Skip to contentSkip to conversation

gpiolib: Deduplicate cleanup for-loop in gpiochip_add_data_with_key()

PATCH Superseded Andy Shevchenko 2 participants 10 messages Feb 21, 2024 – Mar 4, 2024 GPIO

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 gpiolib: Deduplicate cleanup for-loop in gpiochip_add_data_with_key()

There is no need to repeat for-loop twice in the error path in gpiochip_add_data_with_key(). Deduplicate it. While at it, rename loop variable to be more specific and avoid ambguity.

It also properly unwinds the SRCU, i.e. in reversed order of allocating.

---
 drivers/gpio/gpiolib.c | 26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)
drivers/gpio/gpiolib.c +11 -15
index 1706edb3ee3f..60fa7816c799 100644
@@ -861,7 +861,7 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 			       struct lock_class_key *request_key)
 {
 	struct gpio_device *gdev;
-	unsigned int i, j;
+	unsigned int desc_index;
 	int base = 0;
 	int ret = 0;
 
@@ -965,8 +965,8 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 		}
 	}
 
-	for (i = 0; i < gc->ngpio; i++)
-		gdev->descs[i].gdev = gdev;
+	for (desc_index = 0; desc_index < gc->ngpio; desc_index++)
+		gdev->descs[desc_index].gdev = gdev;
 
 	BLOCKING_INIT_NOTIFIER_HEAD(&gdev->line_state_notifier);
 	BLOCKING_INIT_NOTIFIER_HEAD(&gdev->device_notifier);
@@ -992,19 +992,16 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 	if (ret)
 		goto err_cleanup_gdev_srcu;
 
-	for (i = 0; i < gc->ngpio; i++) {
-		struct gpio_desc *desc = &gdev->descs[i];
+	for (desc_index = 0; desc_index < gc->ngpio; desc_index++) {
+		struct gpio_desc *desc = &gdev->descs[desc_index];
 
 		ret = init_srcu_struct(&desc->srcu);
-		if (ret) {
-			for (j = 0; j < i; j++)
-				cleanup_srcu_struct(&gdev->descs[j].srcu);
-			goto err_free_gpiochip_mask;
-		}
+		if (ret)
+			goto err_cleanup_desc_srcu;
 
-		if (gc->get_direction && gpiochip_line_is_valid(gc, i)) {
+		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index)) {
 			assign_bit(FLAG_IS_OUT,
-				   &desc->flags, !gc->get_direction(gc, i));
+				   &desc->flags, !gc->get_direction(gc, desc_index));
 		} else {
 			assign_bit(FLAG_IS_OUT,
 				   &desc->flags, !gc->direction_input);
@@ -1061,9 +1058,8 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 	gpiochip_free_hogs(gc);
 	of_gpiochip_remove(gc);
 err_cleanup_desc_srcu:
-	for (i = 0; i < gdev->ngpio; i++)
-		cleanup_srcu_struct(&gdev->descs[i].srcu);
-err_free_gpiochip_mask:
+	while (desc_index--)
+		cleanup_srcu_struct(&gdev->descs[desc_index].srcu);
 	gpiochip_free_valid_mask(gc);
 err_cleanup_gdev_srcu:
 	cleanup_srcu_struct(&gdev->srcu);
Signature · 1 line
2.43.0.rc1.1.gbec44491f096
#1 lore raw
Quoted 9 lines · Andy Shevchenko

There is no need to repeat for-loop twice in the error path in gpiochip_add_data_with_key(). Deduplicate it. While at it, rename loop variable to be more specific and avoid ambguity.

It also properly unwinds the SRCU, i.e. in reversed order of allocating.

---

This doesn't apply on top of gpio/for-next, I think it depends on one of your earlier patches?

Quoted 54 lines
 drivers/gpio/gpiolib.c | 26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)
drivers/gpio/gpiolib.c +9 -12
index 1706edb3ee3f..60fa7816c799 100644
@@ -861,7 +861,7 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
                               struct lock_class_key *request_key)
 {
        struct gpio_device *gdev;
-       unsigned int i, j;
+       unsigned int desc_index;
        int base = 0;
        int ret = 0;
 
@@ -965,8 +965,8 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
                }
        }
 
-       for (i = 0; i < gc->ngpio; i++)
-               gdev->descs[i].gdev = gdev;
+       for (desc_index = 0; desc_index < gc->ngpio; desc_index++)
+               gdev->descs[desc_index].gdev = gdev;
 
        BLOCKING_INIT_NOTIFIER_HEAD(&gdev->line_state_notifier);
        BLOCKING_INIT_NOTIFIER_HEAD(&gdev->device_notifier);
@@ -992,19 +992,16 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
        if (ret)
                goto err_cleanup_gdev_srcu;
 
-       for (i = 0; i < gc->ngpio; i++) {
-               struct gpio_desc *desc = &gdev->descs[i];
+       for (desc_index = 0; desc_index < gc->ngpio; desc_index++) {
+               struct gpio_desc *desc = &gdev->descs[desc_index];
 
                ret = init_srcu_struct(&desc->srcu);
-               if (ret) {
-                       for (j = 0; j < i; j++)
-                               cleanup_srcu_struct(&gdev->descs[j].srcu);
-                       goto err_free_gpiochip_mask;
-               }
+               if (ret)
+                       goto err_cleanup_desc_srcu;
 
