choi@samsung.com>,
	MyungJoo Ham <myungjoo.ham@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>,
	Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	Qiang Yu <yuq825@gmail.com>, Rob Herring <robh@kernel.org>,
	Tomeu Vizoso <tomeu.vizoso@collabora.com>,
	Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>,
	Nishanth Menon <nm@ti.com>, Stephen Boyd <sboyd@kernel.org>,
	Thierry Reding <thierry.reding@gmail.com>,
	Jonathan Hunter <jonathanh@nvidia.com>,
	linux-pm@vger.kernel.org,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org, linux-samsung-soc@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	dri-devel@lists.freedesktop.org, lima@lists.freedesktop.org,
	linux-tegra@vger.kernel.org
Subject: Re: [PATCH V3 02/20] OPP: Make dev_pm_opp_set_regulators() accept NULL terminated list
Date: Tue, 5 Jul 2022 10:04:39 +0530	[thread overview]
Message-ID: <20220705043439.xlrxusxrhwjupiyt@vireshk-i7> (raw)
In-Reply-To: <48d865e8-6c0d-99c0-a43b-89793d5c3f85@arm.com>

On 04-07-22, 15:35, Steven Price wrote:
> I have to say the 'new improved' list ending with NULL approach doesn't
> work out so well for Panfrost. We already have to have a separate
> 'num_supplies' variable for devm_regulator_bulk_get() /
> regulator_bulk_{en,dis}able(), so the keeping everything in sync
> argument is lost here.
> 
> I would suggest added the NULL on the end of the lists in panfrost_drv.c
> but then it would break the use of ARRAY_SIZE() to automagically keep
> the length correct...

Actually we can still make it work.

> For now the approach isn't too bad because Panfrost doesn't yet support
> enabling devfreq with more than one supply. But that array isn't going
> to work so nicely when that restriction is removed.
> 
> The only sane way I can see of handling this in Panfrost would be
> replicating the loop to count the supplies in the Panfrost code which
> would allow dropping num_supplies from struct panfrost_compatible and
> then supply_names in the same struct could be NULL terminated ready for
> devm_pm_opp_set_regulators().

Or doing this, which will simplify both the cases.

diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
index 7fcbc2a5b6cd..b3b55565b8ef 100644
--- a/drivers/gpu/drm/panfrost/panfrost_drv.c
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -625,24 +625,29 @@ static int panfrost_remove(struct platform_device *pdev)
        return 0;
 }
 
-static const char * const default_supplies[] = { "mali" };
+/*
+ * The OPP core wants the supply names to be NULL terminated, but we need the
+ * correct num_supplies value for regulator core. Hence, we NULL terminate here
+ * and then initialize num_supplies with ARRAY_SIZE - 1.
+ */
+static const char * const default_supplies[] = { "mali", NULL };
 static const struct panfrost_compatible default_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .num_pm_domains = 1, /* optional */
        .pm_domain_names = NULL,
 };
 
 static const struct panfrost_compatible amlogic_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .vendor_quirk = panfrost_gpu_amlogic_quirk,
 };
 
-static const char * const mediatek_mt8183_supplies[] = { "mali", "sram" };
+static const char * const mediatek_mt8183_supplies[] = { "mali", "sram", NULL };
 static const char * const mediatek_mt8183_pm_domains[] = { "core0", "core1", "core2" };
 static const struct panfrost_compatible mediatek_mt8183_data = {
-       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies),
+       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies) - 1,
        .supply_names = mediatek_mt8183_supplies,
        .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains),
        .pm_domain_names = mediatek_mt8183_pm_domains,

