Per datasheet to adjust delay for tw8816

[howard0su]

No description provided.

[howard0su]

Validate on my own Tx

[PhracturedBlue]

well...it is really difficult to review this patch with the style fixes applied.
Also, I don't like some of the formatting changes (like moving curly-braces to the same line...This is one of those things that I think is context specific. I think I want to separate cleanup patches from code-change patches.

[howard0su]

no intention to check this in. This is another testing. I think clang-format overall is useful when we need dealing the whole code base. What's your plan then? Do you want to have one big change to clean all of them? If you like to do so, let's coordinate with the devs who has the pending PR on hand. or your lint gives the warnings and expect we clean it up in a relative long run.

…

On Sun, Jan 20, 2019 at 10:01 PM PhracturedBlue ***@***.***> wrote: well...it is really difficult to review this patch with the style fixes applied. Also, I don't like some of the formatting changes (like moving curly-braces to the same line...This is one of those things that I think is context specific. I think I want to separate cleanup patches from code-change patches. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#619 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAuzRHxgjhJVaPYGNI4WQ5B6tx9gsbCsks5vFHbHgaJpZM4aJcHe> .

-- -Howard

[PhracturedBlue]

what is the plan for this?
My concern is that there are many variants of the tw8816 (I think we've seen 6 different revisions in use by Walkera). They all had different quirks, so I'm very reluctant to make a change that the data-sheet says is ok without having confidence it is ok on multiple actual transmitters. On the other hand it is almost 0.5sec of wait that you can remove which is quite a lot. I need to check my F12E and see what revision of the TW8816 it has.

[howard0su]

i think I should revert the change on msleep.

this change was intent to remove Delay calls. I am thinking to replace Delay with usleep(1) which is almost same as the amount of time Delay actual cause.