5db1a9626a* added compile-core-expr-to-port in order to compile the whole program at once
Abdulaziz Ghuloum
2007-04-28 01:42:12 -0400
f27c2df625* assign-frame-sizes is as complete as could be (time for assigning libcompile.ss dropped from ~5 secs to 1.8 secs).
Abdulaziz Ghuloum
2007-03-11 20:55:43 -0400
068bc38e42* eliminated a few instances of set-rem and set->list.
Abdulaziz Ghuloum
2007-03-11 20:39:21 -0400
2b6bcc324b* when a variable is assigned a location, it does not need to remove itself from the edges of its conflicting nodes.
Abdulaziz Ghuloum
2007-03-11 20:07:00 -0400
c473619084* important realization that set->list is perhaps the most expensive operation on sets. Working on eliminating most uses of set->list.
Abdulaziz Ghuloum
2007-03-11 20:00:08 -0400
2b5b555bac* Just realized that you don't need all var<->var conflicts to be recorded. All is needed is for a var $x$ to know some of its var conflicts as long as the other conflicts know about $x$. I'll be working on this now. I'm hoping for a 50% speedup in the assign-frame-sizes pass.
Abdulaziz Ghuloum
2007-03-11 18:54:15 -0400
aa1c7e1bb7* before changing number of bits to 28
Abdulaziz Ghuloum
2007-03-11 03:40:47 -0400
3cb18c56e9* deleting old list-based set representation.
Abdulaziz Ghuloum
2007-03-10 21:01:41 -0500
fa94d2f6a1* added a few missing instruction sequences where the dest is a memory location. * implemented integer sets (based on tree representation).
Abdulaziz Ghuloum
2007-03-10 19:50:24 -0500
10bdc3dcd6* starting with IntegerSet
Abdulaziz Ghuloum
2007-03-10 16:47:13 -0500
d4559cfd4ccleanup of repository content.
Abdulaziz Ghuloum
2007-03-09 19:28:38 -0500
5bb666bf7f* last commit on alt-cogen and beginning of alt-alt-cogen :-)
Abdulaziz Ghuloum
2007-03-09 14:01:17 -0500
893a2decc8* added bignum? primitive and removed ikrt_isbignum from the runtime system.
Abdulaziz Ghuloum
2007-03-04 13:02:39 -0500
c6861f41d5* added flonum? and removed ikrt_is_flonum from the runtime system.
Abdulaziz Ghuloum
2007-03-04 12:59:28 -0500
b326fbc07a* completed the move to new specify-representation pass infrastructure.
Abdulaziz Ghuloum
2007-03-03 23:17:04 -0500
cf65729971more more more
Abdulaziz Ghuloum
2007-03-02 02:47:36 -0500
e9740fa34f* fixed bug in the graph construction that caused the operand of idivl not to be added to the live set.
Abdulaziz Ghuloum
2007-03-02 00:41:28 -0500
de7c43a16b* optimized symbol calls using the symbol-function field. * added error checks for applying nonprocedures.
Abdulaziz Ghuloum
2007-02-25 21:29:28 -0500
87d8d5a5dd* added a $set-symbol-function! primitive.
Abdulaziz Ghuloum
2007-02-24 15:42:57 -0500
d99c22e8c4* The shortcut form is now two parts: a body and a handler. The evaluation of the body can jump to the handler at any point during evaluation by means of using the interrupt primitive or by overflow.
Abdulaziz Ghuloum
2007-02-22 23:02:50 -0500
819a3ca1c1* bug in collector causing dirty bits for code objects with younger reloc vectors to be unset incorrectly. The bug was in using || instead of | when computing the dirty pattern.
Abdulaziz Ghuloum
2007-02-22 21:58:38 -0500
0bd88991cb* looks not bad.
Abdulaziz Ghuloum
2007-02-19 23:33:29 -0500
30f71b0381* frame moves are not working as nicely as they should.
Abdulaziz Ghuloum
2007-02-19 18:21:35 -0500
b6dd620b94* chaitin can now compile the system.
Abdulaziz Ghuloum
2007-02-17 19:22:14 -0500
0304c85082* pre-spilling call-live variables works for the test cases now.
Abdulaziz Ghuloum
2007-02-17 18:09:03 -0500
85bf359150* psyntax is not working under chaitin.
