On 9/15/2012 22:07, John Marino wrote:
On 9/15/2012 18:01, Francois Tigeot wrote:On Sun, Sep 16, 2012 at 12:20:53AM +0900, OBATA Akio wrote:On Sat, 15 Sep 2012 23:23:52 +0900, John Marino<netbsd%marino.st@localhost> wrote:ftigeot says he had luck building cairo on DragonFly, so I'm going to post the errors I'm seeing on i386 here: http://leaf.dragonflybsd.org/~marino/misc/cairo.log This was not built in a clean environment like Tinderbox or pbulk, but rather the rebuild was triggered by the revbump.I don't know the reason why, but I feel that configure failed to detect pthread.This is not the first time, there's a bug report from 2010 on this very issue: https://bugs.freedesktop.org/show_bug.cgi?id=30910Editing the configure file as outlined in the bug report did not result in pthread getting detected. This may be a similar problem as seen in sysutils/coreutils. I'll take a look at that.
I think this problem is specific to my VM. It's caused by an error that prevents the error handling frame from getting created on the conftest executable, which is probably caused by gcc 4.7 testing a while back. I was able to build cairo just fine on a x86_64 vm. I don't have access to my real i386 to completely confirm, but I'm ok with assuming cairo builds on stock DragonFly for now.
John