All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cristian Marussi <cristian.marussi@arm.com>
To: Oleksii Moisieiev <Oleksii_Moisieiev@epam.com>
Cc: "sudeep.holla@arm.com" <sudeep.holla@arm.com>,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Linus Walleij <linus.walleij@linaro.org>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>
Subject: Re: [PATCH v4 1/4] firmware: arm_scmi: Add optional flags to extended names helper
Date: Wed, 16 Aug 2023 11:27:32 +0100	[thread overview]
Message-ID: <ZNyklIL7qGhJnDiJ@pluto> (raw)
In-Reply-To: <318eb79c7e1ddb1f964a901e778a0475bf18c85b.1691518313.git.oleksii_moisieiev@epam.com>

On Tue, Aug 08, 2023 at 06:25:35PM +0000, Oleksii Moisieiev wrote:
> From: Cristian Marussi <cristian.marussi@arm.com>
> 
> Some recently added SCMI protocols needs an additional flags parameter to
> be able to properly configure the command used to query the extended name
> of a resource.
> 

Hi Oleksii,

> Signed-off-by: Cristian Marussi <cristian.marussi@arm.com>
> ---

You need to apply also your Signed-off as last in V5 for this to be
accepted AND beside this I spotted a bug while re-testing this. (my bad)
See down below a possible fix.

>  drivers/firmware/arm_scmi/clock.c     | 2 +-
>  drivers/firmware/arm_scmi/driver.c    | 7 +++++--
>  drivers/firmware/arm_scmi/perf.c      | 3 ++-
>  drivers/firmware/arm_scmi/power.c     | 2 +-
>  drivers/firmware/arm_scmi/powercap.c  | 2 +-
>  drivers/firmware/arm_scmi/protocols.h | 3 ++-
>  drivers/firmware/arm_scmi/reset.c     | 3 ++-
>  drivers/firmware/arm_scmi/sensors.c   | 2 +-
>  drivers/firmware/arm_scmi/voltage.c   | 2 +-
>  9 files changed, 16 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c
> index 96060bf90a24..e6e087686e8c 100644
> --- a/drivers/firmware/arm_scmi/clock.c
> +++ b/drivers/firmware/arm_scmi/clock.c
> @@ -169,7 +169,7 @@ static int scmi_clock_attributes_get(const struct scmi_protocol_handle *ph,
>  	if (!ret && PROTOCOL_REV_MAJOR(version) >= 0x2) {
>  		if (SUPPORTS_EXTENDED_NAMES(attributes))
>  			ph->hops->extended_name_get(ph, CLOCK_NAME_GET, clk_id,
> -						    clk->name,
> +						    NULL, clk->name,
>  						    SCMI_MAX_STR_SIZE);
>  
>  		if (SUPPORTS_RATE_CHANGED_NOTIF(attributes))
> diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c
> index e7d97b59963b..729201d8f935 100644
> --- a/drivers/firmware/arm_scmi/driver.c
> +++ b/drivers/firmware/arm_scmi/driver.c
> @@ -1438,6 +1438,7 @@ struct scmi_msg_resp_domain_name_get {
>   * @ph: A protocol handle reference.
>   * @cmd_id: The specific command ID to use.
>   * @res_id: The specific resource ID to use.
> + * @flags: A pointer to specific flags to use, if any.
>   * @name: A pointer to the preallocated area where the retrieved name will be
>   *	  stored as a NULL terminated string.
>   * @len: The len in bytes of the @name char array.
> @@ -1445,8 +1446,8 @@ struct scmi_msg_resp_domain_name_get {
>   * Return: 0 on Succcess
>   */
>  static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
> -					 u8 cmd_id, u32 res_id, char *name,
> -					 size_t len)
> +					 u8 cmd_id, u32 res_id, u32 *flags,
> +					 char *name, size_t len)
>  {
>  	int ret;
>  	struct scmi_xfer *t;
> @@ -1458,6 +1459,8 @@ static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
>  		goto out;
>  
>  	put_unaligned_le32(res_id, t->tx.buf);
> +	if (flags)
> +		put_unaligned_le32(*flags, t->tx.buf + sizeof(res_id));
>  	resp = t->rx.buf;
>  
>  	ret = ph->xops->do_xfer(ph, t);

