For EVEX instructions, /is4 can contain a fifth register bit, encoded
in bit 3 of the imm8. Properly generate this case, and simplifiy the
/is4 generation code somewhat.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Improve consistency by allowing contracted forms for EVEX-encoded
instructions when it's allowed for similar VEX-encoded instructions.
Previously the behavior would change depending on the vector size or
the register number which could be somewhat confusing:
vaddps xmm0, xmm1 ; ok
vaddps ymm0, ymm1 ; ok
vaddps zmm0, zmm1 ; error
vaddps xmm0, xmm16 ; error
Also allow contracted forms for a few additional older AVX instructions
where it makes sense.
Signed-off-by: Henrik Gramner <henrik@gramner.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
write_symbolinfo_properties didn't match the S_COMPILE2 record it's
supposed to be writing (the "compiler version" string was emitted
starting in the final "version" field); fix that.
Write version 8.0.50727; the Windows App Certification Kit (WACK)
checks compiler versions as given in app debug info and complains
when the toolchain is too old. 8.0.50727 is the lowest permitted
"MASM" version for WACK to be happy, so that's what we write.
Signed-off-by: Fabian Giesen <fabiang@radgametools.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
On host platforms where $(LN_S) is cp, need to make sure to
link rdf2bin first.
Backport of 6d67dbfa95
Signed-off-by: Fabian Giesen <fabiang@radgametools.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
When symbols are created turing non-first pass we
should at least warn a user since it's a sign of
potential problem.
http://bugzilla.nasm.us/show_bug.cgi?id=3392358
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
In rdoff/rdoff.c, the function declaration for newmembuf is written:
static memorybuf *newmembuf()
The attached diff adds an explicit void parameter.
It seems like it would be nicer with an explicit void parameter
especially since the -Werror build options seem to want to find such
things.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
It is not just DJGPP which has broken header files if __STRICT_ANSI__
is defined. Cygwin has also been confirmed to have problems. It
would be somewhat different if configure didn't still detect those
functions.
Undefine __STRICT_ANSI__ for all gcc platforms. If this breaks then,
well, we'll have to really narrow down the problems.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
In case if there is no environment variable present
we allocated empty string but when working with tokens
we test for second byte for special symbols, accessing
out of memory address (->text[1] for the reference).
http://bugzilla.nasm.us/show_bug.cgi?id=3392333
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
The codeview backend needs to be able to open each source file passed in
so that it can calculate its checksum. In order to take into account
include paths, this patch updates the filename to include the path where
the file was found.
Signed-off-by: Jim Kukunas <james.t.kukunas@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
[ hpa: resolved one conflict in preproc.c ]
This unbreaks checkin 84f6860ed5, which
was broken due to a transcription error of mine. Zenith432 was
faultless in this case.
This fixes bug report 3392355.
Reported-by: Zenith432 <zenith432@users.sourceforge.net>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
For 16-bit and 32-bit x86 code, the size and realsize() always
matches as only jumps, calls and loops uses PC relative
addressing and the address isn't followed by any other opcode
bytes. In 64-bit mode there is RIP relative addressing which
means the fixup location can be followed by an immediate value,
meaning that size > realsize().
When the CPU is calculating the effective address, it takes the
RIP at the end of the instruction and adds the fixed up relative
address value to it.
The linker's point of reference is the end of the fixup location
(which is the end of the instruction for Jcc, CALL, LOOP[cc]).
It is calculating distance between the target symbol and the end
of the fixup location, and add this to the displacement value we
are calculating here and storing at the fixup location.
To get the right effect, we need to _reduce_ the displacement
value by the number of bytes following the fixup.
Example:
data at address 0x100; REL4ADR at 0x050, 4 byte immediate,
end of fixup at 0x054, end of instruction at 0x058.
=> size = 8.
=> realsize() -> 4
=> CPU needs a value of: 0x100 - 0x058 = 0x0a8
=> linker/loader will add: 0x100 - 0x054 = 0x0ac
=> We must add an addend of -4.
