The display controller on Tegra can use syncpoints to count VBLANK
events. syncpoints are 32-bit unsigned integers, so well suited as
VBLANK counters.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Use a sized unsigned 32-bit data type (u32) to store register contents.
The SOR registers are 32 bits wide irrespective of the architecture's
data width.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The Tegra DRM driver uses a single IO virtual address space for buffer
mappings. Provide a table of the address space usage in debugfs.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Check that the desired parent clock is indeed a valid parent for the
display controller clock. This is purely cosmetic at this point since
the parent clocks are specified in DT and all the currently defined
parents are in fact valid parents of the display controller clock.
Signed-off-by: Thierry Reding <treding@nvidia.com>
We need to wait for all fences, not just the exclusive one.
Signed-off-by: Christian König <christian.koenig@amd.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We somehow try to free the SG table twice.
Bugs: https://bugs.freedesktop.org/show_bug.cgi?id=89734
Signed-off-by: Christian König <christian.koenig@amd.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When performing a modeset, use the framebuffer pitch value to set FIMD
IMG_SIZE and Mixer SPAN registers. These are both defined as pitch - the
distance between contiguous lines (bytes for FIMD, pixels for mixer).
Fixes display on Snow (1366x768).
Signed-off-by: Daniel Stone <daniels@collabora.com>
Tested-by: Javier Martinez Canillas <javier.martinez@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
The legcy colorkey ioctls are only implemented for sprite planes, so
reject the ioctl for primary/cursor planes. If we want to support
colorkeying with these planes (assuming we have hw support of course)
we should just move ahead with the colorkey property conversion.
Testcase: kms_legacy_colorkey
Cc: Tommi Rantala <tt.rantala@gmail.com>
Cc: stable@vger.kernel.org
Reference: http://mid.gmane.org/CA+ydwtr+bCo7LJ44JFmUkVRx144UDFgOS+aJTfK6KHtvBDVuAw@mail.gmail.com
Reported-and-tested-by: Tommi Rantala <tt.rantala@gmail.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
This change adds the support in mdp5 kms driver for single
and dual DSI. Dual DSI case depends on the framework API
and sequence change to support dual data path.
v1: Initial change
v2: Address Rob Clark's comment
- Separate command mode encoder to a new file mdp5_cmd_encoder.c
- Rebase to not depend on msm_drm_sub_dev change
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change adds the DSI connector support in msm drm driver.
v1: Initial change
v2:
- Address comments from Archit + minor clean-ups
- Rebase to not depend on msm_drm_sub_dev change [Rob's comment]
v3: Fix issues when initialization is failed
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change is to add an interface to MDP for connector devices
setting split display information.
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change is to make the content in construct_encoder reflect its
name.
Also, DSI connector may be connected to video mode or command mode
encoder, so that 2 different encoders need to be constructed for DSI.
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Normally these are called from within drm core, from the EDID parsing
code. But for dual-dsi in some drivers (at least drm/msm) we need to
call these from the driver. So they should be exported.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Acked-by: Dave Airlie <airlied@redhat.com>
This TODO can now be removed and replaced by the previous patch
"drm/msm/mdp5: Update headers (add CTL flush bits)"
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Some upcoming targets have more bits to set in CTL_FLUSH
registers.
Example: msm8x16 needs to set TIMING1 bit so that some of the
INTF1's interface registers get flushed.
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change adds the hw configuration for msm8x16 chipsets in
mdp5_cfg module.
Note that only one external display interface is present in this
configuration (DSI) but has not been enabled yet. It will be enabled
once drm/msm driver supports DSI connectors.
v2: add CTL flush register's hardware mask [pointed by Archit]
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
SMP blocks are configured for specific client IDs (ports).
These client IDs can be different from one chip to another for a
given pipe.
e.g.: DMA0 pipe fetch Y component is connected to:
- port #10 for MDP5 v1.3
- port #4 for MDP5 v1.6
In order to be compatible for upcoming versions of MDP5, the
client ID list is passed through the MDP5 config module rather
than using a list of hard-coded enum values.
