Alexandru Cornea [Mon, 1 Jul 2013 18:25:03 +0000 (21:25 +0300)]
packaging: resetting manifest requested domain to floor
Change-Id: I7298c87a555157f173f56cbd044f29809bfad6e2
Anas Nashif [Sun, 14 Apr 2013 08:10:29 +0000 (01:10 -0700)]
packaging: Update to 2.10.07
Change-Id: I9e8a1ba87239bca6505fe37e70ca2c3607a012a1
Anas Nashif [Mon, 5 Nov 2012 21:55:04 +0000 (13:55 -0800)]
packaging: add packaging
Change-Id: I0f62fb2bd3f696e6b79ea1a58117fe81ed4e84d7
Anas Nashif [Mon, 5 Nov 2012 20:56:52 +0000 (12:56 -0800)]
packaging: add packaging
Change-Id: Id5d0f3b9482dce77a640467ec475b5841dc23e41
H. Peter Anvin [Wed, 21 May 2014 15:29:41 +0000 (08:29 -0700)]
NASM 2.11.05
H. Peter Anvin [Wed, 21 May 2014 15:28:21 +0000 (08:28 -0700)]
doc: Document --v and duplicate REX prefix fix
Document changes for 2.11.05.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Wed, 21 May 2014 15:19:16 +0000 (08:19 -0700)]
BR 3392279: Fix duplicated REX prefixes
The fix for BR 3392278:
aa29b1d93f5a assemble.c: Don't drop rex prefix from instruction itself
... would cause multiple REX prefixes to be emitted for some
instructions. Create a new flag to indicate that REX has already been
emitted, which can be cleared for each instance of an instruction.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cyrill Gorcunov [Fri, 9 May 2014 18:34:34 +0000 (22:34 +0400)]
options: Add --v option
It's been requested a long ago to handle '--v' option same was as
a regualar '-v'. From initial report
| NASM and yasm are in many respects compatible but yasm uses --v
| instead of -v for version. As often --v is used for version I
| end up using --v initially in NASM. This patch allows me to compile
| Mozilla apps which use yasm with NASM by merely renaming NASM to yasm
| so that the build environment does not have to be updated (Mozilla
| would not accept changes to allow use of NASM).
Reported-by: Andy Willis <abwillis1@gmail.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Fri, 9 May 2014 22:04:42 +0000 (15:04 -0700)]
NASM 2.11.04
Jin Kyu Song [Wed, 7 May 2014 20:55:58 +0000 (13:55 -0700)]
changes: Document bug fixes
- Removed an error checking code for setting evex flags
- Fixed vector length matching bug
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Mon, 5 May 2014 20:58:51 +0000 (13:58 -0700)]
ndisasm: Match vector length with EVEX.b set
With broadcasting, EVEX.L'L should be matched even when EVEX.b is set.
Only in a case of embedded rounding, EVEX.L'L is ignored in matching
function since it becomes EVEX.RC.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Mon, 5 May 2014 20:06:03 +0000 (13:06 -0700)]
AVX512: Remove invalid error checking
An offset-only memref can also have an opmask decorator.
e.g.) vmovdqu32 [0xabcd]{k1}, zmm0
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Mon, 5 May 2014 18:22:02 +0000 (11:22 -0700)]
NASM 2.11.03
H. Peter Anvin [Mon, 5 May 2014 18:21:17 +0000 (11:21 -0700)]
changes: Document TIMES bug
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cyrill Gorcunov [Sun, 4 May 2014 20:30:58 +0000 (00:30 +0400)]
assemble.c: Don't drop rex prefix from instruction itself
emit_rex is supposed to write REX prefix into output stream
if needed, but we happen to drop it off on a first write
which breaks REX required instructions if TIMES directive
is used.
