[PATCH] fixup! test: Add pytest suite for NVMEM framework

[PATCH] fixup! test: Add pytest suite for NVMEM framework

[Ahmad Fatoum]

If CONFIG_DEBUG_PROBES is enabled, barebox will print an info log
whenever a device is probed. This breaks one test case in the newly
introduced nvmem tests, because running the nvmem command results in two
lines instead of one:

      probe-> rmem1
  rmem1

A better solution might be disabling stderr, but for now, let's just
have the test check all lines that appeared after running the command.

This fixes CI breakage in next.

Cc: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
---
 test/py/test_nvmem.py | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/test/py/test_nvmem.py b/test/py/test_nvmem.py
index 0eec93d65c4c..6070877a6868 100644
--- a/test/py/test_nvmem.py
+++ b/test/py/test_nvmem.py
@@ -257,13 +257,7 @@ def test_nvmem_create_dynamic_device_with_var(barebox, barebox_config):
 
     stdout_compound_raw = barebox.run_check(compound_cmd)
 
-    actual_dev_name_in_var = stdout_compound_raw[0].strip() if stdout_compound_raw else "'' (compound cmd gave no stdout lines)"
-
-    assert actual_dev_name_in_var == expected_dev_name_in_var, (
-        f"Environment variable {var_to_set_in_barebox} (from compound cmd) was "
-        f"'{actual_dev_name_in_var}', expected "
-        f"'{expected_dev_name_in_var}'"
-    )
+    assert expected_dev_name_in_var in stdout_compound_raw
 
     # Verify the device is listed by `nvmem`
     stdout_list_a, _, returncode = barebox.run('nvmem')
@@ -271,7 +265,7 @@ def test_nvmem_create_dynamic_device_with_var(barebox, barebox_config):
 
     # The 'nvmem' list might show "rmemX0" while the var stores "rmemX".
     # The pattern checks for this.
-    list_pattern_check = actual_dev_name_in_var
+    list_pattern_check = expected_dev_name_in_var
     pattern = re.compile(f"^{re.escape(list_pattern_check)}0?$")
     assert any(pattern.match(line.strip()) for line in stdout_list_a), (
         f"Device corresponding to '{list_pattern_check}' (pattern: '{pattern.pattern}') "
-- 
2.39.5

Re: [PATCH] fixup! test: Add pytest suite for NVMEM framework

[Oleksij Rempel]

On Tue, Jun 24, 2025 at 10:17:12AM +0200, Ahmad Fatoum wrote:
> If CONFIG_DEBUG_PROBES is enabled, barebox will print an info log
> whenever a device is probed. This breaks one test case in the newly
> introduced nvmem tests, because running the nvmem command results in two
> lines instead of one:
> 
>       probe-> rmem1
>   rmem1
> 
> A better solution might be disabling stderr, but for now, let's just
> have the test check all lines that appeared after running the command.
> 
> This fixes CI breakage in next.
> 
> Cc: Oleksij Rempel <o.rempel@pengutronix.de>
> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>

LGTM

May be better to keep it as separate patch, not as fixup? Since It provides
important information.

Best Regards,
Oleksij
-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Re: [PATCH] fixup! test: Add pytest suite for NVMEM framework

[Ahmad Fatoum]

Hi,

On 6/24/25 10:30, Oleksij Rempel wrote:
> On Tue, Jun 24, 2025 at 10:17:12AM +0200, Ahmad Fatoum wrote:
>> If CONFIG_DEBUG_PROBES is enabled, barebox will print an info log
>> whenever a device is probed. This breaks one test case in the newly
>> introduced nvmem tests, because running the nvmem command results in two
>> lines instead of one:
>>
>>       probe-> rmem1
>>   rmem1
>>
>> A better solution might be disabling stderr, but for now, let's just
>> have the test check all lines that appeared after running the command.
>>
>> This fixes CI breakage in next.
>>
>> Cc: Oleksij Rempel <o.rempel@pengutronix.de>
>> Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> 
> LGTM
> 
> May be better to keep it as separate patch, not as fixup? Since It provides
> important information.

Fine with me either way. If Sascha prefers a separate patch I would
suggest as commit message title:

  test: nvmem: fix test failure with CONFIG_DEBUG_PROBES

Cheers,
Ahmad

> 
> Best Regards,
> Oleksij

-- 
Pengutronix e.K.                  |                             |
Steuerwalder Str. 21              | http://www.pengutronix.de/  |
31137 Hildesheim, Germany         | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686  | Fax:   +49-5121-206917-5555 |

Re: [PATCH] fixup! test: Add pytest suite for NVMEM framework

[Sascha Hauer]

On Tue, 24 Jun 2025 10:17:12 +0200, Ahmad Fatoum wrote:
> If CONFIG_DEBUG_PROBES is enabled, barebox will print an info log
> whenever a device is probed. This breaks one test case in the newly
> introduced nvmem tests, because running the nvmem command results in two
> lines instead of one:
> 
>       probe-> rmem1
>   rmem1
> 
> [...]

Applied, thanks!

[1/1] fixup! test: Add pytest suite for NVMEM framework
      https://git.pengutronix.de/cgit/barebox/commit/?id=b90dad565dbe (link may not be stable)

Best regards,
-- 
Sascha Hauer <s.hauer@pengutronix.de>

Re: [PATCH] fixup! test: Add pytest suite for NVMEM framework

[Sascha Hauer]

On Tue, Jun 24, 2025 at 10:30:14AM +0200, Oleksij Rempel wrote:
> On Tue, Jun 24, 2025 at 10:17:12AM +0200, Ahmad Fatoum wrote:
> > If CONFIG_DEBUG_PROBES is enabled, barebox will print an info log
> > whenever a device is probed. This breaks one test case in the newly
> > introduced nvmem tests, because running the nvmem command results in two
> > lines instead of one:
> > 
> >       probe-> rmem1
> >   rmem1
> > 
> > A better solution might be disabling stderr, but for now, let's just
> > have the test check all lines that appeared after running the command.
> > 
> > This fixes CI breakage in next.
> > 
> > Cc: Oleksij Rempel <o.rempel@pengutronix.de>
> > Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> 
> LGTM
> 
> May be better to keep it as separate patch, not as fixup? Since It provides
> important information.

I agree it provides important information, I just don't see how one
would find it when he needs it, so I decided to keep this as a fixup.

Sascha

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |