was there some weird issue with the second last rawhide kernel? [DIAGNOSED?]

[poma]

On 20.06.2014 17:27, Robert P. J. Day wrote:
> On Thu, 19 Jun 2014, Adam Williamson wrote:
>
>> On Thu, 2014-06-19 at 06:46 -0400, Robert P. J. Day wrote:
>>>    rebooted my rawhide laptop this morning, it totally hung ... powered
>>> down and up, same thing, so i manually backed off to the earlier
>>> kernel release, 3.16.0-0.rc0.git1.1.fc21.x86_64, worked fine, and am
>>> now currently "yum updating" which will bump up my kernel to
>>> kernel-3.16.0-0.rc0.git11.1.fc21.x86_64, and i'll try that one.
>>>
>>>    so was there a known issue with, uh,
>>> 3.16.0-0.rc1.git1.1.fc21.x86_64? it's been a while since a kernel
>>> simply froze my machine in its tracks.
>>
>> they're very early builds, so it's certainly plausible. when you say
>> it "totally hung" - how far did it get? did the screen blank? any
>> error messages?
>
>    ah, a possible diagnosis -- the newest rawhide kernels appear to
> hang if i have an external monitor connected on my HDMI port during
> boot.
>
>    just did a "yum update" and booted with:
>
> $ uname -r
> 3.16.0-0.rc1.git2.1.fc21.x86_64
> $
>
> and as long as i disconnect HDMI first, my asus G74S laptop boots just
> fine, and i can plug in the monitor afterwards and all is well.
>
>    if, however, i reboot with the monitor still connected, i can see
> the little fedora blob slowly filling in with white, followed by the
> appearance of the fedora "f" symbol, which stays up for a couple
> seconds, then ... hang. display goes black, no desktop, no virtual
> consoles.
>
>    i can do more testing but, at this point, it seems that the newest
> rawhide kernels don't play well with a connected HDMI monitor during
> boot. this is relatively new -- i've verified that the earlier
> 3.16.0-0.rc0.git1.1.fc21.x86_64 kernel works just fine booting
> connected to HDMI.
>
>    one more thing ... once i boot with this new kernel, *then* plug in
> the monitor, everything seems to work fine, but when i went to test
> another reboot, the shutdown spewed kernel panic and call trace
> messages all over the screen. i have no idea if this is related, but
> i'm willing to do more testing if you have anything specific in mind.
>
> rday
>

dmesg?
GTX 560M GDDR5?


poma