Here my patch is buggy since when you provide the optional flags they
are after the res_id parameter BUT the previous xfer_get_init still
requires an xfer with a tx_len size of sizeof(res_id) ONLY, so while the
flags will fit into tx.buf that second optional flags field won't be
sent because the core SCMI stack will see t->tx.len == 4 :<
(so this flags extension wont work when needed in Pinctrl)

A possible fix that I tested consist in changing this snippet with:

-->8---
diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c
index b5957cc12fee..06c101edba7f 100644
--- a/drivers/firmware/arm_scmi/driver.c
+++ b/drivers/firmware/arm_scmi/driver.c
@@ -1438,6 +1438,7 @@ struct scmi_msg_resp_domain_name_get {
  * @ph: A protocol handle reference.
  * @cmd_id: The specific command ID to use.
  * @res_id: The specific resource ID to use.
+ * @flags: A pointer to specific flags to use, if any.
  * @name: A pointer to the preallocated area where the retrieved name will be
  *       stored as a NULL terminated string.
  * @len: The len in bytes of the @name char array.
@@ -1445,19 +1446,22 @@ struct scmi_msg_resp_domain_name_get {
  * Return: 0 on Succcess
  */
 static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
-                                        u8 cmd_id, u32 res_id, char *name,
-                                        size_t len)
+                                        u8 cmd_id, u32 res_id, u32 *flags,
+                                        char *name, size_t len)
 {
        int ret;
+       size_t txlen;
        struct scmi_xfer *t;
        struct scmi_msg_resp_domain_name_get *resp;
 
-       ret = ph->xops->xfer_get_init(ph, cmd_id, sizeof(res_id),
-                                     sizeof(*resp), &t);
+       txlen = !flags ? sizeof(res_id) : sizeof(res_id) + sizeof(*flags);
+       ret = ph->xops->xfer_get_init(ph, cmd_id, txlen, sizeof(*resp), &t);
        if (ret)
                goto out;
 
        put_unaligned_le32(res_id, t->tx.buf);
+       if (flags)
+               put_unaligned_le32(*flags, t->tx.buf + sizeof(res_id));
        resp = t->rx.buf;
 
        ret = ph->xops->do_xfer(ph, t);


---8<---

My bad,

Thanks,
Cristian

WARNING: multiple messages have this Message-ID (diff)
From: Cristian Marussi <cristian.marussi@arm.com>
To: Oleksii Moisieiev <Oleksii_Moisieiev@epam.com>
Cc: "sudeep.holla@arm.com" <sudeep.holla@arm.com>,
	Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Linus Walleij <linus.walleij@linaro.org>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>
Subject: Re: [PATCH v4 1/4] firmware: arm_scmi: Add optional flags to extended names helper
Date: Wed, 16 Aug 2023 11:27:32 +0100	[thread overview]
Message-ID: <ZNyklIL7qGhJnDiJ@pluto> (raw)
In-Reply-To: <318eb79c7e1ddb1f964a901e778a0475bf18c85b.1691518313.git.oleksii_moisieiev@epam.com>

On Tue, Aug 08, 2023 at 06:25:35PM +0000, Oleksii Moisieiev wrote:
> From: Cristian Marussi <cristian.marussi@arm.com>
> 
> Some recently added SCMI protocols needs an additional flags parameter to
> be able to properly configure the command used to query the extended name
> of a resource.
> 

Hi Oleksii,

> Signed-off-by: Cristian Marussi <cristian.marussi@arm.com>
> ---

You need to apply also your Signed-off as last in V5 for this to be
accepted AND beside this I spotted a bug while re-testing this. (my bad)
See down below a possible fix.

