Support private objects for stolen memory in psb_gem_create() and
convert users to psb_gem_create(). For stolen memory, psb_gem_create()
now initializes the GEM object via drm_gem_private_object_init().
In the fbdev setup, replace the open-coded initialization of struct
gtt_range with a call to psb_gem_create(). Use drm_gem_object_put()
for release.
In the cursor setup, use psb_gem_create() and get a real GEM object.
Previously the allocated instance of struct gtt_range was only partially
initialized. Release the cursor GEM object in gma_crtc_destroy(). The
release was missing from the original code.
With the conversion of all callers to psb_gem_create(), the extern
declarations of psb_gtt_alloc_range, psb_gtt_free_range and
psb_gem_object_func are not required any longer. Declare them as
static.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211015084053.13708-5-tzimmermann@suse.de
Since commit 43c2de1002 ("drm/rockchip: dsi: move all lane config except
LCDC mux to bind()"), we perform most HW configuration in the bind()
function. This configuration may be lost on suspend/resume, so we
need to call it again. That may lead to errors like this after system
suspend/resume:
dw-mipi-dsi-rockchip ff968000.mipi: failed to write command FIFO
panel-kingdisplay-kd097d04 ff960000.mipi.0: failed write init cmds: -110
Tested on Acer Chromebook Tab 10 (RK3399 Gru-Scarlet).
Note that early mailing list versions of this driver borrowed Rockchip's
downstream/BSP solution, to do HW configuration in mode_set() (which
*is* called at the appropriate pre-enable() times), but that was
discarded along the way. I've avoided that still, because mode_set()
documentation doesn't suggest this kind of purpose as far as I can tell.
Fixes: 43c2de1002 ("drm/rockchip: dsi: move all lane config except LCDC mux to bind()")
Cc: <stable@vger.kernel.org>
Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Tested-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20210928143413.v3.2.I4e9d93aadb00b1ffc7d506e3186a25492bf0b732@changeid
In commit 43c2de1002 ("drm/rockchip: dsi: move all lane config except
LCDC mux to bind()"), we moved most HW configuration to bind(), but we
didn't move the runtime PM management. Therefore, depending on initial
boot state, runtime-PM workqueue delays, and other timing factors, we
may disable our power domain in between the hardware configuration
(bind()) and when we enable the display. This can cause us to lose
hardware state and fail to configure our display. For example:
dw-mipi-dsi-rockchip ff968000.mipi: failed to write command FIFO
panel-innolux-p079zca ff960000.mipi.0: failed to write command 0
or:
dw-mipi-dsi-rockchip ff968000.mipi: failed to write command FIFO
panel-kingdisplay-kd097d04 ff960000.mipi.0: failed write init cmds: -110
We should match the runtime PM to the lifetime of the bind()/unbind()
cycle.
Tested on Acer Chrometab 10 (RK3399 Gru-Scarlet), with panel drivers
built either as modules or built-in.
Side notes: it seems one is more likely to see this problem when the
panel driver is built into the kernel. I've also seen this problem
bisect down to commits that simply changed Kconfig dependencies, because
it changed the order in which driver init functions were compiled into
the kernel, and therefore the ordering and timing of built-in device
probe.
Fixes: 43c2de1002 ("drm/rockchip: dsi: move all lane config except LCDC mux to bind()")
Link: https://lore.kernel.org/linux-rockchip/9aedfb528600ecf871885f7293ca4207c84d16c1.camel@gmail.com/
Reported-by: <aleksandr.o.makarov@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Brian Norris <briannorris@chromium.org>
Tested-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20210928143413.v3.1.Ic2904d37f30013a7f3d8476203ad3733c186827e@changeid
Current code always sets reset line low in .pre_enable callback and
holds it low for 10ms. This is sub-optimal and increases the time
between enablement of the DSI83 and valid LVDS clock.
Rework the reset handling such that the reset line is held low for 10ms
both in probe() of the driver and .disable callback, which guarantees
that the reset line was always held low for more than 10ms and therefore
the reset line timing requirement is satisfied. Furthermore, move the
reset handling into .enable callback so the entire DSI83 initialization
is now in one place.
This reduces DSI83 enablement delay by up to 10ms.
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Robert Foss <robert.foss@linaro.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20211016210402.171595-1-marex@denx.de
I got a null-ptr-deref report:
[drm:drm_dev_init [drm]] *ERROR* Cannot allocate anonymous inode: -12
==================================================================
BUG: KASAN: null-ptr-deref in iput+0x3c/0x4a0
...
