linux-kernel - [PATCH v3] spi: rockchip: Resolve unbalanced runtime PM / system PM handling

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20240827171126.1115748-1-briannorris@chromium.org>
Date: Tue, 27 Aug 2024 10:11:16 -0700
From: Brian Norris <briannorris@...omium.org>
To: Mark Brown <broonie@...nel.org>
Cc: Jon Lin <jon.lin@...k-chips.com>,
	Ondřej Jirman <megi@....cz>,
	Brian Norris <briannorris@...omium.org>,
	stable@...r.kernel.org,
	Heiko Stuebner <heiko@...ech.de>,
	linux-arm-kernel@...ts.infradead.org,
	linux-kernel@...r.kernel.org,
	linux-rockchip@...ts.infradead.org,
	linux-spi@...r.kernel.org,
	shengfei Xu <xsf@...k-chips.com>
Subject: [PATCH v3] spi: rockchip: Resolve unbalanced runtime PM / system PM handling

Commit e882575efc77 ("spi: rockchip: Suspend and resume the bus during
NOIRQ_SYSTEM_SLEEP_PM ops") stopped respecting runtime PM status and
simply disabled clocks unconditionally when suspending the system. This
causes problems when the device is already runtime suspended when we go
to sleep -- in which case we double-disable clocks and produce a
WARNing.

Switch back to pm_runtime_force_{suspend,resume}(), because that still
seems like the right thing to do, and the aforementioned commit makes no
explanation why it stopped using it.

Also, refactor some of the resume() error handling, because it's not
actually a good idea to re-disable clocks on failure.

Fixes: e882575efc77 ("spi: rockchip: Suspend and resume the bus during NOIRQ_SYSTEM_SLEEP_PM ops")
Cc: <stable@...r.kernel.org>
Reported-by: "Ondřej Jirman" <megi@....cz>
Closes: https://lore.kernel.org/lkml/20220621154218.sau54jeij4bunf56@core/
Signed-off-by: Brian Norris <briannorris@...omium.org>
---

Changes in v3:
 - actually CC the appropriate lists (sorry, I accidentally dropped them
   on v2)

Changes in v2:
 - fix unused 'rs' warning

 drivers/spi/spi-rockchip.c | 23 +++++++----------------
 1 file changed, 7 insertions(+), 16 deletions(-)

diff --git a/drivers/spi/spi-rockchip.c b/drivers/spi/spi-rockchip.c
index e1ecd96c7858..0bb33c43b1b4 100644
--- a/drivers/spi/spi-rockchip.c
+++ b/drivers/spi/spi-rockchip.c
@@ -945,14 +945,16 @@ static int rockchip_spi_suspend(struct device *dev)
 {
 	int ret;
 	struct spi_controller *ctlr = dev_get_drvdata(dev);
-	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	ret = spi_controller_suspend(ctlr);
 	if (ret < 0)
 		return ret;
 
-	clk_disable_unprepare(rs->spiclk);
-	clk_disable_unprepare(rs->apb_pclk);
+	ret = pm_runtime_force_suspend(dev);
+	if (ret < 0) {
+		spi_controller_resume(ctlr);
+		return ret;
+	}
 
 	pinctrl_pm_select_sleep_state(dev);
 
@@ -963,25 +965,14 @@ static int rockchip_spi_resume(struct device *dev)
 {
 	int ret;
 	struct spi_controller *ctlr = dev_get_drvdata(dev);
-	struct rockchip_spi *rs = spi_controller_get_devdata(ctlr);
 
 	pinctrl_pm_select_default_state(dev);
 
-	ret = clk_prepare_enable(rs->apb_pclk);
+	ret = pm_runtime_force_resume(dev);
 	if (ret < 0)
 		return ret;
 
-	ret = clk_prepare_enable(rs->spiclk);
-	if (ret < 0)
-		clk_disable_unprepare(rs->apb_pclk);
-
-	ret = spi_controller_resume(ctlr);
-	if (ret < 0) {
-		clk_disable_unprepare(rs->spiclk);
-		clk_disable_unprepare(rs->apb_pclk);
-	}
-
-	return 0;
+	return spi_controller_resume(ctlr);
 }
 #endif /* CONFIG_PM_SLEEP */
 
-- 
2.46.0.295.g3b9ea8a38a-goog