H. Peter Anvin [Tue, 9 Sep 2008 16:54:47 +0000 (09:54 -0700)]
Run "make alldeps"
H. Peter Anvin [Tue, 9 Sep 2008 16:54:03 +0000 (09:54 -0700)]
macros.pl: don't include macros for disabled output formats
Don't include macros for output formats we aren't compiling in.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 9 Sep 2008 16:53:20 +0000 (09:53 -0700)]
outaout.mac: aoutb is an alias for aout
aoutb is an alias format for aout (in the sense they use the same
macro set.)
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 9 Sep 2008 16:43:16 +0000 (09:43 -0700)]
macros.pl: be a bit more careful with open()
Be a bit more strict with open() calls.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 9 Sep 2008 16:37:37 +0000 (09:37 -0700)]
macros: win32 and win64 are COFF, not Mach-O
Somehow the win32 and win64 aliases got listed on Mach-O, not on
COFF. This doesn't have any effect on the current code, but might in
the future. Correct.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Wed, 3 Sep 2008 22:25:33 +0000 (15:25 -0700)]
NASM 2.04rc1
H. Peter Anvin [Wed, 3 Sep 2008 22:22:23 +0000 (15:22 -0700)]
Remove obsolete tagrelease script (duplicate of tag-release)
H. Peter Anvin [Wed, 3 Sep 2008 22:16:44 +0000 (15:16 -0700)]
Update doc/changes.src
H. Peter Anvin [Wed, 3 Sep 2008 22:03:20 +0000 (15:03 -0700)]
Document %use smartalign
H. Peter Anvin [Wed, 3 Sep 2008 21:31:20 +0000 (14:31 -0700)]
Document %unmacro
H. Peter Anvin [Tue, 2 Sep 2008 19:56:39 +0000 (12:56 -0700)]
outcoff: stylistic cleanups
Don't use explicit L's for things which are really size_t; not only is
it unnecessarily ugly, but it's wrong in a lot of ways. Do some other
minor stylistic cleanups.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 2 Sep 2008 19:55:35 +0000 (12:55 -0700)]
outcoff: default output symbols to T_NULL
We would leave the output symbol type uninitialized. Explicitly
initialize it to zero (T_NULL, meaning no symbol type information),
since that's what was effectively done.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 1 Sep 2008 17:56:33 +0000 (10:56 -0700)]
BR 2048950: fix crash due to mmacro list overflow
When allocating the buffer for an mmacro list, we apparently failed to
guarantee space for the terminating NULL. This almost certainly
caused the crash described in BR 2048950, and quite possibly BR
1284169.
H. Peter Anvin [Sat, 30 Aug 2008 01:04:16 +0000 (18:04 -0700)]
test/new: clean up whitespace
H. Peter Anvin [Sat, 30 Aug 2008 00:27:00 +0000 (17:27 -0700)]
Script to create new test case boilerplate
H. Peter Anvin [Fri, 29 Aug 2008 01:05:23 +0000 (18:05 -0700)]
Accept implicit memory size for VMREAD/VMWRITE
H. Peter Anvin [Fri, 29 Aug 2008 01:03:49 +0000 (18:03 -0700)]
BR 2029472: Wrong operand size for VMREAD/VMWRITE in 64-bit mode
Fix the operand size for VMREAD/VMWRITE in 64-bit mode
H. Peter Anvin [Fri, 29 Aug 2008 00:56:17 +0000 (17:56 -0700)]
.gitignore: ignore some editor backup files
Ignore a couple of variants of editor backup files.
H. Peter Anvin [Fri, 29 Aug 2008 00:54:55 +0000 (17:54 -0700)]
BR 2028995: Missing MOVNTI m64, r64
Fix MOVNTI with a 64-bit argument.
H. Peter Anvin [Fri, 29 Aug 2008 00:48:34 +0000 (17:48 -0700)]
Testcase for XCRYPT
H. Peter Anvin [Fri, 29 Aug 2008 00:47:16 +0000 (17:47 -0700)]
BR 2029829: Accept VIA XCRYPT instructions with or without REP
Accept the VIA XCRYPT instructions either with or without a REP
prefix, as documented.