>  drivers/firmware/arm_scmi/clock.c     | 2 +-
>  drivers/firmware/arm_scmi/driver.c    | 7 +++++--
>  drivers/firmware/arm_scmi/perf.c      | 3 ++-
>  drivers/firmware/arm_scmi/power.c     | 2 +-
>  drivers/firmware/arm_scmi/powercap.c  | 2 +-
>  drivers/firmware/arm_scmi/protocols.h | 3 ++-
>  drivers/firmware/arm_scmi/reset.c     | 3 ++-
>  drivers/firmware/arm_scmi/sensors.c   | 2 +-
>  drivers/firmware/arm_scmi/voltage.c   | 2 +-
>  9 files changed, 16 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c
> index 96060bf90a24..e6e087686e8c 100644
> --- a/drivers/firmware/arm_scmi/clock.c
> +++ b/drivers/firmware/arm_scmi/clock.c
> @@ -169,7 +169,7 @@ static int scmi_clock_attributes_get(const struct scmi_protocol_handle *ph,
>  	if (!ret && PROTOCOL_REV_MAJOR(version) >= 0x2) {
>  		if (SUPPORTS_EXTENDED_NAMES(attributes))
>  			ph->hops->extended_name_get(ph, CLOCK_NAME_GET, clk_id,
> -						    clk->name,
> +						    NULL, clk->name,
>  						    SCMI_MAX_STR_SIZE);
>  
>  		if (SUPPORTS_RATE_CHANGED_NOTIF(attributes))
> diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c
> index e7d97b59963b..729201d8f935 100644
> --- a/drivers/firmware/arm_scmi/driver.c
> +++ b/drivers/firmware/arm_scmi/driver.c
> @@ -1438,6 +1438,7 @@ struct scmi_msg_resp_domain_name_get {
>   * @ph: A protocol handle reference.
>   * @cmd_id: The specific command ID to use.
>   * @res_id: The specific resource ID to use.
> + * @flags: A pointer to specific flags to use, if any.
>   * @name: A pointer to the preallocated area where the retrieved name will be
>   *	  stored as a NULL terminated string.
>   * @len: The len in bytes of the @name char array.
> @@ -1445,8 +1446,8 @@ struct scmi_msg_resp_domain_name_get {
>   * Return: 0 on Succcess
>   */
>  static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
> -					 u8 cmd_id, u32 res_id, char *name,
> -					 size_t len)
> +					 u8 cmd_id, u32 res_id, u32 *flags,
> +					 char *name, size_t len)
>  {
>  	int ret;
>  	struct scmi_xfer *t;
> @@ -1458,6 +1459,8 @@ static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
>  		goto out;
>  
>  	put_unaligned_le32(res_id, t->tx.buf);
> +	if (flags)
> +		put_unaligned_le32(*flags, t->tx.buf + sizeof(res_id));
>  	resp = t->rx.buf;
>  
>  	ret = ph->xops->do_xfer(ph, t);

Here my patch is buggy since when you provide the optional flags they
are after the res_id parameter BUT the previous xfer_get_init still
requires an xfer with a tx_len size of sizeof(res_id) ONLY, so while the
flags will fit into tx.buf that second optional flags field won't be
sent because the core SCMI stack will see t->tx.len == 4 :<
(so this flags extension wont work when needed in Pinctrl)

A possible fix that I tested consist in changing this snippet with:

