I am unable to install Mate 16.04 LTS due to two key issues.
(1) I use a UK keyboard, not a US keyboard and so am unable tu try out 16.04 LTS without resorting to a command line to set the keyboard up.
(2) I am unable to get the video driver to work in anything other than 640x480 (by using nomodeset and radeon.modeset=0 options) and most of the windows just don’t work in this resolution. All the buttons to apply things are off the bottom of the display.
The hardware is a Commell LV386 (AMD G-T56N based system).
PCI Devices
-PCI Devices-
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 14h Processor Root Complex
VGA compatible controller : Advanced Micro Devices, Inc. [AMD/ATI] Wrestler [Radeon HD 6320] (prog-if 00 [VGA controller])
PCI bridge : Advanced Micro Devices, Inc. [AMD] Family 14h Processor Root Port (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD] Family 14h Processor Root Port (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD] Family 14h Processor Root Port (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD] Family 14h Processor Root Port (prog-if 00 [Normal decode])
SATA controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 SATA Controller [AHCI mode] (rev 40) (prog-if 01 [AHCI 1.0])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB OHCI0 Controller (prog-if 10 [OHCI])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB EHCI Controller (prog-if 20 [EHCI])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB OHCI0 Controller (prog-if 10 [OHCI])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB EHCI Controller (prog-if 20 [EHCI])
SMBus : Advanced Micro Devices, Inc. [AMD/ATI] SBx00 SMBus Controller (rev 42)
ISA bridge : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 LPC host controller (rev 40)
PCI bridge : Advanced Micro Devices, Inc. [AMD/ATI] SBx00 PCI to PCI Bridge (rev 40) (prog-if 01 [Subtractive decode])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB OHCI2 Controller (prog-if 10 [OHCI])
PCI bridge : Advanced Micro Devices, Inc. [AMD/ATI] SB700/SB800/SB900 PCI to PCI bridge (PCIE port 0) (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD/ATI] SB700/SB800/SB900 PCI to PCI bridge (PCIE port 1) (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD/ATI] SB900 PCI to PCI bridge (PCIE port 2) (prog-if 00 [Normal decode])
PCI bridge : Advanced Micro Devices, Inc. [AMD/ATI] SB900 PCI to PCI bridge (PCIE port 3) (prog-if 00 [Normal decode])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB OHCI0 Controller (prog-if 10 [OHCI])
USB controller : Advanced Micro Devices, Inc. [AMD/ATI] SB7x0/SB8x0/SB9x0 USB EHCI Controller (prog-if 20 [EHCI])
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 0 (rev 43)
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 1
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 2
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 3
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 4
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 6
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 5
Host bridge : Advanced Micro Devices, Inc. [AMD] Family 12h/14h Processor Function 7
Ethernet controller : Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05)
Ethernet controller : Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05)
Ethernet controller : Intel Corporation 82583V Gigabit Network Connection
Ethernet controller : Intel Corporation 82583V Gigabit Network Connection
Serial controller : Oxford Semiconductor Ltd Device c208 (prog-if 02 [16550])
<tt> 4000-40ff </tt> : Advanced Micro Devices, Inc. [AMD/ATI] Wrestler [Radeon HD 6320] (prog-if 00 [VGA controller])
Any help on how to force the video to 1280x1024x24 (The settings I use for uBuntu 10.04 LTS and Mate 14.04 LTS when I got that working by forcing various linux command line settings).
Installation media: 4GB USB Stick created from the MATE install image:
ubuntu-mate-16.04-desktop-amd64.iso
The uBuntu 10.04 LTS GUI app was used to do the creation.
The system is using EFI boot so it skips all the options you get to
changes things that you get when using legacy boot :(.
I will try acpi=off but this may not solve the issue as I need acpi=on
to get some of the other bits of hardware set up properly.
In 14.04 LTS I managed to get things to work without setting nomodeset.
This no-longer works so I am going to look at what has changed in the
Kernel drivers and apply various hardware specific fixes needed for my
hardware. Most of these were developed for the 3.2 series kernels so may
not be applicable but it all looks like a NULL pointer deference in the
radeon DRM code. For the 3.2 series kernels, radeon_dp_getsinktype in
atombios_dp.c failed to validate dig_connector->dp_i2c_bus before use
resulting in a NULL pointer deference.
Turning EFI off is easier said than done as I would have to disconnect
the uBuntu 10.04 LTS SSD that I don’t want to trash just yet :).
It would also not make much difference to the primary issue of the GUI
crashing. It is a NULL pointer dereference bug (probably caused by some
incorrect BIOS code - got at the info via serial debug ).
I am trying a kernel rebuild (not easy on a USB stick with 640x480 only)
with a patch that fixes the symptom (and worked with the 3.2 series
kernels). I will put the updated driver onto the USB stick and see what
happens. If all is the OK I will push the very short patch via launchpad
as it should not break anything and avoids having to work back until I
find out what is actually broken (far to complicated to sort out why the
DDC itc bus pointer is NULL. Better left to a radeon developer to sort
out how it can happen and if required, get me to test various debug
patches).
I eventually got a set up on a USB disk that allowed me to do a proper
kernel build from the 4.4.0_22 sources (using a different computer).
There were four radeon drm files where it was possible to get a NULL
pointer dereference due to the assumption that all connectors have DDC.
After changing these files to eliminate the invalid assumption, I built
up a tweaked generic kernel and installed onto my modified uBuntu-MATE
LiveCD USB stick. This solved the issue I was having so I have edited
the subject line accordingly.
I have attached a patch with the Kernel changes (xz compressed):
Hopefully this will go through to the Kernel and an updated set of ISO
images can be created. Until then I will use my hacked USB install stick.
The next step is to generate a proper modified debian style kernel that
I can install in place of the default generic kernels as there are other
hardware related changes that I need (work around BIOS/ACPI bugs and a
Kernel serial driver bug fix/enhancement for OxPCI/PCIe boards).
).