Skip to content

Add device_name for TB_SENSE_12. #5470

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Dec 20, 2017
Merged

Add device_name for TB_SENSE_12. #5470

merged 2 commits into from
Dec 20, 2017

Conversation

ashok-rao
Copy link
Contributor

Description

device_name key is required to add Cloud Client support / enable bootloader support for this target.

Status

READY

Deploy notes

None.

device_name key is required to add Cloud Client support / enable bootloader support for this target.
@ashok-rao
Copy link
Contributor Author

ashok-rao commented Nov 9, 2017

@stevew817 @0xc0170 ..Could you please check? Thank you.

cc @screamerbg

@0xc0170
Copy link
Contributor

0xc0170 commented Nov 10, 2017

/morph build

@stevew817
Copy link
Contributor

LGTM

@mbed-ci
Copy link

mbed-ci commented Nov 10, 2017

Build : SUCCESS

Build number : 486
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/5470/

Triggering tests

/morph test
/morph uvisor-test
/morph export-build

@mbed-ci
Copy link

mbed-ci commented Nov 10, 2017

@mbed-ci
Copy link

mbed-ci commented Nov 10, 2017

@stevew817
Copy link
Contributor

@0xc0170 The Keil failure above looks weird... Apparently it can't deal with .S files?

@ashok-rao
Copy link
Contributor Author

@theotherjimmy ..does this exporter failure above have something to do with .s and .S as seen in the past? the PR is only adding "device_name" to targets.json..

@0xc0170
Copy link
Contributor

0xc0170 commented Nov 10, 2017

/morph export-build

@amq
Copy link
Contributor

amq commented Nov 10, 2017

There is one more target which needs device_name, both of them are covered in #5253

@mbed-ci
Copy link

mbed-ci commented Nov 10, 2017

@theotherjimmy
Copy link
Contributor

@ashok-rao Those extensions are only distinguished on POSIX, and uvision only runs on Windows. So, I hope not.

I have seen this before, but I don't recall what the solution was. I'll have to take a look in more detail on Monday.

@ashok-rao
Copy link
Contributor Author

@theotherjimmy .. I see.. but I remember I had this issue on Windows some time back and I think others also faced this on their Windows machine.. Issue : #5145 .. Please let us know what you find on Monday! Thanks!

@theotherjimmy theotherjimmy changed the title Adding device_name for TB_SENSE_12. Add device_name for TB_SENSE_12. Nov 13, 2017
@ashok-rao
Copy link
Contributor Author

@theotherjimmy @0xc0170 .. why is only uVision failing for this? I can see others are passing from Jenkins log...??

@theotherjimmy
Copy link
Contributor

@ashok-rao This change only affects the Uvision Exporter. I would be surprised if others failed because of this.

@0xc0170
Copy link
Contributor

0xc0170 commented Nov 14, 2017

@ashok-rao Can you reproduce this issue locally please? Use export -t TB_SENSE_12 -i uvision

@ashok-rao
Copy link
Contributor Author

ashok-rao commented Nov 14, 2017

Hi @0xc0170 . It seems to build fine on my local machine.. Steps I followed:
I cloned the latest master of Mbed OS:

C:\Ashok\blinky\mbed-os-example-blinky\mbed-os>git log
commit 1394bf95f19fdb0cbf569359bbac0b02efec5d87

Made the change..
tb_sense_12_add_device_name

then did mbed export :

C:\Ashok\blinky\mbed-os-example-blinky>mbed export -i uvision -m TB_SENSE_12 -vv
[mbed] Working path "C:\Ashok\blinky\mbed-os-example-blinky" (program)
[mbed] Exec "c:\python27\python.exe -u C:\Ashok\blinky\mbed-os-example-blinky\mbed-os\tools\project.py -i uvision -m TB_SENSE_12 --source ." in C:\Ashok\blinky\mbed-os-example-blinky
Scan: .

