Re: [patch #7911] AX_PTHREAD handles AIX incorrectly

[Daniel Richard G.]

On Mon, 2013 Jan  7 9:45+0100, Thomas Jahns wrote:
>
> > * Keep the not-GCC check
>
> I wasn't sure how that interacted with the gxlc wrapper and
> left it out.

In my view, if the compiler claims to be GCC, we need to take it at its
word; it's too much work to do otherwise. The onus has to be on the tool
to really behave like GCC.

(The Intel C++ compiler also claims to be GCC, and this caused me grief
until I learned of its -no-gcc switch.)

> > * Furthermore, do this only on AIX
>
> Does anyone have access to a machine running Linux on POWER or MacOS X
> on PowerPC with IBM compilers who could comment on whether this is an
> AIX-only issue?

We can always elaborate the case-statement pattern if needed. I'm
concerned that the "use ${CC}_r if it's available" heuristic will go
wrong on some completely unrelated system.

> > * Quoting tweaks
>
> Assignments like
>
> a=$b
>
> don't need quotes even if $b contains white-space, IIRC.

This was more for consistency's sake. (And between with-quotes and
without, I figured, better the way that doesn't require remembering that
special case.)

> > * Sorted the big case pattern and fixed a couple typos
> > * Use AS_EXECUTABLE_P() instead of "test -x"
> 
> Thanks, I wasn't aware of that.

The macro has been around for a long time, but was only documented very
recently. It's new to me too :)


--Daniel


-- 
Daniel Richard G. || address@hidden
My ASCII-art .sig got a bad case of Times New Roman.