diff options
author | Dave Airlie <airlied@redhat.com> | 2013-11-15 12:24:40 +1000 |
---|---|---|
committer | Dave Airlie <airlied@redhat.com> | 2013-11-15 12:24:40 +1000 |
commit | 0846c728e20a0cd1e43fb75a3015f3b176a26466 (patch) | |
tree | a65d8b3a8c4860f79f9bf634d13ff3405fb5e217 /drivers/gpu/drm/nouveau/nouveau_drm.c | |
parent | ad40f83f5a89f6d723fd4db424b531f8dd7d3b49 (diff) | |
parent | d2c7ab32ceb0737c1aba84765c555e0f73d3c935 (diff) |
Merge branch 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6 into drm-next
- Page flipping fixes, with support for syncing them to vblank (finally...).
- Misc other general fixes
* 'drm-nouveau-next' of git://anongit.freedesktop.org/git/nouveau/linux-2.6:
drm/nouveau: do not map evicted vram buffers in nouveau_bo_vma_add
drm/nvc0-/gr: shift wrapping bug in nvc0_grctx_generate_r406800
drm/nouveau/pwr: fix missing mutex unlock in a failure path
drm/nv40/therm: fix slowing down fan when pstate undefined
drm/nv11-: synchronise flips to vblank, unless async flip requested
drm/nvc0-: remove nasty fifo swmthd hack for flip completion method
drm/nv10-: we no longer need to create nvsw object on user channels
drm/nouveau: always queue flips relative to kernel channel activity
drm/nouveau: there is no need to reserve/fence the new fb when flipping
drm/nouveau: when bailing out of a pushbuf ioctl, do not remove previous fence
drm/nouveau: allow nouveau_fence_ref() to be a noop
drm/nvc8/mc: msi rearm is via the nvc0 method
Diffstat (limited to 'drivers/gpu/drm/nouveau/nouveau_drm.c')
-rw-r--r-- | drivers/gpu/drm/nouveau/nouveau_drm.c | 27 |
1 files changed, 27 insertions, 0 deletions
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 2418b0de589e..7a3759f1c41a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -37,6 +37,7 @@ #include <engine/device.h> #include <engine/disp.h> #include <engine/fifo.h> +#include <engine/software.h> #include <subdev/vm.h> @@ -191,6 +192,32 @@ nouveau_accel_init(struct nouveau_drm *drm) return; } + ret = nouveau_object_new(nv_object(drm), NVDRM_CHAN, NVDRM_NVSW, + nouveau_abi16_swclass(drm), NULL, 0, &object); + if (ret == 0) { + struct nouveau_software_chan *swch = (void *)object->parent; + ret = RING_SPACE(drm->channel, 2); + if (ret == 0) { + if (device->card_type < NV_C0) { + BEGIN_NV04(drm->channel, NvSubSw, 0, 1); + OUT_RING (drm->channel, NVDRM_NVSW); + } else + if (device->card_type < NV_E0) { + BEGIN_NVC0(drm->channel, FermiSw, 0, 1); + OUT_RING (drm->channel, 0x001f0000); + } + } + swch = (void *)object->parent; + swch->flip = nouveau_flip_complete; + swch->flip_data = drm->channel; + } + + if (ret) { + NV_ERROR(drm, "failed to allocate software object, %d\n", ret); + nouveau_accel_fini(drm); + return; + } + if (device->card_type < NV_C0) { ret = nouveau_gpuobj_new(drm->device, NULL, 32, 0, 0, &drm->notify); |