Re: Link tests after GCC_NO_EXECUTABLES

[Richard Sandiford <rsandifo@nildram.co.uk>]

Rask Ingemann Lambertsen <rask@sygehus.dk> writes:
> On Fri, Dec 07, 2007 at 11:27:31AM +0000, Richard Sandiford wrote:
>> OK, I'm not doing a good job of staying out of this discussion as
>> promised, but I was hoping someone else would raise this point...
>> 
>> My main concern with applying this patch in its current state is that,
>> while we have (intentionally) only been talking about *-elf so far,
>> newlib != *-elf.  Cygwin is also an important newlib user, and I don't
>> remember anyone explicitly mentioning it so far in this thread.
>
>    What is a good way of detecting the bare-metal targets? *-elf*, *-eabi*?

TBH, I'm not sure.  I wonder if a good starting point might be to take
the toplevel configure.ac stanzas that currently (i.e. after your earlier
patch) set libgloss_dir, and see which of those have an appropriate
"bare-metal-like" configuration in newlib/libgloss.  That might be
easier than a full two-way comparison between toplevel configure
and newlib/libgloss, and it can always be added to later.

(On that note, it might be a good idea to introduce a specific
mips*-*-elf* stanza in the toplevel makefile, and only do the
ibgloss_dir stuff, or its replacement, for that.  The current
mips*-*-* stanza traps almost-certainly-bitrotten targets like
mips*-openbsd* as well.)

I see downthread that you and Mark talked about a new configure option.
Would it be worth defaulting to on for the targets we know are "safe"?

(For the record, I do like the cache idea, and thanks for implementing it.
I hope yesterday's message didn't make it sound otherwise.  I just wanted
to make sure that what went into mainline wouldn't cast the net too wide.
Once we identify targets that work with the cache, I'd personally like
them to use it by default.  And that includes the mips*-elf* family.)

Richard