All of lore.kernel.org
 help / color / mirror / Atom feed
From: Simon South <simon@simonsouth.net>
To: tpiepho@gmail.com, thierry.reding@gmail.com,
	u.kleine-koenig@pengutronix.de, robin.murphy@arm.com,
	lee.jones@linaro.org, heiko@sntech.de, bbrezillon@kernel.org,
	david.wu@rock-chips.com, steven.liu@rock-chips.com,
	linux-pwm@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-rockchip@lists.infradead.org
Cc: simon@simonsouth.net
Subject: [PATCH v4 4/5] pwm: rockchip: Eliminate potential race condition when probing
Date: Tue, 19 Jan 2021 11:12:08 -0500	[thread overview]
Message-ID: <4bf623acf2cae5247f60a6e9765f0755132647e4.1610976129.git.simon@simonsouth.net> (raw)
In-Reply-To: <cover.1610976129.git.simon@simonsouth.net>

Commit 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running
PWMs") introduced a potential race condition in rockchip_pwm_probe(): A
consumer could enable an inactive PWM, or disable a running one, between
rockchip_pwm_probe() registering the device via pwmchip_add() and checking
whether it is enabled (to determine whether it was started by a
bootloader). This could result in a device's PWM clock being either enabled
once more than necessary, potentially causing it to continue running when
no longer needed, or disabled once more than necessary, producing a warning
from the kernel.

Eliminate these possibilities by modifying rockchip_pwm_probe() so it
checks whether a device is enabled before registering it rather than after.

Fixes: 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running PWMs")
Reported-by: Trent Piepho <tpiepho@gmail.com>
Signed-off-by: Simon South <simon@simonsouth.net>
---
 drivers/pwm/pwm-rockchip.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/pwm/pwm-rockchip.c b/drivers/pwm/pwm-rockchip.c
index b5bab427b5de..228147e9bc6e 100644
--- a/drivers/pwm/pwm-rockchip.c
+++ b/drivers/pwm/pwm-rockchip.c
@@ -288,6 +288,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	const struct of_device_id *id;
 	struct rockchip_pwm_chip *pc;
 	u32 enable_conf, ctrl;
+	bool enabled;
 	int ret, count;
 
 	id = of_match_device(rockchip_pwm_dt_ids, &pdev->dev);
@@ -349,6 +350,10 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 		pc->chip.of_pwm_n_cells = 3;
 	}
 
+	enable_conf = pc->data->enable_conf;
+	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
+	enabled = (ctrl & enable_conf) == enable_conf;
+
 	ret = pwmchip_add(&pc->chip);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret);
@@ -356,9 +361,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	}
 
 	/* Keep the PWM clk enabled if the PWM appears to be up and running. */
-	enable_conf = pc->data->enable_conf;
-	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
-	if ((ctrl & enable_conf) != enable_conf)
+	if (!enabled)
 		clk_disable(pc->clk);
 
 	clk_disable(pc->pclk);
-- 
2.30.0


WARNING: multiple messages have this Message-ID (diff)
From: Simon South <simon@simonsouth.net>
To: tpiepho@gmail.com, thierry.reding@gmail.com,
	u.kleine-koenig@pengutronix.de, robin.murphy@arm.com,
	lee.jones@linaro.org, heiko@sntech.de, bbrezillon@kernel.org,
	david.wu@rock-chips.com, steven.liu@rock-chips.com,
	linux-pwm@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-rockchip@lists.infradead.org
Cc: simon@simonsouth.net
Subject: [PATCH v4 4/5] pwm: rockchip: Eliminate potential race condition when probing
Date: Tue, 19 Jan 2021 11:12:08 -0500	[thread overview]
Message-ID: <4bf623acf2cae5247f60a6e9765f0755132647e4.1610976129.git.simon@simonsouth.net> (raw)
In-Reply-To: <cover.1610976129.git.simon@simonsouth.net>

Commit 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running
PWMs") introduced a potential race condition in rockchip_pwm_probe(): A
consumer could enable an inactive PWM, or disable a running one, between
rockchip_pwm_probe() registering the device via pwmchip_add() and checking
whether it is enabled (to determine whether it was started by a
bootloader). This could result in a device's PWM clock being either enabled
once more than necessary, potentially causing it to continue running when
no longer needed, or disabled once more than necessary, producing a warning
from the kernel.

Eliminate these possibilities by modifying rockchip_pwm_probe() so it
checks whether a device is enabled before registering it rather than after.

Fixes: 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running PWMs")
Reported-by: Trent Piepho <tpiepho@gmail.com>
Signed-off-by: Simon South <simon@simonsouth.net>
---
 drivers/pwm/pwm-rockchip.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/pwm/pwm-rockchip.c b/drivers/pwm/pwm-rockchip.c
index b5bab427b5de..228147e9bc6e 100644
--- a/drivers/pwm/pwm-rockchip.c
+++ b/drivers/pwm/pwm-rockchip.c
@@ -288,6 +288,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	const struct of_device_id *id;
 	struct rockchip_pwm_chip *pc;
 	u32 enable_conf, ctrl;
+	bool enabled;
 	int ret, count;
 
 	id = of_match_device(rockchip_pwm_dt_ids, &pdev->dev);
@@ -349,6 +350,10 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 		pc->chip.of_pwm_n_cells = 3;
 	}
 
