Rework fencing workflow.  Stop using ttm helpers, use the
virtio_gpu_array_* helpers instead.

Due to using the gem reservation object it is initialized and ready for
use before calling ttm_bo_init.  So we can simply use the standard
fencing workflow and drop the tricky logic which checks whenever the
command is in flight still.

v6: rewrite most of the patch.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-10-kraxel@redhat.com

Rework fencing workflow, starting with virtio_gpu_execbuffer_ioctl.
Stop using ttm helpers, use the virtio_gpu_array_* helpers (which work
on the reservation objects directly) instead.

Also store the object array in struct virtio_gpu_vbuffer, so we
explicitly keep a reference of all buffers used instead of depending
on ttm_bo_put() checking whenever the object is actually idle before
releasing it.

New workflow:

 (1) All gem objects needed by a command are added to a
     virtio_gpu_object_array.
 (2) All reservation objects will be locked (virtio_gpu_array_lock_resv).
 (3) virtio_gpu_fence_emit() completes fence initialization.
 (4) fence gets added to the objects, reservation objects are unlocked
     (virtio_gpu_array_add_fence, virtio_gpu_array_unlock_resv).
 (5) virtio command is submitted to the host.
 (6) The completion callback (virtio_gpu_dequeue_ctrl_func)
     will drop object references and free virtio_gpu_object_array.

v6: rewrite most of the patch.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-9-kraxel@redhat.com

Some helper functions to manage an array of gem objects.

v9: use dma_resv_lock_interruptible.
v6:
 - add ticket to struct virtio_gpu_object_array.
 - add virtio_gpu_array_{lock,unlock}_resv helpers.
 - add virtio_gpu_array_add_fence helper.
v5: some small optimizations (Chia-I Wu).
v4: make them virtio-private instead of generic helpers.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-8-kraxel@redhat.com

Call reservation_object_* directly instead
of using ttm_bo_{reserve,unreserve}.

v4: check for EINTR only.
v3: check for EINTR too.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-7-kraxel@redhat.com

All callers pass no_wait = false.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-6-kraxel@redhat.com

No users left.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190829103301.3539-5-kraxel@redhat.com

Drop vma_node from ttm_buffer_object, use the gem struct
(base.vma_node) instead.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190805140119.7337-9-kraxel@redhat.com

Drop use of the deprecated drmP.h header file.
Fix fallout by adding missing include files.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Acked-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NEmil Velikov <emil.velikov@collabora.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: virtualization@lists.linux-foundation.org
Link: https://patchwork.freedesktop.org/patch/msgid/20190630061922.7254-28-sam@ravnborg.org

No need to have our own implementation,
atomic helpers can do it for us.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20190611125408.29421-1-kraxel@redhat.com

virtio_gpu_fence_emit() always returns 0, since it
has no error paths.

Consequently no calls for virtio_gpu_fence_emit()
use the return value, and it can be removed.
Signed-off-by: NRobert Foss <robert.foss@collabora.com>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190506091034.30289-1-robert.foss@collabora.comSuggested-by: NEmil Velikov <emil.velikov@collabora.com>
Reviewed-by: NChia-I Wu <olvaffe@gmail.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This is motivated by having meaningful ftrace events, but it also
fixes use cases where dma_fence_is_later is called, such as in
sync_file_merge.

In other drivers, fence creation and cmdbuf submission normally
happen atomically,

  mutex_lock();
  fence = dma_fence_create(..., ++timeline->seqno);
  submit_cmdbuf();
  mutex_unlock();

and have no such issue.  But in our driver, because most ioctls
queue commands into ctrlq, we do not want to grab a lock.  Instead,
we set seqno to 0 when a fence is created, and update it when the
command is finally queued and the seqno is known.
Signed-off-by: NChia-I Wu <olvaffe@gmail.com>
Reviewed-by: NEmil Velikov <emil.velikov@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190429220825.156644-1-olvaffe@gmail.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

This patch does more harm than good, as it breaks both Xwayland and
gnome-shell with X11.

Xwayland requires DRI3 & DRI3 requires PRIME.

X11 crash for obscure double-free reason which are hard to debug
(starting X11 by hand doesn't trigger the crash).

I don't see an apparent problem implementing those stub prime
functions, they may return an error at run-time, and it seems to be
handled fine by GNOME at least.

This reverts commit b318e3ff.
[airlied:
This broke userspace for virtio-gpus, and regressed things from DRI3 to DRI2.

This brings back the original problem, but it's better than regressions.]

Fixes: b318e3ff ("drm/virtio: drop prime import/export callbacks")
Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This patch moves the virtio_gpu_cmd_create_resource() call (which
notifies the host about the new resource created) into the
virtio_gpu_object_create() function.  That way we can call
virtio_gpu_cmd_create_resource() before ttm_bo_init(), so the host
already knows about the object when ttm initializes the object and calls
our driver callbacks.

Specifically the object is already created when the
virtio_gpu_ttm_tt_bind() callback invokes virtio_gpu_object_attach(),
so the extra virtio_gpu_object_attach() calls done after
virtio_gpu_object_create() are not needed any more.

