- 15 Dec, 2000 11 commits
-
-
sewardj authored
temp hack to make contents disappear
-
simonmar authored
Hmm. Merge in stuff from before-ghci-branch that CVS refused to merge before.
-
sewardj authored
more BCO# wibbles
-
sewardj authored
Conversion of unlinked BCOs to linked BCOs.
-
simonmar authored
ASSERT that we have the right number of arguments when building a constructor, to avoid repeats of today's extended debugging session.
-
simonmar authored
Add a special case in doDiscard for the sequence '"'. The bootstrapped compiler can now build ghc/lib/std.
-
simonmar authored
Don't eta reduce something we just saturated. (D'oh!) This fixes the problem which causes a bootstrapped HEAD compiler to crash.
-
sewardj authored
add BCOzh
-
sewardj authored
nuke getBCOPtrszh
-
sewardj authored
fix arg tys for newBCO#, and delete all other BCO related primops
-
sewardj authored
Change from 3-component to 4-component representation for BCOs
-
- 14 Dec, 2000 9 commits
-
-
sewardj authored
itbl renaming wibbles
-
sewardj authored
more interpreter itbl hackery
-
sewardj authored
itbl renaming wibbles
-
sewardj authored
Add info tables to do compiled->interpreted returns and vice versa. Rename various stuff from ...Hugs... to ...Interp...
-
sewardj authored
So long, dudes! It's been Real (tm).
-
sewardj authored
Clarify itbl and linking issues for bcos, and add flag -ddump-bcos.
-
simonmar authored
don't mess with $(MKDEPENDHS)
-
simonmar authored
$(MKDEPENDHS) = $(GHC), not $(HC).
-
sewardj authored
Wrap entire file in #ifdef GHCI, to create link errors if malconfigured
-
- 13 Dec, 2000 5 commits
- 12 Dec, 2000 15 commits
-
-
simonmar authored
Take a pass over the tests, get most things working again. Mostly -package lang changes, and synching up of the expected output. This was a good exercise though: I found at least two real bugs in the process :)
-
simonmar authored
Oops! Don't forget to remove type variables from the binders of a case alternative when converting to STG.
-
sewardj authored
track changes in ByteCodeGen.lhs
-
sewardj authored
More assembler work. Mostly done. Still need to import itbl stuff from old interpreter. Must remember to order new hair to replaced all I tore out today.
-
simonmar authored
error message wibble
-
sewardj authored
Get the assembler infrastructure more or less correct.
-
simonmar authored
Add missing default case from isSpecial
-
simonmar authored
-noC has been removed
-
simonmar authored
wibble: get the exception handlers round the right way
-
simonmar authored
Clean up the error handling a bit; the exception type is moved to Panic, and a new exception for panics has been added.
-
simonmar authored
turn off recompilation checking in here
-
simonmar authored
Allow for compiling to .hc only (substitute for now-defunct -noC option)
-
simonmar authored
Make the foreign export dynamic helper function "exported", so it doesn't get thrown away.
-
simonmar authored
- Add primops for {read,write,index}{Int,Word}{8,16,32,64}OffAddr. This enables us to remove all the _casm_s from Int/Word. - Replace new{Char,Int,etc.}Array# with newByteArray# (save a few primops, at the cost of having to know the size of these types in PrelArr). - Implement MArray/IArray support for sized types. finally. - Move the guts of the sized types into ghc/lib/std, we'll need them for doing more FFIish things in the Prelude.
-
simonmar authored
Treat an unhandled exception in the same way as a panic (i.e. the `impossible' happened), to make it clear that these really are compiler bugs.
-