C:\Ashok\blinky\mbed-os-example-blinky>

And built it with MDK. I'm attaching build log below:
TB_SENSE_12_MDK_build_log.txt

Just checking if CI uses this version of ARMCC: Mine is this version:
*** Using Compiler 'V5.06 update 5 (build 528)', folder: 'C:\Keil_v5\ARM\ARMCC\Bin'

Edit:
Just checking the CI logs & it is using : *** Using Compiler 'V5.06 update 3 (build 300)', folder: 'C:\mbed_tools\Keil_v5\ARM\ARMCC\Bin' .
Does the compiler build version have anything to do with the .s and the .S failures?

@0xc0170
Copy link
Contributor

0xc0170 commented Nov 14, 2017

I would not suspect this is related to .s or .S files. __UVISION_VERSION should not be even present as uvisor is not supported for ARMCC.

I manually fetched this PR, and was able to build for uvision using exported project. After I installed the a pack for this device . 😕

@theotherjimmy
Copy link
Contributor

Does the compiler build version have anything to do with the .s and the .S failures?

I don't recall, so it might.

@theotherjimmy
Copy link
Contributor

@ashok-rao In general, it's more helpful to post code snippets in ``` blocks compared to screenshots. The advantage is that we can copy-paste from those blocks.

@theotherjimmy
Copy link
Contributor

@0xc0170 I don't know what you're on about. The failure is in uVision, which, as far as I know, always emits that macro. the example that is failing is blinky. Where are you getting uvisor? it's just not present anywhere in the logs.

@theotherjimmy
Copy link
Contributor

theotherjimmy commented Nov 14, 2017

After I installed the a pack for this device

What error did you get before installing the pack for the device?


Everyone:
Do not claim you cannot reproduce the problem, after trying to build using the GUI. It may not do exactly the same thing as the batch build. Please use the batch build before you claim that you cannot reproduce the problem.

@ashok-rao
Copy link
Contributor Author

@0xc0170 , same with me, I had to install the DFP for the EFR32MG12P332F1024xx125. Without the pack, MDK was selecting a generic M4_FP target & still builds succesfully. Not sure what you refer to regarding uvisor ?

@theotherjimmy , thanks for the note on screenshots, I just wanted to highlight the line numbers as well since targets.json has changed from my local copy to the latest master..so, just wanted to specify the line numbers where I made the change, hence the screenshot!
Attaching the batch file below that I generated from MDK (Under target options -> build -> generate batch file). It builds fine for me in the local system though & generates an AXF after successful build (I didn't convert that to bin)..
mbed-os-example-blinky.zip

@theotherjimmy
Copy link
Contributor

@ashok-rao Then you could use a git diff, it would include line numbers. For example:

diff --git a/targets/targets.json b/targets/targets.json
index 0a8549f00..44eb9050c 100755
--- a/targets/targets.json
+++ b/targets/targets.json
@@ -2912,6 +2912,7 @@
     },
     "TB_SENSE_12": {
         "inherits": ["EFR32MG12P332F1024GL125"],
+        "device_name": "EFR32MG12P332F1024GL125",
         "device_has": ["ANALOGIN", "I2C", "I2CSLAVE", "I2C_ASYNCH", "INTERRUPTIN", "LOWPOWERTIMER", "PORTIN", "PORTINOUT", "PORTOUT", "PWMOUT", "RTC", "SERIAL", "SERIAL_ASYNCH", "SLEEP", "SPI", "SPISLAVE", "SPI_ASYNCH", "STDIO_MESSAGES", "TRNG", "FLASH"],
         "forced_reset_timeout": 5,
         "config": {

mine is also different, at line 2912 (between the @@). Even better, with a diff I can copy it to git am and git will apply it to my working copy. That said, I don't think line numbers matter much here...


Under target options -> build -> generate batch file

I don't think that's how CI runs the builds. The CI builds issuing the command in the build method. That may produce a different result than the method you used.

That being said, I would be surprised if it did.

