{"thread":{"id":"58","subject":"Re-done kernel archive - real one?","startedAt":"2005-04-16T23:01:45Z","lastAt":"2005-04-19T07:27:56Z","messageCount":60,"participants":["Linus Torvalds","Russell King","Petr Baudis","David A. Wheeler","H. Peter Anvin","Jochen Roemling","David Woodhouse","Randy.Dunlap","randy_dunlap","Martin Schlemmer","Greg KH"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"373","messageId":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","threadId":"58","inReplyTo":null,"subject":"Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-16T23:01:45Z","receivedAt":"2005-04-16T23:01:45Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\nOk, nobody really objected to the notion of leaving the kernel history\nbehind for now, and in fact most people seemed to basically agree. So with\nthat decided, the old kernel testing tree was actually perfectly ok,\nexcept it had been build up with the old-style commit date handling, which\nmade me not want to use it as a base for any real work.\n\nSo I re-created the dang thing (hey, it takes just a few minutes), and\npushed it out, and there's now an archive on kernel.org in my public\n\"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\nof naming git-archive directories as \"*.git\", since that really ends up\nbeing the \".git\" directory for the checked-out thing.\n\nI'm not going to announce it on linux-kernel yet, because I don't think\nit's useful to anybody but a git person anyway. Besides, I don't actually\nknow how happy the kernel.org people are about this distribution method\nand whether it ends up being a horrible disaster for the mirroring setup. \n\nPeter made some noises about /pub/scm, which makes sense, and would be a\nbetter place than my public tree. Apparently there are other places that\nare willing and able to host things too, so we'll see.\n\nNOTE! The roughly 10x expansion of archive size goind from BK to git ends\nup in a similar 10x bandwidth expansion, in addition to just the overhead\nof reading tons of directory entries and comparing them (which is what\nboth a wget and rsync thing ends up doing). I'm sure we can bring that\ndown with smarter synchronization tools, but I also suspect that's some\nway away.\n\nSo is real common usage, though, so maybe it's not that bad at all. Who \nknows. We haven't hit a single real snag so far (except it took several \ndays longer than I expected, but hey, I expect lots of things ;), and I'm \nsure real usage will show lots of them.\n\nSimilarly, we don't really have real merging, which makes tracking harder, \nbut I suspect actually having a tree out there will make people more \nmotivated and have more of a test-case. I'm feeling good enough about the \nplumbing that I think I solved the \"hard\" part of it, and now it's just \nthe boring 95% left - scripting around it.\n\nI think that with the new merge model, the easiest thing to do is to just \ndownload all new objects, and then download the HEAD file under a new \nname.\n\nIe we have two phases to the merge: first get the objects, with something\nlike\n\n\trepo=kernel.org:/pub/kernel/people/torvalds/linux-2.6.git\n\trsync --ignore-existing -acv $(repo)/ .git/\n\nwhich will _not_ download the new HEAD file (since you already have one of \nyour own), and then when you actually decide to merge you do\n\n\trsync -acv $(repo)/HEAD .git/MERGE_WITH\n\nand now you can look at your old HEAD, and the MERGE_WITH thing, look up \nthe parents, and then do\n\n\tread-tree -m \n\twrite-tree\n\tcommit-tree -p -p \n\n(which should actually _work_, assuming that the merge had no file \nconflicts).\n\nThis seems to be a sane way to do merges, and if the scripting starts from \nthere and then becomes smarter...\n\n\t\tLinus\n"},{"id":"480","messageId":"20050417162448.A13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T15:24:48Z","receivedAt":"2005-04-17T15:24:48Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sat, Apr 16, 2005 at 04:01:45PM -0700, Linus Torvalds wrote:\n> So I re-created the dang thing (hey, it takes just a few minutes), and\n> pushed it out, and there's now an archive on kernel.org in my public\n> \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> of naming git-archive directories as \"*.git\", since that really ends up\n> being the \".git\" directory for the checked-out thing.\n\nWe need to work out how we're going to manage to get our git changes to\nyou. At the moment, I've very little idea how to do that. Ideas?\n\nAt the bottom is the script itself. There's probably some aspects of\nit which aren't nice, maybe Petr can advise on this (and maybe increase\nthe functionality of the git shell script to fill in where necessary.)\n\nHowever, I've made a start to generate the necessary emails. How about\nthis format?\n\nI'm not keen on the tree, parent, author and committer objects appearing\nin this - they appear to clutter it up. What're your thoughts?\n\nI'd rather not have the FQDN of the machine where the commit happened\nappearing in the logs. (I've 'xxxx'd it out for the time being, because\nI'd rather not have yet more email-address-like objects get into spammers\ndatabases with which to hammer my 512kbps DSL line.)\n\nLinus,\n\nPlease incorporate the latest ARM changes.\n\nThis will update the following files:\n\n arm/kernel/process.c | 15 +++++++++++----\n arm/kernel/traps.c | 8 ++------\n arm/lib/changebit.S | 11 ++---------\n arm/lib/clearbit.S | 13 ++-----------\n arm/lib/setbit.S | 11 ++---------\n arm/lib/testchangebit.S | 15 ++-------------\n arm/lib/testclearbit.S | 15 ++-------------\n arm/lib/testsetbit.S | 15 ++-------------\n arm/mach-footbridge/dc21285-timer.c | 4 ++--\n arm/mach-sa1100/h3600.c | 2 +-\n asm-arm/ptrace.h | 5 +----\n asm-arm/system.h | 3 +++\n 12 files changed, 32 insertions(+), 85 deletions(-)\n\nthrough these ChangeSets:\n\n\ttree 7c4d75539c29ef7a9dde81acf84a072649f4f394\n\tparent d5922e9c35d21f0b6b82d1fd8b1444cfce57ca34\n\tauthor Russell King 1113749462 +0100\n\tcommitter Russell King 1113749462 +0100\n\t\n\t[PATCH] ARM: bitops\n\t\n\tConvert ARM bitop assembly to a macro. All bitops follow the same\n\tformat, so it's silly duplicating the code when only one or two\n\tinstructions are different.\n\t\n\tSigned-off-by: Russell King \n\ttree fc10d3ffa6062cda10a10cb8262d8df238aea4fb\n\tparent 5d9a545981893629c8f95e2b8b50d15d18c6ddbc\n\tauthor Russell King 1113749436 +0100\n\tcommitter Russell King 1113749436 +0100\n\t\n\t[PATCH] ARM: showregs\n\t\n\tFix show_regs() to provide a backtrace. Provide a new __show_regs()\n\tfunction which implements the common subset of show_regs() and die().\n\tAdd prototypes to asm-arm/system.h\n\t\n\tSigned-off-by: Russell King \n\ttree 5591fced9a2b5f84c6772dcbe2eb4b24e29161fc\n\tparent 488faba31f59c5960aabbb2a5877a0f2923937a3\n\tauthor Russell King 1113748846 +0100\n\tcommitter Russell King 1113748846 +0100\n\t\n\t[PATCH] ARM: h3600_irda_set_speed arguments\n\t\n\th3600_irda_set_speed() had the wrong type for the \"speed\" argument.\n\tFix this.\n\t\n\tSigned-off-by: Russell King \n\ttree 2493491da6e446e48d5443f0a549a10ed3d35b62\n\tparent e7905b2f22eb5d5308c9122b9c06c2d02473dd4f\n\tauthor Russell King 1113748615 +0100\n\tcommitter Russell King 1113748615 +0100\n\t\n\t[PATCH] ARM: footbridge rtc init\n\t\n\tThe footbridge ISA RTC was being initialised before we had setup the\n\tkernel timer. This caused a divide by zero error when the current\n\ttime of day is set. Resolve this by initialising the RTC after\n\tthe kernel timer has been initialised.\n\t\n\tSigned-off-by: Russell King \n\n---\n\n#!/bin/sh\nprev=$(cat .git/heads/origin)\nto=$(cat .git/HEAD)\nwho=Linus\nwhat=ARM\n \ncat << EOT\n${who},\n \nPlease incorporate the latest ${what} changes.\n \nThis will update the following files:\n \nEOT\n \ngit diff $prev $to | diffstat -p1\n \ncat << EOT\n \nthrough these ChangeSets:\n \nEOT\n \nthis=$to\nwhile [ \"$this\" != \"$prev\" ]; do\n cat-file commit $this | sed 's,.*,\\t&,'\n this=$(cat-file commit $this | grep ^parent | cut -d ' ' -f 2)\ndone\n\n\n-- \nRussell King\n\n"},{"id":"489","messageId":"20050417170539.B13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T16:05:40Z","receivedAt":"2005-04-17T16:05:40Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sat, Apr 16, 2005 at 04:01:45PM -0700, Linus Torvalds wrote:\n> So I re-created the dang thing (hey, it takes just a few minutes), and\n> pushed it out, and there's now an archive on kernel.org in my public\n> \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> of naming git-archive directories as \"*.git\", since that really ends up\n> being the \".git\" directory for the checked-out thing.\n\nBTW, there appears to be \"errors\" in the history committed thus far.\nI'm not sure where this came from though. Some of them could be\nUTF8 vs ASCII issues, but there's a number which seem to have extra\nrandom crap in them (\"^M)\" and lots of blank lines).\n\nOne thing which definitely needs to be considered is - what character\nencoding are the comments to be stored as? Currently, it's whatever\nthe committers character encoding is, which will be completely random.\nFor instance, dwmw2 will definitely be using UTF, whereas I'll definitely\nbe using ISO-8859-1 (UTF is far too much of a pain in the ass to use,\nunless _all_ your systems are running UTF, which mine don't yet.)\n\nID: 75f86bac962b7609b0f3c21d25e10647ff8ed280\n[PATCH] intel8x0: AC'97 audio patch for Intel ESB2\n \n This patch adds the Intel ESB2 DID's to the intel8x0.c file for AC'97 audio\n support.\n \n Signed-off-by: Jason Gaston \n Signed-off-by: Andrew Morton \n Signed-off-by: Linus Torvalds \n\nID: baab5c52135dfa2a02c5dcc8422fc5d048acc682\n[PATCH] ppc32: fix compilation error in arch/ppc/kernel/time.c\n \n make defconfig give the following error on ppc (gcc-4):\n \n arch/ppc/kernel/time.c:92: error: static declaration of <80><98>time_offset<80><99>\n follows non-static declaration\n include/linux/timex.h:236: error: previous declaration of <80><98>time_offset<80><99>\n was here\n \n The following patch solves it (time_offset is declared in timer.c).\n \n Signed-Off-By: Benoit Boissinot \n Signed-off-by: Andrew Morton \n Signed-off-by: Linus Torvalds \n\nID: fa357627062fa6f4727638bc4f302b23f5a3acb8\n[PATCH] vmscan: pageout(): remove unneeded test\n \n ^M)\n \n \n \n We only call pageout() for dirty pages, so this test is redundant.\n \n Signed-off-by: Andrew Morton \n Signed-off-by: Linus Torvalds \n\nID: e5fc9026c6a372a67e652c6ba1648e5f8e543600\n[PATCH] arm: fix SIGBUS handling\n \n ^M)\n \n \n From: Russell King \n \n ARM wasn't raising a SIGBUS with a siginfo structure. Fix\n __do_user_fault() to allow us to use it for SIGBUS conditions, and arrange\n for the sigbus path to use this.\n \n We need to prevent the siginfo code being called if we do not have a user\n space context to call it, so consolidate the \"user_mode()\" tests.\n \n Thanks to Ian Campbell who spotted this oversight.\n \n Signed-off-by: Russell King \n Signed-off-by: Andrew Morton \n Signed-off-by: Linus Torvalds \n\n\n-- \nRussell King\n\n"},{"id":"495","messageId":"20050417162829.GK1487@pasky.ji.cz","threadId":"58","inReplyTo":"20050417162448.A13233@flint.arm.linux.org.uk","subject":"Re: Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-17T16:28:29Z","receivedAt":"2005-04-17T16:28:29Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Sun, Apr 17, 2005 at 05:24:48PM CEST, I got a letter\nwhere Russell King told me that...\n> However, I've made a start to generate the necessary emails. How about\n> this format?\n> \n> I'm not keen on the tree, parent, author and committer objects appearing\n> in this - they appear to clutter it up. What're your thoughts?\n\nI think it would look nicer if you used git log format. Actually...\n\n \n> #!/bin/sh\n> prev=$(cat .git/heads/origin)\n> to=$(cat .git/HEAD)\n\nIt is preferred to use the commit-id command instead; when git decides\nto totally dig over its metametadata storage, you won't need to care.\n\n> this=$to\n> while [ \"$this\" != \"$prev\" ]; do\n> cat-file commit $this | sed 's,.*,\\t&,'\n> this=$(cat-file commit $this | grep ^parent | cut -d ' ' -f 2)\n> done\n\nWhat about adding support to git log for not caring about side branches\n(if there are multiple parents, go only over the first one; you might\nadd that as a switch to rev-tree, which would turn it to rev-line ;-)?\n\nThen, you could also easily implement support for passing additional\ncommit ID to git log, and it would then print out the range between\nthese two.\n\nThat would do exactly what you want, and would be probably quite usable\nfor other uses too.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"496","messageId":"Pine.LNX.4.58.0504170926410.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417162448.A13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T16:36:09Z","receivedAt":"2005-04-17T16:36:09Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n>\n> On Sat, Apr 16, 2005 at 04:01:45PM -0700, Linus Torvalds wrote:\n> > So I re-created the dang thing (hey, it takes just a few minutes), and\n> > pushed it out, and there's now an archive on kernel.org in my public\n> > \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> > of naming git-archive directories as \"*.git\", since that really ends up\n> > being the \".git\" directory for the checked-out thing.