{"thread":{"id":"24558","subject":"git and binary files","startedAt":"2010-07-28T14:17:06Z","lastAt":"2010-07-28T20:17:29Z","messageCount":3,"participants":["Jan Dittmer","Matthieu Moy","Avery Pennarun"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"146627","messageId":"AANLkTikceXUDDDcisnJRzg_i8mbi0TGwGiE5znAaK2aM@mail.gmail.com","threadId":"24558","inReplyTo":null,"subject":"git and binary files","fromName":"Jan Dittmer","fromEmail":"jdi@l4x.org","sentAt":"2010-07-28T14:17:06Z","receivedAt":"2010-07-28T14:17:06Z","isPatch":false,"sender":{"key":"jdi@l4x.org","avatar":null},"body":"Hi,\n\nI'm using git to keep track of directories full of jpegs and movies\n(19 GB .git directory, 5825 files, 18+ GB).\n\nI now that this is also possible with rsync and friends, but I like\nto have full version control and history, so please no discussion\nabout that.\n\nThe problem I'm facing is, that 'git status' is very\nvery slow. They seem to read through every file in the directory\ntree. Is it possible to tell git to just use mtime/size information for\nan initial guess? I already tried to turn of rename detection with\ndiff.renames=false, but still the problem remains.\n\nAny hint? git version is 1.7.1.\n\nThanks,\n\nJan\n\n-- \nJan Dittmer <jdi@l4x.org>\n"},{"id":"146629","messageId":"vpqwrsf1xz0.fsf@bauges.imag.fr","threadId":"24558","inReplyTo":"AANLkTikceXUDDDcisnJRzg_i8mbi0TGwGiE5znAaK2aM@mail.gmail.com","subject":"Re: git and binary files","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2010-07-28T14:29:23Z","receivedAt":"2010-07-28T14:29:23Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Jan Dittmer <jdi@l4x.org> writes:\n\n> The problem I'm facing is, that 'git status' is very\n> very slow. They seem to read through every file in the directory\n> tree. Is it possible to tell git to just use mtime/size information for\n> an initial guess?\n\nThis is what it does by default. You probably need an initial \"git\nstatus\" to be slow to get this information right in the index, though.\n\nHave a look at git-bigfiles too.\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"146653","messageId":"AANLkTiktAjui6=PECv8jFx7_0Suqv5sY=R9t4q1Jz+Be@mail.gmail.com","threadId":"24558","inReplyTo":"vpqwrsf1xz0.fsf@bauges.imag.fr","subject":"Re: git and binary files","fromName":"Avery Pennarun","fromEmail":"apenwarr@gmail.com","sentAt":"2010-07-28T20:17:29Z","receivedAt":"2010-07-28T20:17:29Z","isPatch":false,"sender":{"key":"apenwarr@gmail.com","avatar":"https://avatars.githubusercontent.com/u/20592?v=4"},"body":"On Wed, Jul 28, 2010 at 10:29 AM, Matthieu Moy\n<Matthieu.Moy@grenoble-inp.fr> wrote:\n> Jan Dittmer <jdi@l4x.org> writes:\n>> The problem I'm facing is, that 'git status' is very\n>> very slow. They seem to read through every file in the directory\n>> tree. Is it possible to tell git to just use mtime/size information for\n>> an initial guess?\n>\n> This is what it does by default. You probably need an initial \"git\n> status\" to be slow to get this information right in the index, though.\n\nJudging from the description, it sounds like 'git status' is the\nproblem, not the solution.  Although at 18 GB, it was probably pretty\nslow to run, so maybe Jan only ran it once and the second time it\nwould have been fast :)\n\nDepending on what the requirements are (ie. just keeping multiple\nversions for posterity/safety as opposed to actually checking out old\nversions on a regular basis), bup might be more useful than git in\nthis case: http://github.com/apenwarr/bup .\n\nHave fun,\n\nAvery\n"}]}