Conversation
6ac8c76 to
71153f1
Compare
06108ff to
541c23c
Compare
Signed-off-by: Jens Reidel <adrian@travitia.xyz>
541c23c to
45d78ba
Compare
|
This seems to work, but I'm not particularly happy with it yet. It limits us to musl targets that Alpine supports, which isn't particularly great. I'll see if building libgcc and libgcc_s from source is feasible in CI |
|
@Gelbpunkt out of curiosity, any reason not to promote these targets to T2 so we can test them proper? You're effectively acting as the target maintainer already :) (@neuschaefer seems involved in a lot of the same targets, they be interested in co-maintaining) |
For powerpc64-unknown-linux-musl, there is an open MCP. s390x is fun to play with, but I lack the knowledge about the architecture to properly maintain it and it is already maintained by an IBM employee, I'll leave it to them to decide whether to promote it or not. mips64 is working OK, but I have problems with my hardware for it that prevent me from regularly testing it. For RISC-V, I do not have access to hardware that can run the compiler testsuite in adequate time. tl;dr: it's complicated. I'd very much want libc to not regress on these targets, but without tier 2 it's tricky. |
Description
Please don't look at this yet, I'm messing with CI.
Sources
Checklist
libc-test/semverhave been updated*LASTor*MAXareincluded (see #3131)
cd libc-test && cargo test --target mytarget);especially relevant for platforms that may not be checked in CI