Skip to contentSkip to conversation

usb: typec: tcpm: Export partner Source Capabilities

PATCH In review Kyle Tso 4 participants 11 messages Jan 31, 2019 – Feb 14, 2019 1 message not in the archive

Subsystems (3)
lore
Share & tools
lore

Kyle Tso

author
#0 lore raw

PATCH 1/1 usb: typec: tcpm: Export partner Source Capabilities

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
drivers/usb/typec/tcpm/tcpm.c +23 -0
index f1d3e54210df..29cd84ba9960 100644
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free the
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+	unsigned int nr_pdo;
+
+	if (port->nr_source_caps == 0)
+		return -ENODATA;
+
+	*src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);
+	if (!src_pdo)
+		return -ENOMEM;
+
+	mutex_lock(&port->lock);
+	nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+				port->nr_source_caps);
+	mutex_unlock(&port->lock);
+	return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);
+
 /* Power Supply access to expose source power information */
 enum tcpm_psy_online_states {
 	TCPM_PSY_OFFLINE = 0,
include/linux/usb/tcpm.h +1 -0
index 50c74a77db55..fe56d759567c 100644
@@ -173,5 +173,6 @@ void tcpm_pd_transmit_complete(struct tcpm_port *port,
 			       enum tcpm_transmit_status status);
 void tcpm_pd_hard_reset(struct tcpm_port *port);
 void tcpm_tcpc_reset(struct tcpm_port *port);
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **pdo);
 
 #endif /* __LINUX_USB_TCPM_H */
maintainer
#1 lore raw
Quoted 28 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
drivers/usb/typec/tcpm/tcpm.c +12 -0
index f1d3e54210df..29cd84ba9960 100644
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free the
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+	unsigned int nr_pdo;
+
+	if (port->nr_source_caps == 0)
+		return -ENODATA;
+
+	*src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+	if (!src_pdo)
+		return -ENOMEM;
+
+	mutex_lock(&port->lock);
+	nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+				port->nr_source_caps);
+	mutex_unlock(&port->lock);
+	return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

#2 lore raw
Quoted 28 lines · Greg KH
Quoted 27 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
drivers/usb/typec/tcpm/tcpm.c +11 -0
index f1d3e54210df..29cd84ba9960 100644
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free the
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+	unsigned int nr_pdo;
+
+	if (port->nr_source_caps == 0)
+		return -ENODATA;
+
Quoted 8 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4515

Quoted 8 lines
drivers/usb/typec/tcpm/tcpm.c +8 -0
+	*src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);
+	if (!src_pdo)
+		return -ENOMEM;
+
+	mutex_lock(&port->lock);
+	nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+				port->nr_source_caps);
+	mutex_unlock(&port->lock);

The mutex use here suggests that the data can be updated while being copied. But that suggests that port->nr_source_caps can change as well and may no longer be current after mutex_lock().

Quoted 6 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 3 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
+	return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

I am also concerned about the API itself; passing a pointer to be freed by the caller invites memory leaks.

Guenter

author
#3 lore raw
Quoted 34 lines · Guenter Roeck
Quoted 32 lines · Greg KH
Quoted 31 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
drivers/usb/typec/tcpm/tcpm.c +15 -0
index f1d3e54210df..29cd84ba9960 100644
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free the
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+   unsigned int nr_pdo;
+
+   if (port->nr_source_caps == 0)
+           return -ENODATA;
+
+   *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);
+   if (!src_pdo)
+           return -ENOMEM;
+
Quoted 9 lines
Quoted 4 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4515

Quoted 4 lines
drivers/usb/typec/tcpm/tcpm.c +4 -0
+   mutex_lock(&port->lock);
+   nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                           port->nr_source_caps);
+   mutex_unlock(&port->lock);

The mutex use here suggests that the data can be updated while being copied. But that suggests that port->nr_source_caps can change as well and may no longer be current after mutex_lock().

Thanks for pointing out the bug. I will fix this in the next version.

Quoted 11 lines
Quoted 6 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 3 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
+   return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

I am also concerned about the API itself; passing a pointer to be freed by the caller invites memory leaks.

Guenter

What is the better way to implement this? How about providing a function (say "tcpm_put_partner_src_caps") for the caller to free the memory?

author
#4 lore raw
In reply to Greg KH
Quoted 34 lines · Greg KH
Quoted 32 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
drivers/usb/typec/tcpm/tcpm.c +16 -0
index f1d3e54210df..29cd84ba9960 100644
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free the
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
Quoted 9 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +6 -0
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize
the charging process.

