xiaoxiang781216 commented on issue #18849: URL: https://github.com/apache/nuttx/issues/18849#issuecomment-6513373199
> 9\. [@xiaoxiang781216](https://github.com/xiaoxiang781216) yes, time_t/clock_t stay int64_t; no POSIX/API change. > Before proposing code I'm measuring where the size actually goes. Early symbol diff on > nucleo-f072rb:nsh (M0) shows ~80% of the +1.5 KB from [#18840](https://github.com/apache/nuttx/pull/18840) is libgcc's generic 64-bit > divide/multiply (__divdi3, __udivmoddi4, __aeabi_ldivmod...), pulled in by only four > callers: clock_systime_timespec, uptime_read, clock_basetime and long long printf > (__ultoa_invert). Three of them divide by constants. > I'm checking whether avoiding those few divisions removes the libgcc code entirely, and > will post numbers before suggesting a direction. yes, it's right approach to avoid the kernel common link libgcc 64bit functions. -- 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]
