{dfv=} is basically a constant (immediate). Treat it as such during
parsing, except that if "naked" (not in an expression), it has special
matching properties and does not need a terminal comma.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
If TOKEN_BRCCONST are used in an expression (including simply wrapping
them in parentheses), then just treat them as integers. This makes
things like ({dfv=cf}|{dfv=of}) work.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Change the handling of {dfv=} to a more general "braced constant"
expression, to be tagged with an instruction flag to make sure they
match the instruction in question.
This really ought to be an operand flag, but the opflags are precious;
as the CCMP/CTEST instructions can also take an immediate it probably
is necessary to invent a "special immediate" operand type that can
fold these together.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The {dfv=} prefix sequences for the CCMP and CTEST instructions need
special handling in the parser. This means a fair bit of new magic in
the handler of the parser, but it just adds to the fun.
Try to make this as general as possible, so we can use it for other
things.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The parser state does not just necessarily include the position of the
buffer, but make it possible to maintain additional state.
Furthermore, add an explicit ability to push back a token.
All of this might make it easier at some point in the future to keep
track of horizontal position, although that will require lots of
changes to the preprocessor.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Clean up the handling of prefixes in general. Allow a set of braced
prefixes to follow the instruction; this is required for things like
{dfv=} but might also be a nicer syntax for things like {rex}.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
At least three files (asm/assemble.c, disasm/disasm.c, and
x86/insns.pl) depend on the bytecode defintions. It makes a lot more
sense for them to live in an explicit documentation file in the x86/
directory.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The help output has gotten way too long to be shown on a single
command line. It can of course be piped to a pager, but to be a little
nicer to the user, break it up into subtopics that can be individually
displayed. --help all (-h all) can still show all the help information
as a single data dump.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Try to better sort out the necessary dependencies for warning
generation.
Fix regex for cleaning up nasmdoc markup: nasmdoc markup does not
nest, although it may include \} sequences.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
We should always support up to 8 characters, i.e. 64 bits, in a
string-to-numeric conversion.
Reported-by: Aleksandras Krupica <vaikutisasa@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Require the second colon before the grouped parameter count; otherwise
the syntax is ambiguous since an expression can start with (.
Update/complete the documentation and the examples.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add the ability to have fixed arguments in %map. This is extremely
useful for parameterizing the invoked macro using arguments to a
surrounding macro.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Separate out counting and parsing smacro parameters into separate
functions. This not only makes the code *way* easier to read, but
these can be re-used e.g. for %map().
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add the %map() function which can apply arguments to a macro from a
list.
Allow the user to specify the desired radix for an evaluated
parameter. It doesn't make any direct difference, but can be nice for
debugging or turning into strings.
As part of this, split expand_one_smacro() into two parts: parameter
parsing and macro expansion. This is a very straightforward splitting
of two mostly unrelated pieces of functionality.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
When expanding %rep blocks, if any of the %rep blocks are empty, there
may be need to unwind the %rep stack multiple times. The code would
not do so -- there was a break; in the loop, which incidentally turned
it into something that wasn't a loop at all.
Reported-by: E. C. Maslock <pushbx@ulukai.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Make it possible to add a base prefix to %num().
Add the %hex() function, producing hexadecimal values that are
nevertheless valid NASM numeric constants.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The user would generally expect the parameter number to be counted
from 1 for human purposes, and that is also consistent with %1, %2,
... for multi-line macros.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
"LOCK XCHG reg,mem" would issue a warning for being unlockable, which
is incorrect. In this case the RM encoding is simply an alias for the
MR encoding. Add a "LOCK1" bit to deal with that.
However, XCHG is *always* locked, so create a new warning to
explicitly flag a user-specified LOCK XCHG; default off.
Consider optimizing that prefix away in the future, but for now, let's
stick to the user-requested code sequence.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The dependency list needs to be updated for all passes, not just the
last one. The dependency list is already uniquized, so it doesn't
cause problems with multiple entires.
The reasons it needs to be done for all passes is first of all that an
%include could be pass-dependent, and secondly that we only record a
dependency for an %include or %require for the first occurrence of
that file, when pathnames are resolved.