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This patch contains the generated header file of the following
change "drm/msm/mdp5: Get SMP client list from mdp5_cfg".
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
MDP block is actually contained inside the MDSS block. For some
chipsets, the base address of the MDP registers is different from the
current (assumed) 0x100 offset.
Like CTL and LM blocks, this changes introduce a dynamic offset
for the MDP instance, which can be found out at runtime, once the
MDSS HW version is read.
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change contains the generated header file for the following
change "drm/msm/mdp5: Separate MDP5 domain from MDSS domain".
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
The width and height in SSPP_SRC_IMG_SIZE register should be the
size of the entire source framebuffer, not the fetch size.
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
This change adds the registers in mdp5 ping pong blocks
and split display control registers.
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Pingpong register base addresses are different across platforms.
This change adds this information to config table and initialize
the values for 8x74 and 8084.
Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Up until now, we assume that eDP is tight to intf_0 and HDMI to
intf_3. This information shall actually come from the mdp5_cfg
module since it can change from one chip to another.
v2: rename macro to mdp5_cfg_intf_is_virtual() [pointed by Archit]
v3: add sanity check before writing in INTF_TIMING_ENGINE_EN registers
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Some interfaces (WB, DSI Command Mode) need to be kicked off
through a START Signal. This signal needs to be sent at the right
time and requests in some cases to keep track of the pipeline
status (eg: whether pipeline registers are flushed AND output WB
buffers are ready, in case of WB interface).
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
DSI and WB interfaces need a more complex pipeline configuration
than the current mdp5_ctl_set_intf().
For example, memory output connections need to be selected for
WB. Interface mode (Video vs. Command modes) also need to be
configured for DSI.
This change takes care of configuring the whole pipeline as far
as operation mode goes. DSI and WB interfaces will be added
later.
v2: rename macro to mdp5_cfg_intf_is_virtual() [pointed by Archit]
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
[Remove temp bisectability hack -Rob]
Signed-off-by: Rob Clark <robdclark@gmail.com>
Prepare for pipeline operation mode configuration, in particular
for DSI and WB modes.
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
[Throw in a #define temporarily to keep things bisectable -Rob]
Signed-off-by: Rob Clark <robdclark@gmail.com>
MDP5 hardware has some limitation and requires to avoid flushing
registers more than once between two Vblanks.
This change removes all FLUSH operations (except for HW cursor)
beside the one coming from a CRTC's ->atomic_flush().
This avoid this type of behavior (eg: CRTC + 1 plane overlay):
[drm:mdp5_crtc_vblank_irq] vblank
[drm:mdp5_ctl_commit] flush (20048) CTL + LM0 + RGB0
[drm:mdp5_ctl_commit] flush (20040) CTL + LM0
[drm:mdp5_crtc_vblank_irq] blank
[drm:mdp5_ctl_commit] flush (20049) CTL + LM0 + RGB0 + VIG0
[drm:mdp5_crtc_vblank_irq] blank
and replaces it by:
[drm:mdp5_crtc_vblank_irq] vblank
[drm:mdp5_ctl_commit] flush (20048) CTL + LM0 + RGB0
[drm:mdp5_crtc_vblank_irq] blank
[drm:mdp5_ctl_commit] flush (20049) CTL + LM0 + RGB0 + VIG0
[drm:mdp5_crtc_vblank_irq] blank
Only *one* FLUSH is called between Vblanks interrupts.
Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Add support to use the VRAM carveout (if specified in dtb) for fbdev
scanout buffer. This allows drm/msm to take over a bootloader splash-
screen, and avoids corruption on screen that results if the kernel uses
memory that is still being scanned out for itself.
Signed-off-by: Rob Clark <robdclark@gmail.com>
We'll want to extend this a bit to handle also a reserved-memory
("stolen") region, so that drm/msm can take-over bootloader splash
screen. First split it out into it's own fxn to reduce noise in
the following patch.
