linux-toradex.git/drivers/gpu/drm/vc4/vc4_gem.c, branch v4.16-rc6

drm/vc4: Fix NULL pointer dereference in vc4_save_hang_state()

2018-01-18T20:17:03+00:00

When saving BOs in the hang state we skip one entry of the
kernel_state->bo[] array, thus leaving it to NULL. This leads to a NULL
pointer dereference when, later in this function, we iterate over all
BOs to check their ->madv state.

Fixes: ca26d28bbaa3 ("drm/vc4: improve throughput by pipelining binning and rendering jobs")
Cc: 
Signed-off-by: Boris Brezillon 
Signed-off-by: Eric Anholt 
Reviewed-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20180118145821.22344-1-boris.brezillon@free-electrons.com

drm/vc4: Flush the caches before the bin jobs, as well.

2018-01-18T20:17:03+00:00

If the frame samples from a render target that was just written, its
cache flush during the binning step may have occurred before the
previous frame's RCL was completed.  Flush the texture caches again
before starting each RCL job to make sure that the sampling of the
previous RCL's output is correct.

Fixes flickering in the top left of 3DMMES Taiji.

Signed-off-by: Eric Anholt 
Fixes: ca26d28bbaa3 ("drm/vc4: improve throughput by pipelining binning and rendering jobs")
Link: https://patchwork.freedesktop.org/patch/msgid/20171221221722.23809-1-eric@anholt.net
Reviewed-by: Boris Brezillon

drm/vc4: Release fence after signalling

2017-12-08T21:02:22+00:00

We were never releasing the initial fence reference that is obtained
through dma_fence_init.

Link: https://github.com/anholt/linux/issues/122
Fixes: cdec4d361323 ("drm/vc4: Expose dma-buf fences for V3D rendering.")
Signed-off-by: Stefan Schake 
Signed-off-by: Eric Anholt 
Reviewed-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/1512236444-301-1-git-send-email-stschake@gmail.com

drm/vc4: Convert timers to use timer_setup()

2017-11-21T23:46:44+00:00

In preparation for unconditionally passing the struct timer_list pointer to
all timer callbacks, switch to using the new timer_setup() and from_timer()
to pass the timer pointer explicitly.

Cc: Eric Anholt 
Cc: David Airlie 
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Kees Cook 
Signed-off-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20171024151648.GA104538@beast
Reviewed-by: Eric Anholt

drm/vc4: Add the DRM_IOCTL_VC4_GEM_MADVISE ioctl

2017-10-19T17:34:49+00:00

This ioctl will allow us to purge inactive userspace buffers when the
system is running out of contiguous memory.

For now, the purge logic is rather dumb in that it does not try to
release only the amount of BO needed to meet the last CMA alloc request
but instead purges all objects placed in the purgeable pool as soon as
we experience a CMA allocation failure.

Note that the in-kernel BO cache is always purged before the purgeable
cache because those objects are known to be unused while objects marked
as purgeable by a userspace application/library might have to be
restored when they are marked back as unpurgeable, which can be
expensive.

Signed-off-by: Boris Brezillon 
Signed-off-by: Eric Anholt 
Reviewed-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20171019125748.3152-1-boris.brezillon@free-electrons.com

drm/vc4: Add exec flags to allow forcing a specific X/Y tile walk order.

2017-08-08T20:26:44+00:00

This is useful to allow GL to provide defined results for overlapping
glBlitFramebuffer, which X11 in turn uses to accelerate uncomposited
window movement without first blitting to a temporary.  x11perf
-copywinwin100 goes from 1850/sec to 4850/sec.

v2: Default to the same behavior as before when the flags aren't
    passed. (suggested by Boris)

Signed-off-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20170725162733.28007-2-eric@anholt.net
Reviewed-by: Boris Brezillon

drm/vc4: Demote user-accessible DRM_ERROR paths to DRM_DEBUG.

2017-08-08T20:20:11+00:00

Userspace shouldn't be able to spam dmesg by passing bad arguments.
This has particularly become an issues since we started using a bad
argument to set_tiling to detect if set_tiling was supported.

Signed-off-by: Eric Anholt 
Fixes: 83753117f1de ("drm/vc4: Add get/set tiling ioctls.")
Link: https://patchwork.freedesktop.org/patch/msgid/20170725162733.28007-1-eric@anholt.net
Reviewed-by: Boris Brezillon

drm/vc4: switch to drm_*{get,put} helpers

2017-08-08T20:20:11+00:00

drm_*_reference() and drm_*_unreference() functions are just
compatibility alias for drm_*_get() and drm_*_put() adn should not be
used by new code. So convert all users of compatibility functions to use
the new APIs.

Signed-off-by: Cihangir Akturk 
Signed-off-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/1501761585-11757-26-git-send-email-cakturk@gmail.com
Reviewed-by: Eric Anholt

drm/vc4: Fix double destroy of the BO cache on teardown.

2017-08-08T20:20:10+00:00

It's also destroyed from the top level vc4_drv.c initialization, which
is where the cache was actually initialized from.

This used to just involve duplicate del_timer() and cancel_work_sync()
being called, but it started causing kmalloc issues once we
double-freed the new BO label array.

Fixes: 1908a876f909 ("drm/vc4: Add an ioctl for labeling GEM BOs for summary stats")
Signed-off-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20170802203242.12815-1-eric@anholt.net
Tested-by: Noralf Trønnes 
Acked-by: Noralf Trønnes 
Reviewed-by: Boris Brezillon

drm/vc4: Add an ioctl for labeling GEM BOs for summary stats

2017-07-28T23:04:53+00:00

This has proven immensely useful for debugging memory leaks and
overallocation (which is a rather serious concern on the platform,
given that we typically run at about 256MB of CMA out of up to 1GB
total memory, with framebuffers that are about 8MB ecah).

The state of the art without this is to dump debug logs from every GL
application, guess as to kernel allocations based on bo_stats, and try
to merge that all together into a global picture of memory allocation
state.  With this, you can add a couple of calls to the debug build of
the 3D driver and get a pretty detailed view of GPU memory usage from
/debug/dri/0/bo_stats (or when we debug print to dmesg on allocation
failure).

The Mesa side currently labels at the gallium resource level (so you
see that a 1920x20 pixmap has been created, presumably for the window
system panel), but we could extend that to be even more useful with
glObjectLabel() names being sent all the way down to the kernel.

(partial) example of sorted debugfs output with Mesa labeling all
resources:

               kernel BO cache:  16392kb BOs (3)
       tiling shadow 1920x1080:   8160kb BOs (1)
       resource 1920x1080@32/0:   8160kb BOs (1)
scanout resource 1920x1080@32/0:   8100kb BOs (1)
                        kernel:   8100kb BOs (1)

v2: Use strndup_user(), use lockdep assertion instead of just a
    comment, fix an array[-1] reference, extend comment about name
    freeing.

Signed-off-by: Eric Anholt 
Link: https://patchwork.freedesktop.org/patch/msgid/20170725182718.31468-2-eric@anholt.net
Reviewed-by: Chris Wilson