Reported-by <michael@mehlich@com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The fix for BR 3392414 introduced a fairly serious memory
leak. C. Masloch was kind enough to track down the proper root cause
and fix it correctly.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Some target environments may have specific restrictions on what kinds
of relocations are possible or allowed. Allow users to opt-in to
specific warnings as to the relocations they cannot support.
Requested-by: C. Masloch <pushbx@ulukai.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
NASM would try to "eat the comma token" in db expressions, even for
cases where the token was not a comma. Fix that and error out
properly.
To give better error messages, track where in the input string a token
starts or ends. This information is only valid as long as the input
string is kept, but that is just fine for error messages during
parsing.
Reported-by: Peter Cordes <pcordes@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The dependency on the warning files breaks when we are building in a
directory *and* the files already exist from being shipped with the
distribution tarballs. The make VPATH simply isn't sophisticated
enough to deal with it, so let the C compiler handle it by #including
the generated file from a dummy C file.
Reported-by: Rudi Heitbaum <rudi@heitbaum.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Document CPU LATEVEX and the associated prefixes; add CPU EVEX and CPU
VEX flags to further control encodings.
Fix the error message for invalid encodings due to flags.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
For VEX instructions created *after* the corresponding EVEX
instructions, we need the user to either explicitly declare them {vex}
or specifying "cpu latevex".
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
"Why dup_tlist() here? We should own it."
Yes, we own it, but we still need to advance the tail pointer. Create
steal_tlist() for this purpose.
Fixes: https://bugzilla.nasm.us/show_bug.cgi?id=3392774
Reported-and-Debugged-by: C. Masloch <pushbx@ulukai.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Broken %if, %rep and %macro nesting can result in the %exitmacro
unwind overrunning the condition stack. Fix.
Fixes: https://bugzilla.nasm.us/show_bug.cgi?id=3392796
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The argument to nasm_new() is the pointer, not the indirection from
the pointer. This code is only relevant when compiled without token
recycling (TOKEN_BLOCKSIZE not set), but it is still wrong...
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Fix the expansion of the %00 mmacro parameter; broken due to a missing
change of the token type.
Fixes: https://bugzilla.nasm.us/show_bug.cgi?id=3392803
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add the option of having strings only conditionally quoted (&&) -- do
not quote an already quoted string again -- as opposed to always
quoting a string.
This makes a lot of the string functions way simpler to implement, and
removes the need to share ad hoc parsing code with directives.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add the %abs() function, to produce the absolute value as an
always-positive decimal constant.
Change the order of the arguments for %num().
Refactor the handling of optional arguments, to reduce the amount of
redundant code. This is currently only used for builtin functions, but
might be extended in the future.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Add the %num() preprocessor function, which returns a quoted string
with a number formatted in any base between 2 and 64 (using bash
encoding with '@' for 62 and '_' for 63.)
It can specify a fixed number of digits with or without truncation.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Classify all remaining WARN_OTHER warnings in the preprocessor. Move
all preprocessor warnings except "user" under a common pp-* prefix.
Warn for an out-of-range argument to the %sel() function.
Finally, use "dname" in additional places for consistency and future
ease of use.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Implement the %cond() and %sel() functions that expand to a specific
one of the arguments. %cond(x,y,z) is basically a shorthand for
%sel(2-!(x),y,z) used when x is a boolean condition.
Fix a memory leak in %strcat and %strlen.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Allow preprocessor function expansion to recurse. Nearly all the
machinery for recursive smacros was already in place; this merely
activates it for the specific case of preprocessor functions. Making
it a general facility should be deferred to a later relese, though.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Instead of handling conditional instructions ad hoc, generate
individual instruction patterns as normal. This simplifies the code
and makes CMPccXADD support simpler (otherwise it would be necessary
to hack in the handling of a condition code in the middle of an
instruction.)
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Some preprocessor functions have the same name as directives. In those
cases, they should be expanded as functions if and only if they are
followed by a left parenthesis. Although it is not inherently true that
either preprocessor functions require a paren nor that directives
cannot start with one, but it is true and will remain true for all
cases where there is a namespace collision.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
%substr contained a token skip to "skip expanded ID", which is
incorrect, as that has already been skipped at that point. It worked
anyway, accidentally, as this token would always be a whitespace token
-- but we then do skip_white() immediately thereafter.
Delete this to allow this code to be factored.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