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and
"tcpm_update_source_capabilities" are exposed
as well. And no one uses them now.

A message that is not in the archive (sent off-list or not yet received): CAGZ6i=07EOHqd3r9y1MpYn_e1nFeYGt_wp9arhMZWWeTBm-udA@mail.gmail.com

maintainer
#6 lore raw
Quoted 48 lines · Kyle Tso
Quoted 36 lines · Greg KH

Review

Quoted 9 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
 drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
 include/linux/usb/tcpm.h      |  1 +
 2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,

Quoted 5 lines
drivers/usb/typec/tcpm/tcpm.c +2 -0
 }
 EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free

the

Quoted 14 lines
drivers/usb/typec/tcpm/tcpm.c +14 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
Quoted 10 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +6 -0
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 5 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize
the charging process.

And where is that code?

Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and
"tcpm_update_source_capabilities" are exposed
as well. And no one uses them now.

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

thanks,

greg k-h

#7 lore raw
Quoted 49 lines · Greg KH
Quoted 48 lines · Kyle Tso
Quoted 46 lines · Greg KH

Review

Quoted 9 lines · Kyle Tso

Provide a function to get the partner Source Capabilities.

---
  drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
  include/linux/usb/tcpm.h      |  1 +
  2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,

Quoted 5 lines
drivers/usb/typec/tcpm/tcpm.c +2 -0
  }
  EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to free

the

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32 **src_pdo)
+{
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 8 lines
Quoted 5 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize
the charging process.

And where is that code?

Agreed - that code should be sent upstream as well to let us see the entire context.

Quoted 8 lines
Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and
"tcpm_update_source_capabilities" are exposed
as well. And no one uses them now.

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

I sent a patch to do just that. Quite frankly I don't recal why I thought those functions might be needed.

Guenter

#8 lore raw
Quoted 68 lines · Guenter Roeck
Quoted 2 lines · Greg KH
Quoted 1 line · Kyle Tso

On Thu, Jan 31, 2019 at 3:02 PM Greg KH <[email protected]>

Quoted 48 lines · wrote:
Quoted 48 lines
Quoted 48 lines · Kyle Tso

Review

Quoted 9 lines

Provide a function to get the partner Source Capabilities.

---
  drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
  include/linux/usb/tcpm.h      |  1 +
  2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
  }
  EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to
+free

the

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32
+**src_pdo) {
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 7 lines
Quoted 4 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize the charging process.

And where is that code?

Agreed - that code should be sent upstream as well to let us see the entire context.

Quoted 5 lines
Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and "tcpm_update_source_capabilities" are exposed as well. And no one uses them now.

Quoted 6 lines
Quoted 3 lines

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

I sent a patch to do just that. Quite frankly I don't recal why I thought those functions might be needed.

As a mind jog - https://lkml.org/lkml/2017/11/27/1256 :)

maintainer
#9 lore raw
Quoted 70 lines · Adam Thomson
Quoted 68 lines · Guenter Roeck
Quoted 2 lines · Greg KH
Quoted 1 line · Kyle Tso

On Thu, Jan 31, 2019 at 3:02 PM Greg KH <[email protected]>

Quoted 48 lines · wrote:
Quoted 48 lines
Quoted 48 lines · Kyle Tso

Review

Quoted 9 lines

Provide a function to get the partner Source Capabilities.

---
  drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
  include/linux/usb/tcpm.h      |  1 +
  2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
  }
  EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to
+free

the

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32
+**src_pdo) {
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 7 lines
Quoted 4 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize the charging process.

And where is that code?

Agreed - that code should be sent upstream as well to let us see the entire context.

Quoted 5 lines
Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and "tcpm_update_source_capabilities" are exposed as well. And no one uses them now.

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

I sent a patch to do just that. Quite frankly I don't recal why I thought those functions might be needed.

As a mind jog - https://lkml.org/lkml/2017/11/27/1256 :)

Nice find :)

Given that no one used these functions in over a year, we should be safe deleting them. If not, it's trivial to do 'git revert'.

thanks,

greg k-h

#10 lore raw
Quoted 80 lines · Greg KH
Quoted 31 lines · Adam Thomson
Quoted 29 lines · Guenter Roeck
Quoted 3 lines · Greg KH
Quoted 2 lines · Kyle Tso

On Thu, Jan 31, 2019 at 3:02 PM Greg KH
<[email protected]>

Quoted 24 lines · wrote:
Quoted 24 lines
Quoted 24 lines · Kyle Tso

Review

Quoted 9 lines

Provide a function to get the partner Source Capabilities.

