Adi cog candidate #5097

dave-wu · 2017-09-14T01:37:36Z

Notes:

Pull requests will not be accepted until the submitter has agreed to the contributer agreement.
This is just a template, so feel free to use/remove the unnecessary things

Description

Add support for the ADI COG boards: EV-COG-AD3029LZ and EV-COG-AD4050LZ

Status

READY

Migrations

If this PR changes any APIs or behaviors, give a short description of what API users should do when this PR is merged.

NO

Related PRs

None

Todos

None

Deploy notes

None

Steps to test or reproduce

N/A

Committer: Edmund Hsu Changes to be committed: modified: objects.h new file: trng_api.c

Committer: Hsu <[email protected]> Changes to be committed: modified: targets.json

Changes to be committed: modified: .mbedignore

…ectly; - Increased the TRNG counter value to generate better random values; - Moved the TRNG enable and counter value setting calls to the init routine as they only need to be done once.

- Fixed some issues with sleep module; - Modified gpio_api.c to use global gpio driver memory which is now declared in a separate file.

…rge bin files.

…on of large bin files." This reverts commit 24f422e55bd8f1134e84c478a772ba8fd90a3f10.

2 renamed LED pin names for EVAL-ADICUP3029 3 enabled greentea test automation for EVAL-ADICUP3029

…the following test suite: mbed-os-rtos-rtx-target_cortex_m-tests-memory-heap_and_stack

mbed-os-tests-events-timing

…them.

…itions

…ntiate hardware specific configurations

- Changed ADC function to use non-Arduino pin names.

….5.5.

Conflicts: targets/targets.json

…iB set in RTX.

… to the github main trunk.

theotherjimmy · 2017-09-14T01:54:23Z

Hi @dave-wu. Thanks for the contribution! This is HUGE! I have several recommendations to get this in faster:

Split it into multiple, logically distinct/orthogonal, pull requests(PRs).
i) One PR for each platform could be much smaller
ii) Doing baseline support and then adding features in subsequent PRs is generally much easier to review and test
Avoid merge commits, rebase instead
Try to include helpful git commit messages. This is a handy guide for improving commit messages.

Now some information on what we can accept. In order to accept any PR into a patch release, we require that the PR does not contain merge commits. By the looks of it, your git history is somewhere between 10 and 20 percent merge commits. It should be possible to rebase these out. Further, we can only accept PRs based on master. It looks to me like you merged with a release branch at some point.

Phew. I recommend breaking this PR into multiple PRs that introduce one thing at a time. This will both make the rebasing easier by reducing the number of commits in each rebase smaller, and make the PRs much easier to review. If you need help with this, just let us know.

theotherjimmy · 2017-09-14T01:55:32Z

I'm curious, what's the "candidate" in the title supposed to indicate?

dave-wu · 2017-09-15T07:32:01Z

Hi Jimmy,

Thanks for the feedback. Based on the requirements for merging, would the following method be acceptable?

I will fork the latest master branch into my own repository;
Add the files/folders for one of the platforms (i.e. EV-COG-AD3029LZ) and commit to the master branch in the fork;
Create a pull request for this change;
Add the files/folders for the other platform to the master branch in the fork;
Create a second pull request for the code in 4.

I think this may be a lot easier than going back and rebasing all the merges/branches etc. and is less error prone. Do you think this approach is workable?

BTW, the name "candidate" was there just to let myself know this is to be used for the pull request, all the files are working and are in release form.

Thanks

theotherjimmy · 2017-09-15T14:24:59Z

Based on the requirements for merging, would the following method be acceptable?

Yes. The only problem you might encounter is losing your git history. If that does not bother you, then I see no downsides.

Do you think this approach is workable?

Yes.

BTW, the name "candidate" was there just to let myself know this is to be used for the pull request, all the files are working and are in release form.

Thanks for the clarification. For future reference, we take all pull requests as if they were "candidates" by your definition.

dave-wu · 2017-09-17T23:58:19Z

Hi Jimmy,

Is there a way to remove my existing dave-wu/mbed-os fork so I can create a new one from the main mbed-os repo? Or can I just fork on top of it and that will overwrite the existing one? Thanks.

theotherjimmy · 2017-09-18T01:13:35Z

@dave-wu There is no way to do what you are asking for all at once. Since github just uses git underneath, you can just push (or force push, that is `git push -f) over any branch on your fork. If you provide me with some detail about what you are trying to do, I can offer some suggestions on ways to accomplish it. I'm asking for more high level goals, such as "break a single branch into multiple branches, each containing some of the commits" or "Rewrite git history to look like no merges took place".

Explicitly:

Is there a way to remove my existing dave-wu/mbed-os fork so I can create a new one from the main mbed-os repo?

