Ben Skeggs
13d7b06baf
drm/nv50: use a slightly different initial context for nv96
...
I'm not 100% sure that the nv94 one we were using won't work. The context
layouts are identical (well.. same ctxprog, so of course!), only a couple
of registers differ. But, be safe until we actually get some 9xxx chips
working.
2009-02-10 09:07:28 +10:00
Ben Skeggs
8a60b2d6a8
drm/nv50: correct ramfc pointer in channel header
...
Suprisingly the card still worked without this...
2009-02-10 09:02:17 +10:00
Ben Skeggs
bc0e48baa3
nv50: support chipset NV96
...
ctxprog seen in okias' trace identical to one we use on NV94, assuming
the initial context values for NV94 will work here too.
2009-02-10 09:01:00 +10:00
Ben Skeggs
31caf2f9a9
drm/nv50: let the card handle the initial context switch
...
Our PFIFO/PGRAPH context save/load functions don't really work well
(at all?) on nv5x yet. Depending on what random state the card is
in before the drm loads, fbcon probably won't work correctly.
Luckily we've setup the GPU in such a way that it'll actually do a
hw context switch for the first context. Not sure of how successful
this'd be currently on the older chips (actually, pretty sure it won't
work), so NV50 only for now.
2009-02-10 08:57:21 +10:00
Ben Skeggs
2e7db8cc34
drm/nouveau: init fbcon after we flag the card as initialised
2009-02-10 08:56:17 +10:00
Pekka Paalanen
3762c97c82
drm compat: fix euid for >=2.6.28
...
drm_fops.c reads the current process' EUID directly from task_struct.
Apparently starting in 2.6.28-rc4 this fails to build.
In Linus' tree, commit b6dff3ec5e116e3af6f537d4caedcad6b9e5082a
"CRED: Separate task security context from task_struct"
moves the euid field from task_struct to another struct.
Earlier commit 9e2b2dc4133f65272a6d3c5dcb2ce63f8a87cae9
"CRED: Introduce credential access wrappers" implements the wrapper
macros to access e.g. euid. This is in 2.6.27-rc4, and this contains the
definition of current_euid() that will be used in the DRM compatibility header
for kernels before 2.6.27. That commit also creates <linux/cred.h>, which
contains the upstream definition of current_euid().
drm_fops.c is fixed to use current_euid(), and drm_compat.h will offer
the compatibility definition for kernels <2.6.27.
Signed-off-by: Pekka Paalanen <pq@iki.fi>
2009-02-10 08:55:32 +10:00
Ben Skeggs
933995b1da
drm/nouveau: store vram/gart object handles for each channel
2009-02-10 08:55:30 +10:00
Ben Skeggs
417c2c411f
nv50: remove some badness, we bail in these situations now anyway
2009-02-06 05:29:14 +10:00
Ben Skeggs
1952a896d0
nouveau: fix some issues where buffer objects never get freed
2009-02-05 23:03:29 +10:00
Ben Skeggs
95fc1b0b52
nv50/kms: be less verbose without debug=1
2009-02-04 15:03:48 +10:00
Ben Skeggs
15ed426dd4
nv50/kms: use DRM_DEBUG
2009-02-04 15:03:48 +10:00
Ben Skeggs
988b805fd6
nv50/kms: oops, re-enable fbcon unregister, got disabled while debugging
2009-02-04 15:03:47 +10:00
Ben Skeggs
bf8181b78a
nv50/kms: store dcb entry ptr & or in encoder struct, remove nv50_output.h
2009-02-04 15:03:47 +10:00
Ben Skeggs
5923c976fb
nv50/kms: nv50_kms_ -> nv50_
2009-02-04 15:03:47 +10:00
Ben Skeggs
cd2a87a682
nv50/kms: remove the last of nv50_kms_wrapper
2009-02-04 15:03:47 +10:00
Ben Skeggs
69beb9c191
nv50/kms: remove last bit of state mirroring
2009-02-04 15:03:46 +10:00
Ben Skeggs
0d7e613b66
nv50/kms: remove nv50_fb
2009-02-04 15:03:46 +10:00
Ben Skeggs
af8a76bd22
nv50/kms: get rid of nv50_fb_info
2009-02-04 15:03:46 +10:00
Ben Skeggs
f16514cccc
nv50/kms: remove kms_priv
2009-02-04 15:03:46 +10:00
Ben Skeggs
8ed61cf557
nv50/kms: remove nv50_kms_get_scaling_mode()
2009-02-04 15:03:46 +10:00
Ben Skeggs
98e859f2f7
nv50/kms: remove redefinitions of scaling modes..
