get the max io mapping address of system memory to see if it is over
our card accessing range.
v2: move checking later
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NMonk Liu <monk.liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180209024410.1469-2-david1.zhou@amd.com

it will be used to check if the driver needs swiotlb
v2: Don't use inline, instead, move function to drm_memory.c (Michel Daenzer <michel@daenzer.net>)
Signed-off-by: NChunming Zhou <david1.zhou@amd.com>
Reviewed-by: NMonk Liu <monk.liu@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180209024410.1469-1-david1.zhou@amd.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

When we debug print what ioctl we're calling into, we include the pid.
If you have multiple processes rendering simulataneously, the error
return also needs the pid so you can see which of the ioctl calls was
the one to fail.
Signed-off-by: NEric Anholt <eric@anholt.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20180130215643.11016-1-eric@anholt.netReviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Hw older versions support non-alpha color formats derived
from native alpha color formats only on the primary layer.
For instance, RG16 native format without alpha works fine
on 2nd layer but XR24 (derived color format from AR24)
does not work on 2nd layer.
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Reviewed-by: NYannick Fertré <yannick.fertre@st.com>
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180201104243.20726-3-philippe.cornu@st.com

ltdc supports natively some color formats with alpha (like
ARGB8888, ARGB1555, ARGB4444...). Related non-alpha formats are
supported too (ARGB8888->XRGB8888, ARGB4444->XRGB4444...) by
adjusting ltdc blending factors.

Note: Wayland/Weston requests by default the non-alpha XRGB8888
color format.
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Reviewed-by: NYannick Fertré <yannick.fertre@st.com>
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180201104243.20726-2-philippe.cornu@st.com

Add support for the Synopsys DesignWare MIPI DSI version 1.31
Two registers need to be updated/added for supporting 1.31:
* PHY_TMR_CFG 0x9c (updated)
  1.30 [31:24] phy_hs2lp_time
       [23:16] phy_lp2hs_time
       [14: 0] max_rd_time

  1.31 [25:16] phy_hs2lp_time
       [ 9: 0] phy_lp2hs_time

* PHY_TMR_RD_CFG 0xf4 (new)
  1.31 [14: 0] max_rd_time
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Reviewed-by: NAndrzej Hajda <a.hajda@samsung.com>
Signed-off-by: NAndrzej Hajda <a.hajda@samsung.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206084251.303-1-philippe.cornu@st.com

This patch adds the DCS/GENERIC DSI read feature.
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Reviewed-by: NAndrzej Hajda <a.hajda@samsung.com>
Reviewed-by: NBrian Norris <briannorris@chromium.org>
Signed-off-by: NAndrzej Hajda <a.hajda@samsung.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180204213104.17834-1-philippe.cornu@st.com

The Versatile PL110 implementations use multiple endpoints:
from the PL111 port, the lines are routed through a PLD,
and from there forked so the same lines go to a VGA DAC and
an external TFT panel connector. This is discrete wireing
so there is no way to turn of one output, i.e. this is
really two endpoints, not two ports.

We model this with multiple endpoints, so we need to loop
over the available endpoints, check for panel or bridge on
each and accumulate the result before continuing.

The code already will give the panel preference over the
bridge, if present, so the output will be sent to the panel
if both a panel and a bridge is present on two endpoints
of the same port.

If they all return -EPROBE_DEFER we return -EPROBE_DEFER
as well.

If just one endpoint is present on the port, the behaviour
is the same as before.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206093540.8147-5-linus.walleij@linaro.org

The early Integrator CLCD synthesized in the Integrator CP and
IM-PD1 FPGAs are broken: their vertical and next base interrupts
are not functional. Support these variants by simply disabling
the use of the vblank interrupt on these variants.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206093540.8147-4-linus.walleij@linaro.org

The early Integrator CLCD synthesized in the Integrator CP and
IM-PD1 FPGAs are broken: their clock dividers do not work
properly. Support disabling the clock divider and drive the
clock directly from the parent under these circumstances.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206093540.8147-3-linus.walleij@linaro.org

The ARM Versatile series can do RGB/BGR565 with an external
"PLD" (Programmable Logical Device). However the CLCD does not
have control bits for this, so it needs to be set into the
ordinary 16BPP mode, then the RGB/BGR565 handling of the pixel
data is handled by configuring the PLD through the external
register.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206093540.8147-2-linus.walleij@linaro.org

With a bit of refactoring we can contain the variant data for
the strange PL110 versions that is feature-incomplete PL110 for
the ARM Integrator/CP and somewhere inbetween PL110 and PL111
for the ARM Versatile AB and Versatile PB.

