All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] test: boot: Bind bootmeth_{cros,android} only once per run
@ 2024-11-21 15:02 Mattijs Korpershoek
  2024-11-22 13:37 ` [PATCH] test: boot: Bind bootmeth_{cros, android} " Simon Glass
  0 siblings, 1 reply; 3+ messages in thread
From: Mattijs Korpershoek @ 2024-11-21 15:02 UTC (permalink / raw)
  To: Tom Rini, Simon Glass; +Cc: Guillaume La Roque, u-boot, Mattijs Korpershoek

We make fewer calls to dm_test_restore() since
commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")

Because of this some valid test combinations are now broken:

$ ./test/py/test.py --bd sandbox --build -k test_ut
$ ./test/py/test.py --bd sandbox --build -k "bootflow_android or bootflow_cros"

Shows:

  Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
  got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '

Here prep_mmc_bootdev() is called twice and it will bind bootmeth_cros twice.

Since bootmeth_cros is bound twice, 'bootflow scan' will find 2x the
expected bootflows.

Before
commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
this did not happen because a cleanup was called each time.

Check if the bootstd already has a "cros" or "android" driver and only bind it
if it does not exist.

Fixes: fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
---
Initially, this was found when merging some Android bootflow patches
from [1].

However, this can be reproduced on 'next' as well.

Here is the test output that's broken (we have 4 cros bootmethods instead of 2)

=> ut bootstd bootflow_cros
Test: bootflow_cros: bootflow.c
Showing all bootflows
Seq  Method       State   Uclass    Part  Name                      Filename
---  -----------  ------  --------  ----  ------------------------ ----------------
0  extlinux     ready   mmc          1  mmc1.bootdev.part_1 /extlinux/extlinux.conf
1  cros         ready   mmc          2  mmc5.bootdev.part_2
2  cros         ready   mmc          2  mmc5.bootdev.part_2
3  cros         ready   mmc          4  mmc5.bootdev.part_4
4  cros         ready   mmc          4  mmc5.bootdev.part_4
---  -----------  ------  --------  ----  ------------------------ ----------------
(5 bootflows, 5 valid)
test/boot/bootflow.c:1192, bootflow_cros(): console:
Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '
Test bootflow_cros failed 1 times

[1] https://lore.kernel.org/r/all/87mshvxhq3.fsf@baylibre.com/
---
 test/boot/bootflow.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/test/boot/bootflow.c b/test/boot/bootflow.c
index 9397328609d0..835d2e6e35a6 100644
--- a/test/boot/bootflow.c
+++ b/test/boot/bootflow.c
@@ -552,15 +552,21 @@ static int prep_mmc_bootdev(struct unit_test_state *uts, const char *mmc_dev,
 	/* Enable the cros bootmeth if needed */
 	if (IS_ENABLED(CONFIG_BOOTMETH_CROS) && bind_cros_android) {
 		ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
-		ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
-					"cros", 0, ofnode_null(), &dev));
+		/* Only bind the bootmeth once to avoid duplicate scan results */
+		if (device_find_child_by_name(bootstd, "cros", &dev) == -ENODEV) {
+			ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
+						"cros", 0, ofnode_null(), &dev));
+		}
 	}
 
 	/* Enable the android bootmeths if needed */
 	if (IS_ENABLED(CONFIG_BOOTMETH_ANDROID) && bind_cros_android) {
 		ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
-		ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
-					"android", 0, ofnode_null(), &dev));
+		/* Only bind the bootmeth once to avoid duplicate scan results */
+		if (device_find_child_by_name(bootstd, "android", &dev) == -ENODEV) {
+			ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
+						"android", 0, ofnode_null(), &dev));
+		}
 	}
 
 	/* Change the order to include the device */

---
base-commit: dc1859f8d2ac3faaa5e2e1d465ec4bd8980520a5
change-id: 20241121-bootstd-test-fix-multiple-bind-159bd2523414

Best regards,
-- 
Mattijs Korpershoek <mkorpershoek@baylibre.com>


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] test: boot: Bind bootmeth_{cros, android} only once per run
  2024-11-21 15:02 [PATCH] test: boot: Bind bootmeth_{cros,android} only once per run Mattijs Korpershoek
@ 2024-11-22 13:37 ` Simon Glass
  2024-11-22 14:05   ` Mattijs Korpershoek
  0 siblings, 1 reply; 3+ messages in thread
From: Simon Glass @ 2024-11-22 13:37 UTC (permalink / raw)
  To: Mattijs Korpershoek; +Cc: Tom Rini, Guillaume La Roque, u-boot