\n> \n> We need to work out how we're going to manage to get our git changes to\n> you. At the moment, I've very little idea how to do that. Ideas?\n\nTo me, merging is my highest priority. I suspect that once I have a tree \nfrom you (or anybody else) that I actually _test_ merging with, I'll be \nmotivated as hell to make sure that my plumbing actually works. \n\nAfter all, it's not just you who want to have to avoid the pain of \nmerging: it's definitely in my own best interests to make merging as \neasy as possible. You're _the_ most obvious initial candidate, because \nyour merges almost never have any conflicts at all, even on a file level \n(much less within a file).\n\n> However, I've made a start to generate the necessary emails. How about\n> this format?\n> \n> I'm not keen on the tree, parent, author and committer objects appearing\n> in this - they appear to clutter it up. What're your thoughts?\n\nIndeed. I'd almost drop the whole header except for the \"author\" line. \n\nOh, and you need a separator between commits, right now your \n\"Signed-off-by:\" line ends up butting up with the header of the next \ncommit ;)\n\n> I'd rather not have the FQDN of the machine where the commit happened\n> appearing in the logs.\n\nThat's fine. Out short-logs have always tried to have just the real name \nin them, and I do want an email-like thing for tracking the developer, but \nyes, if you remove the email, that's fine. It should be easy enough to do \nwith a simple\n\n\tsed 's/<.*>//'\n\nor similar.\n\nAnd if you replace \"author\" with \"From:\" and do the date conversion, it\nmight look more natural.\n\n\t\tLinus\n"},{"id":"498","messageId":"Pine.LNX.4.58.0504170937020.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417170539.B13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T16:44:14Z","receivedAt":"2005-04-17T16:44:14Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n> \n> BTW, there appears to be \"errors\" in the history committed thus far.\n> I'm not sure where this came from though. Some of them could be\n> UTF8 vs ASCII issues, but there's a number which seem to have extra\n> random crap in them (\"^M)\" and lots of blank lines).\n\nAh, yes. That is actually from the original emails from Andrew. I do not \nknow why, but I see them there. It's his script that does something \nstrange.\n\n(Andrew: in case you care, the first one is\n\n\t[patch 003/198] arm: fix SIGBUS handling\n\nwhich has the email looking like\n\n\t...\n\tFrom: akpm@osdl.org\n\tDate: Tue, 12 Apr 2005 03:30:35 -0700\n\tStatus: \n\tX-Status: \n\tX-Keywords: \n\t\n\t^M)\n\t\n\t\n\tFrom: Russell King \n\t\n\tARM wasn't raising a SIGBUS with a siginfo structure. Fix\n\t__do_user_fault() to allow us to use it for SIGBUS conditions, and arrange\n\tfor the sigbus path to use this.\n\t...\n\n> One thing which definitely needs to be considered is - what character\n> encoding are the comments to be stored as?\n\nTo git, it's just a byte stream, and you can have binary comments if you\nwant to. I personally would prefer to move towards UTF eventually, but I\nreally don't think it matters a whole lot as long as 99.9% of everything\nwe'd see there is still 7-bit ascii.\n\n> ID: 75f86bac962b7609b0f3c21d25e10647ff8ed280\n> [PATCH] intel8x0: AC'97 audio patch for Intel ESB2\n> \n> This patch adds the Intel ESB2 DID's to the intel8x0.c file for AC'97 audio\n> support.\n> \n> Signed-off-by: Jason Gaston \n\nThat is also there in Andrew's original email. It's \"space with the\nhigh bit set\", and I have no idea why.\n\n\t\tLinus\n"},{"id":"510","messageId":"4262A767.4040300@dwheeler.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504170937020.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"David A. Wheeler","fromEmail":"dwheeler@dwheeler.com","sentAt":"2005-04-17T18:13:59Z","receivedAt":"2005-04-17T18:13:59Z","isPatch":false,"sender":{"key":"dwheeler@dwheeler.com","avatar":"https://avatars.githubusercontent.com/u/813150?v=4"},"body":"On Sun, 17 Apr 2005, Russell King wrote:\n>>BTW, there appears to be \"errors\" in the history committed thus far.\n>>I'm not sure where this came from though. Some of them could be\n>>UTF8 vs ASCII issues, ....> \n...\n>>One thing which definitely needs to be considered is - what character\n>>encoding are the comments to be stored as?\n\nLinus Torvalds replied:\n> To git, it's just a byte stream, and you can have binary comments if you\n> want to. I personally would prefer to move towards UTF eventually, but I\n> really don't think it matters a whole lot as long as 99.9% of everything\n> we'd see there is still 7-bit ascii.\n\nI would _heartily_ recommend moving towards UTF-8 as the\ninternal charset for all comments. Alternatives are possible\n(e.g., recording the charset in the header), but they're\nincredibly messy. Even if you don't normally work in UTF-8,\nit's pretty easy to set most editors up to read & write UTF-8.\nHaving the data stored as a constant charset eliminates\na raft of error-prone code.\n\n--- David A. Wheeler\n"},{"id":"511","messageId":"20050417181446.GC1461@pasky.ji.cz","threadId":"58","inReplyTo":"4262A767.4040300@dwheeler.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-17T18:14:46Z","receivedAt":"2005-04-17T18:14:46Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Sun, Apr 17, 2005 at 08:13:59PM CEST, I got a letter\nwhere \"David A. Wheeler\" told me that...\n> On Sun, 17 Apr 2005, Russell King wrote:\n> >>BTW, there appears to be \"errors\" in the history committed thus far.\n> >>I'm not sure where this came from though. Some of them could be\n> >>UTF8 vs ASCII issues, ....> \n> ...\n> >>One thing which definitely needs to be considered is - what character\n> >>encoding are the comments to be stored as?\n> \n> Linus Torvalds replied:\n> >To git, it's just a byte stream, and you can have binary comments if you\n> >want to. I personally would prefer to move towards UTF eventually, but I\n> >really don't think it matters a whole lot as long as 99.9% of everything\n> >we'd see there is still 7-bit ascii.\n> \n> I would _heartily_ recommend moving towards UTF-8 as the\n> internal charset for all comments. Alternatives are possible\n> (e.g., recording the charset in the header), but they're\n> incredibly messy. Even if you don't normally work in UTF-8,\n> it's pretty easy to set most editors up to read & write UTF-8.\n> Having the data stored as a constant charset eliminates\n> a raft of error-prone code.\n\nI tend to agree here. My toilet stuff is what can handle various\nlocale-based conversions at the commit-tree / cat-file tree sides etc,\nbut UTF-8 should be certainly the way to go internally.\n\nNot that the plumbing should actually _care_ at all; anyone who uses it\nshould take the care, so this is more of a \"social\" thing.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"514","messageId":"20050417192000.C13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"4262A767.4040300@dwheeler.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T18:20:00Z","receivedAt":"2005-04-17T18:20:00Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 02:13:59PM -0400, David A. Wheeler wrote:\n> On Sun, 17 Apr 2005, Russell King wrote:\n> >>BTW, there appears to be \"errors\" in the history committed thus far.\n> >>I'm not sure where this came from though. Some of them could be\n> >>UTF8 vs ASCII issues, ....> \n> ...\n> >>One thing which definitely needs to be considered is - what character\n> >>encoding are the comments to be stored as?\n> \n> Linus Torvalds replied:\n> > To git, it's just a byte stream, and you can have binary comments if you\n> > want to. I personally would prefer to move towards UTF eventually, but I\n> > really don't think it matters a whole lot as long as 99.9% of everything\n> > we'd see there is still 7-bit ascii.\n> \n> I would _heartily_ recommend moving towards UTF-8 as the\n> internal charset for all comments. Alternatives are possible\n> (e.g., recording the charset in the header), but they're\n> incredibly messy. Even if you don't normally work in UTF-8,\n> it's pretty easy to set most editors up to read & write UTF-8.\n> Having the data stored as a constant charset eliminates\n> a raft of error-prone code.\n\nExcept, I believe, MicroEMACS, which both Linus and myself use. As\nfar as I know, there aren't any patches to make it UTF-8 compliant.\n\nThe alternative is, I suppose, iconv. However, iconv in _my_ glibc\nseems buggy (segfaults) and my efforts for building glibc 2.3.2 for\nARM have failed. Effectively that means iconv is inaccessible to\nme.\n\n-- \nRussell King\n\n"},{"id":"516","messageId":"4262AE82.1070702@dwheeler.com","threadId":"58","inReplyTo":"20050417192000.C13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"David A. Wheeler","fromEmail":"dwheeler@dwheeler.com","sentAt":"2005-04-17T18:44:18Z","receivedAt":"2005-04-17T18:44:18Z","isPatch":false,"sender":{"key":"dwheeler@dwheeler.com","avatar":"https://avatars.githubusercontent.com/u/813150?v=4"},"body":">>On Sun, 17 Apr 2005, Russell King wrote:\n>>>>One thing which definitely needs to be considered is - what character\n>>>>encoding are the comments to be stored as?\n...\n\nI replied:\n>>I would _heartily_ recommend moving towards UTF-8 as the\n>>internal charset for all comments.\n\nPetr said:\n >Not that the plumbing should actually _care_ at all; anyone who uses it\n >should take the care, so this is more of a \"social\" thing.\n\nThe _plumbing_ shouldn't care, but the stuff above needs to know\nhow to interpret the stuff that the plumbing produces.\n\nRussell King said:\n> Except, I believe, MicroEMACS, which both Linus and myself use. As\n> far as I know, there aren't any patches to make it UTF-8 compliant.\n\nSince plain ASCII is a subset of UTF-8,\nas long as MicroEMACS users only create ASCII comments,\nthen the comments you create in MicroEMACS will still be UTF-8.\nNo big deal.\n\nFor reading comments, if the text is almost entirely\nplain ASCII, you could just ignore the problem and have the\noccasional character scramble. If you need more, you'll\nneed a tool that's more internationalized or a working iconv,\nbut if that's important you'd be motivated.\n\nAgain, I'm looking for more generalized solutions, where\nnon-English comments are more common than in Linux kernel code.\n\n--- David A. Wheeler\n"},{"id":"520","messageId":"20050417195742.D13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504170926410.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T18:57:42Z","receivedAt":"2005-04-17T18:57:42Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 09:36:09AM -0700, Linus Torvalds wrote:\n> On Sun, 17 Apr 2005, Russell King wrote:\n> > On Sat, Apr 16, 2005 at 04:01:45PM -0700, Linus Torvalds wrote:\n> > > So I re-created the dang thing (hey, it takes just a few minutes), and\n> > > pushed it out, and there's now an archive on kernel.org in my public\n> > > \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> > > of naming git-archive directories as \"*.git\", since that really ends up\n> > > being the \".git\" directory for the checked-out thing.\n> > \n> > We need to work out how we're going to manage to get our git changes to\n> > you. At the moment, I've very little idea how to do that. Ideas?\n> \n> To me, merging is my highest priority. I suspect that once I have a tree \n> from you (or anybody else) that I actually _test_ merging with, I'll be \n> motivated as hell to make sure that my plumbing actually works. \n\nOk, I'll throw this tree onto master.kernel.org - how about\nmaster.kernel.org:/home/rmk/linux-2.6-rmk.git ? I think it's in the\nsame format as your trees:\n\nlinux-2.6-rmk.git\n|-- HEAD\n`-- objects\n\nwhere HEAD was copied from my .git/heads/master, and objects from\n.git/objects.\n\n> > However, I've made a start to generate the necessary emails. How about\n> > this format?\n> > \n> > I'm not keen on the tree, parent, author and committer objects appearing\n> > in this - they appear to clutter it up. What're your thoughts?\n> \n> Indeed. I'd almost drop the whole header except for the \"author\" line. \n\nDone.\n\n> Oh, and you need a separator between commits, right now your \n> \"Signed-off-by:\" line ends up butting up with the header of the next \n> commit ;)\n\nDone.\n\n> > I'd rather not have the FQDN of the machine where the commit happened\n> > appearing in the logs.\n> \n> That's fine. Out short-logs have always tried to have just the real name \n> in them, and I do want an email-like thing for tracking the developer, but \n> yes, if you remove the email, that's fine. It should be easy enough to do \n> with a simple\n> \n> \tsed 's/<.*>//'\n> \n> or similar.\n\nDone.\n\n> And if you replace \"author\" with \"From:\" and do the date conversion, it\n> might look more natural.\n\nAlso done. 8)\n\nI still need to work out how to make my noddy script follow different\nbranches which may be present though. However, for my common work\nflow, it fits what I require.\n\nOk, how about this format:\n\nLinus,\n\nPlease incorporate the latest ARM changes.