In the latest opam-repository CI Debian 13 on ARM64 jobs, the following tests fail due __Float32x4_t and __Float64x2_t:
The errors are of the form
# /usr/include/aarch64-linux-gnu/bits/math-vector.h[156:9-22] : syntax error
# Parsing errorFatal error: exception GoblintCil__Frontc.ParseError("Parse error")
# make: *** [Makefile:196: testrunc99/c99-tgmath] Error 2
where the problematic header has the following
typedef __Float32x4_t __f32x4_t;
typedef __Float64x2_t __f64x2_t;
These things are really getting out of hand. These are some kind of vectorized types, which we cannot even handle in much simpler forms (which occurred in some real projects we tried at some point), so we probably don't want to spend time on proper support (at least for Goblint). But losing modern Debian (and I guess soon Ubuntu) will be a problem.
In the latest opam-repository CI Debian 13 on ARM64 jobs, the following tests fail due
__Float32x4_tand__Float64x2_t:testrun/math1testrun/nan-globaltestrunc99/c99-complextestrunc99/c99-float-pragmatestrunc99/c99-tgmathtestrunc99/c99-tgmath2The errors are of the form
where the problematic header has the following
These things are really getting out of hand. These are some kind of vectorized types, which we cannot even handle in much simpler forms (which occurred in some real projects we tried at some point), so we probably don't want to spend time on proper support (at least for Goblint). But losing modern Debian (and I guess soon Ubuntu) will be a problem.