We'll need to access the crtc_state from outside of vc4_crtc.c, so let's
move it to vc4_drv.h
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/1e6e563f9c75961e2885c9d648a3130d3b46b6d1.1590594512.git-series.maxime@cerno.tech

So far the plane creation was done when each CRTC was bound, and those
planes were only tied to the CRTC that was registering them.

This causes two main issues:
  - The planes in the vc4 hardware are actually not tied to any CRTC, but
    can be used with every combination

  - More importantly, so far, we allocate 10 planes per CRTC, with 3 CRTCs.
    However, the next generation of hardware will have 5 CRTCs, putting us
    well above the maximum of 32 planes currently allowed by DRM.

This patch is the first one in a series of patches that will take down both
of these issues so that we can support the next generation of hardware
while keeping a good amount of planes.

We start by changing the way the planes are registered to first registering
the primary planes for each CRTC in the CRTC bind function as we used to,
but moving the overlay and cursor creation to the main driver bind
function, after all the CRTCs have been bound, and make the planes
associated to all CRTCs.

This will slightly change the ID order of the planes, since the primary
planes of all CRTCs will be first, and then a pattern of 8 overlays, 1
cursor plane for each CRTC.

This shouldn't cause any trouble since the ordering between the planes is
preserved though.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/0b85a3fdb20bb4ff85fb62cabd082d5a65e2730b.1590594512.git-series.maxime@cerno.tech

The planes so far were created as part of the CRTC binding code with
each planes created associated only to one CRTC. However, the hardware
in the vc4 doesn't really have such constraint and can be used with any
CRTC.

In order to rework this, let's first move the overlay and cursor planes
creation to a function of its own.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/a378ea56214179f1f25fcd36ecc69511edd1e790.1590594512.git-series.maxime@cerno.tech

vc4_drv.h doesn't have any include guards which prevents it from being
included twice. Let's add them.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/68e89e315c4c35b313efc277c9642eca684e0ade.1590594512.git-series.maxime@cerno.tech

As a result of commit 987d65d0 (drm: debugfs: make
drm_debugfs_create_files() never fail) and changes to various debugfs
functions in drm/core and across various drivers, there is no need for
the drm_driver.debugfs_init() hook to have a return value. Therefore,
declare it as void.

This also includes refactoring all users of the .debugfs_init() hook to
return void across the subsystem.

v2: include changes to the hook and drivers that use it in one patch to
prevent driver breakage and enable individual successful compilation of
this change.

References: https://lists.freedesktop.org/archives/dri-devel/2020-February/257183.htmlSigned-off-by: NWambui Karuga <wambui.karugax@gmail.com>
Reviewed-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20200310133121.27913-18-wambui.karugax@gmail.com

The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:

struct foo {
        int stuff;
        struct boo array[];
};

By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.

Also, notice that, dynamic memory allocations won't be affected by
this change:

"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]

This issue was found with the help of Coccinelle.

[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20200305105707.GA19261@embeddedor

The wait_for macro's for Broadcom VC4 driver used msleep, which is
inappropriate due to its inaccuracy at low values (minimum wait time
is about 30ms on the Raspberry Pi).  This sleep was triggering in
v3d_clean_caches(), causing us to only be able to dispatch ~33 compute
jobs per second.

This patch replaces the macro with the one from the Intel i915 version
which uses usleep_range to provide more accurate waits.

v2: Split from the v3d patch in case this tickles modesetting bugs (by
    anholt)
Signed-off-by: NJames Hughes <james.hughes@raspberrypi.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20200217153145.13780-1-james.hughes@raspberrypi.com

The callback struct drm_driver.get_scanout_position() is deprecated in
favor of struct drm_crtc_helper_funcs.get_scanout_position(). Convert vc4
over.
Signed-off-by: NThomas Zimmermann <tzimmermann@suse.de>
Acked-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20200123135943.24140-18-tzimmermann@suse.de

Drop use of the deprecated header drmP.h.
Fix so vc4_drv.h is now self-contained, and fixed fall-out in remaining
files.