-- 
viresh

WARNING: multiple messages have this Message-ID (diff)
From: Viresh Kumar <viresh.kumar@linaro.org>
To: Steven Price <steven.price@arm.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>,
	Chanwoo Choi <cw00.choi@samsung.com>,
	MyungJoo Ham <myungjoo.ham@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>,
	Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	Qiang Yu <yuq825@gmail.com>, Rob Herring <robh@kernel.org>,
	Tomeu Vizoso <tomeu.vizoso@collabora.com>,
	Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>,
	Nishanth Menon <nm@ti.com>, Stephen Boyd <sboyd@kernel.org>,
	Thierry Reding <thierry.reding@gmail.com>,
	Jonathan Hunter <jonathanh@nvidia.com>,
	linux-pm@vger.kernel.org,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org, linux-samsung-soc@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	dri-devel@lists.freedesktop.org, lima@lists.freedesktop.org,
	linux-tegra@vger.kernel.org
Subject: Re: [PATCH V3 02/20] OPP: Make dev_pm_opp_set_regulators() accept NULL terminated list
Date: Tue, 5 Jul 2022 10:04:39 +0530	[thread overview]
Message-ID: <20220705043439.xlrxusxrhwjupiyt@vireshk-i7> (raw)
In-Reply-To: <48d865e8-6c0d-99c0-a43b-89793d5c3f85@arm.com>

On 04-07-22, 15:35, Steven Price wrote:
> I have to say the 'new improved' list ending with NULL approach doesn't
> work out so well for Panfrost. We already have to have a separate
> 'num_supplies' variable for devm_regulator_bulk_get() /
> regulator_bulk_{en,dis}able(), so the keeping everything in sync
> argument is lost here.
> 
> I would suggest added the NULL on the end of the lists in panfrost_drv.c
> but then it would break the use of ARRAY_SIZE() to automagically keep
> the length correct...

Actually we can still make it work.

> For now the approach isn't too bad because Panfrost doesn't yet support
> enabling devfreq with more than one supply. But that array isn't going
> to work so nicely when that restriction is removed.
> 
> The only sane way I can see of handling this in Panfrost would be
> replicating the loop to count the supplies in the Panfrost code which
> would allow dropping num_supplies from struct panfrost_compatible and
> then supply_names in the same struct could be NULL terminated ready for
> devm_pm_opp_set_regulators().

Or doing this, which will simplify both the cases.

diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
index 7fcbc2a5b6cd..b3b55565b8ef 100644
--- a/drivers/gpu/drm/panfrost/panfrost_drv.c
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -625,24 +625,29 @@ static int panfrost_remove(struct platform_device *pdev)
        return 0;
 }
 
-static const char * const default_supplies[] = { "mali" };
+/*
+ * The OPP core wants the supply names to be NULL terminated, but we need the
+ * correct num_supplies value for regulator core. Hence, we NULL terminate here
+ * and then initialize num_supplies with ARRAY_SIZE - 1.
+ */
+static const char * const default_supplies[] = { "mali", NULL };
 static const struct panfrost_compatible default_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .num_pm_domains = 1, /* optional */
        .pm_domain_names = NULL,
 };
 
 static const struct panfrost_compatible amlogic_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .vendor_quirk = panfrost_gpu_amlogic_quirk,
 };
 
-static const char * const mediatek_mt8183_supplies[] = { "mali", "sram" };
+static const char * const mediatek_mt8183_supplies[] = { "mali", "sram", NULL };
 static const char * const mediatek_mt8183_pm_domains[] = { "core0", "core1", "core2" };
 static const struct panfrost_compatible mediatek_mt8183_data = {
-       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies),
+       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies) - 1,
        .supply_names = mediatek_mt8183_supplies,
        .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains),
        .pm_domain_names = mediatek_mt8183_pm_domains,

