{"thread":{"id":"4186","subject":"1.3.2 git-clone segfaults","startedAt":"2006-05-17T18:32:39Z","lastAt":"2006-05-18T00:54:30Z","messageCount":7,"participants":["Bill Yoder","Fernando J. Pereda","Junio C Hamano","Pavel Roskin","Linus Torvalds"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"20161","messageId":"879BAFDD-87DB-4041-8753-5D63630076B5@cs.utexas.edu","threadId":"4186","inReplyTo":null,"subject":"1.3.2 git-clone segfaults","fromName":"Bill Yoder","fromEmail":"byoder@cs.utexas.edu","sentAt":"2006-05-17T18:32:39Z","receivedAt":"2006-05-17T18:32:39Z","isPatch":false,"sender":{"key":"byoder@cs.utexas.edu","avatar":null},"body":"Dear Git maintainers:\n\nI have twice downloaded git-1.3.2, most recently the git-1.3.2.tar.gz  \nsource package from http://www.t2-project.org/packages/git.html.   \nBoth the prebuilt version and the source version, built with gcc  \n3.2.3 on x86/Linux, run into trouble when executing this command:\n\n% git --exec-path=/usr/local/downloads/git-1.3.2 clone http:// \nwww.denx.de/git/linux-2.6-denx.git linux-2.6-denx\n\nBoth git operations proceed with some 18,000+ lines of output,  \nculminating in this message:\n\ngot 243a15f54002445f5b8b4938981ec90430b73ec6\ngot 03914b7e41b17871aea961f6522ec4ce26a4f8ed\ngot a305ae2e44b6dde305d3afe241768e32c47d8907\ngot 33a653913d942fa35c263edf1d019f36f4e0f5b1\ngot f13f49afe198cc0d59e998fe3a6e721d70fca6b4\nerror: The requested URL returned error: 405 (curl_result = 22,  \nhttp_code = 405, sha1 = b323ff5779672c77b6adfba1c1bdc87f4981f85c)\nerror: Unable to find b323ff5779672c77b6adfba1c1bdc87f4981f85c under  \nhttp://www.denx.de/git/linux-2.6-denx.git/\nCannot obtain needed blob b323ff5779672c77b6adfba1c1bdc87f4981f85c\nwhile processing commit ea989b3245993f95e58e6c0320bf6165a949b072.\nWaiting for http://www.denx.de/git/linux-2.6-denx.git/objects/ \n31/548303ee3767095f86efb47696ce433662450e\n/usr/local/downloads/git-1.3.2/git-clone: line 323: 25972  \nSegmentation fault      git-http-fetch -v -a -w \"$tname\" \"$name\" \"$1/\"\n\nI have also tested git-1.2.3 built from source, and it works dandily.\n\nPlease let me know if I can supply more information or if I'm  \nmisusing the command.\n\nThanks,\n\nBill Yoder\nGit Newbie\n"},{"id":"20163","messageId":"20060517184121.GA17846@ferdyx.org","threadId":"4186","inReplyTo":"879BAFDD-87DB-4041-8753-5D63630076B5@cs.utexas.edu","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Fernando J. Pereda","fromEmail":"ferdy@ferdyx.org","sentAt":"2006-05-17T18:41:21Z","receivedAt":"2006-05-17T18:41:21Z","isPatch":false,"sender":{"key":"ferdy@ferdyx.org","avatar":"https://gravatar.com/avatar/96bf7c1ddf7ccd430255bd12d9d42b212dbc033b28c668a2bdf9c3995aa81e61?d=mp&s=160"},"body":"On Wed, May 17, 2006 at 01:32:39PM -0500, Bill Yoder wrote:\n> Please let me know if I can supply more information or if I'm\n> misusing the command.\n\nI reported this same issue to Nick Hengeveld and he said he'll take a\nlook at it. this is also Gentoo Bug #133412 [1].