{"thread":{"id":"17179","subject":"cygwin git diff crash","startedAt":"2009-01-14T22:09:50Z","lastAt":"2009-01-16T01:47:31Z","messageCount":6,"participants":["Jeremy Ramer","Johannes Schindelin"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"100489","messageId":"b9fd99020901141409w1e0c926fkc762f8709bd1c13f@mail.gmail.com","threadId":"17179","inReplyTo":null,"subject":"cygwin git diff crash","fromName":"Jeremy Ramer","fromEmail":"jdramer@gmail.com","sentAt":"2009-01-14T22:09:50Z","receivedAt":"2009-01-14T22:09:50Z","isPatch":false,"sender":{"key":"jdramer@gmail.com","avatar":null},"body":"Ok this is a weird issue and it's probably cygwin's fault, but I\nhaven't found any way to fix it so I' thought I would throw it out\nhere for comment.\n\nI am using git 1.6.0.4 on cygwin. I have a repo where if any file has\nchanges and git detects as mode 100644 I get this error:\n$git diff\n      3 [main] git 2744 C:\\cygwin\\bin\\git.exe: *** fatal error - could\nnot load user32, Win32 error\n\nIf I change the mode to 100755 git diff will work fine.\n$chmod a+x test.cpp\n$git diff\ndiff --git a/test.cpp b/test.cpp\nold mode 100644\nnew mode 100755\nindex 7c0dfcd..20987a7\n--- a/test.cpp\n+++ b/test.cpp\n@@ -6,9 +6,11 @@ int main()\n\n void func()\n {\n+       int a;^M\n }\n\n void func2()\n {\n+       int b;^M\n }\n\nAnybody have a clue as to why this might occur?  I have seen this in\nmany of the repo's I use, but it is not repeatable.  I tried making a\ntest repo but could not reproduce.\n\nThanks!\nJeremy\n"},{"id":"100498","messageId":"alpine.DEB.1.00.0901142333200.3586@pacific.mpi-cbg.de","threadId":"17179","inReplyTo":"b9fd99020901141409w1e0c926fkc762f8709bd1c13f@mail.gmail.com","subject":"Re: cygwin git diff crash","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2009-01-14T22:33:45Z","receivedAt":"2009-01-14T22:33:45Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Wed, 14 Jan 2009, Jeremy Ramer wrote:\n\n> Ok this is a weird issue and it's probably cygwin's fault, but I\n> haven't found any way to fix it so I' thought I would throw it out\n> here for comment.\n> \n> I am using git 1.6.0.4 on cygwin. I have a repo where if any file has\n> changes and git detects as mode 100644 I get this error:\n> $git diff\n>       3 [main] git 2744 C:\\cygwin\\bin\\git.exe: *** fatal error - could\n> not load user32, Win32 error\n> \n> If I change the mode to 100755 git diff will work fine.\n> $chmod a+x test.cpp\n> $git diff\n> diff --git a/test.cpp b/test.cpp\n> old mode 100644\n> new mode 100755\n> index 7c0dfcd..20987a7\n> --- a/test.cpp\n> +++ b/test.cpp\n> @@ -6,9 +6,11 @@ int main()\n> \n>  void func()\n>  {\n> +       int a;^M\n>  }\n> \n>  void func2()\n>  {\n> +       int b;^M\n>  }\n> \n> Anybody have a clue as to why this might occur?  I have seen this in\n> many of the repo's I use, but it is not repeatable.  I tried making a\n> test repo but could not reproduce.\n\nWow, that _is_ weird.  Does your test suite pass?\n\nCiao,\nDscho\n"},{"id":"100600","messageId":"b9fd99020901150642i2e61bb78yf2dab2b8713592df@mail.gmail.com","threadId":"17179","inReplyTo":"alpine.DEB.1.00.0901142333200.3586@pacific.mpi-cbg.de","subject":"Re: cygwin git diff crash","fromName":"Jeremy Ramer","fromEmail":"jdramer@gmail.com","sentAt":"2009-01-15T14:42:49Z","receivedAt":"2009-01-15T14:42:49Z","isPatch":false,"sender":{"key":"jdramer@gmail.com","avatar":null},"body":"On Wed, Jan 14, 2009 at 3:33 PM, Johannes Schindelin\n<Johannes.Schindelin@gmx.de> wrote:\n> Hi,\n>\n> On Wed, 14 Jan 2009, Jeremy Ramer wrote:\n>\n>> Ok this is a weird issue and it's probably cygwin's fault, but I\n>> haven't found any way to fix it so I' thought I would throw it out\n>> here for comment.