We also accomodate for the custom duct-taped RGB565/BGR565 support
in the Versatile variant.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180206093540.8147-1-linus.walleij@linaro.org

The ARM reference designs "Versatile AB" and "Versatile PB"
contain panel connectors with autodetection of the connected
panel type. This adds a small driver utilizing the MFD syscon
look-up to read the autodetection register and set up the
corresponding panel appropriately.

In the source file there is a bit of elaboration of the
panel types and interfaces on these boards.

This was tested with the PL111 DRM driver on the ARM Versatile
AB with the IB2 daughterboard.
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180205192013.5349-2-linus.walleij@linaro.org

This adds a pretty simple set of device tree bindings for
ARM Versatile panels appearing as child nodes of a system
controller.

Cc: devicetree@vger.kernel.org
Reviewed-by: NRob Herring <robh@kernel.org>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180205192013.5349-1-linus.walleij@linaro.org

Why is the argument to ->mode_valid() called "crtc"?
I guess a copy/paste error. Fix it.
Reviewed-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180129095531.4557-1-linus.walleij@linaro.org

This will make it possible for userspace to know whether reading
will block, without blocking on the fd. This makes it possible to
drain all queued CRC's in blocking mode, without having to reopen
the fd.
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180202142743.68527-1-maarten.lankhorst@linux.intel.comReviewed-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
[mlankhorst: Use POLLIN|POLLRDNORM, based on Ville's suggestion]

Use drm_mode_get_hv_timing() to fill out the plane clip rectangle.

Note that this replaces crtc_state->adjusted_mode usage with
crtc_state->mode. The latter is the correct choice since that's the
mode the user provided and it matches the plane crtc coordinates
the user also provided.

Once everyone agrees on this we can move the clip handling into
drm_atomic_helper_check_plane_state().

Cc: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180123170857.13818-2-ville.syrjala@linux.intel.com
Reviewed-by: Maxime Ripard <maxime.ripard@free-electrons.com> #irc

Use drm_mode_get_hv_timing() to fill out the plane clip rectangle.

No functional changes as the code already uses crtc_state->mode
to populate the clip, which is also what drm_mode_get_hv_timing()
uses.

Once everyone agrees on this we can move the clip handling into
drm_atomic_helper_check_plane_state().

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180123170857.13818-1-ville.syrjala@linux.intel.comReviewed-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>

I'm stepping down, also handing all the drm-misc stuff to the new
team. Plan is that Sean handles 4.17, and Maarten then has fun with
4.18 as his first release.

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Gustavo Padovan <gustavo@padovan.org>
Cc: Sean Paul <seanpaul@chromium.org>
Acked-by: NSean Paul <seanpaul@chromium.org>
Acked-by: NMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: NGustavo Padovan <gustavo.padovan@collabora.com>
Acked-by: NJani Nikula <jani.nikula@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Signed-off-by: NSean Paul <seanpaul@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180131102156.25634-1-daniel.vetter@ffwll.ch

If the wait timeouts, the caps are probably invalid and we shouldn't be
passing them to userspace.
Signed-off-by: NTomeu Vizoso <tomeu.vizoso@collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20171127142126.25765-1-tomeu.vizoso@collabora.comSigned-off-by: NGerd Hoffmann <kraxel@redhat.com>

Assign true or false to boolean variables instead of an integer value.

This issue was detected with the help of Coccinelle.
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/20180130100528.GA7154@training

The parentheses are in the wrong place here so we pass the bits per
pixel as zero.