\n\nThis will update the following files:\n\n arch/arm/kernel/process.c | 15 +++++++++++----\n arch/arm/kernel/traps.c | 8 ++------\n arch/arm/lib/changebit.S | 11 ++---------\n arch/arm/lib/clearbit.S | 13 ++-----------\n arch/arm/lib/setbit.S | 11 ++---------\n arch/arm/lib/testchangebit.S | 15 ++-------------\n arch/arm/lib/testclearbit.S | 15 ++-------------\n arch/arm/lib/testsetbit.S | 15 ++-------------\n arch/arm/mach-footbridge/dc21285-timer.c | 4 ++--\n arch/arm/mach-sa1100/h3600.c | 2 +-\n include/asm-arm/arch-ebsa285/debug-macro.S | 7 +++++--\n include/asm-arm/arch-rpc/debug-macro.S | 5 ++++-\n include/asm-arm/ptrace.h | 5 +----\n include/asm-arm/system.h | 3 +++\n 14 files changed, 41 insertions(+), 88 deletions(-)\n\nthrough these ChangeSets:\n\nAuthor: Russell King: Sun Apr 17 16:28:31 BST 2005\n\n\t[PATCH] ARM: fix debug macros\n\t\n\tFix debug EBSA285 and RiscPC debugging macros to detect whether the\n\tMMU is enabled.\n\t\n\tSigned-off-by: Russell King\n\nAuthor: Russell King: Sun Apr 17 15:51:02 BST 2005\n\n\t[PATCH] ARM: bitops\n\t\n\tConvert ARM bitop assembly to a macro. All bitops follow the same\n\tformat, so it's silly duplicating the code when only one or two\n\tinstructions are different.\n\t\n\tSigned-off-by: Russell King\n\nAuthor: Russell King: Sun Apr 17 15:50:36 BST 2005\n\n\t[PATCH] ARM: showregs\n\t\n\tFix show_regs() to provide a backtrace. Provide a new __show_regs()\n\tfunction which implements the common subset of show_regs() and die().\n\tAdd prototypes to asm-arm/system.h\n\t\n\tSigned-off-by: Russell King\n\nAuthor: Russell King: Sun Apr 17 15:40:46 BST 2005\n\n\t[PATCH] ARM: h3600_irda_set_speed arguments\n\t\n\th3600_irda_set_speed() had the wrong type for the \"speed\" argument.\n\tFix this.\n\t\n\tSigned-off-by: Russell King\n\nAuthor: Russell King: Sun Apr 17 15:36:55 BST 2005\n\n\t[PATCH] ARM: footbridge rtc init\n\t\n\tThe footbridge ISA RTC was being initialised before we had setup the\n\tkernel timer. This caused a divide by zero error when the current\n\ttime of day is set. Resolve this by initialising the RTC after\n\tthe kernel timer has been initialised.\n\t\n\tSigned-off-by: Russell King\n\n\n\n-- \nRussell King\n\n"},{"id":"526","messageId":"Pine.LNX.4.58.0504171226530.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417195742.D13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T19:33:22Z","receivedAt":"2005-04-17T19:33:22Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n> \n> I still need to work out how to make my noddy script follow different\n> branches which may be present though. However, for my common work\n> flow, it fits what I require.\n\nThe way to handle that is that you need to \n\n - remember (or re-fetch) what the latest HEAD was that you merged with in \n my tree.\n\n if you didn't remember, you can just get all my objects and do a\n\n\tmerge-head $(cat .git/HEAD) $linus-current-head\n\n or something (using the current git archive that has a \"merge-head\" \n program. That gives you the most recent common head.\n\n - use \"rev-tree\" to show reachability\n\n\trev-tree $my-current-head $last-merge-head\n\t\t| sort -n\t\t# sort by date rather than sha1\n\t\t| cut -d' ' -f2\t\t# get the sha1 + \"flags\" mask\n\t\t| grep :1\t\t# show the ones that are only\n\t\t\t\t\t# reachable from $my-current-head\n\nand you now have a nice list of sha1's ordered by date.\n\nOr something. I didn't test the above. Testing is for users.\n\n\t\tLinus\n"},{"id":"527","messageId":"Pine.LNX.4.58.0504171242060.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171226530.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T19:45:15Z","receivedAt":"2005-04-17T19:45:15Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Linus Torvalds wrote:\n> \n> - use \"rev-tree\" to show reachability\n\nBtw, you don't even have to really remember the head you merged with. If \nyou have all my objects, you can just use the current head from whatever \nmy repository is, and the \"rev-tree\" will show all your commit objects \nthat aren't in my tree.\n\nThe only reason I said \"remember the last head you merged\" is that that \nway you don't even have to download the objects from my tree, and you can \nmake the decision totally locally without looking at what I've done since.\n\nOf course, at some point you want to get my updated objects _anyway_, \nbut..\n\n\t\t\tLinus\n"},{"id":"530","messageId":"20050417205149.E13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171226530.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T19:51:49Z","receivedAt":"2005-04-17T19:51:49Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"(Dropped HPA from the CC line - I think he was only copied for the\nmaster.kernel.org issues.)\n\nOn Sun, Apr 17, 2005 at 12:33:22PM -0700, Linus Torvalds wrote:\n> On Sun, 17 Apr 2005, Russell King wrote:\n> > I still need to work out how to make my noddy script follow different\n> > branches which may be present though. However, for my common work\n> > flow, it fits what I require.\n> \n> The way to handle that is that you need to \n> \n> - remember (or re-fetch) what the latest HEAD was that you merged with in \n> my tree.\n> \n> if you didn't remember, you can just get all my objects and do a\n> \n> \tmerge-head $(cat .git/HEAD) $linus-current-head\n> \n> or something (using the current git archive that has a \"merge-head\" \n> program. That gives you the most recent common head.\n\nMy script currently sends between two commit-ids, so...\n\n> - use \"rev-tree\" to show reachability\n> \n> \trev-tree $my-current-head $last-merge-head\n> \t\t| sort -n\t\t# sort by date rather than sha1\n> \t\t| cut -d' ' -f2\t\t# get the sha1 + \"flags\" mask\n> \t\t| grep :1\t\t# show the ones that are only\n> \t\t\t\t\t# reachable from $my-current-head\n> \n> and you now have a nice list of sha1's ordered by date.\n\nThis will (and does) do exactly what I want. I'll also read into the\nabove a request that you want it in forward date order. 8)\n\n-- \nRussell King\n\n"},{"id":"533","messageId":"Pine.LNX.4.58.0504171306360.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417205149.E13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T20:08:51Z","receivedAt":"2005-04-17T20:08:51Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n> \n> This will (and does) do exactly what I want. I'll also read into the\n> above a request that you want it in forward date order. 8)\n\nNo, I actually don't _think_ I care. In many ways I'm more used to\n\"reverse date order\", because that's usually how you view a changelog\n(with a pager, and most recent changes at the top).\n\nWhich one makes sense when asking me to merge? I don't know, and I don't\nthink it really even matters, but maybe we can add a \"for now\" to whatever \ndecision you end up coming to?\n\n\t\tLinus\n"},{"id":"534","messageId":"20050417211116.F13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171306360.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T20:11:16Z","receivedAt":"2005-04-17T20:11:16Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 01:08:51PM -0700, Linus Torvalds wrote:\n> On Sun, 17 Apr 2005, Russell King wrote:\n> > This will (and does) do exactly what I want. I'll also read into the\n> > above a request that you want it in forward date order. 8)\n> \n> No, I actually don't _think_ I care. In many ways I'm more used to\n> \"reverse date order\", because that's usually how you view a changelog\n> (with a pager, and most recent changes at the top).\n> \n> Which one makes sense when asking me to merge? I don't know, and I don't\n> think it really even matters, but maybe we can add a \"for now\" to whatever \n> decision you end up coming to?\n\nIt's trivial to change - it's either sort -n or sort -nr !\n\nI'll pick the reverse format so that we generate the changelog in the\norder which people have come to expect from our previous set of tools.\nConsistency with existing practises is a good thing at this point.\n\n-- \nRussell King\n\n"},{"id":"535","messageId":"4262C532.4070203@zytor.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"H. Peter Anvin","fromEmail":"hpa@zytor.com","sentAt":"2005-04-17T20:21:06Z","receivedAt":"2005-04-17T20:21:06Z","isPatch":false,"sender":{"key":"hpa@zytor.com","avatar":null},"body":"Linus Torvalds wrote:\n> Ok, nobody really objected to the notion of leaving the kernel history\n> behind for now, and in fact most people seemed to basically agree. So with\n> that decided, the old kernel testing tree was actually perfectly ok,\n> except it had been build up with the old-style commit date handling, which\n> made me not want to use it as a base for any real work.\n> \n> So I re-created the dang thing (hey, it takes just a few minutes), and\n> pushed it out, and there's now an archive on kernel.org in my public\n> \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> of naming git-archive directories as \"*.git\", since that really ends up\n> being the \".git\" directory for the checked-out thing.\n> \n> I'm not going to announce it on linux-kernel yet, because I don't think\n> it's useful to anybody but a git person anyway. Besides, I don't actually\n> know how happy the kernel.org people are about this distribution method\n> and whether it ends up being a horrible disaster for the mirroring setup. \n> \n> Peter made some noises about /pub/scm, which makes sense, and would be a\n> better place than my public tree. Apparently there are other places that\n> are willing and able to host things too, so we'll see.\n> \n\nI would suggest something like /pub/scm/kernel/git/ on \nkernel.org. This is easy to do, and being outside the \"mirrored \nworldwide\" set shouldn't cause anyone any issues.\n\n/pub/linux/... is mirrored worldwide and that might cause some \nconsternation.\n\nIf this is OK with everyone I'll try to set them up (I'm at LCA at the \nmoment, so I might have limited connectivity at any one particular \nmoment in time.)\n\n\t-hpa\n"},{"id":"537","messageId":"Pine.LNX.4.58.0504171324340.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417211116.F13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T20:26:27Z","receivedAt":"2005-04-17T20:26:27Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n> \n> It's trivial to change - it's either sort -n or sort -nr !\n\nYes.\n\nBtw, it might make sense to do the \"grep\" before the sort, since the sort \ndoesn't scale as nicely with lots of output. On the other hand, then the \ngrep pattern needs to be a bit more complex (so that it doesn't pick up on \nthings that have _parents_ with \":1\"). I think the rev-tree output is \nobvious enough that it's pretty trivial to grep for..\n\n\t\tLinus\n"},{"id":"539","messageId":"20050417214251.G13233@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171324340.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T20:42:51Z","receivedAt":"2005-04-17T20:42:51Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 01:26:27PM -0700, Linus Torvalds wrote:\n> \n> \n> On Sun, 17 Apr 2005, Russell King wrote:\n> > \n> > It's trivial to change - it's either sort -n or sort -nr !\n> \n> Yes.\n> \n> Btw, it might make sense to do the \"grep\" before the sort, since the sort \n> doesn't scale as nicely with lots of output. On the other hand, then the \n> grep pattern needs to be a bit more complex (so that it doesn't pick up on \n> things that have _parents_ with \":1\"). I think the rev-tree output is \n> obvious enough that it's pretty trivial to grep for..\n\nsed works as well - in fact, we can use it to both select lines based\nupon the :1 _and_ cut out the bulk of the stuff we're not interested in.\nSo:\n\nrev-tree $to $prev | \\\n sort -nr | cut -d' ' -f2 | grep :1 | cut -d: -f1\n\nbecomes:\n\nrev-tree $to $prev | \\\n sed -n 's,\\([[:digit:]]\\+\\)[[:space:]]\\+\\([[:xdigit:]]\\+\\):1.*,\\1 \\2,p' | \\\n sort -nr | cut -d' ' -f2\n\n(appologies to those who don't like regexps 8))\n\nOne thing to note about that sed expression though - using \\+ is a GNU\nextension - do we care about that? The portable way is to use \\{1,\\}\nbut that'd clutter it some more.\n\n-- \nRussell King\n\n"},{"id":"546","messageId":"4262DA30.2030500@roemling.net","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Jochen Roemling","fromEmail":"jochen@roemling.net","sentAt":"2005-04-17T21:50:40Z","receivedAt":"2005-04-17T21:50:40Z","isPatch":false,"sender":{"key":"jochen@roemling.net","avatar":"https://gravatar.com/avatar/fd825445a548573b8483227eea8a12d7d3e474693d15bc2b77d9ee7df8870948?d=mp&s=160"},"body":"Linus Torvalds wrote:\n\n>Ie we have two phases to the merge: first get the objects, with something\n>like\n>\n>\trepo=kernel.org:/pub/kernel/people/torvalds/linux-2.6.git\n>\trsync --ignore-existing -acv $(repo)/ .git/\n> \n>\nCould you place a tarball there for people like me who are no \"real\" \nkernel hackers and don't have a kernel.org account? Or is there an \n\"anonymous\" account that I'm just to ignorant to know of?\n\nI'm just somebody who is very interested in the new things happening \nhere, but I guess there will be some month to go and a lot of things to \nlearn until you see the first kernel patch that is signed off by me. :-)\n\nBy the way: Does the git repository include all 60.000 changes of just \nthe latest version of the 17.000 file in the kernel?