platform/kernel/linux-rpi.git
6 years agodrm/omap: Determine connector type directly in omap_connector.c
Laurent Pinchart [Tue, 5 Jun 2018 21:31:57 +0000 (00:31 +0300)]
drm/omap: Determine connector type directly in omap_connector.c

Instead of determining the connector type from the type of the display's
omap_dss_device and passing it to the omap_connector_init() function,
move the type determination code to omap_connector.c and remove the type
argument to the connector init function. This moves code to a more
natural location, making the driver easier to read.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Pass both output and display omap_dss_device to connector init
Laurent Pinchart [Thu, 31 May 2018 20:14:43 +0000 (23:14 +0300)]
drm/omap: Pass both output and display omap_dss_device to connector init

The drm_connector implementation requires access to the omap_dss_device
corresponding to the display, which is passed to its initialization
function and stored internally. Refactoring of the timings operations
will require access to the output omap_dss_device. To prepare for that,
pass it to the connector initialization function and store it internally
as well.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Don't call HDMI mode and infoframe operations recursively
Laurent Pinchart [Thu, 31 May 2018 19:56:59 +0000 (22:56 +0300)]
drm/omap: Don't call HDMI mode and infoframe operations recursively

The HDMI mode (.set_hdmi_mode()) and infoframe (.set_infoframe())
operations are called recursively from the display device back to the
HDMI encoder. This isn't required, as all components other than the HDMI
encoder just forward the operation to the previous component in the
chain. Call the operations directly on the HDMI encoder.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Pass both output and display omap_dss_device to encoder init
Laurent Pinchart [Thu, 31 May 2018 20:14:43 +0000 (23:14 +0300)]
drm/omap: Pass both output and display omap_dss_device to encoder init

The drm_encoder implementation requires access to the omap_dss_device
corresponding to the display, which is passed to its initialization
function and stored internally. Clean up of the HDMI mode and infoframe
handling will require access to the output omap_dss_device. To prepare
for that, pass it to the encoder initialization function and store it
internally as well.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Get from CRTC to display device directly
Laurent Pinchart [Thu, 31 May 2018 20:04:30 +0000 (23:04 +0300)]
drm/omap: Get from CRTC to display device directly

The CRTC mode set implementation needs to access the omap_dss_device for
the pipeline display. To do so, it iterates over all pipelines to find
the one that contains an encoder corresponding to the CRTC, and request
the display device from the encoder. That's a very complicated dance
when the CRTC has a direct pipeline pointer already, and the pipeline
contains a pointer to the display device.

Replace the convoluted code with direct access.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Don't call EDID read operation recursively
Laurent Pinchart [Thu, 31 May 2018 19:09:14 +0000 (22:09 +0300)]
drm/omap: Don't call EDID read operation recursively

Instead of calling the EDID read operation (.read_edid()) recursively
from the display device back to the first device that provides EDID read
support, iterate over the devices manually in the DRM connector code.
This moves the complexity to a single central location and simplifies
the logic in omap_dss_device drivers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Move HPD disconnection handling to omap_connector
Laurent Pinchart [Wed, 30 May 2018 13:53:43 +0000 (16:53 +0300)]
drm/omap: Move HPD disconnection handling to omap_connector

On HDMI outputs, CEC support requires notification of HPD signal
deassertion. The HPD signal can be handled by various omap_dss_device
instances in the pipeline, and all of them forward HPD events to the
OMAP4 internal HDMI encoder.

Knowledge of the DSS internals need to be removed from the
omap_dss_device instances in order to migrate to drm_bridge. To do so,
move HPD handling for CEC to the omap_connector.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Merge HPD enable operation with HPD callback registration
Laurent Pinchart [Wed, 30 May 2018 18:46:44 +0000 (21:46 +0300)]
drm/omap: Merge HPD enable operation with HPD callback registration

The omap_dss_device .enable_hpd() and .disable_hpd() are used to enable
and disable hot-plug detection at omapdrm probe and remove time. This is
required to avoid reporting hot-plug detection events before the DRM
infrastructure is ready to accept them, as that could result in crashes
or other malfunction.

Hot-plug event reporting is conditioned by both HPD being enabled
through the .enable_hpd() operation and by the HPD callback being
registered though the .register_hpd_cb() operation. We thus don't need a
separate enable operation if we can guarantee that callbacks won't be
registered too early.

HPD callbacks are registered at connector initialization time, which is
too early to start reporting HPD events. There's however nothing
blocking a move of callback registration to a later time when the
omapdrm driver calls the HPD enable operations. Do so, and remove the
HPD enable operation completely from omap_dss_device drivers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Remove unneeded safety checks in the HPD operations
Laurent Pinchart [Wed, 30 May 2018 15:51:59 +0000 (18:51 +0300)]
drm/omap: Remove unneeded safety checks in the HPD operations

The HPD-related omap_dss_device operations are now only called when the
device supports HPD. There's no need to duplicate that check in the
omap_dss_device drivers. The .register_hpd_cb() operation can as a
result be turned into a void operation.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Don't call HPD registration operations recursively
Laurent Pinchart [Wed, 30 May 2018 15:49:48 +0000 (18:49 +0300)]
drm/omap: Don't call HPD registration operations recursively

Instead of calling the hot-plug detection callback registration
operations (.register_hpd_cb() and .unregister_hpd_cb()) recursively
from the display device back to the first device that provides hot plug
detection support, iterate over the devices manually in the DRM
connector code. This moves the complexity to a single central location
and simplifies the logic in omap_dss_device drivers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Don't call .detect() operation recursively
Laurent Pinchart [Mon, 28 May 2018 13:49:36 +0000 (16:49 +0300)]
drm/omap: Don't call .detect() operation recursively

Instead of calling the .detect() operation recursively from the display
device back to the first device that provides hot plug detection
support, iterate over the devices manually in the DRM connector
.detect() implementation. This moves the complexity to a single central
location and simplifies the logic in omap_dss_device drivers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Add device operations flags
Laurent Pinchart [Mon, 28 May 2018 13:03:13 +0000 (16:03 +0300)]
drm/omap: dss: Add device operations flags

When an omap_dss_device operation can be implemented in multiple places
in a chain of devices, it is important to find out which device to
address to perfom the operation. This is currently done by calling the
operation on the display device at the end of the chain, and recursively
delagating the operation to the previous device if it can't be performed
locally. The drawback of this approach is an increased complexity in
omap_dss_device drivers.

In order to simplify the drivers, we will switch from a recursive model
to an interative model, centralizing the complexity in a single
location. This requires knowing which operations an omap_dss_device
supports at runtime. We can already test which operations are
implemented by checking the operation pointer, but implemented
operations can require resources whose availability varies between
systems. For instance a hot-plug signal from a connector can be wired to
a GPIO or to a bridge chip.

Add operation flags that can be set in the omap_dss_device structure by
drivers to signal support for operations.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Move most omap_dss_driver operations to omap_dss_device_ops
Laurent Pinchart [Fri, 1 Jun 2018 16:45:01 +0000 (19:45 +0300)]
drm/omap: Move most omap_dss_driver operations to omap_dss_device_ops

omap_dss_device instances have two ops structures, omap_dss_driver and
omap_dss_device_ops. The former is used for devices at the end of the
pipeline (a.k.a. display devices), and the latter for intermediate
devices.

