From: Chen-Yu Tsai <wens@csie.org>
To: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Linus Walleij <linus.walleij@linaro.org>,
Chen-Yu Tsai <wens@csie.org>,
"linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>,
devicetree <devicetree@vger.kernel.org>,
linux-arm-kernel <linux-arm-kernel@lists.infradead.org>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 4/9] pinctrl: sunxi: Deal with configless pins
Date: Tue, 4 Oct 2016 10:28:18 +0800 [thread overview]
Message-ID: <CAGb2v654vv3B-OhHPgdLjsEOf0zPP=Ky1_+o=8mf9dZHMFjR-w@mail.gmail.com> (raw)
In-Reply-To: <8b72a134d957afdee6de48ec2c1891d8bba1b9e3.1475489558.git-series.maxime.ripard@free-electrons.com>
On Mon, Oct 3, 2016 at 6:21 PM, Maxime Ripard
<maxime.ripard@free-electrons.com> wrote:
> Even though the our binding had the assumption that the allwinner,pull and
> allwinner,drive properties were optional, the code never took that into
> account.
>
> Fix that.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
> drivers/pinctrl/sunxi/pinctrl-sunxi.c | 50 +++++++++++++++++++---------
> 1 file changed, 35 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.c b/drivers/pinctrl/sunxi/pinctrl-sunxi.c
> index 6f6f1e0011e2..cec977fcf98c 100644
> --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.c
> +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.c
> @@ -218,20 +218,29 @@ static unsigned long *sunxi_pctrl_build_pin_config(struct device_node *node,
> {
> unsigned long *pinconfig;
> unsigned int configlen = 0, idx = 0;
> + int ret;
>
> if (sunxi_pctrl_has_drive_prop(node))
> configlen++;
> if (sunxi_pctrl_has_bias_prop(node))
> configlen++;
>
> + /*
> + * If we don't have any configuration, bail out
> + */
> + if (!configlen)
> + return NULL;
> +
> pinconfig = kzalloc(configlen * sizeof(*pinconfig), GFP_KERNEL);
> if (!pinconfig)
> - return NULL;
> + return ERR_PTR(-ENOMEM);
>
> if (sunxi_pctrl_has_drive_prop(node)) {
> int drive = sunxi_pctrl_parse_drive_prop(node);
> - if (drive < 0)
> + if (drive < 0) {
> + ret = -EINVAL;
Why not just pass the error code returned from the parse function?
> goto err_free;
> + }
>
> pinconfig[idx++] = pinconf_to_config_packed(PIN_CONFIG_DRIVE_STRENGTH,
> drive);
> @@ -239,8 +248,10 @@ static unsigned long *sunxi_pctrl_build_pin_config(struct device_node *node,
>
> if (sunxi_pctrl_has_bias_prop(node)) {
> int pull = sunxi_pctrl_parse_bias_prop(node);
> - if (pull < 0)
> + if (pull < 0) {
> + ret = -EINVAL;
Same here.
> goto err_free;
> + }
>
> pinconfig[idx++] = pinconf_to_config_packed(pull, 0);
> }
> @@ -251,7 +262,7 @@ static unsigned long *sunxi_pctrl_build_pin_config(struct device_node *node,
>
> err_free:
> kfree(pinconfig);
> - return NULL;
> + return ERR_PTR(ret);
> }
>
> static int sunxi_pctrl_dt_node_to_map(struct pinctrl_dev *pctldev,
> @@ -285,7 +296,10 @@ static int sunxi_pctrl_dt_node_to_map(struct pinctrl_dev *pctldev,
>
> /*
> * We have two maps for each pin: one for the function, one
> - * for the configuration (bias, strength, etc)
> + * for the configuration (bias, strength, etc).
> + *
> + * We might be slightly overshooting, since we might not have
> + * any configuration.
