GitLab

Profiling shows we're wasting a lot of time calling mkErrInfo and
tidying the context, work that is wasted if we're not tracing.

There is more to this compilation performance regression than this,
but this is one bug.

Also give a civilised error message when the data type to vectorised contains no data.

We were never really manipulating the EkCtxt after we had built it,
so it's simpler to just pass the final SDoc.

The previous names were not informative at all, and now we have
named kinds like Constraint and datatype promotion to kind, so
we might as well name these too.

I tried to update some comments to the new names, but certainly
many references to the old names remain.

LLVM doesn't support the OS X system assembler anymore so we must use
their assembler through clang. Also improved error messages when various
LLVM tools can't be run.

We avoid calling "rm -rf" with no file arguments; this fixes cleaning
on Solaris, where that fails.

We also check for suspicious arguments: anything containing "..",
starting "/", or containing a "*" (you need to call $(wildcard ...)
yourself now if you really want globbing). This should make things
a little safer.

The tabs have already disappeared

Now that we don't necessarily put them in the same place as the source,
the directory may not exist.

-ddump-* files are put in the given directory.

-outputdir also sets the dumpdir.

This was pretty straightforward: collect the filenames in the lexer,
and add them in to the tcg_dependent_files list that the typechecker
collects.

Note that we still don't get #included files in the ghc -M output.
Since we don't normally lex the whole file in ghc -M, this same
mechanism can't be used directly.

Currently it is appearing on stderr, whereas the rest of the messages
from --make go to stdout, this change fixes that.