Given that the original symbol versioning patch has shipped
with gcc-6, it would be kind to keep that as a valid base ABI
and make subsequent changes to version 7.1.
At the same time, we must bump the (major) ABI version. This needed to be
done anyway due to ABI breakage in the AArch64 port (see 12cf89ee and the
corresponding GCC PR70024).
The unwind info isn't 100% correct at all points during the epilogue,
and not annotating is just as incorrect as the annotation. This works
better on systems that do not support DW_OP_call_frame_cfa.
Darwin aligns long-double to 16, and thus all of the long double
tests were failing due to not honoring that. We ought to be able
to devise a test case for GCC using __attribute__((aligned)) that
would have failed too.
The Apple assembler defaults to power of two alignment, rather than
byte alignment like everyone else. Force byte alignment by using
the proper directive.
Since libffi currently doesn't allow empty structures, libgo
currently maps them to ffi_type_void. Given that we'll abort
on this case, handle it gracefully.
Unlike ffi_closure_LINUX64, this entry point is called normally,
so we already have the TOC in R2 and the closure in R11.
* powerpc/linux64_closure.S (ffi_closure_LINUX64): Remove a
register dependency chain.
(ffi_go_closure_linux64): Don't load r11 or r2.
This is a port of
http://gcc.gnu.org/viewcvs?rev=207763&root=gcc&view=rev
aka GCC PR libffi/60073, to the rewritten Sparc codebase.
Supposedly, we should have seen failures with the existing
libffi.call/cls_double_va.c testcase, but I hadn't.
Perhaps a gcc newer than 4.6.3 is required to see that...
This eliminates the AM_CONDITIONAL ugliness, which eliminates
just a bit of extra boilerplate for a new target.
At the same time, properly categorize the EXTRA_DIST files
into SOURCES and HEADERS, for the generation of ctags.
The original code, removed in the "rewrite" patch, was incorrect for
large structures, and required dynamic allocation of a trampoline on
every ffi_call.
Instead, allocate a 4k entry table of all possible structure returns.
The table is 80k, but is read-only and dynamically paged, which ought
to be better than allocating the trampoline.
This is difficult to test with gcc. One can only use -O0 at present.
See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63668.
It's impossible to call between v8 and v9 ABIs, because of the stack bias
in the v9 ABI. So let's not pretend it's just not implemented yet. Split
the v9 code out to a separate file.
The register windows prevent ffi_call from setting up the entire stack
frame the assembly, but we needn't make an indirect call back to prep_args.
Assembly to use local labels, .type annotation, hidden annotation.
I do retain the _prefix for the symbols, but given that it wasn't
done consistently across all symbols, I doubt it's actually needed.
Do not modify the ffi_type. Rearrange the tests so that we
quickly eliminate structures that cannot match. Return an
encoded value of element count and base type.
Unties the backend from changes to FFI_TYPE_* constants, and allows
compilation to succeed after the addition of FFI_TYPE_COMPLEX.
Delete the hand-written unwind info.
Move everything into sysv.S, removing win32.S and freebsd.S.
Handle all abis with a single ffi_closure_inner function.
Move complexity of the raw THISCALL trampoline into assembly
instead of the trampoline itself.
Only push the context for the REGISTER abi; let the rest
receive it in a register.
Decouple the assembly from FFI_TYPE_*. Merge prep_args with ffi_call,
passing the frame and the stack to the assembly.
Note that this patch isn't really standalone, as this breaks closures.
The x86_64 unix port only handles one ABI; don't define all of the
other symbols. The UNIX64 symbol retains the same value.
The i386 ports ought to have the same symbols, even if we can't yet
unify the values without incrementing the libffi soname.
Dumps all of the hand-coded unwind info for gas generated. Move jump
table data into .rodata. Adjust ffi_call_unix64 to load the static
chain. Split out sse portions of ffi_closure_unix64 to
ffi_closure_unix64_sse rather than test cif->flags at runtime.
A "ffi_go_closure" is intended to be compatible with the
function descriptors used by Go, and ffi_call_go sets up
the static chain parameter for calling a Go function.
The entry points are disabled when a backend has not been
updated, much like we do for "normal" closures.
Richard Henderson