Daniel Stenberg wrote: >> b95fe985af3c80a2babcaaaf7da69a15b1237c49 is first bad commit > > Oh my, this again. Can we revert this change cleanly without any > nasty side-effects?
I promised to take a look but I haven't had a chance. I will get to it right now and send an update in a bit. > I wish we had a test suite that would catch these errors before we > commit... This is tricky to test for because the failure is dependant on how the operating system behaves, which depends on if it's a local connection or a remote connection, and maybe even which kernel it's running on. //Peter _______________________________________________ libssh2-devel http://cool.haxx.se/cgi-bin/mailman/listinfo/libssh2-devel
