{"thread":{"id":"32865","subject":"git bisect result off by 1 commit","startedAt":"2013-02-09T01:54:13Z","lastAt":"2013-02-11T17:20:30Z","messageCount":3,"participants":["Tim Chen","Junio C Hamano"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"209062","messageId":"1360374853.17632.182.camel@schen9-DESK","threadId":"32865","inReplyTo":null,"subject":"git bisect result off by 1 commit","fromName":"Tim Chen","fromEmail":"tim.c.chen@linux.intel.com","sentAt":"2013-02-09T01:54:13Z","receivedAt":"2013-02-09T01:54:13Z","isPatch":false,"sender":{"key":"tim.c.chen@linux.intel.com","avatar":null},"body":"When I am doing a git bisect to track down a problem commit on the Linux \nkernel tree, I found that git bisect actually led me to a patch that's one\nbefore the problem commit.\n\nIn particular,\n\n$ git bisect replay bisectlog \nPrevious HEAD position was d54b1a9... perf script: Remove use of die/exit\nHEAD is now at a0d271c... Linux 3.6\nBisecting: 0 revisions left to test after this (roughly 0 steps)\n[d54b1a9e0eaca92cde678d19bd82b9594ed00450] perf script: Remove use of die/exit\n\nHowever, the patch that is problematic is the one before the one git bisect indicated.\n[commit 8d3eca20b9f31cf10088e283d704f6a71b9a4ee2].\n\nI am running git 1.7.11.7.\n\nTim\n\n\nThe bisect record is as follow:\n\n------------bisectlog-----------------\ngit bisect start\n# good: [a0d271cbfed1dd50278c6b06bead3d00ba0a88f9] Linux 3.6\ngit bisect good a0d271cbfed1dd50278c6b06bead3d00ba0a88f9\n# bad: [ddffeb8c4d0331609ef2581d84de4d763607bd37] Linux 3.7-rc1\ngit bisect bad ddffeb8c4d0331609ef2581d84de4d763607bd37\n# bad: [24d7b40a60cf19008334bcbcbd98da374d4d9c64] ARM: OMAP2+: PM: MPU DVFS: use generic CPU device for MPU-SS\ngit bisect bad 24d7b40a60cf19008334bcbcbd98da374d4d9c64\n# bad: [d9a807461fc8cc0d6ba589ea0730d139122af012] Merge tag 'usb-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\ngit bisect bad d9a807461fc8cc0d6ba589ea0730d139122af012\n# bad: [06d2fe153b9b35e57221e35831a26918f462db68] Merge tag 'driver-core-3.6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\ngit bisect bad 06d2fe153b9b35e57221e35831a26918f462db68\n# bad: [7e92daaefa68e5ef1e1732e45231e73adbb724e7] Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\ngit bisect bad 7e92daaefa68e5ef1e1732e45231e73adbb724e7\n# good: [620e77533f29796df7aff861e79bd72e08554ebb] Merge branch 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\ngit bisect good 620e77533f29796df7aff861e79bd72e08554ebb\n# bad: [8ad7013b252ba683055df19e657eb03d98f4f312] perf test: Add round trip test for sw and hw event names\ngit bisect bad 8ad7013b252ba683055df19e657eb03d98f4f312\n# good: [f47b58b75f5e2a424834eb15f7565a7458a12f44] perf symbols: Fix builds with NO_LIBELF set\ngit bisect good f47b58b75f5e2a424834eb15f7565a7458a12f44\n# good: [e1aa7c30c599e99b4544f9e5b4c275c8a5325bdc] tools lib traceevent: Fix strerror_r() use in pevent_strerror\ngit bisect good e1aa7c30c599e99b4544f9e5b4c275c8a5325bdc\n# good: [ff1a70e75fd005821ab5f2211312a8aa13bbf959] tools lib traceevent: Modify header to work in C++ programs\ngit bisect good ff1a70e75fd005821ab5f2211312a8aa13bbf959\n# bad: [4592281403e74dc4401d5803ec9948d43bbee7ae] perf tools: Remove the node from rblist in strlist__remove\ngit bisect bad 4592281403e74dc4401d5803ec9948d43bbee7ae\n# good: [cc58482133296f52873be909a2795f6d934ecec9] perf help: Remove use of die and handle errors\ngit bisect good cc58482133296f52873be909a2795f6d934ecec9\n# bad: [8d3eca20b9f31cf10088e283d704f6a71b9a4ee2] perf record: Remove use of die/exit\ngit bisect bad 8d3eca20b9f31cf10088e283d704f6a71b9a4ee2\n"},{"id":"209064","messageId":"7v1ucqxmyd.fsf@alter.siamese.dyndns.org","threadId":"32865","inReplyTo":"1360374853.17632.182.camel@schen9-DESK","subject":"Re: git bisect result off by 1 commit","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2013-02-09T02:03:54Z","receivedAt":"2013-02-09T02:03:54Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Tim Chen <tim.c.chen@linux.intel.com> writes:\n\n> When I am doing a git bisect to track down a problem commit on the Linux \n> kernel tree, I found that git bisect actually led me to a patch that's one\n> before the problem commit.\n>\n> In particular,\n>\n> $ git bisect replay bisectlog \n> Previous HEAD position was d54b1a9... perf script: Remove use of die/exit\n> HEAD is now at a0d271c... Linux 3.6\n> Bisecting: 0 revisions left to test after this (roughly 0 steps)\n> [d54b1a9e0eaca92cde678d19bd82b9594ed00450] perf script: Remove use of die/exit\n>\n> However, the patch that is problematic is the one before the one git bisect indicated.\n> [commit 8d3eca20b9f31cf10088e283d704f6a71b9a4ee2].\n\nLooks perfectly normal to me.  The message above:\n\n> HEAD is now at a0d271c... Linux 3.6\n> Bisecting: 0 revisions left to test after this (roughly 0 steps)\n> [d54b1a9e0eaca92cde678d19bd82b9594ed00450] perf script: Remove use of die/exit\n\nis asking you to test the commit at d54b1a9e and tell it the result\nof the test.  The message says \"0 left to test *after* this\";\ndoesn't it mean you still need to do *this*?\n\nA bisecct session ends when it tells you\n\n\tXXXXXX is the first bad commit\n\nwhich I do not see in the above.  You seem to have stopped before\nthat happens.\n"},{"id":"209278","messageId":"1360603230.17632.183.camel@schen9-DESK","threadId":"32865","inReplyTo":"7v1ucqxmyd.fsf@alter.siamese.dyndns.org","subject":"Re: git bisect result off by 1 commit","fromName":"Tim Chen","fromEmail":"tim.c.chen@linux.intel.com","sentAt":"2013-02-11T17:20:30Z","receivedAt":"2013-02-11T17:20:30Z","isPatch":false,"sender":{"key":"tim.c.chen@linux.intel.com","avatar":null},"body":"\n> \n> Looks perfectly normal to me.  The message above:\n> \n> > HEAD is now at a0d271c... Linux 3.6\n> > Bisecting: 0 revisions left to test after this (roughly 0 steps)\n> > [d54b1a9e0eaca92cde678d19bd82b9594ed00450] perf script: Remove use of die/exit\n> \n> is asking you to test the commit at d54b1a9e and tell it the result\n> of the test.  The message says \"0 left to test *after* this\";\n> doesn't it mean you still need to do *this*?\n> \n> A bisecct session ends when it tells you\n> \n> \tXXXXXX is the first bad commit\n> \n> which I do not see in the above.  You seem to have stopped before\n> that happens.\n\nYes, I should do one more step.  Thanks for catching.\n\nTim\n"}]}