diff options
| author | Ben Skeggs <bskeggs@redhat.com> | 2011-03-07 17:18:04 +1000 |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2011-03-08 07:03:08 +1000 |
| commit | 6f70a4c3d19e8e8e1047a4dbf0ca910fed39f619 (patch) | |
| tree | 53d45ab91ea31167710904f242b2dbdd33b1dc42 /drivers/gpu/drm/nouveau/nv50_vm.c | |
| parent | ef1b287169cd3d1e428c8ed8222e0bbf733d5dbb (diff) | |
drm/nv50-nvc0: prevent multiple vm/bar flushes occuring simultanenously
The per-vm mutex doesn't prevent this completely, a flush coming from the
BAR VM could potentially happen at the same time as one for the channel
VM. Not to mention that if/when we get per-client/channel VM, this will
happen far more frequently.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Diffstat (limited to 'drivers/gpu/drm/nouveau/nv50_vm.c')
| -rw-r--r-- | drivers/gpu/drm/nouveau/nv50_vm.c | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/drivers/gpu/drm/nouveau/nv50_vm.c b/drivers/gpu/drm/nouveau/nv50_vm.c index 459ff08241e..6144156f255 100644 --- a/drivers/gpu/drm/nouveau/nv50_vm.c +++ b/drivers/gpu/drm/nouveau/nv50_vm.c @@ -169,7 +169,11 @@ nv50_vm_flush(struct nouveau_vm *vm) void nv50_vm_flush_engine(struct drm_device *dev, int engine) { + struct drm_nouveau_private *dev_priv = dev->dev_private; + + spin_lock(&dev_priv->ramin_lock); nv_wr32(dev, 0x100c80, (engine << 16) | 1); if (!nv_wait(dev, 0x100c80, 0x00000001, 0x00000000)) NV_ERROR(dev, "vm flush timeout: engine %d\n", engine); + spin_unlock(&dev_priv->ramin_lock); } |