-- 
viresh

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

WARNING: multiple messages have this Message-ID (diff)
From: Viresh Kumar <viresh.kumar@linaro.org>
To: Steven Price <steven.price@arm.com>
Cc: Nishanth Menon <nm@ti.com>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	dri-devel@lists.freedesktop.org,
	Thierry Reding <thierry.reding@gmail.com>,
	Alim Akhtar <alim.akhtar@samsung.com>,
	linux-samsung-soc@vger.kernel.org,
	Vincent Guittot <vincent.guittot@linaro.org>,
	lima@lists.freedesktop.org,
	Kyungmin Park <kyungmin.park@samsung.com>,
	Jonathan Hunter <jonathanh@nvidia.com>,
	Chanwoo Choi <cw00.choi@samsung.com>,
	MyungJoo Ham <myungjoo.ham@samsung.com>,
	Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>,
	linux-pm@vger.kernel.org, linux-tegra@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	Tomeu Vizoso <tomeu.vizoso@collabora.com>,
	Stephen Boyd <sboyd@kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org,
	Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>,
	Qiang Yu <yuq825@gmail.com>
Subject: Re: [PATCH V3 02/20] OPP: Make dev_pm_opp_set_regulators() accept NULL terminated list
Date: Tue, 5 Jul 2022 10:04:39 +0530	[thread overview]
Message-ID: <20220705043439.xlrxusxrhwjupiyt@vireshk-i7> (raw)
In-Reply-To: <48d865e8-6c0d-99c0-a43b-89793d5c3f85@arm.com>

On 04-07-22, 15:35, Steven Price wrote:
> I have to say the 'new improved' list ending with NULL approach doesn't
> work out so well for Panfrost. We already have to have a separate
> 'num_supplies' variable for devm_regulator_bulk_get() /
> regulator_bulk_{en,dis}able(), so the keeping everything in sync
> argument is lost here.
> 
> I would suggest added the NULL on the end of the lists in panfrost_drv.c
> but then it would break the use of ARRAY_SIZE() to automagically keep
> the length correct...

Actually we can still make it work.

> For now the approach isn't too bad because Panfrost doesn't yet support
> enabling devfreq with more than one supply. But that array isn't going
> to work so nicely when that restriction is removed.
> 
> The only sane way I can see of handling this in Panfrost would be
> replicating the loop to count the supplies in the Panfrost code which
> would allow dropping num_supplies from struct panfrost_compatible and
> then supply_names in the same struct could be NULL terminated ready for
> devm_pm_opp_set_regulators().

Or doing this, which will simplify both the cases.

diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c
index 7fcbc2a5b6cd..b3b55565b8ef 100644
--- a/drivers/gpu/drm/panfrost/panfrost_drv.c
+++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
@@ -625,24 +625,29 @@ static int panfrost_remove(struct platform_device *pdev)
        return 0;
 }
 
-static const char * const default_supplies[] = { "mali" };
+/*
+ * The OPP core wants the supply names to be NULL terminated, but we need the
+ * correct num_supplies value for regulator core. Hence, we NULL terminate here
+ * and then initialize num_supplies with ARRAY_SIZE - 1.
+ */
+static const char * const default_supplies[] = { "mali", NULL };
 static const struct panfrost_compatible default_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .num_pm_domains = 1, /* optional */
        .pm_domain_names = NULL,
 };
 
 static const struct panfrost_compatible amlogic_data = {
-       .num_supplies = ARRAY_SIZE(default_supplies),
+       .num_supplies = ARRAY_SIZE(default_supplies) - 1,
        .supply_names = default_supplies,
        .vendor_quirk = panfrost_gpu_amlogic_quirk,
 };
 
-static const char * const mediatek_mt8183_supplies[] = { "mali", "sram" };
+static const char * const mediatek_mt8183_supplies[] = { "mali", "sram", NULL };
 static const char * const mediatek_mt8183_pm_domains[] = { "core0", "core1", "core2" };
 static const struct panfrost_compatible mediatek_mt8183_data = {
-       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies),
+       .num_supplies = ARRAY_SIZE(mediatek_mt8183_supplies) - 1,
        .supply_names = mediatek_mt8183_supplies,
        .num_pm_domains = ARRAY_SIZE(mediatek_mt8183_pm_domains),
        .pm_domain_names = mediatek_mt8183_pm_domains,

