patrickelectric commented on PR #18950:
URL: https://github.com/apache/nuttx/pull/18950#issuecomment-4564096729

   > No other architecture enables performance optimization by default, so why 
should this one? It's better to add an entry in the Documentation regarding 
performance optimization and document it there. Optimization for execution 
speed depends on the application and should not be forced on the user
   
   That's a good point, It's not up to me to device that, but to the project 
maintainers. If that's the overhaul philosophy, I'm ok with it and I'll open a 
new PR to document properly. I know that there is 
   
   **But**, I highly recommend the maintainers to calculate the trade-off of 
this PR. If performance is safe, stable, and delivers a better experience 
overhaul for the users / developers. Why not ? Just for the sake of not doing 
or strong opinions ? As I said is not up to me to decide. I just want to know 
the overhaul decision of the maintainers and follow it. Just giving my two 
cents here.
   
   Now, just to correct the point of:
   > No other architecture enables performance optimization by default
   
   That's not true. As you can see from this list:
   
   
[SMALL_DEFAULT](https://github.com/apache/nuttx/blob/master/Kconfig#L128-L130) 
is disabled by default.
   - Size is not a real concern so.
   
   
[LIBC_NEWLIB_OPTSPEED](https://github.com/apache/nuttx/blob/master/libs/libc/string/Kconfig#L51-L54)
 as consequence this is enabled.
   - The description of this configuration is pretty clear as: "newlib 
optimized function"
   
   
[LIBC_INLINE_QUEUE](https://github.com/apache/nuttx/blob/master/libs/libc/queue/Kconfig#L6-L8)
 also gets enabled.
   - The comment also points: "Inline queue list to improve the performance"
   
   
[NET_TCP_FAST_RETRANSMIT](https://github.com/apache/nuttx/blob/master/net/tcp/Kconfig#L145-L147)
 is enabled.
   - "Enable the Fast Retransmit algorithm"
   
   
[AT45DB_PREWAIT](https://github.com/apache/nuttx/blob/master/drivers/mtd/Kconfig#L616-L618)
 enabled by default.
   - "Enable higher performance write logic"
   
   There are also DMA that are enabled by default and others, but this comment 
is already big enough and I believe that my point is already clear. 


-- 
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