[uClibc] clock() for long-running process

Atsushi Nemoto anemo at mba.ocn.ne.jp
Wed May 12 00:24:25 UTC 2004


>>>>> On Mon, 10 May 2004 11:07:00 -0600, mjn3 at codepoet.org (Manuel Novoa III) said:
mjn3> Thought about it some more and decided to make clock() behavior
mjn3> a build configuration option.  It will either measure absolute
mjn3> cpu time until reaching overflow and then return (clock_t)(-1)
mjn3> after, or it will emulate glibc behavior for applications
mjn3> wanting to measure elapsed times by taking differences.  I still
mjn3> think that latter is bad practice however, as programs doing
mjn3> this are relying on implemenatation defined behavior.

I'm glad to hear that.  I welcome the configuration option.  Thank you.

---
Atsushi Nemoto



More information about the uClibc mailing list