For details on the patches applied, see:
https://sources.debian.org/patches/ltrace/0.7.3-6/
Disabling '-Werror' may be a problem in the future again,
but for now keep things simple now that they're fixed.
Apparently, without this patch `NT_PRSTATUS` is not found. So the patch
adds the include apparently necessary. `NT_PRSTATUS` is also defined in
`<linux/ptrace.h>`, which would likely have been a better name, were it
not in the `linux/` directory, which is a priori not stable.
The need to do that is kind of weird (the change was introduced in [1],
and fedora apparently didn't need this additional import), but I'll try
to upstream it.
[1] https://github.com/SimonKagstrom/kcov/pull/239
Following legacy packing conventions, `isArm` was defined just for
32-bit ARM instruction set. This is confusing to non packagers though,
because Aarch64 is an ARM instruction set.
The official ARM overview for ARMv8[1] is surprisingly not confusing,
given the overall state of affairs for ARM naming conventions, and
offers us a solution. It divides the nomenclature into three levels:
```
ISA: ARMv8 {-A, -R, -M}
/ \
Mode: Aarch32 Aarch64
| / \
Encoding: A64 A32 T32
```
At the top is the overall v8 instruction set archicture. Second are the
two modes, defined by bitwidth but differing in other semantics too, and
buttom are the encodings, (hopefully?) isomorphic if they encode the
same mode.
The 32 bit encodings are mostly backwards compatible with previous
non-Thumb and Thumb encodings, and if so we can pun the mode names to
instead mean "sets of compatable or isomorphic encodings", and then
voilà we have nice names for 32-bit and 64-bit arm instruction sets
which do not use the word ARM so as to not confused either laymen or
experienced ARM packages.
[1]: https://developer.arm.com/products/architecture/a-profile
On GNU/Linux the build references these files, so let's fetch them from
the Chromium repository. I haven't checked whether they are heavily
patched or whether we can use the version from LLVM, but when looking at
the changes, they do seem to divert a bit from upstream LLVM.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer, @stesie
The tarball from upstream seems to be generated on the fly, so the
output is not deterministic and using fetchzip makes this more reliable
as we have a recursively hashed output path without any of the
non-determinisms in tarballs.
Unfortunately, the build still fails on NixOS systems, because we need a
few more stuff in the build tree.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer, @stesie