\n\nBest regards\nJochen Römling\n"},{"id":"547","messageId":"1113774736.3884.4.camel@localhost.localdomain","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504161543590.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"David Woodhouse","fromEmail":"dwmw2@infradead.org","sentAt":"2005-04-17T21:52:15Z","receivedAt":"2005-04-17T21:52:15Z","isPatch":false,"sender":{"key":"dwmw2@infradead.org","avatar":"https://gravatar.com/avatar/7afd4f07e0cf7d7e046ae2d23678296b37777c96488e6f3451e78a5514154ebd?d=mp&s=160"},"body":"On Sat, 2005-04-16 at 16:01 -0700, Linus Torvalds wrote:\n> So I re-created the dang thing (hey, it takes just a few minutes), and\n> pushed it out, and there's now an archive on kernel.org in my public\n> \"personal\" directory called \"linux-2.6.git\". I'll continue the tradition\n> of naming git-archive directories as \"*.git\", since that really ends up\n> being the \".git\" directory for the checked-out thing.\n\nDo you want the commits list running for it yet? Do you want the\nchangesets which are already in it re-mailed without a 'TESTING' tag?\n\n-- \ndwmw2\n\n"},{"id":"550","messageId":"20050417150921.58d6db68.rddunlap@osdl.org","threadId":"58","inReplyTo":"4262DA30.2030500@roemling.net","subject":"Re: Re-done kernel archive - real one?","fromName":"Randy.Dunlap","fromEmail":"rddunlap@osdl.org","sentAt":"2005-04-17T22:09:21Z","receivedAt":"2005-04-17T22:09:21Z","isPatch":false,"sender":{"key":"rddunlap@osdl.org","avatar":null},"body":"On Sun, 17 Apr 2005 23:50:40 +0200 Jochen Roemling wrote:\n\n| Linus Torvalds wrote:\n| \n| >Ie we have two phases to the merge: first get the objects, with something\n| >like\n| >\n| >\trepo=kernel.org:/pub/kernel/people/torvalds/linux-2.6.git\n| >\trsync --ignore-existing -acv $(repo)/ .git/\n| > \n| >\n| Could you place a tarball there for people like me who are no \"real\" \n| kernel hackers and don't have a kernel.org account? Or is there an \n| \"anonymous\" account that I'm just to ignorant to know of?\n\nYou don't need a kernel.org account to rsync it... this works too:\n\nrsync -avz -e ssh --progress --ignore-existing rsync://rsync.kernel.org/pub/linux/kernel/people/torvalds/linux-2.6.git/ .git/\n\n| I'm just somebody who is very interested in the new things happening \n| here, but I guess there will be some month to go and a lot of things to \n| learn until you see the first kernel patch that is signed off by me. :-)\n| \n| By the way: Does the git repository include all 60.000 changes of just \n| the latest version of the 17.000 file in the kernel?\n\nIt's currently just 2.6.12-rc2 (17,000 files) plus changes...\n\nTwo people have announced full history gits if you are interested\nin that... Check the git email archives for mail from Ingo Molnar:\nSubject: full kernel history, in patchset format\n\nand Thomas Gleixner :\nSubject: BK -> git export done\n\n---\n~Randy\n"},{"id":"553","messageId":"Pine.LNX.4.58.0504171511210.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"1113774736.3884.4.camel@localhost.localdomain","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T22:17:50Z","receivedAt":"2005-04-17T22:17:50Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, David Woodhouse wrote:\n> \n> Do you want the commits list running for it yet? Do you want the\n> changesets which are already in it re-mailed without a 'TESTING' tag?\n\nI really don't know. I'm actually very happy where this thing is right\nnow, and completing that first merge successfully was a big milestone to\nme personally. That said, actually _using_ this thing is not for the\nfaint-of-heart, and while I think \"git\" already is showing itself to be\nuseful, I'm very very biased.\n\nIn other words, I really wonder what an outsider that doesn't have the\nsame kind of mental bias thinks of the current git tree. Is it useful, or\nis it still just a toy for Linus to test out his crazy SCM-wannabe.\n\nCan people usefully track my current kernel git repository, or do you have\nto be crazy to do so? That's really the question. You be the judge. Me,\nI'm just giddy from a merge that was clearly done using interfaces that\naren't actually really usable for anybody but me, and barely me at that ;)\n\n\t\tLinus\n\nBtw, I also do want this to show up in the BK trees for people who use \nBitKeeper - the same way we always supported tar-ball + patch users \nbefore. So I'll have to try to come up with some sane way to do that too. \nAny ideas? The first series of 198 patches is obvious enough and can be \njust done that way direcly, but the merge..\n"},{"id":"554","messageId":"20050417231959.A30656@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171511210.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T22:19:59Z","receivedAt":"2005-04-17T22:19:59Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 03:17:50PM -0700, Linus Torvalds wrote:\n> Can people usefully track my current kernel git repository, or do you have\n> to be crazy to do so? That's really the question. You be the judge. Me,\n> I'm just giddy from a merge that was clearly done using interfaces that\n> aren't actually really usable for anybody but me, and barely me at that ;)\n\nI guess I'll have the pleasure to find that out when I update my tree\nwith your latest changes... which I think is a project for tomorrow.\n\n-- \nRussell King\n\n"},{"id":"555","messageId":"4262E119.2000109@zytor.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171511210.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"H. Peter Anvin","fromEmail":"hpa@zytor.com","sentAt":"2005-04-17T22:20:09Z","receivedAt":"2005-04-17T22:20:09Z","isPatch":false,"sender":{"key":"hpa@zytor.com","avatar":null},"body":"I have set up /pub/scm/linux/kernel/git on kernel.org. Everyone who had \ndirectories in /pub/linux/kernel/people now have directories in \n/pub/scm/linux/kernel/git. For non-kernel trees it would probably be \nbetter to have different trees, however; I also would like to request \nthat git itself is moved to /pub/software/scm/git; I have created that \ndirectory and made it owned by Linus.\n\n\t-hpa\n"},{"id":"556","messageId":"20050417152218.3501f2f1.rdunlap@xenotime.net","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171511210.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"randy_dunlap","fromEmail":"rdunlap@xenotime.net","sentAt":"2005-04-17T22:22:18Z","receivedAt":"2005-04-17T22:22:18Z","isPatch":false,"sender":{"key":"rdunlap@xenotime.net","avatar":null},"body":"On Sun, 17 Apr 2005 15:17:50 -0700 (PDT) Linus Torvalds wrote:\n\n| \n| \n| On Mon, 18 Apr 2005, David Woodhouse wrote:\n| > \n| > Do you want the commits list running for it yet? Do you want the\n| > changesets which are already in it re-mailed without a 'TESTING' tag?\n| \n| I really don't know. I'm actually very happy where this thing is right\n| now, and completing that first merge successfully was a big milestone to\n| me personally. That said, actually _using_ this thing is not for the\n| faint-of-heart, and while I think \"git\" already is showing itself to be\n| useful, I'm very very biased.\n| \n| In other words, I really wonder what an outsider that doesn't have the\n| same kind of mental bias thinks of the current git tree. Is it useful, or\n| is it still just a toy for Linus to test out his crazy SCM-wannabe.\n| \n| Can people usefully track my current kernel git repository, or do you have\n| to be crazy to do so? That's really the question. You be the judge. Me,\n| I'm just giddy from a merge that was clearly done using interfaces that\n| aren't actually really usable for anybody but me, and barely me at that ;)\n\nTBD...\n\n| \t\tLinus\n| \n| Btw, I also do want this to show up in the BK trees for people who use \n| BitKeeper - the same way we always supported tar-ball + patch users \n| before. So I'll have to try to come up with some sane way to do that too. \n| Any ideas? The first series of 198 patches is obvious enough and can be \n| just done that way direcly, but the merge..\n\nDavid did the commits-mailing-list script and I'm working on a\ncommits web-page like what was formerly seen at:\nhttp://www.kernel.org/pub/linux/kernel/v2.6/testing/cset/\n(with daily tarball)\n\nbased on some older scripts from David, however I'm wondering if\na variant of the gitlog.sh script wouldn't be a better starting\npoint for it.\n\n---\n~Randy\n"},{"id":"557","messageId":"20050417223006.GN1461@pasky.ji.cz","threadId":"58","inReplyTo":"20050417150921.58d6db68.rddunlap@osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-17T22:30:06Z","receivedAt":"2005-04-17T22:30:06Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Mon, Apr 18, 2005 at 12:09:21AM CEST, I got a letter\nwhere \"Randy.Dunlap\" told me that...\n> On Sun, 17 Apr 2005 23:50:40 +0200 Jochen Roemling wrote:\n> \n> | Linus Torvalds wrote:\n> | \n> | >Ie we have two phases to the merge: first get the objects, with something\n> | >like\n> | >\n> | >\trepo=kernel.org:/pub/kernel/people/torvalds/linux-2.6.git\n> | >\trsync --ignore-existing -acv $(repo)/ .git/\n> | > \n> | >\n> | Could you place a tarball there for people like me who are no \"real\" \n> | kernel hackers and don't have a kernel.org account? Or is there an \n> | \"anonymous\" account that I'm just to ignorant to know of?\n> \n> You don't need a kernel.org account to rsync it... this works too:\n> \n> rsync -avz -e ssh --progress --ignore-existing rsync://rsync.kernel.org/pub/linux/kernel/people/torvalds/linux-2.6.git/ .git/\n\nOr\n\n\tmkdir linux-2.6.git\n\tcd linux-2.6.git\n\tRSYNC_FLAGS=\"--progress --stats\" git init rsync://rsync.kernel.org/pub/linux/kernel/people/torvalds/linux-2.6.git/\n\nwhich also does the initial checkout for you. ;-) (Requires latest\ngit-pasky, though.) Then you can bring the latest and greatest to your\ntree by a mere git pull.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"562","messageId":"20050417235136.B30656@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"20050417231959.A30656@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-17T22:51:36Z","receivedAt":"2005-04-17T22:51:36Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 11:19:59PM +0100, Russell King wrote:\n> On Sun, Apr 17, 2005 at 03:17:50PM -0700, Linus Torvalds wrote:\n> > Can people usefully track my current kernel git repository, or do you have\n> > to be crazy to do so? That's really the question. You be the judge. Me,\n> > I'm just giddy from a merge that was clearly done using interfaces that\n> > aren't actually really usable for anybody but me, and barely me at that ;)\n> \n> I guess I'll have the pleasure to find that out when I update my tree\n> with your latest changes... which I think is a project for tomorrow.\n\nI pulled it tonight into a pristine tree (which of course worked.)\n\nIn doing so, I noticed that I'd messed up one of the commits - there's\na missing new file. Grr. I'll put that down to being a newbie git.\n\n-- \nRussell \"newbie git\" King\n"},{"id":"564","messageId":"1113780105.11910.4.camel@localhost.localdomain","threadId":"58","inReplyTo":"20050417152218.3501f2f1.rdunlap@xenotime.net","subject":"Re: Re-done kernel archive - real one?","fromName":"David Woodhouse","fromEmail":"dwmw2@infradead.org","sentAt":"2005-04-17T23:21:45Z","receivedAt":"2005-04-17T23:21:45Z","isPatch":false,"sender":{"key":"dwmw2@infradead.org","avatar":"https://gravatar.com/avatar/7afd4f07e0cf7d7e046ae2d23678296b37777c96488e6f3451e78a5514154ebd?d=mp&s=160"},"body":"On Sun, 2005-04-17 at 15:22 -0700, randy_dunlap wrote:\n> David did the commits-mailing-list script and I'm working on a\n> commits web-page like what was formerly seen at:\n> http://www.kernel.org/pub/linux/kernel/v2.6/testing/cset/\n> (with daily tarball)\n> \n> based on some older scripts from David, however I'm wondering if\n> a variant of the gitlog.sh script wouldn't be a better starting\n> point for it.\n\nMy commits-list script is in fact based on gitlog.sh. You'll probably\nfind useful things to crib from in both that and the original\nbkexport.sh script.\n\nThe commits script also wants updating to print the date properly now\nthat we've changed how it's stored -- I'll try to find some time this\nweek to update it and set it running on master.kernel.org again, but it\nmay end up waiting till after LCA.\n\n-- \ndwmw2\n\n"},{"id":"565","messageId":"Pine.LNX.4.58.0504171621330.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050417235136.B30656@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-17T23:24:24Z","receivedAt":"2005-04-17T23:24:24Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Sun, 17 Apr 2005, Russell King wrote:\n> \n> I pulled it tonight into a pristine tree (which of course worked.)\n\nGoodie.\n\n> In doing so, I noticed that I'd messed up one of the commits - there's\n> a missing new file. Grr. I'll put that down to being a newbie git.\n\nActually, you should put that down to horribly bad interface tools. With\nBK, we had these nice tools that pointed out that there were files that\nyou might want to commit (ie \"bk citool\"), and made this very obvious.\n\nTools absolutely matter. And it will take time for us to build up that \nkind of helper infrastructure. So being newbie might be part of it, but \nit's the smaller part, I say. Rough interfaces is a big issue.