From: Krzysztof Kozlowski <krzk@kernel.org>
To: Aman Kumar Pandey <aman.kumarpandey@nxp.com>,
linux-kernel@vger.kernel.org, linux-i3c@lists.infradead.org,
alexandre.belloni@bootlin.com, krzk+dt@kernel.org,
robh@kernel.org, conor+dt@kernel.org, devicetree@vger.kernel.org
Cc: vikash.bansal@nxp.com, priyanka.jain@nxp.com,
shashank.rebbapragada@nxp.com, Frank.Li@nxp.com
Subject: Re: [PATCH 2/2] drivers: i3c: Add driver for NXP P3H2x4x i3c-hub device
Date: Wed, 12 Feb 2025 17:53:34 +0100 [thread overview]
Message-ID: <76c991dd-f99b-497c-9d68-9e08541be4f0@kernel.org> (raw)
In-Reply-To: <20250212132227.1348374-2-aman.kumarpandey@nxp.com>
On 12/02/2025 14:22, Aman Kumar Pandey wrote:
> +
> +static void p3h2x4x_of_get_tp_dt_conf(struct device *dev,
> + const struct device_node *node)
> +{
> + struct p3h2x4x *priv = dev_get_drvdata(dev);
> + struct device_node *dev_node;
> + int tp_port;
> +
> + for_each_available_child_of_node(node, dev_node) {
> + if (!dev_node->full_name ||
> + (sscanf(dev_node->full_name, "target-port@%d", &tp_port) != 1))
NAK, undocumented ABI. Also weird code, why are you parsing DT structure
manually? Use proper functions to get the reg.
> + continue;
> +
> + if (tp_port < P3H2x4x_TP_MAX_COUNT) {
> + priv->tp_bus[tp_port].dt_available = true;
> + priv->tp_bus[tp_port].of_node = dev_node;
> + priv->tp_bus[tp_port].tp_mask = BIT(tp_port);
> + priv->tp_bus[tp_port].priv = priv;
> + priv->tp_bus[tp_port].tp_port = tp_port;
> + }
> + }
> +}
> +
> +/* return true when backend node exist */
> +static bool p3h2x4x_is_backend_node_exist(int port, struct p3h2x4x *priv, u32 addr)
> +{
> + struct smbus_device *backend = NULL;
> +
> + list_for_each_entry(backend,
> + &priv->tp_bus[port].tp_device_entry, list) {
> + if (backend->addr == addr)
> + return true;
> + }
> + return false;
> +}
> +
> +static int p3h2x4x_read_backend_from_p3h2x4x_dts(struct device_node *i3c_node_target,
> + struct p3h2x4x *priv)
> +{
> + struct device_node *i3c_node_tp;
> + const char *compatible;
> + int tp_port, ret;
> + u32 addr_dts;
> + struct smbus_device *backend;
> +
> + if (sscanf(i3c_node_target->full_name, "target-port@%d", &tp_port) == 0)
> + return -EINVAL;
> +
> + if (tp_port > P3H2x4x_TP_MAX_COUNT)
> + return -ERANGE;
> +
> + if (tp_port < 0)
> + return -EINVAL;
> +
> + INIT_LIST_HEAD(&priv->tp_bus[tp_port].tp_device_entry);
> +
> + if (priv->settings.tp[tp_port].mode == P3H2x4x_TP_MODE_I3C)
> + return 0;
> +
> + for_each_available_child_of_node(i3c_node_target, i3c_node_tp) {
> +
> + ret = of_property_read_u32(i3c_node_tp, "reg", &addr_dts);
> + if (ret)
> + return ret;
> +
> + if (p3h2x4x_is_backend_node_exist(tp_port, priv, addr_dts))
> + continue;
> +
> + ret = of_property_read_string(i3c_node_tp, "compatible", &compatible);
> + if (ret)
> + return ret;
> +
> + backend = kzalloc(sizeof(*backend), GFP_KERNEL);
> + if (!backend)
> + return -ENOMEM;
> +
> + backend->addr = addr_dts;
> + backend->compatible = compatible;
> + backend->tp_device_dt_node = i3c_node_tp;
> + backend->client = NULL;
> +
> + list_add(&backend->list,
> + &priv->tp_bus[tp_port].tp_device_entry);
> + }
> +
> + return 0;
> +}
> +
> +static void p3h2x4x_parse_dt_tp(struct device *dev,
> + const struct device_node *i3c_node_hub,
> + struct p3h2x4x *priv)
> +{
> + struct device_node *i3c_node_target;
> + int ret;
> +
> + for_each_available_child_of_node(i3c_node_hub, i3c_node_target) {
> + if (!strcmp(i3c_node_target->name, "target-port")) {
> + ret = p3h2x4x_read_backend_from_p3h2x4x_dts(i3c_node_target, priv);
> + if (ret)
> + dev_err(dev, "DTS entry invalid - error %d", ret);
> + }
> + }
> +}
> +
> +static int p3h2x4x_get_tp_local_device_dt_setting(struct device *dev,
> + const struct device_node *node, u32 id)
> +{
> + u8 i;
> + u32 local_dev_count, local_dev;
> + struct p3h2x4x *priv = dev_get_drvdata(dev);
> +
> + if (!of_get_property(node, "local_dev", &local_dev_count))
Oh no, read DTS conding style before upstreaming such code.