Add the missing XCRYPTCTR instruction.
H. Peter Anvin [Fri, 29 Aug 2008 00:35:25 +0000 (17:35 -0700)]
BR 2039212: Handle indirect far jumps in 64-bit mode
Handle indirect far jumps in 64-bit mode. Default to 64 bit unless
overridden, for consistency with other jumps.
H. Peter Anvin [Thu, 28 Aug 2008 20:36:10 +0000 (13:36 -0700)]
Include "nasm.h" in "float.h"
"float.h" references efunc, so it needs "nasm.h".
It also has uint8_t and so need <inttypes.h>, but that's part of
"nasm.h".
H. Peter Anvin [Thu, 28 Aug 2008 01:47:05 +0000 (18:47 -0700)]
Add 256-bit AVX stores per the latest AVX spec.
Add 256-bit forms of VMOVNTPD, VMOVNTPS, and VMOVNT[DQ]Q.
H. Peter Anvin [Thu, 28 Aug 2008 01:42:26 +0000 (18:42 -0700)]
Add AVX forms of the AES instructions (new in the latest AVX spec)
The AES instructions, too, have gotten VEX forms.
H. Peter Anvin [Mon, 25 Aug 2008 01:17:09 +0000 (18:17 -0700)]
test: Fix file with no final newline
Fix test/br2030823.asm, which had no final newline.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 25 Aug 2008 01:12:20 +0000 (18:12 -0700)]
BR 2067820: add the MOVSXD instruction
The official mnemonic for 32-to-64-bit sign extension is MOVSXD for
some idiotic reason. Add support for it while continue to recognize
MOVSX for this as an alias.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 25 Aug 2008 01:09:04 +0000 (18:09 -0700)]
BR 2067837: remove double add of VEX instruction length
We added the length of VEX prefixes twice in calcsize(); this resulted
in the wrong symbol addresses when compiling without the optimizer.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Wed, 20 Aug 2008 16:42:47 +0000 (09:42 -0700)]
BR 2062342: ndisasm: r12 *can* be an index register
R12 can be used as an index register. Special encodings in the modr/m
byte are done *without* consideration for the REX prefix, but special
encodings in the SIB byte *do* take the REX prefix into consideration,
since it doesn't affect the overall instruction format.
H. Peter Anvin [Wed, 13 Aug 2008 23:25:08 +0000 (16:25 -0700)]
BR 2030823: Problem with the 256-bit FMA instructions
Fix the 256-bit FMA instructions per bug report.
H. Peter Anvin [Wed, 13 Aug 2008 23:18:23 +0000 (16:18 -0700)]
BR 2043111: Typo in insns.dat: VCMPFT_OQPD VCMPFT_OQPS
Fix typo
Victor van den Elzen [Wed, 6 Aug 2008 13:15:01 +0000 (15:15 +0200)]
minor bugfix in performtest.pl
Victor van den Elzen [Wed, 6 Aug 2008 12:47:54 +0000 (14:47 +0200)]
Add macro-defaults warning class and documentation.
H. Peter Anvin [Fri, 1 Aug 2008 01:46:11 +0000 (18:46 -0700)]
BR 2034542: fix crash when touching __FILE__
Touching __FILE__ would cause a dereference of an uninitialized
pointer. Fix.
H. Peter Anvin [Thu, 31 Jul 2008 00:30:12 +0000 (17:30 -0700)]
BR 2028910: fix decoding of VEX prefixes in 16- and 32-bit mode
We would incorrectly set a bunch of VEX-related state for C4 and C5
bytes, even though we had already rejected it as not a VEX prefix due
to the top two bits of the following byte not being 11.
H. Peter Anvin [Thu, 31 Jul 2008 00:25:56 +0000 (17:25 -0700)]
BR 2025977: Handle SLDT with a 64-bit register operand
Handle SLDT with a 64-bit register operand. Don't generate a REX.W
prefix in the assembler, since zero-extending is just fine, but do
support it in the disassembler.
