linguini1 commented on PR #16977:
URL: https://github.com/apache/nuttx/pull/16977#issuecomment-3270900366

   > @acassis @linguini1 I don't understand why this feature is so expensive to 
explain. Are you still maintainers? Do you understand the implementation 
details of key features in the system? Can you please look at the code before 
giving a thumbs-up?
   
   Like Alan mentioned, I am not a maintainer, just a committer. I am also 
still relatively new (~1y) to NuttX, and many features are not documented. So I 
won't pretend to have a flawless understanding of the entire system, including 
syslog.
   
   I was mostly looking for a description of the use-case so I could understand 
the reason for this PR better, as the PR description is very minimal. It's not 
my intention to be hostile/slow progress/block your PR for silly reasons, so I 
am sorry if I've come across that way. Mainly I was wondering if there was a 
concrete example of why you decided to implement `nxprintk` now. Was `syslog` 
not working in some debugging scenario you had recently? That would help me 
understand. In my experience, I've been able to use syslog in interrupt 
contexts without issue. I am all for making syslog more robust (like you 
mentioned with LOG_EMERG) but I am little more wary of adding additional APIs 
to solve a problem with the very ubiquitous syslog API.
   
   I will not sit here and block this PR if other committers agree to merge it, 
that would be silly! Many of them know better than I do. However, I also will 
not approve this PR until there is some indication of testing the new API, and 
especially testing that works in the scenarios where you've described syslog 
won't work.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to