Changeset 6548


Ignore:
Timestamp:
May 13, 2007, 12:22:47 AM (15 years ago)
Author:
gb
Message:
 
File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/ccl/ChangeLog

    r6438 r6548  
     12007-05-12 18:27  gb
     2
     3        * release-notes.txt: Update for 070512.
     4
     52007-05-12 18:27  gb
     6
     7        * lisp-kernel/ppc-exceptions.c: Fix typo in new_heap_segment().
     8
     92007-05-12 18:26  gb
     10
     11        * compiler/PPC/ppc2.lisp: Fix typo in ppc2-%i-
     12
     132007-05-12 18:25  gb
     14
     15        * level-1/version.lisp: Bump.
     16
     172007-05-12 18:24  gb
     18
     19        * lisp-kernel/ppc-exceptions.c: new_heap_segment(): don't extend
     20        heap by more than what we need if GC is obviously disabled.
     21
     222007-05-12 18:23  gb
     23
     24        * compiler/PPC/ppc2.lisp: Use *ppc2-target-half-fixnum-type* to
     25        suppress overflow checking in some cases.
     26
     272007-05-12 18:22  gb
     28
     29        * compiler/X86/x862.lisp: Fix indentation.
     30
     312007-05-12 18:22  gb
     32
     33        * lisp-kernel/x86-exceptions.c: new_heap_segment(): don't extend
     34        beyond segment if GC (obviously) disabled.
     35
     362007-05-12 18:21  gb
     37
     38        * level-1/l1-streams.lisp: Fix typo.
     39
     402007-05-12 18:20  gb
     41
     42        * level-1/: l1-streams.lisp, l1-unicode.lisp: Keep bootsrapping
     43        separate character encode/decode limits.
     44
     452007-05-12 18:19  gb
     46
     47        * level-1/l1-unicode.lisp: Start to bootstrap separate
     48        encode/decode limits.
     49
     502007-05-12 18:18  gb
     51
     52        * lisp-kernel/lisp-debug.c: Hide Darwin/Leopard exception context
     53        nonsense when referencing mxcsr in debug_show_fpu().
     54
     552007-05-12 18:17  gb
     56
     57        * lisp-kernel/lisptypes.h: Need alias for __r6 in Tiger and
     58        earlier.  Sheesh.
     59
     602007-05-12 18:17  gb
     61
     62        * level-0/l0-hash.lisp: Fix bad call to %pname-hash.
     63
     642007-05-12 18:16  gb
     65
     66        * level-1/version.lisp: Bump, for now.
     67
     682007-05-12 18:15  gb
     69
     70        * xdump/: faslenv.lisp, heap-image.lisp: Bump fasl versions, image
     71        ABI version.
     72
     732007-05-12 18:14  gb
     74
     75        * lisp-kernel/x86-spentry64.s: nmkunwind.  New calling sequence;
     76        new TRA interpretation.
     77
     782007-05-12 18:13  gb
     79
     80        * lisp-kernel/x86-spjump64.s: nmkunwind.
     81
     822007-05-12 18:12  gb
     83
     84        * lisp-kernel/x86-macros.s: Make_Catch uses %imm2, not %temp2.
     85        nMake_Catch handles return addresses on stack.  repret, for use
     86        when a one-byte ret instruction is a branch target or is preceded
     87        by a conditional branch.
     88
     892007-05-12 18:11  gb
     90
     91        * lisp-kernel/x86-gc.c: New TRA handling; new node regs in
     92        exception contexts.
     93
     942007-05-12 18:11  gb
     95
     96        * lisp-kernel/x86-exceptions.h: New stuff for finding FP state in
     97        exception context; new TRA handling.
     98
     992007-05-12 18:10  gb
     100
     101        * lisp-kernel/x86-exceptions.c: New tra handling.
