4 This, like the AUTHORS file, is intended for easy readability by both human
5 and machine, thus the format.
8 V: version you should expect it by
9 R: responsible person or - if unassigned
12 D: maybe on multiple lines
14 Anything that doesn't start with /^[FVRCD]:/ should be ignored.
16 F:-line triggers new entry.
17 Empty V,R,C assume: V: ?, R: -, C: 0%
21 F: Extended x64 Support
22 D: Full FPU/MMX/SSE* instruction support for x64
24 F: ELF64 output format
25 D: Support for assembling code to the ELF64 output format
27 F: NDISASM x64 Support
28 D: Ability to disassemble respective x64 code
30 F: General x64 Support
34 D: Support for assembling 64-bit code to various output formats
36 F: win64 (x86-64 COFF) output format
40 D: Support for assembling code to the win64 output format
42 F: c99 data-type compliance
46 D: Revamped entire source-code base data-types for compliance
47 D: with c99 (inttypes.h)
49 F: __BITS__ Standard Macro
53 D: __BITS__ standard macro that returns current [BITS XX] mode
56 D: kkanios: be careful about that, stick to UTF-8 if anything
58 F: Convert shallow code model to deep code model
59 D: Tired of messing between lots of unrelated files (especially .c/.h stuff)
61 F: Automated dependency generation for Makefile
62 D: Current looks awful and will break if anything changes.
64 F: Move output modules out*.c to output/ subdir
68 == THESE ARE FROM old NASM's Wishlist
69 == THEY NEED SEVERE REVISING (seems they weren't updated for a couple of years or so)
71 F: Check misc/ide.cfg into RCS as Watcom IDE enhancement thingy
75 F: Package the Linux Assembler HOWTO
78 F: 3DNow!, SSE and other extensions need documenting
80 D: hpa: Does it really make sense to have a whole instruction set
81 D: reference packaged with the assembler?
82 D: kkanios: Yes, for me it was a great help... and still is.
84 F: prototypes of lrotate don't match in test/*. Fix.
87 F: Build djgpp binaries for 0.98 onwards. Look into PMODE/W as a stub
89 D: it might be a lot better than CWSDPMI. It's in PMW133.ZIP.
91 F: %undef operator that goes along with %define
95 F: Fix `%error' giving error messages twice.
97 D: Not especially important, as changes planned for 1.1x below will make
98 D: the preprocessor be only called once.
100 F: Sort out problems with OBJ
102 D: * TLINK32 doesn't seem to like SEGDEF32 et al. So for that, we
103 D: should avoid xxx32 records wherever we can.
104 D: * However, didn't we change _to_ using xxx32 at some stage? Try
105 D: to remember why and when.
106 D: * Apparently Delphi's linker has trouble with two or more
107 D: globals being defined inside a PUBDEF32. Don't even know if it
108 D: _can_ cope with a PUBDEF16.
109 D: * Might need extra flags. *sigh*
111 F: Symbol table output may possibly be useful.
113 D: Ken Martwick (kenm@efn.org) wants the following format:
114 D: labelname type offset(hex) repetition count
115 D: Possibly include xref addresses after repetition count?
119 D: There are various other bugs in outelf.c that make certain kinds
120 D: of relocation not work. See zbrown.asm. Looks like we may have to do
121 D: a major rewrite of parts of it. Compare some NASM code output with
122 D: equivalent GAS code output. Look at the ELF spec. Generally fix things.
126 D: NASM is currently using a kludge in ELF that involves defining
127 D: a symbol at a zero absolute offset. This isn't needed, as the
128 D: documented solution to the problem that this solves is to use
131 F: Debug information, in all formats it can be usefully done in.
133 D: * including line-number record support.
134 D: * "George C. Lindauer" <gclind01@starbase.spd.louisville.edu>
135 D: wants to have some say in how this goes through.
136 D: * Andrew Crabtree <andrewc@rosemail.rose.hp.com> wants to help out.
138 F: Think about a line-continuation character.
141 F: Consider allowing declaration of two labels on the same line,
143 D: syntax 'label1[:] label2[:] ... instruction'.
144 D: Need to investigate feasibility.
146 F: Quoting of quotes by doubling them, in string and char constants.
149 F: Two-operand syntax for SEGMENT/SECTION macro to avoid warnings
150 D: of ignored section parameters on reissue of __SECT__.
151 D: Or maybe skip the warning if the given parameters are identical to
152 D: what was actually stored. Investigate.
155 F: Apparently we are not missing a PSRAQ instruction, because it
156 D: doesn't exist. Check that it doesn't exist as an undocumented
157 D: instruction, or something stupid like that.
160 F: Any assembled form starting 0x80 can also start 0x82.
162 D: ndisasm should know this. New special code in instruction encodings, probably.
164 F: Pointing an EQU at an external symbol now generates an error.
166 D: There may be a better way of handling this; we should look into it.
167 D: Ideally, the label mechanism should be changed to cope with one
168 D: label being declared relative to another - that may work, but could be
169 D: a pain to implement (or is it? it may be easy enough that you just
170 D: need to declare a new offset in the same segment...) This should be done
171 D: before v1.0 is released. There is a comment regarding this in labels.c,
172 D: towards the end of the file, which discusses ways of fixing this.
174 F: nested %rep used to cause a panic.
176 D: Now a more informative error message is produced. This problem whould
177 D: be fixed before v1.0.
178 D: See comment in switch() statement block for PP_REP in do_directive()
179 D: in preproc.c (line 1585, or thereabouts)
182 D: zgraeme.tar contains improved hash table routines
183 D: contributed by Graeme Defty <graeme@HK.Super.NET> for use in the
187 D: zsyntax.zip contains a syntax-highlighting mode for
188 D: NASM, for use with the Aurora text editor (??).
191 D: zvim.zip contains a syntax-highlighting mode for NASM, for use with vim.
194 D: zkendal1.zip and zkendal2.zip contain Kendall
195 D: Bennett's (<KendallB@scitechsoft.com>) alternative syntax stuff,
196 D: providing an alternative syntax mode for NASM which allows a macro
197 D: set to be written that allows the same source files to be
198 D: assembled with NASM and TASM.
202 F: Add the UD2 instruction.
205 F: Add the four instructions documented in 24368901.pdf (Intel's own document).
208 F: Some means of avoiding MOV memoffs,EAX which apparently the
209 D: Pentium pairing detector thinks modifies EAX. Similar means of
210 D: choosing instruction encodings where necessary.
213 F: The example of ..@ makes it clear that a ..@ label isn't just
214 D: local, but doesn't make it clear that it isn't just global either.
216 F: hpa wants an evaluator operator for ceil(log2(x)).
218 F: Extra reloc types in ELF
219 D: R_386_16 type 20, PC16 is 21, 8 is 22, PC8 is 23.
220 D: Add support for the 16s at least.
222 F: Lazy section creation or selective section output
223 D: in COFF/win32 at least and probably other formats: don't bother to emit a section
224 D: if it contains no data. Particularly the default auto-created
225 D: section. We believe zero-length sections crash at least WLINK (in win32).
227 F: Make the flags field in `struct itemplate' in insns.h a long instead of an int.
230 F: Implement %ifref to check whether a single-line macro has ever been expanded since (last re) definition. Or maybe not. We'll see.
232 F: add pointer to \k{insLEAVE} and \k{insENTER} in chapters about mixed-language programming.
234 F: Some equivalent to TASM's GLOBAL directive
235 D: ie something which defines a symbol as external if it doesn't end up being defined
236 D: but defines it as public if it does end up being defined.
238 F: Documentation doesn't explain about C++ name mangling.
240 F: see if BITS can be made to do anything sensible in obj (eg set the default new-segment property to Use32).
242 F: OBJ: coalesce consecutive offset and segment fixups for the same location into full-32bit-pointer fixups.
243 D: This is apparently necessary because some twazzock in the PowerBASIC development
244 D: team didn't design to support the OMF spec the way the rest of the
247 F: Allow % to be separated from the rest of a preproc directive, for alternative directive indentation styles.
249 F: __DATE__, __TIME__, and text variants of __NASM_MAJOR__ and __NASM_MINOR__.
251 F: Warn on TIMES combined with multi-line macros.
253 D: TIMES gets applied to first line only - should bring to users' attention.
255 F: Re-work the evaluator, again, with a per-object-format fixup
256 D: routine, so as to be able to cope with section offsets "really"
257 D: being pure numbers; should be able to allow at _least_ the two
259 D: TIMES 510-$ DB 0 ; bootsector
260 D: MOV AX,(PROG_END-100H)/16 ; .COM TSR
261 D: Would need to call the fixup throughout the evaluator, and the
262 D: fixup would have to be allowed to return UNKNOWN on pass one if it
263 D: had to. (_Always_ returning UNKNOWN on pass one, though a lovely
264 D: clean design, breaks the first of the above examples.)
267 F: Preprocessor identifier concatenation?
270 F: Arbitrary section names in `bin'.
272 D: Is this necessary? Is it even desirable?
273 D: hpa: Desirable, yes. Necessary? Probably not, but there are definitely cases where it becomes quite useful.
277 F: Ability to read from a pipe.
279 D: Obviously not useful under dos, so memory problems with storing
280 D: entire input file aren't a problem either.
282 F: File caching under DOS/32 bit...
284 D: maybe even implement discardable buffers that get thrown away
285 D: when we get a NULL returned from malloc(). Only really useful under
286 D: DOS. Think about it.
288 F: possibly spool out the pre-processed stuff to a file, to avoid having to re-process it.
290 D: Possible problems with preprocessor values not known on pass 1? Have a look...
292 F: Or maybe we can spool out a pre-parsed version...?
294 D: Need to investigate feasibility. Does the results from the parser
295 D: change from pass 1 to pass 2? Would it be feasible to alter it so that
296 D: the parser returns an invariant result, and this is then processed
297 D: afterwards to resolve label references, etc?
299 F: Subsection support?
301 F: A good ALIGN mechanism, similar to GAS's.
303 D: GAS pads out space by means of the following (32-bit) instructions:
304 D: 8DB42600000000 lea esi,[esi+0x0]
305 D: 8DB600000000 lea esi,[esi+0x0]
306 D: 8D742600 lea esi,[esi+0x0]
307 D: 8D7600 lea esi,[esi+0x0]
308 D: 8D36 lea esi,[esi]
310 D: It uses up to two of these instructions to do up to 14-byte pads;
311 D: when more than 14 bytes are needed, it issues a (short) jump to
312 D: the end of the padded section and then NOPs the rest. Come up with
313 D: a similar scheme for 16 bit mode, and also come up with a way to
314 D: use it - internal to the assembler, so that programs using ALIGN
315 D: don't knock over preprocess-only mode.
316 D: Also re-work the macro form so that when given one argument in a
317 D: code section it calls this feature.
321 F: Possibly a means whereby FP constants can be specified as immediate operands to non-FP instructions.
322 D: * Possible syntax: MOV EAX,FLOAT 1.2 to get a single-precision FP
323 D: constant. Then maybe MOV EAX,HI_FLOAT 1.2 and MOV EAX,LO_FLOAT
324 D: 1.2 to get the two halves of a double-precision one. Best to
325 D: ignore extended-precision in case it bites.
326 D: * Alternatively, maybe MOV EAX,FLOAT(4,0-4,1.2) to get bytes 0-4
327 D: (ie 0-3) of a 4-byte constant. Then HI_FLOAT is FLOAT(8,4-8,x)
328 D: and LO_FLOAT is FLOAT(8,0-4,x). But this version allows two-byte
329 D: chunks, one-byte chunks, even stranger chunks, and pieces of
330 D: ten-byte reals to be bandied around as well.
332 F: A UNION macro might be quite cool
333 D: now that ABSOLUTE is sane enough to be able to handle it.
335 F: An equivalent to gcc's ## stringify operator, plus string concatenation
336 D: somehow implemented without undue ugliness, so as
337 D: to be able to do `%include "/my/path/%1"' in a macro, or something
340 F: Actually _do_ something with the processor, privileged and
341 D: undocumented flags in the instruction table. When this happens,
342 D: consider allowing PMULHRW to map to either of the Cyrix or AMD
344 D: hpa: The -p option to ndisasm now uses this to some extent.
347 F: Maybe NEC V20/V30 instructions? ?
348 D: hpa: What are they? Should be trivial to implement.
350 F: Yet more object formats.
351 D: * Possibly direct support for .EXE files?
354 F: Symbol map in binary format. Format-specific options...
357 F: REDESIGN: Think about EQU dependency, and about start-point specification in OBJ. Possibly re-think directive support.
360 F: Think about a wrapper program like gcc?
362 D: Possibly invent a _patch_ for gcc so that it can take .asm files on the command line?
363 D: If a wrapper happens, think about adding an option to cause the
364 D: resulting executable file to be executed immediately, thus
365 D: allowing NASM source files to have #!... (probably silly)
367 F: Multi-platform support?
368 D: If so: definitely Alpha; possibly Java byte code;
369 D: probably ARM/StrongARM; maybe Sparc; maybe Mips; maybe
370 D: Vax. Perhaps Z80 and 6502, just for a laugh?
372 F: Consider a 'verbose' option that prints information about the resulting object file onto stdout.
374 F: Line numbers in the .lst file don't match the line numbers in the input.
375 D: They probably should, rather than the current matching of the post-preprocessor line numbers.