Best regards,
Krzysztof
--
linux-i3c mailing list
linux-i3c@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-i3c
WARNING: multiple messages have this Message-ID (diff)
From: Krzysztof Kozlowski <krzk@kernel.org>
To: Aman Kumar Pandey <aman.kumarpandey@nxp.com>,
linux-kernel@vger.kernel.org, linux-i3c@lists.infradead.org,
alexandre.belloni@bootlin.com, krzk+dt@kernel.org,
robh@kernel.org, conor+dt@kernel.org, devicetree@vger.kernel.org
Cc: vikash.bansal@nxp.com, priyanka.jain@nxp.com,
shashank.rebbapragada@nxp.com, Frank.Li@nxp.com
Subject: Re: [PATCH 2/2] drivers: i3c: Add driver for NXP P3H2x4x i3c-hub device
Date: Wed, 12 Feb 2025 17:53:34 +0100 [thread overview]
Message-ID: <76c991dd-f99b-497c-9d68-9e08541be4f0@kernel.org> (raw)
In-Reply-To: <20250212132227.1348374-2-aman.kumarpandey@nxp.com>
On 12/02/2025 14:22, Aman Kumar Pandey wrote:
> +
> +static void p3h2x4x_of_get_tp_dt_conf(struct device *dev,
> + const struct device_node *node)
> +{
> + struct p3h2x4x *priv = dev_get_drvdata(dev);
> + struct device_node *dev_node;
> + int tp_port;
> +
> + for_each_available_child_of_node(node, dev_node) {
> + if (!dev_node->full_name ||
> + (sscanf(dev_node->full_name, "target-port@%d", &tp_port) != 1))
NAK, undocumented ABI. Also weird code, why are you parsing DT structure
manually? Use proper functions to get the reg.
> + continue;
> +
> + if (tp_port < P3H2x4x_TP_MAX_COUNT) {
> + priv->tp_bus[tp_port].dt_available = true;
> + priv->tp_bus[tp_port].of_node = dev_node;
> + priv->tp_bus[tp_port].tp_mask = BIT(tp_port);
> + priv->tp_bus[tp_port].priv = priv;
> + priv->tp_bus[tp_port].tp_port = tp_port;
> + }
> + }
> +}
> +
> +/* return true when backend node exist */
> +static bool p3h2x4x_is_backend_node_exist(int port, struct p3h2x4x *priv, u32 addr)
> +{
> + struct smbus_device *backend = NULL;
> +
> + list_for_each_entry(backend,
> + &priv->tp_bus[port].tp_device_entry, list) {
> + if (backend->addr == addr)
> + return true;
> + }
> + return false;
> +}
> +
> +static int p3h2x4x_read_backend_from_p3h2x4x_dts(struct device_node *i3c_node_target,
> + struct p3h2x4x *priv)
> +{
> + struct device_node *i3c_node_tp;
> + const char *compatible;
> + int tp_port, ret;
> + u32 addr_dts;
> + struct smbus_device *backend;
> +
> + if (sscanf(i3c_node_target->full_name, "target-port@%d", &tp_port) == 0)
> + return -EINVAL;
> +
> + if (tp_port > P3H2x4x_TP_MAX_COUNT)
> + return -ERANGE;
> +
> + if (tp_port < 0)
> + return -EINVAL;
> +
> + INIT_LIST_HEAD(&priv->tp_bus[tp_port].tp_device_entry);
> +
> + if (priv->settings.tp[tp_port].mode == P3H2x4x_TP_MODE_I3C)
> + return 0;
> +
> + for_each_available_child_of_node(i3c_node_target, i3c_node_tp) {
> +
> + ret = of_property_read_u32(i3c_node_tp, "reg", &addr_dts);
> + if (ret)
> + return ret;
> +
> + if (p3h2x4x_is_backend_node_exist(tp_port, priv, addr_dts))
> + continue;
> +
> + ret = of_property_read_string(i3c_node_tp, "compatible", &compatible);
> + if (ret)
> + return ret;
> +
> + backend = kzalloc(sizeof(*backend), GFP_KERNEL);
> + if (!backend)
> + return -ENOMEM;
> +