\n\n\t\tLinus\n"},{"id":"595","messageId":"20050417183344.50145c5b.rdunlap@xenotime.net","threadId":"58","inReplyTo":"1113780105.11910.4.camel@localhost.localdomain","subject":"Re: Re-done kernel archive - real one?","fromName":"randy_dunlap","fromEmail":"rdunlap@xenotime.net","sentAt":"2005-04-18T01:33:44Z","receivedAt":"2005-04-18T01:33:44Z","isPatch":false,"sender":{"key":"rdunlap@xenotime.net","avatar":null},"body":"On Mon, 18 Apr 2005 09:21:45 +1000 David Woodhouse wrote:\n\n| On Sun, 2005-04-17 at 15:22 -0700, randy_dunlap wrote:\n| > David did the commits-mailing-list script and I'm working on a\n| > commits web-page like what was formerly seen at:\n| > http://www.kernel.org/pub/linux/kernel/v2.6/testing/cset/\n| > (with daily tarball)\n| > \n| > based on some older scripts from David, however I'm wondering if\n| > a variant of the gitlog.sh script wouldn't be a better starting\n| > point for it.\n| \n| My commits-list script is in fact based on gitlog.sh. You'll probably\n| find useful things to crib from in both that and the original\n| bkexport.sh script.\n\nYep, already had all of that in progress.\n\n| The commits script also wants updating to print the date properly now\n| that we've changed how it's stored -- I'll try to find some time this\n| week to update it and set it running on master.kernel.org again, but it\n| may end up waiting till after LCA.\n\nHere's a working gitgenwebpage.sh script, although it may be possible\nto do it better/faster, and merge it with the commits mail script\n(as you mentioned).\n\n\n---\n~Randy\n"},{"id":"636","messageId":"20050418102332.A21081@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171621330.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-18T09:23:32Z","receivedAt":"2005-04-18T09:23:32Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> On Sun, 17 Apr 2005, Russell King wrote:\n> > I pulled it tonight into a pristine tree (which of course worked.)\n> \n> Goodie.\n\nNote the \"pristine\". Now comes the real test...\n\n> > In doing so, I noticed that I'd messed up one of the commits - there's\n> > a missing new file. Grr. I'll put that down to being a newbie git.\n> \n> Actually, you should put that down to horribly bad interface tools. With\n> BK, we had these nice tools that pointed out that there were files that\n> you might want to commit (ie \"bk citool\"), and made this very obvious.\n> \n> Tools absolutely matter. And it will take time for us to build up that \n> kind of helper infrastructure. So being newbie might be part of it, but \n> it's the smaller part, I say. Rough interfaces is a big issue.\n\nOk, I just tried pulling your tree into the tree you pulled from, and\ngot this:\n\nTree change: e7905b2f22eb5d5308c9122b9c06c2d02473dd4f ee423ea56280512778a5961ee58a785a73acb7d1\n...\n*100644->100644 blob 46f0a3caae02b4bb8f903d7ac86456aa0c37954b->ba4afd7956173b6f89eb6b0b9ad23b392d5c0aee arch/arm/kernel/process.c\n*100644->100644 blob 4a36fa7192e11df36f5e0928b064239dabe1e305->ec0bc8f315ab5d78a4220e176e7aee76d52d1c74 arch/arm/kernel/traps.c\n*100644->100644 blob 311d19ee00208faf02359f9e7c5394577a40f253->bf923a953703c6ca0c88eac3b2850cf07b838996 arch/arm/lib/changebit.S\n*100644->100644 blob c07afa31695654e6489ec59c3f837183b325e9da->41f89b3a393d5af939f04f63c5bf4991b2bf6599 arch/arm/lib/clearbit.S\n...\nTracked branch, applying changes...\nMerging e7905b2f22eb5d5308c9122b9c06c2d02473dd4f -> ee423ea56280512778a5961ee58a785a73acb7d1\n to df4449813c900973841d0fa5a9e9bc7186956e1e...\nCOPYING: needs update\nCREDITS: needs update\nDocumentation/00-INDEX: needs update\nDocumentation/BK-usage/00-INDEX: needs update\n...\npatching file arch/arm/kernel/process.c\nReversed (or previously applied) patch detected! Skipping patch.\n2 out of 2 hunks ignored -- saving rejects to file arch/arm/kernel/process.c.rejpatching file arch/arm/kernel/traps.c\nReversed (or previously applied) patch detected! Skipping patch.\n3 out of 3 hunks ignored -- saving rejects to file arch/arm/kernel/traps.c.rej\npatching file arch/arm/lib/changebit.S\nReversed (or previously applied) patch detected! Skipping patch.\n2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/changebit.S.rej\npatching file arch/arm/lib/clearbit.S\nReversed (or previously applied) patch detected! Skipping patch.\n2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/clearbit.S.rej\n\nso obviously git pull isn't able to indentify what's already in the\nlocal repository.\n\nInterestingly, the files listed above as having rejects are excluded\nfrom the list of \"needs update\". And I don't know why git is staying\nthat these files need updating, because they haven't changed since\nthey were initially checked out.\n\nThis was with some random version of git-pasky-0.04. Unfortunately,\nthis version doesn't have the sha1 ID appended, so I couldn't say\ndefinitively that it's the latest and greatest. It might be a day\nold.\n\n-- \nRussell King\n\n"},{"id":"646","messageId":"1113822890.16288.19.camel@nosferatu.lan","threadId":"58","inReplyTo":"20050418102332.A21081@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Martin Schlemmer","fromEmail":"azarah@nosferatu.za.org","sentAt":"2005-04-18T11:14:50Z","receivedAt":"2005-04-18T11:14:50Z","isPatch":false,"sender":{"key":"azarah@nosferatu.za.org","avatar":null},"body":"On Mon, 2005-04-18 at 10:23 +0100, Russell King wrote:\n> On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> > On Sun, 17 Apr 2005, Russell King wrote:\n> > > I pulled it tonight into a pristine tree (which of course worked.)\n> > \n> > Goodie.\n> \n> Note the \"pristine\". Now comes the real test...\n> \n> > > In doing so, I noticed that I'd messed up one of the commits - there's\n> > > a missing new file. Grr. I'll put that down to being a newbie git.\n> > \n> > Actually, you should put that down to horribly bad interface tools. With\n> > BK, we had these nice tools that pointed out that there were files that\n> > you might want to commit (ie \"bk citool\"), and made this very obvious.\n> > \n> > Tools absolutely matter. And it will take time for us to build up that \n> > kind of helper infrastructure. So being newbie might be part of it, but \n> > it's the smaller part, I say. Rough interfaces is a big issue.\n> \n> Ok, I just tried pulling your tree into the tree you pulled from, and\n> got this:\n> \n> Tree change: e7905b2f22eb5d5308c9122b9c06c2d02473dd4f ee423ea56280512778a5961ee58a785a73acb7d1\n> ...\n> *100644->100644 blob 46f0a3caae02b4bb8f903d7ac86456aa0c37954b->ba4afd7956173b6f89eb6b0b9ad23b392d5c0aee arch/arm/kernel/process.c\n> *100644->100644 blob 4a36fa7192e11df36f5e0928b064239dabe1e305->ec0bc8f315ab5d78a4220e176e7aee76d52d1c74 arch/arm/kernel/traps.c\n> *100644->100644 blob 311d19ee00208faf02359f9e7c5394577a40f253->bf923a953703c6ca0c88eac3b2850cf07b838996 arch/arm/lib/changebit.S\n> *100644->100644 blob c07afa31695654e6489ec59c3f837183b325e9da->41f89b3a393d5af939f04f63c5bf4991b2bf6599 arch/arm/lib/clearbit.S\n> ...\n> Tracked branch, applying changes...\n> Merging e7905b2f22eb5d5308c9122b9c06c2d02473dd4f -> ee423ea56280512778a5961ee58a785a73acb7d1\n> to df4449813c900973841d0fa5a9e9bc7186956e1e...\n> COPYING: needs update\n> CREDITS: needs update\n> Documentation/00-INDEX: needs update\n> Documentation/BK-usage/00-INDEX: needs update\n> ...\n> patching file arch/arm/kernel/process.c\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/kernel/process.c.rejpatching file arch/arm/kernel/traps.c\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 3 out of 3 hunks ignored -- saving rejects to file arch/arm/kernel/traps.c.rej\n> patching file arch/arm/lib/changebit.S\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/changebit.S.rej\n> patching file arch/arm/lib/clearbit.S\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/clearbit.S.rej\n> \n> so obviously git pull isn't able to indentify what's already in the\n> local repository.\n> \n> Interestingly, the files listed above as having rejects are excluded\n> from the list of \"needs update\". And I don't know why git is staying\n> that these files need updating, because they haven't changed since\n> they were initially checked out.\n> \n> This was with some random version of git-pasky-0.04. Unfortunately,\n> this version doesn't have the sha1 ID appended, so I couldn't say\n> definitively that it's the latest and greatest. It might be a day\n> old.\n> \n\ngitmerge.sh does not yet have support for the new merge stuff as far as\nI know, and if it does, then its a very recent version (ie, one that\nhave the sha1 ID appended).\n\n\n-- \nMartin Schlemmer\n\n"},{"id":"647","messageId":"1113822925.16288.22.camel@nosferatu.lan","threadId":"58","inReplyTo":"20050417192000.C13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Martin Schlemmer","fromEmail":"azarah@nosferatu.za.org","sentAt":"2005-04-18T11:15:25Z","receivedAt":"2005-04-18T11:15:25Z","isPatch":false,"sender":{"key":"azarah@nosferatu.za.org","avatar":null},"body":"On Sun, 2005-04-17 at 19:20 +0100, Russell King wrote:\n> On Sun, Apr 17, 2005 at 02:13:59PM -0400, David A. Wheeler wrote:\n> > On Sun, 17 Apr 2005, Russell King wrote:\n> > >>BTW, there appears to be \"errors\" in the history committed thus far.\n> > >>I'm not sure where this came from though. Some of them could be\n> > >>UTF8 vs ASCII issues, ....> \n> > ...\n> > >>One thing which definitely needs to be considered is - what character\n> > >>encoding are the comments to be stored as?\n> > \n> > Linus Torvalds replied:\n> > > To git, it's just a byte stream, and you can have binary comments if you\n> > > want to. I personally would prefer to move towards UTF eventually, but I\n> > > really don't think it matters a whole lot as long as 99.9% of everything\n> > > we'd see there is still 7-bit ascii.\n> > \n> > I would _heartily_ recommend moving towards UTF-8 as the\n> > internal charset for all comments. Alternatives are possible\n> > (e.g., recording the charset in the header), but they're\n> > incredibly messy. Even if you don't normally work in UTF-8,\n> > it's pretty easy to set most editors up to read & write UTF-8.\n> > Having the data stored as a constant charset eliminates\n> > a raft of error-prone code.\n> \n> Except, I believe, MicroEMACS, which both Linus and myself use. As\n> far as I know, there aren't any patches to make it UTF-8 compliant.\n> \n> The alternative is, I suppose, iconv. However, iconv in _my_ glibc\n> seems buggy (segfaults) and my efforts for building glibc 2.3.2 for\n> ARM have failed. Effectively that means iconv is inaccessible to\n> me.\n> \n\nOT, and probably not much help, but glibc-2.3.5 is out ...\n\n\n-- \nMartin Schlemmer\n\n"},{"id":"648","messageId":"20050418111525.GN1461@pasky.ji.cz","threadId":"58","inReplyTo":"20050418102332.A21081@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T11:15:25Z","receivedAt":"2005-04-18T11:15:25Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Mon, Apr 18, 2005 at 11:23:32AM CEST, I got a letter\nwhere Russell King told me that...\n> Ok, I just tried pulling your tree into the tree you pulled from, and\n> got this:\n> \n> Tree change: e7905b2f22eb5d5308c9122b9c06c2d02473dd4f ee423ea56280512778a5961ee58a785a73acb7d1\n> ...\n> *100644->100644 blob 46f0a3caae02b4bb8f903d7ac86456aa0c37954b->ba4afd7956173b6f89eb6b0b9ad23b392d5c0aee arch/arm/kernel/process.c\n> *100644->100644 blob 4a36fa7192e11df36f5e0928b064239dabe1e305->ec0bc8f315ab5d78a4220e176e7aee76d52d1c74 arch/arm/kernel/traps.c\n> *100644->100644 blob 311d19ee00208faf02359f9e7c5394577a40f253->bf923a953703c6ca0c88eac3b2850cf07b838996 arch/arm/lib/changebit.S\n> *100644->100644 blob c07afa31695654e6489ec59c3f837183b325e9da->41f89b3a393d5af939f04f63c5bf4991b2bf6599 arch/arm/lib/clearbit.S\n> ...\n> Tracked branch, applying changes...\n> Merging e7905b2f22eb5d5308c9122b9c06c2d02473dd4f -> ee423ea56280512778a5961ee58a785a73acb7d1\n> to df4449813c900973841d0fa5a9e9bc7186956e1e...\n> COPYING: needs update\n> CREDITS: needs update\n> Documentation/00-INDEX: needs update\n> Documentation/BK-usage/00-INDEX: needs update\n> ...\n> patching file arch/arm/kernel/process.c\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/kernel/process.c.rejpatching file arch/arm/kernel/traps.c\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 3 out of 3 hunks ignored -- saving rejects to file arch/arm/kernel/traps.c.rej\n> patching file arch/arm/lib/changebit.S\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/changebit.S.rej\n> patching file arch/arm/lib/clearbit.S\n> Reversed (or previously applied) patch detected! Skipping patch.\n> 2 out of 2 hunks ignored -- saving rejects to file arch/arm/lib/clearbit.S.rej\n> \n> so obviously git pull isn't able to indentify what's already in the\n> local repository.\n\nFor unknown reason, patch sucks at this, even if I pass it\n\n-N or --forward\n Ignore patches that seem to be reversed or already applied. See also -R.\n\n:-(\n\ngit merge does not use the in-GIT merging capabilities yet.\n\nCould you do\n\n\tmerge-base $(tree-id ee423ea56280512778a5961ee58a785a73acb7d1) \\\n\t\t$(tree-id df4449813c900973841d0fa5a9e9bc7186956e1e)\n\nand check if it returns e7905b2f22eb5d5308c9122b9c06c2d02473dd4f, please?\nI guess it won't. So you could try to wipe the -b \"$orig_head\" from\ngitmerge.sh invocation in gitpull.sh.