Re: X 4.0.1 what's the deal?

[Adam Lazur <adam@lazur.org>]

Decklin Foster (decklin@red-bean.com) said:
> Branden Robinson writes:
> > I've tested with 2.2.17 and 2.2.18pre15, and could not reproduce it.
> 
> It appears to be specific to recent 2.4test kernels. I could not
> reproduce on 2.2.17, but I was able to hang my system with test9/test10.
> I'm at a loss as to how to debug this (I couldn't log in remotely), or
> which upstream it should be forwarded to.

I reproduced it on my 2.4.0-test10 box and _could_ log in remotely.

strace on the X process shows the following looping forever:

uname({sys="Linux", node="bamf", ...})  = 0
uname({sys="Linux", node="bamf", ...})  = 0
connect(6, {sin_family=AF_UNIX, path="/tmp/.font-unix/fs7100"}, 24) = -1 ENOENT (No such file or directory)

I let it run for about 5 minutes doing this, then killed it (with a -9).

The hang didn't occur with 2.2.18pre15, and my ultra super strace test on
/usr/X11/bin/xinit shows:

uname({sys="Linux", node="bamf", ...})  = 0
uname({sys="Linux", node="bamf", ...})  = 0
connect(3, {sin_family=AF_UNIX, path="/tmp/.X11-unix/X0"}, 19) = 0

it continues after this. So, the syscalls are different for the differing
kernels... not much more data, but it's at least a little more.

.adam

-- 
[                <adam@lazur.org> <laz@clustermonkey.org>                ]
[              icq #3354423 | lazur.org | clustermonkey.org              ]