H. Peter Anvin [Wed, 23 Jul 2008 14:49:26 +0000 (10:49 -0400)]
preproc: restore correct break; in do_directive()
Checkin
a26433db6805cf1f1d711eb820f6a50a98f54f36 incorrectly changed a
few break;s in do_directive() that were *inside loops* to returns.
This broke single-line macros as well as %exitrep; fix.
Victor van den Elzen [Wed, 23 Jul 2008 13:14:22 +0000 (15:14 +0200)]
BR 560960: warn about trailing garbage in %macro/%ifmacro
Victor van den Elzen [Wed, 23 Jul 2008 11:21:29 +0000 (13:21 +0200)]
Improve checking and documentation for %ifctx
H. Peter Anvin [Sun, 20 Jul 2008 21:59:18 +0000 (14:59 -0700)]
BR 2023036: MOV reg32,dreg and vice versa are NOLONG
MOV reg32,dreg and MOV dreg,reg32 are NOLONG; in 64-bit mode we always
move to/from reg64.
H. Peter Anvin [Sun, 20 Jul 2008 16:40:14 +0000 (09:40 -0700)]
Err, "alignr" was really the same as the previous "alignb"
Replace "alignb" with the (slightly more optimized) "alignr" macro,
but keep the name "alignb".
H. Peter Anvin [Sun, 20 Jul 2008 16:32:31 +0000 (09:32 -0700)]
Add "alignr" standard macro (using resb)
Add an "alignr" standard macro, for use in BSS (nobits) and absolute
segments.
H. Peter Anvin [Sun, 20 Jul 2008 04:44:26 +0000 (21:44 -0700)]
Avoid redundant "const" for macros_t
Don't use a redundant "const" for macros_t (which is const unsigned
char), since OpenWatcom doesn't like it, and I believe it is incorrect
per the C standard.
H. Peter Anvin [Sun, 20 Jul 2008 04:40:07 +0000 (21:40 -0700)]
BR 2003451: add test case
Add test case for BR 2003451: forwardness leakage between operands.
H. Peter Anvin [Sun, 20 Jul 2008 04:38:56 +0000 (21:38 -0700)]
BR 2003451: avoid "forwardness" leaks between operands
Any use of ins->forw_ref that isn't related to control of the
optimizer is fundamentally broken. Use
operand->opflags & OPERAND_FORWARD instead. This even has the nice
side benefit of simplifying the code.
Slavik Gnatenko [Sun, 20 Jul 2008 02:27:41 +0000 (19:27 -0700)]
BR 2010180: outobj: Garbage may be written in a last PUBDEF
The testcase illustrates the problem. After "nasm -f obj
alonesym.nasm"
let's look to dump:
======
PUBDEF386(91) recnum:5, offset:0000005bh, len:03f9h, chksum:bbh(bb)
Group: 0, Seg: 1
00020000h - 'sym0000' Type:0
00020004h - 'sym0001' Type:0
....
00020134h - 'sym0077' Type:0
PUBDEF(90) recnum:6, offset:00000457h, len:000ah, chksum:b6h(b6)
Group: 0, Seg: 1
00000138h - 's' Type:2
0000b600h - '' Type:0
======
The problem is while 's' offset is 20138h it is marked as type 90h not
91h. The root cause is located in obj_x():
static ObjRecord *obj_x(ObjRecord * orp, uint32_t val)
{
if (orp->type & 1)
orp->x_size = 32;
if (val > 0xFFFF)
orp = obj_force(orp, 32);
if (orp->x_size == 32)
return (obj_dword(orp, val));
orp->x_size = 16;
return (obj_word(orp, val));
}
It sets up x_size and than writes data. In the testcase data are the
offset and this offset overflows a record. In this case the record is
emitted and its x_size is cleared. Because this is last PUBDEF the new
record with only 's' symbol is emitted also but its x_size is not 32
(it's still zero) so obj_fwrite doesn't switch to 91h type.
The problem seems to be very generic and expected to be occurred on
many other record types as well.