Call Trace:
dump_stack_lvl+0x6c/0x8b
kasan_report.cold+0x64/0xdb
__asan_load8+0x69/0x90
iput+0x3c/0x4a0
drm_dev_init_release+0x39/0xb0 [drm]
drm_managed_release+0x158/0x2d0 [drm]
drm_dev_init+0x3a7/0x4c0 [drm]
__devm_drm_dev_alloc+0x55/0xd0 [drm]
mi0283qt_probe+0x8a/0x2b5 [mi0283qt]
spi_probe+0xeb/0x130
...
entry_SYSCALL_64_after_hwframe+0x44/0xae
If drm_fs_inode_new() fails in drm_dev_init(), dev->anon_inode will point
to PTR_ERR(...) instead of NULL. This will result in null-ptr-deref when
drm_fs_inode_free(dev->anon_inode) is called.
drm_dev_init()
drm_fs_inode_new() // fail, dev->anon_inode = PTR_ERR(...)
drm_managed_release()
drm_dev_init_release()
drm_fs_inode_free() // access non-existent anon_inode
Define a temp variable and assign it to dev->anon_inode if the temp
variable is not PTR_ERR.
Fixes: 2cbf7fc671 ("drm: Use drmm_ for drm_dev_init cleanup")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20211013114139.4042207-1-wanghai38@huawei.com
The model and make of the LCD panel of the Vivax TPC-9150 is unknown,
hence the panel settings that were retrieved with a FEX dump are named
after the device NOT the actual panel.
The LCD in question is a 50 pin MISO TFT LCD panel of the resolution
1024x600 used by the aforementioned device.
Version 2, as Thierry kindly suggested that I fix the order in which the
panel was ordered compared to others.
Version 3, filling in the required info suggested by Sam. Plus some
factual issues that I've corrected myself (tested working)
Version 4, rearranged the display parameters and fix invalid bit format
issue. (Thanks Sam)
Version 5, referred to FEX file instead of manual debugging for
information.
Version 6, same as above. This time, it'll be documented.
A bit of context first: I experimented with this a long time ago whilst
I was first learning how to get Linux running on Allwinner boards, I
didn't have many resources at hand so this was quite slow. Anyways, I
stumbled upon this guide (https://linux-sunxi.org/LCD) and was reading
about how to setup the LCD for my tablet. Since I was able to make a
proper FEX dump, I was also able to read the correct parameters for
myself without relying on leaked documents or part numbers and whatnot.
In the FEX dump the value lcd_frm IS SET to 1, which means, at least
according to the document, that this display is INDEED an 18 bit per
pixel panel. Compiling U-Boot and seeing the tux in proper colors
confirmed this. As per Sam Ravnborg's suggestion, I've changed the panel
to his format "MEDIA_BUS_FMT_RGB666_1X7X3_SPWG", however this does not
lead to any actual change in regards to the functionality since the sunxi
panel driver just ignores this value. However, hopefully this clears up
any errors down the road as either the driver becomes advanced enough to
not ignore this value or that some other piece of software relies on
this value being known. PS: Apologies to the maintainers that have to
endure my misjudgement about how these things work.
As for the concerns about a single patch series, I wasn't sure where to
send the patches as they clearly aren't dt-bindings related and my
previous patches have ended up in drm-misc-fixes anyway. So I'm guessing
I'll be fine if I just post them in the list from last time???
Signed-off-by: Nikola Pavlica <pavlica.nikola@gmail.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20211011212731.77763-1-pavlica.nikola@gmail.com
Add 2 drm_connector privacy-screen helper functions:
1. drm_connector_attach_privacy_screen_provider(), this function creates
and attaches the standard privacy-screen properties and registers a
generic notifier for generating sysfs-connector-status-events on external
changes to the privacy-screen status.
2. drm_connector_update_privacy_screen(), update the privacy-screen's
sw_state if the connector has a privacy-screen.
Changes in v2:
- Do not update connector->state->privacy_screen_sw_state on
atomic-commits.
- Change drm_connector_update_privacy_screen() to take drm_connector_state
as argument instead of a full drm_atomic_state. This allows the helper
to be called by drivers when they are enabling crtcs/encoders/connectors.
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005202322.700909-6-hdegoede@redhat.com
Add support for privacy-screen consumers to register a notifier to
be notified of external (e.g. done by the hw itself on a hotkey press)
state changes.