Fixes: abbee623 ("drm/mgag200: Added resolution and bandwidth limits for various G200e products.")
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20180125142655.GA23885@mwanda

drm_encoder_slave is the old way to write bridge drivers, for i2c
bridges only. It's deprecated, and definitely should not be used in
new drivers. This has absolutely nothing to do with the new bridge
driver infrastructure implemented by drm_bridge.

What's even strange is that arcpgu doesn't even use any of this, it
really only wants a plain normal drm_encoder. Nuke all the surplus
real estate.

v2: Actually git add after compile testing ...

v3: Clarify commit message and stop including drm_encoder_slave.h.

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Reviewed-by: NLaurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180117141755.16933-1-daniel.vetter@ffwll.ch

The structure should really be sw_sync_create_fence_data rather than
sw_sync_ioctl_create_fence which is the function name.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1515988079-8677-1-git-send-email-shawn.guo@linaro.org

Userspace can set a FB_ID on a plane without setting CRTC_ID, which
will fail with -EINVAL, but the kernel shouldn't warn about that.

Same for !FB_ID and CRTC_ID being set.
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20180130102704.28016-1-maarten.lankhorst@linux.intel.comReviewed-by: NHarry Wentland <harry.wentland@amd.com>

The dw_mipi_dsi_host_transfer() must return the number of
bytes transmitted/received on success instead of 0.
Note: As the read feature is not implemented, only the
transmitted number of bytes is returned for the moment.
Reviewed-by: NBrian Norris <briannorris@chromium.org>
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Signed-off-by: NArchit Taneja <architt@codeaurora.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180125103800.1999-3-philippe.cornu@st.com

The dcs/generic dsi read feature is not yet implemented so it
is important to warn the host_transfer() caller in case of
read operation requests.
Reviewed-by: NBrian Norris <briannorris@chromium.org>
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Signed-off-by: NArchit Taneja <architt@codeaurora.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180125103800.1999-2-philippe.cornu@st.com

To optimize data transfers, align pitch on 128 bytes & height
on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by: NYannick Fertre <yannick.fertre@st.com>
Signed-off-by: NVincent Abriou <vincent.abriou@st.com>
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180130104200.21602-1-philippe.cornu@st.com

Add support for the stm dsi phy/wrapper version 1.31.
Only lane capabilities need to be modified.
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180122164628.24728-1-philippe.cornu@st.com

In the dsi panel example, clock names in the "clock-names"
field have been swapped:
* "pclk" (peripheral clock) is <&rcc 1 CLK_F469_DSI> on stm32f4
* "ref" (dsi phy pll ref clock) is <&clk_hse> on stm32f4
Signed-off-by: NPhilippe Cornu <philippe.cornu@st.com>
Reviewed-by: NRob Herring <robh@kernel.org>
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20180122153547.23470-1-philippe.cornu@st.com

During a non-blocking commit, it is possible to return before the
commit_tail work is queued (-ERESTARTSYS, for example).

Since a reference on the crtc commit object is obtained for the pending
vblank event when preparing the commit, the above situation will leave
us with an extra reference.

Therefore, if the commit_tail worker has not consumed the event at the
end of a commit, release it's reference.

Changes since v1:
- Also check for state->event->base.completion being set, to
  handle the case where stall_checks() fails in setup_crtc_commit().
Changes since v2:
- Add a flag to drm_crtc_commit, to prevent dereferencing a freed event.
  i915 may unreference the state in a worker.

Fixes: 24835e44 ("drm: reference count event->completion")
Cc: <stable@vger.kernel.org> # v4.11+
Signed-off-by: NLeo (Sunpeng) Li <sunpeng.li@amd.com>
Acked-by: Harry Wentland <harry.wentland@amd.com> #v1
Signed-off-by: NMaarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180117115108.29608-1-maarten.lankhorst@linux.intel.comReviewed-by: NSean Paul <seanpaul@chromium.org>

We never support certain mode flags etc. Reject those early on in the
mode_config.mode_valid() hook. That allows us to remove some duplicated
checks from the connector .mode_valid() hooks, and it guarantees that
we never see those flags even from user mode as the
mode_config.mode_valid() hooks gets executed for those as well.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-11-ville.syrjala@linux.intel.comReviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Allow drivers to provide a device wide .mode_valid() hook in addition to
the already existing crtc/encoder/bridge/connector hooks. This can be
used to validate device/driver wide constraings without having to add
those to the other hooks. And since we call this hook also for user
modes later on in the modeset we don't have to worry about anything the
hook has already rejected.

