Skip to contentSkip to conversation

Fix the iio-gts-helpers available times table sorting

PATCH In review Matti Vaittinen 3 participants 9 messages Apr 30, 2024 – May 10, 2024

Subsystems (2)
lore
Share & tools
lore

Matti Vaittinen

author maintainer
#0 lore raw

v2 cover Fix the iio-gts-helpers available times table sorting

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

---

Chenyuan Yang (1):
iio: Fix the sorting functionality in iio_gts_build_avail_time_table

Matti Vaittinen (1):
iio: test: gts: test available times and gains sorting

Chenyuan Yang (1):
iio: Fix the sorting functionality in iio_gts_build_avail_time_table

Matti Vaittinen (1):
iio: test: gts: test available times and gains sorting

 drivers/iio/industrialio-gts-helper.c | 7 +++++--
 drivers/iio/test/iio-test-gts.c       | 8 +++++---
 2 files changed, 10 insertions(+), 5 deletions(-)

base-commit: 4cece764965020c22cff7665b18a012006359095

Signature · 14 lines
2.44.0


-- 
Matti Vaittinen, Linux device drivers
ROHM Semiconductors, Finland SWDC
Kiviharjunlenkki 1E
90220 OULU
FINLAND

~~~ "I don't think so," said Rene Descartes. Just then he vanished ~~~
Simon says - in Latin please.
~~~ "non cogito me" dixit Rene Descarte, deinde evanescavit ~~~
Thanks to Simon Glass for the translation =] 
author maintainer
#1 lore raw

PATCH v2 1/2 iio: Fix the sorting functionality in iio_gts_build_avail_time_table

From: Chenyuan Yang <[email protected]>

The sorting in iio_gts_build_avail_time_table is not working as intended. It could result in an out-of-bounds access when the time is zero.

Here are more details:

1. When the gts->itime_table[i].time_us is zero, e.g., the time sequence is 3, 0, 1, the inner for-loop will not terminate and do out-of-bound writes. This is because once times[j] > new, the value new will be added in the current position and the times[j] will be moved to j+1 position, which makes the if-condition always hold. Meanwhile, idx will be added one, making the loop keep running without termination and out-of-bound write.
2. If none of the gts->itime_table[i].time_us is zero, the elements will just be copied without being sorted as described in the comment "Sort times from all tables to one and remove duplicates".

For more details, please refer to
https://lore.kernel.org/all/[email protected].

Fixes 38416c28e168 ("iio: light: Add gain-time-scale helpers")

---
Revision history:
v1 => v2:
- Fix the sender and From: tag
- Drop the Co-developed-by tag from original author

Original commit (by Chenyuan Yang) was amended by me to remove duplicates so that the user-space callers do not get multiple instances of same time as was discussed here:
https://lore.kernel.org/all/[email protected]/

---
 drivers/iio/industrialio-gts-helper.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/iio/industrialio-gts-helper.c +5 -2
index b51eb6cb766f..59d7615c0f56 100644
@@ -362,17 +362,20 @@ static int iio_gts_build_avail_time_table(struct iio_gts *gts)
 	for (i = gts->num_itime - 1; i >= 0; i--) {
 		int new = gts->itime_table[i].time_us;
 
-		if (times[idx] < new) {
+		if (idx == 0 || times[idx - 1] < new) {
 			times[idx++] = new;
 			continue;
 		}
 
-		for (j = 0; j <= idx; j++) {
+		for (j = 0; j < idx; j++) {
+			if (times[j] == new)
+				break;
 			if (times[j] > new) {
 				memmove(&times[j + 1], &times[j],
 					(idx - j) * sizeof(int));
 				times[j] = new;
 				idx++;
+				break;
 			}
 		}
 	}
Signature · 14 lines
2.44.0


-- 
Matti Vaittinen, Linux device drivers
ROHM Semiconductors, Finland SWDC
Kiviharjunlenkki 1E
90220 OULU
FINLAND

~~~ "I don't think so," said Rene Descartes. Just then he vanished ~~~
Simon says - in Latin please.
~~~ "non cogito me" dixit Rene Descarte, deinde evanescavit ~~~
Thanks to Simon Glass for the translation =] 
author maintainer
#2 lore raw

PATCH v2 2/2 iio: test: gts: test available times and gains sorting

In reply to Matti Vaittinen

The iio_gts helpers build available times and scales tables based on the times and gains arrays given from the driver. The driver should be able to list all valid register values so that conversion from register valu to correct gain/time works for all supported register values.

