[llvm] r307852 - [Solaris] Detect Solaris LD, use detection results to pass Solaris-ld options

[Davide Italiano via llvm-commits]

On Tue, Sep 5, 2017 at 11:40 AM, Rainer Orth
<ro at cebitec.uni-bielefeld.de> wrote:
> Davide Italiano via llvm-commits <llvm-commits at lists.llvm.org> writes:
>
>> Picking a random Solaris commit.
>> I'd like to bring this up sooner rather than later.
>> It seems there's been a decent amount of interest/support to make
>> clang working on Solaris.
>> It's also not a mystery anymore that Solaris has been discontinued
>> and, as it stands, the LLVM/clang support for the OS is not completed.
>> I'm not entirely sure what are the plans, but in case nobody is
>> interested, I'd rather avoid the maintenance burden and remove the
>> support from the tree as it happened in similar circumstances (see
>> e.g. https://reviews.llvm.org/D35707).
>
> While this is true for Oracle Solaris, there are several OpenSolaris
> derivatives that are very much alive and just as interested as Oracle
> has been for a while.  They also do use Solaris ld instead of GNU ld, so
> the needs are almost exactly the same.
>
>         Rainer (also wearing the Solaris GCC maintainer hat ;-)
>

OK :)
I don't have strong opinions on the matter, I just wanted to collect
people thoughts.
>From what I see (or at least from what I remember), large part of the
work to support clang in Solaris has been done by the same corp that
decided to kill the OS.
I wasn't aware of Illumos trying to pick clang, but I guess I wasn't
following very closely :)
My only worry was that of having unmaintained code (and so prone to
bitrot) in tree, but it seems that there's enough interest to keep
this port alive.

Thank you for the replies, guys.

-- 
Davide

"There are no solved problems; there are only problems that are more
or less solved" -- Henri Poincare