Having two sets of operations isn't convenient as code that iterates
over omap_dss_device instances need to take them both into account.
There's currently a reasonably small amount of such code, but more will
be introduced to move the driver away from recursive operations. To
simplify current and future code, move all operations that are not
specific to the display device to the omap_dss_device_ops.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: panel-tpo-td043mtea1: Convert to the GPIO descriptors API
Laurent Pinchart [Mon, 28 May 2018 14:18:26 +0000 (17:18 +0300)]
drm/omap: panel-tpo-td043mtea1: Convert to the GPIO descriptors API

The GPIO descriptor API is favoured over the plain GPIO API for consumer
drivers. Using it simplifies the driver code.

As the descriptor API handles the active-low flag internally we need to
invert the polarity of all GPIO operations in the driver. Rename the
nreset_gpio field to reset_gpio to reflect that.

The reset GPIO is mandatory, so drop conditional tests through the
driver.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: panel-tpo-td028ttec1: Drop unneeded linux/gpio.h header
Laurent Pinchart [Mon, 28 May 2018 14:16:52 +0000 (17:16 +0300)]
drm/omap: panel-tpo-td028ttec1: Drop unneeded linux/gpio.h header

The driver doesn't use GPIOs and thus doesn't need to include the
linux/gpio.h header.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: panel-sony-acx565akm: Convert to the GPIO descriptors API
Laurent Pinchart [Mon, 28 May 2018 14:18:26 +0000 (17:18 +0300)]
drm/omap: panel-sony-acx565akm: Convert to the GPIO descriptors API

The GPIO descriptor API is favoured over the plain GPIO API for consumer
drivers. Using it simplifies the driver code.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: panel-nec-nl8048hl11: Convert to the GPIO descriptors API
Laurent Pinchart [Mon, 28 May 2018 14:18:26 +0000 (17:18 +0300)]
drm/omap: panel-nec-nl8048hl11: Convert to the GPIO descriptors API

The GPIO descriptor API is favoured over the plain GPIO API for consumer
drivers. Using it simplifies the driver code.

The reset GPIO is mandatory, so drop conditional tests through the
driver. The qvga GPIO is unused, so drop it completely.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: encoder-tfp410: Convert to the GPIO descriptors API
Laurent Pinchart [Mon, 28 May 2018 14:18:26 +0000 (17:18 +0300)]
drm/omap: encoder-tfp410: Convert to the GPIO descriptors API

The GPIO descriptor API is favoured over the plain GPIO API for consumer
drivers. Using it simplifies the driver code.

As the descriptor API handles the active-low flag internally we need to
invert the polarity of all GPIO operations in the driver.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: connector-hdmi: Convert to the GPIO descriptors API
Laurent Pinchart [Mon, 28 May 2018 14:18:26 +0000 (17:18 +0300)]
drm/omap: connector-hdmi: Convert to the GPIO descriptors API

The GPIO descriptor API is favoured over the plain GPIO API for consumer
drivers. Using it simplifies the driver code.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Check omap_dss_device type based on the output_type field
Laurent Pinchart [Sun, 27 May 2018 19:01:24 +0000 (22:01 +0300)]
drm/omap: Check omap_dss_device type based on the output_type field

Various functions that need to differentiate between omap_dss_device
instances corresponding to displays and to internal encoders use the
omap_dss_device.driver field, which is only set for display instances.
This gets in the way of the omap_dss_device operations refactoring.
Replace that with a check based on the output_type field which is set
for all omap_dss_device instances but displays.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Remove unnecessary display output sanity checks
Laurent Pinchart [Thu, 29 Mar 2018 11:51:04 +0000 (14:51 +0300)]
drm/omap: Remove unnecessary display output sanity checks

The omapdrm driver checks at suspend and resume time whether the
displays it operates on have their driver operations set. This check is
unneeded, as all display drivers set the driver operations field at
probe time and never touch it afterwards. This is furthermore proven by
the dereferencing of the driver field without checking it first in
several locations.

The omapdss driver performs a similar check at shutdown time. This is
unneeded as well, as the for_each_dss_display() macro it uses to iterate
over displays locates the displays by checking the driver field
internally.

As those checks are unnecessary, remove them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove omap_dss_driver .[gs]et_mirror operations
Laurent Pinchart [Thu, 29 Mar 2018 11:08:12 +0000 (14:08 +0300)]
drm/omap: dss: Remove omap_dss_driver .[gs]et_mirror operations

The .get_mirror() and .set_mirror() omap_dss_driver operations are
implemented by the panel-tpo-td043mtea1 driver but are never used.
Remove them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove unused omap_dss_driver operations
Laurent Pinchart [Thu, 29 Mar 2018 11:08:12 +0000 (14:08 +0300)]
drm/omap: dss: Remove unused omap_dss_driver operations

The .probe(), .remove(), .run_test(), .get_rotate() and .set_rotate()
omap_dss_driver operations are not used. Remove them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove the dss_mgr_(dis)connect() operations
Laurent Pinchart [Tue, 6 Mar 2018 21:34:53 +0000 (23:34 +0200)]
drm/omap: dss: Remove the dss_mgr_(dis)connect() operations

The dss_mgr .connect() and .disconnect() are implemented as no-op in
omapdrm. The operations are unneeded, remove them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Set dispc_channel_connect from DSS output connect handlers
Laurent Pinchart [Tue, 6 Mar 2018 22:28:18 +0000 (00:28 +0200)]
drm/omap: Set dispc_channel_connect from DSS output connect handlers

The omap_dss_device.dispc_channel_connect field is used by DSS outputs
to fail the .enable() operation if they're not connected. Set the field
directly from the (dis)connect handlers of the DSS outputs instead of
going through the CRTC dss_mgr operations.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Remove supported output check in CRTC connect handler
Laurent Pinchart [Tue, 6 Mar 2018 22:22:10 +0000 (00:22 +0200)]
drm/omap: Remove supported output check in CRTC connect handler

The CRTC connect handler checks whether the DSS output supports the
DISPC channel assigned to it. As the channel is assigned to the output
by the output driver a failure there could only result from a driver
bug. All the output drivers have been verified and they are always
assigned a DISPC channel that is supported on the SoC they run on. The
check can thus be removed.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Remove omap_crtc_output global array
Laurent Pinchart [Tue, 6 Mar 2018 21:38:21 +0000 (23:38 +0200)]
drm/omap: Remove omap_crtc_output global array

The omap_crtc_output global array is used to look up the DSS output
device by channel. We can replace that by accessing the output device
from the pipeline if we store the pipeline pointer in the omap_crtc
structure.