-- 
viresh

  reply	other threads:[~2022-07-05  4:34 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-04 12:07 [PATCH V3 00/20] OPP: Add new configuration interface: dev_pm_opp_set_config() Viresh Kumar
2022-07-04 12:07 ` Viresh Kumar
2022-07-04 12:07 ` Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 01/20] OPP: Track if clock name is configured by platform Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 02/20] OPP: Make dev_pm_opp_set_regulators() accept NULL terminated list Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 14:35   ` Steven Price
2022-07-04 14:35     ` Steven Price
2022-07-04 14:35     ` Steven Price
2022-07-05  4:34     ` Viresh Kumar [this message]
2022-07-05  4:34       ` Viresh Kumar
2022-07-05  4:34       ` Viresh Kumar
2022-07-06  8:09       ` Steven Price
2022-07-06  8:09         ` Steven Price
2022-07-06  8:09         ` Steven Price
2022-07-06  8:18   ` [PATCH V3.1 " Viresh Kumar
2022-07-06  8:18     ` Viresh Kumar
2022-07-06  8:18     ` Viresh Kumar
2022-07-06  9:50     ` Steven Price
2022-07-06  9:50       ` Steven Price
2022-07-06  9:50       ` Steven Price
2022-07-07 19:04   ` [PATCH V3 " Chanwoo Choi
2022-07-07 19:04     ` Chanwoo Choi
2022-07-07 19:04     ` Chanwoo Choi
2022-07-04 12:07 ` [PATCH V3 03/20] OPP: Add dev_pm_opp_set_config() and friends Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 04/20] cpufreq: qcom-nvmem: Migrate to dev_pm_opp_set_config() Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 05/20] cpufreq: sti: " Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 06/20] cpufreq: ti: " Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 07/20] drm/lima: " Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 08/20] soc/tegra: Add comment over devm_pm_opp_set_clkname() Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 09/20] soc/tegra: Migrate to dev_pm_opp_set_config() Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 10/20] OPP: Migrate set-regulators API to use set-config helpers Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-07 19:29   ` Chanwoo Choi
2022-07-07 19:29     ` Chanwoo Choi
2022-07-04 12:07 ` [PATCH V3 11/20] OPP: Migrate set-supported-hw " Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 12/20] OPP: Migrate set-clk-name " Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 13/20] OPP: Migrate set-opp-helper " Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 14/20] OPP: Migrate attach-genpd " Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 15/20] OPP: Migrate set-prop-name helper " Viresh Kumar
2022-07-04 12:07   ` Viresh Kumar
2022-07-04 18:59   ` Samuel Holland
2022-07-04 18:59     ` Samuel Holland
2022-07-04 12:07 ` [PATCH V3 16/20] OPP: Add support for config_regulators() helper Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 17/20] OPP: Make _generic_set_opp_regulator() a config_regulators() interface Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 18/20] OPP: Add dev_pm_opp_get_supplies() Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 19/20] OPP: ti: Migrate to dev_pm_opp_set_config_regulators() Viresh Kumar
2022-07-04 12:07 ` [PATCH V3 20/20] OPP: Remove custom OPP helper support Viresh Kumar

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=20220705043439.xlrxusxrhwjupiyt@vireshk-i7 \
    --to=viresh.kumar@linaro.org \
    --cc=alim.akhtar@samsung.com \
    --cc=alyssa.rosenzweig@collabora.com \
    --cc=cw00.choi@samsung.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=jonathanh@nvidia.com \
    --cc=krzysztof.kozlowski@linaro.org \
    --cc=kyungmin.park@samsung.com \
    --cc=lima@lists.freedesktop.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=linux-tegra@vger.kernel.org \
    --cc=myungjoo.ham@samsung.com \
    --cc=nm@ti.com \
    --cc=rafael@kernel.org \
    --cc=robh@kernel.org \
    --cc=sboyd@kernel.org \
    --cc=steven.price@arm.com \
    --cc=thierry.reding@gmail.com \
    --cc=tomeu.vizoso@collabora.com \
    --cc=vincent.guittot@linaro.org \
    --cc=yuq825@gmail.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.