     102        new_heap_segment: try to avoid requesting too much when user sets
     103        lisp_heap_gc_threshold to inappropriate value.  Pass old_valence
     104        down to fault handler, don't call out to lisp on faults in foreign
     105        code.  Handle USE_MACH_EXCEPTION_LOCK conditionalization.
     106
     1072007-05-12 18:08  gb
     108
     109        * lisp-kernel/: x86-constants64.h, x86-constants64.s: New register
     110        assignments.
     111
     1122007-05-12 18:07  gb
     113
     114        * lisp-kernel/x86-subprims64.s: New tra handling, new calling
     115        sequence.
     116
     1172007-05-12 18:06  gb
     118
     119        * lisp-kernel/: xlbt.c, x86_print.c: New tra handling.
     120
     1212007-05-12 18:05  gb
     122
     123        * lisp-kernel/x86-asmutils64.s: Use repret for RET instructions
     124        that're branch targets or that follow conditional branches.
     125
     1262007-05-12 18:04  gb
     127
     128        * lisp-kernel/thread_manager.c: Conditionalize for
     129        USE_MACH_EXCEPTION_LOCK.
     130
     1312007-05-12 18:03  gb
     132
     133        * lisp-kernel/ppc-spjump.s: .SPnmkunwind.
     134
     1352007-05-12 18:02  gb
     136
     137        * lisp-kernel/ppc-spentry.s: .SPnmkunwind.  Old catch/unwind: do
     138        explicit blr.
     139
     1402007-05-12 18:01  gb
     141
     142        * lisp-kernel/darwinppc/Makefile: Use -Wno-deprecated-declarations.
     143         Gee, I hope that works.
     144
     1452007-05-12 18:00  gb
     146
     147        * lisp-kernel/ppc-macros.s: Make_Catch macro: don't include a blr.
     148
     1492007-05-12 17:59  gb
     150
     151        * lisp-kernel/ppc-exceptions.h: Only declare mach_exception_lock if
     152        using the mechanism.
     153
     1542007-05-12 17:59  gb
     155
     156        * lisp-kernel/ppc-exceptions.c: new_heap_segment: if we can't
     157        extend with (possibly very large) lisp_heap_gc_threshold, try
     158        smaller values.
     159       
     160        Pass "old_valence" to exception handlers; don't call out to lisp on
     161        memory faults from foreign code.
     162
     1632007-05-12 17:58  gb
     164
     165        * lisp-kernel/lisptypes.h: xmm register alias nonsense.
     166
     1672007-05-12 17:57  gb
     168
     169        * lisp-kernel/lisp-exceptions.h: Change some function prototypes.
     170        Don't use a MACH_EXCEPTION_LOCK on Darwin, since it doesn't seem to
     171        be correct/necessary.
     172
     1732007-05-12 17:56  gb
     174
     175        * lisp-kernel/lisp-debug.c: lisp register display changes for
     176        x86-64.  Get FPU display working on x86-64.
     177
     1782007-05-12 17:55  gb
     179
     180        * lisp-kernel/linuxx8664/.gdbinit: gtra, just in time for the
     181        calling-sequence change.
     182
     1832007-05-12 17:54  gb
     184
     185        * lisp-kernel/image.h: New ABI version.
     186
     1872007-05-12 17:54  gb
     188
     189        * lisp-kernel/darwinx8664/.gdbinit: Ignore SIGSYS.  If only I
     190        could; if only I could ...
     191
     1922007-05-12 17:53  gb
     193
     194        * lib/x8664env.lisp: More imm temps! (one more ...).
     195
     1962007-05-12 17:52  gb
     197
     198        * lib/nfcomp.lisp: Bump current fasl version.
     199
     2002007-05-12 17:51  gb
     201
     202        * lib/macros.lisp: RLET, MAKE-RECORD, etc: don't set pointer type
     203        in cross-compiled code.
     204
     2052007-05-12 17:50  gb
     206
     207        * lib/foreign-types.lisp: Don't use #. to initialize *host-ftd*
     208        interface dir, because of cross-compilation issues.