The global array is also used to protect against double connection of an
output. This can't happen with the connection handling mechanism going
from DSS outputs to displays. We can thus drop that check, allowing
removal of the global array.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Store CRTC lookup by channel table in omap_drm_private
Laurent Pinchart [Tue, 6 Mar 2018 22:01:33 +0000 (00:01 +0200)]
drm/omap: Store CRTC lookup by channel table in omap_drm_private

The omap_crtcs global array is used to store pointers to omap_crtc
indexed by DISPC channel number, in order to look them up in the dss_mgr
operations. Store the information in the omap_drm_private structure in
the form of an array of omap_drm_pipeline pointers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Pass pipe pointer to omap_crtc_init()
Laurent Pinchart [Tue, 6 Mar 2018 21:37:25 +0000 (23:37 +0200)]
drm/omap: Pass pipe pointer to omap_crtc_init()

Replace the dss display device pointer by a pipe pointer that will allow
the omap_crtc_init() function to access both the display and the DSS
output. As a result we can remove the omapdss_device_get_dispc_channel()
function that is now unneeded.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Merge two disconnection helpers
Laurent Pinchart [Mon, 5 Mar 2018 23:51:31 +0000 (01:51 +0200)]
drm/omap: dss: Merge two disconnection helpers

To simplify the pipeline disconnection handling merge the
omapdss_device_disconnect() and omapdss_output_unset_device() functions.
The device state check is now called for every device in the pipeline,
extending this sanity check coverage.

There is no need to return an error from omapdss_device_disconnect()
when the check fails, as omapdss_output_unset_device() used to do, given
that we can't prevent disconnection due to device unbinding (the return
value of omapdss_output_unset_device() is never checked in the current
code for that reason).

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move display type validation to initialization time
Laurent Pinchart [Mon, 5 Mar 2018 23:25:13 +0000 (01:25 +0200)]
drm/omap: dss: Move display type validation to initialization time

The display type is validated when the display is connected to the DSS
output. We already have all the information we need for validation when
initializing the outputs. Move validation to output initialization to
simplify pipeline connection handling.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move connection checks to omapdss_device_(dis)connect
Laurent Pinchart [Sat, 26 May 2018 17:25:10 +0000 (20:25 +0300)]
drm/omap: dss: Move connection checks to omapdss_device_(dis)connect

When a DSS output is (dis)connected the omapdss_output_(un)set_device()
function performs a sanity check to ensure that the output isn't already
(dis)connected. The check is unnecessary as those situations should
never happen, but can nonetheless be useful to catch driver bugs. To
prepare for removal of the omapdss_output_(un)set_device() functions
move the connection check to the omapdss_device_connect() function. The
omapdss_device_disconnect() already contains a corresponding check.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Reverse direction of DSS device (dis)connect operations
Laurent Pinchart [Sun, 4 Mar 2018 21:42:36 +0000 (23:42 +0200)]
drm/omap: Reverse direction of DSS device (dis)connect operations

The omapdrm and omapdss drivers are architectured based on display
pipelines made of multiple components handled from sink (display) to
source (DSS output). This is incompatible with the DRM bridge and panel
APIs that handle components from source to sink.

To reconcile the omapdrm and omapdss drivers with the DRM bridge and
panel model, we need to reverse the direction of the DSS device
operations. Start with the connect and disconnect operations.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Group CRTC, encoder, connector and dssdev in a structure
Laurent Pinchart [Mon, 5 Mar 2018 13:02:22 +0000 (15:02 +0200)]
drm/omap: Group CRTC, encoder, connector and dssdev in a structure

Create an omap_drm_pipeline structure to model display pipelines, made
of a CRTC, an encoder, a connector and a DSS display device. This allows
grouping related parameters together instead of storing them in
independent arrays and thus improves code readability.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Create all planes before CRTCs
Laurent Pinchart [Mon, 5 Mar 2018 17:11:30 +0000 (19:11 +0200)]
drm/omap: Create all planes before CRTCs

Creating all the planes in a single location instead of creating them
per-CRTC with remaining planes then created in a second step simplifies
the logic.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Remove unneeded variable assignments in omap_modeset_init
Laurent Pinchart [Mon, 5 Mar 2018 12:47:47 +0000 (14:47 +0200)]
drm/omap: Remove unneeded variable assignments in omap_modeset_init

The crtc_idx and plane_idw variables in the main loop are always equal
to the loop counter i, use it instead. Don't unnecessarily initialize
dssdev to NULL.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Get regulators at probe time
Laurent Pinchart [Sun, 4 Mar 2018 22:10:55 +0000 (00:10 +0200)]
drm/omap: dss: Get regulators at probe time

Regulators for the DPI, DSI, HDMI, SDI and VENC outputs are all looked
up when connecting the output omap_dss_device. There's no need to delay
regulator handling to that time, get the regulators at probe time.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove duplicated parameter to dss_mgr_(dis)connect()
Laurent Pinchart [Sun, 4 Mar 2018 21:55:56 +0000 (23:55 +0200)]
drm/omap: dss: Remove duplicated parameter to dss_mgr_(dis)connect()

The dss_mgr_connect() and dss_mgr_disconnect() functions take two
omap_dss_device pointers as parameters, which are always set to the same
value by all callers. Remove the duplicated pointer.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Add function to retrieve display for an output
Laurent Pinchart [Mon, 5 Mar 2018 12:28:06 +0000 (14:28 +0200)]
drm/omap: dss: Add function to retrieve display for an output

Add a new omapdss_display_get() function to retrieve the omap_dss_device
for a given DSS output. This will be used when reversing the direction
of the DSS pipeline handling logic.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Add for_each_dss_output() macro
Laurent Pinchart [Sun, 4 Mar 2018 20:28:25 +0000 (22:28 +0200)]
drm/omap: dss: Add for_each_dss_output() macro

Similarly to for_each_dss_display(), the for_each_dss_output() macro
iterates over all the DSS connected outputs.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Acquire next dssdev at probe time
Laurent Pinchart [Fri, 2 Mar 2018 20:13:06 +0000 (22:13 +0200)]
drm/omap: dss: Acquire next dssdev at probe time

Look up the next dssdev at probe time based on device tree links for all
DSS outputs and encoders. This will be used to reverse the order of the
dssdev connect and disconnect call chains.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: venc: Move initialization code from bind to probe
Laurent Pinchart [Sat, 3 Mar 2018 16:52:59 +0000 (18:52 +0200)]
drm/omap: dss: venc: Move initialization code from bind to probe

There's no reason to delay initialization of most of the driver (such as
mapping memory I/O or enabling runtime PM) to the component bind
handler. Perform as much of the initialization as possible at probe
time, initializing at bind time only the parts that depends on the DSS.
The cleanup code is moved from unbind to remove in a similar way.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: hdmi5: Move initialization code from bind to probe
Laurent Pinchart [Sat, 3 Mar 2018 16:52:59 +0000 (18:52 +0200)]
drm/omap: dss: hdmi5: Move initialization code from bind to probe

There's no reason to delay initialization of most of the driver (such as
mapping memory I/O or enabling runtime PM) to the component bind
handler. Perform as much of the initialization as possible at probe
time, initializing at bind time only the parts that depends on the DSS.
The cleanup code is moved from unbind to remove in a similar way.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: hdmi4: Move initialization code from bind to probe
Laurent Pinchart [Sat, 3 Mar 2018 16:52:59 +0000 (18:52 +0200)]
drm/omap: dss: hdmi4: Move initialization code from bind to probe

