drm/msm: shrinker support

For a first step, only purge obj->madv==DONTNEED objects. We could be
more agressive and next try unpinning inactive objects.. but that is
only useful if you have swap.

drm/msm: shrinker support

For a first step, only purge obj->madv==DONTNEED objects. We could be
more agressive and next try unpinning inactive objects.. but that is
only useful if you have swap.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: add put_iova() helper

We'll need this too for shrinker/purging.

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/msm: add madvise ioctl

Doesn't do anything too interesting until we wire up shrinker. Pretty
much lifted from i915.

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/msm: deal with exhausted vmap space better

Some, but not all, callers of obj->vmap() would check if return
IS_ERR(). So let's actually return an error if vmap() fails. And fixup
the call-sites

drm/msm: deal with exhausted vmap space better

Some, but not all, callers of obj->vmap() would check if return
IS_ERR(). So let's actually return an error if vmap() fails. And fixup
the call-sites that were not handling this properly.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm: Remove unused drm_device from drm_gem_object_lookup()

drm_gem_object_lookup() has never required the drm_device for its file
local translation of the user handle to the GEM object. Let's remove

drm: Remove unused drm_device from drm_gem_object_lookup()

drm_gem_object_lookup() has never required the drm_device for its file
local translation of the user handle to the GEM object. Let's remove the
unused parameter and save some space.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: dri-devel@lists.freedesktop.org
Cc: Dave Airlie <airlied@redhat.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
[danvet: Fixup kerneldoc too.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

show more ...

drm/msm: 'struct fence' conversion

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/msm: remove fence_cbs

This was only used for atomic commit these days. So instead just give
atomic it's own work-queue where we can do a block on each bo in turn.
Simplifies things a whole bunc

drm/msm: remove fence_cbs

This was only used for atomic commit these days. So instead just give
atomic it's own work-queue where we can do a block on each bo in turn.
Simplifies things a whole bunch and makes the 'struct fence' conversion
easier.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: introduce msm_fence_context

Better encapsulate the per-timeline stuff into fence-context. For now
there is just a single fence-context, but eventually we'll also have one
per-CRTC to enabl

drm/msm: introduce msm_fence_context

Better encapsulate the per-timeline stuff into fence-context. For now
there is just a single fence-context, but eventually we'll also have one
per-CRTC to enable fully explicit fencing.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: use imported dmabuf's reservation object

This was always the intention, but somehow it was never wired up
properly.

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/msm: move fence code to it's own file

Signed-off-by: Rob Clark <robdclark@gmail.com>

mm, dax, gpu: convert vm_insert_mixed to pfn_t

Convert the raw unsigned long 'pfn' argument to pfn_t for the purpose of
evaluating the PFN_MAP and PFN_DEV flags. When both are set it triggers
_PAGE

mm, dax, gpu: convert vm_insert_mixed to pfn_t

Convert the raw unsigned long 'pfn' argument to pfn_t for the purpose of
evaluating the PFN_MAP and PFN_DEV flags. When both are set it triggers
_PAGE_DEVMAP to be set in the resulting pte.

There are no functional changes to the gpu drivers as a result of this
conversion.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave@sr71.net>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

show more ...

drm/msm: change to uninterruptible wait in atomic commit

The atomic commit cannot easily undo and return an error once the
state is swapped. Change to uninterruptible wait, and ignore the
timeout er

drm/msm: change to uninterruptible wait in atomic commit

The atomic commit cannot easily undo and return an error once the
state is swapped. Change to uninterruptible wait, and ignore the
timeout error.

Signed-off-by: Wentao Xu <wentaox@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: change to uninterruptible wait in atomic commit

The atomic commit cannot easily undo and return an error once the
state is swapped. Change to uninterruptible wait, and ignore the
timeout er

drm/msm: change to uninterruptible wait in atomic commit

The atomic commit cannot easily undo and return an error once the
state is swapped. Change to uninterruptible wait, and ignore the
timeout error.

Signed-off-by: Wentao Xu <wentaox@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: fix timeout calculation

The 'timeout' value comes from userspace (CLOCK_MONOTONIC), but
converting this directly to jiffies doesn't take into account the
initial jiffies count at boot, whic

drm/msm: fix timeout calculation

The 'timeout' value comes from userspace (CLOCK_MONOTONIC), but
converting this directly to jiffies doesn't take into account the
initial jiffies count at boot, which may differ from the base time
of CLOCK_MONOTONIC.

TODO: add ktime_delta_jiffies() when rebasing on 4.1 and use that
instead of ktime_sub/ktime_to_timespec/timespec_to_jiffies combo (as
suggested by Arnd)

v2: switch over from 'struct timespec' to ktime_t throughout, since
'struct timespec' will be deprecated (as suggested by Arnd)
v3: minor cosmetic tweaks

Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: Call drm_prime_gem_destroy to clean up imported GEM object

If the GEM object is imported, drm_prime_gem_destroy needs to be
called to clean up dma buffer related information.

