{"thread":{"id":"1784","subject":"Add note about IANA confirmation","startedAt":"2005-09-12T18:23:00Z","lastAt":"2005-09-12T22:40:58Z","messageCount":3,"participants":["Linus Torvalds","Junio C Hamano"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"8405","messageId":"Pine.LNX.4.58.0509121057520.3242@g5.osdl.org","threadId":"1784","inReplyTo":null,"subject":"Add note about IANA confirmation","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-09-12T18:23:00Z","receivedAt":"2005-09-12T18:23:00Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\nThe git port (9418) is officially listed by IANA now.\n\nSo document it.\n\nSigned-off-by: Linus Torvalds \n---\n\nBtw, another thing happened yesterday too: I'm now officially using a\npacked archive for my core kernel stuff. The format hasn't changed in\nI-don't-know-how-long, and I can't recall any pack-specific bug either\n(apart from the obvious lack of support by some of the client-side stuff).\n\nGoodie. Me happy.\n\ndiff --git a/cache.h b/cache.h\n--- a/cache.h\n+++ b/cache.h\n@@ -50,6 +50,17 @@\n *\tYour search - \"port 9418\" - did not match any documents.\n *\n * as www.google.com puts it.\n+ *\n+ * This port has been properly assigned for git use by IANA:\n+ * git (Assigned-9418) [I06-050728-0001].\n+ *\n+ *\tgit 9418/tcp git pack transfer service\n+ *\tgit 9418/udp git pack transfer service\n+ *\n+ * with Linus Torvalds as the point of\n+ * contact. September 2005.\n+ *\n+ * See http://www.iana.org/assignments/port-numbers\n */\n #define DEFAULT_GIT_PORT 9418\n \n"},{"id":"8433","messageId":"7vk6hm7ymd.fsf@assigned-by-dhcp.cox.net","threadId":"1784","inReplyTo":"Pine.LNX.4.58.0509121057520.3242@g5.osdl.org","subject":"Re: Add note about IANA confirmation","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2005-09-12T21:41:14Z","receivedAt":"2005-09-12T21:41:14Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Linus Torvalds writes:\n\n> The git port (9418) is officially listed by IANA now.\n\nThanks. Applied.\n\n> Btw, another thing happened yesterday too: I'm now officially using a\n> packed archive for my core kernel stuff. The format hasn't changed in\n> I-don't-know-how-long, and I can't recall any pack-specific bug either\n> (apart from the obvious lack of support by some of the client-side stuff).\n\nYou are brave ;-). \n\nSeriously, packed archive seems to be really stable. I've been\non packed archive for git.git for quite some time now, and all\nof my recent read-tree and merge experiments have been done on a\nlinux-2.6 repository that has several maintainer trees rolled\ninto a single big ball of wax, packed into one pack that weighs\nabout 76MB, with a few handred unpacked objects.\n\nI noticed one thing that is a bit curious. Packing the same set\nof objects with --window=50 seems to give worse packing result\n(space-wise) than with --window=10, which is the default if I\nrecall correctly.\n\n: siamese; git-rev-list --objects v2.6.12 >++object-list-2.6.12\n: siamese; git-pack-objects --window=50 --stdout \\\n <./++object-list-2.6.12 >./++pack-w50.pack\nPacking 31617 objects\n: siamese; git-pack-objects --window=10 --stdout \\\n <./++object-list-2.6.12 >./++pack-w10.pack\nPacking 31617 objects\n: siamese; ls -l ./++pack-w*.pack\n-rw-rw-r-- 1 junio src 57240724 Sep 12 14:39 ./++pack-w10.pack\n-rw-rw-r-- 1 junio src 57411879 Sep 12 14:37 ./++pack-w50.pack\n\nBTW, I happen to know one pathological case where 'git repack'\nwould do a wrong thing and abort, but it is really of \"doctor if\nI do this it hurts -- do not do it then\" kind, so it is not on\nmy priority list right now.\n"},{"id":"8441","messageId":"Pine.LNX.4.58.0509121523170.3266@g5.osdl.org","threadId":"1784","inReplyTo":"7vk6hm7ymd.fsf@assigned-by-dhcp.cox.net","subject":"Re: Add note about IANA confirmation","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-09-12T22:40:58Z","receivedAt":"2005-09-12T22:40:58Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 12 Sep 2005, Junio C Hamano wrote:\n> \n> I noticed one thing that is a bit curious. Packing the same set\n> of objects with --window=50 seems to give worse packing result\n> (space-wise) than with --window=10, which is the default if I\n> recall correctly.\n\nThe window doesn't seem to make much of a difference any more, now that we \nsort object blobs by filename. I suspect that it's just being unlucky: it \nfinds a locally smaller diff and uses it, but then that ends up causing \nsome other delta to _not_ be generated, because that would overflow the \n\"--depth\".\n\nAt least I _think_ that's the case. Maybe there's something really wrong \nin the window handling.\n\nGenerally, it's cheaper and more efficient to just increase the \"depth\" \nrather than the \"window\". Of course, that has a possibly very bad run-time\noverhead at unpacking time, so increasing it too much is bad.\n\nQuick test (maybe I did it right, maybe I didn't):\n\n\t\t\twindow=10\twindow=50\n\n\t depth=10\t 57240724\t 57411879\n\t depth=50\t 56700053\t 56808913\n\t depth=inf\t 56603101\t 56541663\n\nie \"--window=50\" is actually worse than \"--window=10\" for depths 10 and\n50, but better for the \"infinite\" depth case (approximated by using\n\"--depth=50000\" which is practically infinite, since I just packed the \n2.6.12 history with less than 50,000 objects ;)\n\nOf course, the 2.6.12 tree is not a good pack object. Half the objects \nin that tree have no history at all. It's not clear that it's particularly \nrepresentative of any real long-term history.\n\n\t\tLinus\n"}]}