Divided include files in blocks.
Sorted include files within their blocks.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Acked-by: NEmil Velikov <emil.velikov@collabora.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190716064220.18157-7-sam@ravnborg.org

The idea is that gem_prime_export is deprecated in favor of
obj_funcs.export. That's much easier to do if both have matching
function signatures.
Reviewed-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEmil Velikov <emil.velikov@collabora.com>
Acked-by: NChristian König <christian.koenig@amd.com>
Acked-by: NThierry Reding <treding@nvidia.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <maxime.ripard@bootlin.com>
Cc: Sean Paul <sean@poorly.run>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Zhi Wang <zhi.a.wang@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Eric Anholt <eric@anholt.net>
Cc: "Michel Dänzer" <michel.daenzer@amd.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Huang Rui <ray.huang@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Cc: Feifei Xu <Feifei.Xu@amd.com>
Cc: Jim Qu <Jim.Qu@amd.com>
Cc: Evan Quan <evan.quan@amd.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Thomas Zimmermann <tdz@users.sourceforge.net>
Cc: Kate Stewart <kstewart@linuxfoundation.org>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: Jilayne Lovejoy <opensource@jilayne.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Junwei Zhang <Jerry.Zhang@amd.com>
Cc: intel-gvt-dev@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Link: https://patchwork.freedesktop.org/patch/msgid/20190614203615.12639-10-daniel.vetter@ffwll.ch

Based on 2 normalized pattern(s):

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license version 2 as
  published by the free software foundation

  this program is free software you can redistribute it and or modify
  it under the terms of the gnu general public license version 2 as
  published by the free software foundation #

extracted by the scancode license scanner the SPDX license identifier

  GPL-2.0-only

has been chosen to replace the boilerplate/reference in 4122 file(s).
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NEnrico Weigelt <info@metux.net>
Reviewed-by: NKate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: NAllison Randal <allison@lohutok.net>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190604081206.933168790@linutronix.deSigned-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The binner BO is not required until the V3D is in use, so avoid
allocating it at probe and do it on the first non-dumb BO allocation.

Keep track of which clients are using the V3D and liberate the buffer
when there is none left, using a kref. Protect the logic with a
mutex to avoid race conditions.

The binner BO is created at the time of the first render ioctl and is
destroyed when there is no client and no exec job using it left.

The Out-Of-Memory (OOM) interrupt also gets some tweaking, to avoid
enabling it before having allocated a binner bo.

We also want to keep the BO alive during runtime suspend/resume to avoid
failing to allocate it at resume. This happens when the CMA pool is
full at that point and results in a hard crash.
Signed-off-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190516145544.29051-5-paul.kocialkowski@bootlin.com

One might want to use the VC4 display stack without using Mesa.
Similar to the debugfs fixes for not having all of the possible
display bits enabled, make sure you can't oops in vc4 if v3d isn't
enabled.

v2: Fix matching against other v3d variants (review by Paul), don't
    forget to set irq_enabled so that the vblank uapi works
v3: Use -ENODEV instead of -EINVAL on Paul's suggestion.
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190401183559.3823-2-eric@anholt.netReviewed-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>

The global list of all debugfs entries for the driver was painful: the
list couldn't see into the components' structs, so each component had
its own debugs show function to find the component, then find the
regset and dump it.  The components also had to be careful to check
that they were actually registered in vc4 before dereferencing
themselves, in case they weren't probed on a particular platform.
They routinely failed at that.

Instead, we can have the components add their debugfs callbacks to a
little list in vc4 to be registered at drm_dev_register() time, which
gets vc4_debugfs.c out of the business of knowing the whole list of
components.

Thanks to this change, dsi0 (if it existed) would register its node.

v2: Rebase on hvs_underrun addition.
v3: whitespace fixup
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190401183559.3823-1-eric@anholt.netReviewed-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>

This makes sure the vc4_reset doesn't hit an obscure race with the
GET_PARAM ioctl, fixes a decrement outside of the lock, and prevents
future code from making mistakes with the weird return value of
pm_runtime_get_sync().
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190220210343.28157-6-eric@anholt.netReviewed-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>

This removes a bunch of duplicated boilerplate for the debugfs vs
runtime printk debug dumping.
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20190220210343.28157-2-eric@anholt.netReviewed-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>

In order to test whether the load tracker is working as expected, we
need the ability to compare the commit result with the underrun
indication. With the load tracker always enabled, commits that are
expected to trigger an underrun are always rejected, so userspace
cannot get the actual underrun indication from the hardware.

Add a debugfs entry to disable/enable the load tracker, so that a DRM
commit expected to trigger an underrun can go through with the load
tracker disabled. The underrun indication is then available to
userspace and can be checked against the commit result with the load
tracker enabled.
Signed-off-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime.ripard@bootlin.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190220155124.25022-4-paul.kocialkowski@bootlin.com

The HVS block is supposed to fill the pixelvalve FIFOs fast enough to
meet the requested framerate. The problem is, the HVS and memory bus
bandwidths are limited, and if we don't take these limitations into
account we might end up with HVS underflow errors.

This patch is trying to model the per-plane HVS and memory bus bandwidth
consumption and take a decision at atomic_check() time whether the
estimated load will fit in the HVS and membus budget.