It might be more convenient for drivers to list these times and gains in the order where they're listed in the data-sheet than ascending order.

However, for user who requests the supported scales / times it is more convenient to get the results in asscending order. Also, listing duplicated values is not meaning for the user.

Hence the GTS heler should do sorting and deduplication of the scales and times when it builds the tables listing the available times/scales. Note, currently duplicated gain values aren't handled by GTS-helpers.

Unsort the gain and time arrays in the test code, and add duplicates to time array in order to test the sorting and deduplicating works.

---
Revision history:
- No changes

 drivers/iio/test/iio-test-gts.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/iio/test/iio-test-gts.c +5 -3
index cf7ab773ea0b..5f16a7b5e6d4 100644
@@ -70,6 +70,7 @@
  */
 static struct iio_gts gts;
 
+/* Keep the gain and time tables unsorted to test the sorting */
 static const struct iio_gain_sel_pair gts_test_gains[] = {
 	GAIN_SCALE_GAIN(1, TEST_GSEL_1),
 	GAIN_SCALE_GAIN(4, TEST_GSEL_4),
@@ -79,16 +80,17 @@ static const struct iio_gain_sel_pair gts_test_gains[] = {
 	GAIN_SCALE_GAIN(256, TEST_GSEL_256),
 	GAIN_SCALE_GAIN(512, TEST_GSEL_512),
 	GAIN_SCALE_GAIN(1024, TEST_GSEL_1024),
-	GAIN_SCALE_GAIN(2048, TEST_GSEL_2048),
 	GAIN_SCALE_GAIN(4096, TEST_GSEL_4096),
+	GAIN_SCALE_GAIN(2048, TEST_GSEL_2048),
 #define HWGAIN_MAX 4096
 };
 
 static const struct iio_itime_sel_mul gts_test_itimes[] = {
-	GAIN_SCALE_ITIME_US(400 * 1000, TEST_TSEL_400, 8),
-	GAIN_SCALE_ITIME_US(200 * 1000, TEST_TSEL_200, 4),
 	GAIN_SCALE_ITIME_US(100 * 1000, TEST_TSEL_100, 2),
+	GAIN_SCALE_ITIME_US(400 * 1000, TEST_TSEL_400, 8),
+	GAIN_SCALE_ITIME_US(400 * 1000, TEST_TSEL_400, 8),
 	GAIN_SCALE_ITIME_US(50 * 1000, TEST_TSEL_50, 1),
+	GAIN_SCALE_ITIME_US(200 * 1000, TEST_TSEL_200, 4),
 #define TIMEGAIN_MAX 8
 };
 #define TOTAL_GAIN_MAX	(HWGAIN_MAX * TIMEGAIN_MAX)
Signature · 14 lines
2.44.0


-- 
Matti Vaittinen, Linux device drivers
ROHM Semiconductors, Finland SWDC
Kiviharjunlenkki 1E
90220 OULU
FINLAND

~~~ "I don't think so," said Rene Descartes. Just then he vanished ~~~
Simon says - in Latin please.
~~~ "non cogito me" dixit Rene Descarte, deinde evanescavit ~~~
Thanks to Simon Glass for the translation =] 
maintainer
#3 lore raw
In reply to Matti Vaittinen
Quoted 5 lines · On Tue, 30 Apr 2024 15:44:26 +0300 Matti Vaittinen <[email protected]>…

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Quoted 7 lines

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

Am I right in thinking this doesn't matter for existing drivers? As such not high priority for back porting?

I'll assume that and queue it up for 6.11. If someone shouts I can pull the fix forwards, but then we have the mess of chasing the testing in later.

Applied to the togreg branch of iio.git and pushed out as testing for 0-day to poke at it.