-->8---
diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c
index b5957cc12fee..06c101edba7f 100644
--- a/drivers/firmware/arm_scmi/driver.c
+++ b/drivers/firmware/arm_scmi/driver.c
@@ -1438,6 +1438,7 @@ struct scmi_msg_resp_domain_name_get {
  * @ph: A protocol handle reference.
  * @cmd_id: The specific command ID to use.
  * @res_id: The specific resource ID to use.
+ * @flags: A pointer to specific flags to use, if any.
  * @name: A pointer to the preallocated area where the retrieved name will be
  *       stored as a NULL terminated string.
  * @len: The len in bytes of the @name char array.
@@ -1445,19 +1446,22 @@ struct scmi_msg_resp_domain_name_get {
  * Return: 0 on Succcess
  */
 static int scmi_common_extended_name_get(const struct scmi_protocol_handle *ph,
-                                        u8 cmd_id, u32 res_id, char *name,
-                                        size_t len)
+                                        u8 cmd_id, u32 res_id, u32 *flags,
+                                        char *name, size_t len)
 {
        int ret;
+       size_t txlen;
        struct scmi_xfer *t;
        struct scmi_msg_resp_domain_name_get *resp;
 
-       ret = ph->xops->xfer_get_init(ph, cmd_id, sizeof(res_id),
-                                     sizeof(*resp), &t);
+       txlen = !flags ? sizeof(res_id) : sizeof(res_id) + sizeof(*flags);
+       ret = ph->xops->xfer_get_init(ph, cmd_id, txlen, sizeof(*resp), &t);
        if (ret)
                goto out;
 
        put_unaligned_le32(res_id, t->tx.buf);
+       if (flags)
+               put_unaligned_le32(*flags, t->tx.buf + sizeof(res_id));
        resp = t->rx.buf;
 
        ret = ph->xops->do_xfer(ph, t);


---8<---

My bad,

Thanks,
Cristian

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

  reply	other threads:[~2023-08-16 10:28 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-08 18:25 [PATCH v4 0/4] firmware: arm_scmi: Add SCMI v3.2 pincontrol protocol basic support Oleksii Moisieiev
2023-08-08 18:25 ` Oleksii Moisieiev
2023-08-08 18:25 ` [PATCH v4 2/4] " Oleksii Moisieiev
2023-08-08 18:25   ` Oleksii Moisieiev
2023-08-16 12:21   ` Cristian Marussi
2023-08-16 12:21     ` Cristian Marussi
2023-08-08 18:25 ` [PATCH v4 1/4] firmware: arm_scmi: Add optional flags to extended names helper Oleksii Moisieiev
2023-08-08 18:25   ` Oleksii Moisieiev
2023-08-16 10:27   ` Cristian Marussi [this message]
2023-08-16 10:27     ` Cristian Marussi
2023-08-08 18:25 ` [PATCH v4 3/4] pinctrl: Implementation of the generic scmi-pinctrl driver Oleksii Moisieiev
2023-08-08 18:25   ` Oleksii Moisieiev
2023-08-10  8:43   ` Linus Walleij
2023-08-10  8:43     ` Linus Walleij
2023-08-16 12:25   ` Cristian Marussi
2023-08-16 12:25     ` Cristian Marussi
2023-08-23  4:13   ` AKASHI Takahiro
2023-08-23  4:13     ` AKASHI Takahiro
2023-08-23 13:34     ` Cristian Marussi
2023-08-23 13:34       ` Cristian Marussi
2023-08-08 18:25 ` [PATCH v4 4/4] dt-bindings: firmware: arm,scmi: Add support for pinctrl protocol Oleksii Moisieiev
2023-08-08 18:25   ` Oleksii Moisieiev
2023-08-21 16:57   ` Rob Herring
2023-08-21 16:57     ` Rob Herring
2023-08-25  1:03   ` AKASHI Takahiro
2023-08-25  1:03     ` AKASHI Takahiro
2023-08-16 12:30 ` [PATCH v4 0/4] firmware: arm_scmi: Add SCMI v3.2 pincontrol protocol basic support Cristian Marussi
2023-08-16 12:30   ` Cristian Marussi

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=ZNyklIL7qGhJnDiJ@pluto \
    --to=cristian.marussi@arm.com \
    --cc=Oleksii_Moisieiev@epam.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=krzysztof.kozlowski+dt@linaro.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=robh+dt@kernel.org \
    --cc=sudeep.holla@arm.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.