There's no reason to delay initialization of most of the driver (such as
mapping memory I/O or enabling runtime PM) to the component bind
handler. Perform as much of the initialization as possible at probe
time, initializing at bind time only the parts that depends on the DSS.
The cleanup code is moved from unbind to remove in a similar way.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: dsi: Move initialization code from bind to probe
Laurent Pinchart [Sat, 3 Mar 2018 16:52:59 +0000 (18:52 +0200)]
drm/omap: dss: dsi: Move initialization code from bind to probe

There's no reason to delay initialization of most of the driver (such as
mapping memory I/O or enabling runtime PM) to the component bind
handler. Perform as much of the initialization as possible at probe
time, initializing at bind time only the parts that depends on the DSS.
The cleanup code is moved from unbind to remove in a similar way.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Cleanup error paths in output init functions
Laurent Pinchart [Fri, 2 Mar 2018 19:38:21 +0000 (21:38 +0200)]
drm/omap: dss: Cleanup error paths in output init functions

Rename the jump labels according to the cleanup they perform, not the
location they're accessed from, and move functions from error checks to
cleanup paths, and move reference handling to simplify cleanup.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: displays: Don't cast dssdev to panel data unnecessarily
Laurent Pinchart [Fri, 2 Mar 2018 20:05:41 +0000 (22:05 +0200)]
drm/omap: displays: Don't cast dssdev to panel data unnecessarily

The connect handle of the analog TV and HDMI connectors casts the dssdev
to panel data only to then access fields of the panel data that are also
present in the dssdev. Remove the cast and use dssdev directly.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Extend omapdss_of_find_source_for_first_ep() to sinks
Laurent Pinchart [Fri, 2 Mar 2018 19:11:06 +0000 (21:11 +0200)]
drm/omap: dss: Extend omapdss_of_find_source_for_first_ep() to sinks

The omapdss_of_find_source_for_first_ep() function locates the source
corresponding to the first endpoint of the first port of a device node.
We can easily extend it to locate sinks as well by passing the port
number as a parameter. This will be useful to find sinks in encoders
drivers.

Extend the function and rename it to omapdss_of_find_connected_device()
to reflect its new extended purpose.

Additionally, it is useful to differentiate between failures to return
the connected device because no link exists in the device tree for the
requested port, or because the connected device as described in the
device tree is invalid or not probed yet. Return NULL in the first case
and an error code in the second case, and update the callers
accordingly.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Replace omap_dss_device port number with bitmask
Laurent Pinchart [Sun, 4 Mar 2018 19:49:28 +0000 (21:49 +0200)]
drm/omap: dss: Replace omap_dss_device port number with bitmask

The omap_dss_device port_num field stores the DT port number associated
with the device. The field is used in different ways depending on the
device type:

- For DPI outputs, the port number is used as an identifier of the DPI
instance

- For sources, the port number is used to look up the omap_dss_device by
DT port node

As omap_dss_device instances are only looked up as sources by sinks,
setting the field to the number of the source port works for both use
cases.

However, to enable looking up sinks, we need to record all the ports
associated with an omap_dss_device. Do so by turning the port_num field
into an of_ports bitmask. For DPI outputs the port number is
additionally stored in the dpi_data structure as the output ID.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Modify omapdss_find_output_from_display() to return channel
Laurent Pinchart [Fri, 2 Mar 2018 01:11:49 +0000 (03:11 +0200)]
drm/omap: dss: Modify omapdss_find_output_from_display() to return channel

The omapdss_find_output_from_display() function is only used to retrieve
the dispc channel corresponding to the display. Return the dispc channel
directly, and rename the function to omapdss_device_get_dispc_channel()
to match its new purpose.

The dssdev->id check is removed as the dssdev is guaranteed to be an
output and have a non-zero id, as proved by the lack of crash despite
the caller never checking the returned pointer before dereferencing it.

As the function is not specific to outputs anymore, move it from
output.c to base.c.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move DSS mgr ops and private data to dss_device
Laurent Pinchart [Fri, 2 Mar 2018 01:05:10 +0000 (03:05 +0200)]
drm/omap: dss: Move DSS mgr ops and private data to dss_device

The DSS manager ops and private data pointer are specific to a DSS
instance. Store them in the dss_device structure instead of global
variable.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Store dss_device pointer in omap_dss_device
Laurent Pinchart [Fri, 2 Mar 2018 00:54:16 +0000 (02:54 +0200)]
drm/omap: dss: Store dss_device pointer in omap_dss_device

Storing the dss_device pointer in the omap_dss_device structure will
allow accessing the dss_device from the dss_mgr API functions.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move and rename omap_dss_(get|put)_device()
Laurent Pinchart [Fri, 2 Mar 2018 00:43:45 +0000 (02:43 +0200)]
drm/omap: dss: Move and rename omap_dss_(get|put)_device()

The functions operate on any omap_dss_device, move them from display.c
to base.c. While at it rename them to match the naming of the other
functions operating on struct omap_dss_device.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove panel devices list
Laurent Pinchart [Thu, 1 Mar 2018 23:25:32 +0000 (01:25 +0200)]
drm/omap: dss: Remove panel devices list

The panel devices list isn't used anymore, all panel devices are
accessed through the global devices list. Remove it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Split omapdss_register_display()
Laurent Pinchart [Fri, 2 Mar 2018 00:15:36 +0000 (02:15 +0200)]
drm/omap: dss: Split omapdss_register_display()

Split the function into omapdss_display_init() to perform
display-specific initialization of the omap_dss_device, and
omapdss_register_display() to register the device. The latter will then
be replaced by more generic registration.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Make omap_dss_get_next_device() more generic
Laurent Pinchart [Fri, 2 Mar 2018 00:07:34 +0000 (02:07 +0200)]
drm/omap: dss: Make omap_dss_get_next_device() more generic

Despite its name, the omap_dss_get_next_device() function operates on
display devices only. Make it more generic by allowing operation on all
devices, with a parameter to specify the device type.

While at it rename the function to omapdss_device_get_next() to match
the naming of the other functions operating on struct omap_dss_device.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Rename for_each_dss_dev macro to for_each_dss_display
Laurent Pinchart [Thu, 1 Mar 2018 23:40:48 +0000 (01:40 +0200)]
drm/omap: dss: Rename for_each_dss_dev macro to for_each_dss_display

The macro iterates over displays only, rename it accordingly.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove output devices list
Laurent Pinchart [Thu, 1 Mar 2018 23:25:32 +0000 (01:25 +0200)]
drm/omap: dss: Remove output devices list

The output devices list isn't used anymore, all output devices are
accessed through the global devices list. Remove it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Move DSI debugfs clocks dump to dsi%u_clks files
Laurent Pinchart [Thu, 1 Mar 2018 22:40:04 +0000 (00:40 +0200)]
drm/omap: Move DSI debugfs clocks dump to dsi%u_clks files

The DSI clocks are dumped in the DSS-level debugfs clocks file. This
complicates the implementation as the DSI private data has to be looked
up through the outputs list. Simplify it by creating two debugfs files,
dsi1_clks and dsi2_clks, to dump the DSI clocks.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dsi: Simplify debugfs implementation
Laurent Pinchart [Thu, 1 Mar 2018 22:40:04 +0000 (00:40 +0200)]
drm/omap: dsi: Simplify debugfs implementation