Thanks,

Jonathan

Quoted 22 lines

---

Chenyuan Yang (1):
iio: Fix the sorting functionality in iio_gts_build_avail_time_table

Matti Vaittinen (1):
iio: test: gts: test available times and gains sorting

Chenyuan Yang (1):
iio: Fix the sorting functionality in iio_gts_build_avail_time_table

Matti Vaittinen (1):
iio: test: gts: test available times and gains sorting

 drivers/iio/industrialio-gts-helper.c | 7 +++++--
 drivers/iio/test/iio-test-gts.c       | 8 +++++---
 2 files changed, 10 insertions(+), 5 deletions(-)

base-commit: 4cece764965020c22cff7665b18a012006359095

author maintainer
#4 lore raw
Quoted 16 lines · Jonathan Cameron
Quoted 5 lines · On Tue, 30 Apr 2024 15:44:26 +0300 Matti Vaittinen <[email protected]>…

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Quoted 7 lines

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

Quoted 1 line

Am I right in thinking this doesn't matter for existing drivers?

I think this is right. Only couple of in-tree drivers are using these helpers for now, and all of them sorted the tables already in driver.

Quoted 1 line

As such not high priority for back porting?

The bug is pretty nasty as it causes invalid memory accesses. Hence I'd like to see this landing in the longterm kernels. It seems to me the GTS helpers got merged in 6.4, so getting the fix backported to 6.6 might make sense.

Quoted 2 lines

I'll assume that and queue it up for 6.11. If someone shouts I can pull the fix forwards, but then we have the mess of chasing the testing in later.

I am sorry Jonathan but I'm not quite sure what you mean by "pulling fix forward", or what is the "mess of chasing the testing in later" :)

Quoted 2 lines

Applied to the togreg branch of iio.git and pushed out as testing for 0-day to poke at it.

Thanks! Appreciate your work as always!

Yours,
-- Matti

Signature · 5 lines
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~
maintainer
#5 lore raw
Quoted 29 lines · On Mon, 6 May 2024 08:09:27 +0300 Matti Vaittinen <[email protected]>…
Quoted 16 lines · Jonathan Cameron
Quoted 5 lines · On Tue, 30 Apr 2024 15:44:26 +0300 Matti Vaittinen <[email protected]>…

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Quoted 7 lines

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

Quoted 1 line

Am I right in thinking this doesn't matter for existing drivers?

I think this is right. Only couple of in-tree drivers are using these helpers for now, and all of them sorted the tables already in driver.

Quoted 1 line

As such not high priority for back porting?

The bug is pretty nasty as it causes invalid memory accesses. Hence I'd like to see this landing in the longterm kernels. It seems to me the GTS helpers got merged in 6.4, so getting the fix backported to 6.6 might make sense.

Quoted 5 lines
Quoted 2 lines

I'll assume that and queue it up for 6.11. If someone shouts I can pull the fix forwards, but then we have the mess of chasing the testing in later.

I am sorry Jonathan but I'm not quite sure what you mean by "pulling fix forward", or what is the "mess of chasing the testing in later" :)

Hmm. That was an odd choice of words :) I just meant that I could send the fix in the first set of fixes after 6.10-rc1 rather than waiting for 6.11.

For now I'll leave it queued for 6.11 on the basis there are a lot of ways a driver writer can cause similar out of bounds accesses and they should notice it not working during testing. So it 'should' not be a problem to not rush this in.

J

Quoted 9 lines
Quoted 3 lines

Applied to the togreg branch of iio.git and pushed out as testing for 0-day to poke at it.

Thanks! Appreciate your work as always!

Yours,
-- Matti

author maintainer
#6 lore raw
Quoted 32 lines · Jonathan Cameron
Quoted 29 lines · On Mon, 6 May 2024 08:09:27 +0300 Matti Vaittinen <[email protected]>…
Quoted 16 lines · Jonathan Cameron
Quoted 5 lines · On Tue, 30 Apr 2024 15:44:26 +0300 Matti Vaittinen <[email protected]>…

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Quoted 7 lines

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

Quoted 1 line

Am I right in thinking this doesn't matter for existing drivers?

I think this is right. Only couple of in-tree drivers are using these helpers for now, and all of them sorted the tables already in driver.

Quoted 1 line

As such not high priority for back porting?

The bug is pretty nasty as it causes invalid memory accesses. Hence I'd like to see this landing in the longterm kernels. It seems to me the GTS helpers got merged in 6.4, so getting the fix backported to 6.6 might make sense.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

I'll assume that and queue it up for 6.11. If someone shouts I can pull the fix forwards, but then we have the mess of chasing the testing in later.