Signed-off-by: Rob Clark <robdclark@gmail.com>
This patch implements the virtual GEM driver with PRIME sharing which
allows vgem to import a gem object from other drivers for the purpose
of mmap-ing them to userspace. The mmap is done using the mmap
operation exported by other drivers.
v2: remove platform_device and do not attach to dma bufs
v3: use drm helpers for get/put pages
v4: correct dumb create pitch
Reviewed-by: Rob Clark <robdclark@gmail.com> (v3)
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org> (v3)
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Zach Reizner <zachr@google.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
When pixel clock less than 148.5MHz, make sloopboost=2 tklvl=20
cklvl=13 increase rasing/falling time and increase data & clock
voltage driver.
When pixel clock less than 74.25MHz, make sloopboost=0 tklvl=19
cklvl=18, increase data and clock voltage driver.
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Because of iMX6 & Rockchip have differnet mpll config parameter,
the VLEVCTRL parameter would be different. In this case we should
separate VLEVCTRL setting from the common dw_hdmi driver, config
this parameter in platform driver(dw_hdmi-imx and dw_hdmi-rockchip)
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Using a local struct pointer to reduce one level of indirection
makes the code slightly more readable.
Signed-off-by: Yakir Yang <ykk@rock-chips.com>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Final drm-misc pull for 4.0, just various things all over, including a few
more important atomic fixes. btw I didn't pick up the vmwgfx patch from
Ville's series, but one patch has one hunk touching vmwgfx and
Thomas/Jakob didn't get around to ack it. I figured it's simple enough to
be ok though.
* tag 'topic/drm-misc-2015-03-31' of git://anongit.freedesktop.org/drm-intel:
drm: line wrap DRM_IOCTL_DEF* macros
drm/atomic: Don't try to free a NULL state
drm/atomic: Clear crtcs, connectors and planes when clearing state
drm: Rewrite drm_ioctl_flags() to resemble the new drm_ioctl() code
drm: Use max() to make the ioctl alloc size code cleaner
drm: Simplify core vs. drv ioctl handling
drm: Drop ioctl->cmd_drv
drm: Fix DRM_IOCTL_DEF_DRV()
drm/atomic-helpers: Properly avoid full modeset dance
drm: atomic: Allow setting CRTC active property
drm: atomic: Expose CRTC active property
drm: crtc_helper: Update hwmode before mode_set call
drm: mode: Allow NULL modes for equality check
drm: fb_helper: Simplify exit condition
drm: mode: Fix typo in kerneldoc
drm/dp: Print the number of bytes processed for aux nacks
This backmerges 4.0-rc6 due to the recent fixes in rc5/6
- DP link rate refactoring from Ville
- byt/bsw rps tuning from Chris
- kerneldoc for the shrinker code
- more dynamic ppgtt pte work (Michel, Ben, ...)
- vlv dpll code refactoring to prep fro bxt (Imre)
- refactoring the sprite colorkey code (Ville)
- rotated ggtt view support from Tvrtko
- roll out struct drm_atomic_state to prep for atomic update (Ander)
* tag 'drm-intel-next-2015-03-27-merge' of git://anongit.freedesktop.org/drm-intel: (473 commits)
Linux 4.0-rc6
arm64: juno: Fix misleading name of UART reference clock
drm/i915: Update DRIVER_DATE to 20150327
drm/i915: Skip allocating shadow batch for 0-length batches
drm/i915: Handle error to get connector state when staging config
drm/i915: Compare GGTT view structs instead of types
drm/i915: fix simple_return.cocci warnings
drm/i915: Add module param to test the load detect code
drm/i915: Remove usage of encoder->new_crtc from clock computations
drm/i915: Don't look at staged config crtc when changing DRRS state
drm/i915: Convert intel_pipe_will_have_type() to using atomic state
drm/i915: Pass an atomic state to modeset_global_resources() functions
drm/i915: Add dynamic page trace events
drm/i915: Finish gen6/7 dynamic page table allocation
drm/i915: Remove unnecessary gen6_ppgtt_unmap_pages
drm/i915: Fix i915_dma_map_single positive error code
drm/i915: Prevent out of range pt in gen6_for_each_pde
drm/i915: fix definition of the DRM_IOCTL_I915_GET_SPRITE_COLORKEY ioctl
drm/i915: Rip out GET_SPRITE_COLORKEY ioctl
watchdog: imgpdc: Fix default heartbeat
...
