Created attachment 34737 [details]
modeset=0, Xorg.0.log

Created attachment 34738 [details]
modeset=1, dmesg

Created attachment 34739 [details]
modeset=1, Xorg.0.log

This is with the latest 2.3.32 ubuntu kernels as well as vanilla 2.3.32.11

The drm is failing to load when kms is enabled likely due to radeonfb claiming the device.  try blacklisting radeonfb.

Based on Michael's comments, it looks like radeonfb and radeon can live side-by-side, and radeonfb is needed for console access:

http://lists.x.org/archives/xorg-devel/2009-May/000814.html

I'd rather not blacklist radeonfb and loose console.

Does the KMS support in 2.6.32's radeon driver allow console access without radeonfb, and if so, is there anything special I need to do to use it?

(In reply to comment #6)
> Based on Michael's comments, it looks like radeonfb and radeon can live
> side-by-side, and radeonfb is needed for console access:
> 
> http://lists.x.org/archives/xorg-devel/2009-May/000814.html

before KMS, the drm coexisted with kernel fb driver since it didn't provide an fb interface itself.  Now the drm is a real driver.

> 
> I'd rather not blacklist radeonfb and loose console.
> 
> Does the KMS support in 2.6.32's radeon driver allow console access without
> radeonfb, and if so, is there anything special I need to do to use it?

The drm provides a kernel fb interface for the console just like radeonfb.  So it's a replacement for radeonfb and then some.

Although make sure your radeon drm is built with kms support.  It was in staging until 2.6.33, so you have to enable staging drivers to use it.

Thinking about it more, I suspect you don't have a kms enabled drm, but the drm is failing to load since you've passed the module an invalid option (modeset, which is invalid if your drm doesn't have kms support).

(In reply to comment #9)
> Thinking about it more, I suspect you don't have a kms enabled drm, but the drm
> is failing to load since you've passed the module an invalid option (modeset,
> which is invalid if your drm doesn't have kms support).

No, that isn't the case because it works successfully when I give it the option modeset=0

Also modinfo reports:

parm:           no_wb:Disable AGP writeback for scratch registers (int)
parm:           modeset:Disable/Enable modesetting (int)
parm:           dynclks:Disable/Enable dynamic clocks (int)
parm:           r4xx_atom:Enable ATOMBIOS modesetting for R4xx (int)
parm:           vramlimit:Restrict VRAM for testing (int)
parm:           agpmode:AGP Mode (-1 == PCI) (int)
parm:           gartsize:Size of PCIE/IGP gart to setup in megabytes (32,64, etc)
 (int)
parm:           benchmark:Run benchmark (int)
parm:           test:Run tests (int)
parm:           connector_table:Force connector table (int)
parm:           tv:TV enable (0 = disable) (int)
parm:           new_pll:Select new PLL code for AVIVO chips (int)
parm:           audio:Audio enable (0 = disable) (int)

And I do have staging enabled:

CONFIG_STAGING=y
CONFIG_STAGING_EXCLUDE_BUILD=y

(In reply to comment #10)
> No, that isn't the case because it works successfully when I give it the option
> modeset=0
>

Ah right.  It's probably radeonfb then.

Created attachment 34745 [details]
2.6.33.2,radeon,no_radeonfb dmesg

Created attachment 34746 [details]
2.6.33.2,radeon,no_radeonfb Xorg.0.log

Ok, I built 2.6.33.2 without radeonfb and enabled KMS on the radeon drm driver.  lspci -v reports:

0000:00:10.0 VGA compatible controller: ATI Technologies Inc RV350 [Mobility Radeon 9600 M10]
	Subsystem: ATI Technologies Inc RV350 [Mobility Radeon 9600 M10]
	Flags: bus master, 66MHz, medium devsel, latency 255, IRQ 48
	Memory at b8000000 (32-bit, prefetchable) [size=128M]
	I/O ports at 0400 [size=256]
	Memory at b0000000 (32-bit, non-prefetchable) [size=64K]
	Expansion ROM at b0020000 [disabled] [size=128K]
	Capabilities: [58] AGP version 2.0
	Capabilities: [50] Power Management version 2
	Kernel driver in use: radeon

The fb mode is completely garbage.  I waited long enough, and eventually X loaded, but I could not read any of the garbled output on the display in console mode.

/dev/dri is present:

root@aerith:~# ls -l /dev/dri*
total 0
crw-rw----+ 1 root video 226,  0 2010-04-06 19:19 card0
crw-rw----  1 root video 226, 64 2010-04-06 19:19 controlD64

but glxinfo shows swrast being used

[    0.392633] platform radeon_cp.0: firmware: requesting radeon/R300_cp.bin
[   60.398776] radeon_cp: Failed to load firmware "radeon/R300_cp.bin"
[   60.411715] radeon 0000:00:10.0: failled initializing CP (-2).
[   60.418207] radeon 0000:00:10.0: Disabling GPU acceleration

# sha1sum /lib/firmware/radeon/R300_cp.bin 
76925e062bc2c06b23ead294bf3fdaedc4a10bdf  /lib/firmware/radeon/R300_cp.bin

# dpkg -S /lib/firmware/radeon/R300_cp.bin 
linux-firmware: /lib/firmware/radeon/R300_cp.bin

# aptitude show linux-firmware
Package: linux-firmware
State: installed
Automatically installed: no
Version: 1.33
Priority: optional
Section: misc
Maintainer: Ubuntu Kernel Team <kernel-team@lists.ubuntu.com>
Uncompressed Size: 16.1M
Conflicts: atmel-firmware
Replaces: atmel-firmware, linux-restricted-common
Provides: atmel-firmware
Description: Firmware for Linux kernel drivers
 This package provides firmware used by Linux kernel drivers.

> --- Comment #16 from Jeremy Huddleston <jeremyhu@freedesktop.org> 2010-04-06 19:31:20 PDT ---
> [    0.392633] platform radeon_cp.0: firmware: requesting radeon/R300_cp.bin
> [   60.398776] radeon_cp: Failed to load firmware "radeon/R300_cp.bin"
> [   60.411715] radeon 0000:00:10.0: failled initializing CP (-2).
> [   60.418207] radeon 0000:00:10.0: Disabling GPU acceleration

If drm & co modules are in initrd you need the firmware in initrd too.
Updating the initrd should be enough.

If you compiled modules into kernel then you need the firmware in kernel too.

alright.  thanks.  it now reports the hw rendering... although it does not work.  I'll followup in other bugs for cleanliness