Re: [PATCH, v3] Potential solution to librt issue.

[Paolo Carlini <paolo dot carlini at oracle dot com>]

Hi,

first, as you may have noticed, I have just reverted our first attempt at solving this issue.

> So it works on my arch (x86_64) w/o librt and pthread... I
> must be missing something.

I think the only unspected behavior here is that nothing crashes. Because you are not linking the library containing clock_gettime and thus __clock_gettime should still be zero at run time. Anyway, whatever value it has, I don't think the actual clock_gettime is being called: just check in the debugger, no?

Anyway, what we really want is playing a bit with Paolo's draft, nothing is going to wrong, otherwise, because nothing in the compiler + library takes care of linking librt on demand. If the compiler bits are ok, then the library bits, which you have essentially ready, are very simple.

Paolo.