1 Git v1.7.7.1 Release Notes
2 ==========================
7 * On some BSD systems, adding +s bit on directories is detrimental
8 (it is not necessary on BSD to begin with). The installation
9 procedure has been updated to take this into account.
11 * After incorrectly written third-party tools store a tag object in
12 HEAD, git diagnosed it as a repository corruption and refused to
13 proceed in order to avoid spreading the damage. We now gracefully
14 recover from such a situation by pretending as if the commit that
15 is pointed at by the tag were in HEAD.
17 * "git apply --whitespace=error" did not bother to report the exact
18 line number in the patch that introduced new blank lines at the end
21 * "git apply --index" did not check corrupted patch.
23 * "git checkout $tree $directory/" resurrected paths locally removed or
24 modified only in the working tree in $directory/ that did not appear
25 in $directory of the given $tree. They should have been kept intact.
27 * "git diff $tree $path" used to apply the pathspec at the output stage,
28 reading the whole tree, wasting resources.
30 * The code to check for updated submodules during a "git fetch" of the
31 superproject had an unnecessary quadratic loop.
33 * "git fetch" from a large bundle did not enable the progress output.
35 * When "git fsck --lost-and-found" found that an empty blob object in the
36 object store is unreachable, it incorrectly reported an error after
37 writing the lost blob out successfully.
39 * "git filter-branch" did not refresh the index before checking that the
40 working tree was clean.
42 * "git grep $tree" when run with multiple threads had an unsafe access to
43 the object database that should have been protected with mutex.
45 * The "--ancestry-path" option to "git log" and friends misbehaved in a
46 history with complex criss-cross merges and showed an uninteresting
49 * Test t1304 assumed LOGNAME is always set, which may not be true on
52 * Tests with --valgrind failed to find "mergetool" scriptlets.
54 * "git patch-id" miscomputed the patch-id in a patch that has a line longer
57 * When an "exec" insn failed after modifying the index and/or the working
58 tree during "rebase -i", we now check and warn that the changes need to