2 \# NASM revision history in nasmdoc format
5 \H{cl-2.xx} NASM 2 Series
7 The NASM 2 series support x86-64, and is the production version of NASM
11 \S{cl-2.05} Version 2.05
13 \b Fix redundant REX.W prefix on \c{JMP reg64}.
15 \b Make the behaviour of \c{-O0} match NASM 0.98 legacy behavior.
18 \b \c{-w-user} can be used to suppress the output of \c{%warning} directives.
21 \b Fix bug where \c{ALIGN} would issue a full alignment datum instead of
24 \b Fix offsets in list files.
26 \b Fix \c{%include} inside multi-line macros or loops.
28 \b Fix error where NASM would generate a spurious warning on valid
29 optimizations of immediate values.
31 \b Fix arguments to a number of the \c{CVT} SSE instructions.
33 \b Fix RIP-relative offsets when the instruction carries an immediate.
35 \b Fix the generation of RIP-relative GOT relocations in ELF64.
38 \S{cl-2.04} Version 2.04
40 \b Sanitize macro handing in the \c{%error} directive.
42 \b New \c{%warning} directive to issue user-controlled warnings.
44 \b \c{%error} directives are now deferred to the final assembly phase.
46 \b New \c{%fatal} directive to immediately terminate assembly.
48 \b New \c{%strcat} directive to join quoted strings together.
50 \b New \c{%use} macro directive to support standard macro directives. See
53 \b Excess default parameters to \c{%macro} now issues a warning by default.
56 \b Fix \c{%ifn} and \c{%elifn}.
58 \b Fix nested \c{%else} clauses.
60 \b Correct the handling of nested \c{%rep}s.
62 \b New \c{%unmacro} directive to undeclare a multi-line macro.
65 \b Builtin macro \c{__PASS__} which expands to the current assembly pass.
68 \b \c{__utf16__} and \c{__utf32__} operators to generate UTF-16 and UTF-32
69 strings. See \k{unicode}.
71 \b Fix bug in case-insensitive matching when compiled on platforms that
72 don't use the \c{configure} script. Of the official release binaries,
73 that only affected the OS/2 binary.
75 \b Support for x87 packed BCD constants. See \k{bcdconst}.
77 \b Correct the \c{LTR} and \c{SLDT} instructions in 64-bit mode.
79 \b Fix unnecessary REX.W prefix on indirect jumps in 64-bit mode.
81 \b Add AVX versions of the AES instructions (\c{VAES}...).
83 \b Fix the 256-bit FMA instructions.
85 \b Add 256-bit AVX stores per the latest AVX spec.
87 \b VIA XCRYPT instructions can now be written either with or without
88 \c{REP}, apparently different versions of the VIA spec wrote them
91 \b Add missing 64-bit \c{MOVNTI} instruction.
93 \b Fix the operand size of \c{VMREAD} and \c{VMWRITE}.
95 \b Numerous bug fixes, especially to the AES, AVX and VTX instructions.
97 \b The optimizer now always runs until it converges. It also runs even
98 when disabled, but doesn't optimize. This allows most forward references
99 to be resolved properly.
102 \S{cl-2.03.01} Version 2.03.01
104 \b Fix buffer overflow in the listing module.
106 \b Fix the handling of hexadecimal escape codes in `...` strings.
108 \b The Postscript/PDF documentation has been reformatted.
110 \b The \c{-F} option now implies \c{-g}.
113 \S{cl-2.03} Version 2.03
115 \b Add support for Intel AVX, CLMUL and FMA instructions,
116 including YMM registers.
118 \b \c{dy}, \c{resy} and \c{yword} for 32-byte operands.
120 \b Fix some SSE5 instructions.
122 \b Intel \c{INVEPT}, \c{INVVPID} and \c{MOVBE} instructions.
124 \b Fix checking for critical expressions when the optimizer is enabled.
126 \b Support the DWARF debugging format for ELF targets.
128 \b Fix optimizations of signed bytes.
130 \b Fix operation on bigendian machines.
132 \b Fix buffer overflow in the preprocessor.
134 \b \c{SAFESEH} support for Win32, \c{IMAGEREL} for Win64 (SEH).
136 \b \c{%?} and \c{%??} to refer to the name of a macro itself. In particular,
137 \c{%idefine keyword $%?} can be used to make a keyword "disappear".
139 \b New options for dependency generation: \c{-MD}, \c{-MF},
140 \c{-MP}, \c{-MT}, \c{-MQ}.
142 \b New preprocessor directives \c{%pathsearch} and \c{%depend}; INCBIN
143 reimplemented as a macro.
145 \b \c{%include} now resolves macros in a sane manner.
147 \b \c{%substr} can now be used to get other than one-character substrings.
149 \b New type of character/string constants, using backquotes (\c{`...`}),
150 which support C-style escape sequences.
152 \b \c{%defstr} and \c{%idefstr} to stringize macro definitions before
155 \b Fix forward references used in \c{EQU} statements.
157 \S{cl-2.02} Version 2.02
160 \b Additional fixes for MMX operands with explicit \c{qword}, as well as
161 (hopefully) SSE operands with \c{oword}.
163 \b Fix handling of truncated strings with \c{DO}.
165 \b Fix segfaults due to memory overwrites when floating-point constants
168 \b Fix segfaults due to missing include files.
170 \b Fix OpenWatcom Makefiles for DOS and OS/2.
172 \b Add autogenerated instruction list back into the documentation.
174 \b ELF: Fix segfault when generating stabs, and no symbols have been
177 \b ELF: Experimental support for DWARF debugging information.
179 \b New compile date and time standard macros.
181 \b \c{%ifnum} now returns true for negative numbers.
183 \b New \c{%iftoken} test for a single token.
185 \b New \c{%ifempty} test for empty expansion.
187 \b Add support for the \c{XSAVE} instruction group.
189 \b Makefile for Netware/gcc.
191 \b Fix issue with some warnings getting emitted way too many times.
193 \b Autogenerated instruction list added to the documentation.
195 \S{cl-2.01} Version 2.01
198 \b Fix the handling of MMX registers with explicit \c{qword} tags on
199 memory (broken in 2.00 due to 64-bit changes.)
201 \b Fix the PREFETCH instructions.
203 \b Fix the documentation.
205 \b Fix debugging info when using \c{-f elf}
206 (backwards compatibility alias for \c{-f elf32}).
208 \b Man pages for rdoff tools (from the Debian project.)
210 \b ELF: handle large numbers of sections.
212 \b Fix corrupt output when the optimizer runs out of passes.
215 \S{cl-2.00} Version 2.00
217 \b Added c99 data-type compliance.
219 \b Added general x86-64 support.
221 \b Added win64 (x86-64 COFF) output format.
223 \b Added \c{__BITS__} standard macro.
225 \b Renamed the \c{elf} output format to \c{elf32} for clarity.
227 \b Added \c{elf64} and \c{macho} (MacOS X) output formats.
229 \b Added Numeric constants in \c{dq} directive.
231 \b Added \c{oword}, \c{do} and \c{reso} pseudo operands.
233 \b Allow underscores in numbers.
235 \b Added 8-, 16- and 128-bit floating-point formats.
237 \b Added binary, octal and hexadecimal floating-point.
239 \b Correct the generation of floating-point constants.
241 \b Added floating-point option control.
243 \b Added Infinity and NaN floating point support.
245 \b Added ELF Symbol Visibility support.
247 \b Added setting OSABI value in ELF header directive.
249 \b Added Generate Makefile Dependencies option.
251 \b Added Unlimited Optimization Passes option.
