Re: [PATCH 0/3] Fix xmm_t to rte_xmm_t scalar conversion

["Stanisław Kardach" <kda@semihalf.com>]

On Tue, Jun 21, 2022 at 1:53 PM Thomas Monjalon <thomas@monjalon.net> wrote:
> <snip>
> If it is made intentionally hard, it is just a wrong design.
> A toolchain prefix is just a name.
> We can have 2 toolchains compiled with the same name and different behaviours.
> And we can have 2 similar toolchains with a different name.
I don't think meson will allow it anytime soon (see [1]). The
reasoning being that it's easy to screw up the environment and not
notice it where as files are persistent.
>
> > So should the direction be environment or rather separating
> > cross-files into arch-part and toolchain-parts and letting user create
> > his own toolchain part while maintaining a matrix of supported
> > combinations for CI? I'm not advocating either, just want to wrap my
> > head around it.
>
> We should be able to use a toolchain compiled anywhere
> without modifying the cross files, just because a "-gnu-" is missing
> or any other irrelevant detail.
I've checked that if I remove the binaries from a cross-file, then
specifying CC/CXX/AR/STRIP environment variables is picked up by
meson:
  AR=riscv64-unknown-linux-gnu-ar \
    STRIP=riscv64-unknown-linux-gnu-strip \
    CC=riscv64-unknown-linux-gnu-gcc \
    CXX=riscv64-unknown-linux-gnu-g++ \
    meson build-rv-test --cross-file=config/riscv/riscv64_linux_gcc
But then there are no default values.

A suggested frankenstein-like solution in [1] is to use a script that
generates a cross-file with [constants] section and launches meson
with it.

[1] https://github.com/mesonbuild/meson/issues/9#issuecomment-381410972