I am sorry Jonathan but I'm not quite sure what you mean by "pulling fix forward", or what is the "mess of chasing the testing in later" :)

Hmm. That was an odd choice of words :) I just meant that I could send the fix in the first set of fixes after 6.10-rc1 rather than waiting for 6.11.

Oh, right :)

Quoted 5 lines

For now I'll leave it queued for 6.11 on the basis there are a lot of ways a driver writer can cause similar out of bounds accesses and they should notice it not working during testing. So it 'should' not be a problem to not rush this in.

I guess this means the 6.10 won't have the fix? I believe this is fine - assuming the 6.10 is not going to be an LTS. Thanks for taking care of this! :)

Yours,
-- Matti

Signature · 5 lines
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~
#7 lore raw
Quoted 50 lines · On Tue, 7 May 2024 09:14:15 +0300 Matti Vaittinen <[email protected]>…
Quoted 32 lines · Jonathan Cameron
Quoted 29 lines · On Mon, 6 May 2024 08:09:27 +0300 Matti Vaittinen <[email protected]>…
Quoted 16 lines · Jonathan Cameron
Quoted 5 lines · On Tue, 30 Apr 2024 15:44:26 +0300 Matti Vaittinen <[email protected]>…

Fix the available times table sorting in iio-gts-helpers

This series contains a fix and test for the sorting of the available times in IIO-gts helpers. Fix was originally developed and posted by Chenyuan Yang.

Quoted 7 lines

Revision history:
v1 => v2:
- Fix the sender for patch 1/2 (Sic!)
- Fix Co-Developed-by tag (drop this from Chenyuan who is the original author)
- Fix the From: tag as instructed in:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html

Quoted 1 line

Am I right in thinking this doesn't matter for existing drivers?

I think this is right. Only couple of in-tree drivers are using these helpers for now, and all of them sorted the tables already in driver.

Quoted 1 line

As such not high priority for back porting?

The bug is pretty nasty as it causes invalid memory accesses. Hence I'd like to see this landing in the longterm kernels. It seems to me the GTS helpers got merged in 6.4, so getting the fix backported to 6.6 might make sense.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

I'll assume that and queue it up for 6.11. If someone shouts I can pull the fix forwards, but then we have the mess of chasing the testing in later.

I am sorry Jonathan but I'm not quite sure what you mean by "pulling fix forward", or what is the "mess of chasing the testing in later" :)

Hmm. That was an odd choice of words :) I just meant that I could send the fix in the first set of fixes after 6.10-rc1 rather than waiting for 6.11.

Oh, right :)

Quoted 5 lines

For now I'll leave it queued for 6.11 on the basis there are a lot of ways a driver writer can cause similar out of bounds accesses and they should notice it not working during testing. So it 'should' not be a problem to not rush this in.

Quoted 3 lines

I guess this means the 6.10 won't have the fix? I believe this is fine - assuming the 6.10 is not going to be an LTS. Thanks for taking care of this! :)

It may well get backported anyway, but after 6.11 merge window.

J

Quoted 5 lines

Yours,
-- Matti

author maintainer
#8 lore raw
Quoted 9 lines · Jonathan Cameron
Quoted 7 lines · On Tue, 7 May 2024 09:14:15 +0300 Matti Vaittinen <[email protected]>…
Quoted 6 lines · Jonathan Cameron
Quoted 4 lines · On Mon, 6 May 2024 08:09:27 +0300 Matti Vaittinen <[email protected]>…
Quoted 3 lines · Jonathan Cameron

On Tue, 30 Apr 2024 15:44:26 +0300
Matti Vaittinen <[email protected]> wrote:
>>> For now I'll leave it queued for 6.11 on the basis there are a lot

of ways

Quoted 9 lines
Quoted 8 lines
Quoted 4 lines

a driver writer can cause similar out of bounds accesses and they should notice it not working during testing. So it 'should' not be a problem to not rush this in.

I guess this means the 6.10 won't have the fix? I believe this is fine - assuming the 6.10 is not going to be an LTS. Thanks for taking care of this! :)

It may well get backported anyway, but after 6.11 merge window.

This sounds good. Thanks for clarifying!

Yours,
-- Matti

Signature · 5 lines
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~