This patch makes the fsl,data-width and fsl,data-mapping device tree
properties optional if a panel is connected to the ldb output port
via the of_graph bindings. The data mapping is determined from the
display_info.bus_format field provided by the panel.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
The LDB driver changes the attached display interface's input clock mux
to the LDB_DI clock reference. Change it back again when disabling the
LVDS display. Changing back to the PLL5 for example allows to configure
the same DI for HDMI output later.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
This patch allows to optionally attach the lvds-channel to a panel
supported by a drm_panel driver using of-graph bindings, instead of
supplying the modes via display-timings in the device tree.
This depends on of_graph_get_port_by_id and uses the OF graph to
link the optional DRM panel to the LDB lvds-channel. The output
port number is 1 on devices without the 4-port input multiplexer
(i.MX5) and 4 on devices with the mux (i.MX6).
Before:
ldb {
...
lvds-channel@0 {
...
display-timings {
native-timing = <&timing1>;
timing1: etm0700g0dh6 {
hactive = <800>;
vactive = <480>;
clock-frequency = <33260000>;
hsync-len = <128>;
hback-porch = <88>;
hfront-porch = <40>;
vsync-len = <2>;
vback-porch = <33>;
vfront-porch = <10>;
hsync-active = <0>;
vsync-active = <0>;
...
};
};
...
};
};
After:
ldb {
...
lvds-channel@0 {
...
port@4 {
reg = <4>;
lvds_out: endpoint {
remote_endpoint = <&panel_in>;
};
};
};
};
panel {
compatible = "edt,etm0700g0dh6", "simple-panel";
...
port {
panel_in: endpoint {
remote-endpoint = <&lvds_out>;
};
};
};
[Fixed build error due to missing select on DRM_PANEL --rmk]
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
This patch consolidates the different interface_pix_fmt, pixel_fmt, pix_fmt,
and pixfmt variables to a common name "bus_format" wherever they describe the
pixel format on the bus between display controller and encoder hardware.
At the same time, it renames imx_drm_panel_format to imx_drm_set_bus_format.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Emil Renner Berthing <kernel@esmil.dk>
imx-drm internally misused the V4L2_PIX_FMT constants, which are supposed to
describe the pixel format of frame buffers in memory, to describe the pixel
format on the bus between the display controller and the encoder hardware.
Now that MEDIA_BUS_FMT constants are available to drm drivers, use those
instead.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Emil Renner Berthing <kernel@esmil.dk>
This patch allows the IPU to divide the 27 MHz input clock from
the TVE by two to obtain the 13.5 MHz pixel clock needed for
NTSC/PAL SD modes.
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Improve the readability and keeps the lines shorter than 80 columns.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
- Allow amdkfd to work with multiple kgd instances. This is in preparation for
AMD's new open source kernel graphic driver (amdgpu), and for the new
AMD APU, Carrizo.
- Convert timestamping to use 64bit time accessors
- Three other minor changes.
* tag 'drm-amdkfd-next-2015-03-25' of git://people.freedesktop.org/~gabbayo/linux:
drm/amdkfd: Add multiple kgd support
drm/amdkfd: Convert timestamping to use 64bit time accessors
drm/amdkfd: add debug prints for process teardown
drm/amdkfd: Remove unused field from struct qcm_process_device
drm/amdkfd: rename fence_wait_timeout
Patch derived from one from Yakir Yang. Yakir Yang says:
For Designerware HDMI, the following write sequence is recommended:
1. aud_n3 (set bit ncts_atomic_write if desired)
2. aud_cts3 (set CTS_manual and CTS value if desired/enabled)
3. aud_cts2 (required in CTS_manual)
4. aud_cts1 (required in CTS_manual)
5. aud_n3 (bit ncts_atomic_write with same value as in step 1.)