Note that we take an extra margin on the memory bus consumption to let
the system run smoothly when other blocks are doing heavy use of the
memory bus. Same goes for the HVS limit, except the margin is smaller in
this case, since the HVS is not used by external components.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime.ripard@bootlin.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190220155124.25022-3-paul.kocialkowski@bootlin.com

Add a debugfs entry and helper for reporting HVS underrun errors as
well as helpers for masking and unmasking the underrun interrupts.
Add an IRQ handler and initial IRQ configuration.
Rework related register definitions to take the channel number.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Signed-off-by: NPaul Kocialkowski <paul.kocialkowski@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NMaxime Ripard <maxime.ripard@bootlin.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190220155124.25022-2-paul.kocialkowski@bootlin.com

Now that the base struct drm_gem_object has a reservation_object, use it
and remove the private BO one.

Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: NEric Anholt <eric@anholt.net>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: NRob Herring <robh@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20190202154158.10443-6-robh@kernel.orgSigned-off-by: NMaxime Ripard <maxime.ripard@bootlin.com>

Move drm_can_sleep() out of drmP.h to allow users
to get rid of the drmP.h include.

There was no header file that was a good match for this helper function.
So add this to drm_util with the relevant includes.

Add include of drm_util.h to all users.

v2:
- Update comments to use kernel-doc style (Daniel)
- Add FIXME to drm_can_sleep and add note that this
  function should not be used in new code (Daniel)

v3:
- Fix kernel-doc syntax (Daniel)
- Plug drm_util.h into drm-internels.rst (Daniel)
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <maxime.ripard@bootlin.com>
Cc: Sean Paul <sean@poorly.run>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: "David (ChunMing) Zhou" <David1.Zhou@amd.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: Eric Anholt <eric@anholt.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20190112193251.20450-2-sam@ravnborg.org

Applyin margins is just a matter of scaling all planes appropriately
and adjusting the CRTC X/Y offset to account for the
left/right/top/bottom borders.

Create a vc4_plane_margins_adj() function doing that and call it from
vc4_plane_setup_clipping_and_scaling() so that we are ready to attach
margins properties to the HDMI connector.
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20181206142439.10441-5-boris.brezillon@bootlin.com

We are about to use vc4_plane_mode_set() in the async check path, but
async check can decide that async update is not possible and force the
driver to fallback to a sync update.

All the checks that have been done on the plane state during async check
stay valid, and checking it again is not necessary. Add a ->checked
field to vc4_plane_state, and use it to track the status of the state
(checked or not).
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20181130090254.594-3-boris.brezillon@bootlin.com

We are about to use vc4_plane_mode_set() in the async check path, and
async updates require that LBM size stay the same since they reuse the
LBM from the previous state. So we definitely don't want to allocate a
new LBM region that we know for sure will be free right away.

Move the LBM allocation out of vc4_plane_mode_set() and call the new
function (vc4_plane_update_lbm()) from vc4_plane_atomic_check().
Signed-off-by: NBoris Brezillon <boris.brezillon@bootlin.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20181130090254.594-2-boris.brezillon@bootlin.com

The transposer block is providing support for mem-to-mem composition,
which is exposed as a drm_writeback connector in DRM.

Add a driver to support this feature.
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20180703075022.15138-9-boris.brezillon@bootlin.com

Use new return type vm_fault_t for fault handler.
Signed-off-by: NSouptick Joarder <jrdr.linux@gmail.com>
Reviewed-by: NMatthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>

Allow userland to specify a syncobj that is waited on before a render job
starts processing.

v2: Use 0 as invalid syncobj to drop flag (Eric)
    Drop extra newline (Eric)
Signed-off-by: NStefan Schake <stschake@gmail.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/1524607427-12876-2-git-send-email-stschake@gmail.com

The hardware has a single block for applying a CTM prior to gamma lut.
It can be fed with pixels from one of our CRTC at a time and uses a
matrix with S0.9 scalars. Use private atomic state to reject attempts
from userland to apply CTM for more than one CRTC at a time and reject
matrices with scalars that we can't approximate without integer bits.
Signed-off-by: NStefan Schake <stschake@gmail.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/218067/

We need to access the channel for configuring our CTM hardware.
Signed-off-by: NStefan Schake <stschake@gmail.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/1523479755-20812-4-git-send-email-stschake@gmail.com

We need to reference it from the CRTC to make a decision for enabling
background color fill.
Signed-off-by: NStefan Schake <stschake@gmail.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/1520556817-97297-4-git-send-email-stschake@gmail.com

The V3D engine has various hardware counters which might be interesting
to userspace performance analysis tools.

Expose new ioctls to create/destroy a performance monitor object and
query the counter values of this perfmance monitor.