     209
     2102007-05-12 17:49  gb
     211
     212        * lib/db-io.lisp: Support "queries" in FFI reader macros: #_?foo,
     213        #$?foo, and #&?foo return T if foo is defined, NIL otherwise.
     214
     2152007-05-12 17:49  gb
     216
     217        * lib/compile-ccl.lisp: Lose some stuff leftover from 68k->PPC MCL
     218        transition.
     219
     2202007-05-12 17:48  gb
     221
     222        * lib/ccl-export-syms.lisp: Mapped-vector stuff.
     223
     2242007-05-12 17:47  gb
     225
     226        * lib/encapsulate.lisp: Add a comment.
     227
     2282007-05-12 17:46  gb
     229
     230        * library/hash-cons.lisp: Very small lap change.
     231
     2322007-05-12 17:45  gb
     233
     234        * level-1/ppc-threads-utils.lisp: Move %FIXNUM-REF-MACPTR and
     235        %FIXNUM-SET-MACPTR elsewhere.
     236
     2372007-05-12 17:44  gb
     238
     239        * level-1/linux-files.lisp: File-mapped vectors.
     240
     2412007-05-12 17:43  gb
     242
     243        * level-1/l1-streams.lisp: OPEN : :if-exists defaults to
     244        :new-version if pathname-version of filename is :NEWEST.
     245
     2462007-05-12 17:43  gb
     247
     248        * level-1/l1-readloop-lds.lisp: %BREAK-IN-FRAME might be called
     249        with a CONDITION object.
     250
     2512007-05-12 17:42  gb
     252
     253        * level-1/l1-processes.lisp: *HOST-PAGE-SIZE* initialized earlier.
     254
     2552007-05-12 17:41  gb
     256
     257        * level-1/l1-pathnames.lisp: If there's no host component in a
     258        namestring argument, LOGICAL-PATHNAME signals a TYPE-ERROR.
     259
     2602007-05-12 17:40  gb
     261
     262        * level-1/l1-lisp-threads.lisp: Initial-lisp-thread, initial
     263        listener stack sizes now static.
     264
     2652007-05-12 17:39  gb
     266
     267        * level-1/l1-files.lisp: Pathname arcana.
     268
     2692007-05-12 17:38  gb
     270
     271        * level-1/l1-events.lisp: Optionally, invoke debugger-hook on
     272        SIGINT.
     273
     2742007-05-12 17:38  gb
     275
     276        * level-1/l1-dcode.lisp: Clear dispatch table cache if GF APO
     277        changes.
     278
     2792007-05-12 17:37  gb
     280
     281        * level-0/l0-misc.lisp: Don't cons when clearing spin lock.
     282       
     283        Commented-out (#+bad-idea) GC-lock-via-exception-lock idea.
     284
     2852007-05-12 17:36  gb
     286
     287        * level-0/l0-hash.lisp: Use %pname-hash to hash ... pnames.
     288
     2892007-05-12 17:35  gb
     290
     291        * level-0/l0-def.lisp: %FIXNUM-REF-MACPTR, %FIXNUM-SET-MACPTR: move
     292        here.
     293
     2942007-05-12 17:34  gb
     295
     296        * level-0/l0-cfm-support.lisp: Remove an old declaration.
     297
     2982007-05-12 17:33  gb
     299
     300        * level-0/X86/x86-utils.lisp: New calling sequence.
     301
     3022007-05-12 17:32  gb
     303
     304        * level-0/X86/x86-pred.lisp: New calling sequence. Use a CMOV in
     305        EQL.
     306
     3072007-05-12 17:32  gb
     308
     309        * level-0/X86/x86-numbers.lisp: New calling sequence.  No need to
     310        be careful with %rcx (now %imm2) in shift instructions.
     311
     3122007-05-12 17:30  gb
     313
     314        * level-0/X86/: x86-float.lisp, x86-misc.lisp: New calling
     315        sequence.
