libyasm/md5.c:166:31: warning: argument to 'sizeof' in 'memset' call is the same expression as the destination; did you mean to dereference it? [-Wsizeof-pointer-memaccess]
memset(ctx, 0, sizeof(ctx)); /* In case it's sensitive */
~~~ ^~~
Due to the svn import structure, a special case is currently implemented
to look for the 1.1.0 branchpoint instead of the most recent tag on the
master branch. This will be removed after the first release is tagged
on the master branch in git.
Specific details:
autogen.sh: More aggressively clean autoconf cache.
This is needed to ensure the version number is actually regenerated.
Don't generate PACKAGE_PATCHLEVEL or PACKAGE_BUILD variables. The
genversion program now parses PACKAGE_VERSION directly.
For Mkfiles builds, YASM-VERSION.h is generated and included by the custom
config.h. This avoids the need to edit config.h for versioning.
dwarf2_loc.discriminator wasn't being initialized by
dwarf2_generate_line_section, leading to incorrect output, including
intermittent test failures in gen64/dwarf64_pathname.
This allows signaling to the caller (expr_level_op) that the int_term
was destroyed. Without this, the new expr-simplify-identity testcase
has a use-after-free.
[#232 state:resolved]
These instructions use "VSIB" encoding, which takes the place of the
usual SIB encoding. Several tests cover various legal and illegal
modes.
Last part of [#227 state:resolved].
Reference: http://www.intel.com/software/avx rev11 spec
This is all AVX2 instructions except for VGATHER*/VPGATHER*, which
require additional ModRM handling.
Portions contributed by: Mark Charney <mark.charney@intel.com>
Part of [#227].
Reference: http://www.intel.com/software/avx rev11 spec
Also add appropriate CPU bits and directive handling for these.
Currently we have no good way of handling an "or" of instruction bits
(in this case needed for LZCNT, where it's either AMD or LZCNT). For
now, make it LZCNT only.
Contributed by: Mark Charney <mark.charney@intel.com>
Part of [#227].
Previously a line such as "times 4 mov rax, [rel foobar]" would result
in incorrect relocations being generated.
Patch by: bird-yasm@anduin.net
[#211 state:resolved]
As dwarf2 has a directory table in addition to a filename table, we split
the provided pathname when generating the filename table. This correctly
combined both "./foo" and "foo" into a single filename table entry.
However, we were only matching on pathname instead of dir+filename when
actually generating the dwarf opcodes, resulting in not finding the "./foo"
version. Fix to check both dir and filename separately, splitting the same
way we do when generating the filename table.
svn path=/trunk/yasm/; revision=2367
- Add support for deferred symbols (those which depend on other symbols
that have not yet been .set)
- Add support for re-defining symbols
- Add support for 0x hex notation in literals
Contributed by: Alexei Svitkine <alexei.svitkine@gmail.com>
svn path=/trunk/yasm/; revision=2366
Previously this was implemented the opposite way, contrary to its
documentation. This resulted in clearing any already set yasm_value
relative fields.
svn path=/trunk/yasm/; revision=2364
The register versions worked okay due to size matching, but the memory
versions relied on suffix matching, which wasn't being generated correctly.
Reported by: Tony Goelz <cag@absoft.com>
svn path=/trunk/yasm/; revision=2358
- strcpy() was being used with overlapping memory ranges; switched to memmove().
- bline->line_number was not set in one location.
Exact causes identified using valgrind.
svn path=/trunk/yasm/; revision=2348