Hello Yuta thanks for your patch and interest.
> On 17 Sep 2021, at 16:26, Yuta Saito via Gcc-patches > <gcc-patches@gcc.gnu.org> wrote: > Currently, building gcc for aarch64-darwin host fails due to missing > host_hooks definition. > > This patch adds host_hooks definition for aarch64-darwin. > aarch64-darwin is not supported as a target yet, but this allows using > gcc cross-compiler on aarch64-darwin. 1. The development prototype for aarch64-darwin is here: https://github.com/iains/gcc-darwin-arm64 we need to phase the work into master with the approval of the Arm maintainers - so I would recommend that if you have fixes or improvements in the short-term, to make pull requests against the development branch. 2. The patch you have presented is identical in the host-only content to the existing development one: https://github.com/iains/gcc-darwin-arm64/commit/2190f7bda7bc0e6d0b74c7bd41c97510a685b06b So, that aspect has already been handled in the development. > I confirmed linking gcc-cross succeed on aarch64 darwin. Patches need more testing than that - specifically, that they do not regress other targets (unlikely, in this case, but it is good form to test at least that aarch64-linux-gnu is unaffected). ==== There are two other small changes I’d like to check before enabling host support 1/ https://github.com/iains/gcc-darwin-arm64/commit/98c8f79929db1bf29ac52f748137b08c21976483 might be needed for cross tools too. 2/ We need to ensure that PCH is defaulted to “off” and that there is a proper warning if the user tries to configure it “on” see: https://github.com/iains/gcc-darwin-arm64/issues/2 (which we do not plan to fix in the short term). ==== There is clearly interest in building cross-compilers on aarch64-darwin, so I will try to phase the host support sooner rather than later, thanks again for the patch, Iain