Hi Mattijs,

On Thu, 21 Nov 2024 at 08:03, Mattijs Korpershoek
<mkorpershoek@baylibre.com> wrote:
>
> We make fewer calls to dm_test_restore() since
> commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
>
> Because of this some valid test combinations are now broken:
>
> $ ./test/py/test.py --bd sandbox --build -k test_ut
> $ ./test/py/test.py --bd sandbox --build -k "bootflow_android or bootflow_cros"
>
> Shows:
>
>   Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
>   got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '
>
> Here prep_mmc_bootdev() is called twice and it will bind bootmeth_cros twice.
>
> Since bootmeth_cros is bound twice, 'bootflow scan' will find 2x the
> expected bootflows.
>
> Before
> commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
> this did not happen because a cleanup was called each time.
>
> Check if the bootstd already has a "cros" or "android" driver and only bind it
> if it does not exist.
>
> Fixes: fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
> Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
> ---
> Initially, this was found when merging some Android bootflow patches
> from [1].
>
> However, this can be reproduced on 'next' as well.
>
> Here is the test output that's broken (we have 4 cros bootmethods instead of 2)
>
> => ut bootstd bootflow_cros
> Test: bootflow_cros: bootflow.c
> Showing all bootflows
> Seq  Method       State   Uclass    Part  Name                      Filename
> ---  -----------  ------  --------  ----  ------------------------ ----------------
> 0  extlinux     ready   mmc          1  mmc1.bootdev.part_1 /extlinux/extlinux.conf
> 1  cros         ready   mmc          2  mmc5.bootdev.part_2
> 2  cros         ready   mmc          2  mmc5.bootdev.part_2
> 3  cros         ready   mmc          4  mmc5.bootdev.part_4
> 4  cros         ready   mmc          4  mmc5.bootdev.part_4
> ---  -----------  ------  --------  ----  ------------------------ ----------------
> (5 bootflows, 5 valid)
> test/boot/bootflow.c:1192, bootflow_cros(): console:
> Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
> got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '
> Test bootflow_cros failed 1 times
>
> [1] https://lore.kernel.org/r/all/87mshvxhq3.fsf@baylibre.com/
> ---
>  test/boot/bootflow.c | 14 ++++++++++----
>  1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/test/boot/bootflow.c b/test/boot/bootflow.c
> index 9397328609d0..835d2e6e35a6 100644
> --- a/test/boot/bootflow.c
> +++ b/test/boot/bootflow.c
> @@ -552,15 +552,21 @@ static int prep_mmc_bootdev(struct unit_test_state *uts, const char *mmc_dev,
>         /* Enable the cros bootmeth if needed */
>         if (IS_ENABLED(CONFIG_BOOTMETH_CROS) && bind_cros_android) {
>                 ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
> -               ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
> -                                       "cros", 0, ofnode_null(), &dev));
> +               /* Only bind the bootmeth once to avoid duplicate scan results */
> +               if (device_find_child_by_name(bootstd, "cros", &dev) == -ENODEV) {
> +                       ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
> +                                               "cros", 0, ofnode_null(), &dev));
> +               }
>         }
>
>         /* Enable the android bootmeths if needed */
>         if (IS_ENABLED(CONFIG_BOOTMETH_ANDROID) && bind_cros_android) {
>                 ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
> -               ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
> -                                       "android", 0, ofnode_null(), &dev));
> +               /* Only bind the bootmeth once to avoid duplicate scan results */
> +               if (device_find_child_by_name(bootstd, "android", &dev) == -ENODEV) {
> +                       ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
> +                                               "android", 0, ofnode_null(), &dev));
> +               }
>         }
>
>         /* Change the order to include the device */
>
> ---
> base-commit: dc1859f8d2ac3faaa5e2e1d465ec4bd8980520a5
> change-id: 20241121-bootstd-test-fix-multiple-bind-159bd2523414
>
> Best regards,
> --
> Mattijs Korpershoek <mkorpershoek@baylibre.com>
>

Instead of this, could you add flags to the test to tell it to reset
driver mode?

UTF_DM | UTF_SCAN_FDT

Regards,
Simon

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] test: boot: Bind bootmeth_{cros, android} only once per run
  2024-11-22 13:37 ` [PATCH] test: boot: Bind bootmeth_{cros, android} " Simon Glass
@ 2024-11-22 14:05   ` Mattijs Korpershoek
  0 siblings, 0 replies; 3+ messages in thread
From: Mattijs Korpershoek @ 2024-11-22 14:05 UTC (permalink / raw)
  To: Simon Glass; +Cc: Tom Rini, Guillaume La Roque, u-boot

Hi Simon,

On ven., nov. 22, 2024 at 06:37, Simon Glass <sjg@chromium.org> wrote:

> Hi Mattijs,
>
> On Thu, 21 Nov 2024 at 08:03, Mattijs Korpershoek
> <mkorpershoek@baylibre.com> wrote:
>>
>> We make fewer calls to dm_test_restore() since
>> commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
>>
>> Because of this some valid test combinations are now broken:
>>
>> $ ./test/py/test.py --bd sandbox --build -k test_ut
>> $ ./test/py/test.py --bd sandbox --build -k "bootflow_android or bootflow_cros"
>>
>> Shows:
>>
>>   Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
>>   got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '
>>
>> Here prep_mmc_bootdev() is called twice and it will bind bootmeth_cros twice.
>>
>> Since bootmeth_cros is bound twice, 'bootflow scan' will find 2x the
>> expected bootflows.
>>
>> Before
>> commit fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
>> this did not happen because a cleanup was called each time.
>>
>> Check if the bootstd already has a "cros" or "android" driver and only bind it
>> if it does not exist.
>>
>> Fixes: fbdac8155c89 ("test: Expand implementation of ut_list_has_dm_tests()")
>> Signed-off-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
>> ---
>> Initially, this was found when merging some Android bootflow patches
>> from [1].
>>
>> However, this can be reproduced on 'next' as well.
>>
>> Here is the test output that's broken (we have 4 cros bootmethods instead of 2)
>>
>> => ut bootstd bootflow_cros
>> Test: bootflow_cros: bootflow.c
>> Showing all bootflows
>> Seq  Method       State   Uclass    Part  Name                      Filename
>> ---  -----------  ------  --------  ----  ------------------------ ----------------
>> 0  extlinux     ready   mmc          1  mmc1.bootdev.part_1 /extlinux/extlinux.conf
>> 1  cros         ready   mmc          2  mmc5.bootdev.part_2
>> 2  cros         ready   mmc          2  mmc5.bootdev.part_2
>> 3  cros         ready   mmc          4  mmc5.bootdev.part_4
>> 4  cros         ready   mmc          4  mmc5.bootdev.part_4
>> ---  -----------  ------  --------  ----  ------------------------ ----------------
>> (5 bootflows, 5 valid)
>> test/boot/bootflow.c:1192, bootflow_cros(): console:
>> Expected '  2  cros         ready   mmc          4 mmc5.bootdev.part_4       ',
>> got '  2  cros         ready   mmc          2 mmc5.bootdev.part_2       '
>> Test bootflow_cros failed 1 times
>>
>> [1] https://lore.kernel.org/r/all/87mshvxhq3.fsf@baylibre.com/
>> ---
>>  test/boot/bootflow.c | 14 ++++++++++----
>>  1 file changed, 10 insertions(+), 4 deletions(-)
>>
>> diff --git a/test/boot/bootflow.c b/test/boot/bootflow.c
>> index 9397328609d0..835d2e6e35a6 100644
>> --- a/test/boot/bootflow.c
>> +++ b/test/boot/bootflow.c
>> @@ -552,15 +552,21 @@ static int prep_mmc_bootdev(struct unit_test_state *uts, const char *mmc_dev,
>>         /* Enable the cros bootmeth if needed */
>>         if (IS_ENABLED(CONFIG_BOOTMETH_CROS) && bind_cros_android) {
>>                 ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
>> -               ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
>> -                                       "cros", 0, ofnode_null(), &dev));
>> +               /* Only bind the bootmeth once to avoid duplicate scan results */
>> +               if (device_find_child_by_name(bootstd, "cros", &dev) == -ENODEV) {
>> +                       ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_cros),
>> +                                               "cros", 0, ofnode_null(), &dev));
>> +               }
>>         }
>>
>>         /* Enable the android bootmeths if needed */
>>         if (IS_ENABLED(CONFIG_BOOTMETH_ANDROID) && bind_cros_android) {
>>                 ut_assertok(uclass_first_device_err(UCLASS_BOOTSTD, &bootstd));
>> -               ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
>> -                                       "android", 0, ofnode_null(), &dev));
>> +               /* Only bind the bootmeth once to avoid duplicate scan results */
>> +               if (device_find_child_by_name(bootstd, "android", &dev) == -ENODEV) {
>> +                       ut_assertok(device_bind(bootstd, DM_DRIVER_REF(bootmeth_android),
>> +                                               "android", 0, ofnode_null(), &dev));
>> +               }
>>         }
>>
>>         /* Change the order to include the device */
>>
>> ---
>> base-commit: dc1859f8d2ac3faaa5e2e1d465ec4bd8980520a5
>> change-id: 20241121-bootstd-test-fix-multiple-bind-159bd2523414
>>
>> Best regards,
>> --
>> Mattijs Korpershoek <mkorpershoek@baylibre.com>
>>
>
> Instead of this, could you add flags to the test to tell it to reset
> driver mode?
>
> UTF_DM | UTF_SCAN_FDT

That works as well, and seems like a cleaner approach. Thanks a lot for
the suggestion.

Will send a v2.

>
> Regards,
> Simon

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2024-11-22 14:06 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-11-21 15:02 [PATCH] test: boot: Bind bootmeth_{cros,android} only once per run Mattijs Korpershoek
2024-11-22 13:37 ` [PATCH] test: boot: Bind bootmeth_{cros, android} " Simon Glass
2024-11-22 14:05   ` Mattijs Korpershoek

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.