nouveau_ttm_tt_unpopulate() is supposed to return right after calling
ttm_dma_unpopulate() in the case of a coherent buffer. The return
statement was omitted, leading to the pages being unmapped twice. Fix
this.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Preparation for transition to planes, which use framebuffers for the
cursor image. We've always done copies from the userspace buffer up
until now for legacy reasons, there's no good reason to do so on the
chipsets this code covers.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
We have the ability to move buffers around in the kernel if necessary,
and should probably use it rather than failing if userspace passes us
a non-contig buffer for a plane.
The NOUVEAU_GEM_TILE_NONCONTIG flag from userspace will become a mere
initial placement hint once all the relevant paths have been updated.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
The voltage value are calculated by the hardware characterized
result.
Signed-off-by: Vince Hsu <vinceh@nvidia.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Acked-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
For GK20A we need the GPU speedo value to calculate voltage levels.
Signed-off-by: Vince Hsu <vinceh@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Move the vbios parsing out of init() and call it conditionally if the
platform has a vbios. Non-vbios platforms can use the ctor() to init the
data structures.
Signed-off-by: Vince Hsu <vinceh@nvidia.com>
Acked-by: Alexandre Courbot <acourbot@nvidia.com>
Acked-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
A machine has been spotted where the ramcfg strap is "8", and the ramcfg
xlat table goes 0-7,0-7, resulting in us selecting config 0 for memory
items. On this particular system, config "8" is available and supposed
to be used. It appears that starting from GT21x (where Mv2 appears),
we're supposed to use the value in this table instead.
One concern here is that not all the places we currently use ramcfg xlat
are supposed to be treated the same now. The strap xlat table wasn't
removed from the vbios either, presumably for some kind of good reason.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
We only support one kind of matching here (ramcfg strap), but it appears
alternate methods are possible. I wrote a tool to scan our vbios repo
for other types, but did not see any used. Hopefully this means there
aren't any in the wild that will now break.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
For some reason max_vco was set to a lower value that it can support,
which prevented some clock states to be applied. Fix this by setting it
to the same value as downstream.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
On architectures for which access to GPU memory is non-coherent,
caches need to be flushed and invalidated explicitly when BO control
changes between CPU and GPU.
This patch adds buffer synchronization functions which invokes the
correct API (PCI or DMA) to ensure synchronization is effective.
Based on the TTM DMA cache helper patches by Lucas Stach.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Specify TTM_PL_FLAG_UNCACHED when allocating GPFIFOs and fences to
allow them to be safely accessed by the kernel without being synced
on non-coherent architectures.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Allow nouveau_bo_new() to recognize the TTM_PL_FLAG_UNCACHED flag, which
means that we want the allocated BO to be perfectly coherent between the
CPU and GPU. This is useful on non-coherent architectures for which we
do not want to manually sync some rarely-accessed buffers: typically,
fences and pushbuffers.
A TTM BO allocated with the TTM_PL_FLAG_UNCACHED on a non-coherent
architecture will be populated using the DMA API, and accesses to it
performed using the coherent mapping performed by dma_alloc_coherent().
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Add a function allowing us to know whether a device is CPU-coherent,
i.e. accesses performed by the CPU on GPU-mapped buffers will
be immediately visible on the GPU side and vice-versa.
For now, a device is considered to be coherent if it uses the PCI bus on
a non-ARM architecture.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Pinned BOs are supposed to remain in their current location until
unpinned. Display a warning for the supposedly-erroneous case where we
are trying to move such objects.
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Probably missing something here, doesn't make a lot of sense to write
or+link data into a register whose offset is calculated by the same
or+link info..
This is the all I've witnessed the binary driver and vbios doing so
far, so it'll do.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
