Radeon driver on HD-7660D - ok so far

[Graham Bradley]

On 13.02.2014 00:34, Graham Bradley wrote:
>> The transient ripple on filling big rectangles is visible.
> 
> I've looked again.... To clarify my earlier comment about 
> rectangles, it may well be coming from the blue 'ncurses' -type
> graphics screen you get when you do a 'make config' in the process of
> installing a port. The procession of character-size colour blocks is
> quite apparent as it forms into a menu (rectangle with text in it)
> but (I guess) it has nothing to do with graphics acceleration. I have
> not observed anything similar in any other context. What terminal are you using?
-- 
Jean-S?bastien P?dron

The terminal is a BenQ LCD 24-inch 1920x1080, connected directly to the DVI output.  Claims a quick response.

I looked back at my previous version... 9.2-RELEASE with the VESA driver, and it seemed much sharper at redrawing xterms (i.e. text)
but of course way slower on the rendering.  That's a subjective comment - it could be down to (I guess) scheduler changes I don't know about.

Incidentally, I just noticed what may be an error message in X.org.0.log (already supplied to you)
from which I snip this....

PITCAIRN, PITCAIRN, PITCAIRN, PITCAIRN, PITCAIRN, PITCAIRN, PITCAIRN,
VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE,
VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE, VERDE,
VERDE, VERDE, VERDE, OLAND, OLAND, OLAND, OLAND, OLAND, OLAND, OLAND,
OLAND, OLAND, OLAND, OLAND, OLAND, OLAND, HAINAN, HAINAN, HAINAN,
HAINAN, HAINAN, HAINAN, BONAIRE, BONAIRE, BONAIRE, BONAIRE, BONAIRE,
BONAIRE, BONAIRE, BONAIRE, KABINI, KABINI, KABINI, KABINI, KABINI,
KABINI, KABINI, KABINI, KABINI, KABINI, KABINI, KABINI, KABINI,
KABINI, KABINI, KABINI
[  1099.353] (--) Using syscons driver with X support (version 2.0)
[  1099.354] (--) using VT number 9

[  1099.365] (II) [KMS] Kernel modesetting enabled.
[  1099.367] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[  1099.368] (**) RADEON(0): Depth 24, (--) framebuffer bpp 32
[  1099.370] (II) RADEON(0): Pixel depth = 24 bits stored in 4 bytes (32 bpp pix
maps)
[  1099.370] (==) RADEON(0): Default visual is TrueColor
[  1099.371] (==) RADEON(0): RGB weight 888
[  1099.372] (II) RADEON(0): Using 8 bits per RGB (8 bit DAC)
[  1099.373] (--) RADEON(0): Chipset: "ARUBA" (ChipID = 0x9901)
[  1099.374] drmOpenDevice: node name is /dev/dri/card0
[  1099.375] Failed to change owner or group for file /dev/dri! 2: No such file 
or directory
[  1099.375] Failed to change owner or group for file /dev/dri/card0! 2: No such
file or directory
[  1099.376] drmOpenDevice: open result is -1, (No such file or directory)
[  1099.377] Failed to change owner or group for file /dev/dri/card0! 2: No such
file or directory
[  1099.377] drmOpenDevice: open result is -1, (No such file or directory)
[  1099.379] drmOpenDevice: Open failed
[  1101.130] drmOpenByBusid: Searching for BusID pci:0000:00:01.0
[  1101.130] drmOpenDevice: node name is /dev/dri/card0
[  1101.131] drmOpenDevice: open result is 9, (OK)
[  1101.132] drmOpenByBusid: drmOpenMinor returns 9
[  1101.133] drmOpenByBusid: drmGetBusid reports pci:0000:00:01.0
[  1101.134] (II) Loading sub module "dri2"
[  1101.135] (II) LoadModule: "dri2"
[  1101.136] (II) Loading /usr/local/lib/xorg/modules/extensions/libdri2.so
[  1101.137] (II) Module dri2: vendor="X.Org Foundation"
[  1101.138]    compiled for 1.12.4, module version = 1.2.0
[  1101.139]    ABI class: X.Org Server Extension, version 6.0
[  1101.140] (II) Loading sub module "exa"
[  1101.140] (II) LoadModule: "exa"
[  1101.141] (II) Loading /usr/local/lib/xorg/modules/libexa.so
[  1101.143] (II) Module exa: vendor="X.Org Foundation"
[  1101.144]    compiled for 1.12.4, module version = 2.5.0

now the "Failed" messages in the "drmOpen" lines refer to a directory and file /dev/dri/ and /dev/dri/card0
and then the following messages suggest it's ok, so I don't know which to believe!  Does it matter?

but I have the evidence that the location is there and accessible (the user of startx is in the wheel group) ....

[graham at tigger ~]$ ls -ld /dev/dri
dr-xr-xr-x  2 root  wheel  512 Feb 13 18:31 /dev/dri
[graham at tigger ~]$ ls -l /dev/dri/card0 
crw-rw----  1 root  wheel  0x72 Feb 13 18:31 /dev/dri/card0
[graham at tigger ~]$ groups
users wheel firefox usenet
[graham at tigger ~]$ 

Apologies if this is an irrelevance,
regards
Graham