For example the code like
| times 4 movq xmm11, xmm11
compiles into
|
0000000000000000 <.text>:
| 0: f3 45 0f 7e db movq %xmm11,%xmm11
| 5: f3 0f 7e db movq %xmm3,%xmm3
| 9: f3 0f 7e db movq %xmm3,%xmm3
| d: f3 0f 7e db movq %xmm3,%xmm3
instead of proper
|
0000000000000000 <.text>:
| 0: f3 45 0f 7e db movq %xmm11,%xmm11
| 5: f3 45 0f 7e db movq %xmm11,%xmm11
| a: f3 45 0f 7e db movq %xmm11,%xmm11
| f: f3 45 0f 7e db movq %xmm11,%xmm11
http://bugzilla.nasm.us/show_bug.cgi?id=3392278
Reported-by: Javier <elpochodelagente@gmail.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Wed, 19 Feb 2014 23:50:26 +0000 (15:50 -0800)]
NASM 2.11.02
H. Peter Anvin [Wed, 19 Feb 2014 23:39:57 +0000 (15:39 -0800)]
Add CLFLUSHOPT instruction
Add the CLFLUSHOPT instruction from the Intel Instruction Set
Architecture Extensions document version 319433-018 (Feb 2014).
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Wed, 19 Feb 2014 22:55:54 +0000 (14:55 -0800)]
insns: add XSAVEC, XSAVES and XRSTORS instructions
Add the XSAVEC, XSAVES, and XRSTORS instructions from the Intel SDM
release 253665-050US (Feb 2014).
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 18 Feb 2014 22:05:52 +0000 (14:05 -0800)]
NASM 2.11.01
H. Peter Anvin [Tue, 18 Feb 2014 22:05:14 +0000 (14:05 -0800)]
changes.src: changelog for 2.11.01
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 18 Feb 2014 21:23:27 +0000 (13:23 -0800)]
outelf: Error out on "section align" without value
If someone specifies "section align" without =value, error out.
Reported-by: Ilya Albrekht <ilya.albrekht@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Sun, 16 Feb 2014 18:25:25 +0000 (10:25 -0800)]
BR 3392275: Don't require xmm0 to be specified when implicit
BR 3392275 complains about xmm0 having to be explicitly included in
the assembly syntax when it is implicit in the encoding. In the
interest of "be liberal in what you accept", accept either form in the
input.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cyrill Gorcunov [Sat, 15 Feb 2014 14:40:12 +0000 (18:40 +0400)]
BR3392274: output: Elf -- Don't crash on erronious syntax
Elf align section attribute requires syntax "align=value",
but in case if '=' is missed we pass nil pointer into
atoi function which cause libc to crash.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
H. Peter Anvin [Tue, 31 Dec 2013 18:40:10 +0000 (10:40 -0800)]
NASM 2.11
H. Peter Anvin [Tue, 31 Dec 2013 18:38:51 +0000 (10:38 -0800)]
doc: Document DEFAULT BND/NOBND in changes, add use case
Add DEFAULT BND/NOBND to the change history, and explain the use case.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 31 Dec 2013 18:35:12 +0000 (10:35 -0800)]
changes: Document change in [nosplit reg]
Document that [nosplit reg] as opposed to [nosplit reg*1] will no
longer force an index register.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Jin Kyu Song [Thu, 19 Dec 2013 06:01:14 +0000 (22:01 -0800)]
nosplit: Generate index-only EA only when a multiplier is used.
[nosplit eax] has been encoded as [eax*1+0] since 0.98.34.
But this seems like unexpected behavior.
So only when a register is multiplied, that will be treated
as an index. ([nosplit eax*1] -> [eax*1+0])
Document is updated accordingly.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 19 Dec 2013 05:31:51 +0000 (21:31 -0800)]
mpx: Clean up instruction data
Cleaned up unneccessary size specifiers in the instruction data.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 19 Dec 2013 05:28:41 +0000 (21:28 -0800)]
nosplit: Limit the effect of NOSPLIT
[nosplit eax+eax] was encoded [eax*2] previously but
this seems against the user's intention.
So in this case, nosplit is ignored now and [eax+eax] will be
generated.
