davids5 commented on issue #2203:
URL: 
https://github.com/apache/incubator-nuttx/issues/2203#issuecomment-721691926


   > @davids5 you are using a workaround in PX4, was this reported here? 
@xiaoxiang781216 what can we do to fix this properly on NuttX?
   
   @v01d Yes - years before ASF. But for a different reason. The one you found 
printf had a side effect. It would write to a file if  0,1,2 are not stdxxxx. 
Then 0,1,2 becomes the first, second and third file opened in a group. printf 
corrupts the file. @patacongo rejected my fix at the time. He expected that we 
should ifdef all the printfs. I disagreed and implemented what I posted.  
   
   When https://github.com/apache/incubator-nuttx/pull/1755 came in it broke 
systems with out a /dev/console. as shown 
https://github.com/apache/incubator-nuttx/issues/2203#issuecomment-721146799 
   
   All the nuttx bootloders in PX4 do not use a console and where hardfaulting 
in `sercon` 's printf.  
   
   Since my concept was rejected, I saw no point in rehashing the issue.  
   
   I agree @xiaoxiang781216 soult fix it properly and we need to test better 
(we being me too, i missed the bootloader test case when I approved the #1755) 
   
   I agree that CONFIG_CONSOLE_* is confusing and we should have a doc that 
explains all the console options and what is mutually exclusive and why. There 
should also be code to enforce it.  I think this was cased by organic growth. 
Unfortunately I have lost context and would have to reverse engineer it and 
still might misconstrue the intent.
   


----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Reply via email to