GitLab

Also, print that message on stdout.

Pick longest flag when more than one matches in findArg.

This fixes an issue where -ignore-dot-ghci wasn't honored, because the
flag was parsed as "-i gnore-dot-ghci".

mad.one@gmail.com authored Aug 05, 2012 and pcapriotti committed Aug 13, 2012

In particular, many of the restrictions on using unboxed tuples went
away thanks to Max Bolingbroke, and they can now be bound to variables
without pattern matching, and passed as arguments.

Also fix a little formatting in glasgow-exts.xml. Some other parts
of this section may be out of date.
Signed-off-by: Austin Seipp <mad.one@gmail.com>

mad.one@gmail.com authored Aug 05, 2012 and pcapriotti committed Aug 13, 2012

Supresses an ugly warning from gcc 4.6+ saying this is a C/ObjC flag
only.
Signed-off-by: Austin Seipp <mad.one@gmail.com>

Simon Marlow authored Aug 09, 2012 and chak@cse.unsw.edu.au. committed Aug 09, 2012

When refactoring this recently I accidentally put the
dead-strip-preventer symbol (only used on OS X) in the wrong section.

Now that it's included in settings, the bindist configure script needs
to know it.

The problem occurred when the idle GC was turned off with +RTS -I0.
Then the scheduler would go into the state ACTIVITY_DONE_GC directly
without doing a GC, and a subsequent GC would put it back to
ACTIVITY_YES but without turning the timer back on.  Instead if the GC
finds the state is ACTIVITY_DONE_GC it should leave it there.

See comments.

We weren't passing the arguments correctly to the GC functions, which
usually happened to work because the arguments were in the right
registers already.

After this fix the profiling tests go through with the new code
generator.

When calling newCAF, refer to the closure using its LocalReg rather
than R1.  Using R1 here was preventing the register allocator from
coalescing the assignment x=R1 at the beginning of the function.

aargh.

This saves compile time and can make a big difference in some
pathological cases (T4801)

This means that we now generate the same code whatever platform we are
on, which should help avoid changes on one platform breaking the build
on another.

It's also another step towards full cross-compilation.

To explicitly choose whether you want an unregisterised build you now
need to use the "--enable-unregisterised"/"--disable-unregisterised"
configure flags.

This is a bit odd by itself, but it's a stepping stone on the way to
putting "target unregisterised" into the settings file.

No functional differences yet

We weren't defining it in the other places that MachRegs.h gets
imported, which seems a little suspicious. And if it's not defined
then it defaults to 4 anyway, so this define doesn't seem necessary.

The new code generator now apparently generates a working stage2
compiler.

See Note [foreign calls clobber GlobalRegs]