Document is also updated accordingly.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 19 Dec 2013 05:28:17 +0000 (21:28 -0800)]
mib: Avoid RIP-relative addressing in mib
Using RIP relative for mib operands causes #UD exception.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Cyrill Gorcunov [Fri, 13 Dec 2013 06:59:39 +0000 (10:59 +0400)]
insns: Mark LOADALL, LOADALL286 with ND flag
Otherwise disassembler treat syscall, sysret incorrectly.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Jin Kyu Song [Wed, 11 Dec 2013 00:24:45 +0000 (16:24 -0800)]
mib: Handle MIB EA in a different way from regular EA's
In mib operands, users' intention should be preserved.
e.g.) [eax + eax*1] and [eax*2] must be distinguished and encoded differently.
So a new EA flag EAF_MIB for mib operands is added.
And a new EA hint EAH_SUMMED for the case of [eax+eax*4] being parsed
as [eax*5] is also added.
NOSPLIT specifier does not have an effect in mib, so [nosplit eax + eax*1]
will be encoded as [eax, eax] rather than [eax*2] as in a regular EA.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Wed, 11 Dec 2013 20:09:09 +0000 (12:09 -0800)]
misc/release: Generate manpages
asciidoc/xmlto are not tools we require every users to have, so each
tarball should contain them. That means the release script needs to
know about them.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cyrill Gorcunov [Tue, 10 Dec 2013 07:09:49 +0000 (11:09 +0400)]
iflag: Don't use c99 array initialization
It's sad but not all compilers support c99 features, so drop
off IFLAG_INIT helper.
Reported-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 7 Dec 2013 12:15:03 +0000 (16:15 +0400)]
insns-iflags: Drop occasionally introduced \Tab's
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 7 Dec 2013 12:14:00 +0000 (16:14 +0400)]
iflag: \Tabs -> \Space
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 7 Dec 2013 12:12:07 +0000 (16:12 +0400)]
iflag: Introduce IFLAG_INIT helper
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
H. Peter Anvin [Thu, 5 Dec 2013 08:06:45 +0000 (00:06 -0800)]
NASM 2.11rc4
Jin Kyu Song [Thu, 5 Dec 2013 04:51:13 +0000 (20:51 -0800)]
doc: Update nasmdoc
Added bnd warning and nobnd prefix. DEFAULT directive section
has got more description about BND-related settings.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 5 Dec 2013 04:05:55 +0000 (20:05 -0800)]
bnd: Add a new nobnd prefix
bnd and nobnd prifixes can be used for each instruction line to
direct whether bnd registers should be preserved or not.
And those are also added as options for DEFAULT directive.
Once bnd is set with default, DEFAULT BND, all bnd-prefix
available instructions are prefixed with bnd. To override it,
nobnd prefix can be used.
In the other way, DEFAULT NOBND can disable DEFAULT BND and
have nasm encode in the normal way.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 29 Nov 2013 08:38:29 +0000 (00:38 -0800)]
bnd: Show warning when bnd prefix is dropped
When bnd prefix is dropped as jmp is encoded as jmp short,
nasm shows a warning message, which can be suppressed with a new
command line option, -w-bnd.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Sun, 1 Dec 2013 00:38:42 +0000 (16:38 -0800)]
pfmask: Limit the preferred mask to the vendor specific flags
In ndisasm, the priority follows the order of instructions in insns.dat.
Other iflags could affect this mechanism when a proper instruction form
had a higher iflag bit set.
The preferred mask bits are now limited to vendor flags (Cyrix and AMD)
and other flags do not affect disassembler any more.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Fri, 29 Nov 2013 05:26:26 +0000 (21:26 -0800)]
doc: Make the bit about mib operands a bit clearer
Clean up the text about what a mib is.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Fri, 29 Nov 2013 05:24:09 +0000 (21:24 -0800)]
doc: Clean up changelog for 2.11
- We don't need to list internal infrastructure improvements.