\n\n- ferdy\n\n[1] https://bugs.gentoo.org/show_bug.cgi?id=133412\n\n-- \nFernando J. Pereda Garcimartín\nGentoo Developer (Alpha,net-mail,mutt,git)\n20BB BDC3 761A 4781 E6ED  ED0B 0A48 5B0C 60BD 28D4\n"},{"id":"20165","messageId":"7vwtckcwve.fsf@assigned-by-dhcp.cox.net","threadId":"4186","inReplyTo":"879BAFDD-87DB-4041-8753-5D63630076B5@cs.utexas.edu","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Junio C Hamano","fromEmail":"junkio@cox.net","sentAt":"2006-05-17T18:46:45Z","receivedAt":"2006-05-17T18:46:45Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Bill Yoder <byoder@cs.utexas.edu> writes:\n\n> I have twice downloaded git-1.3.2, most recently the git-1.3.2.tar.gz\n> source package from http://www.t2-project.org/packages/git.html.\n\nCould you try 1.3.3?\n"},{"id":"20173","messageId":"1871789F-06D0-423F-A87D-1BFF9E0FC8E1@cs.utexas.edu","threadId":"4186","inReplyTo":"7vwtckcwve.fsf@assigned-by-dhcp.cox.net","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Bill Yoder","fromEmail":"byoder@cs.utexas.edu","sentAt":"2006-05-17T19:27:09Z","receivedAt":"2006-05-17T19:27:09Z","isPatch":false,"sender":{"key":"byoder@cs.utexas.edu","avatar":null},"body":"Dear Junio (and Fernando):\n\n> Could you try 1.3.3?\n\nHere are the last few lines of output from 1.3.3, cooked with gcc 3.2.3:\n\nGetting alternates list for http://www.denx.de/git/linux-2.6-denx.git/\ngot b603ba9e9b5482d3e80fc8e0fa96bb9a943502ff\ngot 1635ee25918fc19ea613d1e8dbcb672075220efb\ngot dd7d627bf66f306b4ee9401f06ed4fb574896a85\ngot c771a7db9871bfa3f3c76b78c1369111c4be767b\ngot 374e20ad8b0d02f15fbcaa5315e272eabd6c4f76\ngot a8bef1d1371cc999ce6882d355c7554ca7738173\ngot ab08f35cbc355f4b2058d88ff289552f202ea5b4\nGetting pack list for http://www.denx.de/git/linux-2.6-denx.git/\ngot 92297ff24e8525de2617dff728b3420a4649f66a\ngot 93dcbe1abb4c83b65cc6af59fb84c3c5e16effbb\ngot 41ecbb847f32f301a7ecd30b6438fea702886a33\ngot 94f557fa46369ec94ec718d25616ceb0b73fd2d2\ngot 09e00433c78e267b37b3f485c0d877de780a0674\ngot da7c09e4ede6e83198bf1ab5f16c571b0cc214ee\ngot 85533ec5aaa17f4146452a16ef61ca40fc601c80\ngot 50e338d2ffda9cb5a835d67849e38ae0ceba1647\n/usr/local/downloads/git-1.3.3/git-clone: line 323:  1124  \nSegmentation fault      git-http-fetch -v -a -w \"$tname\" \"$name\" \"$1/\"\n\nRegards,\nBill\n"},{"id":"20174","messageId":"1147894165.16654.10.camel@dv","threadId":"4186","inReplyTo":"879BAFDD-87DB-4041-8753-5D63630076B5@cs.utexas.edu","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2006-05-17T19:29:25Z","receivedAt":"2006-05-17T19:29:25Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"On Wed, 2006-05-17 at 13:32 -0500, Bill Yoder wrote:\n> /usr/local/downloads/git-1.3.2/git-clone: line 323: 25972  \n> Segmentation fault      git-http-fetch -v -a -w \"$tname\" \"$name\" \"$1/\"\n\nI've seen git-http-fetch segfaults many times when cloning qgit, but\nit's hard to reproduce on demand.\n\nI think you should compile git without optimizations and allow coredumps\n(ulimit -c  unlimited), then load git-http-fetch in gdb with the core\n(gdb --core=core git-http-fetch) and run bt to see the backtrace.