253 \b Added \c{%IFN} and \c{%ELIFN} support.
255 \b Added Logical Negation Operator.
257 \b Enhanced Stack Relative Preprocessor Directives.
259 \b Enhanced ELF Debug Formats.
261 \b Enhanced Send Errors to a File option.
263 \b Added SSSE3, SSE4.1, SSE4.2, SSE5 support.
265 \b Added a large number of additional instructions.
267 \b Significant performance improvements.
270 \H{cl-0.98.xx} NASM 0.98 Series
272 The 0.98 series was the production versions of NASM from 1999 to 2007.
275 \S{cl-0.98.39} Version 0.98.39
277 \b fix buffer overflow
279 \b fix outas86's \c{.bss} handling
281 \b "make spotless" no longer deletes config.h.in.
283 \b \c{%(el)if(n)idn} insensitivity to string quotes difference (#809300).
285 \b (nasm.c)\c{__OUTPUT_FORMAT__} changed to string value instead of symbol.
287 \S{cl-0.98.38} Version 0.98.38
290 \b Add Makefile for 16-bit DOS binaries under OpenWatcom, and modify
291 \c{mkdep.pl} to be able to generate completely pathless dependencies, as
292 required by OpenWatcom wmake (it supports path searches, but not
295 \b Fix the \c{STR} instruction.
297 \b Fix the ELF output format, which was broken under certain
298 circumstances due to the addition of stabs support.
300 \b Quick-fix Borland format debug-info for \c{-f obj}
302 \b Fix for \c{%rep} with no arguments (#560568)
304 \b Fix concatenation of preprocessor function call (#794686)
306 \b Fix long label causes coredump (#677841)
308 \b Use autoheader as well as autoconf to keep configure from generating
309 ridiculously long command lines.
311 \b Make sure that all of the formats which support debugging output
312 actually will suppress debugging output when \c{-g} not specified.
314 \S{cl-0.98.37} Version 0.98.37
317 \b Paths given in \c{-I} switch searched for \c{incbin}-ed as
318 well as \c{%include}-ed files.
320 \b Added stabs debugging for the ELF output format, patch from
323 \b Fix \c{output/outbin.c} to allow origin > 80000000h.
325 \b Make \c{-U} switch work.
327 \b Fix the use of relative offsets with explicit prefixes, e.g.
330 \b Remove \c{backslash()}.
332 \b Fix the \c{SMSW} and \c{SLDT} instructions.
334 \b \c{-O2} and \c{-O3} are no longer aliases for \c{-O10} and \c{-O15}.
335 If you mean the latter, please say so! :)
337 \S{cl-0.98.36} Version 0.98.36
340 \b Update rdoff - librarian/archiver - common rec - docs!
342 \b Fix signed/unsigned problems.
344 \b Fix \c{JMP FAR label} and \c{CALL FAR label}.
346 \b Add new multisection support - map files - fix align bug
348 \b Fix sysexit, movhps/movlps reg,reg bugs in insns.dat
350 \b \c{Q} or \c{O} suffixes indicate octal
352 \b Support Prescott new instructions (PNI).
354 \b Cyrix \c{XSTORE} instruction.
357 \S{cl-0.98.35} Version 0.98.35
359 \b Fix build failure on 16-bit DOS (Makefile.bc3 workaround for compiler bug.)
361 \b Fix dependencies and compiler warnings.
363 \b Add "const" in a number of places.
365 \b Add -X option to specify error reporting format (use -Xvc to
366 integrate with Microsoft Visual Studio.)
368 \b Minor changes for code legibility.
370 \b Drop use of tmpnam() in rdoff (security fix.)
373 \S{cl-0.98.34} Version 0.98.34
375 \b Correct additional address-size vs. operand-size confusions.
377 \b Generate dependencies for all Makefiles automatically.
379 \b Add support for unimplemented (but theoretically available)
380 registers such as tr0 and cr5. Segment registers 6 and 7 are called
381 segr6 and segr7 for the operations which they can be represented.
383 \b Correct some disassembler bugs related to redundant address-size prefixes.
384 Some work still remains in this area.
386 \b Correctly generate an error for things like "SEG eax".
388 \b Add the JMPE instruction, enabled by "CPU IA64".
390 \b Correct compilation on newer gcc/glibc platforms.
392 \b Issue an error on things like "jmp far eax".
395 \S{cl-0.98.33} Version 0.98.33
397 \b New __NASM_PATCHLEVEL__ and __NASM_VERSION_ID__ standard macros to
398 round out the version-query macros. version.pl now understands
399 X.YYplWW or X.YY.ZZplWW as a version number, equivalent to
400 X.YY.ZZ.WW (or X.YY.0.WW, as appropriate).
402 \b New keyword "strict" to disable the optimization of specific
405 \b Fix the handing of size overrides with JMP instructions
406 (instructions such as "jmp dword foo".)
408 \b Fix the handling of "ABSOLUTE label", where "label" points into a
411 \b Fix OBJ output format with lots of externs.
413 \b More documentation updates.
415 \b Add -Ov option to get verbose information about optimizations.
417 \b Undo a braindead change which broke \c{%elif} directives.
422 \S{cl-0.98.32} Version 0.98.32
424 \b Fix NASM crashing when \c{%macro} directives were left unterminated.
426 \b Lots of documentation updates.
428 \b Complete rewrite of the PostScript/PDF documentation generator.
430 \b The MS Visual C++ Makefile was updated and corrected.
432 \b Recognize .rodata as a standard section name in ELF.
434 \b Fix some obsolete Perl4-isms in Perl scripts.
436 \b Fix configure.in to work with autoconf 2.5x.
438 \b Fix a couple of "make cleaner" misses.
440 \b Make the normal "./configure && make" work with Cygwin.
443 \S{cl-0.98.31} Version 0.98.31
445 \b Correctly build in a separate object directory again.
447 \b Derive all references to the version number from the version file.
449 \b New standard macros __NASM_SUBMINOR__ and __NASM_VER__ macros.
451 \b Lots of Makefile updates and bug fixes.
453 \b New \c{%ifmacro} directive to test for multiline macros.
455 \b Documentation updates.
457 \b Fixes for 16-bit OBJ format output.
459 \b Changed the NASM environment variable to NASMENV.
462 \S{cl-0.98.30} Version 0.98.30
464 \b Changed doc files a lot: completely removed old READMExx and
465 Wishlist files, incorporating all information in CHANGES and TODO.
467 \b I waited a long time to rename zoutieee.c to (original) outieee.c
469 \b moved all output modules to output/ subdirectory.
471 \b Added 'make strip' target to strip debug info from nasm & ndisasm.
473 \b Added INSTALL file with installation instructions.
475 \b Added -v option description to nasm man.
477 \b Added dist makefile target to produce source distributions.
479 \b 16-bit support for ELF output format (GNU extension, but useful.)
482 \S{cl-0.98.28} Version 0.98.28
484 \b Fastcooked this for Debian's Woody release:
485 Frank applied the INCBIN bug patch to 0.98.25alt and called
486 it 0.98.28 to not confuse poor little apt-get.
489 \S{cl-0.98.26} Version 0.98.26
491 \b Reorganised files even better from 0.98.25alt
494 \S{cl-0.98.25alt} Version 0.98.25alt
496 \b Prettified the source tree. Moved files to more reasonable places.
498 \b Added findleak.pl script to misc/ directory.
500 \b Attempted to fix doc.
503 \S{cl-0.98.25} Version 0.98.25
505 \b Line continuation character \c{\\}.
507 \b Docs inadvertantly reverted - "dos packaging".
510 \S{cl-0.98.24p1} Version 0.98.24p1
512 \b FIXME: Someone, document this please.
515 \S{cl-0.98.24} Version 0.98.24
517 \b Documentation - Ndisasm doc added to Nasm.doc.
520 \S{cl-0.98.23} Version 0.98.23
522 \b Attempted to remove rdoff version1
524 \b Lino Mastrodomenico's patches to preproc.c (%$$ bug?).
527 \S{cl-0.98.22} Version 0.98.22
529 \b Update rdoff2 - attempt to remove v1.
532 \S{cl-0.98.21} Version 0.98.21
534 \b Optimization fixes.
537 \S{cl-0.98.20} Version 0.98.20
539 \b Optimization fixes.
542 \S{cl-0.98.19} Version 0.98.19
544 \b H. J. Lu's patch back out.
547 \S{cl-0.98.18} Version 0.98.18
549 \b Added ".rdata" to "-f win32".
552 \S{cl-0.98.17} Version 0.98.17
554 \b H. J. Lu's "bogus elf" patch. (Red Hat problem?)
557 \S{cl-0.98.16} Version 0.98.16
559 \b Fix whitespace before "[section ..." bug.
562 \S{cl-0.98.15} Version 0.98.15
564 \b Rdoff changes (?).
566 \b Fix fixes to memory leaks.
569 \S{cl-0.98.14} Version 0.98.14
574 \S{cl-0.98.13} Version 0.98.13
576 \b There was no 0.98.13
579 \S{cl-0.98.12} Version 0.98.12
581 \b Update optimization (new function of "-O1")
583 \b Changes to test/bintest.asm (?).
586 \S{cl-0.98.11} Version 0.98.11
588 \b Optimization changes.
593 \S{cl-0.98.10} Version 0.98.10
595 \b There was no 0.98.10
598 \S{cl-0.98.09} Version 0.98.09
600 \b Add multiple sections support to "-f bin".
602 \b Changed GLOBAL_TEMP_BASE in outelf.c from 6 to 15.
604 \b Add "-v" as an alias to the "-r" switch.
606 \b Remove "#ifdef" from Tasm compatibility options.
608 \b Remove redundant size-overrides on "mov ds, ex", etc.
610 \b Fixes to SSE2, other insns.dat (?).
612 \b Enable uppercase "I" and "P" switches.
614 \b Case insinsitive "seg" and "wrt".
616 \b Update install.sh (?).
618 \b Allocate tokens in blocks.
620 \b Improve "invalid effective address" messages.
623 \S{cl-0.98.08} Version 0.98.08
625 \b Add "\c{%strlen}" and "\c{%substr}" macro operators
627 \b Fixed broken c16.mac.
629 \b Unterminated string error reported.
631 \b Fixed bugs as per 0.98bf
634 \S{cl-0.98.09b with John Coffman patches released 28-Oct-2001} Version 0.98.09b with John Coffman patches released 28-Oct-2001
636 Changes from 0.98.07 release to 98.09b as of 28-Oct-2001
638 \b More closely compatible with 0.98 when -O0 is implied
639 or specified. Not strictly identical, since backward
640 branches in range of short offsets are recognized, and signed
641 byte values with no explicit size specification will be
642 assembled as a single byte.
644 \b More forgiving with the PUSH instruction. 0.98 requires
645 a size to be specified always. 0.98.09b will imply the size
646 from the current BITS setting (16 or 32).
648 \b Changed definition of the optimization flag:
650 -O0 strict two-pass assembly, JMP and Jcc are
651 handled more like 0.98, except that back-
652 ward JMPs are short, if possible.
654 -O1 strict two-pass assembly, but forward
655 branches are assembled with code guaranteed
656 to reach; may produce larger code than
657 -O0, but will produce successful assembly
658 more often if branch offset sizes are not
661 -O2 multi-pass optimization, minimize branch
662 offsets; also will minimize signed immed-
663 iate bytes, overriding size specification.
665 -O3 like -O2, but more passes taken, if needed
668 \S{cl-0.98.07 released 01/28/01} Version 0.98.07 released 01/28/01
670 \b Added Stepane Denis' SSE2 instructions to a *working*
671 version of the code - some earlier versions were based on
672 broken code - sorry 'bout that. version "0.98.07"
678 \b Cosmetic modifications to nasm.c, nasm.h,
682 \S{cl-0.98.06f released 01/18/01} Version 0.98.06f released 01/18/01
685 \b - Add "metalbrain"s jecxz bug fix in insns.dat
686 - alter nasmdoc.src to match - version "0.98.06f"
689 \S{cl-0.98.06e released 01/09/01} Version 0.98.06e released 01/09/01
692 \b Removed the "outforms.h" file - it appears to be
693 someone's old backup of "outform.h". version "0.98.06e"
697 \b fbk - finally added the fix for the "multiple %includes bug",
698 known since 7/27/99 - reported originally (?) and sent to
699 us by Austin Lunnen - he reports that John Fine had a fix
700 within the day. Here it is...
702 \b Nelson Rush resigns from the group. Big thanks to Nelson for
703 his leadership and enthusiasm in getting these changes
704 incorporated into Nasm!
706 \b fbk - [list +], [list -] directives - ineptly implemented, should
707 be re-written or removed, perhaps.
709 \b Brian Raiter / fbk - "elfso bug" fix - applied to aoutb format
710 as well - testing might be desirable...
714 \b James Seter - -postfix, -prefix command line switches.
716 \b Yuri Zaporogets - rdoff utility changes.
719 \S{cl-0.98p1} Version 0.98p1
721 \b GAS-like palign (Panos Minos)
723 \b FIXME: Someone, fill this in with details
726 \S{cl-0.98bf (bug-fixed)} Version 0.98bf (bug-fixed)
728 \b Fixed - elf and aoutb bug - shared libraries
729 - multiple "%include" bug in "-f obj"
731 - unrecognized option bug in ndisasm
733 \S{cl-0.98.03 with John Coffman's changes released 27-Jul-2000} Version 0.98.03 with John Coffman's changes released 27-Jul-2000
735 \b Added signed byte optimizations for the 0x81/0x83 class
736 of instructions: ADC, ADD, AND, CMP, OR, SBB, SUB, XOR:
737 when used as 'ADD reg16,imm' or 'ADD reg32,imm.' Also
738 optimization of signed byte form of 'PUSH imm' and 'IMUL
739 reg,imm'/'IMUL reg,reg,imm.' No size specification is needed.
741 \b Added multi-pass JMP and Jcc offset optimization. Offsets
742 on forward references will preferentially use the short form,
743 without the need to code a specific size (short or near) for
744 the branch. Added instructions for 'Jcc label' to use the
745 form 'Jnotcc $+3/JMP label', in cases where a short offset
746 is out of bounds. If compiling for a 386 or higher CPU, then
747 the 386 form of Jcc will be used instead.
749 This feature is controlled by a new command-line switch: "O",
750 (upper case letter O). "-O0" reverts the assembler to no
751 extra optimization passes, "-O1" allows up to 5 extra passes,
752 and "-O2"(default), allows up to 10 extra optimization passes.
754 \b Added a new directive: 'cpu XXX', where XXX is any of:
755 8086, 186, 286, 386, 486, 586, pentium, 686, PPro, P2, P3 or
756 Katmai. All are case insensitive. All instructions will
757 be selected only if they apply to the selected cpu or lower.
758 Corrected a couple of bugs in cpu-dependence in 'insns.dat'.
760 \b Added to 'standard.mac', the "use16" and "use32" forms of
761 the "bits 16/32" directive. This is nothing new, just conforms
762 to a lot of other assemblers. (minor)
764 \b Changed label allocation from 320/32 (10000 labels @ 200K+)
765 to 32/37 (1000 labels); makes running under DOS much easier.
766 Since additional label space is allocated dynamically, this
767 should have no effect on large programs with lots of labels.
768 The 37 is a prime, believed to be better for hashing. (minor)
771 \S{cl-0.98.03} Version 0.98.03
773 "Integrated patchfile 0.98-0.98.01. I call this version 0.98.03 for
774 historical reasons: 0.98.02 was trashed." --John Coffman
775 <johninsd@san.rr.com>, 27-Jul-2000
777 \b Kendall Bennett's SciTech MGL changes
779 \b Note that you must define "TASM_COMPAT" at compile-time
780 to get the Tasm Ideal Mode compatibility.
782 \b All changes can be compiled in and out using the TASM_COMPAT macros,
783 and when compiled without TASM_COMPAT defined we get the exact same
784 binary as the unmodified 0.98 sources.
786 \b standard.mac, macros.c: Added macros to ignore TASM directives before
789 \b nasm.h: Added extern declaration for tasm_compatible_mode
791 \b nasm.c: Added global variable tasm_compatible_mode
793 \b Added command line switch for TASM compatible mode (-t)
795 \b Changed version command line to reflect when compiled with TASM additions
797 \b Added response file processing to allow all arguments on a single
798 line (response file is @resp rather than -@resp for NASM format).
800 \b labels.c: Changes islocal() macro to support TASM style @@local labels.
802 \b Added islocalchar() macro to support TASM style @@local labels.
804 \b parser.c: Added support for TASM style memory references (ie: mov
805 [DWORD eax],10 rather than the NASM style mov DWORD [eax],10).
807 \b preproc.c: Added new directives, \c{%arg}, \c{%local}, \c{%stacksize} to directives
810 \b Added support for TASM style directives without a leading % symbol.
812 \b Integrated a block of changes from Andrew Zabolotny <bit@eltech.ru>:
814 \b A new keyword \c{%xdefine} and its case-insensitive counterpart \c{%ixdefine}.
815 They work almost the same way as \c{%define} and \c{%idefine} but expand
816 the definition immediately, not on the invocation. Something like a cross
817 between \c{%define} and \c{%assign}. The "x" suffix stands for "eXpand", so
818 "xdefine" can be deciphered as "expand-and-define". Thus you can do
824 \c %xdefine %1 dword [esp+ofs]
828 \b Changed the place where the expansion of %$name macros are expanded.
829 Now they are converted into ..@ctxnum.name form when detokenizing, so
830 there are no quirks as before when using %$name arguments to macros,
831 in macros etc. For example:
840 Now last line will be expanded into "hello" as expected. This also allows
841 for lots of goodies, a good example are extended "proc" macros included
844 \b Added a check for "cstk" in smacro_defined() before calling get_ctx() -
845 this allows for things like:
850 to work without warnings even in no context.
852 \b Added a check for "cstk" in %if*ctx and %elif*ctx directives -
853 this allows to use \c{%ifctx} without excessive warnings. If there is
854 no active context, \c{%ifctx} goes through "false" branch.
856 \b Removed "user error: " prefix with \c{%error} directive: it just clobbers the
857 output and has absolutely no functionality. Besides, this allows to write
858 macros that does not differ from built-in functions in any way.
860 \b Added expansion of string that is output by \c{%error} directive. Now you
863 \c %define hello(x) Hello, x!
865 \c %define %$name andy
866 \c %error "hello(%$name)"
868 Same happened with \c{%include} directive.
870 \b Now all directives that expect an identifier will try to expand and
871 concatenate everything without whitespaces in between before usage.
872 For example, with "unfixed" nasm the commands
874 \c %define %$abc hello
875 \c %define __%$abc goodbye
878 would produce "incorrect" output: last line will expand to
880 \c hello goodbyehello
882 Not quite what you expected, eh? :-) The answer is that preprocessor
883 treats the \c{%define} construct as if it would be
885 \c %define __ %$abc goodbye
887 (note the white space between __ and %$abc). After my "fix" it
888 will "correctly" expand into
892 as expected. Note that I use quotes around words "correct", "incorrect"
893 etc because this is rather a feature not a bug; however current behaviour
894 is more logical (and allows more advanced macro usage :-).
896 Same change was applied to:
897 \c{%push},\c{%macro},\c{%imacro},\c{%define},\c{%idefine},\c{%xdefine},\c{%ixdefine},
898 \c{%assign},\c{%iassign},\c{%undef}
900 \b A new directive [WARNING {+|-}warning-id] have been added. It works only
901 if the assembly phase is enabled (i.e. it doesn't work with nasm -e).
903 \b A new warning type: macro-selfref. By default this warning is disabled;
904 when enabled NASM warns when a macro self-references itself; for example
905 the following source:
907 \c [WARNING macro-selfref]
918 will produce a warning, but if we remove the first line we won't see it
919 anymore (which is The Right Thing To Do {tm} IMHO since C preprocessor
920 eats such constructs without warnings at all).
922 \b Added a "error" routine to preprocessor which always will set ERR_PASS1
923 bit in severity_code. This removes annoying repeated errors on first
924 and second passes from preprocessor.
926 \b Added the %+ operator in single-line macros for concatenating two
927 identifiers. Usage example:
929 \c %define _myfunc _otherfunc
930 \c %define cextern(x) _ %+ x
933 After first expansion, third line will become "_myfunc". After this
934 expansion is performed again so it becomes "_otherunc".
936 \b Now if preprocessor is in a non-emitting state, no warning or error
937 will be emitted. Example:
942 \c put anything you want between these two brackets,
943 \c even macro-parameter references %1 or local
944 \c labels %$zz or macro-local labels %%zz - no
945 \c warning will be emitted.
948 \b Context-local variables on expansion as a last resort are looked up
949 in outer contexts. For example, the following piece:
959 will expand correctly the fourth line to [esp]; if we'll define another
960 %$a inside the "inner" context, it will take precedence over outer
961 definition. However, this modification has been applied only to
962 expand_smacro and not to smacro_define: as a consequence expansion
963 looks in outer contexts, but \c{%ifdef} won't look in outer contexts.
965 This behaviour is needed because we don't want nested contexts to
966 act on already defined local macros. Example:
968 \c %define %$arg1 [esp+4]
974 In this example the "if" mmacro enters into the "if" context, so %$arg1
975 is not valid anymore inside "if". Of course it could be worked around
976 by using explicitely %$$arg1 but this is ugly IMHO.
978 \b Fixed memory leak in \c{%undef}. The origline wasn't freed before
981 \b Fixed trap in preprocessor when line expanded to empty set of tokens.
982 This happens, for example, in the following case:
988 \S{cl-0.98} Version 0.98
990 All changes since NASM 0.98p3 have been produced by H. Peter Anvin <hpa@zytor.com>.
992 \b The documentation comment delimiter is \# not #.
994 \b Allow EQU definitions to refer to external labels; reported by
997 \b Re-enable support for RDOFF v1; reported by Pedro Gimeno.
999 \b Updated License file per OK from Simon and Julian.
1002 \S{cl-0.98p9} Version 0.98p9
1004 \b Update documentation (although the instruction set reference will
1005 have to wait; I don't want to hold up the 0.98 release for it.)
1007 \b Verified that the NASM implementation of the PEXTRW and PMOVMSKB
1008 instructions is correct. The encoding differs from what the Intel
1009 manuals document, but the Pentium III behaviour matches NASM, not
1012 \b Fix handling of implicit sizes in PSHUFW and PINSRW, reported by
1015 \b Resurrect the -s option, which was removed when changing the
1016 diagnostic output to stdout.
1019 \S{cl-0.98p8} Version 0.98p8
1021 \b Fix for "DB" when NASM is running on a bigendian machine.
1023 \b Invoke insns.pl once for each output script, making Makefile.in
1024 legal for "make -j".
1026 \b Improve the Unix configure-based makefiles to make package
1029 \b Included an RPM .spec file for building RPM (RedHat Package Manager)
1030 packages on Linux or Unix systems.
1032 \b Fix Makefile dependency problems.
1034 \b Change src/rdsrc.pl to include sectioning information in info
1035 output; required for install-info to work.
1037 \b Updated the RDOFF distribution to version 2 from Jules; minor
1038 massaging to make it compile in my environment.
1040 \b Split doc files that can be built by anyone with a Perl interpreter off
1041 into a separate archive.
1043 \b "Dress rehearsal" release!
1046 \S{cl-0.98p7} Version 0.98p7
1048 \b Fixed opcodes with a third byte-sized immediate argument to not
1049 complain if given "byte" on the immediate.
1051 \b Allow \c{%undef} to remove single-line macros with arguments. This
1052 matches the behaviour of #undef in the C preprocessor.
1054 \b Allow -d, -u, -i and -p to be specified as -D, -U, -I and -P for
1055 compatibility with most C compilers and preprocessors. This allows
1056 Makefile options to be shared between cc and nasm, for example.
1060 \b Went through the list of Katmai instructions and hopefully fixed the
1061 (rather few) mistakes in it.
1063 \b (Hopefully) fixed a number of disassembler bugs related to ambiguous
1064 instructions (disambiguated by -p) and SSE instructions with REP.
1066 \b Fix for bug reported by Mark Junger: "call dword 0x12345678" should
1067 work and may add an OSP (affected CALL, JMP, Jcc).
1069 \b Fix for environments when "stderr" isn't a compile-time constant.
1072 \S{cl-0.98p6} Version 0.98p6
1075 \b Took officially over coordination of the 0.98 release; so drop
1076 the p3.x notation. Skipped p4 and p5 to avoid confusion with John
1077 Fine's J4 and J5 releases.
1079 \b Update the documentation; however, it still doesn't include
1080 documentation for the various new instructions. I somehow wonder if
1081 it makes sense to have an instruction set reference in the assembler
1082 manual when Intel et al have PDF versions of their manuals online.
1084 \b Recognize "idt" or "centaur" for the -p option to ndisasm.
1086 \b Changed error messages back to stderr where they belong, but add an
1087 -E option to redirect them elsewhere (the DOS shell cannot redirect
1090 \b -M option to generate Makefile dependencies (based on code from Alex
1093 \b \c{%undef} preprocessor directive, and -u option, that undefines a
1096 \b OS/2 Makefile (Mkfiles/Makefile.os2) for Borland under OS/2; from
1099 \b Various minor bugfixes (reported by):
1100 - Dangling \c{%s} in preproc.c (Martin Junker)
1102 \b THERE ARE KNOWN BUGS IN SSE AND THE OTHER KATMAI INSTRUCTIONS. I am
1103 on a trip and didn't bring the Katmai instruction reference, so I
1104 can't work on them right now.
1106 \b Updated the License file per agreement with Simon and Jules to
1107 include a GPL distribution clause.
1110 \S{cl-0.98p3.7} Version 0.98p3.7
1112 \b (Hopefully) fixed the canned Makefiles to include the outrdf2 and
1115 \b Renamed changes.asm to changed.asm.
1118 \S{cl-0.98p3.6} Version 0.98p3.6
1120 \b Fixed a bunch of instructions that were added in 0.98p3.5 which had
1121 memory operands, and the address-size prefix was missing from the
1122 instruction pattern.
1125 \S{cl-0.98p3.5} Version 0.98p3.5
1127 \b Merged in changes from John S. Fine's 0.98-J5 release. John's based
1128 0.98-J5 on my 0.98p3.3 release; this merges the changes.
1130 \b Expanded the instructions flag field to a long so we can fit more
1131 flags; mark SSE (KNI) and AMD or Katmai-specific instructions as
1134 \b Fix the "PRIV" flag on a bunch of instructions, and create new
1135 "PROT" flag for protected-mode-only instructions (orthogonal to if
1136 the instruction is privileged!) and new "SMM" flag for SMM-only
1139 \b Added AMD-only SYSCALL and SYSRET instructions.
1141 \b Make SSE actually work, and add new Katmai MMX instructions.
1143 \b Added a -p (preferred vendor) option to ndisasm so that it can
1144 distinguish e.g. Cyrix opcodes also used in SSE. For example:
1146 \c ndisasm -p cyrix aliased.bin
1147 \c 00000000 670F514310 paddsiw mm0,[ebx+0x10]
1148 \c 00000005 670F514320 paddsiw mm0,[ebx+0x20]
1149 \c ndisasm -p intel aliased.bin
1150 \c 00000000 670F514310 sqrtps xmm0,[ebx+0x10]
1151 \c 00000005 670F514320 sqrtps xmm0,[ebx+0x20]
1153 \b Added a bunch of Cyrix-specific instructions.
1156 \S{cl-0.98p3.4} Version 0.98p3.4
1158 \b Made at least an attempt to modify all the additional Makefiles (in
1159 the Mkfiles directory). I can't test it, but this was the best I
1162 \b DOS DJGPP+"Opus Make" Makefile from John S. Fine.
1164 \b changes.asm changes from John S. Fine.
1167 \S{cl-0.98p3.3} Version 0.98p3.3
1169 \b Patch from Conan Brink to allow nesting of \c{%rep} directives.
1171 \b If we're going to allow INT01 as an alias for INT1/ICEBP (one of
1172 Jules 0.98p3 changes), then we should allow INT03 as an alias for INT3
1175 \b Updated changes.asm to include the latest changes.
1177 \b Tried to clean up the <CR>s that had snuck in from a DOS/Windows
1178 environment into my Unix environment, and try to make sure than
1179 DOS/Windows users get them back.
1181 \b We would silently generate broken tools if insns.dat wasn't sorted
1182 properly. Change insns.pl so that the order doesn't matter.
1184 \b Fix bug in insns.pl (introduced by me) which would cause conditional
1185 instructions to have an extra "cc" in disassembly, e.g. "jnz"
1186 disassembled as "jccnz".
1189 \S{cl-0.98p3.2} Version 0.98p3.2
1191 \b Merged in John S. Fine's changes from his 0.98-J4 prerelease; see
1192 http://www.csoft.net/cz/johnfine/
1194 \b Changed previous "spotless" Makefile target (appropriate for distribution)
1195 to "distclean", and added "cleaner" target which is same as "clean"
1196 except deletes files generated by Perl scripts; "spotless" is union.
1198 \b Removed BASIC programs from distribution. Get a Perl interpreter
1199 instead (see below.)
1201 \b Calling this "pre-release 3.2" rather than "p3-hpa2" because of
1202 John's contributions.
1204 \b Actually link in the IEEE output format (zoutieee.c); fix a bunch of
1205 compiler warnings in that file. Note I don't know what IEEE output
1206 is supposed to look like, so these changes were made "blind".
1209 \S{cl-0.98p3-hpa} Version 0.98p3-hpa
1211 \b Merged nasm098p3.zip with nasm-0.97.tar.gz to create a fully
1212 buildable version for Unix systems (Makefile.in updates, etc.)
1214 \b Changed insns.pl to create the instruction tables in nasm.h and
1215 names.c, so that a new instruction can be added by adding it *only*
1218 \b Added the following new instructions: SYSENTER, SYSEXIT, FXSAVE,
1219 FXRSTOR, UD1, UD2 (the latter two are two opcodes that Intel
1220 guarantee will never be used; one of them is documented as UD2 in
1221 Intel documentation, the other one just as "Undefined Opcode" --
1222 calling it UD1 seemed to make sense.)
1224 \b MAX_SYMBOL was defined to be 9, but LOADALL286 and LOADALL386 are 10
1225 characters long. Now MAX_SYMBOL is derived from insns.dat.
1227 \b A note on the BASIC programs included: forget them. insns.bas is
1228 already out of date. Get yourself a Perl interpreter for your
1229 platform of choice at
1230 \W{http://www.cpan.org/ports/index.html}{http://www.cpan.org/ports/index.html}.
1233 \S{cl-0.98p3} Version 0.98 pre-release 3
1235 \b added response file support, improved command line handling, new layout
1238 \b fixed limit checking bug, 'OUT byte nn, reg' bug, and a couple of rdoff
1239 related bugs, updated Wishlist; 0.98 Prerelease 3.
1242 \S{cl-0.98p2} Version 0.98 pre-release 2
1244 \b fixed bug in outcoff.c to do with truncating section names longer
1245 than 8 characters, referencing beyond end of string; 0.98 pre-release 2
1248 \S{cl-0.98p1} Version 0.98 pre-release 1
1250 \b Fixed a bug whereby STRUC didn't work at all in RDF.
1252 \b Fixed a problem with group specification in PUBDEFs in OBJ.
1254 \b Improved ease of adding new output formats. Contribution due to
1257 \b Fixed a bug in relocations in the `bin' format: was showing up when
1258 a relocatable reference crossed an 8192-byte boundary in any output
1261 \b Fixed a bug in local labels: local-label lookups were inconsistent
1262 between passes one and two if an EQU occurred between the definition
1263 of a global label and the subsequent use of a local label local to
1266 \b Fixed a seg-fault in the preprocessor (again) which happened when
1267 you use a blank line as the first line of a multi-line macro
1268 definition and then defined a label on the same line as a call to
1271 \b Fixed a stale-pointer bug in the handling of the NASM environment
1272 variable. Thanks to Thomas McWilliams.
1274 \b ELF had a hard limit on the number of sections which caused
1275 segfaults when transgressed. Fixed.
1277 \b Added ability for ndisasm to read from stdin by using `-' as the
1280 \b ndisasm wasn't outputting the TO keyword. Fixed.
1282 \b Fixed error cascade on bogus expression in \c{%if} - an error in
1283 evaluation was causing the entire \c{%if} to be discarded, thus creating
1284 trouble later when the \c{%else} or \c{%endif} was encountered.
1286 \b Forward reference tracking was instruction-granular not operand-
1287 granular, which was causing 286-specific code to be generated
1288 needlessly on code of the form `shr word [forwardref],1'. Thanks to
1289 Jim Hague for sending a patch.
1291 \b All messages now appear on stdout, as sending them to stderr serves
1292 no useful purpose other than to make redirection difficult.
1294 \b Fixed the problem with EQUs pointing to an external symbol - this
1295 now generates an error message.
1297 \b Allowed multiple size prefixes to an operand, of which only the first
1298 is taken into account.
1300 \b Incorporated John Fine's changes, including fixes of a large number
1301 of preprocessor bugs, some small problems in OBJ, and a reworking of
1302 label handling to define labels before their line is assembled, rather
1305 \b Reformatted a lot of the source code to be more readable. Included
1306 'coding.txt' as a guideline for how to format code for contributors.
1308 \b Stopped nested \c{%reps} causing a panic - they now cause a slightly more
1309 friendly error message instead.
1311 \b Fixed floating point constant problems (patch by Pedro Gimeno)
1313 \b Fixed the return value of insn_size() not being checked for -1, indicating
1316 \b Incorporated 3Dnow! instructions.
1318 \b Fixed the 'mov eax, eax + ebx' bug.
1320 \b Fixed the GLOBAL EQU bug in ELF. Released developers release 3.
1322 \b Incorporated John Fine's command line parsing changes
1324 \b Incorporated David Lindauer's OMF debug support
1326 \b Made changes for LCC 4.0 support (\c{__NASM_CDecl__}, removed register size
1327 specification warning when sizes agree).
1330 \H{cl-0.9x} NASM 0.9 Series
1332 Revisions before 0.98.
1335 \S{cl-0.97} Version 0.97 released December 1997
1337 \b This was entirely a bug-fix release to 0.96, which seems to have got
1340 \b Fixed stupid mistake in OBJ which caused `MOV EAX,<constant>' to
1341 fail. Caused by an error in the `MOV EAX,<segment>' support.
1343 \b ndisasm hung at EOF when compiled with lcc on Linux because lcc on
1344 Linux somehow breaks feof(). ndisasm now does not rely on feof().
1346 \b A heading in the documentation was missing due to a markup error in
1347 the indexing. Fixed.
1349 \b Fixed failure to update all pointers on realloc() within extended-
1350 operand code in parser.c. Was causing wrong behaviour and seg faults
1351 on lines such as `dd 0.0,0.0,0.0,0.0,...'
1353 \b Fixed a subtle preprocessor bug whereby invoking one multi-line
1354 macro on the first line of the expansion of another, when the second
1355 had been invoked with a label defined before it, didn't expand the
1358 \b Added internal.doc back in to the distribution archives - it was
1359 missing in 0.96 *blush*
1361 \b Fixed bug causing 0.96 to be unable to assemble its own test files,
1362 specifically objtest.asm. *blush again*
1364 \b Fixed seg-faults and bogus error messages caused by mismatching
1365 \c{%rep} and \c{%endrep} within multi-line macro definitions.
1367 \b Fixed a problem with buffer overrun in OBJ, which was causing
1368 corruption at ends of long PUBDEF records.
1370 \b Separated DOS archives into main-program and documentation to reduce
1374 \S{cl-0.96} Version 0.96 released November 1997
1376 \b Fixed a bug whereby, if `nasm sourcefile' would cause a filename
1377 collision warning and put output into `nasm.out', then `nasm
1378 sourcefile -o outputfile' still gave the warning even though the
1380 Fixed name pollution under Digital UNIX: one of its header files
1381 defined R_SP, which broke the enum in nasm.h.
1383 \b Fixed minor instruction table problems: FUCOM and FUCOMP didn't have
1384 two-operand forms; NDISASM didn't recognise the longer register
1385 forms of PUSH and POP (eg FF F3 for PUSH BX); TEST mem,imm32 was
1386 flagged as undocumented; the 32-bit forms of CMOV had 16-bit operand
1387 size prefixes; `AAD imm' and `AAM imm' are no longer flagged as
1388 undocumented because the Intel Architecture reference documents
1391 \b Fixed a problem with the local-label mechanism, whereby strange
1392 types of symbol (EQUs, auto-defined OBJ segment base symbols)
1393 interfered with the `previous global label' value and screwed up
1396 \b Fixed a bug whereby the stub preprocessor didn't communicate with
1397 the listing file generator, so that the -a and -l options in
1398 conjunction would produce a useless listing file.
