{"thread":{"id":"4811","subject":"What *have* I done?","startedAt":"2006-07-08T22:48:22Z","lastAt":"2006-07-09T22:25:21Z","messageCount":3,"participants":["Nix","Petr Baudis"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"23445","messageId":"87u05ru30p.fsf@hades.wkstn.nix","threadId":"4811","inReplyTo":null,"subject":"What *have* I done?","fromName":"Nix","fromEmail":"nix@esperi.org.uk","sentAt":"2006-07-08T22:48:22Z","receivedAt":"2006-07-08T22:48:22Z","isPatch":false,"sender":{"key":"nix@esperi.org.uk","avatar":"https://avatars.githubusercontent.com/u/6503005?v=4"},"body":"I am just a git newbie, still reliant on cogito 0.17.3 for most things\n(cogito has a couple of cherry-picked patches from the head for\nfaster cg-seek and for support of lightweight tags.)\n\nI tried my first conflicted cherry-pick a while back; resolved the\nconflict, committed, ran git-update-index over the conflicted file as\ngit-cherry-pick requested, yet now the branch the cherry-pick was done\nto seems to be oddly `half-stuck'.\n\nHere's an example:\n\nloki 590 /usr/packages/linux/linux% cg-status -g\nHeads:\n R 2.6.17-stable 4f9619cdd90ac846fa0ca6e9e8a9d87a0d6b4f57\n esperi-guest 764415bf7dc0388d9974218fb71b3c63f19f3828\n master 578997683c8d788f80f7a11fc27fcdf66ac8fd9f\n R origin 120bda20c6f64b32e8bfbdd7b34feafaa5f5332e\n v2.6.17 e519ad6612b8ab69cbf45b59dda45f88cada41f2\n >x86-host e2a9e65e9d5f7e76dbe0a8fdd0f1356266146c22\nloki 591 /usr/packages/linux/linux% cg-diff\nloki 592 /usr/packages/linux/linux% cg-switch v2.6.17\nSwitching to branch v2.6.17...\n\nNow one would expect that after switching, I'd have no uncommitted\nchanges, right, since I had none before switching and they shouldn't\nappear out of thin air? You'd be wrong:\n\nloki 593 /usr/packages/linux/linux% cg-diff | diffstat\n b/Makefile | 4 +-\n b/arch/sparc64/Makefile | 15 +++----\n b/arch/sparc64/kernel/Makefile | 2 -\n b/arch/um/kernel/dyn.lds.S | 1\n b/kernel/pid.c | 2 -\n b/scripts/kconfig/Makefile | 7 ++-\n b/usr/gen_init_cpio.c | 2 -\n usr/init | 79 -----------------------------------------\n usr/initramfs | 45 -----------------------\n 9 files changed, 15 insertions(+), 142 deletions(-)\n\nThose miraculously materialized uncommitted changes are the *reverse* of\nevery commit I made after that failed cherry-pick! (If I switch branches\nback, or run `cg-seek', it tries to take these `uncommitted local\nchanges' with me, with amusing results.)\n\nSo, has it forgotten what the branch head is? No: cg-log reports correct\ndata, cg-reset works fine:\n\nloki 594 /usr/packages/linux/linux% cg-reset\nloki 595 /usr/packages/linux/linux% cg-diff\nloki 596 /usr/packages/linux/linux%\n\n\nSo, um, what on earth has gone wrong and how do I fix it? (If you need\nmore info, just ask!)\n"},{"id":"23493","messageId":"20060709212113.GS29115@pasky.or.cz","threadId":"4811","inReplyTo":"87u05ru30p.fsf@hades.wkstn.nix","subject":"Re: What *have* I done?","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2006-07-09T21:21:13Z","receivedAt":"2006-07-09T21:21:13Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":" Hi,\n\nDear diary, on Sun, Jul 09, 2006 at 12:48:22AM CEST, I got a letter\nwhere Nix said that...\n> loki 593 /usr/packages/linux/linux% cg-diff | diffstat\n\n Note that cg-diff -s might be useful.\n\n> loki 594 /usr/packages/linux/linux% cg-reset\n> loki 595 /usr/packages/linux/linux% cg-diff\n> loki 596 /usr/packages/linux/linux%\n\n ...but when you switch away and switch back, the uncommitted changes\nappear again?\n\n Just in case, don't the tree happen to simply stay the same as in the\nbranch you switched from? (Try cg-diff -r yourpreviousbranch.)\n\n And most importantly, does this also happen with just stock 0.17.3\ncg-switch? With stock master cg-switch? (If both is no, you probably\nforgot to cherrypick some associated cg-switch bugfix or something.)\n\n-- \n\t\t\t\tPetr \"Pasky\" Baudis\nStuff: http://pasky.or.cz/\nSnow falling on Perl. White noise covering line noise.\nHides all the bugs too. -- J. Putnam\n"},{"id":"23502","messageId":"87hd1ql8ku.fsf@hades.wkstn.nix","threadId":"4811","inReplyTo":"20060709212113.GS29115@pasky.or.cz","subject":"Re: What *have* I done?","fromName":"Nix","fromEmail":"nix@esperi.org.uk","sentAt":"2006-07-09T22:25:21Z","receivedAt":"2006-07-09T22:25:21Z","isPatch":false,"sender":{"key":"nix@esperi.org.uk","avatar":"https://avatars.githubusercontent.com/u/6503005?v=4"},"body":"On Sun, 9 Jul 2006, Petr Baudis yowled:\n> Hi,\n> \n> Dear diary, on Sun, Jul 09, 2006 at 12:48:22AM CEST, I got a letter\n> where Nix said that...\n>> loki 593 /usr/packages/linux/linux% cg-diff | diffstat\n> \n> Note that cg-diff -s might be useful.\n\nIndeed :)\n\n>> loki 594 /usr/packages/linux/linux% cg-reset\n>> loki 595 /usr/packages/linux/linux% cg-diff\n>> loki 596 /usr/packages/linux/linux%\n> \n> ...but when you switch away and switch back, the uncommitted changes\n> appear again?\n\nYes.\n\n> Just in case, don't the tree happen to simply stay the same as in the\n> branch you switched from? (Try cg-diff -r yourpreviousbranch.)\n\nYes. Also, that branch (which was forked recently from the misbehaving\none) claims that all changes are merged from that branch, but this is\ndemonstrably untrue.\n\n> And most importantly, does this also happen with just stock 0.17.3\n> cg-switch? With stock master cg-switch? (If both is no, you probably\n> forgot to cherrypick some associated cg-switch bugfix or something.)\n\nThis is 0.17.3 cg-switch plus the seek speedup 40f8a28806a1c5a7cb1f7d137f1bb271b71f890f.\n(I doubt that's to blame, but I've been wrong before.)\n\nI think the oddity is related to git-cherry-pick somehow, but I'm\nnot sure how: I took the original upstream pack again, rebuilt the\nbranches from scratch and now I can't make the problem happen anymore.\n\nIf it happens again I'll keep the tree around and collect more data.\n(I'll have to dig some more disk space up...)\n\n-- \n`She *is*, however, one of the few sf authors with a last name ending in O,\n which adds some extra appeal to those of us who obsess about things like\n having a book review of an author for each letter in the alphabet.' -- rra\n"}]}