\n>>\n>> I am using git 1.6.0.4 on cygwin. I have a repo where if any file has\n>> changes and git detects as mode 100644 I get this error:\n>> $git diff\n>>       3 [main] git 2744 C:\\cygwin\\bin\\git.exe: *** fatal error - could\n>> not load user32, Win32 error\n>>\n>> If I change the mode to 100755 git diff will work fine.\n>> $chmod a+x test.cpp\n>> $git diff\n>> diff --git a/test.cpp b/test.cpp\n>> old mode 100644\n>> new mode 100755\n>> index 7c0dfcd..20987a7\n>> --- a/test.cpp\n>> +++ b/test.cpp\n>> @@ -6,9 +6,11 @@ int main()\n>>\n>>  void func()\n>>  {\n>> +       int a;^M\n>>  }\n>>\n>>  void func2()\n>>  {\n>> +       int b;^M\n>>  }\n>>\n>> Anybody have a clue as to why this might occur?  I have seen this in\n>> many of the repo's I use, but it is not repeatable.  I tried making a\n>> test repo but could not reproduce.\n>\n> Wow, that _is_ weird.  Does your test suite pass?\n>\n> Ciao,\n> Dscho\n>\n\nForgive my ignorance, but I am not sure how to run the test suite.  I\ndid a quick google search but didn't find anything.  I am using\ncygwin's packaged version of git and am not building it from source,\nso maybe the suite is not available.  If I get a chance I will try\nbuilding from source.\n\nJeremy\n"},{"id":"100613","messageId":"alpine.DEB.1.00.0901151631280.3586@pacific.mpi-cbg.de","threadId":"17179","inReplyTo":"b9fd99020901150642i2e61bb78yf2dab2b8713592df@mail.gmail.com","subject":"Re: cygwin git diff crash","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2009-01-15T15:34:18Z","receivedAt":"2009-01-15T15:34:18Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Thu, 15 Jan 2009, Jeremy Ramer wrote:\n\n> On Wed, Jan 14, 2009 at 3:33 PM, Johannes Schindelin\n> <Johannes.Schindelin@gmx.de> wrote:\n> > Hi,\n> >\n> > On Wed, 14 Jan 2009, Jeremy Ramer wrote:\n> >\n> >> Ok this is a weird issue and it's probably cygwin's fault, but I\n> >> haven't found any way to fix it so I' thought I would throw it out\n> >> here for comment.\n> >>\n> >> I am using git 1.6.0.4 on cygwin. I have a repo where if any file has\n> >> changes and git detects as mode 100644 I get this error:\n> >> $git diff\n> >>       3 [main] git 2744 C:\\cygwin\\bin\\git.exe: *** fatal error - could\n> >> not load user32, Win32 error\n> >>\n> >> If I change the mode to 100755 git diff will work fine.\n> >> $chmod a+x test.cpp\n> >> $git diff\n> >> diff --git a/test.cpp b/test.cpp\n> >> old mode 100644\n> >> new mode 100755\n> >> index 7c0dfcd..20987a7\n> >> --- a/test.cpp\n> >> +++ b/test.cpp\n> >> @@ -6,9 +6,11 @@ int main()\n> >>\n> >>  void func()\n> >>  {\n> >> +       int a;^M\n> >>  }\n> >>\n> >>  void func2()\n> >>  {\n> >> +       int b;^M\n> >>  }\n> >>\n> >> Anybody have a clue as to why this might occur?  I have seen this in\n> >> many of the repo's I use, but it is not repeatable.  I tried making a\n> >> test repo but could not reproduce.\n> >\n> > Wow, that _is_ weird.  Does your test suite pass?\n> >\n> > Ciao,\n> > Dscho\n> >\n> \n> Forgive my ignorance, but I am not sure how to run the test suite.  I\n> did a quick google search but didn't find anything.  I am using\n> cygwin's packaged version of git and am not building it from source,\n> so maybe the suite is not available.  If I get a chance I will try\n> building from source.\n\nIt is really easy: just make sure that make, gcc, libiconv-dev, \nopenssl-dev and libcurl-dev are installed (that's it AFAIR), then download \na tarball, e.g.\n\n\thttp://repo.or.cz/w/git.git?a=snapshot;h=next;sf=tgz\n\nunpack it, cd to it and run \"make\".  (I would _not_ run configure...)