{"thread":{"id":"3252","subject":"git-bisect doesnt record bad commits","startedAt":"2006-02-06T21:54:42Z","lastAt":"2006-02-06T22:24:48Z","messageCount":3,"participants":["Olaf Hering","Linus Torvalds"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"15642","messageId":"20060206215442.GA12209@suse.de","threadId":"3252","inReplyTo":null,"subject":"git-bisect doesnt record bad commits","fromName":"Olaf Hering","fromEmail":"olh@suse.de","sentAt":"2006-02-06T21:54:42Z","receivedAt":"2006-02-06T21:54:42Z","isPatch":false,"sender":{"key":"olh@suse.de","avatar":null},"body":"\nIs there a reason why git-bisect does only record good ones? I would\nreally like to know which commits did crash for me. :(\n\n 2006-02-05 14:07 .git/refs/bisect/good-a020ff412f0ecbb1e4aae1681b287e5785dd77b5\n 2006-02-05 21:25 .git/refs/bisect/good-f61ea1b0c825a20a1826bb43a226387091934586\n 2006-02-06 07:02 .git/refs/bisect/good-f2e46561cc1afa82b18b2fc6efc8510ec57c7d7d\n 2006-02-06 10:24 .git/refs/bisect/good-d779188d2baf436e67fe8816fca2ef53d246900f\n 2006-02-06 14:37 .git/refs/bisect/good-1cb9e8e01d2c73184e2074f37cd155b3c4fdaae6\n 2006-02-06 17:43 .git/refs/bisect/good-54e08a2392e99ba9e48ce1060e0b52a39118419c\n 2006-02-06 19:52 .git/refs/bisect/bad\n 2006-02-06 21:57 .git/refs/bisect/good-4a4efbdee278b2f4ed91aad2db5c006ff754276e\n\n-- \nshort story of a lazy sysadmin:\n alias appserv=wotan\n"},{"id":"15643","messageId":"Pine.LNX.4.64.0602061417510.3854@g5.osdl.org","threadId":"3252","inReplyTo":"20060206215442.GA12209@suse.de","subject":"Re: git-bisect doesnt record bad commits","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2006-02-06T22:21:12Z","receivedAt":"2006-02-06T22:21:12Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 6 Feb 2006, Olaf Hering wrote:\n> \n> Is there a reason why git-bisect does only record good ones? I would\n> really like to know which commits did crash for me. :(\n\nIt does. Look in the bisection log (\".git/BISECT_LOG\").\n\nNow, the reason it doesn't actually save them away as _tags_ is because \ntechnically, it doesn't need to. The algorithm itself only really cares \nabout the most recently marked-bad commit, and the others are totally \nuninteresting from an algorithmic standpoint.\n\nThat said, if you want the visualization tools (for example) to show them, \nyou could certainly _create_ the tags, if you want to. The fact that they \nwouldn't be used by the bisection algorithm doesn't mean that you can't \ncreate the \".git/refs/bisect/bad-<sha1>\" file in _addition_ to the one \nthat marks the latest one.\n\n\t\tLinus\n"},{"id":"15644","messageId":"20060206222448.GA9196@suse.de","threadId":"3252","inReplyTo":"Pine.LNX.4.64.0602061417510.3854@g5.osdl.org","subject":"Re: git-bisect doesnt record bad commits","fromName":"Olaf Hering","fromEmail":"olh@suse.de","sentAt":"2006-02-06T22:24:48Z","receivedAt":"2006-02-06T22:24:48Z","isPatch":false,"sender":{"key":"olh@suse.de","avatar":null},"body":" On Mon, Feb 06, Linus Torvalds wrote:\n\n> \n> \n> On Mon, 6 Feb 2006, Olaf Hering wrote:\n> > \n> > Is there a reason why git-bisect does only record good ones? I would\n> > really like to know which commits did crash for me. :(\n> \n> It does. Look in the bisection log (\".git/BISECT_LOG\").\n\nThats good enough for me to restore my steps, I think. Thanks.\n\n-- \nshort story of a lazy sysadmin:\n alias appserv=wotan\n"}]}