- We don't list rc releases separately.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Thu, 28 Nov 2013 20:22:05 +0000 (12:22 -0800)]
NASM 2.11rc3
H. Peter Anvin [Thu, 28 Nov 2013 20:11:24 +0000 (12:11 -0800)]
Add {vex3} and {vex2} prefixes by analogy with {evex}
Allow specifying {vex3} or {vex2} (the latter is currently always
redundant, unless we end up with instructions at some point can be
specified with legacy prefixes or VEX) to select a specific encoding
of VEX-encoded instructions.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Thu, 28 Nov 2013 19:36:26 +0000 (11:36 -0800)]
NASM 2.11rc2
H. Peter Anvin [Thu, 28 Nov 2013 19:33:28 +0000 (11:33 -0800)]
Remove "high 16" register class macros for xmm/ymm/zmm
The "high 16" register class macros were actually incorrect, as they
simply aliased the corresponding whole set class. In oder to keep
someone from getting confused and making mistakes, remove them.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Jin Kyu Song [Thu, 28 Nov 2013 04:52:22 +0000 (20:52 -0800)]
testcase: Remove escape characters - '\'
Since the multi-line macro preprocessor is modified to expand
grouped parameter with braces. The escape character is not needed
any more.
The testcase converter script is also modified not to generate '\'.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 28 Nov 2013 04:52:16 +0000 (20:52 -0800)]
preproc: Handle curly braces in multi-line macro parameters
Multi-line macro uses curly braces for enclosing a parameter
containing comma(s). Passing curly braces as a part of a parameter
which is already enclosed with braces confuses the macro expander.
The number of braces in a group parameter is counted and any brace
in the outmost enclosing braces is treated as a part of parameter.
e.g.) mmacro {1,2,3}, {4,{5,6}}
mmacro gets 2 parameters of '1,2,3' and '4,{5,6}'
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 28 Nov 2013 04:23:53 +0000 (20:23 -0800)]
Revert "AVX-512: Handle curly braces in multi-line macro parameters"
This reverts commit
a800aed7b75d56114f2e1e4928cbc48ecf96a4a0.
As recommended by the community, braces inside a group parameter
of multi-line macro should be parsed without a need of a leading
escape character such as "\{ab,c\}".
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 27 Nov 2013 22:10:40 +0000 (14:10 -0800)]
stdscan: Rework curly brace parsing routines
As recommended by the community, a comma-separated decorators ({k1,z})
and nested braces ({{k1},{z}}) are dropped out. So only standard syntax
is supported from now.
This rework made source code neat and easy to maintain. Most of the codes
for handling corner cases are removed.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 27 Nov 2013 01:27:48 +0000 (17:27 -0800)]
iflags: Add IF_EVEX for checking {evex} availability
For checking the availability of {evex} prefix, AVX512 iflag
has been used. But this is a flag for an instruction set
not for an encoding scheme. And there are some AVX512 instructions
encoded with VEX prefix.
So a new instruction flag (IF_EVEX) is added for the instructions
which are actually encoded with EVEX prefix.
This flag is automatically added by insns.pl, so no need to add manually
in insns.dat.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 27 Nov 2013 01:14:07 +0000 (17:14 -0800)]
opflags: Separate vector registers into low-16 and high-16
Since only EVEX supports all 32 vector registers encoding for now,
VEX/REX encoded instructions should not take high-16 registers as operands.
This filtering had been done using instruction flag so far, but
using the opflags makes more sense.
[XYZ]MMREG operands used for non-EVEX instructions are automatically
converted to [XYZ]MM_L16 in insns.pl
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Tue, 26 Nov 2013 01:59:25 +0000 (17:59 -0800)]
doc: Update the change history and nasmdoc
Added the list of features added since 2.10 release.