----
And the fix is simple:
if (orp->x_size == 32)
{
ObjRecord * nxt = obj_dword(orp, val);
nxt->x_size = 32; /* x_size is cleared when a record overflows */
return nxt;
}
H. Peter Anvin [Thu, 17 Jul 2008 21:29:07 +0000 (14:29 -0700)]
test: more smart alignment test
H. Peter Anvin [Thu, 17 Jul 2008 21:28:29 +0000 (14:28 -0700)]
smartalign: use context-local label
Use a context-local label in the smart align macro.
H. Peter Anvin [Thu, 17 Jul 2008 21:22:10 +0000 (14:22 -0700)]
smartalign: adjust the alignment threshold
Apparently the current recommendation is for a smaller threshold when
using the "generic"-style alignment macros (short jumps are cheaper on
newer CPUs.)
Also change the alignment threshold definition to reflect the maximum
number of padding instead of when to start using jumps.
H. Peter Anvin [Thu, 17 Jul 2008 21:20:06 +0000 (14:20 -0700)]
smartalign: 16-bit generic alignment macros
Smart alignment content for 16-bit "generic" mode
H. Peter Anvin [Thu, 17 Jul 2008 21:13:53 +0000 (14:13 -0700)]
smartalign: 16-bit P6 NOPs
Add 16-bit P6 NOPs
H. Peter Anvin [Wed, 16 Jul 2008 21:41:39 +0000 (14:41 -0700)]
smartalign.mac: smart alignments macro package
"%use smartalign" followed by an optional "alignmode" can be used to
enable smart macros.
H. Peter Anvin [Wed, 16 Jul 2008 21:40:01 +0000 (14:40 -0700)]
preproc: add %un[i]macro, add cleanups
Add %un[i]macro, and a few stylistic cleanups.
Note: unlike %undef, %un[i]macro takes an argument specification,
which must *exactly* match the macro being undefined. Similarly,
%unimacro has to be used to undefine a macro defined with %imacro, and
vice versa.
H. Peter Anvin [Wed, 16 Jul 2008 21:38:58 +0000 (14:38 -0700)]
standard.mac: allow non-power-of-2 alignments
Allow aligning to a non-power-of-2 boundary. It's probably useless,
but doesn't really hurt.
H. Peter Anvin [Wed, 16 Jul 2008 21:38:24 +0000 (14:38 -0700)]
preproc.c: fix %ifn, %elifn
The sense of %ifn and %elifn was reversed due to a bogus nonstandard
return sequence.
Victor van den Elzen [Wed, 16 Jul 2008 13:20:56 +0000 (15:20 +0200)]
Fix multipass inline warning (dis/en)abling
Also add a new form: resetting warnings to their original value.
Victor van den Elzen [Wed, 16 Jul 2008 11:41:37 +0000 (13:41 +0200)]
Fix %rep ... %endmacro crash
Also improved a comment and an error message.
Victor van den Elzen [Wed, 25 Jun 2008 10:00:21 +0000 (12:00 +0200)]
update tests
Victor van den Elzen [Wed, 25 Jun 2008 09:41:40 +0000 (11:41 +0200)]
fix unitialized variable in eval_strfunc
Victor van den Elzen [Wed, 4 Jun 2008 13:24:20 +0000 (15:24 +0200)]
Fix fclose bug on error.
Contrary to the comments, the fclose is needed.
Failure to close the file caused remove to fail on Windows.
Victor van den Elzen [Wed, 4 Jun 2008 10:44:31 +0000 (12:44 +0200)]
Add a 'make test' target.
Victor van den Elzen [Wed, 28 May 2008 12:02:37 +0000 (14:02 +0200)]
Improve performtest.pl
Improve arguments and documentation of performtest.pl
Remove carriage returns in .stdout/.stderr so *nix can
read Windows test results
H. Peter Anvin [Mon, 14 Jul 2008 06:54:00 +0000 (02:54 -0400)]
changes.src: put \c{...} around macro directives
Put \c{...} around macro directives. Not the only ones that should
have that, of course, but they were easy to do with search and replace.
H. Peter Anvin [Mon, 14 Jul 2008 06:49:52 +0000 (02:49 -0400)]
changes.src: remove double entry for %warning
H. Peter Anvin [Mon, 14 Jul 2008 06:45:57 +0000 (02:45 -0400)]
Move the revision history into the documentation
Clumsily convert the revision history to nasmdoc format, so it can be
included in the documentation as Appendix C.
