[PATCH 1/2] i2c: amd: Switch to guard(mutex)

[PATCH 1/2] i2c: amd: Switch to guard(mutex)

[Shyam Sundar S K]

Instead of using the 'goto label; mutex_unlock()' pattern use
'guard(mutex)' which will release the mutex when it goes out of scope.

Co-developed-by: Sanket Goswami <Sanket.Goswami@amd.com>
Signed-off-by: Sanket Goswami <Sanket.Goswami@amd.com>
Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
---
 drivers/i2c/busses/i2c-designware-amdpsp.c | 26 ++++++++--------------
 1 file changed, 9 insertions(+), 17 deletions(-)

diff --git a/drivers/i2c/busses/i2c-designware-amdpsp.c b/drivers/i2c/busses/i2c-designware-amdpsp.c
index 8fbd2a10c31a..404571ad61a8 100644
--- a/drivers/i2c/busses/i2c-designware-amdpsp.c
+++ b/drivers/i2c/busses/i2c-designware-amdpsp.c
@@ -151,19 +151,16 @@ static void release_bus(void)
 
 static void psp_release_i2c_bus_deferred(struct work_struct *work)
 {
-	mutex_lock(&psp_i2c_access_mutex);
+	guard(mutex)(&psp_i2c_access_mutex);
 
 	/*
 	 * If there is any pending transaction, cannot release the bus here.
 	 * psp_release_i2c_bus() will take care of this later.
 	 */
 	if (psp_i2c_access_count)
-		goto cleanup;
+		return;
 
 	release_bus();
-
-cleanup:
-	mutex_unlock(&psp_i2c_access_mutex);
 }
 static DECLARE_DELAYED_WORK(release_queue, psp_release_i2c_bus_deferred);
 
@@ -171,11 +168,11 @@ static int psp_acquire_i2c_bus(void)
 {
 	int status;
 
-	mutex_lock(&psp_i2c_access_mutex);
+	guard(mutex)(&psp_i2c_access_mutex);
 
 	/* Return early if mailbox malfunctioned */
 	if (psp_i2c_mbox_fail)
-		goto cleanup;
+		return 0;
 
 	psp_i2c_access_count++;
 
@@ -184,11 +181,11 @@ static int psp_acquire_i2c_bus(void)
 	 * reservation period.
 	 */
 	if (psp_i2c_sem_acquired)
-		goto cleanup;
+		return 0;
 
 	status = psp_send_i2c_req(PSP_I2C_REQ_ACQUIRE);
 	if (status)
-		goto cleanup;
+		return 0;
 
 	psp_i2c_sem_acquired = jiffies;
 
@@ -201,18 +198,16 @@ static int psp_acquire_i2c_bus(void)
 	 * communication with PSP. At any case i2c bus is granted to the caller,
 	 * thus always return success.
 	 */
-cleanup:
-	mutex_unlock(&psp_i2c_access_mutex);
 	return 0;
 }
 
 static void psp_release_i2c_bus(void)
 {
-	mutex_lock(&psp_i2c_access_mutex);
+	guard(mutex)(&psp_i2c_access_mutex);
 
 	/* Return early if mailbox was malfunctioned */
 	if (psp_i2c_mbox_fail)
-		goto cleanup;
+		return;
 
 	/*
 	 * If we are last owner of PSP semaphore, need to release arbitration
@@ -220,7 +215,7 @@ static void psp_release_i2c_bus(void)
 	 */
 	psp_i2c_access_count--;
 	if (psp_i2c_access_count)
-		goto cleanup;
+		return;
 
 	/*
 	 * Send a release command to PSP if the semaphore reservation timeout
@@ -228,9 +223,6 @@ static void psp_release_i2c_bus(void)
 	 */
 	if (!delayed_work_pending(&release_queue))
 		release_bus();
-
-cleanup:
-	mutex_unlock(&psp_i2c_access_mutex);
 }
 
 /*
-- 
2.34.1

[PATCH 2/2] i2c: dw: Update the master_xfer callback name

[Shyam Sundar S K]

In light of the recent updates to the i2c subsystem, ensure to use the
correct callback names. Specifically, replace '.master_xfer' with '.xfer'.

Co-developed-by: Sanket Goswami <Sanket.Goswami@amd.com>
Signed-off-by: Sanket Goswami <Sanket.Goswami@amd.com>
Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
---
 drivers/i2c/busses/i2c-designware-master.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-designware-master.c b/drivers/i2c/busses/i2c-designware-master.c
index 2569bf1a72e0..c5394229b77f 100644
--- a/drivers/i2c/busses/i2c-designware-master.c
+++ b/drivers/i2c/busses/i2c-designware-master.c
@@ -907,7 +907,7 @@ i2c_dw_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
 }
 
 static const struct i2c_algorithm i2c_dw_algo = {
-	.master_xfer = i2c_dw_xfer,
+	.xfer = i2c_dw_xfer,
 	.functionality = i2c_dw_func,
 };
 
-- 
2.34.1

Re: [PATCH 1/2] i2c: amd: Switch to guard(mutex)

[Andy Shevchenko]

On Mon, Feb 17, 2025 at 06:17:08PM +0530, Shyam Sundar S K wrote:
> Instead of using the 'goto label; mutex_unlock()' pattern use
> 'guard(mutex)' which will release the mutex when it goes out of scope.

Both LGTM,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

But next time either sent them separately (seems like they are independent),
or add a cover letter explaining what the big picture behind all the patches
as a whole.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 1/2] i2c: amd: Switch to guard(mutex)

[Shyam Sundar S K]

On 2/18/2025 01:49, Andy Shevchenko wrote:
> On Mon, Feb 17, 2025 at 06:17:08PM +0530, Shyam Sundar S K wrote:
>> Instead of using the 'goto label; mutex_unlock()' pattern use
>> 'guard(mutex)' which will release the mutex when it goes out of scope.
> 
> Both LGTM,
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> 
> But next time either sent them separately (seems like they are independent),
> or add a cover letter explaining what the big picture behind all the patches
> as a whole.
> 

Thanks! and.. for sure next time..

Thanks,
Shyam

Re: [PATCH 1/2] i2c: amd: Switch to guard(mutex)

[Jarkko Nikula]

On 2/17/25 10:19 PM, Andy Shevchenko wrote:
> On Mon, Feb 17, 2025 at 06:17:08PM +0530, Shyam Sundar S K wrote:
>> Instead of using the 'goto label; mutex_unlock()' pattern use
>> 'guard(mutex)' which will release the mutex when it goes out of scope.
> 
> Both LGTM,
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> 
> But next time either sent them separately (seems like they are independent),
> or add a cover letter explaining what the big picture behind all the patches
> as a whole.
> 
To both:

Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>