\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"20196","messageId":"1147909920.32050.29.camel@dv","threadId":"4186","inReplyTo":"1147894165.16654.10.camel@dv","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2006-05-17T23:52:00Z","receivedAt":"2006-05-17T23:52:00Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"On Wed, 2006-05-17 at 15:29 -0400, Pavel Roskin wrote:\n> On Wed, 2006-05-17 at 13:32 -0500, Bill Yoder wrote:\n> > /usr/local/downloads/git-1.3.2/git-clone: line 323: 25972  \n> > Segmentation fault      git-http-fetch -v -a -w \"$tname\" \"$name\" \"$1/\"\n> \n> I've seen git-http-fetch segfaults many times when cloning qgit, but\n> it's hard to reproduce on demand.\n> \n> I think you should compile git without optimizations and allow coredumps\n> (ulimit -c  unlimited), then load git-http-fetch in gdb with the core\n> (gdb --core=core git-http-fetch) and run bt to see the backtrace.\n\nAlso comment out both \"trap\" invocations in git-clone, or the coredump\nwill be deleted.\n\nThat's what I've got on Fedora Core 5 x86_64 with glibc and curl debug\ninfo installed:\n\n#0  __strncasecmp (s1=Variable \"s1\" is not available.\n) at strncase.c:68\n68        while ((result = TOLOWER (*p1) - TOLOWER (*p2++)) == 0)\n(gdb) where\n#0  __strncasecmp (s1=Variable \"s1\" is not available.\n) at strncase.c:68\n#1  0x00000031f3e26c09 in curl_strnequal (first=Variable \"first\" is not available.\n) at strequal.c:60\n#2  0x00000031f3e0f43a in checkheaders (data=Variable \"data\" is not available.\n) at http.c:119\n#3  0x00000031f3e10cf9 in Curl_http (conn=0x1c421c0, done=Variable \"done\" is not available.\n) at http.c:1580\n#4  0x00000031f3e1a858 in Curl_do (connp=0x83af88, done=0x7fff29c97ebb \"\\001\\001\")\n    at url.c:3841\n#5  0x00000031f3e28f22 in curl_multi_perform (multi_handle=0x53b590, \n    running_handles=0x7fff29c97ef8) at multi.c:526\n#6  0x00000000004040c0 in step_active_slots () at http.c:376\n#7  0x000000000040412c in run_active_slot (slot=0x546690) at http.c:400\n#8  0x0000000000403e44 in http_cleanup () at http.c:275\n#9  0x00000000004077d7 in main (argc=7, argv=0x7fff29c98258) at http-fetch.c:1274\n(gdb) p p1\n$1 = (const unsigned char *) 0x0\n(gdb) p p2\n$2 = (const unsigned char *) 0x31f3e2e817 \"User-Agent:\"\n(gdb)\n\nLooks like a curl bug to me.  curl 7.15.1, glibc 2.4, git master branch.\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"20197","messageId":"Pine.LNX.4.64.0605171752190.10823@g5.osdl.org","threadId":"4186","inReplyTo":"1147909920.32050.29.camel@dv","subject":"Re: 1.3.2 git-clone segfaults","fromName":"Linus Torvalds","fromEmail":"torvalds@osdl.org","sentAt":"2006-05-18T00:54:30Z","receivedAt":"2006-05-18T00:54:30Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Wed, 17 May 2006, Pavel Roskin wrote:\n> \n> Looks like a curl bug to me.  curl 7.15.1, glibc 2.4, git master branch.\n\nIf the thing is fixed by turning off DAV support (and I thought somebody \nreported it was), maybe we should turn that off by default? Ie, make \nNO_EXPAT be the default, and you have to explicitly turn it off.\n\n\t\tLinus\n"}]}