diff options
author | simonmar <unknown> | 1999-11-02 15:06:05 +0000 |
---|---|---|
committer | simonmar <unknown> | 1999-11-02 15:06:05 +0000 |
commit | f6692611aad945e46ffb615bde1df7def3fc742f (patch) | |
tree | 04e2e2af9c43eba1b60312b89eb3ac8f34209e2c /ghc/includes/Stg.h | |
parent | 947d2e363f75e9e230d535c876ecdafba45174b5 (diff) | |
download | haskell-f6692611aad945e46ffb615bde1df7def3fc742f.tar.gz |
[project @ 1999-11-02 15:05:38 by simonmar]
This commit adds in the current state of our SMP support. Notably,
this allows the new way 's' to be built, providing support for running
multiple Haskell threads simultaneously on top of any pthreads
implementation, the idea being to take advantage of commodity SMP
boxes.
Don't expect to get much of a speedup yet; due to the excessive
locking required to synchronise access to mutable heap objects, you'll
see a slowdown in most cases, even on a UP machine. The best I've
seen is a 1.6-1.7 speedup on an example that did no locking (two
optimised nfibs in parallel).
- new RTS -N flag specifies how many pthreads to start.
- new driver -smp flag, tells the driver to use way 's'.
- new compiler -fsmp option (not for user comsumption)
tells the compiler not to generate direct jumps to
thunk entry code.
- largely rewritten scheduler
- _ccall_GC is now done by handing back a "token" to the
RTS before executing the ccall; it should now be possible
to execute blocking ccalls in the current thread while
allowing the RTS to continue running Haskell threads as
normal.
- you can only call thread-safe C libraries from a way 's'
build, of course.
Pthread support is still incomplete, and weird things (including
deadlocks) are likely to happen.
Diffstat (limited to 'ghc/includes/Stg.h')
-rw-r--r-- | ghc/includes/Stg.h | 37 |
1 files changed, 29 insertions, 8 deletions
diff --git a/ghc/includes/Stg.h b/ghc/includes/Stg.h index 9b2ab0d5c0..756e8fb51a 100644 --- a/ghc/includes/Stg.h +++ b/ghc/includes/Stg.h @@ -1,5 +1,5 @@ /* ----------------------------------------------------------------------------- - * $Id: Stg.h,v 1.17 1999/07/06 09:42:39 sof Exp $ + * $Id: Stg.h,v 1.18 1999/11/02 15:05:52 simonmar Exp $ * * (c) The GHC Team, 1998-1999 * @@ -16,6 +16,17 @@ #define _POSIX_SOURCE #endif +/* If we include "Stg.h" directly, we're in STG code, and we therefore + * get all the global register variables, macros etc. that go along + * with that. If "Stg.h" is included via "Rts.h", we're assumed to + * be in vanilla C. + */ +#ifdef NOT_IN_STG_CODE +#define NO_REGS /* don't define fixed registers */ +#else +#define IN_STG_CODE +#endif + /* Configuration */ #include "config.h" #ifdef __HUGS__ /* vile hack till the GHC folks come on board */ @@ -33,13 +44,17 @@ * For now, do lazy and not eager. */ -#define LAZY_BLACKHOLING -/* #define EAGER_BLACKHOLING */ - -#ifdef TICKY_TICKY -/* TICKY_TICKY needs EAGER_BLACKHOLING to verify no double-entries of single-entry thunks. */ -# undef LAZY_BLACKHOLING -# define EAGER_BLACKHOLING +/* TICKY_TICKY needs EAGER_BLACKHOLING to verify no double-entries of + * single-entry thunks. + * + * SMP needs EAGER_BLACKHOLING because it has to lock thunks + * synchronously, in case another thread is trying to evaluate the + * same thunk simultaneously. + */ +#if defined(SMP) || defined(TICKY_TICKY) +# define EAGER_BLACKHOLING +#else +# define LAZY_BLACKHOLING #endif /* ToDo: Set this flag properly: COMPILER and INTERPRETER should not be mutually exclusive. */ @@ -96,8 +111,10 @@ void _stgAssert (char *, unsigned int); #include "ClosureTypes.h" #include "InfoTables.h" #include "TSO.h" +#include "Block.h" /* STG/Optimised-C related stuff */ +#include "SMP.h" #include "MachRegs.h" #include "Regs.h" #include "TailCalls.h" @@ -121,6 +138,10 @@ void _stgAssert (char *, unsigned int); #include <unistd.h> #endif +#ifdef SMP +#include <pthread.h> +#endif + /* GNU mp library */ #include "gmp.h" |