LLVM versions >= 12 cannot currently correctly compile gcc stage 1 because
GCC assumes references to certain symbols like "ix86_isa_flags" will be
optimized away.
Details: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=255760
Changelog:
* Bugfixes.
Target Specific Changes
AArch64
A bug with the Random Number intrinsics in the arm_acle.h header
that resulted in an incorrect status result being returned has
been fixed.
GCC now supports the Fujitsu A64FX. The associated -mcpu and
-mtune options are -mcpu=a64fx and -mtune=a64fx respectively.
In particular, -mcpu=a64fx generates code for Armv8.2-A with
SVE and tunes the code for the A64FX. This includes tuning the
SVE code, although by default the code is still length-agnostic
and so works for all SVE implementations. Adding -msve-vector-bits=512
makes the code specific to 512-bit SVE.
While here point out that the aarch64 equivalent patch was sent upstream.
Bump PKGREVISION. fix gcc*-libs PKGREVISION accordingly.
Fixes PR pkg/55992: math/blas fails on NetBSD/sparc64
Fixes report by Connor McLaughlan on pkgsrc-users