\n\nHth,\nDscho\n"},{"id":"100658","messageId":"b9fd99020901151308u3e901602ya66f170da2b37045@mail.gmail.com","threadId":"17179","inReplyTo":"alpine.DEB.1.00.0901151631280.3586@pacific.mpi-cbg.de","subject":"Re: cygwin git diff crash","fromName":"Jeremy Ramer","fromEmail":"jdramer@gmail.com","sentAt":"2009-01-15T21:08:20Z","receivedAt":"2009-01-15T21:08:20Z","isPatch":false,"sender":{"key":"jdramer@gmail.com","avatar":null},"body":"On Thu, Jan 15, 2009 at 8:34 AM, Johannes Schindelin\n<Johannes.Schindelin@gmx.de> wrote:\n> Hi,\n>\n> On Thu, 15 Jan 2009, Jeremy Ramer wrote:\n>\n>> On Wed, Jan 14, 2009 at 3:33 PM, Johannes Schindelin\n>> <Johannes.Schindelin@gmx.de> wrote:\n>> > Hi,\n>> >\n>> > On Wed, 14 Jan 2009, Jeremy Ramer wrote:\n>> >\n>> >> Ok this is a weird issue and it's probably cygwin's fault, but I\n>> >> haven't found any way to fix it so I' thought I would throw it out\n>> >> here for comment.\n>> >>\n>> >> I am using git 1.6.0.4 on cygwin. I have a repo where if any file has\n>> >> changes and git detects as mode 100644 I get this error:\n>> >> $git diff\n>> >>       3 [main] git 2744 C:\\cygwin\\bin\\git.exe: *** fatal error - could\n>> >> not load user32, Win32 error\n>> >>\n>> >> If I change the mode to 100755 git diff will work fine.\n>> >> $chmod a+x test.cpp\n>> >> $git diff\n>> >> diff --git a/test.cpp b/test.cpp\n>> >> old mode 100644\n>> >> new mode 100755\n>> >> index 7c0dfcd..20987a7\n>> >> --- a/test.cpp\n>> >> +++ b/test.cpp\n>> >> @@ -6,9 +6,11 @@ int main()\n>> >>\n>> >>  void func()\n>> >>  {\n>> >> +       int a;^M\n>> >>  }\n>> >>\n>> >>  void func2()\n>> >>  {\n>> >> +       int b;^M\n>> >>  }\n>> >>\n>> >> Anybody have a clue as to why this might occur?  I have seen this in\n>> >> many of the repo's I use, but it is not repeatable.  I tried making a\n>> >> test repo but could not reproduce.\n>> >\n>> > Wow, that _is_ weird.  Does your test suite pass?\n>> >\n>> > Ciao,\n>> > Dscho\n>> >\n>>\n>> Forgive my ignorance, but I am not sure how to run the test suite.  I\n>> did a quick google search but didn't find anything.  I am using\n>> cygwin's packaged version of git and am not building it from source,\n>> so maybe the suite is not available.  If I get a chance I will try\n>> building from source.\n>\n> It is really easy: just make sure that make, gcc, libiconv-dev,\n> openssl-dev and libcurl-dev are installed (that's it AFAIR), then download\n> a tarball, e.g.\n>\n>        http://repo.or.cz/w/git.git?a=snapshot;h=next;sf=tgz\n>\n> unpack it, cd to it and run \"make\".  (I would _not_ run configure...)\n>\n> Hth,\n> Dscho\n>\n\nWell, after compiling from source the error seems to have disappeared.\nStrange. I have seen it disappear before and come back later so I will\nwatch out for it, but for now it's not an issue.\n\nJust a note, the cygwin names for the packages necessary to compile are\ncurl-devel\ngcc\nlibiconv\nmake\nopenssl-devel\n\nJeremy\n"},{"id":"100680","messageId":"alpine.DEB.1.00.0901160247040.3586@pacific.mpi-cbg.de","threadId":"17179","inReplyTo":"b9fd99020901151308u3e901602ya66f170da2b37045@mail.gmail.com","subject":"Re: cygwin git diff crash","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2009-01-16T01:47:31Z","receivedAt":"2009-01-16T01:47:31Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Thu, 15 Jan 2009, Jeremy Ramer wrote:\n\n> Well, after compiling from source the error seems to have disappeared. \n> Strange. I have seen it disappear before and come back later so I will \n> watch out for it, but for now it's not an issue.\n\nWell, at least it is gone for now.\n\n> Just a note, the cygwin names for the packages necessary to compile are\n> curl-devel\n> gcc\n> libiconv\n> make\n> openssl-devel\n\nThanks,\nDscho\n"}]}