Nasmdoc is also updated with those new features.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Wed, 27 Nov 2013 21:43:45 +0000 (13:43 -0800)]
disasm: Don't rely on iflag_cmp() returning +/-1
It is safer to just rely on the sign, for future options.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Wed, 27 Nov 2013 21:41:50 +0000 (13:41 -0800)]
iflag: Do the equality test in iflag_cmp() first
The equality test indicates how long we spin, so do that first.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 26 Nov 2013 23:21:15 +0000 (15:21 -0800)]
iflag: In iflag_cmp() scan from the most significant word down
In order for iflag_cmp() to return an ordering that makes sense, we
need to scan from the most significant word downward. That way the
bits with the higher index consistently are the more significant.
This fixes the disassembler vendor selection algorithm. While we are
doing that, make that dependency more explicit in the comments.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 25 Nov 2013 01:14:34 +0000 (17:14 -0800)]
iflag: Drop the use of double underscores
Double underscores are reserved for the implementation, i.e. the C
compiler and its libraries. NASM is an application and should not use
this namespace.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 25 Nov 2013 01:13:20 +0000 (17:13 -0800)]
iflag: Make the insns_flags array const
Make the insns_flags array const, and change the helper functions to
match.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 25 Nov 2013 01:07:33 +0000 (17:07 -0800)]
insns.dat: Fix a typo in a comment
isnsn-iflags.pl not insns-iflag.pl.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Sun, 24 Nov 2013 19:52:32 +0000 (11:52 -0800)]
iflag: Move instruction flag comment
Move the instruction flag comment to the Perl file where they are
defined.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Sun, 24 Nov 2013 19:49:24 +0000 (11:49 -0800)]
iflag: remove C99 constructs, don't hardcode the number of words
Avoid using C99 constructs when not necessary. Don't hardcode the
number of words when we can autodiscover them.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Sun, 24 Nov 2013 19:18:50 +0000 (11:18 -0800)]
"make alldeps"
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Sun, 24 Nov 2013 19:13:10 +0000 (11:13 -0800)]
iflag: Fix dependencies, factor out static components of iflag.h
Multi-dependencies don't work as expected, especially not across Make
versions, this is why we don't use them and read the instructions list
multiple times.
iflag.h has a lot of static content, so factor out the static content.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Sun, 24 Nov 2013 18:56:49 +0000 (10:56 -0800)]
"make alldeps"
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cyrill Gorcunov [Sun, 10 Nov 2013 07:45:24 +0000 (11:45 +0400)]
insns: Restore back MMX,FPU flags
In commit
9bb987d8e0330429afba42015b1fc7c7ca0d1b16
we had to drop some flags due to flags type used
in nasm code.
Since now flags internal structure is reworked, we
can restore them back.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 9 Nov 2013 18:28:52 +0000 (22:28 +0400)]
Update .gitignore
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 9 Nov 2013 18:28:05 +0000 (22:28 +0400)]
make: Update deps
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 9 Nov 2013 18:16:11 +0000 (22:16 +0400)]
iflag: Start using new instruction flags engine
Here we start using instruction flags generator.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sat, 26 Oct 2013 21:20:42 +0000 (01:20 +0400)]
insns: Introduce insns-flags.pl
It been found that 64 bits for instruction flags is too small,
so instead we start using indirect addressing scheme to keep
instruction flags in bitvectors instead.
Using one bitvector per instruction template entry is wastefull
(especially if vector grow in future, at moment it's 128 bit length),
so we use indirect addressing, which is generated as follow
- read instruction flags from insns.dat
- flag sequence sorted and joined into one key string
- this key string become a hash index
- all hash entries are compacted into one array
- every instruction template uses array offset instead
of flags bitfield
Just for info, at moment we have 195 unique flags combination,
but since instruction template will use index as unsigned
integer, we can use a way more wider combination of flags
in future.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cyrill Gorcunov [Sun, 24 Nov 2013 08:26:35 +0000 (12:26 +0400)]
make: Add 'golden' target
We've 'test' target in toplevel Makefile, lets be
complete and allow to generate "golden" tests from
toplevel as well.