     316
     3172007-05-12 17:29  gb
     318
     319        * level-0/X86/x86-def.lisp: Handle new-style tras, other
     320        new-calling-sequence changes.
     321
     3222007-05-12 17:27  gb
     323
     324        * level-0/X86/: X8664/x8664-bignum.lisp, x86-array.lisp,
     325        x86-clos.lisp: New calling sequence.  %rcx is %imm2 now, not
     326        %temp2, and no need to be so careful with it.
     327
     3282007-05-12 17:26  gb
     329
     330        * compiler/vinsn.lisp: global label-valued temporaries.
     331
     3322007-05-12 17:25  gb
     333
     334        * compiler/optimizers.lisp: FLOAT with no protototype argument.
     335
     3362007-05-12 17:24  gb
     337
     338        * compiler/nx0.lisp: Try to do a little type inference for more
     339        flavors of AREF.
     340
     3412007-05-12 17:23  gb
     342
     343        * compiler/nx-basic.lisp: FUNCTION-INFORMATION might return an
     344        afunc (or equivalent) as a non-nil second value.
     345
     3462007-05-12 17:22  gb
     347
     348        * compiler/X86/x862.lisp: Support the new (call/ret) calling
     349        sequence, new tra/talign scheme.  One more imm reg, so 3d aset is
     350        less nasty.
     351
     3522007-05-12 17:21  gb
     353
     354        * compiler/X86/x86-lapmacros.lisp: Support the new (call/ret)
     355        calling sequence, new tra/talign scheme.
     356
     3572007-05-12 17:21  gb
     358
     359        * compiler/X86/x86-lap.lisp: Support :talign, (@ (:^ label) (%
     360        rip)).
     361
     3622007-05-12 17:20  gb
     363
     364        * compiler/X86/x86-disassemble.lisp: Recognize CALL label operands,
     365        rip-relative label refs.
     366
     3672007-05-12 17:19  gb
     368
     369        * compiler/X86/x86-backend.lisp: Allow label-valued temporaries,
     370        support :talign directive in vinsn templates.
     371
     3722007-05-12 17:18  gb
     373
     374        * compiler/X86/x86-asm.lisp: Add call instruction; note that %rip
     375        is a register ...
     376
     3772007-05-12 17:17  gb
     378
     379        * compiler/X86/X8664/x8664-vinsns.lisp: Use "tail-aligned" calls
     380        for most things, except calls to subprims that manipulate the
     381        stack.
     382
     3832007-05-12 17:16  gb
     384
     385        * compiler/X86/X8664/x8664-backend.lisp: Try to make
     386        cross-compilation with x86-64 targets a little easier, by using a
     387        cross-ftd interface directory.
     388
     3892007-05-12 17:16  gb
     390
     391        * compiler/X86/X8664/x8664-arch.lisp: move away from lea/jmp,
     392        towards call/ret.  rcx is now an imm reg; ra0 = temp2.  new tra
     393        scheme.
     394
     3952007-05-12 17:15  gb
     396
     397        * compiler/PPC/ppc2.lisp: %immediate-set-xxx: get sign right.
     398        unwind-protect uses nmkunwind (more happens out-of-line.)
     399
     4002007-05-12 17:14  gb
     401
     402        * compiler/PPC/PPC64/ppc64-vinsns.lisp: nmkunwind.
     403
     4042007-05-12 17:13  gb
     405
     406        * compiler/PPC/PPC32/ppc32-vinsns.lisp: getu32/gets32, nmkunwind.
     407
     4082007-05-12 17:12  gb
     409
     410        * compiler/PPC/ppc-arch.lisp: New .SPnmkunwind.
     411
     4122007-05-12 17:11  gb
     413
     414        * compiler/PPC/: PPC32/ppc32-arch.lisp, PPC64/ppc64-arch.lisp: Make
     415        %GET-KERNEL-GLOBAL-PTR actually work.
     416
    14172007-05-07 19:10  gb
    2418
Note: See TracChangeset for help on using the changeset viewer.