The DSI debugfs regs and irqs show handlers received a pointer to the
DSI private data. There's no need to look it up from the list of DSS
outputs. Use the pointer directly, this allows simplifying the
implementation of the handlers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: displays: Remove input omap_dss_device from panel data
Laurent Pinchart [Wed, 28 Feb 2018 15:30:30 +0000 (17:30 +0200)]
drm/omap: displays: Remove input omap_dss_device from panel data

All connectors, encoders and panels store a pointer to their input
omap_dss_device in the panel driver data structure. This duplicates the
src field in the omap_dss_device structure. Remove the private copy and
use the src field.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move src and dst check and set to connection handlers
Laurent Pinchart [Wed, 28 Feb 2018 15:30:30 +0000 (17:30 +0200)]
drm/omap: dss: Move src and dst check and set to connection handlers

The encoders duplicate the same omap_dss_device src and dst fields set
and checks in their connect and disconnect handlers. Move the code to
the connect and disconnect wrappers.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: displays: Don't call disconnect handlers directly
Laurent Pinchart [Sat, 4 Aug 2018 19:10:44 +0000 (22:10 +0300)]
drm/omap: displays: Don't call disconnect handlers directly

In preparation for the move of checks from the disconnect handlers to
the omapdss_device_disconnect() function, replace direct calls to the
disconnect handlers at remove time with calls to
omapdss_device_disconnect().

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move debug message and checks to connection handlers
Laurent Pinchart [Wed, 28 Feb 2018 15:30:30 +0000 (17:30 +0200)]
drm/omap: dss: Move debug message and checks to connection handlers

The connectors, encoders and display duplicate the same debug messages
and connection checks in their omap_dss_device connect and disconnect
handlers. Move the code to the connect and disconnect wrappers.

To simplify the code the connect function returns -EBUSY unconditionally
if the device is already connected. This doesn't cause any change in
practice: the connect handler of displays is never called on a connected
device as it is only invoked during omapdrm initialization.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Add functions to connect and disconnect devices
Laurent Pinchart [Wed, 28 Feb 2018 15:30:30 +0000 (17:30 +0200)]
drm/omap: dss: Add functions to connect and disconnect devices

The omap_dss_device objects model display components and are connected
at runtime to create display pipelines. The connect and disconnect
operations implemented by each component contain lots of duplicate code.
As a first step towards fixing this, create new functions to wrap the
direct calls to those operations and use them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move common device operations to common structure
Laurent Pinchart [Wed, 28 Feb 2018 13:58:13 +0000 (15:58 +0200)]
drm/omap: dss: Move common device operations to common structure

The various types of omapdss_*_ops structures define multiple operations
that are not specific to a bus type. To simplify the code and remove
dependencies on specific bus types move those operations to a common
structure. Operations that are specific to a bus type are kept in the
specialized ops structures.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Allow looking up any device by port
Laurent Pinchart [Thu, 1 Mar 2018 21:35:55 +0000 (23:35 +0200)]
drm/omap: dss: Allow looking up any device by port

The omap_dss_find_output_by_port() function looks up an omap_dss_device
by port from the list of devices registered as outputs. In preparation
for looking up sinks in addition to sources, allow the function to look
up any registered device. Rename it to omap_dss_find_device_by_port() to
match its new purpose.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Rework output lookup by port node
Laurent Pinchart [Thu, 1 Mar 2018 21:35:55 +0000 (23:35 +0200)]
drm/omap: dss: Rework output lookup by port node

The omap_dss_find_output_by_port_node() function defined in output.c
looks up an output from its port node. To do so it needs to call helper
functions from dss-of.c to lookup the port parent and the port number.
As omap_dss_find_output_by_port_node() is only called by
omapdss_of_find_source_for_first_ep() from dss-of.c this goes back and
forth between the to source files and isn't very clear.

Simplify the code by passing both the parent and the port number to
omap_dss_find_output_by_port_node() instead of the port node, and rename
the function to omap_dss_find_output_by_port().

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Create and use omapdss_device_is_registered()
Laurent Pinchart [Wed, 28 Feb 2018 21:53:16 +0000 (23:53 +0200)]
drm/omap: dss: Create and use omapdss_device_is_registered()

The omapdss_component_is_loaded() function test whether a component is
loaded by checking whether it is present in the displays list or the
outputs list. Simplify the implementation by checking for the component
in the global omap_dss_device list.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Create global list of all omap_dss_device instances
Laurent Pinchart [Wed, 28 Feb 2018 21:49:24 +0000 (23:49 +0200)]
drm/omap: dss: Create global list of all omap_dss_device instances

The omap_dss_device instances are stored in two separate lists,
depending on whether they are panels or outputs. Create a third list
that stores all omap_dss_device instances to allow generic code to
operate on all instances.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Rename omap_dss_device list field to output_list
Laurent Pinchart [Wed, 28 Feb 2018 21:48:35 +0000 (23:48 +0200)]
drm/omap: dss: Rename omap_dss_device list field to output_list

For coherency with the panel_list field, rename list to output_list.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove omap_dss_device panel fields
Laurent Pinchart [Wed, 28 Feb 2018 12:48:45 +0000 (14:48 +0200)]
drm/omap: dss: Remove omap_dss_device panel fields

The omap_dss_device panel.dsi_pix_fmt and panel.dsi_mode fields are
unused. Remove them.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: displays: Remove videomode from omap_dss_device structure
Laurent Pinchart [Tue, 27 Feb 2018 18:11:52 +0000 (20:11 +0200)]
drm/omap: displays: Remove videomode from omap_dss_device structure

The omap_dss_device structure stores a videomode. All the connector and
panel drivers that use omap_dss_device also store the videomode in their
own panel_drv_data structures. There's no need to duplicate, remove the
videomode field from omap_dss_device.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Constify omap_dss_driver operations structure
Laurent Pinchart [Wed, 14 Feb 2018 01:12:37 +0000 (03:12 +0200)]
drm/omap: dss: Constify omap_dss_driver operations structure

The structure contains function pointers that don't need to be modified.
Make all its instances const to improve security.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove unused omapdss_default_get_timings()
Laurent Pinchart [Wed, 14 Feb 2018 01:26:52 +0000 (03:26 +0200)]
drm/omap: dss: Remove unused omapdss_default_get_timings()

All omap_dss_driver instances provide the get_timings operation. Remove
the default function.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove DSS encoders get_timings operation
Laurent Pinchart [Tue, 27 Feb 2018 18:29:21 +0000 (20:29 +0200)]
drm/omap: dss: Remove DSS encoders get_timings operation

The get_timings operation from DSS encoders (not to be confused with the
identically named operation in omap_dss_driver) is never called. Remove
it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove omapdss_atv_ops get_wss and set_wss operations
Laurent Pinchart [Wed, 14 Feb 2018 00:22:42 +0000 (02:22 +0200)]
drm/omap: dss: Remove omapdss_atv_ops get_wss and set_wss operations

The operations are never used, remove them. If the need to set wide
screen signaling data arises later, it should be implemented by
extending the DRM bridge API.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Handle DPI and SDI port initialization failures
Laurent Pinchart [Wed, 7 Mar 2018 18:34:42 +0000 (20:34 +0200)]
drm/omap: dss: Handle DPI and SDI port initialization failures

The dpi_init_port() and sdi_init_port() functions can return errors but
their return value is ignored. This prevents both probe failures and
probe deferral from working correctly. Propagate the errors up the call
stack.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Move platform_device_register from core.c to dss.c probe
Jyri Sarha [Fri, 16 Feb 2018 11:25:07 +0000 (13:25 +0200)]
drm/omap: dss: Move platform_device_register from core.c to dss.c probe

Register the omapdrm device when we know that dss device probe going
to succeed. This avoids DSS6 and DSS2 omapdrm device registration from
colliding with each other.

