{"thread":{"id":"10777","subject":"Reducing the memory footprint","startedAt":"2007-11-09T23:38:00Z","lastAt":"2007-11-10T06:07:05Z","messageCount":4,"participants":["Jon Smirl","Brian Downing"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"59146","messageId":"9e4733910711091538h260fbcd2s5783f01e7db4b19a@mail.gmail.com","threadId":"10777","inReplyTo":null,"subject":"Reducing the memory footprint","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2007-11-09T23:38:00Z","receivedAt":"2007-11-09T23:38:00Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"I'm using this config file:\n\n[core]\n repositoryformatversion = 0\n filemode = true\n bare = false\n logallrefupdates = true\n packedGitWindowSize = 1M\n packedGitLimit = 1M\n deltaBaseCacheLimit = 1M\n[pack]\n windowMemory = 1M\n deltaCacheSize = 1M\n\nAnd I have NO_MMAP compiled in.\n\ngit is still using over 200MB of memory or address space, my process\ngets killed either way.\n\nIdeas on how to reduce the foot print more?\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"59157","messageId":"20071110005327.GH6212@lavos.net","threadId":"10777","inReplyTo":"9e4733910711091538h260fbcd2s5783f01e7db4b19a@mail.gmail.com","subject":"Re: Reducing the memory footprint","fromName":"Brian Downing","fromEmail":"bdowning@lavos.net","sentAt":"2007-11-10T00:53:27Z","receivedAt":"2007-11-10T00:53:27Z","isPatch":false,"sender":{"key":"bdowning@lavos.net","avatar":"https://avatars.githubusercontent.com/u/366426?v=4"},"body":"On Fri, Nov 09, 2007 at 06:38:00PM -0500, Jon Smirl wrote:\n> I'm using this config file:\n> \n> [pack]\n> windowMemory = 1M\n> deltaCacheSize = 1M\n> \n> And I have NO_MMAP compiled in.\n> \n> git is still using over 200MB of memory or address space, my process\n> gets killed either way.\n\nI'm assuming it's dying on repacking since you included the pack\nparameters.\n\nHow big is your biggest object? Even with pack.windowMemory, it still\nkeeps the last object around to try and delta against (in other words,\nthe window only shrinks to size 1), which means you have to have room\nfor it and its delta index.\n\n-bcd\n"},{"id":"59158","messageId":"9e4733910711091705i6f77d05uc5ba04f668796a73@mail.gmail.com","threadId":"10777","inReplyTo":"20071110005327.GH6212@lavos.net","subject":"Re: Reducing the memory footprint","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2007-11-10T01:05:30Z","receivedAt":"2007-11-10T01:05:30Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On 11/9/07, Brian Downing wrote:\n> On Fri, Nov 09, 2007 at 06:38:00PM -0500, Jon Smirl wrote:\n> > I'm using this config file:\n> >\n> > [pack]\n> > windowMemory = 1M\n> > deltaCacheSize = 1M\n> >\n> > And I have NO_MMAP compiled in.\n> >\n> > git is still using over 200MB of memory or address space, my process\n> > gets killed either way.\n>\n> I'm assuming it's dying on repacking since you included the pack\n> parameters.\n>\n> How big is your biggest object? Even with pack.windowMemory, it still\n> keeps the last object around to try and delta against (in other words,\n> the window only shrinks to size 1), which means you have to have room\n> for it and its delta index.\n\nIt's a Linux kernel repository. Git receive-pack is going over 200MB\nand getting zapped. I don't understand why the process is so large. I\nam compiled with -DNO_MMAP.\n\nI think I have a achieved a work around. I rsync'd in my last several\nweeks of changes. Now I can 'git push' small amounts of changes\nwithout getting killed.\n\nI'm begging dreamhost to simply install git. Installed commands don't\nget zapped.\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"59171","messageId":"9e4733910711092207p6d997750i5db997dfa125ef5d@mail.gmail.com","threadId":"10777","inReplyTo":"9e4733910711091705i6f77d05uc5ba04f668796a73@mail.gmail.com","subject":"Re: Reducing the memory footprint","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2007-11-10T06:07:05Z","receivedAt":"2007-11-10T06:07:05Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On 11/9/07, Jon Smirl wrote:\n> On 11/9/07, Brian Downing wrote:\n> > On Fri, Nov 09, 2007 at 06:38:00PM -0500, Jon Smirl wrote:\n> > > I'm using this config file:\n> > >\n> > > [pack]\n> > > windowMemory = 1M\n> > > deltaCacheSize = 1M\n> > >\n> > > And I have NO_MMAP compiled in.\n> > >\n> > > git is still using over 200MB of memory or address space, my process\n> > > gets killed either way.\n> >\n> > I'm assuming it's dying on repacking since you included the pack\n> > parameters.\n> >\n> > How big is your biggest object? Even with pack.windowMemory, it still\n> > keeps the last object around to try and delta against (in other words,\n> > the window only shrinks to size 1), which means you have to have room\n> > for it and its delta index.\n>\n> It's a Linux kernel repository. Git receive-pack is going over 200MB\n> and getting zapped. I don't understand why the process is so large. I\n> am compiled with -DNO_MMAP.\n\nI believe I must not have installed everything correctly with my\nNO_MMAP build. After debugging for a while and fixing things I'm able\nto do a push now in about 80MB of memory.\n\n> I think I have a achieved a work around. I rsync'd in my last several\n> weeks of changes. Now I can 'git push' small amounts of changes\n> without getting killed.\n>\n> I'm begging dreamhost to simply install git. Installed commands don't\n> get zapped.\n>\n> --\n> Jon Smirl\n> jonsmirl@gmail.com\n>\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"}]}