drm/nouveau: fix regression on original nv50 board

Xorg (and any non-DRM client really) doesn't have permission to directly
touch VRAM on nv50 and up, which the fence code prior to g84 depends on.

It's less invasive to temporarily grant it premission to do so, as it
previously did, than it is to rework fencenv50 to use the VM.  That
will come later on.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index 99cd9e4..3440fc9 100644 (file)
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -285,6 +285,7 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
        struct nouveau_software_chan *swch;
        struct nv_dma_v0 args = {};
        int ret, i;
+       bool save;
 
        nvif_object_map(chan->object);
 
@@ -386,7 +387,11 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
        }
 
        /* initialise synchronisation */
-       return nouveau_fence(chan->drm)->context_new(chan);
+       save = cli->base.super;
+       cli->base.super = true; /* hack until fencenv50 fixed */
+       ret = nouveau_fence(chan->drm)->context_new(chan);
+       cli->base.super = save;
+       return ret;
 }
 
 int