Signed-off-by: Jyri Sarha <jsarha@ti.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Gather OMAP DSS components at probe time
Laurent Pinchart [Thu, 1 Mar 2018 19:51:43 +0000 (21:51 +0200)]
drm/omap: dss: Gather OMAP DSS components at probe time

The omapdss_gather_components() function walks the OF graph to create a
list of all components part of the display device. There's no need to
delay this operation until DSS bind time as we have all the information
we need at probe time.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: dss: Remove display ordering from dss/display.c
Peter Ujfalusi [Mon, 12 Feb 2018 09:44:40 +0000 (11:44 +0200)]
drm/omap: dss: Remove display ordering from dss/display.c

As ordering of the dss_devices based on DT aliases is now implemented in
omap_drm.c, there is no need to do the ordering in dss/display.c
anymore.

At the same time remove the alias member of the omap_dss_device struct
since it is no longer needed. The only place it was used is in the
omapdss_register_display() function.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Do dss_device (display) ordering in omap_drv.c
Peter Ujfalusi [Mon, 12 Feb 2018 09:44:39 +0000 (11:44 +0200)]
drm/omap: Do dss_device (display) ordering in omap_drv.c

Sort the dssdev array based on DT aliases.

With this change we can remove the panel ordering from dss/display.c and
have all sorting related to dssdevs in one place.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Manage the usable omap_dss_device list within omap_drm_private
Peter Ujfalusi [Mon, 12 Feb 2018 09:44:37 +0000 (11:44 +0200)]
drm/omap: Manage the usable omap_dss_device list within omap_drm_private

Instead of reaching back to DSS to iterate through the dss_devices every
time, use an internal array where we store the available and usable
dss_devices.

At the same time remove the omapdss_device_is_connected() check from
omap_modeset_init() as it became irrelevant: We are not adding dssdevs
if their connect failed.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agodrm/omap: Allocate drm_device earlier and unref it as last step
Peter Ujfalusi [Mon, 12 Feb 2018 09:44:36 +0000 (11:44 +0200)]
drm/omap: Allocate drm_device earlier and unref it as last step

If we allocate the drm_device earlier we can just return the error code
without the need to use goto.
Do the unref of the drm_device as a last step when cleaning up. This will
make the drm_device available longer for us and makes sure that we only
free up the memory when all other cleanups have been already done.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Sebastian Reichel <sebastian.reichel@collabora.co.uk>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
6 years agoLinux 4.19-rc1
Linus Torvalds [Sun, 26 Aug 2018 21:11:59 +0000 (14:11 -0700)]
Linux 4.19-rc1

6 years agoMerge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 20:39:05 +0000 (13:39 -0700)]
Merge branch 'timers-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull timer update from Thomas Gleixner:
 "New defines for the compat time* types so they can be shared between
  32bit and 64bit builds. Not used yet, but merging them now allows the
  actual conversions to be merged through different maintainer trees
  without dependencies

  We still have compat interfaces for 32bit on 64bit even with the new
  2038 safe timespec/val variants because pointer size is different. And
  for the old style timespec/val interfaces we need yet another 'compat'
  interface for both 32bit native and 32bit on 64bit"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  y2038: Provide aliases for compat helpers

6 years agoMerge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax
Linus Torvalds [Sun, 26 Aug 2018 18:48:42 +0000 (11:48 -0700)]
Merge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax

Pull IDA updates from Matthew Wilcox:
 "A better IDA API:

      id = ida_alloc(ida, GFP_xxx);
      ida_free(ida, id);

  rather than the cumbersome ida_simple_get(), ida_simple_remove().

  The new IDA API is similar to ida_simple_get() but better named.  The
  internal restructuring of the IDA code removes the bitmap
  preallocation nonsense.

  I hope the net -200 lines of code is convincing"

* 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax: (29 commits)
  ida: Change ida_get_new_above to return the id
  ida: Remove old API
  test_ida: check_ida_destroy and check_ida_alloc
  test_ida: Convert check_ida_conv to new API
  test_ida: Move ida_check_max
  test_ida: Move ida_check_leaf
  idr-test: Convert ida_check_nomem to new API
  ida: Start new test_ida module
  target/iscsi: Allocate session IDs from an IDA
  iscsi target: fix session creation failure handling
  drm/vmwgfx: Convert to new IDA API
  dmaengine: Convert to new IDA API
  ppc: Convert vas ID allocation to new IDA API
  media: Convert entity ID allocation to new IDA API
  ppc: Convert mmu context allocation to new IDA API
  Convert net_namespace to new IDA API
  cb710: Convert to new IDA API
  rsxx: Convert to new IDA API
  osd: Convert to new IDA API
  sd: Convert to new IDA API
  ...

6 years agoMerge tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:41:08 +0000 (11:41 -0700)]
Merge tag 'gcc-plugins-v4.19-rc1-fix' of git://git./linux/kernel/git/kees/linux

Pull gcc plugin fix from Kees Cook:
 "Lift gcc test into Kconfig. This is for better behavior when the
  kernel is built with Clang, reported by Stefan Agner"

* tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  gcc-plugins: Disable when building under Clang

6 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:25:21 +0000 (11:25 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf updates from Thomas Gleixner:
 "Kernel:
   - Improve kallsyms coverage
   - Add x86 entry trampolines to kcore
   - Fix ARM SPE handling
   - Correct PPC event post processing

  Tools:
   - Make the build system more robust
   - Small fixes and enhancements all over the place
   - Update kernel ABI header copies
   - Preparatory work for converting libtraceevnt to a shared library
   - License cleanups"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (100 commits)
  tools arch: Update arch/x86/lib/memcpy_64.S copy used in 'perf bench mem memcpy'
  tools arch x86: Update tools's copy of cpufeatures.h
  perf python: Fix pyrf_evlist__read_on_cpu() interface
  perf mmap: Store real cpu number in 'struct perf_mmap'
  perf tools: Remove ext from struct kmod_path
  perf tools: Add gzip_is_compressed function
  perf tools: Add lzma_is_compressed function
  perf tools: Add is_compressed callback to compressions array
  perf tools: Move the temp file processing into decompress_kmodule
  perf tools: Use compression id in decompress_kmodule()
  perf tools: Store compression id into struct dso
  perf tools: Add compression id into 'struct kmod_path'
  perf tools: Make is_supported_compression() static
  perf tools: Make decompress_to_file() function static
  perf tools: Get rid of dso__needs_decompress() call in __open_dso()
  perf tools: Get rid of dso__needs_decompress() call in symbol__disassemble()
  perf tools: Get rid of dso__needs_decompress() call in read_object_code()
  tools lib traceevent: Change to SPDX License format
  perf llvm: Allow passing options to llc in addition to clang
  perf parser: Improve error message for PMU address filters
  ...

