On Oct 3, 2013, at 1:49 PM, Tom Browder <[email protected]> wrote:

> No, I was wrong--it's working as advertised.  And the standard tests
> have different results with empty flags (the current case)  versus
> using the $CMAKE_C_FLAGS as in my test case.  See the differences in
> the two configure logs attached.

Not nearly as many differences as I would've expected, but the few changes do 
look right.  Very reassuring actually.  It means we can simplify our macros and 
add a few more tests now.

> I can't yet get my head around how CMake does such a generic
> compile check and I'm not sure I want to.

It's important to realize that there are different types of tests.  There are 
tests whether a symbol is declared in a header, whether it exists in a library 
as a symbol, whether it actually .. links, whether it actually .. runs, all 
different tests.  Some tests call for a certain type of specific check while 
others are more of an integration test answering "can I actually use this 
thing."

> I will try it on my system.  I notice they have a cmake plugin.

Their extensive available plugins and use by projects similar in complexity and 
size to BRL-CAD were aspects that faired better than Buildbot.

> <config-cmake-required-flags-OFF.log>
> <config-cmake-required-flags-ON.log>

They do indicate more tests are needed.  with -std=, knowing a symbol exists 
isn't sufficient.  We also have to check the header.

Cheers!
Sean


------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________
BRL-CAD Developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-devel

Reply via email to