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Jin Kyu Song [Fri, 22 Nov 2013 03:40:42 +0000 (19:40 -0800)]
bnd: Drop bnd prefix for relaxed short jmp instructions
Reverted the redundant branch instruction patterns for bnd prefix.
And when a relaxed jmp instruction becomes a short (Jb) form,
bnd prefix is not needed because it does not initialize bnd registers.
So in that case, bnd prefix is silently dropped.
BND JMP foo -> drops bnd prefix
BND JMP short foo -> shows an explicit error
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 20 Nov 2013 23:32:52 +0000 (15:32 -0800)]
disp8: Consolidate a logic to get compressed displacement
Consolidated two separate but similar functions in nasm and ndisasm
into a commonly linked source code.
To encode and decode the compressed displacement (disp8*N) for EVEX,
N value should be derived using various conditions.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 20 Nov 2013 02:44:13 +0000 (18:44 -0800)]
disasm: Add suport for bnd registers
MPX uses a new bnd registers and a new mib syntax.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 20 Nov 2013 02:43:04 +0000 (18:43 -0800)]
disasm: Add EVEX decorator syntax
Broadcasting, opmask, embedded rounding and SAE decorators
are not added at proper position.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 15 Nov 2013 21:20:55 +0000 (13:20 -0800)]
disasm: Add ZMM vsib
ZMM registers can be used for vsib index and the number of registers
has been increased to 32.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 15 Nov 2013 21:18:32 +0000 (13:18 -0800)]
disasm: Support EVEX compressed displacement
When tuple type is indicated in the bytecode, disp8 is
treated as a compressed disp and gets calculated back to disp8*N.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Mon, 11 Nov 2013 19:49:10 +0000 (11:49 -0800)]
disasm: Add basic AVX512 support
Disassembler can translate EVEX prefix, count up to 32 vector registers
and recognize new ZMM / opmask registers.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Sat, 9 Nov 2013 01:22:24 +0000 (17:22 -0800)]
disasm: style cleanup
Fix indentation.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 30 Oct 2013 10:12:45 +0000 (03:12 -0700)]
match: Check the number of elements in broadcasting operands
The broadcasting decorator {1to##} must describe exactly how many times
the memory element is repeated in order to clearly match the correct
instruction format.
For example,
vaddpd zmm30,zmm29,QWORD [rdx+0x3f8]{1to8} ; good
vaddpd zmm30,zmm29,QWORD [rdx+0x3f8]{1to16} ; fail qword * 16 = 1024b
vaddps zmm30,zmm29,DWORD [rcx]{1to16} ; good
vaddps zmm30,zmm29,DWORD [rcx]{1to8} ; fail dword * 8 = 256b
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 30 Oct 2013 10:00:12 +0000 (03:00 -0700)]
match: Improve broadcast opsize matching
Broadcasting operand size is different from the original
operand size because 32b or 64b element is repeated to form a vector.
So when matching a broadcasting operand, opsize should be treated
differently.
The broadcasting element size is specified in the decorator information.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Tue, 29 Oct 2013 00:07:57 +0000 (17:07 -0700)]
Build: Suppress warning messages
Giving a correct printf format specifier supresses the warning message.
And a local pointer variable is initialized with NULL.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Sat, 26 Oct 2013 02:29:53 +0000 (19:29 -0700)]
AVX-512: Add {evex} instruction prefix
For instructions that can be encoded either in VEX or EVEX,
{evex} forces nasm to encode in EVEX.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Thu, 24 Oct 2013 01:39:03 +0000 (18:39 -0700)]
PREFETCHWT1: Add a new instruction flag
PREFETCHWT1 instruction's CPUID was TBD before.
