[OpenIndiana-discuss] Nobody for whom OI/Hipster Mate on i915/KMS is working fine wants to report?

[Мартин Бохниг]

Hi Apostolos,

don't let yourself get confused by the line:

> [ 185.362] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support


Solaris 11 ships 

/11.3/platform/i86pc/kernel/drv/amd64/vga_arbiter
/11.3/platform/i86pc/kernel/drv/vga_arbiter.conf

But you don't necessarily need it only to get multicard support, especially not as long as you do not depend on having multiple heads functioning simultaneously.

I thought for some time to add vga_arbiter to Illumos, but now that they thanked me for my 11 years of unpaid OpenSolaris enthusisam like *that* and finally also in full day-light declared their "free multipurpose OS" to be a closed commercial group's x64-server-only cash-cow, they are equally as free to mess with it themselves (and they probably don't even want to, as they proudly state themselves that only MS/NSA-Windows can be used for Desktops, interesting statement for the "community" "leader"ship of a "free" UNIX kernel). Well, everybody saw it.

But as said, you can safely ignore this line in Xorg.0.log.
Here more info about it:

https://www.kernel.org/doc/Documentation/vgaarbiter.txt  


"""""VGA Arbiter
===========

Graphic devices are accessed through ranges in I/O or memory space. While most
modern devices allow relocation of such ranges, some "Legacy" VGA devices
implemented on PCI will typically have the same "hard-decoded" addresses as
they did on ISA. For more details see "PCI Bus Binding to IEEE Std 1275-1994
Standard for Boot (Initialization Configuration) Firmware Revision 2.1"
Section 7, Legacy Devices.

The Resource Access Control (RAC) module inside the X server [0] existed for
the legacy VGA arbitration task (besides other bus management tasks) when more
than one legacy device co-exists on the same machine. But the problem happens
when these devices are trying to be accessed by different userspace clients
(e.g. two server in parallel). Their address assignments conflict. Moreover,
ideally, being a userspace application, it is not the role of the X server to
control bus resources. Therefore an arbitration scheme outside of the X server
is needed to control the sharing of these resources. This document introduces
the operation of the VGA arbiter implemented for the Linux kernel.
"""""


Sample diff:
i915: Add module option to support VGA arbiter on HD devices
https://patchwork.freedesktop.org/patch/25700/ 


So, this is _not_ the reason for the problems we see on your laptop.

Because it _is_ of course possible to use several GPU's on the same system, be it Intel versus NVidia (most frequent use scenario) or anything else.

It depends heavily on the system's BIOS/FW settings which GPU's x86-BIOS codes get executed during POST and which device gets chosen as boot-console etc (it can be more than one, but BIOS handles it proprietarily, so luck is also involved).

Derived from that it influences how successful you later are on such dual- or multi-GPU systems after the kernel was booted, to start Xorg on one or more of these GPU's, via customizing xorg.conf .


Add to this complexity, that in such cases 

/lib/opengl/ogl_select/nvidia_vendor_select
/lib/opengl/ogl_select/mesa_vendor_select

may frequently get confused, too.

If you tell me more about your systems, we can hopefully get DRM working for you, be it with my Intel-DRM/KMS S12 backport for Intel and/or with NVidia's closed src drivers for NVidia.



best regards,
%martin bochnig, East-German from East-Berlin




>Четверг,  8 сентября 2016, 8:33 UTC от Apostolos Syropoulos <asyropoulos at yahoo.com>:
>
>
>Also, this is the part of Xorg.0.log that reports about the
>nVIDIA card. Note that for some reason it cannot probe for
>an alternative video card.
>
> 
>[   185.362] (WW) NVIDIA(0): The NVIDIA GeForce 6200 TurboCache(TM) GPU installed in this
>[   185.362] (WW) NVIDIA(0):     system is supported through the NVIDIA 304.xx Legacy
>[   185.362] (WW) NVIDIA(0):     drivers. Please visit
>[   185.362] (WW) NVIDIA(0):  http://www.nvidia.com/object/unix.html for more
>[   185.362] (WW) NVIDIA(0):     information.  The 340.96 NVIDIA driver will ignore this
>[   185.362] (WW) NVIDIA(0):     GPU.  Continuing probe...
>[   185.362] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card suppor
>