Re: BUG: git log: fatal: internal error in diff-resolve-rename-copy

From: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Date: Aug 13, 2010, 13:38 UTC
Message-ID: <AANLkTikGE-m=Tht2tNbP5=Bej78284_ouLBfdk_aEDSO@mail.gmail.com>
In-Reply-To: <AANLkTikPhWgeeLBV3dbLZ5UM3UDmkOmpqrmwqPmGfn7Z@mail.gmail.com>

On Fri, Aug 13, 2010 at 11:25, Constantine Plotnikov <constantine.plotnikov@gmail.com> wrote:

Show 5 quoted lines

> Somewhere between the git 1.7.0.2 and the git 1.7.2.0 the rename
> detection started to fail with fatal error on some files in our
> repository. The bug could be seen on the public IntelliJ IDEA
> repository (about 760M in size), but our users have reported it as
> well.

>From a bisect:

    1da6175d438a9849db07a68326ee05f291510074 is the first bad commit
    commit 1da6175d438a9849db07a68326ee05f291510074
    Author: Bo Yang <struggleyb.nku@gmail.com>
    Date:   Thu May 6 21:52:28 2010 -0700

        Make diffcore_std only can run once before a diff_flush

        When file renames/copies detection is turned on, the
        second diffcore_std will degrade a 'C' pair to a 'R' pair.

        And this may happen when we run 'git log --follow' with
        hard copies finding. That is, the try_to_follow_renames()
        will run diffcore_std to find the copies, and then
        'git log' will issue another diffcore_std, which will reduce
        'src->rename_used' and recognize this copy as a rename.
        This is not what we want.

        So, I think we really don't need to run diffcore_std more
        than one time.

        Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
        Signed-off-by: Junio C Hamano <gitster@pobox.com>

Bisect script:

    #!/bin/sh

    cd ~/g/git
    make clean
    make -j 10 CC=clang CFLAGS=-O0

    ./git --git-dir=/tmp/idea/.git log -M --follow --name-only --
platform/lang-api/src/com/intellij/lang/documentation/CompositeDocumentationProvider.java

> /dev/null

    ret=$?
    test $ret -gt 127 && ret=127
    exit $ret