Clevo P150EM HCL - kernel 3.9.2-1 and kernel 3.9.2-2

[Andrew Sorensen]

What's the difference between the packaged kernels 3.9.2-1 and 3.9.2-2?
I remember being stuck with kernel-devel-3.9.2-1 in R2B2 (which was
missing files I needed), but now I'm able to use the kernel-devel from
3.9.2-2 with 3.9.2-1 and build my modules.

Here's my current summary of kernels on my machine (Clevo P150EM) -
interrupt and DMA remapping enabled (VT-d):

3.7.x-3.9: USB support broken (disconnect event not triggered - prevents
system sleep and USB from working after unplugging devices). Upstream
issue fixed in 3.9.

3.9.2-1: Intel HD 4000 graphics works, no issues to report (I am
building my kernel modules against kernel-devel-3.9.2-2).

3.9.2-2: Intel HD 4000 graphics completely broken, can not start KDM.

3.11.1: Intel HD 4000 graphics partially broken (only two displays work,
3rd has EDID decoding trouble).

I'll update the HCL when I get a chance.

[Marek Marczykowski-Górecki]

On 15.12.2013 21:50, Andrew Sorensen wrote:
> What's the difference between the packaged kernels 3.9.2-1 and 3.9.2-2?
> I remember being stuck with kernel-devel-3.9.2-1 in R2B2 (which was
> missing files I needed), but now I'm able to use the kernel-devel from
> 3.9.2-2 with 3.9.2-1 and build my modules.

Looking at the git log the only difference is kernel-devel package, which was
broken in 3.9.2-1.

> Here's my current summary of kernels on my machine (Clevo P150EM) -
> interrupt and DMA remapping enabled (VT-d):
>
> 3.7.x-3.9: USB support broken (disconnect event not triggered - prevents
> system sleep and USB from working after unplugging devices). Upstream
> issue fixed in 3.9.
>
> 3.9.2-1: Intel HD 4000 graphics works, no issues to report (I am
> building my kernel modules against kernel-devel-3.9.2-2).
>
> 3.9.2-2: Intel HD 4000 graphics completely broken, can not start KDM.

Strange, there shouldn't be any difference in kernel itself between those two
versions. Did you get any error messages? Do you have text console access?

> 3.11.1: Intel HD 4000 graphics partially broken (only two displays work,
> 3rd has EDID decoding trouble).

This is probably not Qubes-related problem... Did you searched the web for any
hints? Perhaps someone can tell on which kernel this device is working.

> I'll update the HCL when I get a chance.
>

--
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

[Andrew Sorensen]

On 12/15/13 13:10, Marek Marczykowski-Górecki wrote:
> On 15.12.2013 21:50, Andrew Sorensen wrote:
>> What's the difference between the packaged kernels 3.9.2-1 and 3.9.2-2?
>> I remember being stuck with kernel-devel-3.9.2-1 in R2B2 (which was
>> missing files I needed), but now I'm able to use the kernel-devel from
>> 3.9.2-2 with 3.9.2-1 and build my modules.
> Looking at the git log the only difference is kernel-devel package, which was
> broken in 3.9.2-1.
>
>> Here's my current summary of kernels on my machine (Clevo P150EM) -
>> interrupt and DMA remapping enabled (VT-d):
>>
>> 3.7.x-3.9: USB support broken (disconnect event not triggered - prevents
>> system sleep and USB from working after unplugging devices). Upstream
>> issue fixed in 3.9.
>>
>> 3.9.2-1: Intel HD 4000 graphics works, no issues to report (I am
>> building my kernel modules against kernel-devel-3.9.2-2).
>>
>> 3.9.2-2: Intel HD 4000 graphics completely broken, can not start KDM.
> Strange, there shouldn't be any difference in kernel itself between those two
> versions. Did you get any error messages? Do you have text console access?

Xorg.log:

"Fatal server error:
fatal error, fail to get element buffer. GL context may not be created
correctly."

Since Xorg already took control of the keyboard it is not possible to
regain access to the system (magic keys are disabled).

>> 3.11.1: Intel HD 4000 graphics partially broken (only two displays work,
>> 3rd has EDID decoding trouble).
> This is probably not Qubes-related problem... Did you searched the web for any
> hints? Perhaps someone can tell on which kernel this device is working.
>

Yes, I agree. Triple monitor support on the HD 4000 has been a hit and
miss since it was released. This is more of a note for anyone else who
stumbles on this thread while using Qubes.