Changes in v2:
- Drop WARN_ON(mutex_is_locked(&priv->lock)) check in
drm_privacy_screen_call_notifier_chain() it may be locked by
another thread, which would lead to a false-positive triggering
of the check
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005202322.700909-5-hdegoede@redhat.com
On some new laptops the LCD panel has a builtin electronic privacy-screen.
We want to export this functionality as a property on the drm connector
object. But often this functionality is not exposed on the GPU but on some
other (ACPI) device.
This commit adds a privacy-screen class allowing the driver for these
other devices to register themselves as a privacy-screen provider; and
allowing the drm/kms code to get a privacy-screen provider associated
with a specific GPU/connector combo.
Changes in v2:
- Make CONFIG_DRM_PRIVACY_SCREEN a bool which controls if the drm_privacy
code gets built as part of the main drm module rather then making it
a tristate which builds its own module.
- Add a #if IS_ENABLED(CONFIG_DRM_PRIVACY_SCREEN) check to
drm_privacy_screen_consumer.h and define stubs when the check fails.
Together these 2 changes fix several dependency issues.
- Remove module related code now that this is part of the main drm.ko
- Use drm_class as class for the privacy-screen devices instead of
adding a separate class for this
Changes in v3:
- Make the static inline drm_privacy_screen_get_state() stub set sw_state
and hw_state to PRIVACY_SCREEN_DISABLED to squelch an uninitialized
variable warning when CONFIG_DRM_PRIVICAY_SCREEN is not set
Changes in v4:
- Make drm_privacy_screen_set_sw_state() skip calling out to the hw if
hw_state == new_sw_state
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005202322.700909-3-hdegoede@redhat.com
Add support for generic electronic privacy screen properties, that
can be added by systems that have an integrated EPS.
Changes in v2 (Hans de Goede)
- Create 2 properties, "privacy-screen sw-state" and
"privacy-screen hw-state", to deal with devices where the OS might be
locked out of making state changes
- Write kerneldoc explaining how the 2 properties work together, what
happens when changes to the state are made outside of the DRM code's
control, etc.
Changes in v3 (Hans de Goede)
- Some small tweaks to the kerneldoc describing the 2 properties
Changes in v4 (Hans de Goede)
- Change the "Enabled, locked" and "Disabled, locked" hw-state enum value
names to "Enabled-locked" and "Disabled-locked". The xrandr command shows
all possible enum values separated by commas in its output, so having a
comma in an enum name is not a good idea.
- Do not add a privacy_screen_hw_state member to drm_connector_state
since this property is immutable its value must be directly stored in the
obj->properties->values array
Signed-off-by: Rajat Jain <rajatja@google.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.com>
Reviewed-by: Mario Limonciello <Mario.limonciello@dell.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Co-developed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211005202322.700909-2-hdegoede@redhat.com
If drm_modeset_lock() returns -EDEADLK, the caller is supposed to drop
all currently held locks using drm_modeset_backoff(). Failing to do so
will result in warnings and backtraces on the paths trying to lock a
contended lock. Add support for optionally printing the backtrace on the
path that hit the deadlock and didn't gracefully handle the situation.
For example, the patch [1] inadvertently dropped the return value check
and error return on replacing calc_watermark_data() with
intel_compute_global_watermarks(). The backtraces on the subsequent
locking paths hitting WARN_ON(ctx->contended) were unhelpful, but adding
the backtrace to the deadlock path produced this helpful printout:
<7> [98.002465] drm_modeset_lock attempting to lock a contended lock without backoff:
drm_modeset_lock+0x107/0x130
drm_atomic_get_plane_state+0x76/0x150
skl_compute_wm+0x251d/0x2b20 [i915]
intel_atomic_check+0x1942/0x29e0 [i915]
drm_atomic_check_only+0x554/0x910
drm_atomic_nonblocking_commit+0xe/0x50
drm_mode_atomic_ioctl+0x8c2/0xab0
drm_ioctl_kernel+0xac/0x140
Add new CONFIG_DRM_DEBUG_MODESET_LOCK to enable modeset lock debugging
with stack depot and trace.
[1] https://lore.kernel.org/r/20210924114741.15940-4-jani.nikula@intel.com
v2:
- default y if DEBUG_WW_MUTEX_SLOWPATH (Daniel)
- depends on DEBUG_KERNEL
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Dave Airlie <airlied@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211001091444.8177-1-jani.nikula@intel.com