6. aud_n2
7. aud_n1
However, avoid the ncts_atomic_write_bit and CTS_manual settings in this
patch, both of which are marked reserved in the iMX6 documentation. All
iMX6 code in the wild seems to want CTS_manual cleared.
Having requested clarification from FSL, it appears that neither of
these bits are implemented in their version of the IP.
Acked-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The HDMI n/cts settings need to be updated whenever the audio sample
rate or the video pixel clock changes. This needs to be protected
against concurrency as there is no synchronisation between these two
operations. Introduce a mutex (called audio_mutex) to protect against
two threads trying to update the video clock rate and pixel clock
simultaneously.
Acked-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Combine these two functions into a single implementation. These two
functions are called consecutively anyway. Idea from a patch by
Yakir Yang.
Acked-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Acked-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Since
commit 17cabf571e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed Jan 14 11:20:57 2015 +0000
drm/i915: Trim the command parser allocations
we may then try to allocate a zero-sized object and attempt to extract
its pages. Understandably this fails.
Note that the real offender seems to be
commit b9ffd80ed6
Author: Brad Volkin <bradley.d.volkin@intel.com>
Date: Thu Dec 11 12:13:10 2014 -0800
drm/i915: Use batch length instead of object size in command parser
Testcase: igt/gem_exec_nop #ivb,byt,hsw
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[cherry picked from commit 743e78c1d7
from drm-intel-next because 4.0 seems to be affected by this too,
despite that the obvious culprit is definitely not in 4.0. Whatever,
if fixes a bug.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Consistently with other free functions, handle the NULL case without
oopsing.
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJVGHwjAAoJEHm+PkMAQRiG8rcIAJ6cEJ6mbqLpyz5XrGf4yNp0
+wG/QlEpT8rgrxe9wSjB3lfW3kR2Pe69b9fVVCdiklygdkmva5vfmDrVGGzYfe3M
QrFSSlMVBplvh6IiM/L1mVMtr3DSmCO23YZZ9R5b7FoEYatNHRpNWBCBpuXpd4aD
sLuIvO3L/S7LqeOAFkkYWv6AuL9umicmjR8u+nsmCSRJom7At/aJ6R66WIp9vxho
Rn7r6wcUk6B2Q/gYNjdSE8SIwdyKhuBGyvqQ9U9s6Btg9DQfM/b0vG5kw9hqeAq/
9445jqVDP1whA2vz6GjnvltidxrqRvuDPBwzOnFmY5U+KZz4lS3x2mnWAAJ3xWs=
=TqVJ
-----END PGP SIGNATURE-----
Merge tag 'v4.0-rc6' into drm-intel-next
Backmerge Linux 4.0-rc6 because conflicts are (again) getting out of
hand. To make sure we don't lose any bugfixes from the 4.0-rc5-rc6
flurry of patches we've applied them all to -next too.
Conflicts:
drivers/gpu/drm/i915/intel_display.c
Always take the version from -next, we've already handled all
conflicts with explicit cherrypicking.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Users of the atomic state assume that if the pointer to a crtc, plane or
connector is not NULL in the respective object vector, than the state
for that object in *_states vector also won't be NULL. That assumption
was broken by drm_atomic_state_clear(), which would clear the state
pointer but leave the pointer to the object still set.
This fixes a NULL pointer dereference in i915 caused by the use of
drm_atomic_state_clear().
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Use the same logic when checking for valid ioctl range in
drm_ioctl_flags() that is used in drm_ioctl() to avoid
confusion.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Use max() to make the code to determine the allocation size for
the ioctl data easier to read.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Now that cmd_drv is gone the handling for core and driver ioctls is
mostly identical, so eliminate the duplication. Also take the
opportunity to simplify the range checks to be less cluttered.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
ioctl->cmd_drv is pointless and we can just as well stick the full ioctl
definition into ioctl->cmd.