=> realsize() - size = -4.
The code used to do size - realsize() at least since v0.90,
probably because it wasn't needed...
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Concatenating the cwd with the name of the output file is incorrect
for filenames which are specified as absolute. We already have
nasm_realpath() for this purpose, use it.
Cc: Jim Kukunas <james.t.kukunas@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
There is no reason why the warning-generating ilog2 has to be only the
floor variant. However, I am pretty sure we can simply implement the
ilog2cw() as a macro only; we can always fix that if that turns out to
be incorrect.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Instead of walking a linear list of files for every line, make a
simple comparison for the common case of the same file, and otherwise
use a hash table.
Cc: Jim Kukunas <james.t.kukunas@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
This essentially reverts 6503051dcc since
that workaround is no longer needed thanks to support for multiple source
files
Signed-off-by: Jim Kukunas <james.t.kukunas@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Handle the existence of multiple source files, as is normal when using
include files.
Signed-of-by: Jim Kukunas <james.t.kukunas@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
clang doesn't error out on unknown -W options unless
-Werror=unknown-warning-option is specified first, so do that so the
configure script can do its job.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The code to handle building in a separate directory had seriously
bitrotted. This contains a number of fixes to make it possible,
including bits like the documentation which never worked in the past.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
The old code for keeping track of source file name and line was
confused as hell about ownership of the strings, and it is pretty
clear we leaked that information all over the place.
Instead, use a hash table to keep a copy of each string as necessary,
and simply make references to a string pool that we keep until the end
of the assembly session.
This pool probably should be unified with the list of dependency
files, and so on, but that is for the development branch.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Whereas nasm_unquote() unquotes the argument in place, nasm_quote()
has to allocate a new string (since the new string will *always* be
longer than the old string!) Make the old string const since we're
making a copy anyway.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Previously, debug info would refer to the first file seen, even
when it did not actually generate line numbers (e.g. segto=-1).
Fix it so we only lock in the file name the first time we actually
produce a line number record. Not as good as proper support for
debug info referencing multiple source files but much more useful
than the current behavior.
Signed-off-by: Fabian Giesen <fabiang@radgametools.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
When assembling on Windows machines with CRLF line endings, computing
the MD5 hash from the file read in "text" mode (transforms CRLF->LF)
gives incorrect results.
Signed-off-by: Fabian Giesen <fabiang@radgametools.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
The hash calculation in calc_md5 tries to open the source file via
"filename" again. For %includes, this is the file name that was
specified in the %include directive, not the actual name of the file
that was opened by the preprocessor. In other words, this fails if the
include file is not in the current working directory.
Add pp_input_fopen that uses the preprocessor include path lookup
code to resolve a file name and open it, and use that in codeview.c.
Signed-off-by: Fabian Giesen <fabiang@radgametools.com>
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Fix the printing of the macro stack: we need to follow the
mstk->next_active list, not mstk->next, and we need to reverse the
order so that the highest-level inclusion comes first.
Since this should be a rare or at least performance-insensitive
operation, do it using simple function recursion.
Finally, add an ellipsis before the "from macro" message; it greatly
enhances readability.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
It can be hard to find errors inside potentially nested macros.
Show the mmacro expansion stack when printing diagnostics.
Note that a list file doesn't help for errors that are detected
before the code-generation pass.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
If we get an invalid warning name passed to the -w/-W option or the
[warning] directive, ignore it. This may be a warning name enabled in
a future version of NASM, and it is rather pointless to error out on
it.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Issue a specific suppressible warning if we encounter the PTR keyword.
This usually indicates someone mistakenly using MASM syntax in NASM.
This introduces a generic infrastructure for issuing warnings for such
keywords.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Fix pass1 warnings so they actually display.
When issuing suppressible warnings, display the option that controls
them, as gcc has been doing for a while.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
