Changeset 5012
- Timestamp:
- Aug 22, 2006, 1:22:24 AM (18 years ago)
- File:
-
- 1 edited
-
trunk/ccl/ChangeLog (modified) (1 diff)
Legend:
- Unmodified
- Added
- Removed
-
trunk/ccl/ChangeLog
r4987 r5012 1 2006-08-22 02:08 gb 2 3 * lisp-kernel/x86_print.c: Try not to use sprintf() (at least not 4 for printing integers), since it seems to need access to pthread 5 data (and calling print_lisp_object() from GDB might fail if %gs is 6 pointing at the lisp TCR.) 7 8 2006-08-22 02:06 gb 9 10 * lisp-kernel/x86-spentry64.s: Unbox %arg_z (boxed entrypoint) 11 earlier in _SPffcall, before entering foreign context. (Need to 12 check PPC version(s) as well.) 13 14 Use DARWIN_GS_HACK to swap %gs memory around when transitioning 15 between foreign code and lisp. 16 17 2006-08-22 02:04 gb 18 19 * lisp-kernel/x86-macros.s: Fix a typo. 20 21 2006-08-22 02:04 gb 22 23 * lisp-kernel/x86-exceptions.h: Define a few more things ... 24 25 2006-08-22 02:03 gb 26 27 * lisp-kernel/x86-exceptions.c: Lots of changes, especially to Mach 28 exception handling code. 29 30 Do the DARWIN_GS_HACK things to overload %gs. 31 32 2006-08-22 02:02 gb 33 34 * lisp-kernel/x86-asmutils64.s: atomic_and(), pseudo_sigreturn() 35 for DarwinX8664. 36 37 2006-08-22 02:01 gb 38 39 * lisp-kernel/thread_manager.c: Some PPC64 conditionalizations 40 should really have tested WORD_SIZE. 41 42 DARWIN_GS_HACK stuff around suspend/resume signal handlers. (May 43 not be necessary.) 44 45 Ensure that tcr.linear gets set on x86, even if not HAVE_TLS. 46 47 2006-08-22 01:59 gb 48 49 * lisp-kernel/pmcl-kernel.c: Don't really flush cache lines on 50 x8664. 51 52 Some PPC64 conditionalizations should really have tested WORD_SIZE. 53 54 2006-08-22 01:58 gb 55 56 * lisp-kernel/m4macros.m4: Define SYSCALL_SETS_CARRY_ON_ERROR for 57 FreeBSDX664, DarwinX8664. 58 59 2006-08-22 01:57 gb 60 61 * lisp-kernel/darwinx8664/Makefile: Much smaller pagezero, much 62 lower seg1addr. 63 64 Something in the (m4-generated) symbolic info in some .s files is 65 screwy and causes ld64 to segfault; it's currently necesssary to 66 use "strip -S" on the assembler-generated .o files. 67 68 2006-08-22 01:55 gb 69 70 * scripts/openmcl64: Untested changes for Darwinx8664. 71 72 2006-08-22 01:54 gb 73 74 * xdump/xx8664-fasload.lisp: Fix typo in darwinx8664 75 xload-backend-info (compiler target is :darwinx8664, not 76 :linuxx8664). 77 78 2006-08-22 01:53 gb 79 80 * lib/nfcomp.lisp: Dump concatenated functions from xfunctions when 81 cross-compiling to x86 from x86. 82 83 2006-08-22 01:52 gb 84 85 * lib/misc.lisp: MACHINE-VERSION: conditionalize on :DARWIN-TARGET 86 (not :DARWINPPC-TARGET). 87 88 2006-08-22 01:50 gb 89 90 * lib/foreign-types.lisp: :DARWIN-TARGET (not :DARWINPPC-TARGET). 91 92 2006-08-22 01:50 gb 93 94 * lib/compile-ccl.lisp: :DARWINX8664 stuff. 95 96 2006-08-22 01:49 gb 97 98 * level-1/version.lisp: Bump. 99 100 2006-08-22 01:49 gb 101 102 * level-1/l1-sysio.lisp: FILE-STREAM-EXTERNAL-FORMAT stuff for 103 BASIC-STREAMs. 104 105 2006-08-22 01:48 gb 106 107 * level-1/: l1-sockets.lisp, linux-files.lisp: Conditionalize on 108 DARWIN-TARGET, not DARWINPPC-TARGET. 109 110 2006-08-22 01:47 gb 111 112 * level-0/l0-init.lisp: DarwinX8664 *features*. 113 114 2006-08-22 01:47 gb 115 116 * level-0/l0-cfm-support.lisp: Some DARWINPPC-TARGET conditionals 117 really want to be DARWIN-TARGET. 118 119 2006-08-22 01:46 gb 120 121 * compiler/X86/X8664/x8664-arch.lisp: YIELD-SYSCALL for 122 DarwinX8664. 123 124 2006-08-22 01:45 gb 125 126 * compiler/X86/x862.lisp: Use CROSS-CREATE-X86-FUNCTION if not 127 compiling natively. 128 129 Fix X862-FIXUP-FWD-REFS if not native function on x86. 130 131 2006-08-22 01:44 gb 132 133 * compiler/X86/x86-lap.lisp: Ensure that CROSS-CREATE-X86-FUNCTION 134 is defined, even on x86. Use it if not compiling natively on x86. 135 1 136 2006-08-20 23:59 gb 2 137
Note:
See TracChangeset
for help on using the changeset viewer.
