So we did *not* have the stgCallocBytes prototype, and subsequently
the C compiler defaulted to `int` as a return value. Thus generating
sxtw instructions for the return value of stgCalloBytes to produce
the expected void *.

Ben Gamari authored 5 years ago and Moritz Angermann committed 4 years ago

Due to #17607.

markLiveObject is called by GC worker threads and therefore must be
thread-safe. This was a rather egregious oversight which the testsuite
missed.

We don't pass MAKE_ARGS for windows builds, so this should unbreak
them.

Ben Gamari authored 4 years ago and Moritz Angermann committed 4 years ago

We now have a proper periodic clean-up script installed on the runners.

See ghc/ghc#17018

This drops allocateExec for darwin, and replaces it with
a alloc, write, mark executable strategy instead. This prevents
us from trying to allocate an executable range and then write to
it, which X^W will prohibit on darwin.

This will *only* work if we can use mmap.

_AC_PROG_CC_99 isn't available anymore in recent autoconf versions.
This broke on aarch64-darwin.

This breaks lots of tests unnecessarily. The LLVM IR
has been pretty stable since ~7.  This stops GHC from
emitting the warnings during the testsuite run.

This may be necessary when building stage2 cross compiled
compilers.

Ben Gamari authored 4 years ago and Moritz Angermann committed 4 years ago

To support proper parsing of arm64 targets, we needed to adjust
the GHC_LLVM_TARGET function to allow parsing arm64-apple-darwin
into aarch64. This however discared the proper os detection.

To rectify this, we'll pull the os detection into separate block.

Fixes #19173.

This is a pre-requisite for making aarch64-darwin work.

This is needed so that the codegen can produce C ABI calls
that require knowledge about the actual size of arguments.

Specifically aarch64-darwin will require arguments (in exess
of available registers) to be passed *packed* on the stack.

this prevents the testlib/driver to be overly noisy, and will also
kill some noise produiced by the aarch64-darwin cc (for now).

Fixing sysctl, will allow us to run the test's properly in a nix-shell
on aarch64-darwin

this is effectively now required.

arm64 is really only a name apple uses, and we should
refrain from using it across multiple architectures.
Let's call aarch64-darwin, arm64 if needed, but otherwise
stick to aarch64.

     int
     memcmp(const void *s1, const void *s2, size_t n);

     size_t
     strlen(const char *s);

This is required for aarch64-darwin.
Bump the integer-gmp submodule as well.

This will fail for now. But allows us to add aarch64-darwin
machines to CI.

Previously we only built Linux bindists with hyperlinked sources.

Ensure that deb10-dwarf artifacts are preserved.

(cherry picked from commit a4f50cd8)

The motivation is given in Note [tcFamTyPats: zonking the result kind].

Fixes #19250  --  the fix is easy.

(cherry picked from commit 69cab37a)

Tamar Christina authored 4 years ago and Ben Gamari committed 4 years ago

(cherry picked from commit 4517a382)

Tamar Christina authored 4 years ago and Ben Gamari committed 4 years ago

(cherry picked from commit f065b6b0)

We do not support foreign "C" imports of varargs functions. While this
works on amd64, in general the platform's calling convention may need
more type information that our Cmm representation can currently provide.
For instance, this is the case with Darwin's AArch64 calling convention.
Document this fact in the users guide and fix T5423 which makes use of a
disallowed foreign import.

Closes #18854.

(cherry picked from commit 0b772221)