clean up document makefiles.

add -s option for extra obj suffixes, like mkdependHS

Default WithGhcHc is now just "ghc".  Too many people complained.

fixup tests for 4.00

fix conc004 for new type of forkIO.
add conc007: killThread test.

mangle percent symbols properly.

update for 4.00, correct untruths

AT_GLASGOW went away.

fix a couple of bugs in the text/data segment end symbol tests.

back out last commit on trunk (should have gone to new-rts branch).

back out revision 1.2 (change made on wrong branch).

Add support for FreeBSD 3.0 (ELFish).
Rename i386-unknown-freebsd to i386-unknown-freebsd2, and add
i386-unknown-freebsd3.

NOTE: this won't bootstrap an ELF GHC using an a.out GHC, some laying
on of hands is required for that.

sync with recent changes to configure.in.

Generalise --exlude-module=M handling slightly, it now
will ignore any appearances of M along the user import
path too.

Included fixity info declared elsewhere; added 'seq' infix decl

tweaked Integral.Int.div to cope with overflows

Protect uses of ForeignObj with #ifndef __PARALLEL_HASKELL__

Added showIntAtBase to export list

For posterity: some older entries

-Added entries for library changes + new FFI support.

new helper: openAlphaTyVars

wibble

merged from new-rts: permit unsafeCoerce# to be applied to values with an unboxed type

Added CODE_BEFORE_DATA, DATA_SECTION_END_MARKER, TEXT_SECTION_END_MARKER

added use of FPTOOLS_CODE_BEFORE_DATA, FPTOOLS_END_DATA_SECTION, FPTOOLS_END_TEXT_SECTION

Added feature tests for etext and end plus text vs. data section placement in executable

merged from HEAD

disable posix subdir, one of the tests hangs on Linux.

quick fix for multi-file sgml docs.

Some Numeric lib tests

remove socket, these tests don't seem to be suitable for
non-interactive use.

In an interface file, don't allow data and newtype decls to have deriving clauses (not needed.)

add #ifndef __PARALLEL_HASKELL__ around ForeignObj bits.

fix typo

Bomb out if we encounter a _ccall_GC_; can't handle it.

Did I say first argument in the previous commit msg, I meant
first component of the result tuple :-)

collectArgs: return function expr. as first argument, not the whole
expression.

floatToDigits bugfix

Surplus to requirements (cf. last IO commit)

Spell out which IO functions the Prelude provides