X server still crazy

Chuck Ebbert cebbert at redhat.com
Mon Oct 8 17:01:33 UTC 2007


On 10/08/2007 09:29 AM, Adam Jackson wrote:
> On Mon, 2007-10-08 at 14:14 +0200, antonio montagnani wrote:
> 
>> I find that Xorg.0.log is very short and last lines are:
>> Module pcidata etc, etc
>> using VT number 7
> 
> This means we're stalled waiting for the VT switch from rhgb to gdm to
> complete.  Which is very weird; we shouldn't be.
> 
> I'm looking into it.
> 

Hmm...

Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=70cb97935b8859f27296772885104b599f560576
Commit:     70cb97935b8859f27296772885104b599f560576
Parent:     fc8b28a65d81a6fdf58ef81ce5b8ac7a35304e68
Author:     Linus Torvalds <torvalds at woody.linux-foundation.org>
AuthorDate: Sun Oct 7 16:02:55 2007 -0700
Committer:  Linus Torvalds <torvalds at woody.linux-foundation.org>
CommitDate: Sun Oct 7 16:02:55 2007 -0700

    VT_WAITACTIVE: Avoid returning EINTR when not necessary
    
    We should generally prefer to return ERESTARTNOHAND rather than EINTR,
    so that processes with unhandled signals that get ignored don't return
    EINTR.
    
    This can help with X startup issues:
    
        Fatal server error:
        xf86OpenConsole: VT_WAITACTIVE failed: Interrupted system call
    
    although the real fix is having the X server always retry EINTR
    regardless (since EINTR does happen for signals that have handlers
    installed). Keithp has a patch for that.




More information about the fedora-test-list mailing list