The fence support for the create ioctl becomes a bit more tricky though.
The code moved into virtio_gpu_object_create() too.  We first submit the
(fenced) virtio_gpu_cmd_create_resource() command, then initialize the
ttm object, and finally attach just created object to the fence for the
command in case it didn't finish yet.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190318113332.10900-6-kraxel@redhat.com

Add 3d resource parameters to virtio_gpu_object_params struct.  With
that in place we can use it for virtio_gpu_cmd_resource_create_3d()
calls.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190318113332.10900-5-kraxel@redhat.com

Add format, width and height fields to the virtio_gpu_object_params
struct.  With that in place we can use the parameter struct for
virtio_gpu_cmd_create_resource() calls too.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190318113332.10900-4-kraxel@redhat.com

Create virtio_gpu_object_params, use that to pass object parameters to
virtio_gpu_object_create.  This is just the first step, followup patches
will add more parameters to the struct.  The plan is to use the struct
for all object parameters.

Drop unused "kernel" parameter for virtio_gpu_alloc_object(), it is
unused and always false.

Also drop "pinned" parameter.  virtio-gpu doesn't shuffle around
objects, so effecively they all are pinned anyway.  Hardcode
TTM_PL_FLAG_NO_EVICT so ttm knows.  Doesn't change much for the moment
as virtio-gpu supports TTM_PL_FLAG_TT only so there is no opportunity to
move around objects.  That'll probably change in the future though.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190318113332.10900-3-kraxel@redhat.com

Just run drm_prime_pages_to_sg() on the ttm pages list to get an
sg_table for export.  The pages list is created at object initialization
time, so there should be no need to handle an unpopulated page list.
Add a sanity check nevertheless.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190227144441.6755-4-kraxel@redhat.com

virtio-gpu objects never move around, so effectively they are
pinned all the time.  Therefore we don't need the (optional)
pin/unpin callbacks.  Remove them.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NNoralf Trønnes <noralf@tronnes.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20190227144441.6755-3-kraxel@redhat.com

Having the probe helper stuff (which pretty much everyone needs) in
the drm_crtc_helper.h file (which atomic drivers should never need) is
confusing. Split them out.

To make sure I actually achieved the goal here I went through all
drivers. And indeed, all atomic drivers are now free of
drm_crtc_helper.h includes.

v2: Make it compile. There was so much compile fail on arm drivers
that I figured I'll better not include any of the acks on v1.

v3: Massive rebase because i915 has lost a lot of drmP.h includes, but
not all: Through drm_crtc_helper.h > drm_modeset_helper.h -> drmP.h
there was still one, which this patch largely removes. Which means
rolling out lots more includes all over.

This will also conflict with ongoing drmP.h cleanup by others I
expect.

v3: Rebase on top of atomic bochs.

v4: Review from Laurent for bridge/rcar/omap/shmob/core bits:
- (re)move some of the added includes, use the better include files in
  other places (all suggested from Laurent adopted unchanged).
- sort alphabetically

v5: Actually try to sort them, and while at it, sort all the ones I
touch.

v6: Rebase onto i915 changes.

v7: Rebase once more.
Acked-by: NHarry Wentland <harry.wentland@amd.com>
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Acked-by: NJani Nikula <jani.nikula@intel.com>
Acked-by: NNeil Armstrong <narmstrong@baylibre.com>
Acked-by: NOleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
Acked-by: NCK Hu <ck.hu@mediatek.com>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Reviewed-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: NLiviu Dudau <liviu.dudau@arm.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: virtualization@lists.linux-foundation.org
Cc: etnaviv@lists.freedesktop.org
Cc: linux-samsung-soc@vger.kernel.org
Cc: intel-gfx@lists.freedesktop.org
Cc: linux-mediatek@lists.infradead.org
Cc: linux-amlogic@lists.infradead.org
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: spice-devel@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-renesas-soc@vger.kernel.org
Cc: linux-rockchip@lists.infradead.org
Cc: linux-stm32@st-md-mailman.stormreply.com
Cc: linux-tegra@vger.kernel.org
Cc: xen-devel@lists.xen.org
Link: https://patchwork.freedesktop.org/patch/msgid/20190117210334.13234-1-daniel.vetter@ffwll.ch

Also set prime_handle_to_fd and prime_fd_to_handle to NULL,
so drm will not advertive DRM_PRIME_CAP_{IMPORT,EXPORT} to
userspace.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NOleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190110111545.26768-1-kraxel@redhat.com

Move the code around so the driver is probed the bus
.probe and removed from the bus .remove callbacks.
This commit is just a cleanup and shouldn't affect
functionality.
Signed-off-by: NEzequiel Garcia <ezequiel@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20190108145930.15080-1-ezequiel@collabora.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Just call drm_fence_put directly instead.
Also set vgfb->fence to NULL after dropping the reference.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NEzequiel Garcia <ezequiel@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181219122708.4586-4-kraxel@redhat.com

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181213134915.24722-1-kraxel@redhat.com

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NRobert Foss <robert.foss@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181128151021.29565-3-kraxel@redhat.com

Since commit "9fdd90c0 drm/virtio: add virtio_gpu_alloc_fence()"
fences are not allocated any more by virtio_gpu_fence_emit().  So there
is no need to pass down a reference to the fence pointer, a plain
pointer is enough now.