You can technically do this, but I do not recommend it. It is generally more instructive to try to recover the situation you are it. You would have to go to the settings panel and delete the repo.

Or can I just fork on top of it and that will overwrite the existing one?

You can't do this at all. You would have to either force push a branch to match the original repo or delete your fork (as above) and re-fork it. I recommend using git to force push a branch back to it's original state, in preference to the other option I outlined, as it's less destructive in that it only affects one branch.

These questions indicate to me that you want to rewrite your git history to change the commits in this Pull Request. How about you just try doing that directly with git rebase -i?

Some resources on rebasing can be found here.

dave-wu · 2017-09-18T05:06:17Z

Hi Jimmy,

Since the current fork dave-wu/mbed-os.git has all the undesirable merges and branches etc. that you mentioned so I am thinking of:

Getting a fresh fork off the main mbed-os repo;
Add the code for my platforms to the main branch so I can do a compliant pull request
(I am aware that I'll lose the change history on my code but I am OK with that)

So based on what you said, my understanding is that the best way to get the existing forked repo (i.e. dave-wu/mbed-os.git) to match exactly the main mbed-os trunk (without any of my previous merges and branches) is to clone a fresh copy of the main mbed-os repo and then do a 'push -f ' it to the existing fork to overwrite it, as opposed to deleting the fork and re-forking, is this correct?

Thanks.

theotherjimmy · 2017-09-18T14:03:30Z

Getting a fresh fork off the main mbed-os repo;

This is an unnecessary step, and not productive

Add the code for my platforms to the main branch so I can do a compliant pull request

I'm just confused by this sentence.

What's a complain PR?
What would you be complaining about?
What "main branch" are you talking about?

my understanding is that the best way to get the existing forked repo (i.e. dave-wu/mbed-os.git) to match exactly the main mbed-os trunk (without any of my previous merges and branches) is to clone a fresh copy of the main mbed-os repo and then do a 'push -f ' it to the existing fork to overwrite it, as opposed to deleting the fork and re-forking, is this correct?

Yes. Although, there is no such branch called "trunk". Generally SVN and CVS calls things "trunk" and git calls them "master".

dave-wu · 2017-09-19T07:41:25Z

@theotherjimmy: I have just made a new pull request PR #5137 for just the EV-COG-AD3029LZ board, this should be clean enough to be merged. I haven't yet made a pull request for the EV-COG-AD4050LZ. Should I hold off the pull request for the EV-COG-AD4050LZ until you have merged the EV-COG-AD3029LZ code? Thanks.

theotherjimmy · 2017-09-19T14:20:16Z

Should I hold off the pull request for the EV-COG-AD4050LZ until you have merged the EV-COG-AD3029LZ code?

That's up to you. It may be easier for you to deal with one PR at once. It may get in faster to do both at the same time.

dave-wu · 2017-09-25T00:38:00Z

Closing this PR as it is replaced by 2 separate PRs: #5137 and #5144.

U-ANALOG\dwu and others added 30 commits March 23, 2017 16:41

Initial check-in for ADuCM3029 target

f5e1152

added ADuCM3029 target information

ffe6e59

add trng_api.c for ADuCM3029

fd1752c

Committer: Edmund Hsu Changes to be committed: modified: objects.h new file: trng_api.c

add "TRNG" to ADuCM3029's "device_has"

463c75f

Committer: Hsu <[email protected]> Changes to be committed: modified: targets.json

Convert '\' to '/' since '\' cannot be compiled in linux

4d7c475

Changes to be committed: modified: .mbedignore

- Changed pin_function routine in pinmap to write the mux values corr…

a2b07e5

…ectly; - Increased the TRNG counter value to generate better random values; - Moved the TRNG enable and counter value setting calls to the init routine as they only need to be done once.

Added sleep functions.

3374298

- Added GPIO interrupt support;

9aeac30

- Fixed some issues with sleep module; - Modified gpio_api.c to use global gpio driver memory which is now declared in a separate file.

Removed declaration of IVT table in RAM to avoid the generation of la…

99f68c7

…rge bin files.

Revert "Removed declaration of IVT table in RAM to avoid the generati…

d56228a

…on of large bin files." This reverts commit 24f422e55bd8f1134e84c478a772ba8fd90a3f10.

Use NOLOAD in LD to prevent the generation of large bin files.