\n\n> Interestingly, the files listed above as having rejects are excluded\n> from the list of \"needs update\". And I don't know why git is staying\n> that these files need updating, because they haven't changed since\n> they were initially checked out.\n\nBecause it is checking out only the files which changed (pass -a to git\nmerge to check out everything; I will probably make git pull take a -m\nargument which will take merge arguments ;). And then it does\nupdate-cache --refresh which complains about missing files. :/\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"658","messageId":"20050418142243.GR1461@pasky.ji.cz","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171621330.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T14:22:43Z","receivedAt":"2005-04-18T14:22:43Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Mon, Apr 18, 2005 at 01:24:24AM CEST, I got a letter\nwhere Linus Torvalds told me that...\n> \n> \n> On Sun, 17 Apr 2005, Russell King wrote:\n> > \n> > I pulled it tonight into a pristine tree (which of course worked.)\n> \n> Goodie.\n> \n> > In doing so, I noticed that I'd messed up one of the commits - there's\n> > a missing new file. Grr. I'll put that down to being a newbie git.\n> \n> Actually, you should put that down to horribly bad interface tools. With\n> BK, we had these nice tools that pointed out that there were files that\n> you might want to commit (ie \"bk citool\"), and made this very obvious.\n> \n> Tools absolutely matter. And it will take time for us to build up that \n> kind of helper infrastructure. So being newbie might be part of it, but \n> it's the smaller part, I say. Rough interfaces is a big issue.\n\nI just committed some simple git status, which is equivalent to svn\nstatus or cvs update (except it does no update). So it shows all the\nfiles not tracked by git with a question mark in front of them.\n\nThis will need some ignore rules, though (currently it just ignores *.o\nand the tags file). Now it turns out that it is rather unfortunate that\ngit ignores hidden files, since this would be a perfect object for that\n- I think it is useful to have the ignore list tracked by git. I think I\nwill just name it git-ignores to be found in the working directory for\nnow.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"659","messageId":"20050418150456.GC12750@kroah.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504171621330.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Greg KH","fromEmail":"greg@kroah.com","sentAt":"2005-04-18T15:04:57Z","receivedAt":"2005-04-18T15:04:57Z","isPatch":false,"sender":{"key":"greg@kroah.com","avatar":"https://gravatar.com/avatar/5bb5aa0cc2e01c00ec899d11130c07796bc186e465bae57bc34873b13b72c7c8?d=mp&s=160"},"body":"On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> \n> Tools absolutely matter. And it will take time for us to build up that \n> kind of helper infrastructure. So being newbie might be part of it, but \n> it's the smaller part, I say. Rough interfaces is a big issue.\n\nSpeaking of tools, you had a \"dotest\" program to apply patches in email\nform to a bk tree. And from what I can gather, you've changed that to\nhandle git archives, right? Any pointers to where I can find this so I\ncould try to build up some git trees for you to merge with? I think I\ncan even make a tree with a merge issue if you want to test that out :)\n\nthanks,\n\ngreg k-h\n"},{"id":"660","messageId":"Pine.LNX.4.58.0504180802060.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418102332.A21081@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T15:23:16Z","receivedAt":"2005-04-18T15:23:16Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Russell King wrote:\n> \n> Ok, I just tried pulling your tree into the tree you pulled from, and\n> got this:\n\nNo, that can't work. The pesky tools are helpful, but they really don't do \nmerges worth cr*p right now, excuse my french. \n\nThe _real_ way to pull is to do the (horribly complex) thing I described\nby the merge, but noticing that one of the commits you are merging is a\nproper subset of the other one, and just updating the head instead of\nactually doing a real merge (ie skipping the \"read-tree -m\" and\n\"write-tree\" phases).\n\n> This was with some random version of git-pasky-0.04. Unfortunately,\n> this version doesn't have the sha1 ID appended, so I couldn't say\n> definitively that it's the latest and greatest. It might be a day\n> old.\n\nI'm afraid that until Pasky's tools script this properly, a \"pull\" really \nends up being something like this (which _can_ be scripted, never fear):\n\nNOTE NOTE NOTE! This is untested! I'm writing this within the email \neditor, so do _not_ do this on a tree that you care about.\n\n\t#!/bin/sh\n\t#\n\t# use \"$1\" or something in a real script, this \n\t# just hard-codes it.\n\t#\n\tmerge_repo=master.kernel.org:/pub/linux/kernel/people/torvalds/linux-2.6.git\n\n\techo \"Getting object database\"\n\trsync -avz --ignore-existing $merge_repo/ .git/\n\n\techo \"Getting remote head\"\n\trsync -avz $merge_repo/HEAD .git/MERGE_HEAD\n\n\thead=$(cat .git/HEAD)\n\tmerge_head=$(cat .git/MERGE-HEAD)\n\tcommon=$(merge-base $head $merge_head)\n\tif [ -z \"$common\" ]; then\n\t\techo \"Unable to find common commit between\" $merge_head $head\n\t\texit 1\n\tfi\n\n\t# Get the trees associated with those commits\n\tcommon_tree=tree=$(cat-file commit $common | sed 's/tree //;q')\n\thead_tree=tree=$(cat-file commit $head | sed 's/tree //;q')\n\tmerge_tree=tree=$(cat-file commit $merge | sed 's/tree //;q')\n\n\tif [ \"$common\" == \"$merge_head\" ]; then\n\t\techo \"Already up-to-date. Yeeah!\"\n\t\texit 0\n\tfi\n\tif [ \"$common\" == \"$head\" ]; then\n\t\techo \"Updating from $head to $merge_head.\"\n\t\techo \"Destroying all noncommitted data!\"\n\t\techo \"Kill me within 3 seconds..\"\n\t\tsleep 3\n\t\tread-tree $merge_tree && checkout-cache -f -a\n\t\techo $merge_head > .git/HEAD\n\t\texit 0\n\tfi\n\techo \"Trying to merge $merge_head into $head\"\n\tread-tree -m $common_tree $head_tree $merge_tree\n\tresult_tree=$(write-tree) || exit 1\n\tresult_commit=$(echo \"Merge $merge_repo\" | commit-tree $result_tree -p $head -p $merge_head)\n\techo \"Committed merge $result_commit\"\n\techo $result_commit > .git/HEAD\n\tread-tree $result_tree && checkout-cache -f -a\n\nThe above looks like it might work, but I also warn you: it's not only\nuntested, but it's pretty fragile in that if something breaks, you are\nprobably left with a mess. I _tried_ to do the right thing, but... So it\nobviously will need testing, tweaking and just general tender loving care.\n\nAnd if the merge isn't clean, it will exit early thanks to the\n\n\twrite-tree || exit 1\n\nand now you have to resolve the merge yourself. There are tools to help\nyou do so automatically, but that's really a separate script.\n\nYou shouldn't hit the \"merge\" case at all right now, you should hit the \n\"Updating from $head to $merge_head\" thing.\n\nIf Pesky wants to take the above script, test it, and see if it works,\nthat would be good. It's definitely a much better \"pull\" than trying to\napply the patches forward..\n\n\t\tLinus\n"},{"id":"662","messageId":"20050418082551.69a20aeb.rddunlap@osdl.org","threadId":"58","inReplyTo":"20050418150456.GC12750@kroah.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Randy.Dunlap","fromEmail":"rddunlap@osdl.org","sentAt":"2005-04-18T15:25:51Z","receivedAt":"2005-04-18T15:25:51Z","isPatch":false,"sender":{"key":"rddunlap@osdl.org","avatar":null},"body":"On Mon, 18 Apr 2005 08:04:57 -0700 Greg KH wrote:\n\n| On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n| > \n| > Tools absolutely matter. And it will take time for us to build up that \n| > kind of helper infrastructure. So being newbie might be part of it, but \n| > it's the smaller part, I say. Rough interfaces is a big issue.\n| \n| Speaking of tools, you had a \"dotest\" program to apply patches in email\n| form to a bk tree. And from what I can gather, you've changed that to\n| handle git archives, right? Any pointers to where I can find this so I\n| could try to build up some git trees for you to merge with? I think I\n| can even make a tree with a merge issue if you want to test that out :)\n\nit's at\nhttp://www.kernel.org/pub/linux/kernel/people/torvalds/git-tools.git/\n\n---\n~Randy\n"},{"id":"665","messageId":"Pine.LNX.4.58.0504180840370.7211@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418150456.GC12750@kroah.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T15:42:14Z","receivedAt":"2005-04-18T15:42:14Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Greg KH wrote:\n>\n> On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> > \n> > Tools absolutely matter. And it will take time for us to build up that \n> > kind of helper infrastructure. So being newbie might be part of it, but \n> > it's the smaller part, I say. Rough interfaces is a big issue.\n> \n> Speaking of tools, you had a \"dotest\" program to apply patches in email\n> form to a bk tree. And from what I can gather, you've changed that to\n> handle git archives, right?\n\nYup.\n\nIt's a git archive at \n\n\tkernel.org:/pub/linux/kernel/people/torvalds/git-tools.git\n\nand it seems to work. It's what I've used for all the kernel patches \n(except for the merge), and it's what I use for the git stuff that shows \nup as authored by others.\n\n\t\tLinus\n"},{"id":"671","messageId":"Pine.LNX.4.58.0504181003030.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504180802060.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T17:05:19Z","receivedAt":"2005-04-18T17:05:19Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Linus Torvalds wrote:\n> \n> No, that can't work. The pesky tools are helpful [...]\n> I'm afraid that until Pasky's tools script this properly, [... ]\n> If Pesky wants to take the above script, test it, [...]\n\nOk, one out of three isn't too bad, is it? Pesky/Pasky, so close yet so \nfar. Sorry,\n\n\t\tLinus\n"},{"id":"673","messageId":"20050418180737.GA5554@pasky.ji.cz","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504181003030.15725@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T18:07:37Z","receivedAt":"2005-04-18T18:07:37Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Mon, Apr 18, 2005 at 07:05:19PM CEST, I got a letter\nwhere Linus Torvalds told me that...\n> \n> \n> On Mon, 18 Apr 2005, Linus Torvalds wrote:\n> > \n> > No, that can't work. The pesky tools are helpful [...]\n> > I'm afraid that until Pasky's tools script this properly, [... ]\n> > If Pesky wants to take the above script, test it, [...]\n> \n> Ok, one out of three isn't too bad, is it? Pesky/Pasky, so close yet so \n> far. Sorry,\n\nNo problem. :-) Or you can just call me Petr if you want. ;-)\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"697","messageId":"20050418223359.A16789@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"20050418102332.A21081@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-18T21:33:59Z","receivedAt":"2005-04-18T21:33:59Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Mon, Apr 18, 2005 at 10:23:32AM +0100, Russell King wrote:\n> On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> > On Sun, 17 Apr 2005, Russell King wrote:\n> > > I pulled it tonight into a pristine tree (which of course worked.)\n> > \n> > Goodie.\n> \n> Note the \"pristine\". Now comes the real test...\n> \n> > > In doing so, I noticed that I'd messed up one of the commits - there's\n> > > a missing new file. Grr. I'll put that down to being a newbie git.\n> > \n> > Actually, you should put that down to horribly bad interface tools. With\n> > BK, we had these nice tools that pointed out that there were files that\n> > you might want to commit (ie \"bk citool\"), and made this very obvious.\n> > \n> > Tools absolutely matter. And it will take time for us to build up that \n> > kind of helper infrastructure. So being newbie might be part of it, but \n> > it's the smaller part, I say. Rough interfaces is a big issue.\n> \n> Ok, I just tried pulling your tree into the tree you pulled from, and\n> got this:\n>...\n\nSince this happened, I've been working out what state my tree is in,\nand I restored it back to a state where I had one dangling commit head,\nwhich was _my_ head.\n\nI then checked whether my objects matched the objects which I uploaded\nto master.kernel.org, and discovered I'd removed some extra ones. With\nthem restored, I have an additional dangling commit.\n\nNow, I'm pretty sure that I had an up to date tree when I did the\noriginal commits, so I'm a little confused.\n\nWhat I'm seeing is:\n\nb4a9a5114b3c6da131a832a8e2cd1941161eb348\n+- e7905b2f22eb5d5308c9122b9c06c2d02473dd4f\n +- dc90c0db0dd5214aca5304fd17ccd741031e5493 <-- extra dangling head\n +- 488faba31f59c5960aabbb2a5877a0f2923937a3\n +- 5d9a545981893629c8f95e2b8b50d15d18c6ddbc\n +- d5922e9c35d21f0b6b82d1fd8b1444cfce57ca34\n +- ff219d69be01af1fd04ada305b5fe7cd4c563cc6\n +- df4449813c900973841d0fa5a9e9bc7186956e1e <-- my head\n\nIt's very much like I somehow committed against the _parent_ of the\nhead, rather than the head itself.\n\nHowever, I've lost the state that this tree was in when I did the initial\ncommit, so who knows why this happened... I think it's something to\nkeep an eye out for though.