2009-02-04 15:03:45 +10:00
Ben Skeggs
140a2da135
nv50/kms: remove or_counter, unused
2009-02-04 15:03:45 +10:00
Ben Skeggs
9f20d45d0a
nv50/kms: clean up dcb_table usage a little
2009-02-04 15:03:45 +10:00
Ben Skeggs
d2b4a1fc8b
nv50/kms: remove now unused fields from structs
2009-02-04 15:03:45 +10:00
Ben Skeggs
12f2125f11
nv50/kms: remove nv50_connector.type, also redundant
2009-02-04 15:03:44 +10:00
Ben Skeggs
c45ad75cd5
nv50/kms: remove nv50_output.type, it's redundant
2009-02-04 15:03:44 +10:00
Ben Skeggs
d84972f0eb
nv50/kms: get rid of struct nouveau_mode
2009-02-04 15:03:44 +10:00
Ben Skeggs
ef42af67ce
nv50/kms: merge nv50_kms_crtc and nv50_crtc
2009-02-04 15:03:44 +10:00
Ben Skeggs
1b6ff5a4db
nv50/kms: merge nv50_output and nv50_kms_output
2009-02-04 15:03:44 +10:00
Ben Skeggs
4736d2172f
nv50/kms: merge nv50_connector and nv50_kms_connector
2009-02-04 15:03:43 +10:00
Ben Skeggs
8e96ee5341
nouveau: bring in current mm implementation
2009-02-04 15:03:43 +10:00
Ben Skeggs
68891ae2ef
nouveau: install libdrm_nouveau with libdrm
2009-02-04 15:03:34 +10:00
Ben Skeggs
6ef7b032ea
nouveau: bring in new mm api definitions, without the actual mm code
...
Use of the new bits is guarded with a mm_enabled=0 hardcode.
2009-02-04 13:28:39 +10:00
Ben Skeggs
34b74c06b6
nv50: ack nsource to prevent continuous protection fault irqs
2009-02-04 13:28:19 +10:00
Ben Skeggs
91a86bdc12
ttm: fix drm_fence_wait_polling timeout value
2009-02-04 13:27:54 +10:00
Dave Airlie
bbd5e64216
radeon: make dword writes be inlined
...
This moves the size checks to the begin section and makes dword
writes inline for all CS people.
Gets me back some of my CPU
2009-02-04 11:11:55 +10:00
Dave Airlie
98738555d1
radeon: add space accounting to cs code
2009-01-31 01:50:23 +10:00
Dave Airlie
e9298a02ac
radeon: add bo_wait functionality
2009-01-13 16:48:04 +10:00
Dave Airlie
2b0c9c01c2
radeon: add new getparam to get pcidevice
2009-01-11 09:49:00 +10:00
Dave Airlie
d587c6f150
radeon: make name consistent with libdrm_intel
2008-12-22 12:26:56 +10:00
Dave Airlie
b3d4e7433d
radeon: size the relocs correctly
2008-12-22 11:57:15 +10:00
Dave Airlie
c93f1c88ce
radeon: drop more cs2
2008-12-19 16:44:22 +10:00
Dave Airlie
8c40a5c648
radeon: more cs2 removal
2008-12-19 16:43:01 +10:00
Dave Airlie
4b10ddff78
radeon: drop CS2 into CS, remove start/end offset
2008-12-19 16:31:55 +10:00
Jesse Barnes
12e68f8059
Merge branch 'master' into modesetting-gem
2008-12-03 11:54:07 -08:00
Jesse Barnes
d5d5aca7f9
Merge branch 'master' into modesetting-gem
2008-12-03 11:53:36 -08:00
Eric Anholt
b0d93c74d8
Bump version to 2.4.2 to signal new libdrm_intel functions.
2008-12-02 12:19:37 -08:00
Eric Anholt
6fb1ad767d
intel: Add a function for setting (GTT,GTT) domain, for use by UXA.
...
This function can also serve the role that the bo_wait_rendering did, when
write_enable is unset.
2008-12-02 12:19:34 -08:00
Ben Skeggs
7e4e0fbbb8
nv50: support NV94 chipset
2008-11-23 18:49:09 +11:00
Ben Skeggs
52232ad702
nv50: update context-related tables for original 8800GTS
...
I either messed up when I pulled these from a mmio-trace last time, or
the previous values didn't work on my card. Hopefully it's the former!
In any case, at least one of the original NV50 chipsets work now.
2008-11-21 13:18:58 +11:00