Cc: Jakob Bornecrantz <jakob@vmware.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since
commit 17cabf571e
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed Jan 14 11:20:57 2015 +0000
drm/i915: Trim the command parser allocations
we may then try to allocate a zero-sized object and attempt to extract
its pages. Understandably this fails.
Testcase: igt/gem_exec_nop #ivb,byt,hsw
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Otherwise the VCE firmware needs to be in the first 256MB of VRAM.
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Dumping is still possible if a ring isn't ready, only when it
isn't allocated at all we need to abort here.
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use readb() and memcpy_fromio() accessors instead.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
The return value of one of the calls to drm_atomic_get_connector_state()
in intel_modeset_stage_output_state() wasn't checked for errors.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
To allow for views where the view type is not defined by the view type only,
like it is in stereo or rotated 90 degree view, change the semantic to require
the whole view structure for comparison when we match a GGTT view.
This allows including parameters like offset to be included in the view which
is useful for eg. partial views.
v3:
- Rely on ggtt_view type being 0 for non-GGTT vma's, which equals to
I915_GGTT_VIEW_NORMAL. (Daniel Vetter)
- Do not use potentially slower comparison when we only want to know if
something is or is not a normal view.
- Rebase on top of rotated view patches. Add rotated view singleton.
- If one view is missing in comparison they're equal only if both are missing.
v4:
- Use comparison helper in obj_to_ggtt_view too. (Tvrtko Ursulin)
- Do WARN_ON if one view is NULL. (Tvrtko Ursulin)
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
drivers/gpu/drm/i915/i915_gem_gtt.c:1349:1-4: WARNING: end returns can be simpified and declaration on line 1347 can be dropped
Simplify a trivial if-return sequence. Possibly combine with a
preceding function call.
Generated by: scripts/coccinelle/misc/simple_return.cocci
CC: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Legacy setCrtc has a nice fastpath for just updating the frontbuffer
when the output routing doesn't change. Which I of course tried to
keep working, except that I fumbled the job: The helpers correctly
compute ->mode_changed, CRTC updates get correctly skipped but
connector functions are called unconditionally.
Fix this.
v2: For the disable sided connector->state->crtc might be NULL.
Instead look at the old_connector_state->crtc, but still use the new
crtc state for that old crtc. Reported by Thierry.
Reported-and-Tested-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk> (v1)
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Gustavo Padovan <gustavo.padovan@collabora.co.uk> (v1)
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
If the user supplies EDID through firmware or debugfs override, the
driver callbacks are bypassed and the connector ELD does not get
updated, and audio fails. Set ELD for firmware and debugfs EDIDs too.
There should be no harm in gratuitously doing this for non HDMI/DP
connectors, as it's still up to the driver to use the ELD, if any.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=82349
Reference: https://bugs.freedesktop.org/show_bug.cgi?id=80691
Reported-by: Emil <emilsvennesson@gmail.com>
Reported-by: Rob Engle <grenoble@gmail.com>
Tested-by: Jolan Luff <jolan@gormsby.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
This is useful for writing igts to make sure we don't break this,
without being forced to own a one of these dinosaurs.
Suggested-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Ander Conselvan de Oliveira <conselvan2@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Some of the crtc_compute_clock() still depended on encoder->new_crtc
since they didn't use intel_pipe_will_have_type() and used an open
coded version of that function instead. This patch replaces those with
the appropriate code that checks the atomic state intead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Separate the if (!connector) continue to facility easier
extraction of a loop iterator for all of these (there's lots more in
i915 and atomic helpers).]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The function intel_dp_set_drrs_state() would decide which pipe to
downclock based on the staged config for the given connector. However,
the result of that function is immediate, and it uses input values from
crtc->config, so it should be looking at the current crtc instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Pass a crtc_state to it and find whether the pipe has an encoder of a
given type by looking at the drm_atomic_state the crtc_state points to.