@ashok-rao
Copy link
Contributor Author

Thanks for the git diff notes! Will try to use where applicable.

Regarding batch files, I don't suspect either generation method would produce major differences though in the batch file generated. That being said, I don't have reach into the CI builds, so I cannot check that end. Someone with access could help us here!

@cmonr
Copy link
Contributor

cmonr commented Dec 1, 2017

@ashok-rao I was just able to reproduce the error locally, and am now working to root cause the issue.

@ashok-rao
Copy link
Contributor Author

Nice. Thank you @cmonr . Awaiting your inputs.

@cmonr
Copy link
Contributor

cmonr commented Dec 1, 2017

@ashok-rao It looks like the issue is on our side, and will have to wait until Monday for a fix to be tested.

No action on your end is needed at the moment.

@ashok-rao
Copy link
Contributor Author

@cmonr .. any news on this, please?

@cmonr
Copy link
Contributor

cmonr commented Dec 11, 2017

@ashok-rao Looks like the modification made it in this weekend. I'm going to rerun the job to verify.

/morph export-build

@mbed-ci
Copy link

mbed-ci commented Dec 11, 2017

@kegilbert
Copy link
Contributor

/morph export-build

@mbed-ci
Copy link

mbed-ci commented Dec 11, 2017

@ashok-rao
Copy link
Contributor Author

?? Same error again..!! @cmonr ..

@ashok-rao
Copy link
Contributor Author

@cmonr @kegilbert ..bump!

@mbed-ci
Copy link

mbed-ci commented Dec 15, 2017

@ashok-rao
Copy link
Contributor Author

Hmm, I don't see what failed here.. Looks like all tests have passed from the CI logs.. @0xc0170 @cmonr @kegilbert can you see what has failed?

@kegilbert
Copy link
Contributor

We've been seeing some Jenkins related failures intermittently, sorry for the delay.
/morph export-build

@cmonr
Copy link
Contributor

cmonr commented Dec 15, 2017

@ashok-rao I ran a limited manual build, and things finally look good on our side.

Barring any further strange Jenkins issues, this last build should do it!

@mbed-ci
Copy link

mbed-ci commented Dec 15, 2017

@ashok-rao
Copy link
Contributor Author

@cmonr @kegilbert .. crashed again!! And this time it's a different platform altogether.

@kegilbert
Copy link
Contributor

kegilbert commented Dec 15, 2017

This looks like an actual compilation error, running it by @geky since it looks like a filesystem error. That it only failed on that one platform makes me think it's another one-off error though...

@kegilbert
Copy link
Contributor

kegilbert commented Dec 15, 2017

My local compiler version is slightly off from the CI version but builds the export correctly. Trying again to see if it replicates and allocate the build node to debug further if it happens again.
/morph export-build

@kegilbert
Copy link
Contributor

Offline static node got assigned, this time for real.
/morph export-build

@mbed-ci
Copy link

mbed-ci commented Dec 15, 2017

Exporter Build : SUCCESS

Build number : 362
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/exporter/5470/

@kegilbert
Copy link
Contributor

/morph build

@mbed-ci
Copy link

mbed-ci commented Dec 17, 2017

Build : SUCCESS

Build number : 709
Build artifacts/logs : http://mbed-os.s3-website-eu-west-1.amazonaws.com/?prefix=builds/5470/

Triggering tests

/morph test

@mbed-ci
Copy link

mbed-ci commented Dec 17, 2017

Test : SUCCESS

Build number : 541
Test logs :http://mbed-os-logs.s3-website-us-west-1.amazonaws.com/?prefix=logs/5470/541

@ashok-rao
Copy link
Contributor Author

Awesome! Finally..! @0xc0170 @adbridge ..over to you to take this PR further. Thank you @kegilbert @cmonr ..

@kegilbert
Copy link
Contributor

No problem, sorry for the pain and delay @ashok-rao .

@0xc0170 0xc0170 merged commit 9ebefcd into ARMmbed:master Dec 20, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

8 participants