-               if (gc->get_direction && gpiochip_line_is_valid(gc, i)) {
+               if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index)) {
                        assign_bit(FLAG_IS_OUT,
-                                  &desc->flags, !gc->get_direction(gc, i));
+                                  &desc->flags, !gc->get_direction(gc, desc_index));
                } else {
                        assign_bit(FLAG_IS_OUT,
                                   &desc->flags, !gc->direction_input);

drivers/gpio/gpiolib.c int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data, · line 1061

Quoted 7 lines
drivers/gpio/gpiolib.c +1 -3
@@ -1061,9 +1058,8 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
        gpiochip_free_hogs(gc);
        of_gpiochip_remove(gc);
 err_cleanup_desc_srcu:
-       for (i = 0; i < gdev->ngpio; i++)
-               cleanup_srcu_struct(&gdev->descs[i].srcu);
-err_free_gpiochip_mask:
+       while (desc_index--)

What about gdev->descs[0]?

Quoted 7 lines
drivers/gpio/gpiolib.c +1 -1
+               cleanup_srcu_struct(&gdev->descs[desc_index].srcu);
        gpiochip_free_valid_mask(gc);
 err_cleanup_gdev_srcu:
        cleanup_srcu_struct(&gdev->srcu);
--

2.43.0.rc1.1.gbec44491f096

Bart

#2 lore raw
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Andy Shevchenko

There is no need to repeat for-loop twice in the error path in gpiochip_add_data_with_key(). Deduplicate it. While at it, rename loop variable to be more specific and avoid ambguity.

It also properly unwinds the SRCU, i.e. in reversed order of allocating.

...

Quoted 2 lines

This doesn't apply on top of gpio/for-next, I think it depends on one of your earlier patches?

Yes, on the fix with error path.

...

Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

The difference is what the value will i get _after_ the loop.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw
Quoted 26 lines · Andy Shevchenko
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Andy Shevchenko

There is no need to repeat for-loop twice in the error path in gpiochip_add_data_with_key(). Deduplicate it. While at it, rename loop variable to be more specific and avoid ambguity.

It also properly unwinds the SRCU, i.e. in reversed order of allocating.

...

Quoted 2 lines

This doesn't apply on top of gpio/for-next, I think it depends on one of your earlier patches?

Yes, on the fix with error path.

...

Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

Quoted 7 lines

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

Bart

Quoted 6 lines

--
With Best Regards,
Andy Shevchenko

#4 lore raw
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 10 lines
Quoted 10 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

Quoted 7 lines
Quoted 3 lines

are all equivalents.

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#5 lore raw
Quoted 22 lines · Andy Shevchenko
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 12 lines
Quoted 12 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

Quoted 9 lines
Quoted 5 lines
Quoted 1 line

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

Whatever, I don't have a strong opinion, just rebase it and resend.

Bart

#6 lore raw
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 14 lines
Quoted 14 lines
Quoted 14 lines
Quoted 13 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

The difference is what the value will i get _after_ the loop.

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

Whatever, I don't have a strong opinion, just rebase it and resend.

Sure (just will wait to the fix to be settled down first), thanks for review!

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#7 lore raw
Quoted 31 lines · Andy Shevchenko
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 18 lines
Quoted 18 lines
Quoted 17 lines
Quoted 6 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

Quoted 7 lines

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

Whatever, I don't have a strong opinion, just rebase it and resend.

Sure (just will wait to the fix to be settled down first), thanks for review!

I realized you haven't resent it after all, do you still want to change this?

Bart

#8 lore raw
Quoted 11 lines · Bartosz Golaszewski
Quoted 9 lines · Andy Shevchenko
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 18 lines
Quoted 18 lines
Quoted 18 lines
Quoted 18 lines
Quoted 17 lines
Quoted 6 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

Quoted 7 lines

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

Whatever, I don't have a strong opinion, just rebase it and resend.

Sure (just will wait to the fix to be settled down first), thanks for review!

I realized you haven't resent it after all, do you still want to change this?

Yes. U can prepare a new version later today.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#9 lore raw
In reply to Andy Shevchenko
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · Bartosz Golaszewski
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Bartosz Golaszewski

On Wed, Feb 21, 2024 at 8:36 PM Andy Shevchenko
<[email protected]> wrote:

...

Quoted 12 lines
Quoted 12 lines
Quoted 12 lines
Quoted 3 lines
Quoted 1 line

+ while (desc_index--)

What about gdev->descs[0]?

What about it? :-)

for (i = i - 1; i >= 0; i--)
while (--i >= 0)
while (i--)

are all equivalents.

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

The difference is what the value will i get _after_ the loop.

Ugh of course. But the first one is more readable given I got tricked by variant #3 at a quick glance but the for loop says out loud what it does.

I disagree. while (i--) is very well known cleanup pattern. Less letters to parse, easier to understand.

$ git grep -n 'while (i--)' | wc -l
298
$ git grep -n 'while (--i >= 0)' | wc -l
246
$ git grep -n 'for (--i; i >= 0; i--)' | wc -l
29
$ git grep -n 'for (i = i - 1; i >= 0; i--)' | wc -l
17
Signature · 2 lines
With Best Regards,
Andy Shevchenko