On 28 Jan 2015, at 18:16, Richard Henderson wrote: > On 01/28/2015 10:10 AM, Dominique d'Humières wrote: >>> I can't think of any reason they shouldn't work. Were they not running >>> before, >>> or did something else change? >> >> AFAIU the commit, the tests were not run on x86_64-*-*, so the tests and the >> corresponding failures are new. > > Well, the commit fixed the fact that they weren't running on x86_64 for the > -m32 multilib. If you had built a i686-darwin before the commit, the tests > should have already been running (and, presumably, failing).
I don't think (To Be Confirmed) this mode has ever been tested on Darwin in previous incarnations of libffi. (i.e darwin + windows call markers) ISTM that the first occassion was following your commit 12th Jan. >> If you think the failures are a bug, I’ll open a PR for them. > > I do think the failures are a bug. If Darwin wasn't supposed to support these > abis, they ought to have failed test compilation, not test execution. Well, somewhat confusingly "x86/darwin_c.c" is shared between darwin and the various Windows 32 bit implementations. It might be nice to rename it (but that can wait ...). A quick look suggests that FFI_THISCALL and friends will result in an abort or a return of "invalid ABI" on Darwin, but I don't (yet) see a compile-time error. We need to look into adding that - and disabling any inappropriate tests on Darwin. Iain