6 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 17:13:21 +0000 (10:13 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:

 - Correct the L1TF fallout on 32bit and the off by one in the 'too much
   RAM for protection' calculation.

 - Add a helpful kernel message for the 'too much RAM' case

 - Unbreak the VDSO in case that the compiler desides to use indirect
   jumps/calls and emits retpolines which cannot be resolved because the
   kernel uses its own thunks, which does not work for the VDSO. Make it
   use the builtin thunks.

 - Re-export start_thread() which was unexported when the 32/64bit
   implementation was unified. start_thread() is required by modular
   binfmt handlers.

 - Trivial cleanups

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/speculation/l1tf: Suggest what to do on systems with too much RAM
  x86/speculation/l1tf: Fix off-by-one error when warning that system has too much RAM
  x86/kvm/vmx: Remove duplicate l1d flush definitions
  x86/speculation/l1tf: Fix overflow in l1tf_pfn_limit() on 32bit
  x86/process: Re-export start_thread()
  x86/mce: Add notifier_block forward declaration
  x86/vdso: Fix vDSO build if a retpoline is emitted

6 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 16:55:28 +0000 (09:55 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq update from Thomas Gleixner:
 "A small set of updats/fixes for the irq subsystem:

   - Allow GICv3 interrupts to be configured as wake-up sources to
     enable wakeup from suspend

   - Make the error handling of the STM32 irqchip init function work

   - A set of small cleanups and improvements"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic-v3: Allow interrupt to be configured as wake-up sources
  irqchip/tango: Set irq handler and data in one go
  dt-bindings: irqchip: renesas-irqc: Document r8a774a1 support
  irqchip/s3c24xx: Remove unneeded comparison of unsigned long to 0
  irqchip/stm32: Fix init error handling
  irqchip/bcm7038-l1: Hide cpu offline callback when building for !SMP

6 years agoMerge branch 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 16:47:00 +0000 (09:47 -0700)]
Merge branch 'locking-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull licking update from Thomas Gleixner:
 "Mark the switch cases which fall through to the next case with the
  proper comment so the fallthrough compiler checks can be enabled"

* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  futex: Mark expected switch fall-throughs

6 years agoMerge tag 'libnvdimm-for-4.19_dax-memory-failure' of gitolite.kernel.org:pub/scm...
Linus Torvalds [Sun, 26 Aug 2018 01:43:59 +0000 (18:43 -0700)]
Merge tag 'libnvdimm-for-4.19_dax-memory-failure' of gitolite.pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm memory-failure update from Dave Jiang:
 "As it stands, memory_failure() gets thoroughly confused by dev_pagemap
  backed mappings. The recovery code has specific enabling for several
  possible page states and needs new enabling to handle poison in dax
  mappings.

  In order to support reliable reverse mapping of user space addresses:

   1/ Add new locking in the memory_failure() rmap path to prevent races
      that would typically be handled by the page lock.

   2/ Since dev_pagemap pages are hidden from the page allocator and the
      "compound page" accounting machinery, add a mechanism to determine
      the size of the mapping that encompasses a given poisoned pfn.

   3/ Given pmem errors can be repaired, change the speculatively
      accessed poison protection, mce_unmap_kpfn(), to be reversible and
      otherwise allow ongoing access from the kernel.

  A side effect of this enabling is that MADV_HWPOISON becomes usable
  for dax mappings, however the primary motivation is to allow the
  system to survive userspace consumption of hardware-poison via dax.
  Specifically the current behavior is:

     mce: Uncorrected hardware memory error in user-access at af34214200
     {1}[Hardware Error]: It has been corrected by h/w and requires no further action
     mce: [Hardware Error]: Machine check events logged
     {1}[Hardware Error]: event severity: corrected
     Memory failure: 0xaf34214: reserved kernel page still referenced by 1 users
     [..]
     Memory failure: 0xaf34214: recovery action for reserved kernel page: Failed
     mce: Memory error not recovered
     <reboot>

  ...and with these changes:

     Injecting memory failure for pfn 0x20cb00 at process virtual address 0x7f763dd00000
     Memory failure: 0x20cb00: Killing dax-pmd:5421 due to hardware memory corruption
     Memory failure: 0x20cb00: recovery action for dax page: Recovered

  Given all the cross dependencies I propose taking this through
  nvdimm.git with acks from Naoya, x86/core, x86/RAS, and of course dax
  folks"

* tag 'libnvdimm-for-4.19_dax-memory-failure' of gitolite.kernel.org:pub/scm/linux/kernel/git/nvdimm/nvdimm:
  libnvdimm, pmem: Restore page attributes when clearing errors
  x86/memory_failure: Introduce {set, clear}_mce_nospec()
  x86/mm/pat: Prepare {reserve, free}_memtype() for "decoy" addresses
  mm, memory_failure: Teach memory_failure() about dev_pagemap pages
  filesystem-dax: Introduce dax_lock_mapping_entry()
  mm, memory_failure: Collect mapping size in collect_procs()
  mm, madvise_inject_error: Let memory_failure() optionally take a page reference
  mm, dev_pagemap: Do not clear ->mapping on final put
  mm, madvise_inject_error: Disable MADV_SOFT_OFFLINE for ZONE_DEVICE pages
  filesystem-dax: Set page->index
  device-dax: Set page->index
  device-dax: Enable page_mapping()
  device-dax: Convert to vmf_insert_mixed and vm_fault_t

6 years agoMerge tag 'libnvdimm-for-4.19_misc' of gitolite.kernel.org:pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 26 Aug 2018 01:13:10 +0000 (18:13 -0700)]
Merge tag 'libnvdimm-for-4.19_misc' of gitolite.pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm updates from Dave Jiang:
 "Collection of misc libnvdimm patches for 4.19 submission:

   - Adding support to read locked nvdimm capacity.

   - Change test code to make DSM failure code injection an override.

   - Add support for calculate maximum contiguous area for namespace.

   - Add support for queueing a short ARS when there is on going ARS for
     nvdimm.

   - Allow NULL to be passed in to ->direct_access() for kaddr and pfn
     params.

   - Improve smart injection support for nvdimm emulation testing.

   - Fix test code that supports for emulating controller temperature.

   - Fix hang on error before devm_memremap_pages()

   - Fix a bug that causes user memory corruption when data returned to
     user for ars_status.