> */
> nmaps = npins * 2;
> *map = kmalloc(nmaps * sizeof(struct pinctrl_map), GFP_KERNEL);
> @@ -293,8 +307,8 @@ static int sunxi_pctrl_dt_node_to_map(struct pinctrl_dev *pctldev,
> return -ENOMEM;
>
> pinconfig = sunxi_pctrl_build_pin_config(node, &configlen);
> - if (!pinconfig) {
> - ret = -EINVAL;
> + if (IS_ERR(pinconfig)) {
> + ret = PTR_ERR(pinconfig);
> goto err_free_map;
> }
>
> @@ -321,15 +335,16 @@ static int sunxi_pctrl_dt_node_to_map(struct pinctrl_dev *pctldev,
>
> i++;
>
> - (*map)[i].type = PIN_MAP_TYPE_CONFIGS_GROUP;
> - (*map)[i].data.configs.group_or_pin = group;
> - (*map)[i].data.configs.configs = pinconfig;
> - (*map)[i].data.configs.num_configs = configlen;
> -
> - i++;
> + if (pinconfig) {
> + (*map)[i].type = PIN_MAP_TYPE_CONFIGS_GROUP;
> + (*map)[i].data.configs.group_or_pin = group;
> + (*map)[i].data.configs.configs = pinconfig;
> + (*map)[i].data.configs.num_configs = configlen;
> + i++;
> + }
> }
>
> - *num_maps = nmaps;
> + *num_maps = i;
Thought: should we do a krealloc to shrink the array?
>
> return 0;
>
> @@ -342,8 +357,13 @@ static void sunxi_pctrl_dt_free_map(struct pinctrl_dev *pctldev,
> struct pinctrl_map *map,
> unsigned num_maps)
> {
> + unsigned long *pinconfig;
> +
> /* All the maps have the same pin config, free only the first one */
> - kfree(map[0].data.configs.configs);
> + pinconfig = map[0].data.configs.configs;
> + if (pinconfig)
> + kfree(pinconfig);
Passing NULL to kfree is allowed. (It becomes a no-op.)
So you could leave this function alone.
ChenYu
> +
> kfree(map);
> }
>
> --
> git-series 0.8.10
next prev parent reply other threads:[~2016-10-04 2:28 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-03 10:20 [PATCH 0/9] pinctrl: sunxi: Generic bindings rework Maxime Ripard
2016-10-03 10:20 ` [PATCH 1/9] pinctrl: sunxi: Rework the pin config building code Maxime Ripard
2016-10-03 10:21 ` [PATCH 2/9] pinctrl: sunxi: Add bindings define Maxime Ripard
2016-10-04 1:57 ` Chen-Yu Tsai
2016-10-03 10:21 ` [PATCH 4/9] pinctrl: sunxi: Deal with configless pins Maxime Ripard
2016-10-04 2:28 ` Chen-Yu Tsai [this message]
[not found] ` <CAGb2v654vv3B-OhHPgdLjsEOf0zPP=Ky1_+o=8mf9dZHMFjR-w-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-10-07 9:02 ` Maxime Ripard
2016-10-03 10:21 ` [PATCH 5/9] pinctrl: sunxi: Support generic binding Maxime Ripard
2016-10-04 2:29 ` Chen-Yu Tsai
[not found] ` <cover.26c6f76b361d146fb079d744251f020ef6d2db7d.1475489558.git-series.maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
2016-10-03 10:21 ` [PATCH 3/9] pinctrl: sunxi: Handle bias disable Maxime Ripard
[not found] ` <f4de32e1b124b25c27e34efcadc1d332cb93ee59.1475489558.git-series.maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
2016-10-04 2:14 ` Chen-Yu Tsai
2016-10-03 10:21 ` [PATCH 6/9] dt-bindings: pinctrl: Deprecate sunxi pinctrl bindings Maxime Ripard
[not found] ` <fad5a977d8ecdd43b37d7028b1bc32d8667b65a2.1475489558.git-series.maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
2016-10-04 2:32 ` Chen-Yu Tsai
[not found] ` <CAGb2v67wk_DANTyzaxTM6Stgh-u2kZXn_kjb0vbWDYUd-DUxGA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2016-10-07 9:04 ` Maxime Ripard
2016-10-03 10:21 ` [PATCH 8/9] ARM: sunxi: Remove useless allwinner,pull property Maxime Ripard
2016-10-04 2:35 ` Chen-Yu Tsai
2016-10-03 10:21 ` [PATCH 9/9] ARM: sunxi: Convert pinctrl nodes to generic bindings Maxime Ripard
2016-10-04 2:37 ` Chen-Yu Tsai
2016-10-03 10:21 ` [PATCH 7/9] ARM: sunxi: Remove useless allwinner,drive property Maxime Ripard
2016-10-04 2:34 ` Chen-Yu Tsai
2016-10-07 9:11 ` Maxime Ripard
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='CAGb2v654vv3B-OhHPgdLjsEOf0zPP=Ky1_+o=8mf9dZHMFjR-w@mail.gmail.com' \
--to=wens@csie.org \
--cc=devicetree@vger.kernel.org \
--cc=linus.walleij@linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-gpio@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maxime.ripard@free-electrons.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 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).