> + backend->addr = addr_dts;
> + backend->compatible = compatible;
> + backend->tp_device_dt_node = i3c_node_tp;
> + backend->client = NULL;
> +
> + list_add(&backend->list,
> + &priv->tp_bus[tp_port].tp_device_entry);
> + }
> +
> + return 0;
> +}
> +
> +static void p3h2x4x_parse_dt_tp(struct device *dev,
> + const struct device_node *i3c_node_hub,
> + struct p3h2x4x *priv)
> +{
> + struct device_node *i3c_node_target;
> + int ret;
> +
> + for_each_available_child_of_node(i3c_node_hub, i3c_node_target) {
> + if (!strcmp(i3c_node_target->name, "target-port")) {
> + ret = p3h2x4x_read_backend_from_p3h2x4x_dts(i3c_node_target, priv);
> + if (ret)
> + dev_err(dev, "DTS entry invalid - error %d", ret);
> + }
> + }
> +}
> +
> +static int p3h2x4x_get_tp_local_device_dt_setting(struct device *dev,
> + const struct device_node *node, u32 id)
> +{
> + u8 i;
> + u32 local_dev_count, local_dev;
> + struct p3h2x4x *priv = dev_get_drvdata(dev);
> +
> + if (!of_get_property(node, "local_dev", &local_dev_count))
Oh no, read DTS conding style before upstreaming such code.
Best regards,
Krzysztof
next prev parent reply other threads:[~2025-02-12 17:11 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-12 13:22 [PATCH 1/2] dt-bindings: i3c: Add NXP P3H2x4x i3c-hub support Aman Kumar Pandey
2025-02-12 13:22 ` Aman Kumar Pandey
2025-02-12 13:22 ` [PATCH 2/2] drivers: i3c: Add driver for NXP P3H2x4x i3c-hub device Aman Kumar Pandey
2025-02-12 13:22 ` Aman Kumar Pandey
2025-02-12 16:42 ` Frank Li
2025-02-12 16:42 ` Frank Li
2025-02-12 16:53 ` Krzysztof Kozlowski [this message]
2025-02-12 16:53 ` Krzysztof Kozlowski
2025-02-12 18:20 ` Alexandre Belloni
2025-02-12 18:20 ` Alexandre Belloni
2025-02-13 11:49 ` kernel test robot
2025-02-13 11:49 ` kernel test robot
2025-02-13 17:46 ` kernel test robot
2025-02-13 17:46 ` kernel test robot
2025-02-14 4:21 ` kernel test robot
2025-02-14 4:21 ` kernel test robot
2025-02-16 15:17 ` Dan Carpenter
2025-02-16 15:17 ` Dan Carpenter
2025-02-19 4:52 ` kernel test robot
2025-02-19 4:52 ` kernel test robot
2025-02-12 15:25 ` [PATCH 1/2] dt-bindings: i3c: Add NXP P3H2x4x i3c-hub support Rob Herring (Arm)
2025-02-12 15:25 ` Rob Herring (Arm)
2025-02-12 16:23 ` Frank Li
2025-02-12 16:23 ` Frank Li
2025-02-12 16:49 ` Krzysztof Kozlowski
2025-02-12 16:49 ` Krzysztof Kozlowski
-- strict thread matches above, loose matches on Subject: below --
2025-02-15 1:02 [PATCH 2/2] drivers: i3c: Add driver for NXP P3H2x4x i3c-hub device kernel test robot
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=76c991dd-f99b-497c-9d68-9e08541be4f0@kernel.org \
--to=krzk@kernel.org \
--cc=Frank.Li@nxp.com \
--cc=alexandre.belloni@bootlin.com \
--cc=aman.kumarpandey@nxp.com \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=krzk+dt@kernel.org \
--cc=linux-i3c@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=priyanka.jain@nxp.com \
--cc=robh@kernel.org \
--cc=shashank.rebbapragada@nxp.com \
--cc=vikash.bansal@nxp.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 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.