   - Maintainer updates for Ross Zwisler emails and adding Jan Kara to
     fsdax"

* tag 'libnvdimm-for-4.19_misc' of gitolite.kernel.org:pub/scm/linux/kernel/git/nvdimm/nvdimm:
  libnvdimm: fix ars_status output length calculation
  device-dax: avoid hang on error before devm_memremap_pages()
  tools/testing/nvdimm: improve emulation of smart injection
  filesystem-dax: Do not request kaddr and pfn when not required
  md/dm-writecache: Don't request pointer dummy_addr when not required
  dax/super: Do not request a pointer kaddr when not required
  tools/testing/nvdimm: kaddr and pfn can be NULL to ->direct_access()
  s390, dcssblk: kaddr and pfn can be NULL to ->direct_access()
  libnvdimm, pmem: kaddr and pfn can be NULL to ->direct_access()
  acpi/nfit: queue issuing of ars when an uc error notification comes in
  libnvdimm: Export max available extent
  libnvdimm: Use max contiguous area for namespace size
  MAINTAINERS: Add Jan Kara for filesystem DAX
  MAINTAINERS: update Ross Zwisler's email address
  tools/testing/nvdimm: Fix support for emulating controller temperature
  tools/testing/nvdimm: Make DSM failure code injection an override
  acpi, nfit: Prefer _DSM over _LSR for namespace label reads
  libnvdimm: Introduce locked DIMM capacity support

6 years agoMerge tag 'armsoc-late' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Linus Torvalds [Sat, 25 Aug 2018 21:12:36 +0000 (14:12 -0700)]
Merge tag 'armsoc-late' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC late updates from Olof Johansson:
 "A couple of late-merged changes that would be useful to get in this
  merge window:

   - Driver support for reset of audio complex on Meson platforms. The
     audio driver went in this merge window, and these changes have been
     in -next for a while (just not in our tree).

   - Power management fixes for IOMMU on Rockchip platforms, getting
     closer to kexec working on them, including Chromebooks.

   - Another pass updating "arm,psci" -> "psci" for some properties that
     have snuck in since last time it was done"

* tag 'armsoc-late' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  iommu/rockchip: Move irq request past pm_runtime_enable
  iommu/rockchip: Handle errors returned from PM framework
  arm64: rockchip: Force CONFIG_PM on Rockchip systems
  ARM: rockchip: Force CONFIG_PM on Rockchip systems
  arm64: dts: Fix various entry-method properties to reflect documentation
  reset: imx7: Fix always writing bits as 0
  reset: meson: add meson audio arb driver
  reset: meson: add dt-bindings for meson-axg audio arb

6 years agoMerge tag 'kbuild-v4.19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy...
Linus Torvalds [Sat, 25 Aug 2018 20:40:38 +0000 (13:40 -0700)]
Merge tag 'kbuild-v4.19-2' of git://git./linux/kernel/git/masahiroy/linux-kbuild

Pull more Kbuild updates from Masahiro Yamada:

 - add build_{menu,n,g,x}config targets for compile-testing Kconfig

 - fix and improve recursive dependency detection in Kconfig

 - fix parallel building of menuconfig/nconfig

 - fix syntax error in clang-version.sh

 - suppress distracting log from syncconfig

 - remove obsolete "rpm" target

 - remove VMLINUX_SYMBOL(_STR) macro entirely

 - fix microblaze build with CONFIG_DYNAMIC_FTRACE

 - move compiler test for dead code/data elimination to Kconfig

 - rename well-known LDFLAGS variable to KBUILD_LDFLAGS

 - misc fixes and cleanups

* tag 'kbuild-v4.19-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  kbuild: rename LDFLAGS to KBUILD_LDFLAGS
  kbuild: pass LDFLAGS to recordmcount.pl
  kbuild: test dead code/data elimination support in Kconfig
  initramfs: move gen_initramfs_list.sh from scripts/ to usr/
  vmlinux.lds.h: remove stale <linux/export.h> include
  export.h: remove VMLINUX_SYMBOL() and VMLINUX_SYMBOL_STR()
  Coccinelle: remove pci_alloc_consistent semantic to detect in zalloc-simple.cocci
  kbuild: make sorting initramfs contents independent of locale
  kbuild: remove "rpm" target, which is alias of "rpm-pkg"
  kbuild: Fix LOADLIBES rename in Documentation/kbuild/makefiles.txt
  kconfig: suppress "configuration written to .config" for syncconfig
  kconfig: fix "Can't open ..." in parallel build
  kbuild: Add a space after `!` to prevent parsing as file pattern
  scripts: modpost: check memory allocation results
  kconfig: improve the recursive dependency report
  kconfig: report recursive dependency involving 'imply'
  kconfig: error out when seeing recursive dependency
  kconfig: add build-only configurator targets
  scripts/dtc: consolidate include path options in Makefile

6 years agoMerge tag 'for-linus-20180825' of git://git.kernel.dk/linux-block
Linus Torvalds [Sat, 25 Aug 2018 20:30:23 +0000 (13:30 -0700)]
Merge tag 'for-linus-20180825' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A few small fixes for this merge window:

   - Locking imbalance fix for bcache (Shan Hai)

   - A few small fixes for wbt. One is a cleanup/prep, one is a fix for
     an existing issue, and the last two are fixes for changes that went
     into this merge window (me)"

* tag 'for-linus-20180825' of git://git.kernel.dk/linux-block:
  blk-wbt: don't maintain inflight counts if disabled
  blk-wbt: fix has-sleeper queueing check
  blk-wbt: use wq_has_sleeper() for wq active check
  blk-wbt: move disable check into get_limit()
  bcache: release dc->writeback_lock properly in bch_writeback_thread()

6 years agoMerge tag 'upstream-4.19-rc1-fix' of git://git.infradead.org/linux-ubifs
Linus Torvalds [Sat, 25 Aug 2018 20:27:35 +0000 (13:27 -0700)]
Merge tag 'upstream-4.19-rc1-fix' of git://git.infradead.org/linux-ubifs

Pull UBIFS fix from Richard Weinberger:
 "Remove an empty file from UBIFS source"

* tag 'upstream-4.19-rc1-fix' of git://git.infradead.org/linux-ubifs:
  ubifs: Remove empty file.h

6 years agoMerge tag '4.19-rc-smb3' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Sat, 25 Aug 2018 20:17:53 +0000 (13:17 -0700)]
Merge tag '4.19-rc-smb3' of git://git.samba.org/sfrench/cifs-2.6

Pull cifs fixes from Steve French:
 "Three small SMB3 fixes, one for stable"

* tag '4.19-rc-smb3' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: update internal module version number for cifs.ko to 2.12
  cifs: check kmalloc before use
  cifs: check if SMB2 PDU size has been padded and suppress the warning
  cifs: create a define for how many iovs we need for an SMB2_open()

6 years agomm/cow: don't bother write protecting already write-protected pages
Linus Torvalds [Mon, 9 Jul 2018 20:19:49 +0000 (13:19 -0700)]
mm/cow: don't bother write protecting already write-protected pages

This is not normally noticeable, but repeated forks are unnecessarily
expensive because they repeatedly dirty the parent page tables during
the page table copy operation.

It's trivial to just avoid write protecting the page table entry if it
was already not writable.

This patch was inspired by

    https://bugzilla.kernel.org/show_bug.cgi?id=200447

which points to an ancient "waste time re-doing fork" issue in the
presence of lots of signals.

That bug was fixed by Eric Biederman's signal handling series
culminating in commit c3ad2c3b02e9 ("signal: Don't restart fork when
signals come in"), but the unnecessary work for repeated forks is still
work just fixing, particularly since the fix is trivial.

Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>