From: Linus Walleij <linus.walleij@linaro.org>
To: linux-gpio@vger.kernel.org
Cc: Linus Walleij <linus.walleij@linaro.org>,
"Rafael J . Wysocki" <rafael.j.wysocki@intel.com>,
Ulf Hansson <ulf.hansson@linaro.org>
Subject: [PATCH] pinctrl: Enforce device links
Date: Thu, 12 Dec 2019 11:11:34 +0100 [thread overview]
Message-ID: <20191212101134.42420-1-linus.walleij@linaro.org> (raw)
Instead of opt-in device links, enforce it across the board.
Everyone probably needs this anyway, lest runtime[_pm] suspend
order will be haphazard.
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
As there is no progress on opting in drivers for this we can just
enforce it.
I am a bit concerned that with every pin control state change the
link reference count will just go up, but does it really matter?
---
drivers/pinctrl/core.c | 25 ++++++++++++++-----------
drivers/pinctrl/pinctrl-stmfx.c | 1 -
drivers/pinctrl/stm32/pinctrl-stm32.c | 1 -
include/linux/pinctrl/pinctrl.h | 5 -----
4 files changed, 14 insertions(+), 18 deletions(-)
diff --git a/drivers/pinctrl/core.c b/drivers/pinctrl/core.c
index 2bbd8ee93507..1d2cdeebb316 100644
--- a/drivers/pinctrl/core.c
+++ b/drivers/pinctrl/core.c
@@ -1220,15 +1220,6 @@ struct pinctrl_state *pinctrl_lookup_state(struct pinctrl *p,
}
EXPORT_SYMBOL_GPL(pinctrl_lookup_state);
-static void pinctrl_link_add(struct pinctrl_dev *pctldev,
- struct device *consumer)
-{
- if (pctldev->desc->link_consumers)
- device_link_add(consumer, pctldev->dev,
- DL_FLAG_PM_RUNTIME |
- DL_FLAG_AUTOREMOVE_CONSUMER);
-}
-
/**
* pinctrl_commit_state() - select/activate/program a pinctrl state to HW
* @p: the pinctrl handle for the device that requests configuration
@@ -1276,8 +1267,20 @@ static int pinctrl_commit_state(struct pinctrl *p, struct pinctrl_state *state)
}
/* Do not link hogs (circular dependency) */
- if (p != setting->pctldev->p)
- pinctrl_link_add(setting->pctldev, p->dev);
+ if (p != setting->pctldev->p) {
+ /*
+ * Create a device link to the consumer such that
+ * it will enforce that runtime PM suspend/resume
+ * is done first on consumers before we get to
+ * the pin controller itself. As some devices get
+ * their pin control state even before probe() it is
+ * important to use DL_FLAG_AUTOREMOVE_CONSUMER.
+ */
+ device_link_add(p->dev,
+ setting->pctldev->dev,
+ DL_FLAG_PM_RUNTIME |
+ DL_FLAG_AUTOREMOVE_CONSUMER);
+ }
}
p->state = state;
diff --git a/drivers/pinctrl/pinctrl-stmfx.c b/drivers/pinctrl/pinctrl-stmfx.c
index 16723797fa7c..4306b8444188 100644
--- a/drivers/pinctrl/pinctrl-stmfx.c
+++ b/drivers/pinctrl/pinctrl-stmfx.c
@@ -638,7 +638,6 @@ static int stmfx_pinctrl_probe(struct platform_device *pdev)
pctl->pctl_desc.pins = stmfx_pins;
pctl->pctl_desc.npins = ARRAY_SIZE(stmfx_pins);
pctl->pctl_desc.owner = THIS_MODULE;
- pctl->pctl_desc.link_consumers = true;
ret = devm_pinctrl_register_and_init(pctl->dev, &pctl->pctl_desc,
pctl, &pctl->pctl_dev);
diff --git a/drivers/pinctrl/stm32/pinctrl-stm32.c b/drivers/pinctrl/stm32/pinctrl-stm32.c
index 2d5e0435af0a..ec59a58600ce 100644
--- a/drivers/pinctrl/stm32/pinctrl-stm32.c
+++ b/drivers/pinctrl/stm32/pinctrl-stm32.c
@@ -1439,7 +1439,6 @@ int stm32_pctl_probe(struct platform_device *pdev)
pctl->pctl_desc.owner = THIS_MODULE;
pctl->pctl_desc.pins = pins;
pctl->pctl_desc.npins = pctl->npins;
- pctl->pctl_desc.link_consumers = true;
pctl->pctl_desc.confops = &stm32_pconf_ops;
pctl->pctl_desc.pctlops = &stm32_pctrl_ops;
pctl->pctl_desc.pmxops = &stm32_pmx_ops;
diff --git a/include/linux/pinctrl/pinctrl.h b/include/linux/pinctrl/pinctrl.h
index 7ce23450a1cb..c6159f041f4e 100644
--- a/include/linux/pinctrl/pinctrl.h
+++ b/include/linux/pinctrl/pinctrl.h
@@ -122,10 +122,6 @@ struct pinctrl_ops {
* the hardware description
* @custom_conf_items: Information how to print @params in debugfs, must be
* the same size as the @custom_params, i.e. @num_custom_params
- * @link_consumers: If true create a device link between pinctrl and its
- * consumers (i.e. the devices requesting pin control states). This is
- * sometimes necessary to ascertain the right suspend/resume order for
- * example.
*/
struct pinctrl_desc {
const char *name;
@@ -140,7 +136,6 @@ struct pinctrl_desc {
const struct pinconf_generic_params *custom_params;
const struct pin_config_item *custom_conf_items;
#endif
- bool link_consumers;
};
/* External interface to pin controller */
--
2.23.0
next reply other threads:[~2019-12-12 10:11 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-12 10:11 Linus Walleij [this message]
2019-12-12 10:56 ` [PATCH] pinctrl: Enforce device links Ulf Hansson
2019-12-12 13:19 ` Benjamin GAIGNARD
2019-12-12 13:47 ` Ulf Hansson
2019-12-12 15:29 ` Benjamin GAIGNARD
-- strict thread matches above, loose matches on Subject: below --
2019-12-13 13:20 Linus Walleij
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=20191212101134.42420-1-linus.walleij@linaro.org \
--to=linus.walleij@linaro.org \
--cc=linux-gpio@vger.kernel.org \
--cc=rafael.j.wysocki@intel.com \
--cc=ulf.hansson@linaro.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;
as well as URLs for NNTP newsgroup(s).