[ANNOUNCE] Git 1.7.0.2

[ANNOUNCE] Git 1.7.0.2

[Junio C Hamano]

The latest maintenance release Git 1.7.0.2 is available at the
usual places:

  http://www.kernel.org/pub/software/scm/git/

  git-1.7.0.2.tar.{gz,bz2}			(source tarball)
  git-htmldocs-1.7.0.2.tar.{gz,bz2}		(preformatted docs)
  git-manpages-1.7.0.2.tar.{gz,bz2}		(preformatted docs)

The RPM binary packages for a few architectures are found in:

  RPMS/$arch/git-*-1.7.0.2-1.fc11.$arch.rpm	(RPM)

Git v1.7.0.2 Release Notes
==========================

Fixes since v1.7.0.1
--------------------

 * GIT_PAGER was not honored consistently by some scripted Porcelains, most
   notably "git am".

 * updating working tree files after telling git to add them to the
   index and while it is still working created garbage object files in
   the repository without diagnosing it as an error.

 * "git bisect -- pathspec..." did not diagnose an error condition properly when
   the simplification with given pathspec made the history empty.

 * "git rev-list --cherry-pick A...B" now has an obvious optimization when the
   histories haven't diverged (i.e. when one end is an ancestor of the other).

 * "git diff --quiet -w" did not work as expected.

 * "git fast-import" didn't work with a large input, as it lacked support
   for producing the pack index in v2 format.

 * "git imap-send" didn't use CRLF line endings over the imap protocol
   when storing its payload to the draft box, violating RFC 3501.

 * "git log --format='%w(x,y,z)%b'" and friends that rewrap message
   has been optimized for utf-8 payload.

 * Error messages generated on the receiving end did not come back to "git
   push".

 * "git status" in 1.7.0 lacked the optimization we used to have in 1.6.X series
   to speed up scanning of large working tree.

 * "gitweb" did not diagnose parsing errors properly while reading tis configuration
   file.

And other minor fixes and documentation updates.

----------------------------------------------------------------

Changes since v1.7.0.1 are as follows:

Bert Wesarg (1):
      unset GREP_OPTIONS in test-lib.sh

Christian Couder (1):
      bisect: error out when passing bad path parameters

Dmitry Potapov (1):
      hash-object: don't use mmap() for small files

Erik Faye-Lund (1):
      run-command: support custom fd-set in async

Gabriel Filion (1):
      require_work_tree broken with NONGIT_OK

Hitoshi Mitake (1):
      git-imap-send: Convert LF to CRLF before storing patch to draft box

Jakub Narebski (2):
      gitweb: Die if there are parsing errors in config file
      gitweb: Fix project-specific feature override behavior

Jonathan Nieder (8):
      Fix 'git var' usage synopsis
      Make 'git var GIT_PAGER' always print the configured pager
      git.1: Clarify the behavior of the --paginate option
      git svn: Fix launching of pager
      am: Fix launching of pager
      tests: Add tests for automatic use of pager
      t7006-pager: if stdout is not a terminal, make a new one
      tests: Fix race condition in t7006-pager

Junio C Hamano (6):
      t6000lib: Fix permission
      lib-patch-mode.sh: Fix permission
      status: preload index to optimize lstat(2) calls
      Start preparing for 1.7.0.2
      Update draft release notes to 1.7.0.2
      Git 1.7.0.2

Larry D'Anna (2):
      git diff --quiet -w: check and report the status
      git-diff: add a test for git diff --quiet -w

Matthieu Moy (1):
      stash: suggest the correct command line for unknown options.

Michal Sojka (1):
      Remove extra '-' from git-am(1)

Nicolas Pitre (8):
      fast-import: start using struct pack_idx_entry
      fast-import: use sha1write() for pack data
      fast-import: use write_idx_file() instead of custom code
      fast-import: make default pack size unlimited
      fast-import: honor pack.indexversion and pack.packsizelimit config vars
      fast-import: use the diff_delta() max_delta_size argument
      sha1_file: don't malloc the whole compressed result when writing out objects
      sha1_file: be paranoid when creating loose objects

René Scharfe (4):
      utf8.c: remove print_wrapped_text()
      utf8.c: remove print_spaces()
      utf8.c: remove strbuf_write()
      utf8.c: speculatively assume utf-8 in strbuf_add_wrapped_text()

Shawn O. Pearce (7):
      run-command: Allow stderr to be a caller supplied pipe
      send-pack: demultiplex a sideband stream with status data
      receive-pack: Refactor how capabilities are shown to the client
      receive-pack: Wrap status reports inside side-band-64k
      receive-pack: Send hook output over side band #2
      t5401: Use a bare repository for the remote peer
      receive-pack: Send internal errors over side-band #2

Stephen Boyd (1):
      t7406: Fix submodule init config tests

Sylvain Rabot (1):
      gitweb multiple project roots documentation

Tay Ray Chuan (1):
      transport: add got_remote_refs flag

Thomas Rast (1):
      cherry_pick_list: quit early if one side is empty

Re: [ANNOUNCE] Git 1.7.0.2

[Mike Galbraith]

On Sun, 2010-03-07 at 14:29 -0800, Junio C Hamano wrote:

>  * "git bisect -- pathspec..." did not diagnose an error condition properly when
>    the simplification with given pathspec made the history empty.