Until recently i9xx_get_refclk() used to be called indirectly from
vlv_force_pll_on() with a dummy crtc_state. That dummy crtc state is not
converted to be part of a full drm atomic state, so add a WARN in case
someone decides to call that again with a such dummy state. This was
removed in
commit 9cbe40c15a
Author: Vijay Purushothaman <vijay.a.purushothaman@linux.intel.com>
Date: Thu Mar 5 19:33:08 2015 +0530
drm/i915: Update prop, int co-eff and gain threshold for CHV
v2: Warn if there is no connectors for a given crtc. (Daniel)
Replace comment i9xx_get_refclk() with a WARN_ON(). (Ander)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Add commit reference for when i9xx_get_refclk was removed.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Follow up patches will convert some functions called from there to use
the atomic state, instead of directly accessing the new or current
config. This patch just changes the parameters, but shouldn't have any
functional changes.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Traces for page directories and tables allocation and map.
v2: Removed references to teardown.
v3: bitmap_scnprintf has been deprecated.
v4: Replace bitmap_scnprintf with scnprintf correctly, and get right
range lengths. (Mika)
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This patch continues on the idea from "Track GEN6 page table usage".
From here on, in the steady state, PDEs are all pointing to the scratch
page table (as recommended in the spec). When an object is allocated in
the VA range, the code will determine if we need to allocate a page for
the page table. Similarly when the object is destroyed, we will remove,
and free the page table pointing the PDE back to the scratch page.
Following patches will work to unify the code a bit as we bring in GEN8
support. GEN6 and GEN8 are different enough that I had a hard time to
get to this point with as much common code as I do.
The aliasing PPGTT must pre-allocate all of the page tables. There are a
few reasons for this. Two trivial ones: aliasing ppgtt goes through the
ggtt paths, so it's hard to maintain, we currently do not restore the
default context (assuming the previous force reload is indeed
necessary). Most importantly though, the only way (it seems from
empirical evidence) to invalidate the CS TLBs on non-render ring is to
either use ring sync (which requires actually stopping the rings in
order to synchronize when the sync completes vs. where you are in
execution), or to reload DCLV. Since without full PPGTT we do not ever
reload the DCLV register, there is no good way to achieve this. The
simplest solution is just to not support dynamic page table
creation/destruction in the aliasing PPGTT.
We could always reload DCLV, but this seems like quite a bit of excess
overhead only to save at most 2MB-4k of memory for the aliasing PPGTT
page tables.
v2: Make the page table bitmap declared inside the function (Chris)
Simplify the way scratching address space works.
Move the alloc/teardown tracepoints up a level in the call stack so that
both all implementations get the trace.
v3: Updated trace event to spit out a name
v4: Aliasing ppgtt is now initialized differently (in setup global gtt)
v5: Rebase to latest code. Also removed unnecessary aliasing ppgtt check
for trace, as it is no longer possible after the PPGTT cleanup patch series
of a couple of months ago (Daniel).
v6: Implement changes from code review (Daniel):
- allocate/teardown_va_range calls added.
- Add a scratch page allocation helper (only need the address).
- Move trace events to a new patch.
- Use updated mark_tlbs_dirty.
- Moved pt preallocation for aliasing ppgtt into gen6_ppgtt_init.
v7: teardown_va_range removed (Daniel).
In init, gen6_ppgtt_clear_range call is only needed for aliasing ppgtt.
v8: Rebase after s/page_tables/page_table/.
v9: Remove unnecessary scratch flag in page_table struct, future patches
can just compare against ppgtt->scratch_pt, and alloc_pt_scratch becomes
redundant. Initialize scratch_pt and pt. (Mika)
v10: Clean up aliasing ppgtt init error path and prevent leaking the
ppgtt obj when init fails. (Mika)
Updated commit author. (Daniel)
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v4+)
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We are already unmapping them in gen6_ppgtt_free. This function became
redundant since commit 06fda602db
("drm/i915: Create page table allocators").
