[jallib build] buildbot failure in jallib on jallib-standard

[bu...@sirloon.net]

Hi guys,

This is buildbot speaking. I have finished a build of jallib-standard on jallib.
Buildslave for this Build: sebbot

Build Reason: The web-page 'force build' button was pressed by 'seb': force

Build Source Stamp: HEAD
Blamelist:

BUILD FAILED: failed validate
Logs are attached.

sincerely,
-The Buildbot

[Rob Hamerling]

Hi Seb,

On 4-11-12 10:33, bu...@sirloon.net wrote:

> This is buildbot speaking. I have finished a build of jallib-standard on jallib.
> Buildslave for this Build: sebbot
>
> Build Reason: The web-page 'force build' button was pressed by 'seb': force
>
> Build Source Stamp: HEAD
> Blamelist:
>
> BUILD FAILED: failed validate
> Logs are attached.

Thanks for correcting pwm_common (my fault, forgot once again to validate!).

I see that buildbot still uses compiler version 2.4o, while 2.4p is
available since 2 months. Please update buildbot.

Regards, Rob.



--
R. Hamerling, Netherlands --- http://www.robh.nl

[Sebastien Lelong]

Hi Rob,

Sure. I replaced with last compiler, re-run a standard build then will trigger a weekly bee build.

PS: server is still aving troubles, buildbot can't seem to run properly anymore (frequent reboot it seems, unstable)

Cheers

Seb

--
You received this message because you are subscribed to the Google Groups "jallib" group.
To post to this group, send email to jal...@googlegroups.com.
To unsubscribe from this group, send email to jallib+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/jallib?hl=en.

--
Sébastien Lelong

[Rob Hamerling]

On 4-11-12 10:51, Sebastien Lelong wrote:

> I replaced with last compiler, re-run a standard build then will
> trigger a weekly bee build.

I don't understand: when I compile the samples 48 of them report
warnings from the new PWM libaries (because of using deprecated
procedures). Does buildbot ignore these warnings?

I see that I have to update some of my samples, but most of the PWM
samples are created from board and test files by Joep. Could you have a
look Joep?

[bu...@sirloon.net]

Hi guys,

This is buildbot speaking. I have finished a build of jallib-standard on jallib.
Buildslave for this Build: sebbot

Build Reason: The web-page 'force build' button was pressed by 'seb': force with last compiler

Build Source Stamp: HEAD
Blamelist:

BUILD FAILED: failed unittest

[Sebastien Lelong]

this is due to a timeout (it's taking too much time, at least more than usual). I'll modify timeout setting and rerun again

seb

--
You received this message because you are subscribed to the Google Groups "jallib" group.
To post to this group, send email to jal...@googlegroups.com.

To unsubscribe from this group, send email to jallib+un...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/jallib?hl=en.

--
Sébastien Lelong

[Sebastien Lelong]

IIRC buildbot won't complain about warnings during compilation, because exit code is no error.

cheers

seb

--
You received this message because you are subscribed to the Google Groups "jallib" group.
To post to this group, send email to jal...@googlegroups.com.
To unsubscribe from this group, send email to jallib+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/jallib?hl=en.

--
Sébastien Lelong

[bu...@sirloon.net]

Hi guys,

This is buildbot speaking. I have finished a build of jallib-standard on jallib.
Buildslave for this Build: sebbot

Build Reason: The web-page 'force build' button was pressed by 'seb': rereforce

[Joep Suijs]

Hi Rob,

warnings are gone as far as I can see.
Didn't actually run any code (as a matter of fact, I have not run any
code from the new compiler yet), but the changes are small & compile ,
hence most likely okay.

Regards,
Joep

2012/11/4 Rob Hamerling <robham...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups
> "jallib" group.
> To post to this group, send email to jal...@googlegroups.com.
> To unsubscribe from this group, send email to

> jallib+un...@googlegroups.com.

[Rob Hamerling]

Hi Joep,

On 14-11-12 22:47, Joep Suijs wrote:
>
> warnings are gone as far as I can see.

