Commit Graph

82 Commits

Author SHA1 Message Date
Abdulaziz Ghuloum fb48ef12bd * disabled gzipped fasl files. too slow :-( 2007-05-22 19:01:07 -04:00
Abdulaziz Ghuloum 580481d8fc * The runtime system can now open gzipped bootfiles in addition to
uncompressed fasl files.
2007-05-22 17:56:15 -04:00
Abdulaziz Ghuloum 890dd348b2 * fasl files now may contain ascii strings as a special type in
order to reduce the size of the fasl file.  Gains: reduces the
  size of the current boot file from 2.9M down to 2.5M.
2007-05-21 23:09:45 -04:00
Abdulaziz Ghuloum 1abce54167 * string transition completed. string data is now 32-bit
tagged characters.
2007-05-19 14:38:45 -04:00
Abdulaziz Ghuloum 182de12428 * cleaned up some uses of naked strings in the runtime system. 2007-05-19 14:13:51 -04:00
Abdulaziz Ghuloum 267da9e77c * Writing strings to fasl files now uses 4-byte words for each
character.
2007-05-19 13:54:13 -04:00
Abdulaziz Ghuloum 1c4d3c4b40 * uuid now uses a bytevector for ffi. 2007-05-18 22:31:00 -04:00
Abdulaziz Ghuloum 027d50e84a * marked env and environ procedures busted. 2007-05-18 22:26:22 -04:00
Abdulaziz Ghuloum e499eb0f5a * delete-file now uses bytevectors for filename rep. 2007-05-18 22:21:36 -04:00
Abdulaziz Ghuloum cc500489b7 * file-exists? now uses bytevectors for filenames. 2007-05-18 22:18:51 -04:00
Abdulaziz Ghuloum b54cc9a762 * system now passes a bytevector to ik_system. 2007-05-18 22:15:52 -04:00
Abdulaziz Ghuloum 862871a787 * bignum->string now uses bytevectors as intermediate rep. 2007-05-18 22:08:45 -04:00
Abdulaziz Ghuloum 691a42fe4e * string->flonum now uses bytevectors as intermediate rep. 2007-05-18 22:01:04 -04:00
Abdulaziz Ghuloum 7ca0ddfa81 * flonum->string now uses a bytevector as the intermediate buffer. 2007-05-18 21:52:04 -04:00
Abdulaziz Ghuloum 179063c117 * open-output-file now uses a bytevector instead of a string to
represent the filename during ffi.
2007-05-18 21:06:25 -04:00
Abdulaziz Ghuloum 1586986dbe * open-input-file now uses a bytevector representing the filename
during ffi.
2007-05-18 21:03:14 -04:00
Abdulaziz Ghuloum 498084a6e8 * date-string now uses bytevectors instead of strings during ffi. 2007-05-18 20:59:10 -04:00
Abdulaziz Ghuloum 85944a2cff * Added utf8-bytevector->string decoder 2007-05-18 20:18:55 -04:00
Abdulaziz Ghuloum 10f5e53338 * foreign procedures are now represented as bytevectors instead of
strings.
2007-05-18 18:55:20 -04:00
Abdulaziz Ghuloum dfe8e0e1c1 * ikarus-fasl.c now recognizes the "v" marker for bytevectors 2007-05-18 18:16:52 -04:00
Abdulaziz Ghuloum 35fa003558 * Added $write-byte to ikarus.io-primitives.unsafe.ss 2007-05-18 08:15:51 -04:00
Abdulaziz Ghuloum 52a28f8332 * input ports now use bytevectors fully. 2007-05-17 06:44:51 -04:00
Abdulaziz Ghuloum 49dc13d5ee * make string operations safe in all io layers. 2007-05-17 05:41:00 -04:00
Abdulaziz Ghuloum 5249a8ec31 * The gc now knows about bytevectors 2007-05-17 04:34:52 -04:00
Abdulaziz Ghuloum 97f59ad1ee * transition of symbols to secondary type done. 2007-05-15 11:37:43 -04:00
Abdulaziz Ghuloum 6bdb50004a * incremental step towards loading a fresh compiler for
bootstrapping.
