Modify target_write_u32() to make 32-bit data output easier

[kaeteyaruyo]

• Modify target_write_u32() function to make 32-bit data output easier
  • User can now output integer or any data type in 32-bit by assigning to (volatile <type> *)0x40000008

[jserv]

Please read this carefully: https://chris.beams.io/posts/git-commit/
The subject of git commit message and pull request should summarize the motivation why you would like to make the modification.

[kaeteyaruyo]

Sorry I put my purpose in the body but not title... It seems that I cannot modify a commit after I created it. Should I close this PR and create another one with an appropriate commit message?

[jserv]

Sorry I put my purpose in the body but not title... It seems that I cannot modify a commit after I created it. Should I close this PR and create another one with an appropriate commit message?

Check this: https://docs.github.com/en/free-pro-team@latest/github/committing-changes-to-your-project/changing-a-commit-message
Then, you can do force-push.

[jserv]

I suppose the change is improving the verbosity of data read operations, right? If so, please consider to wrap it with DEBUG_EXTRA.

[kaeteyaruyo]

I was intended to mimic line 822, which output a character when UART receives data.
Because UART outputs every byte received as character, even if I divide an integer by 4 and send them respectively, the processor still cannot display it correctly. I think we need a way to display data in any data type, and 4-byte integer is a very commonly used one. So at least I want the processor to display integer when it detects integer write. So I put a printf() in target_write_u32(), and it works.
But maybe this is not an appropriate way to deal with data output... (indeed this output is more like debug message rather than standard output) Should I use try other ways to output data > 8 bits?

[jserv]

You can manipulate the serialized data with a buffer and then flush later.