I'm afraid the following samples have some issues still:

12hv615_blink
12hv615_temperature_fan
16f648a_backlight
16f648a_pwm_led
16f723_pwm_led
16f73_pwm2
16f73_pwm_led
16f877a_backlight
16f877a_pwm2
16f877a_pwm_led
16f877a_servo_rc_slave
16f877_backlight
16f877_pwm2
16f877_pwm_led
16f877_servo_rc_slave
16f886_pwm_adc_res
16f88_backlight
16f88_ir_ranger_gp2d02
16f88_pwm_led
18f14k50_pwm_led
18f2450_pwm_led
18f2620_pwm_led
18f27j53_sd_card_minix_read
18f4525_servo_rc_slave
18f452_servo_rc_slave
18f4550_pwm_led
18f4550_servo_rc_slave
18f4620_servo_rc_slave

I'll try to analyse these ASAP. If possible I'll correct 'm.
I've no objections against help by the original author....

> Didn't actually run any code (as a matter of fact, I have not run any

> code from the new compiler yet), but the changes are small& compile ,
> hence most likely okay.

There are probably a few samples which have to be modified.
Some samples use a (byte) ADC-value (0..255) and feed it directly to a
(now deprecated) PWM procedure. This byte value has to be changed to one
of these:
- adjusted properly to a word value 0..1023 for ...pwm_ratio()
- converted to a percentage 0..100 for ...pwm_precent()

Regards, Rob

[Joep Suijs]

Hi Rob,

I had some difficulty in finding the actual issues. What I did is
regenerate all samples and check for 'deprecated' in the redirected
output.

If there are still warnings in the re-generated samples, I'll check
again. A reference to the test-file involved would be helpful.

I've given up on samples that have been generated in the past, but are
not generated any more. It's quite time-consuming to manage them in
the current structure and various attempts to adapt the structure of
samples have been unsuccessful.
I suggest we remove all samples in this category that have warnings.

> There are probably a few samples which have to be modified.
> Some samples use a (byte) ADC-value (0..255) and feed it directly to a (now
> deprecated) PWM procedure. This byte value has to be changed to one of
> these:
> - adjusted properly to a word value 0..1023 for ...pwm_ratio()
> - converted to a percentage 0..100 for ...pwm_precent()

I made those changes in the process, so that should be okay.

Joep

[Rob Hamerling]

Hi Joep, Matt,

[Joep]

> I had some difficulty in finding the actual issues. What I did is
> regenerate all samples and check for 'deprecated' in the redirected
> output.
>
> If there are still warnings in the re-generated samples, I'll check
> again. A reference to the test-file involved would be helpful.

I work similarly: compile the samples with compiler-log file enabled and
check the log.

> I've given up on samples that have been generated in the past, but are
> not generated any more. It's quite time-consuming to manage them in
> the current structure and various attempts to adapt the structure of
> samples have been unsuccessful.
> I suggest we remove all samples in this category that have warnings.

Fine with me! As far as I'm concerned: remove 'm.


Could you have a look at library lcd_backlight_pwm? It looks like you
revised pwm1 (which still has an error) but you did not not revise the
parts for pwm2, 3, and 4.


Matt,
I see a few warnings in your servo samples, but when I look in the
sample program I would have expected much more warnings. It seems some
code is 'optimized' out, which hides possible issues. Could you have a
look please?


Regards, Rob

[Joep Suijs]

Hi Rob,

>> I've given up on samples that have been generated in the past, but are
>> not generated any more. It's quite time-consuming to manage them in
>> the current structure and various attempts to adapt the structure of
>> samples have been unsuccessful.
>> I suggest we remove all samples in this category that have warnings.
>
>
> Fine with me! As far as I'm concerned: remove 'm.

I removed them from my svn but but when I addressed:

> Could you have a look at library lcd_backlight_pwm? It looks like you
> revised pwm1 (which still has an error) but you did not not revise the parts
> for pwm2, 3, and 4.

it turned out that all samples are regenerated and all seem to compile
without warning.

Joep