H. Peter Anvin [Sun, 13 Jul 2008 22:55:55 +0000 (15:55 -0700)]
doc: document packed BCD constants
H. Peter Anvin [Sun, 13 Jul 2008 22:44:25 +0000 (15:44 -0700)]
doc: update NASM Version Macros
Clean up and slightly update the section on NASM version macros.
H. Peter Anvin [Sun, 13 Jul 2008 22:41:36 +0000 (15:41 -0700)]
doc: move %error/%warning to a separate section
They don't really belong in the section on conditional assembly.
H. Peter Anvin [Sun, 13 Jul 2008 22:35:07 +0000 (15:35 -0700)]
doc: Document %strcat
H. Peter Anvin [Sun, 13 Jul 2008 22:25:54 +0000 (15:25 -0700)]
Update CHANGES to current delta from 2.03.x.
H. Peter Anvin [Sun, 13 Jul 2008 22:21:01 +0000 (15:21 -0700)]
BR 2017453: indirect jumps in 64-bit mode are implicitly 64 bits
Indirect jumps in 64-bit mode implicitly have 64-bit operand size.
Fix this; the disassembly is still unnecessarily ugly, however.
H. Peter Anvin [Sun, 13 Jul 2008 22:06:55 +0000 (15:06 -0700)]
test: add test of nested %rep, BCD constants, and %warning
Add a test case which has smoked out errors in the handling of nested
%rep, BCD constants, and %warning...
H. Peter Anvin [Sun, 13 Jul 2008 22:05:53 +0000 (15:05 -0700)]
preproc.c: make %warning actually issue a warning...
The calculation of "severity" was buggered up, with the result that
%warning actually issued an error.
H. Peter Anvin [Sun, 13 Jul 2008 20:54:47 +0000 (13:54 -0700)]
version.mak for the version Makefile fragment
Be consistent about the naming of the version Makefile fragment. We
use .mak elsewhere for Makefiles, so use that.
Charles Crayne [Sun, 13 Jul 2008 19:52:02 +0000 (12:52 -0700)]
Fix Bugs item #2017455 (LTR in long mode)
LTR is valid in long (64-bit) mode, but still uses
16-bit operand, so remove NOLONG restriction.
Charles Crayne [Sat, 12 Jul 2008 23:42:33 +0000 (16:42 -0700)]
%EXITREP inside nested %REPs
Apply updated version of fix submitted with feature request 803785.
This fix causes %exitrep to terminate only the innermost %rep block,
and also allows the count for nested blocks to be calculated in the
containing block.
H. Peter Anvin [Sun, 6 Jul 2008 00:45:01 +0000 (17:45 -0700)]
Add version.make to PERLREQ
version.make is produced by a Perl script, and therefore should be in
PERLREQ.
H. Peter Anvin [Fri, 4 Jul 2008 03:16:40 +0000 (20:16 -0700)]
test: simple test of packed BCD.
H. Peter Anvin [Fri, 4 Jul 2008 03:16:07 +0000 (20:16 -0700)]
For consistency, allow 0p.. prefix for packed BCD
Allow 0p... to be used as a prefix, for analogy with base conversion.
H. Peter Anvin [Fri, 4 Jul 2008 03:12:37 +0000 (20:12 -0700)]
float: support packed-BCD constants in 'dt' statements
Support packed-BCD constants in 'dt' statements (and elsewhere 80-bit
floating point is handled), using MASM syntax (terminal 'p').
H. Peter Anvin [Fri, 4 Jul 2008 03:11:30 +0000 (20:11 -0700)]
float: fix buffer overrun
Fix a buffer overrun; generally causing hexadecimal constants to be
incorrectly rejected.
H. Peter Anvin [Thu, 3 Jul 2008 01:11:49 +0000 (18:11 -0700)]
standard.mac: use anonymous contexts
Use anonymous %push instead of giving a context name.