1400 \b Merged `os2' object file format back into `obj', after discovering
1401 that `obj' _also_ shouldn't have a link pass separator in a module
1402 containing a non-trivial MODEND. Flat segments are now declared
1403 using the FLAT attribute. `os2' is no longer a valid object format
1406 \b Removed the fixed-size temporary storage in the evaluator. Very very
1407 long expressions (like `mov ax,1+1+1+1+...' for two hundred 1s or
1408 so) should now no longer crash NASM.
1410 \b Fixed a bug involving segfaults on disassembly of MMX instructions,
1411 by changing the meaning of one of the operand-type flags in nasm.h.
1412 This may cause other apparently unrelated MMX problems; it needs to
1413 be tested thoroughly.
1415 \b Fixed some buffer overrun problems with large OBJ output files.
1416 Thanks to DJ Delorie for the bug report and fix.
1418 \b Made preprocess-only mode actually listen to the \c{%line} markers as it
1419 prints them, so that it can report errors more sanely.
1421 \b Re-designed the evaluator to keep more sensible track of expressions
1422 involving forward references: can now cope with previously-nightmare
1429 \b Added the ALIGN and ALIGNB standard macros.
1431 \b Added PIC support in ELF: use of WRT to obtain the four extra
1432 relocation types needed.
1434 \b Added the ability for output file formats to define their own
1435 extensions to the GLOBAL, COMMON and EXTERN directives.
1437 \b Implemented common-variable alignment, and global-symbol type and
1438 size declarations, in ELF.
1440 \b Implemented NEAR and FAR keywords for common variables, plus
1441 far-common element size specification, in OBJ.
1443 \b Added a feature whereby EXTERNs and COMMONs in OBJ can be given a
1444 default WRT specification (either a segment or a group).
1446 \b Transformed the Unix NASM archive into an auto-configuring package.
1448 \b Added a sanity-check for people applying SEG to things which are
1449 already segment bases: this previously went unnoticed by the SEG
1450 processing and caused OBJ-driver panics later.
1452 \b Added the ability, in OBJ format, to deal with `MOV EAX,<segment>'
1453 type references: OBJ doesn't directly support dword-size segment
1454 base fixups, but as long as the low two bytes of the constant term
1455 are zero, a word-size fixup can be generated instead and it will
1458 \b Added the ability to specify sections' alignment requirements in
1459 Win32 object files and pure binary files.
1461 \b Added preprocess-time expression evaluation: the \c{%assign} (and
1462 \c{%iassign}) directive and the bare \c{%if} (and \c{%elif}) conditional. Added
1463 relational operators to the evaluator, for use only in \c{%if}
1464 constructs: the standard relationals = < > <= >= <> (and C-like
1465 synonyms == and !=) plus low-precedence logical operators &&, ^^ and
1468 \b Added a preprocessor repeat construct: \c{%rep} / \c{%exitrep} / \c{%endrep}.
1470 \b Added the __FILE__ and __LINE__ standard macros.
1472 \b Added a sanity check for number constants being greater than
1473 0xFFFFFFFF. The warning can be disabled.
1475 \b Added the %0 token whereby a variadic multi-line macro can tell how
1476 many parameters it's been given in a specific invocation.
1478 \b Added \c{%rotate}, allowing multi-line macro parameters to be cycled.
1480 \b Added the `*' option for the maximum parameter count on multi-line
1481 macros, allowing them to take arbitrarily many parameters.
1483 \b Added the ability for the user-level forms of EXTERN, GLOBAL and
1484 COMMON to take more than one argument.
1486 \b Added the IMPORT and EXPORT directives in OBJ format, to deal with
1489 \b Added some more preprocessor \c{%if} constructs: \c{%ifidn} / \c{%ifidni} (exact
1490 textual identity), and \c{%ifid} / \c{%ifnum} / \c{%ifstr} (token type testing).
1492 \b Added the ability to distinguish SHL AX,1 (the 8086 version) from
1493 SHL AX,BYTE 1 (the 286-and-upwards version whose constant happens to
1496 \b Added NetBSD/FreeBSD/OpenBSD's variant of a.out format, complete
1497 with PIC shared library features.
1499 \b Changed NASM's idiosyncratic handling of FCLEX, FDISI, FENI, FINIT,
1500 FSAVE, FSTCW, FSTENV, and FSTSW to bring it into line with the
1501 otherwise accepted standard. The previous behaviour, though it was a
1502 deliberate feature, was a deliberate feature based on a
1503 misunderstanding. Apologies for the inconvenience.
1505 \b Improved the flexibility of ABSOLUTE: you can now give it an
1506 expression rather than being restricted to a constant, and it can
1507 take relocatable arguments as well.
1509 \b Added the ability for a variable to be declared as EXTERN multiple
1510 times, and the subsequent definitions are just ignored.
1512 \b We now allow instruction prefixes (CS, DS, LOCK, REPZ etc) to be
1513 alone on a line (without a following instruction).
1515 \b Improved sanity checks on whether the arguments to EXTERN, GLOBAL
1516 and COMMON are valid identifiers.
1518 \b Added misc/exebin.mac to allow direct generation of .EXE files by
1519 hacking up an EXE header using DB and DW; also added test/binexe.asm
1520 to demonstrate the use of this. Thanks to Yann Guidon for
1521 contributing the EXE header code.
1523 \b ndisasm forgot to check whether the input file had been successfully
1524 opened. Now it does. Doh!
1526 \b Added the Cyrix extensions to the MMX instruction set.
1528 \b Added a hinting mechanism to allow [EAX+EBX] and [EBX+EAX] to be
1529 assembled differently. This is important since [ESI+EBP] and
1530 [EBP+ESI] have different default base segment registers.
1532 \b Added support for the PharLap OMF extension for 4096-byte segment
1536 \S{cl-0.95 released July 1997} Version 0.95 released July 1997
1538 \b Fixed yet another ELF bug. This one manifested if the user relied on
1539 the default segment, and attempted to define global symbols without
1540 first explicitly declaring the target segment.
1542 \b Added makefiles (for NASM and the RDF tools) to build Win32 console
1543 apps under Symantec C++. Donated by Mark Junker.
1545 \b Added `macros.bas' and `insns.bas', QBasic versions of the Perl
1546 scripts that convert `standard.mac' to `macros.c' and convert
1547 `insns.dat' to `insnsa.c' and `insnsd.c'. Also thanks to Mark
1550 \b Changed the diassembled forms of the conditional instructions so
1551 that JB is now emitted as JC, and other similar changes. Suggested
1552 list by Ulrich Doewich.
1554 \b Added `@' to the list of valid characters to begin an identifier
1557 \b Documentary changes, notably the addition of the `Common Problems'
1558 section in nasm.doc.
1560 \b Fixed a bug relating to 32-bit PC-relative fixups in OBJ.
1562 \b Fixed a bug in perm_copy() in labels.c which was causing exceptions
1563 in cleanup_labels() on some systems.
1565 \b Positivity sanity check in TIMES argument changed from a warning to
1566 an error following a further complaint.
1568 \b Changed the acceptable limits on byte and word operands to allow
1569 things like `~10111001b' to work.
1571 \b Fixed a major problem in the preprocessor which caused seg-faults if
1572 macro definitions contained blank lines or comment-only lines.
1574 \b Fixed inadequate error checking on the commas separating the
1575 arguments to `db', `dw' etc.
1577 \b Fixed a crippling bug in the handling of macros with operand counts
1578 defined with a `+' modifier.