---
  drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
  include/linux/usb/tcpm.h      |  1 +
  2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct · line 4499

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
  }
  EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs
+to free

the

Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct · line 4509

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32
+**src_pdo) {
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32),

GFP_KERNEL);

Quoted 37 lines
Quoted 36 lines
Quoted 14 lines
Quoted 14 lines
Quoted 14 lines

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct · line 4519

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 7 lines
Quoted 4 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize the charging process.

And where is that code?

Agreed - that code should be sent upstream as well to let us see the entire context.

Quoted 8 lines
Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and "tcpm_update_source_capabilities" are exposed as well. And no one uses them now.

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

I sent a patch to do just that. Quite frankly I don't recal why I thought those functions might be needed.

Quoted 6 lines
Quoted 1 line

As a mind jog - https://lkml.org/lkml/2017/11/27/1256 :)

Nice find :)

Given that no one used these functions in over a year, we should be safe deleting them. If not, it's trivial to do 'git revert'.

Yep, agreed.

#11 lore raw
In reply to Adam Thomson
Quoted 70 lines · Adam Thomson
Quoted 68 lines · Guenter Roeck
Quoted 2 lines · Greg KH
Quoted 1 line · Kyle Tso

On Thu, Jan 31, 2019 at 3:02 PM Greg KH <[email protected]>

Quoted 48 lines · wrote:
Quoted 48 lines
Quoted 48 lines · Kyle Tso

Review

Quoted 9 lines

Provide a function to get the partner Source Capabilities.

---
   drivers/usb/typec/tcpm/tcpm.c | 23 +++++++++++++++++++++++
   include/linux/usb/tcpm.h      |  1 +
   2 files changed, 24 insertions(+)
(quoted diff) +0 -0

b/drivers/usb/typec/tcpm/tcpm.c

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct

Quoted 4 lines

index f1d3e54210df..29cd84ba9960 100644

drivers/usb/typec/tcpm/tcpm.c +0 -0
@@ -4494,6 +4494,29 @@ int tcpm_update_sink_capabilities(struct

tcpm_port *port, const u32 *pdo,

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo,

Quoted 6 lines
drivers/usb/typec/tcpm/tcpm.c +3 -0
   }
   EXPORT_SYMBOL_GPL(tcpm_update_sink_capabilities);
 
+/*
+ * Don't call this function in interrupt context. Caller needs to
+free

the

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+ * memory itself.
+ */
+int tcpm_get_partner_src_caps(struct tcpm_port *port, u32
+**src_pdo) {
+     unsigned int nr_pdo;
+
+     if (port->nr_source_caps == 0)
+             return -ENODATA;
+
+     *src_pdo = kcalloc(port->nr_source_caps, sizeof(u32), GFP_KERNEL);

drivers/usb/typec/tcpm/tcpm.c int tcpm_update_sink_capabilities(struct tcpm_port *port, const u32 *pdo, · line 4518

Quoted 10 lines
drivers/usb/typec/tcpm/tcpm.c +10 -0
+     if (!src_pdo)
+             return -ENOMEM;
+
+     mutex_lock(&port->lock);
+     nr_pdo = tcpm_copy_pdos(*src_pdo, port->source_caps,
+                             port->nr_source_caps);
+     mutex_unlock(&port->lock);
+     return nr_pdo;
+}
+EXPORT_SYMBOL_GPL(tcpm_get_partner_src_caps);

We don't add new functions that no one uses :(

Quoted 7 lines
Quoted 4 lines

This function is useful if the PD Device Policy Manager is implemented outside of TCPM.
In this situation, Device Policy Manager needs to know the partner capabilities to optimize the charging process.

And where is that code?

Agreed - that code should be sent upstream as well to let us see the entire context.

Quoted 5 lines
Quoted 4 lines

Take existing functions in TCPM for example:
Function "tcpm_update_sink_capabilities" and "tcpm_update_source_capabilities" are exposed as well. And no one uses them now.

Quoted 9 lines
Quoted 6 lines
Quoted 3 lines

Great, let's go delete them now, we should not have apis that no one uses. This isn't a new thing...

I sent a patch to do just that. Quite frankly I don't recal why I thought those functions might be needed.

As a mind jog - https://lkml.org/lkml/2017/11/27/1256 :)

Nice find. However, since the code using it was never implemented, I think it is prudent at this point to remove the code. It can be reintroduced if/when needed, and at that point be tested properly.

Thanks,
Guenter