Convert virtio_gpu_fence_emit() and callers.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NRobert Foss <robert.foss@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181128151021.29565-2-kraxel@redhat.com

linux guest driver implementation of the VIRTIO_GPU_F_EDID feature.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Acked-by: NDaniel Vetter <daniel@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20181030063206.19528-3-kraxel@redhat.com

To reflect the (backward compatible) changes in the uabi we are bumping
the driver's version.
Signed-off-by: NGustavo Padovan <gustavo.padovan@collabora.co>
Signed-off-by: NRobert Foss <robert.foss@collabora.com>
Reviewed-by: NEmil Velikov <emil.velikov@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181112165157.32765-5-robert.foss@collabora.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Refactor fence creation, add fences to relevant GPU
operations and add cursor helper functions.

This removes the potential for allocation failures from the
cmd_submit and atomic_commit paths.
Now a fence will be allocated first and only after that
will we proceed with the rest of the execution.
Signed-off-by: NGustavo Padovan <gustavo.padovan@collabora.com>
Signed-off-by: NRobert Foss <robert.foss@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181112165157.32765-2-robert.foss@collabora.comSuggested-by: NRob Herring <robh@kernel.org>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Make sure that the global BO state is always correctly initialized.

This allows removing all the device code to initialize it.

v2: fix up vbox (Alex)
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

As the name says we only need one global instance of ttm_mem_global.

Drop all the driver initialization and just use a single exported
instance which is initialized during BO global initialization.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NJunwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

These IDRs were only being used to allocate unique numbers, not to look
up pointers, so they can use the more space-efficient IDA instead.
Signed-off-by: NMatthew Wilcox <willy@infradead.org>
Link: http://patchwork.freedesktop.org/patch/msgid/20180926160031.15721-2-willy@infradead.org

[ kraxel: resolve conflict ]
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Move virtio_gpu_resource_id_{get,put} to virtgpu_object.c and make them
static.  Allocate and free the id on creation and destroy, drop all
other calls.  That way objects have a valid handle for the whole
lifetime of the object.

Also fixes ids leaking.  Worst offender are dumb buffers, and I think
some error paths too.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181019061847.18958-7-kraxel@redhat.com

We pass the obj anyway, so obj->hw_res_handle can be used instead
in virtio_gpu_object_attach() and virtio_gpu_cmd_create_resource().
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181019061847.18958-6-kraxel@redhat.com

Track whenever the virtio_gpu_object is already created (i.e. host knows
about it) in a new variable.  Add checks to virtio_gpu_object_attach()
to do nothing on objects not created yet.

Make virtio_gpu_ttm_bo_destroy() use the new variable too, instead of
expecting hw_res_handle indicating the object state.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20181019061847.18958-2-kraxel@redhat.com

Currently, virtio_gpu_object_kmap() is only called by
virtio_gpufb_create(), when a DRM framebuffer is created.

Thus, instead of returning the vmap'ed address, emit a warning
if virtio_gpu_object_kmap is called on an already mapped
object. With this change, kmap/kunmap calls are now balanced.
Signed-off-by: NEzequiel Garcia <ezequiel@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180925161606.17980-3-ezequiel@collabora.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Implement a virtio_gpu_object_kunmap() to unmap the kernel
mapping, and use it in the TTM object destroy path.
Signed-off-by: NEzequiel Garcia <ezequiel@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180925161606.17980-2-ezequiel@collabora.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Pass virtio_gpu_object down to virtio_gpu_cmd_transfer_to_host_2d and
virtio_gpu_cmd_transfer_to_host_3d functions, instead of passing just
the virtio resource handle.

This is needed to lookup the scatter list of the object, for dma sync.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NJiandi An <jiandi.an@amd.com>
Tested-by: NJiandi An <jiandi.an@amd.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180920062924.6514-1-kraxel@redhat.com

With virtio gpu ttm-pages being dma mapped, dma sync is needed when
swiotlb is used as bounce buffers, before TRANSFER_TO_HOST_2D/3D
commands are sent.
Signed-off-by: NJiandi An <jiandi.an@amd.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180919070931.91168-1-jiandi.an@amd.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Use the dma mapping api and properly add iommu mappings for
objects, unless virtio is in iommu quirk mode.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180829122026.27012-3-kraxel@redhat.com

The new function balances virtio_gpu_object_attach().

Also make virtio_gpu_cmd_resource_inval_backing() static and switch
call sites to the new virtio_gpu_object_attach() function.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Reviewed-by: NDave Airlie <airlied@redhat.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20180829122026.27012-2-kraxel@redhat.com