H. Peter Anvin [Thu, 3 Jul 2008 01:11:04 +0000 (18:11 -0700)]
preproc: Allow anonymous contexts
Allow %push and %repl without a context name. For a lot of uses, it
is only a potential source of namespace pollution.
H. Peter Anvin [Wed, 2 Jul 2008 04:42:08 +0000 (21:42 -0700)]
Permit commas in %strcat
H. Peter Anvin [Wed, 2 Jul 2008 04:26:27 +0000 (21:26 -0700)]
preproc: %strcat directive to concatenate quoted strings
I noticed there was no sane way to concatenate the contents of quoted
strings, so add the %strcat directive.
These really need to become preprocessor functions at some stage.
H. Peter Anvin [Sun, 29 Jun 2008 01:53:55 +0000 (18:53 -0700)]
Update the INSTALL file to match current reality
H. Peter Anvin [Sun, 29 Jun 2008 01:39:13 +0000 (18:39 -0700)]
Document case-insensitivity bug.
H. Peter Anvin [Sun, 29 Jun 2008 01:32:16 +0000 (18:32 -0700)]
test: add a test for %imacro
Add a test for case-insensitive matching of %imacro.
H. Peter Anvin [Sun, 29 Jun 2008 01:31:08 +0000 (18:31 -0700)]
nasmlib: fix nasm_str[n]icmp()
Fix nasm_str[n]icmp() on platforms which don't have this function
natively.
XXX: Given the new nasm_tolower() implementation, we should consider
if this might actually be a faster function than the platform-native
one.
H. Peter Anvin [Sun, 29 Jun 2008 01:30:27 +0000 (18:30 -0700)]
preproc: MMacro.finishes is a pointer, not a boolean
MMacro.finishes is a pointer, not a boolean, so set it to "false", not
"NULL".
H. Peter Anvin [Fri, 27 Jun 2008 18:41:59 +0000 (11:41 -0700)]
AES instructions are WESTMERE, not NEHALEM
Still need to make this crap saner...
H. Peter Anvin [Wed, 25 Jun 2008 21:54:14 +0000 (14:54 -0700)]
Make the macros table "unsigned char"
It gets less ugly if we make the macros table "unsigned char".
H. Peter Anvin [Sun, 22 Jun 2008 01:19:13 +0000 (18:19 -0700)]
Run "make alldeps"
H. Peter Anvin [Sun, 22 Jun 2008 01:18:41 +0000 (18:18 -0700)]
pp_directives_len can be uint8_t
Save a few hundred bytes...
H. Peter Anvin [Sat, 21 Jun 2008 22:15:40 +0000 (15:15 -0700)]
Drop the index tables from the canned macros
Instead of an array of strings, just have a character array; that
reduces the size of canned macros by up to 30%, and we only did
sequential access anyway.
H. Peter Anvin [Sat, 21 Jun 2008 18:03:51 +0000 (11:03 -0700)]
Fix a few more <ctype.h> instances
A few isolated instances of isalpha() and isxdigit().
H. Peter Anvin [Sat, 21 Jun 2008 17:23:17 +0000 (10:23 -0700)]
ctype.h: wrapper ctype functions with a cast to (unsigned char)
ctype functions take an *int*, which the user is expected to have
taken the input character from getc() and friends, or taken a
character and cast it to (unsigned char).
We don't care about EOF (-1), so use macros that cast to (unsigned
char) for us.
H. Peter Anvin [Fri, 20 Jun 2008 22:55:45 +0000 (15:55 -0700)]
preproc: free the include path and the final filename
Memory leaks: free the include path, and the final used filename.
H. Peter Anvin [Fri, 20 Jun 2008 22:20:16 +0000 (15:20 -0700)]
Move the output format macros into the macros.pl mechanism
Move the handling of "extra" macros (i.e. output format macros) into
the macros.pl mechanism. This allows us to change the format of the
internal macro store in the future - e.g. to a single byte store
without redundant pointers.
Also, stop using indicies into a long array when there is no good
reason to not just use different arrays.
H. Peter Anvin [Fri, 20 Jun 2008 06:19:20 +0000 (23:19 -0700)]
raa: clean up indentation
Clean up in indentation in the RAA code.