\n\n-- \nRussell King\n\n"},{"id":"699","messageId":"20050418225356.B16789@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504180802060.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-18T21:53:57Z","receivedAt":"2005-04-18T21:53:57Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Mon, Apr 18, 2005 at 08:23:16AM -0700, Linus Torvalds wrote:\n> NOTE NOTE NOTE! This is untested! I'm writing this within the email \n> editor, so do _not_ do this on a tree that you care about.\n\nIt did the right thing for me at least!\n\n> \t#!/bin/sh\n> \t#\n> \t# use \"$1\" or something in a real script, this \n> \t# just hard-codes it.\n> \t#\n> \tmerge_repo=master.kernel.org:/pub/linux/kernel/people/torvalds/linux-2.6.git\n> \n> \techo \"Getting object database\"\n> \trsync -avz --ignore-existing $merge_repo/ .git/\n> \n> \techo \"Getting remote head\"\n> \trsync -avz $merge_repo/HEAD .git/MERGE_HEAD\n> \n> \thead=$(cat .git/HEAD)\n> \tmerge_head=$(cat .git/MERGE-HEAD)\n> \tcommon=$(merge-base $head $merge_head)\n> \tif [ -z \"$common\" ]; then\n> \t\techo \"Unable to find common commit between\" $merge_head $head\n> \t\texit 1\n> \tfi\n> \n> \t# Get the trees associated with those commits\n> \tcommon_tree=tree=$(cat-file commit $common | sed 's/tree //;q')\n> \thead_tree=tree=$(cat-file commit $head | sed 's/tree //;q')\n> \tmerge_tree=tree=$(cat-file commit $merge | sed 's/tree //;q')\n\nThis wants to be:\n\n\tcommon_tree=$(cat-file commit $common | sed 's/tree //;q')\n\thead_tree=$(cat-file commit $head | sed 's/tree //;q')\n\tmerge_tree=$(cat-file commit $merge_head | sed 's/tree //;q')\n\n> \tif [ \"$common\" == \"$merge_head\" ]; then\n> \t\techo \"Already up-to-date. Yeeah!\"\n> \t\texit 0\n> \tfi\n> \tif [ \"$common\" == \"$head\" ]; then\n> \t\techo \"Updating from $head to $merge_head.\"\n> \t\techo \"Destroying all noncommitted data!\"\n> \t\techo \"Kill me within 3 seconds..\"\n> \t\tsleep 3\n> \t\tread-tree $merge_tree && checkout-cache -f -a\n\nDon't we want to do an update-cache --refresh here?\n\n> \t\techo $merge_head > .git/HEAD\n> \t\texit 0\n> \tfi\n> \techo \"Trying to merge $merge_head into $head\"\n> \tread-tree -m $common_tree $head_tree $merge_tree\n> \tresult_tree=$(write-tree) || exit 1\n> \tresult_commit=$(echo \"Merge $merge_repo\" | commit-tree $result_tree -p $head -p $merge_head)\n> \techo \"Committed merge $result_commit\"\n> \techo $result_commit > .git/HEAD\n> \tread-tree $result_tree && checkout-cache -f -a\n> \n> The above looks like it might work, but I also warn you: it's not only\n> untested, but it's pretty fragile in that if something breaks, you are\n> probably left with a mess. I _tried_ to do the right thing, but... So it\n> obviously will need testing, tweaking and just general tender loving care.\n\nMaybe Petr can improve the error handling, and incorporate it (or at\nleast some of it) into git-pasky\n\n> You shouldn't hit the \"merge\" case at all right now, you should hit the \n> \"Updating from $head to $merge_head\" thing.\n\nExactly what happened. Thanks.\n\n-- \nRussell King\n\n"},{"id":"700","messageId":"Pine.LNX.4.58.0504181440020.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418223359.A16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T21:56:47Z","receivedAt":"2005-04-18T21:56:47Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Russell King wrote:\n> \n> Since this happened, I've been working out what state my tree is in,\n> and I restored it back to a state where I had one dangling commit head,\n> which was _my_ head.\n\nFor the future, if your tree gets messed up to the point where you say \n\"screw it\" and just want to go back in time, you can do this (it's \nequivalent to \"undo\" in BK speak):\n\n\tgit log | less -S\n\n\t.. find which HEAD it was that you trusted..\n\nIn this case your HEAD before I merged with it was this one:\n\n\tdf4449813c900973841d0fa5a9e9bc7186956e1e\n\nSo to get back to that one, you can do\n\n\techo df4449813c900973841d0fa5a9e9bc7186956e1e > .git/HEAD\n\nand now\n\n\tcat-file commit $(cat .git/HEAD) | head -1\n\ngives you\n\n\ttree a43c4447b2edc9fb01a6369f10c1165de4494c88\n\nso you can restore your checked-out state with\n\n\tread-tree a43c4447b2edc9fb01a6369f10c1165de4494c88\n\tcheckout-cache -f -a\n\tupdate-cache --refresh\n\nand your tree should be valid again.\n\nNow, to remove any bogus objects, you can then run my \"git-prune-script\"\n(look at it carefully first to make sure you realize what you are doing).\n\nNOTE NOTE NOTE! This will _revert_ everything you had done after the \n\"trusted\" point. So you may not actually want to do this. Instead:\n\n> It's very much like I somehow committed against the _parent_ of the\n> head, rather than the head itself.\n\nThat's very common if you just forget to update your new \".git/HEAD\" when \nyou do a commit.\n\nAgain, it's the tools that make it a bit too easy to mess up. The \n\"commit-tree\" thing is supposed to really only be used from scripts (which \nwould do something like\n\n\tresult=$(commit-tree ...) && echo $result > .git/HEAD\n\nbut when doing things by hand, if you forget to update your HEAD, your \nnext commit will be done against the wrong head, and you get dangling \ncommits.\n\nThe good news is that this is not that hard to fix up. The _trees_ are all\ncorrect, and the objects are all correct, so what you can do is just\ngenerate a few new (proper) commit objects, with the right parents. Then\nyou can do the \"git-prune-script\" thing that will throw away the old\nbroken commits, since they won't be reachable from your new commits (even\nthough their _trees_ will be there and be the same).\n\nSo in this case:\n\n\tb4a9a5114b3c6da131a832a8e2cd1941161eb348\n\t+- e7905b2f22eb5d5308c9122b9c06c2d02473dd4f\n\t +- dc90c0db0dd5214aca5304fd17ccd741031e5493 <-- extra dangling head\n\t +- 488faba31f59c5960aabbb2a5877a0f2923937a3\n\nyou can do\n\n\tcat-file commit dc90c0db0dd5214aca5304fd17ccd741031e5493\n\nto remind you what your old tree and commit message was, and then just \nre-commit that tree with the same message but with the proper parent:\n\n\tcommit-tree xxxx -p 488faba31f59c5960aabbb2a5877a0f2923937a3\n\nand then you need to do the same thing for the other commits (which will \nnow need to be re-based to have the new commit-chain as their parents).\n\nThen, when you fixed up the final one, remember to update .git/HEAD with \nits commit ID, and now the prune-thing will get rid of the old dangling \ncommits that you just created new duplicates of.\n\n\t\tLinus\n"},{"id":"702","messageId":"Pine.LNX.4.58.0504181459500.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418225356.B16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T22:01:08Z","receivedAt":"2005-04-18T22:01:08Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Russell King wrote:\n> > \n> > \t# Get the trees associated with those commits\n> > \tcommon_tree=tree=$(cat-file commit $common | sed 's/tree //;q')\n> > \thead_tree=tree=$(cat-file commit $head | sed 's/tree //;q')\n> > \tmerge_tree=tree=$(cat-file commit $merge | sed 's/tree //;q')\n> \n> This wants to be:\n> \n> \tcommon_tree=$(cat-file commit $common | sed 's/tree //;q')\n> \thead_tree=$(cat-file commit $head | sed 's/tree //;q')\n> \tmerge_tree=$(cat-file commit $merge_head | sed 's/tree //;q')\n\nYup, I already fixed that in the current git version.\n\n> > \t\tread-tree $merge_tree && checkout-cache -f -a\n> \n> Don't we want to do an update-cache --refresh here?\n\nYes. Thanks, added.\n\n\t\tLinus\n"},{"id":"703","messageId":"20050418220541.GB19744@kroah.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504180840370.7211@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Greg KH","fromEmail":"greg@kroah.com","sentAt":"2005-04-18T22:05:41Z","receivedAt":"2005-04-18T22:05:41Z","isPatch":false,"sender":{"key":"greg@kroah.com","avatar":"https://gravatar.com/avatar/5bb5aa0cc2e01c00ec899d11130c07796bc186e465bae57bc34873b13b72c7c8?d=mp&s=160"},"body":"On Mon, Apr 18, 2005 at 08:42:14AM -0700, Linus Torvalds wrote:\n> \n> \n> On Mon, 18 Apr 2005, Greg KH wrote:\n> >\n> > On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> > > \n> > > Tools absolutely matter. And it will take time for us to build up that \n> > > kind of helper infrastructure. So being newbie might be part of it, but \n> > > it's the smaller part, I say. Rough interfaces is a big issue.\n> > \n> > Speaking of tools, you had a \"dotest\" program to apply patches in email\n> > form to a bk tree. And from what I can gather, you've changed that to\n> > handle git archives, right?\n> \n> Yup.\n> \n> It's a git archive at \n> \n> \tkernel.org:/pub/linux/kernel/people/torvalds/git-tools.git\n> \n> and it seems to work. It's what I've used for all the kernel patches \n> (except for the merge), and it's what I use for the git stuff that shows \n> up as authored by others.\n\nHm, have you pushed all of the recent changes public? I get the\nfollowing when trying to apply a patch:\n$ ../git-tools/dotest < ~/linux/patches/usb/usb-visor-tapwave_zodiac.patch \n mailsplit \n$ \n\nI set my PATH to point to the git-tools/ directory:\n$ which mailsplit\n/home/greg/linux/git/git-tools/mailsplit\n\nAny thoughts?\n\nthanks,\n\ngreg k-h\n"},{"id":"705","messageId":"20050418221407.GA20290@kroah.com","threadId":"58","inReplyTo":"20050418220541.GB19744@kroah.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Greg KH","fromEmail":"greg@kroah.com","sentAt":"2005-04-18T22:14:07Z","receivedAt":"2005-04-18T22:14:07Z","isPatch":false,"sender":{"key":"greg@kroah.com","avatar":"https://gravatar.com/avatar/5bb5aa0cc2e01c00ec899d11130c07796bc186e465bae57bc34873b13b72c7c8?d=mp&s=160"},"body":"On Mon, Apr 18, 2005 at 03:05:41PM -0700, Greg KH wrote:\n> On Mon, Apr 18, 2005 at 08:42:14AM -0700, Linus Torvalds wrote:\n> > \n> > \n> > On Mon, 18 Apr 2005, Greg KH wrote:\n> > >\n> > > On Sun, Apr 17, 2005 at 04:24:24PM -0700, Linus Torvalds wrote:\n> > > > \n> > > > Tools absolutely matter. And it will take time for us to build up that \n> > > > kind of helper infrastructure. So being newbie might be part of it, but \n> > > > it's the smaller part, I say. Rough interfaces is a big issue.\n> > > \n> > > Speaking of tools, you had a \"dotest\" program to apply patches in email\n> > > form to a bk tree. And from what I can gather, you've changed that to\n> > > handle git archives, right?\n> > \n> > Yup.\n> > \n> > It's a git archive at \n> > \n> > \tkernel.org:/pub/linux/kernel/people/torvalds/git-tools.git\n> > \n> > and it seems to work. It's what I've used for all the kernel patches \n> > (except for the merge), and it's what I use for the git stuff that shows \n> > up as authored by others.\n> \n> Hm, have you pushed all of the recent changes public? I get the\n> following when trying to apply a patch:\n> $ ../git-tools/dotest < ~/linux/patches/usb/usb-visor-tapwave_zodiac.patch \n> mailsplit \n\nDoh, you want it as a command line argument now. Stupid me...\n\nAnyway, I try it this way and get:\n\n$ dotest ~/linux/patches/usb/usb-visor-tapwave_zodiac.patch \n\nApplying USB: visor Tapwave Zodiac support patch\n\nfatal: preparing to update file 'drivers/usb/serial/visor.c' not uptodate in cache\n\n\nWhat did I forget to do?\n\nthanks,\n\ngreg k-h\n"},{"id":"706","messageId":"20050418231652.C16789@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"20050417195742.D13233@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-18T22:16:52Z","receivedAt":"2005-04-18T22:16:52Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"Ok, since the last one was soo successful, and I'm up for more\npunishment, here's another attempt. The diffstat is rather\ninteresting in this one, claiming no changes. It should look\nlike this:\n\n arch/arm/lib/bitops.h | 33 +++++++++++++++++++++++++++++++++\n 1 files changed, 33 insertions(+)\n\nHowever, it seems that git diff can't handle new files appearing\nyet.\n\nThe other interesting thing to note is that patches are generated\nfor '-p0' rather than '-p1' application, which is contary to our\nhistorical requirements. This is going to confuse people - can\nwe make it generate -p1 patches please?\n\nLinus - assuming I un-messed-up my tree properly (it appears to\nbe correct and fsck-cache $(commit-id) is happy) please merge\nthis. Thanks.\n\nLinus,\n\nPlease incorporate the latest ARM changes, which can\nbe found at:\n\n\tmaster.kernel.org:/home/rmk/linux-2.6-rmk.git\n\nThis will update the following files:\n\n 0 files changed\n\nthrough these ChangeSets:\n\nFrom: Russell King: Mon Apr 18 22:50:01 BST 2005\n\t\n\t[PATCH] ARM: Add missing new file for bitops patch\n\t\n\tSigned-off-by: Russell King\n\n\n-- \nRussell King\n\n"},{"id":"711","messageId":"20050418223306.GJ5554@pasky.ji.cz","threadId":"58","inReplyTo":"20050418231652.C16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T22:33:06Z","receivedAt":"2005-04-18T22:33:06Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Tue, Apr 19, 2005 at 12:16:52AM CEST, I got a letter\nwhere Russell King told me that...\n> Ok, since the last one was soo successful, and I'm up for more\n> punishment, here's another attempt. The diffstat is rather\n> interesting in this one, claiming no changes. It should look\n> like this:\n> \n> arch/arm/lib/bitops.h | 33 +++++++++++++++++++++++++++++++++\n> 1 files changed, 33 insertions(+)\n> \n> However, it seems that git diff can't handle new files appearing\n> yet.