[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ee5c8637-b8b2-491b-b011-e399942691dc@moroto.mountain>
Date: Wed, 1 May 2024 16:32:40 +0300
From: Dan Carpenter <dan.carpenter@...aro.org>
To: "Peng Fan (OSS)" <peng.fan@....nxp.com>
Cc: Linus Walleij <linus.walleij@...aro.org>,
Thierry Reding <thierry.reding@...il.com>,
Jonathan Hunter <jonathanh@...dia.com>,
Dvorkin Dmitry <dvorkin@...bo.com>, Wells Lu <wellslutw@...il.com>,
Maxime Coquelin <mcoquelin.stm32@...il.com>,
Alexandre Torgue <alexandre.torgue@...s.st.com>,
Emil Renner Berthing <kernel@...il.dk>,
Jianlong Huang <jianlong.huang@...rfivetech.com>,
Hal Feng <hal.feng@...rfivetech.com>,
Orson Zhai <orsonzhai@...il.com>,
Baolin Wang <baolin.wang@...ux.alibaba.com>,
Chunyan Zhang <zhang.lyra@...il.com>,
Viresh Kumar <vireshk@...nel.org>,
Shiraz Hashim <shiraz.linux.kernel@...il.com>, soc@...nel.org,
Krzysztof Kozlowski <krzk@...nel.org>,
Sylwester Nawrocki <s.nawrocki@...sung.com>,
Alim Akhtar <alim.akhtar@...sung.com>,
Geert Uytterhoeven <geert+renesas@...der.be>,
Patrice Chotard <patrice.chotard@...s.st.com>,
Heiko Stuebner <heiko@...ech.de>,
Damien Le Moal <dlemoal@...nel.org>,
Ludovic Desroches <ludovic.desroches@...rochip.com>,
Nicolas Ferre <nicolas.ferre@...rochip.com>,
Alexandre Belloni <alexandre.belloni@...tlin.com>,
Claudiu Beznea <claudiu.beznea@...on.dev>,
Dong Aisheng <aisheng.dong@....com>,
Fabio Estevam <festevam@...il.com>, Shawn Guo <shawnguo@...nel.org>,
Jacky Bai <ping.bai@....com>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
Chester Lin <chester62515@...il.com>,
Matthias Brugger <mbrugger@...e.com>,
Ghennadi Procopciuc <ghennadi.procopciuc@....nxp.com>,
Sean Wang <sean.wang@...nel.org>,
Matthias Brugger <matthias.bgg@...il.com>,
AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>,
Sascha Hauer <s.hauer@...gutronix.de>,
Andrew Jeffery <andrew@...econstruct.com.au>,
Joel Stanley <joel@....id.au>, linux-gpio@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-tegra@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org,
linux-stm32@...md-mailman.stormreply.com,
linux-samsung-soc@...r.kernel.org,
linux-renesas-soc@...r.kernel.org,
linux-rockchip@...ts.infradead.org, linux-riscv@...ts.infradead.org,
linux-mediatek@...ts.infradead.org, imx@...ts.linux.dev,
linux-aspeed@...ts.ozlabs.org, openbmc@...ts.ozlabs.org,
Peng Fan <peng.fan@....com>
Subject: Re: [PATCH 01/21] pinctrl: ti: iodelay: Use scope based
of_node_put() cleanups
On Wed, May 01, 2024 at 08:55:59PM +0800, Peng Fan (OSS) wrote:
> @@ -879,16 +874,12 @@ static int ti_iodelay_probe(struct platform_device *pdev)
> ret = pinctrl_register_and_init(&iod->desc, dev, iod, &iod->pctl);
> if (ret) {
> dev_err(dev, "Failed to register pinctrl\n");
> - goto exit_out;
> + return ret;
> }
>
> platform_set_drvdata(pdev, iod);
>
> return pinctrl_enable(iod->pctl);
> -
> -exit_out:
> - of_node_put(np);
> - return ret;
> }
This will call of_node_put() on the success path so it's a behavior
change. The original code is buggy, it's supposed to call of_node_put()
on the success path here or in ti_iodelay_remove().
If it's supposed to call of_node_put() here, then fine, this is bugfix
but if it's supposed to call it in ti_iodelay_remove() then we need to
save the pointer somewhere using no_free_ptr(). Probably saving ->np
is the safest choice?
The original code is already a little bit buggy because it doesn't
check for pinctrl_enable() errors and cleanup.
diff --git a/drivers/pinctrl/ti/pinctrl-ti-iodelay.c b/drivers/pinctrl/ti/pinctrl-ti-iodelay.c
index 040f2c46a868..f40a1476e4ff 100644
--- a/drivers/pinctrl/ti/pinctrl-ti-iodelay.c
+++ b/drivers/pinctrl/ti/pinctrl-ti-iodelay.c
@@ -156,6 +156,7 @@ struct ti_iodelay_device {
const struct ti_iodelay_reg_data *reg_data;
struct ti_iodelay_reg_values reg_init_conf_values;
+ struct device_node *np;
};
/**
@@ -884,7 +885,12 @@ static int ti_iodelay_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, iod);
- return pinctrl_enable(iod->pctl);
+ ret = pinctrl_enable(iod->pctl);
+ if (ret)
+ goto exit_out;
+
+ iod->np = no_free_ptr(np);
+ return 0;
exit_out:
of_node_put(np);
@@ -903,6 +909,7 @@ static void ti_iodelay_remove(struct platform_device *pdev)
pinctrl_unregister(iod->pctl);
ti_iodelay_pinconf_deinit_dev(iod);
+ of_node_put(iod->np);
/* Expect other allocations to be freed by devm */
}
Powered by blists - more mailing lists