Signed-off-by

drm/msm: Call drm_prime_gem_destroy to clean up imported GEM object

If the GEM object is imported, drm_prime_gem_destroy needs to be
called to clean up dma buffer related information.

Signed-off-by: Jilai Wang <jilaiw@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: Fix a couple of 64-bit build warnings

Avoid casts from pointers to fixed-size integers to prevent the compiler
from warning. Print virtual memory addresses using %p instead. Also turn
a cou

drm/msm: Fix a couple of 64-bit build warnings

Avoid casts from pointers to fixed-size integers to prevent the compiler
from warning. Print virtual memory addresses using %p instead. Also turn
a couple of %d/%x specifiers into %zu/%zd/%zx to avoid further warnings
due to mismatched format strings.

Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: add support for "stolen" mem

Add support to use the VRAM carveout (if specified in dtb) for fbdev
scanout buffer. This allows drm/msm to take over a bootloader splash-
screen, and avoids c

drm/msm: add support for "stolen" mem

Add support to use the VRAM carveout (if specified in dtb) for fbdev
scanout buffer. This allows drm/msm to take over a bootloader splash-
screen, and avoids corruption on screen that results if the kernel uses
memory that is still being scanned out for itself.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: Deletion of unnecessary checks before two function calls

The functions framebuffer_release() and vunmap() perform also input
parameter validation. Thus the test around the call is not neede

drm/msm: Deletion of unnecessary checks before two function calls

The functions framebuffer_release() and vunmap() perform also input
parameter validation. Thus the test around the call is not needed.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: small mmap offset cleanups

Use pre-computed iova when unmapping, to reduce the places we assume iova
and mmap offset are (at the moment) the same. And get rid of an extra
drm_gem_free_mmap

drm/msm: small mmap offset cleanups

Use pre-computed iova when unmapping, to reduce the places we assume iova
and mmap offset are (at the moment) the same. And get rid of an extra
drm_gem_free_mmap_offset() call (since it is already called from
drm_gem_object_release())

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: fb prepare/cleanup

Atomic wants to split the prepare/pin from where we actually program the
scanout address (so that any part that can fail is done synchronously).
Add some fb/gem apis to m

drm/msm: fb prepare/cleanup

Atomic wants to split the prepare/pin from where we actually program the
scanout address (so that any part that can fail is done synchronously).
Add some fb/gem apis to make this easier to use from the kms parts.

Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: small fence cleanup

Give ourselves a way to wait for certain fence #.. makes it easier to
wait on a set of bo's, which we'll need for atomic.

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/msm: no mmu is only error if not using vram carveout

Signed-off-by: Rob Clark <robdclark@gmail.com>

drm/gem: remove misleading gfp parameter to get_pages()

drm_gem_get_pages() currently allows passing a 'gfp' parameter that is
passed to shmem combined with mapping_gfp_mask(). Given that the defaul

drm/gem: remove misleading gfp parameter to get_pages()

drm_gem_get_pages() currently allows passing a 'gfp' parameter that is
passed to shmem combined with mapping_gfp_mask(). Given that the default
mapping_gfp_mask() is GFP_HIGHUSER, it is _very_ unlikely that anyone will
ever make use of that parameter. In fact, all drivers currently pass
redundant flags or 0.

This patch removes the 'gfp' parameter. The only reason to keep it is to
remove flags like __GFP_WAIT. But in its current form, it can only be used
to add flags. So to remove __GFP_WAIT, you'd have to drop it from the
mapping_gfp_mask, which again is stupid as this mask is used by shmem-core
for other allocations, too.

If any driver ever requires that parameter, we can introduce a new helper
that takes the raw 'gfp' parameter. The caller'd be responsible to combine
it with mapping_gfp_mask() in a suitable way. The current
drm_gem_get_pages() helper would then simply use mapping_gfp_mask() and
call the new helper. This is what shmem_read_mapping_pages{_gfp,} does
right now.

Moreover, the gfp-zone flag-usage is not obvious: If you pass a modified
zone, shmem core will WARN() or even BUG(). In other words, the following
must be true for 'gfp' passed to shmem_read_mapping_pages_gfp():
gfp_zone(mapping_gfp_mask(mapping)) == gfp_zone(gfp)
Add a comment to drm_gem_read_pages() explaining that constraint.

Signed-off-by: David Herrmann <dh.herrmann@gmail.com>

show more ...

drm/msm: fix IOMMU cleanup for -EPROBE_DEFER

If probe fails after IOMMU is attached, we need to detach in order to
clean up properly. Before this change, IOMMU faults would occur if the
probe faile

drm/msm: fix IOMMU cleanup for -EPROBE_DEFER

If probe fails after IOMMU is attached, we need to detach in order to
clean up properly. Before this change, IOMMU faults would occur if the
probe failed (-EPROBE_DEFER).

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

show more ...

drm/msm: fix memory leak

Signed-off-by: Micah Richert <richert@braincorporation.com>
Signed-off-by: Rob Clark <robdclark@gmail.com>