1580 \b Fixed a bug whereby object file formats which stored the input file
1581 name in the output file (such as OBJ and COFF) weren't doing so
1582 correctly when the output file name was specified on the command
1585 \b Removed [INC] and [INCLUDE] support for good, since they were
1588 \b Fixed a bug in OBJ which caused all fixups to be output in 16-bit
1589 (old-format) FIXUPP records, rather than putting the 32-bit ones in
1590 FIXUPP32 (new-format) records.
1592 \b Added, tentatively, OS/2 object file support (as a minor variant on
1595 \b Updates to Fox Cutter's Borland C makefile, Makefile.bc2.
1597 \b Removed a spurious second fclose() on the output file.
1599 \b Added the `-s' command line option to redirect all messages which
1600 would go to stderr (errors, help text) to stdout instead.
1602 \b Added the `-w' command line option to selectively suppress some
1603 classes of assembly warning messages.
1605 \b Added the `-p' pre-include and `-d' pre-define command-line options.
1607 \b Added an include file search path: the `-i' command line option.
1609 \b Fixed a silly little preprocessor bug whereby starting a line with a
1610 `%!' environment-variable reference caused an `unknown directive'
1613 \b Added the long-awaited listing file support: the `-l' command line
1616 \b Fixed a problem with OBJ format whereby, in the absence of any
1617 explicit segment definition, non-global symbols declared in the
1618 implicit default segment generated spurious EXTDEF records in the
1621 \b Added the NASM environment variable.
1623 \b From this version forward, Win32 console-mode binaries will be
1624 included in the DOS distribution in addition to the 16-bit binaries.
1625 Added Makefile.vc for this purpose.
1627 \b Added `return 0;' to test/objlink.c to prevent compiler warnings.
1629 \b Added the __NASM_MAJOR__ and __NASM_MINOR__ standard defines.
1631 \b Added an alternative memory-reference syntax in which prefixing an
1632 operand with `&' is equivalent to enclosing it in square brackets,
1633 at the request of Fox Cutter.
1635 \b Errors in pass two now cause the program to return a non-zero error
1636 code, which they didn't before.
1638 \b Fixed the single-line macro cycle detection, which didn't work at
1639 all on macros with no parameters (caused an infinite loop). Also
1640 changed the behaviour of single-line macro cycle detection to work
1641 like cpp, so that macros like `extrn' as given in the documentation
1644 \b Fixed the implementation of WRT, which was too restrictive in that
1645 you couldn't do `mov ax,[di+abc wrt dgroup]' because (di+abc) wasn't
1646 a relocatable reference.
1649 \S{cl-0.94 released April 1997} Version 0.94 released April 1997
1652 \b Major item: added the macro processor.
1654 \b Added undocumented instructions SMI, IBTS, XBTS and LOADALL286. Also
1655 reorganised CMPXCHG instruction into early-486 and Pentium forms.
1656 Thanks to Thobias Jones for the information.
1658 \b Fixed two more stupid bugs in ELF, which were causing `ld' to
1659 continue to seg-fault in a lot of non-trivial cases.
1661 \b Fixed a seg-fault in the label manager.
1663 \b Stopped FBLD and FBSTP from _requiring_ the TWORD keyword, which is
1664 the only option for BCD loads/stores in any case.
1666 \b Ensured FLDCW, FSTCW and FSTSW can cope with the WORD keyword, if
1667 anyone bothers to provide it. Previously they complained unless no
1668 keyword at all was present.
1670 \b Some forms of FDIV/FDIVR and FSUB/FSUBR were still inverted: a
1671 vestige of a bug that I thought had been fixed in 0.92. This was
1672 fixed, hopefully for good this time...
1674 \b Another minor phase error (insofar as a phase error can _ever_ be
1675 minor) fixed, this one occurring in code of the form
1677 \c rol ax,forward_reference
1678 \c forward_reference equ 1
1680 \b The number supplied to TIMES is now sanity-checked for positivity,
1681 and also may be greater than 64K (which previously didn't work on
1684 \b Added Watcom C makefiles, and misc/pmw.bat, donated by Dominik Behr.
1686 \b Added the INCBIN pseudo-opcode.
1688 \b Due to the advent of the preprocessor, the [INCLUDE] and [INC]
1689 directives have become obsolete. They are still supported in this
1690 version, with a warning, but won't be in the next.
1692 \b Fixed a bug in OBJ format, which caused incorrect object records to
1693 be output when absolute labels were made global.
1695 \b Updates to RDOFF subdirectory, and changes to outrdf.c.
1698 \S{cl-0.93 released January 1997} Version 0.93 released January 1997
1700 This release went out in a great hurry after semi-crippling bugs
1703 \b Really \e{did} fix the stack overflows this time. *blush*
1705 \b Had problems with EA instruction sizes changing between passes, when
1706 an offset contained a forward reference and so 4 bytes were
1707 allocated for the offset in pass one; by pass two the symbol had
1708 been defined and happened to be a small absolute value, so only 1
1709 byte got allocated, causing instruction size mismatch between passes
1710 and hence incorrect address calculations. Fixed.
1712 \b Stupid bug in the revised ELF section generation fixed (associated
1713 string-table section for .symtab was hard-coded as 7, even when this
1714 didn't fit with the real section table). Was causing `ld' to
1715 seg-fault under Linux.
1717 \b Included a new Borland C makefile, Makefile.bc2, donated by Fox
1718 Cutter <lmb@comtch.iea.com>.
1721 \S{cl-0.92 released January 1997} Version 0.92 released January 1997
1723 \b The FDIVP/FDIVRP and FSUBP/FSUBRP pairs had been inverted: this was
1724 fixed. This also affected the LCC driver.
1726 \b Fixed a bug regarding 32-bit effective addresses of the form
1727 \c{[other_register+ESP]}.
1729 \b Documentary changes, notably documentation of the fact that Borland
1730 Win32 compilers use `obj' rather than `win32' object format.
1732 \b Fixed the COMENT record in OBJ files, which was formatted
1735 \b Fixed a bug causing segfaults in large RDF files.
1737 \b OBJ format now strips initial periods from segment and group
1738 definitions, in order to avoid complications with the local label
1741 \b Fixed a bug in disassembling far calls and jumps in NDISASM.
1743 \b Added support for user-defined sections in COFF and ELF files.
1745 \b Compiled the DOS binaries with a sensible amount of stack, to
1746 prevent stack overflows on any arithmetic expression containing
1749 \b Fixed a bug in handling of files that do not terminate in a newline.
1752 \S{cl-0.91 released November 1996} Version 0.91 released November 1996
1754 \b Loads of bug fixes.
1756 \b Support for RDF added.
1758 \b Support for DBG debugging format added.
1760 \b Support for 32-bit extensions to Microsoft OBJ format added.
1762 \b Revised for Borland C: some variable names changed, makefile added.
1764 \b LCC support revised to actually work.
1766 \b JMP/CALL NEAR/FAR notation added.
1768 \b `a16', `o16', `a32' and `o32' prefixes added.
1770 \b Range checking on short jumps implemented.
1772 \b MMX instruction support added.
1774 \b Negative floating point constant support added.
1776 \b Memory handling improved to bypass 64K barrier under DOS.
1778 \b \c{$} prefix to force treatment of reserved words as identifiers added.
1780 \b Default-size mechanism for object formats added.
1782 \b Compile-time configurability added.
1784 \b \c{#}, \c{@}, \c{~} and c\{?} are now valid characters in labels.
1786 \b \c{-e} and \c{-k} options in NDISASM added.
1789 \S{cl-0.90 released October 1996} Version 0.90 released October 1996
1791 First release version. First support for object file output. Other
1792 changes from previous version (0.3x) too numerous to document.