Note that a perfomance monitor is given an ID that is only valid on the
file descriptor it has been allocated from. A performance monitor can be
attached to a CL submission and the driver will enable HW counters for
this request and update the performance monitor values at the end of the
job.
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20180112090926.12538-1-boris.brezillon@free-electrons.com

Use drm_fb_cma_fbdev_init() and drm_fb_cma_fbdev_fini() which relies on
the fact that drm_device holds a pointer to the drm_fb_helper structure.
This means that the driver doesn't have to keep track of that.
Also use the drm_fb_helper functions directly.

Cc: Eric Anholt <eric@anholt.net>
Signed-off-by: NNoralf Trønnes <noralf@tronnes.org>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20171115142001.45358-18-noralf@tronnes.org

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: NBoris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20171019125748.3152-1-boris.brezillon@free-electrons.com

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: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20170725182718.31468-2-eric@anholt.netReviewed-by: NChris Wilson <chris@chris-wilson.co.uk>

It is no longer used as of commit 34c8ea40 ("drm/vc4: Mimic
drm_atomic_helper_commit() behavior")
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: http://patchwork.freedesktop.org/patch/msgid/20170621185002.28563-4-eric@anholt.netAcked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NBoris Brezillon <boris.brezillon@free-electrons.com>

This allows mesa to set the tiling format for a BO and have that
tiling format be respected by mesa on the other side of an
import/export (and by vc4 scanout in the kernel), without defining a
protocol to pass the tiling through userspace.
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: http://patchwork.freedesktop.org/patch/msgid/20170608001336.12842-2-eric@anholt.netAcked-by: NDave Airlie <airlied@redhat.com>

Fixes a copy&paste error.
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
Reviewed-by: NEric Anholt <eric@anholt.net>
Link: http://patchwork.freedesktop.org/patch/msgid/1495550187-525-1-git-send-email-boris.brezillon@free-electrons.com

Include <drm/*.h> instead of relative path from include/drm, then
remove the -Iinclude/drm compiler flag.

While we are here, use <...> instead of "..." for include/linux/*.h
and include/sound/*.h headers too.
Signed-off-by: NMasahiro Yamada <yamada.masahiro@socionext.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1495081793-9707-2-git-send-email-yamada.masahiro@socionext.com

If we restrict this helper to only kms drivers (which is the case) we
can look up the correct mode easily ourselves. But it's a bit tricky:

- All legacy drivers look at crtc->hwmode. But that is updated already
  at the beginning of the modeset helper, which means when we disable
  a pipe. Hence the final timestamps might be a bit off. But since
  this is an existing bug I'm not going to change it, but just try to
  be bug-for-bug compatible with the current code. This only applies
  to radeon&amdgpu.

- i915 tries to get it perfect by updating crtc->hwmode when the pipe
  is off (i.e. vblank->enabled = false).

- All other atomic drivers look at crtc->state->adjusted_mode. Those
  that look at state->requested_mode simply don't adjust their mode,
  so it's the same. That has two problems: Accessing crtc->state from
  interrupt handling code is unsafe, and it's updated before we shut
  down the pipe. For nonblocking modesets it's even worse.

For atomic drivers try to implement what i915 does. To do that we add
a new hwmode field to the vblank structure, and update it from
drm_calc_timestamping_constants(). For atomic drivers that's called
from the right spot by the helper library already, so all fine. But
for safety let's enforce that.

For legacy driver this function is only called at the end (oh the
fun), which is broken, so again let's not bother and just stay
bug-for-bug compatible.

The  benefit is that we can use drm_calc_vbltimestamp_from_scanoutpos
directly to implement ->get_vblank_timestamp in every driver, deleting
a lot of code.

v2: Completely new approach, trying to mimick the i915 solution.

v3: Fixup kerneldoc.

v4: Drop the WARN_ON to check that the vblank is off, atomic helpers
currently unconditionally call this. Recomputing the same stuff should
be harmless.

v5: Fix typos and move misplaced hunks to the right patches (Neil).

v6: Undo hunk movement (kbuild).

Cc: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
Cc: Eric Anholt <eric@anholt.net>
Cc: Rob Clark <robdclark@gmail.com>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: NNeil Armstrong <narmstrong@baylibre.com>
Acked-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170509140329.24114-4-daniel.vetter@ffwll.ch

It's overkill to have a flag parameter which is essentially used just
as a boolean. This takes care of core + adjusting drivers.

Adjusting the scanout position callback is a bit harder, since radeon
also supplies it's own driver-private flags in there.

v2: Fixup misplaced hunks (Neil).

v3: kbuild says v1 was better ...

Cc: Mario Kleiner <mario.kleiner@tuebingen.mpg.de>
Cc: Eric Anholt <eric@anholt.net>
Cc: Rob Clark <robdclark@gmail.com>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Reviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NNeil Armstrong <narmstrong@baylibre.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170509140329.24114-2-daniel.vetter@ffwll.ch