2007-05-15 08:56:22 -04:00
Abdulaziz Ghuloum 7dc4f7617b * ikarus-runtime no longer assigns the value of $base-rtd to the
base-rtd object.  The reader now accesses the base-rtd from the
  pcb directly.
2007-05-09 19:45:41 -04:00
Abdulaziz Ghuloum 8226619438 * install-library now takes an extra visibility flag.
* installed-libraries now takes an optional "all?" flag:
  - if all? is true, it returns all libraries.
  - if all? is false, it returns only the visible libraries
  - all? defaults to #t.
2007-05-07 04:52:22 -04:00
Abdulaziz Ghuloum 48e062dfc4 * base-rtd is now placed in the pcb.
* renamed:
    src/library-manager.ss => src/ikarus.library-manager.ss
2007-05-05 22:42:26 -04:00
Abdulaziz Ghuloum 5bb666bf7f * last commit on alt-cogen and beginning of alt-alt-cogen :-) 2007-03-09 14:01:17 -05:00
Abdulaziz Ghuloum 893a2decc8 * added bignum? primitive and removed ikrt_isbignum from the runtime
system.
2007-03-04 13:02:39 -05:00
Abdulaziz Ghuloum c6861f41d5 * added flonum? and removed ikrt_is_flonum from the runtime system. 2007-03-04 12:59:28 -05:00
Abdulaziz Ghuloum b326fbc07a * completed the move to new specify-representation pass infrastructure. 2007-03-03 23:17:04 -05:00
Abdulaziz Ghuloum cf65729971 more more more 2007-03-02 02:47:36 -05:00
Abdulaziz Ghuloum de7c43a16b * optimized symbol calls using the symbol-function field.
* added error checks for applying nonprocedures.
2007-02-25 21:29:28 -05:00
Abdulaziz Ghuloum 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.
2007-02-22 21:58:38 -05:00
Abdulaziz Ghuloum 8294a8dee9 * fixed an assembler bug in "addl reg -> mem" 2007-02-13 17:24:00 -05:00
Abdulaziz Ghuloum 297e47db32 * adding some asm tests in asm-tests.ss 2007-02-13 05:08:48 -05:00
Abdulaziz Ghuloum d6c2d5dd66 * representation (tags) of bignums/flonums changed successfully. 2007-01-21 20:36:22 -05:00
Abdulaziz Ghuloum 1c37a4ee54 * changing representation for all non-fixnum numbers. 2007-01-21 19:20:37 -05:00
Abdulaziz Ghuloum 783beb990b * exact->inexact is now implemnted.
* +, -, and * now handles flonums by converting exact arguments to
  inexact if the other argument is a flonum.
2007-01-20 19:26:17 -05:00
Abdulaziz Ghuloum 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).
2007-01-20 18:12:12 -05:00
Abdulaziz Ghuloum de910a3544 * added flonum?, string->flonum, and flonum->string. 2007-01-20 17:49:25 -05:00
Abdulaziz Ghuloum c6bc77bc9d * Many numeric predicates are added (exact?, inexact?, real?,
complex?, etc.).
* min and max are added.
2007-01-20 16:52:22 -05:00
Abdulaziz Ghuloum 679af26761 * Added benchmarks from the larceny source distribution. 2007-01-20 14:32:30 -05:00
Abdulaziz Ghuloum 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.
2007-01-19 18:13:44 -05:00
Abdulaziz Ghuloum 13e8f76f13 * intermediate step:
gc has tcbucket_size=16
  compile allocates tcbucket_size=24
2007-01-17 14:34:25 -05:00
Abdulaziz Ghuloum 872a5e2e9e * Added even? and odd? primitives. 2007-01-13 22:32:54 -05:00
Abdulaziz Ghuloum 212224b4cf * Added the primitive (environ):
returns an alist of (string x string) as obtained from the
    system environ.
* Added the primitive env:
  (env "key") retrieves the env value associated with key
  (env "key" "val") sets the env value assocuated with key to val
  (env "key" "val" overwrite?) same as above but it does not
  overwrite an existing value of overwrite? is #f.
2007-01-13 00:42:37 -05:00
Abdulaziz Ghuloum 63575db860 * makefile.ss is updated to use fork and waitpid in order to make
use of the dual-core capability of my macbook :-)  
  Compile-time is now 1.5secs real-time.
2007-01-01 02:48:17 +03:00