+	enable_conf = pc->data->enable_conf;
+	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
+	enabled = (ctrl & enable_conf) == enable_conf;
+
 	ret = pwmchip_add(&pc->chip);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret);
@@ -356,9 +361,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	}
 
 	/* Keep the PWM clk enabled if the PWM appears to be up and running. */
-	enable_conf = pc->data->enable_conf;
-	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
-	if ((ctrl & enable_conf) != enable_conf)
+	if (!enabled)
 		clk_disable(pc->clk);
 
 	clk_disable(pc->pclk);
-- 
2.30.0


_______________________________________________
Linux-rockchip mailing list
Linux-rockchip@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-rockchip

WARNING: multiple messages have this Message-ID (diff)
From: Simon South <simon@simonsouth.net>
To: tpiepho@gmail.com, thierry.reding@gmail.com,
	u.kleine-koenig@pengutronix.de, robin.murphy@arm.com,
	lee.jones@linaro.org, heiko@sntech.de, bbrezillon@kernel.org,
	david.wu@rock-chips.com, steven.liu@rock-chips.com,
	linux-pwm@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-rockchip@lists.infradead.org
Cc: simon@simonsouth.net
Subject: [PATCH v4 4/5] pwm: rockchip: Eliminate potential race condition when probing
Date: Tue, 19 Jan 2021 11:12:08 -0500	[thread overview]
Message-ID: <4bf623acf2cae5247f60a6e9765f0755132647e4.1610976129.git.simon@simonsouth.net> (raw)
In-Reply-To: <cover.1610976129.git.simon@simonsouth.net>

Commit 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running
PWMs") introduced a potential race condition in rockchip_pwm_probe(): A
consumer could enable an inactive PWM, or disable a running one, between
rockchip_pwm_probe() registering the device via pwmchip_add() and checking
whether it is enabled (to determine whether it was started by a
bootloader). This could result in a device's PWM clock being either enabled
once more than necessary, potentially causing it to continue running when
no longer needed, or disabled once more than necessary, producing a warning
from the kernel.

Eliminate these possibilities by modifying rockchip_pwm_probe() so it
checks whether a device is enabled before registering it rather than after.

Fixes: 48cf973cae33 ("pwm: rockchip: Avoid glitches on already running PWMs")
Reported-by: Trent Piepho <tpiepho@gmail.com>
Signed-off-by: Simon South <simon@simonsouth.net>
---
 drivers/pwm/pwm-rockchip.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/pwm/pwm-rockchip.c b/drivers/pwm/pwm-rockchip.c
index b5bab427b5de..228147e9bc6e 100644
--- a/drivers/pwm/pwm-rockchip.c
+++ b/drivers/pwm/pwm-rockchip.c
@@ -288,6 +288,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	const struct of_device_id *id;
 	struct rockchip_pwm_chip *pc;
 	u32 enable_conf, ctrl;
+	bool enabled;
 	int ret, count;
 
 	id = of_match_device(rockchip_pwm_dt_ids, &pdev->dev);
@@ -349,6 +350,10 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 		pc->chip.of_pwm_n_cells = 3;
 	}
 
+	enable_conf = pc->data->enable_conf;
+	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
+	enabled = (ctrl & enable_conf) == enable_conf;
+
 	ret = pwmchip_add(&pc->chip);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "pwmchip_add() failed: %d\n", ret);
@@ -356,9 +361,7 @@ static int rockchip_pwm_probe(struct platform_device *pdev)
 	}
 
 	/* Keep the PWM clk enabled if the PWM appears to be up and running. */
-	enable_conf = pc->data->enable_conf;
-	ctrl = readl_relaxed(pc->base + pc->data->regs.ctrl);
-	if ((ctrl & enable_conf) != enable_conf)
+	if (!enabled)
 		clk_disable(pc->clk);
 
 	clk_disable(pc->pclk);
-- 
2.30.0


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

  parent reply	other threads:[~2021-01-19 18:27 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-19 16:12 [PATCH v4 0/5] pwm: rockchip: Eliminate potential race condition when probing Simon South
2021-01-19 16:12 ` Simon South
2021-01-19 16:12 ` Simon South
2021-01-19 16:12 ` [PATCH v4 1/5] pwm: rockchip: Enable APB clock during register access while probing Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12 ` [PATCH v4 2/5] pwm: rockchip: rockchip_pwm_probe(): Remove superfluous clk_unprepare() Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12 ` [PATCH v4 3/5] pwm: rockchip: Replace "bus clk" with "PWM clk" Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12 ` Simon South [this message]
2021-01-19 16:12   ` [PATCH v4 4/5] pwm: rockchip: Eliminate potential race condition when probing Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12 ` [PATCH v4 5/5] pwm: rockchip: Enable clock before calling clk_get_rate() Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 16:12   ` Simon South
2021-01-19 19:22   ` Uwe Kleine-König
2021-01-19 19:22     ` Uwe Kleine-König
2021-01-19 19:22     ` Uwe Kleine-König

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=4bf623acf2cae5247f60a6e9765f0755132647e4.1610976129.git.simon@simonsouth.net \
    --to=simon@simonsouth.net \
    --cc=bbrezillon@kernel.org \
    --cc=david.wu@rock-chips.com \
    --cc=heiko@sntech.de \
    --cc=lee.jones@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-pwm@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=robin.murphy@arm.com \
    --cc=steven.liu@rock-chips.com \
    --cc=thierry.reding@gmail.com \
    --cc=tpiepho@gmail.com \
    --cc=u.kleine-koenig@pengutronix.de \
    /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.