5990e64

1 removed misplaced .mbedignore

5f50d76

2 renamed LED pin names for EVAL-ADICUP3029 3 enabled greentea test automation for EVAL-ADICUP3029

fixed incorrect INITIAL_SP address in mbed_rtx.h which cause fail in …

2e5cbc7

…the following test suite: mbed-os-rtos-rtx-target_cortex_m-tests-memory-heap_and_stack

enable EVAL-ADICUP3029

893934e

Fixed misplaced TIMER_CLK_FREQ_MHZ and TIMER_PRESCALER in us_ticker.c

df4e158

Real Time Clock API for ADuCM3029

608ebe2

Added macro RTC

3cc600a

move files to right folders for future BSP upgrade

09dd5fc

Upgrade BSP to released ADuCM302x_DFP.1.0.4

3e7b437

Added low power ticker.

df3da5d

changes for SPI on the aducm3029

37cf887

fixed timestamp backward drift issues in test suite:

06e95be

mbed-os-tests-events-timing

Set the clock source to 32 kHz internal before calling adi_rtc_Open()

b5aaf09

added aducm4050 target

331c937

Added files for ADUCM4050.

bc63da7

Fixed incorrect driver state buffer sizing.

d108869

Now clear the appropriate GPIO mux configuration bits before setting …

d8e5c22

…them.

added UART_1 to pinmap

dd11542

update serial_api.c to support UART_0 and UART_1

9d2736c

add trng for ADuCM4050

7d00bb1

mtrommer and others added 14 commits August 15, 2017 12:18

Updated arduino pin map macros

47f562e

updated arduino pin macros for ADICUP3029

c826383

Fixed LED2, PB0 and PB1 pin mappings, Added system wake INT pin defin…

a860f79

…itions

Updated ADI platforms - added hardware name to macros list to differe…

9dbe818

…ntiate hardware specific configurations

- Updated the pin names for all platforms;

4d3385a

- Changed ADC function to use non-Arduino pin names.

Merge branch 'master' of https://github.com/ARMmbed/mbed-os version 5…

1445d59

….5.5.

Fixed security options for all supported toolchains

a127ab0

Merge https://github.com/ARMmbed/mbed-os mbed-os v5.5.6

44a7977

Conflicts: targets/targets.json

Add iar export support for ADuCM3029 and ADuCM4050 platforms

0c551f3

removed averaging from the init() function

127b028

Merge remote-tracking branch 'origin/master'

607b34c

removed averqaging from the init() function

326ad74

Added default thread stack size definition, set to 1KiB instead of 4K…

be4d72e

…iB set in RTX.

Created a new branch and cleaned up unneeded files for a pull request…

324fcde

… to the github main trunk.

theotherjimmy added the needs: work label Sep 14, 2017

dave-wu mentioned this pull request Sep 19, 2017

Add support for ADI EV-COG-AD3029LZ platform #5137

Merged

dave-wu mentioned this pull request Sep 20, 2017

Add support for ADI EV-COG-AD4050LZ platform. #5144

Merged

dave-wu closed this Sep 25, 2017

sg- removed the needs: work label Sep 25, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Adi cog candidate #5097

Adi cog candidate #5097

Uh oh!

dave-wu commented Sep 14, 2017

Uh oh!

theotherjimmy commented Sep 14, 2017

Uh oh!

theotherjimmy commented Sep 14, 2017

Uh oh!

dave-wu commented Sep 15, 2017

Uh oh!

theotherjimmy commented Sep 15, 2017

Uh oh!

dave-wu commented Sep 17, 2017

Uh oh!

theotherjimmy commented Sep 18, 2017 •

edited

Loading

Uh oh!

dave-wu commented Sep 18, 2017

Uh oh!

theotherjimmy commented Sep 18, 2017

Uh oh!

dave-wu commented Sep 19, 2017

Uh oh!

theotherjimmy commented Sep 19, 2017

Uh oh!

dave-wu commented Sep 25, 2017

Uh oh!

Uh oh!

Adi cog candidate #5097

Adi cog candidate #5097

Uh oh!

Conversation

dave-wu commented Sep 14, 2017

Description

Status

Migrations

Related PRs

Todos

Deploy notes

Steps to test or reproduce

Uh oh!

theotherjimmy commented Sep 14, 2017

Uh oh!

theotherjimmy commented Sep 14, 2017

Uh oh!

dave-wu commented Sep 15, 2017

Uh oh!

theotherjimmy commented Sep 15, 2017

Uh oh!

dave-wu commented Sep 17, 2017

Uh oh!

theotherjimmy commented Sep 18, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dave-wu commented Sep 18, 2017

Uh oh!

theotherjimmy commented Sep 18, 2017

Uh oh!

dave-wu commented Sep 19, 2017

Uh oh!

theotherjimmy commented Sep 19, 2017

Uh oh!

dave-wu commented Sep 25, 2017

Uh oh!

Uh oh!

theotherjimmy commented Sep 18, 2017 •

edited

Loading