Sparse 0.1 released
I have tagged and tarballed a 0.1 release of Sparse, now available from http://ftp.be.debian.org/pub/software/devel/sparse/dist/, with sha1sum 9e0a4d5abb8e8a4be4cf8d9fe632c69dbec3e242.
As discussed in Re: Official releases?, I've taken maintainership of sparse. Thanks to Linus Torvalds for his previous maintainership. As a result, this release comes from my sparse Git repository. You can find more information about obtaining sparse via Git at the new sparse homepage.
In addition to all the work in the previous Sparse repository, this release includes the following changes:
Adam DiCarlo (1): Add type information to enum mismatch warning Al Viro (2): added a bunch of gcc builtins switch to hash-based get_one_special() Josh Triplett (15): "Initializer entry defined twice" should not trigger with zero-size fields Fix incorrect symbol in comment on #endif for multiple-inclusion guard Add -Wno-uninitialized graph: Show position in basic block nodes bb_terminated: Use boundary values rather than specific opcodes Turn on -Wcontext by default Merge branch 'fix-defined-twice-error-on-empty-struct' into staging Merge branch 'graph' into staging merge branch 'more-warning-flags' into staging and fix conflicts merge branch 'no-semantic-h' into staging and fix conflicts Merge branch 'Wcontext-default' into staging Add test cases to validation/context.c for the Linux __cond_lock macro Merge branch 'context-test-cases-for-cond-lock' into josh Rename test case bad-assignement.c to bad-assignment.c, fixing the typo. Stop building and installing libsparse.so Josh Triplett and Pavel Roskin (1): Recognize and ignore __alias__ and __visibility__ Pavel Roskin (4): Compile sparse executable under it's own name, not as "check" Add support for __builtin_strpbrk() Typo fixes Install cgcc on "make install", refactor installation code
Known issue with this release:
- Sparse does not produce the expected set of warnings for several of the validation programs, included in the sparse source in the directory validation/. Some scripts should provoke warnings but don't, and others provoke warnings they shouldn't.
-- Josh Triplett