For PASS1 warnings, only do them when pass0 == 1. The prior passes
are to be considered training passes. This is a bit awkward if we
then hit an error, but it's better than n repeated warnings.
The five-pass-minimum was a hack for a bug which I think is identified
now. Doesn't really change the fact that if you want the optimizer,
you probably want -Ox.
We have a number of bug reports about things not working properly when
the optimizer is running out of passes. I suspect the reason is
simply that we don't properly execute the final passes (pass0 = 1, 2)
when hitting the limit. Make sure we advance pass0 the last few
times.
Avoid redundant error messages:
./nasm
nasm: error: no input file specified
nasm: fatal: file `' is both input and output file
type `nasm -h' for help
... which is more than a wee bit confusing to the user.
Add gcc-style -Wxxx -Wno-xxx warning selection as an alternative to
-w+xxx/-w-xxx.
Add "all" as an alias for all (actual) warnings.
Add "error" to treat warnings as errors.
Address data is always int64_t even if the size itself is smaller;
this was broken on bigendian hosts (still need testing!)
Create simple "write sized object" macros.
Actually generate the appropriate floating-point warnings, and only
one per assembly, pretty please.
Correct the round-to-overflow condition; as written all numbers with a
positive exponent were considered overflows!
Proper use of bool and enum makes code easier to debug. Do more of
it. In particular, we really should stomp out any residual uses of
magic constants that aren't enums or, in some cases, even #defines.
Both C and C++ have "bool", "true" and "false" in lower case; C
requires <stdbool.h> for this, in C++ it is an inherent type built
into the compiler. Use those instead of the old macros; emulate with
a simple typedef enum if unavailable.
SAA's were never intended to allow random access, but several backends
do random or semirandom access via saa_fread() and saa_fwrite()
anyway. Rewrite the SAA system to allow for efficient random access.
On "label.pl 10000000" this improves performance by a factor of 12.
Change cloc_t to struct location, and reorder the members so that it
should fit in 16 bytes instead of needing 8 bytes of extra padding on
64-bit machines.
Change loc_t to cloc_t to avoid AIX conflict.
We really shouldn't use _t names at all; they are usually considered
platform types, but worry about that later.
Concentrate compiler dependencies to compiler.h; make sure compiler.h
is included first in every .c file (since some prototypes may depend
on the presence of feature request macros.)
Actually use the conditional inclusion of various functions (totally
broken in previous releases.)
Implement the -MG option, to generate dependencies in the presence of
generated files. In the end, we probably need to support the full
gamut of GCC-like dependency-generation options.
Finish the perfect hash tokenizer, and actually enable it.
Move stdscan() et al to a separate file, since it's not needed in any
of the clients of nasmlib other than nasm itself.
Run make alldeps.
Implement "REL" and "ABS" modifiers for offsets in 64-bit mode. This
replaces "rip+XXX" type addressing. The infrastructure to set the default
mode is there, but there is nothing to throw the switch just yet.
- MOV gpr,CRx or MOV CRx,gpr can access high control registers with a LOCK
prefix; handle that in both the assembler and disassembler.
- Get a saner error message when trying to access high resources in
non-64-bit mode.
- Remove obsolete types like "uint32"; use "uint32_t" consistently.
- Make sure we include <inttypes.h> where needed.
- Header file guards should be FOO_H or SUBDIR_FOO_H; _FOO_H infringes
on the C implementation's namespace and should only be used when
writing libc!
- Change a few "int8_t" back to "char" where appropriate. There are
a lot more places where that should be done, though.
- Clean up the check for getuid/getgid in rdoff/rdlar.h.
