{"thread":{"id":"36977","subject":"autopacking twice?","startedAt":"2014-06-22T05:51:54Z","lastAt":"2014-06-23T10:57:31Z","messageCount":4,"participants":["Michael S. Tsirkin","Matthieu Moy","Duy Nguyen"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"244832","messageId":"20140622055154.GA13819@redhat.com","threadId":"36977","inReplyTo":null,"subject":"autopacking twice?","fromName":"Michael S. Tsirkin","fromEmail":"mst@redhat.com","sentAt":"2014-06-22T05:51:54Z","receivedAt":"2014-06-22T05:51:54Z","isPatch":false,"sender":{"key":"mst@kernel.org","avatar":null},"body":"I noticed this:\nremote: Counting objects: 302, done.\nremote: Compressing objects: 100% (195/195), done.\nremote: Total 209 (delta 169), reused 15 (delta 14)\nReceiving objects: 100% (209/209), 42.83 KiB | 0 bytes/s, done.\nResolving deltas: 100% (169/169), completed with 67 local objects.\nFrom git://git.qemu.org/qemu\n 6baa963..427e175 master -> origin/master\nAuto packing the repository in background for optimum performance.\nSee \"git help gc\" for manual housekeeping.\nFirst, rewinding head to replay your work on top of it...\nApplying: vhost: block migration if backend does not log memory\nApplying: vhost: fix resource leak in error handling\nApplying: qapi/hmp: use 'backend' instead of 'device' with memory\nbackend\nApplying: libqemustub: add more stubs for qemu-char\nApplying: qtest: fix qtest for vhost-user\nApplying: qtest: fix vhost-user-test unbalanced mutex locks\nApplying: e1000: emulate auto-negotiation during external link status\nchange\nApplying: e1000: improve auto-negotiation reporting via mii-tool\nApplying: e1000: signal guest on successful link auto-negotiation\nApplying: e1000: move e1000_autoneg_timer() to after set_ics()\nApplying: e1000: factor out checking for auto-negotiation availability\nApplying: fixup! libqemustub: add more stubs for qemu-char\nApplying: qapi/string-output-visitor: fix human output\nApplying: tests: add human format test for string output visitor\nApplying: Revert \"fixup! libqemustub: add more stubs for qemu-char\"\nApplying: fixup! libqemustub: add more stubs for qemu-char\nwarning: notes ref refs/notes/commits is invalid\nAuto packing the repository in background for optimum performance.\nSee \"git help gc\" for manual housekeeping.\n\nWhy did it auto-pack twice in a single pull?\nNone of the changes applied are very large.\n\nGuess: auto-packing was started in background, did not\ncomplete in time, and was restarted for the second time?\nIf true, some kind of lock file would be useful\nto prevent this.\n\n-- \nMST\n"},{"id":"244835","messageId":"vpqbntlq893.fsf@anie.imag.fr","threadId":"36977","inReplyTo":"20140622055154.GA13819@redhat.com","subject":"Re: autopacking twice?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2014-06-22T15:22:32Z","receivedAt":"2014-06-22T15:22:32Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"\"Michael S. Tsirkin\" writes:\n\n> Guess: auto-packing was started in background, did not\n> complete in time, and was restarted for the second time?\n\nProbably once for fetch, and another after rebase.\n\n> If true, some kind of lock file would be useful\n> to prevent this.\n\nIt is the case with recent versions of Git (>= 1.8.5). Which version are\nyou running?\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"244838","messageId":"20140622164357.GA31997@redhat.com","threadId":"36977","inReplyTo":"vpqbntlq893.fsf@anie.imag.fr","subject":"Re: autopacking twice?","fromName":"Michael S. Tsirkin","fromEmail":"mst@redhat.com","sentAt":"2014-06-22T16:43:57Z","receivedAt":"2014-06-22T16:43:57Z","isPatch":false,"sender":{"key":"mst@kernel.org","avatar":null},"body":"On Sun, Jun 22, 2014 at 05:22:32PM +0200, Matthieu Moy wrote:\n> \"Michael S. Tsirkin\" writes:\n> \n> > Guess: auto-packing was started in background, did not\n> > complete in time, and was restarted for the second time?\n> \n> Probably once for fetch, and another after rebase.\n> \n> > If true, some kind of lock file would be useful\n> > to prevent this.\n> \n> It is the case with recent versions of Git (>= 1.8.5). Which version are\n> you running?\n\nSorry, forgot to mention this:\n\n$ git --version\ngit version 2.0.0.542.g41d13ef\n\nthis is git master with some patches by my own.\n\n> -- \n> Matthieu Moy\n> http://www-verimag.imag.fr/~moy/\n"},{"id":"244865","messageId":"CACsJy8D3cz9w9RrVLcRgz+xgw0+2B9G_BAotvVzUoJ9Uf-Zoag@mail.gmail.com","threadId":"36977","inReplyTo":"20140622055154.GA13819@redhat.com","subject":"Re: autopacking twice?","fromName":"Duy Nguyen","fromEmail":"pclouds@gmail.com","sentAt":"2014-06-23T10:57:31Z","receivedAt":"2014-06-23T10:57:31Z","isPatch":false,"sender":{"key":"pclouds@gmail.com","avatar":"https://avatars.githubusercontent.com/u/720?v=4"},"body":"On Sun, Jun 22, 2014 at 12:51 PM, Michael S. Tsirkin wrote:\n> Why did it auto-pack twice in a single pull?\n> None of the changes applied are very large.\n>\n> Guess: auto-packing was started in background, did not\n> complete in time, and was restarted for the second time?\n> If true, some kind of lock file would be useful\n> to prevent this.\n\nThere is a lock, $GIT_DIR/gc.pid. However we are prepared to deal with\nstale locks, and may have slipped somehow. The test is in\nbuiltin/gc.c, lock_repo_for_gc(). Oh wait, we check for lock (and exit\nif lock exists) _after_ we print the message (in cmd_gc). That might\nbe it.. If it so, maybe we can reorder some code to prevent this bogus\nmessage, but has to be careful and do it before daemonizing..\n-- \nDuy\n"}]}