Would that be what led to this somewhat less than perfect tip bisection
result?  I had to repeat, leaving off the pathspec, in order to bisect
to the troublesome commit.

b4dad12cdd22f61eb093d23f11a7b09620e53aba is the first bad commit

git bisect start '--' 'tools/perf'
# good: [64ba9926759792cf7b95f823402e2781edd1b5d4] Merge branch 'for-linus' of git://git.open-osd.org/linux-open-osd
git bisect good 64ba9926759792cf7b95f823402e2781edd1b5d4
# bad: [a8313ee914acb061cacb44d0884e448760e33392] Merge branch 'perf/pebs'
git bisect bad a8313ee914acb061cacb44d0884e448760e33392
# good: [eaa5eec739637f32f8733d528ff0b94fd62b1214] Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp
git bisect good eaa5eec739637f32f8733d528ff0b94fd62b1214
# bad: [b9377fc0bf1cd0b7b494337eacf8ccf0f7b52e1f] Merge branch 'perf/probes'
git bisect bad b9377fc0bf1cd0b7b494337eacf8ccf0f7b52e1f
# bad: [318794793ce30dc7000ed6992f4f81f308ae3958] manual merge of linus
git bisect bad 318794793ce30dc7000ed6992f4f81f308ae3958
# good: [617f757a1b45395ca3b11b162138ce514afc5b29] Merge branch 'perf/core'
git bisect good 617f757a1b45395ca3b11b162138ce514afc5b29
# good: [c93025db5f6d093fd00e96fde8e47d5c791a03fd] Merge branch 'perf/core'
git bisect good c93025db5f6d093fd00e96fde8e47d5c791a03fd
# good: [eb3addcb8fe9b81f47ad0f14ed0ae220f18ae83b] Merge branch 'perf/urgent'
git bisect good eb3addcb8fe9b81f47ad0f14ed0ae220f18ae83b
# bad: [b4dad12cdd22f61eb093d23f11a7b09620e53aba] Merge branch 'core/iommu'
git bisect bad b4dad12cdd22f61eb093d23f11a7b09620e53aba

Re: [ANNOUNCE] Git 1.7.0.2

[Junio C Hamano]

Mike Galbraith <efault@gmx.de> writes:

> On Sun, 2010-03-07 at 14:29 -0800, Junio C Hamano wrote:
>
>>  * "git bisect -- pathspec..." did not diagnose an error condition properly when
>>    the simplification with given pathspec made the history empty.
>
> Would that be what led to this somewhat less than perfect tip bisection
> result?

The bugfix was about the case where there was no change between the bad
and good commit with respect to the given pathspec, so I suspect your case
was irrelevant.

Your case probably was coming from that you originally suspected one of
the commits that touch tools/perf was a culprit, but because that initial
suspicion was wrong, bisection didn't find anything useful.

Re: [ANNOUNCE] Git 1.7.0.2

[Mike Galbraith]

On Sun, 2010-03-07 at 21:13 -0800, Junio C Hamano wrote:
> Mike Galbraith <efault@gmx.de> writes:
> 
> > On Sun, 2010-03-07 at 14:29 -0800, Junio C Hamano wrote:
> >
> >>  * "git bisect -- pathspec..." did not diagnose an error condition properly when
> >>    the simplification with given pathspec made the history empty.
> >
> > Would that be what led to this somewhat less than perfect tip bisection
> > result?
> 
> The bugfix was about the case where there was no change between the bad
> and good commit with respect to the given pathspec, so I suspect your case
> was irrelevant.

Oh, ok.  Thanks.

> Your case probably was coming from that you originally suspected one of
> the commits that touch tools/perf was a culprit, but because that initial
> suspicion was wrong, bisection didn't find anything useful.

Hm, the successfully bisected and verified culprit was pure tools/perf.

marge:..git/linux-2.6 # git show 804b36068eccd8163ccea420c662fb5d1a21b141|grep diff
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
diff --git a/tools/perf/builtin-probe.c b/tools/perf/builtin-probe.c
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h

Oh well.  Something went wonky.. I'll survive :)

	-Mike