\n\nPatches welcome. :^)\n\nActually it shouldn't be hard. Just for through add-queue/rm-queue and\ndiff it with /dev/null.\n\n> The other interesting thing to note is that patches are generated\n> for '-p0' rather than '-p1' application, which is contary to our\n> historical requirements. This is going to confuse people - can\n> we make it generate -p1 patches please?\n\nThis should be fixed with latest Junio's show-diff changes. I might\nchange it to be identical with gitdiff-do output later though, using\ntree ID and \"uncommitted\" as the directories instead of 'a' and 'b'.\nIt will stay -p1 now, though.\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"717","messageId":"20050418224852.GM5554@pasky.ji.cz","threadId":"58","inReplyTo":"20050418225356.B16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T22:48:52Z","receivedAt":"2005-04-18T22:48:52Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Mon, Apr 18, 2005 at 11:53:57PM CEST, I got a letter\nwhere Russell King told me that...\n> Maybe Petr can improve the error handling, and incorporate it (or at\n> least some of it) into git-pasky\n\nThis does not need to touch git pull at all now; all the relevant logic\ncan change in git merge (and git commit), and I'm hacking that now. It\nshould be rather easy, I think.\n\nI think I won't make git merge commit automatically - I think the user\nshould get a chance to do a git diff on what is getting merged to check\nif everything is all right.\n\nWhat is actually a little annoying is having to cd ,,merge and then\nback, though. I don't know, but the current pull-merge script does not\nbother with the temporary merge directory neither, even though Linus\nwanted it. Linus, do you still do? ;-)\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"718","messageId":"20050418235951.D16789@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"20050418224852.GM5554@pasky.ji.cz","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-18T22:59:52Z","receivedAt":"2005-04-18T22:59:52Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Tue, Apr 19, 2005 at 12:48:52AM +0200, Petr Baudis wrote:\n> Dear diary, on Mon, Apr 18, 2005 at 11:53:57PM CEST, I got a letter\n> where Russell King told me that...\n> > Maybe Petr can improve the error handling, and incorporate it (or at\n> > least some of it) into git-pasky\n> \n> This does not need to touch git pull at all now; all the relevant logic\n> can change in git merge (and git commit), and I'm hacking that now. It\n> should be rather easy, I think.\n> \n> I think I won't make git merge commit automatically - I think the user\n> should get a chance to do a git diff on what is getting merged to check\n> if everything is all right.\n> \n> What is actually a little annoying is having to cd ,,merge and then\n> back, though. I don't know, but the current pull-merge script does not\n> bother with the temporary merge directory neither, even though Linus\n> wanted it. Linus, do you still do? ;-)\n\nIn the case I highlighted, we don't want to end up having to require\nuser intervention. This is a common case here, and was one which was\nentirely scripted with BK.\n\nEssentially, with BK, at 7am localtime each morning, I'd:\n\n- update my baseline linux 2.6 tree\n- for each working tree which may be pulled from\n - if the baseline is a superset\n - update working tree from baseline\n\nThe net result is that my workflow consisted entirely of:\n\n1. commit whatever into working tree\n2. test\n3. send linus a pull request\n4. repeat next day\n\nThe tree resynchronisation happened completely and entirely in the\nbackground with no user intervention required at all.\n\nWith your suggested requirement for user intervention whenever there's\na merge, it means that this just isn't possible - you could automate\nthe pulls, but you need to ensure that you'd visited each and every\nunmerged tree before the next day, or you don't script it at all and\ndo the whole thing manually.\n\nHey, I'm lazy, and that means that just won't get done, and my trees\nwill end up being horrendously out of date all the time. But isn't\nthis precisely what we have computers and scripts for?\n\n-- \nRussell King\n\n"},{"id":"720","messageId":"20050418230941.GN5554@pasky.ji.cz","threadId":"58","inReplyTo":"20050418235951.D16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T23:09:42Z","receivedAt":"2005-04-18T23:09:42Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Tue, Apr 19, 2005 at 12:59:52AM CEST, I got a letter\nwhere Russell King told me that...\n> In the case I highlighted, we don't want to end up having to require\n> user intervention. This is a common case here, and was one which was\n> entirely scripted with BK.\n\nWell, you can script that\n\n\tcd ,,merge && echo 'Merge with Linus' | git ci && cd ..\n\ntoo. ;-)\n\nBut it seems like a good idea to just proceed with commit in case of no\nconflicts (and possibly have a switch which will tell git merge to just\nmerge, not commit).\n\n> Essentially, with BK, at 7am localtime each morning, I'd:\n> \n> - update my baseline linux 2.6 tree\n> - for each working tree which may be pulled from\n> - if the baseline is a superset\n> - update working tree from baseline\n> \n> The net result is that my workflow consisted entirely of:\n> \n> 1. commit whatever into working tree\n> 2. test\n> 3. send linus a pull request\n> 4. repeat next day\n> \n> The tree resynchronisation happened completely and entirely in the\n> background with no user intervention required at all.\n\nAnd in the case of conflicts...?\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"719","messageId":"Pine.LNX.4.58.0504181610230.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418220541.GB19744@kroah.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T23:10:41Z","receivedAt":"2005-04-18T23:10:41Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Greg KH wrote:\n> \n> Hm, have you pushed all of the recent changes public?\n\nOops. Obviously not. Will fix.\n\n\t\tLinus\n"},{"id":"722","messageId":"Pine.LNX.4.58.0504181613490.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418221407.GA20290@kroah.com","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T23:16:45Z","receivedAt":"2005-04-18T23:16:45Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Greg KH wrote:\n> \n> Anyway, I try it this way and get:\n\nYou should update to the newest version anyway..\n\n> $ dotest ~/linux/patches/usb/usb-visor-tapwave_zodiac.patch \n> \n> Applying USB: visor Tapwave Zodiac support patch\n> \n> fatal: preparing to update file 'drivers/usb/serial/visor.c' not uptodate in cache\n> \n> What did I forget to do?\n\nThe most common reason is that the scripts _really_ want the index to \nmatch your current tree exactly. Run \"update-cache --refresh\". And if you \nhave any uncommitted information, make sure to commit it first.\n\n(Not _strictly_ true - you can leave edited files in your directory, and \njust hope the patch never touches them. The thing you should _not_ do is \nto do an \"update-cache xxxx.c\" to commit any changes to the 'index', \nbecause then the patch applicator will actually commit that one too).\n\n\t\tLinus\n"},{"id":"725","messageId":"20050418232607.GA20907@kroah.com","threadId":"58","inReplyTo":"Pine.LNX.4.58.0504181613490.15725@ppc970.osdl.org","subject":"Re: Re-done kernel archive - real one?","fromName":"Greg KH","fromEmail":"greg@kroah.com","sentAt":"2005-04-18T23:26:08Z","receivedAt":"2005-04-18T23:26:08Z","isPatch":false,"sender":{"key":"greg@kroah.com","avatar":"https://gravatar.com/avatar/5bb5aa0cc2e01c00ec899d11130c07796bc186e465bae57bc34873b13b72c7c8?d=mp&s=160"},"body":"On Mon, Apr 18, 2005 at 04:16:45PM -0700, Linus Torvalds wrote:\n> \n> \n> On Mon, 18 Apr 2005, Greg KH wrote:\n> > \n> > Anyway, I try it this way and get:\n> \n> You should update to the newest version anyway..\n> \n> > $ dotest ~/linux/patches/usb/usb-visor-tapwave_zodiac.patch \n> > \n> > Applying USB: visor Tapwave Zodiac support patch\n> > \n> > fatal: preparing to update file 'drivers/usb/serial/visor.c' not uptodate in cache\n> > \n> > What did I forget to do?\n> \n> The most common reason is that the scripts _really_ want the index to \n> match your current tree exactly. Run \"update-cache --refresh\". And if you \n> have any uncommitted information, make sure to commit it first.\n\nAh, that was the step I was missing, thanks, it's working now.\n\ngreg k-h\n"},{"id":"726","messageId":"Pine.LNX.4.58.0504181617010.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418231652.C16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T23:29:43Z","receivedAt":"2005-04-18T23:29:43Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Mon, 18 Apr 2005, Russell King wrote:\n>\n> Ok, since the last one was soo successful, and I'm up for more\n> punishment, here's another attempt. The diffstat is rather\n> interesting in this one, claiming no changes. It should look\n> like this:\n> \n> arch/arm/lib/bitops.h | 33 +++++++++++++++++++++++++++++++++\n> 1 files changed, 33 insertions(+)\n> \n> However, it seems that git diff can't handle new files appearing\n> yet.\n\nIt should definitely be able to do that. \n\nDo a \"git log | less\" to look up the trees involved, and do a \"git diff\n \" to see the output. If you don't see your new\nfile, then either you have an old \"git diff\" that doesn't like the new\ntools (and you need to add a \"-z\" flag to diff-tree), or you didn't \ncheck in the new file successfully ;)\n\nYou can also always do \"tree-diff -r old-tree new-tree\" which will show\nyou the tree-level changes. That's the low-level plumbing stuff: it \ndoesn't show you the actual file contents, just how the tree changed.\n\n> The other interesting thing to note is that patches are generated\n> for '-p0' rather than '-p1' application, which is contary to our\n> historical requirements. This is going to confuse people - can\n> we make it generate -p1 patches please?\n\nThat should already be the case now after the latest diffs from Junio.\n\n> Linus - assuming I un-messed-up my tree properly (it appears to\n> be correct and fsck-cache $(commit-id) is happy) please merge\n> this. \n\nLooks ok, which seems to mean that your scripts are buggered since they \ndidn't pick up the new file.\n\nMerge pushed out.\n\n\t\t\tLinus\n"},{"id":"728","messageId":"Pine.LNX.4.58.0504181630410.15725@ppc970.osdl.org","threadId":"58","inReplyTo":"20050418224852.GM5554@pasky.ji.cz","subject":"Re: Re-done kernel archive - real one?","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2005-04-18T23:31:26Z","receivedAt":"2005-04-18T23:31:26Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Tue, 19 Apr 2005, Petr Baudis wrote:\n> \n> What is actually a little annoying is having to cd ,,merge and then\n> back, though. I don't know, but the current pull-merge script does not\n> bother with the temporary merge directory neither, even though Linus\n> wanted it. Linus, do you still do? ;-)\n\nNo, now that the merge is done entirely in the index file, I don't care \nany more. The index file _is_ the temporary directory as far as I'm \nconcerned.\n\n\t\tLinus\n"},{"id":"731","messageId":"20050418235327.GQ5554@pasky.ji.cz","threadId":"58","inReplyTo":"20050418231652.C16789@flint.arm.linux.org.uk","subject":"Re: Re-done kernel archive - real one?","fromName":"Petr Baudis","fromEmail":"pasky@ucw.cz","sentAt":"2005-04-18T23:53:27Z","receivedAt":"2005-04-18T23:53:27Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"Dear diary, on Tue, Apr 19, 2005 at 12:16:52AM CEST, I got a letter\nwhere Russell King told me that...\n> However, it seems that git diff can't handle new files appearing\n> yet.\n\nFixed. :-)\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nC++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor\n"},{"id":"787","messageId":"20050419082756.A13105@flint.arm.linux.org.uk","threadId":"58","inReplyTo":"20050418230941.GN5554@pasky.ji.cz","subject":"Re: Re-done kernel archive - real one?","fromName":"Russell King","fromEmail":"rmk@arm.linux.org.uk","sentAt":"2005-04-19T07:27:56Z","receivedAt":"2005-04-19T07:27:56Z","isPatch":false,"sender":{"key":"rmk@arm.linux.org.uk","avatar":null},"body":"On Tue, Apr 19, 2005 at 01:09:42AM +0200, Petr Baudis wrote:\n> > Essentially, with BK, at 7am localtime each morning, I'd:\n> > \n> > - update my baseline linux 2.6 tree\n> > - for each working tree which may be pulled from\n> > - if the baseline is a superset\n> > - update working tree from baseline\n> > \n> > The net result is that my workflow consisted entirely of:\n> > \n> > 1. commit whatever into working tree\n> > 2. test\n> > 3. send linus a pull request\n> > 4. repeat next day\n> > \n> > The tree resynchronisation happened completely and entirely in the\n> > background with no user intervention required at all.\n> \n> And in the case of conflicts...?\n\nIf the baseline is a superset of the working tree, there will never be\nany conflicts. Note that as I said above, this is a condition on doing\nthe pull in the first place.\n\nHow we determine that with git is another matter though. 8)\n\n-- \nRussell King\n\n"}]}