Now it has its new CPUID bit : PREFETCHWT1
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 18 Oct 2013 03:37:26 +0000 (20:37 -0700)]
AVX512: Update instruction group
CPUID bit for vptestnm[dq] changed.
vptestnm[dq]: AVX512CD -> AVX512F
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 8 Nov 2013 21:31:58 +0000 (13:31 -0800)]
MPX: Move BND prefix indication from bytecode to iflags
As BND prefix validity check conflicts with jcc8 prefix,
IF_BND is added for the instruction templates which can have
bnd prefix for preserving the content of bound register.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Fri, 8 Nov 2013 19:41:12 +0000 (11:41 -0800)]
MPX: Adapt GAS's mib syntax with an index reg only
GAS uses *1 multiplier for explicitly marking an index register in mib operand.
e.g.) [rdx * 1 + 3] is equivalent to [3, rdx] in NASM's split EA format
So only for mib operands, this is encoded same as gas does.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 02:59:10 +0000 (19:59 -0700)]
iflags: Use UINT64_C() for 64bit values
UINT64_C() macro is used for iflags as it becomes 64bit.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 03:49:54 +0000 (20:49 -0700)]
SHA: SHA test cases
32bit and 64bit test asm files.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 02:44:03 +0000 (19:44 -0700)]
SHA: Add SHA instructions
New instruction extensions of SHA family are added.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 02:42:41 +0000 (19:42 -0700)]
MPX: Add test cases for MPX
MPX test asm files are added. These include all three different styles of
mib syntax (NASM, ICC and gas).
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 02:38:51 +0000 (19:38 -0700)]
MPX: Add BND prefix for branch instructions
BND prefix is used for adding bounds checking protection
across flow control changes such as call, ret, jmp and jcc calls.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Wed, 16 Oct 2013 02:10:13 +0000 (19:10 -0700)]
MPX: Add MPX instructions
Added MPX instructions and corresponding parser and encoder.
ICC style mib - base + disp and index are separate - is supported.
E.g. bndstx [ebx+3], bnd2, edx -> ebx+3 : base+disp, edx : index
As a supplement to NASM style mib - split EA - parser,
omitted base+disp is now treated as 0 displacement.
E.g. bndstx [,edx], bnd2 -> bndstx [0,edx], bnd2
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Fri, 27 Sep 2013 23:53:51 +0000 (16:53 -0700)]
disasm: add support for emitting split EA format
Add support for emitting split EA format in the disassembler,
indicated by the MIB instruction flag.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Fri, 27 Sep 2013 23:39:16 +0000 (16:39 -0700)]
parser: support split base,index effective address
Mostly intended for the "mib" expressions in BNDLDX/BNDSTX.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Fri, 27 Sep 2013 00:28:39 +0000 (17:28 -0700)]
parse: factor out mref parsing
Factor out the parsing of an mref from parse_line(). In order to
support the [base,index] syntax we need to be able to parse an mref in
two steps.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
H. Peter Anvin [Fri, 27 Sep 2013 00:23:08 +0000 (17:23 -0700)]
parser: simplify code by keeping a pointer to the current operand
There is an awful lot of syntax-heavy "result->oprs[operand]" going
on; simplify by making a pointer to the current operand.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Jin Kyu Song [Sat, 21 Sep 2013 03:52:11 +0000 (20:52 -0700)]
AVX-512: Add perfomtest-compliant headers to test cases
test/avx512*.asm files are now tested by using perfomtest.pl
Refer to pefomtest help message for the usage.
Signed-off-by: Jin Kyu Song <jin.kyu.song@intel.com>
Cyrill Gorcunov [Sat, 9 Nov 2013 08:02:15 +0000 (12:02 +0400)]
BR3392270: preproc: Handle all token chains in mmacro params range
A typical example is
|
| %macro m0 0-*
| %rep %0
| m0 arg is %1
| %rotate 1
| %endrep
| %endmacro
|
| %macro m1 0-*
| m0 %{1:-1}
| %endmacro
|
| m1 a=b, c=d
If passed with nasm -E the output must be like
m0 arg is a=b
m0 arg is c=d
http://bugzilla.nasm.us/show_bug.cgi?id=3392270
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>