I also have some further ideas for this hook. Eg. we could replace the
drm_mode_set_crtcinfo(HALVE_V) call in drm_mode_convert_umode()/etc.
with a driver specific variant via this hook. At least on i915 we would
like to pass CRTC_STEREO_DOUBLE to that function instead, and then
we could safely use the crtc_ timings in all our .mode_valid() hooks,
which would allow us to reuse those hooks for validating the
adjusted_mode during a modeset.

v2: Fix the language fails in the kernel docs (Daniel)
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-10-ville.syrjala@linux.intel.comReviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

BUILTIN, CRTC_C, CLOCK_C, and DEFULT mode types are unused. Let's
refuse to generate them or accept them from userspace either. A
cursory check didn't reveal any userspace code that would depend
on these.

v2: Recommend DRIVER instead of BUILTIN (ajax)

Cc: Jose Abreu <Jose.Abreu@synopsys.com>
Cc: Adam Jackson <ajax@redhat.com>
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: NAdam Jackson <ajax@redhat.com>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171115154504.14338-1-ville.syrjala@linux.intel.comReviewed-by: NJose Abreu <joabreu@synopsys.com>

No idea what the DRM_MODE_TYPE_CLOCK_CRTC_C define is supposed to
achieve. Totally unused so kill if off.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-8-ville.syrjala@linux.intel.comReviewed-by: NAlex Deucher <alexander.deucher@amd.com>

For some reason drm_mode_set_crtcinfo() does nothing if the mode has
the DRM_MODE_TYPE_BUILTIN flag set without the other bit from
DRM_MODE_TYPE_CRTC_C also set. I have zero idea what that is supposed
to achieve, but since we have no users for neither flag bit let's kill
this nonsense off.

v2: Fix typo in commit message
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-7-ville.syrjala@linux.intel.comReviewed-by: NAlex Deucher <alexander.deucher@amd.com>

These days DRM_MODE_TYPE_USERDEF is used to flag modes defined via the
kernel command line. Update the docs to reflect that fact.
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-6-ville.syrjala@linux.intel.comAcked-by: NAlex Deucher <alexander.deucher@amd.com>

Reject any mode with DRM_MODE_FLAG_BCAST. We have no code that even
checks for this flag hence it can't possibly do any good.

I think this maybe originated from fbdev where it was supposed to
indicate PAL/NTSC broadcast timings. I have no idea why those would
have to be identified by a flag rather than by just the timings
themselves. And then I assume it got copied into xfree86 for
fbdevhw, and later on it leaked into the randr protocol and kms uapi.

Since kms fbdev emulation never uses the corresponding fbdev flag
there should be no sane way for this to come back into kms via
userspace either.

Cc: Jose Abreu <Jose.Abreu@synopsys.com>
Cc: Adam Jackson <ajax@redhat.com>
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-5-ville.syrjala@linux.intel.comReviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NJose Abreu <joabreu@synopsys.com>

Reject any mode with DRM_MODE_FLAG_PIXMUX. We have no code that even
checks for this flag hence it can't possibly do any good.

Looks like this flag had something to do the the controller<->ramdac
interface with some ancient S3 graphics adapters. Why someone though
it would be a good idea to expose it directly to users I don't know.
And later on it got copied into the randr protocol and kms uapi.

Cc: Jose Abreu <Jose.Abreu@synopsys.com>
Cc: Adam Jackson <ajax@redhat.com>
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171114183258.16976-4-ville.syrjala@linux.intel.comReviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NJose Abreu <joabreu@synopsys.com>