>Patches for GCC has been updated to work against Musl, GCC-9.
You could take it from Ataraxia, I did them after 9.1.0 release. But I
haven't checked gdc :D

On Wed, Sep 25, 2019, 12:24 AM Matías Fonzo <selk@dragora.org> wrote:

> I am pleased to announce the release of darkcrusade (2019Sep24).
>
> Darkcrusade is the codename given to deploy a set of independent
> cross compilers built upon on the standard C library Musl.  The full
> toolchain is composed by GNU Binutils, GNU Linux-libre, GCC, and Musl.
> Darkcrusade is ready to be used anywhere under general purposes.
>
> The homepage is at http://www.dragora.org
>
> The sources can be downloaded from
> http://rsync.dragora.org/current/compilers/darkcrusade_2019Sep24/
>  or from
> http://mirror.cedia.org.ec/dragora/current/compilers/darkcrusade_2019Sep24/
> (please use this ^ second link which is a mirror).
>
> This release contains the following changes:
>
> Version 2019Sep24 released.
>
>     * Binutils 2.32 + updates from 2_32-branch (version: 20190914).
>
>     * Linux-libre 4.19.75 (headers).
>
>     * GCC-9-20190921 (from snapshot, post gcc-9.2.0).
>
>       - GMP 6.1.2.
>
>       - MPFR 4.0.2 + cumulative patch (latest update: 2019-06-02).
>
>       - MPC 1.1.0.
>
>       - ISL 0.21.
>
>     * Musl updated to the snapshot 20190913_f5eee489 (post musl-1.1.23).
>
>     * Targets:
>
>         armv7-a:
>
>         - This soft float target has been renamed to 'armv7a_fp'.
>
>         - The CPU name for the target triplet has been changed to be
>           more generic (from armv7a to arm).
>
>         armv7a_hf:
>
>         - New target added.  This is almost the same target as
>           'armv7a_fp' with the difference that this target is hard
>           float, plus FPU 'vfpv3-d16'.
>
>         armv7hl:
>
>         - New target added.  This is a hard float target for the armv7-a
>           architecture (hard float) but with the FPU tuned for 'neon'.
>
>         i586:
>
>         - The option --enable-cld for the CLD instruction is now used by
>           default (just in case).
>
>         powerpc:
>
>         - Option for decimal support has been disabled here.
>
>         powerpc64:
>
>         - The option --with-abi=elfv2 is now passed by default (just in
>           case).
>
>         - The target name has been renamed to 'powerpc64le' for
>           clarification.
>
>         riscv64:
>
>         - New target added with multilib support.  This is a hard float
>           target for the architecture 'rv64gc' with the ABI 'lp64d'.
>
>         s390x:
>
>         - Directory to look into the corresponding (kernel) headers for
>           the architecture has been fixed. (s390x -> s390).
>
>     * Toolchain:
>
>       - Option for binutils --disable-separate-code is now given by
>         default. See: http://www.openwall.com/lists/musl/2019/08/12/4
>
>       - Patches for GCC has been updated to work against Musl, GCC-9.
>         Thanks to Szabolcs Nagy for provide it.  :-)
>
>       - config.sub and config.guess have been updated in order to
>         detect more target triplets, for example those containing
>         "armv7hl"...
>
>     * Darkcrusade is now distributed using the tarlz format.  For more
>       information, visit http://lzip.nongnu.org/tarlz.html
>
>     * Minor changes.
>
> The following (target) architectures are currently supported:
>
> - aarch64: The 64-bit ARM architecture.
> - armv7a_fp: ARM softfloat v7.
> - armv7a_hf: ARM hardfloat v7 (FPU vfpv3-d16).
> - armv7hl: ARM hardfloat v7 (FPU neon).
> - i586: Intel 586 optimized for Intel 686+ processors.
> - microblaze: Soft microprocessor core designed for Xilinx FPGAs
>   (32-bit).
> - mips: MIPS r4k big endian.
> - powerpc: PowerPC32.
> - powerpc: PowerPC64 (for Talos II systems).
> - riscv64: "risk-five" (64-bit long and pointers, hard-float).
> - s390x: z/Arch 64-bit.
> - x32: x86_64 ABI using 32-bit integers, longs and pointers.
> - x86_64: x86_64 (generic).
>
> Please try and report any errors you find.
>
>
> Best regards,
> Matías Fonzo, darkcrusade author and maintainer.
>