Abdulaziz Ghuloum
2007-02-16 10:11:21 -0500
649e7f022a* libchezio and libhash are remaining
Abdulaziz Ghuloum
2007-02-15 23:54:39 -0500
1a4cdcb7b0* libcore now compiles with chaitin
Abdulaziz Ghuloum
2007-02-14 19:42:36 -0500
434ebe9525* The $apply macro and all the appcall forms were removed from the compiler after realizing that I was stupid. ($appcall rator rand*) was just (jmpcal SL_apply rator rand*)
Abdulaziz Ghuloum
2007-02-14 15:50:34 -0500
8294a8dee9* fixed an assembler bug in "addl reg -> mem"
Abdulaziz Ghuloum
2007-02-13 17:24:00 -0500
297e47db32* adding some asm tests in asm-tests.ss
Abdulaziz Ghuloum
2007-02-13 05:08:48 -0500
a123a77bbc* nothing much
Abdulaziz Ghuloum
2007-02-13 02:05:58 -0500
826adfe9dd* can now compile libtoplevel
Abdulaziz Ghuloum
2007-02-12 23:19:56 -0500
f88e3a8b65* libnumeric can be compiled now with chaitin
Abdulaziz Ghuloum
2007-02-12 23:03:41 -0500
2e477881b6* split libcontrol into primitive libcontrol0 and extended libcontrol1.
Abdulaziz Ghuloum
2007-02-12 19:17:31 -0500
fad2c4e999* fixed bug in assembler * assembler now handles orl imm, mem
Abdulaziz Ghuloum
2007-02-12 17:59:58 -0500
f766ca1148* passes tests, but fails to compile psyntax.
Abdulaziz Ghuloum
2007-02-12 13:58:04 -0500
fa6e499b22* before attempting to use chaitin to compile the system.
Abdulaziz Ghuloum
2007-02-11 21:42:01 -0500
126b7aa8fa* call-with-values where the second argument is a single-value procedure is now direct-call-optimized.
Abdulaziz Ghuloum
2007-01-26 10:23:07 -0500
8139a91a61* Modified the reader and writer to accept "->[subsequent]*" as an identifier.
Abdulaziz Ghuloum
2007-01-26 09:05:07 -0500
82375a3ddc* new-cogen now removes all primcalls before proceeding. Ideally, no primcalls should be introduced at that point.
Abdulaziz Ghuloum
2007-01-21 22:24:12 -0500
f114fc2f78* Started work on new cogen
Abdulaziz Ghuloum
2007-01-21 21:20:39 -0500
1c37a4ee54* changing representation for all non-fixnum numbers.
Abdulaziz Ghuloum
2007-01-21 19:20:37 -0500
783beb990b* exact->inexact is now implemnted. * +, -, and * now handles flonums by converting exact arguments to inexact if the other argument is a flonum.
Abdulaziz Ghuloum
2007-01-20 19:26:17 -0500
96ad8a04a4* flonum->string now always returns a string with a decimal point. * number?, exact?, inexact?, ... now know about flonums. * number->string also handles flonums, and so is the writer (automatically).
Abdulaziz Ghuloum
2007-01-20 18:12:12 -0500
c6bc77bc9d* Many numeric predicates are added (exact?, inexact?, real?, complex?, etc.). * min and max are added.
Abdulaziz Ghuloum
2007-01-20 16:52:22 -0500
679af26761* Added benchmarks from the larceny source distribution.
Abdulaziz Ghuloum
2007-01-20 14:32:30 -0500
a47856614f* The procedure member is now defined.
Abdulaziz Ghuloum
2007-01-20 00:09:11 -0500
c8111df150* fixed bug in handling tcbuckets. Summary of the bug: if ik_alloc is called during gc, and if it causes a heap overflow, then the pages are deallocated before returning to scheme.
Abdulaziz Ghuloum
2007-01-19 18:13:44 -0500
8b1236e031* removed all dlink stuff from hash table.
Abdulaziz Ghuloum
2007-01-17 14:56:45 -0500
fe89891d96updated tests in libpp.ss
Abdulaziz Ghuloum
2007-01-13 21:58:14 -0500
92e278281c* pretty-print is now added to ikarus. * new-cafe now uses pretty-print instead of write for printing to the interaction repl.
Abdulaziz Ghuloum
2007-01-13 21:33:04 -0500
bf23a8a1b7* pretty-print now handles vectors
Abdulaziz Ghuloum
2007-01-13 20:37:54 -0500
38082fc489pretty-print now handles read-macros.
Abdulaziz Ghuloum
2007-01-13 18:51:23 -0500
Abdulaziz Ghuloum