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
i915_dma_map_single relies on dma_mapping_error, which returns positive
error codes. Found by static checker.
Introduced by commit 678d96fbb3
("drm/i915: Track GEN6 page table usage").
v2: Return negative error code and renamed commit title. (Dan)
v3: Missing reported-by tag (Daniel)
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Found by static analysis tool, this was harmless as the pt was not
used out of scope though.
Introduced by commit 678d96fbb3
("drm/i915: Track GEN6 page table usage").
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
It's completely unused and Tommi noticed that the #define is borked
since forever. I've done a git search in userspace and only found
broken definitions and no users anywhere.
Cc: Tommi Rantala <tt.rantala@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
This patch makes STI driver use the atomic helpers.
I have fix the comments done by Daniel on the first version and get
his ack with this second version.
* 'drm-st-next-2015-03-19' of git://git.linaro.org/people/benjamin.gaignard/kernel:
drm: sti: convert driver to atomic modeset
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Instead of using connector->new_encoder, get the same information from
the pipe_config, thus making the function ready for the atomic
conversion.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Keep the if (!connector) continue; separate so that it's
easier to eventually extract a for_each_connector_in_state iterator.
And because of the upcast it's also safer.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Move towards atomic by using the legacy modeset's drm_atomic_state
instead.
v2: Move call to drm_atomic_add_affected_connectors() to
intel_modeset_compute_config(). (Daniel)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Resurrect the ret local variable which I've dropped from an
earlier patch and which is now needed.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
With this in place, we can start converting pieces of the modeset code
to look at the connector atomic state instead of the staged config.
v2: Handle the load detect staged config changes too. (Ander)
Remove unnecessary blank line. (Daniel)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Keep that state updated so that we can write code that depends on it on
the follow up patches.
v2: Fix BUG due to stale connector_state->crtc value. (Chandra)
v3: Update comment about dummy state connectors. (Chandra)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
So that we can add connector states to the drm_atomic_state used in the
legacy modeset.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
For consistency, allocate a new crtc_state for a crtc that is being
disabled. Previously only the enabled value of the current state would
change.
v2: Rebase on v5 of previous patch. (Ander)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Resolve rebase conflict.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
For the atomic conversion, the mode set paths need to be changed to rely
on an atomic state instead of using the staged config. By using an
atomic state for the legacy code, we will be able to convert the code
base in small chunks.
v2: Squash patch that adds stat argument to intel_set_mode(). (Ander)
Make every caller of intel_set_mode() allocate state. (Daniel)
Call drm_atomic_state_clear() in set config's error path. (Daniel)
v3: Copy staged config to atomic state in force restore path. (Ander)
v4: Don't update ->new_config for disabled pipes in __intel_set_mode(),
since it is expected to be NULL in that case. (Ander)
v5: Don't change return type of intel_modeset_pipe_config(). (Chandra)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
[danvet: Remove spurious ret local variable due to changes in v5.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
For now this is not necessary since intel_set_mode() doesn't acquire any
new locks. However, once that function is converted to atomic, that will
change, since we'll pass an atomic state to it, and that needs to have
the right acquire context set.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The pattern of getting the crtc state with drm_atomic_get_crtc_state()
and then converting it to intel_crtc_state will repeat quite often in
the following patches, so add a helper function to save some typing.
v2: Fix upcasting so that crtc_state base field could be moved. (Daniel)
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Currently we only set preserve_bios_swizzling when the initial fb is
shared and totally miss the single-screen case. Fix this by
consolidating all the logic for both cases.
This seems to go back to when swizzle preservation was originally
merged in
commit d9ceb81633
Author: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Thu Oct 9 12:57:43 2014 -0700
drm/i915: preserve swizzle settings if necessary v4
Cc: Kristian Høgsberg <hoegsberg@gmail.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
In spirit with
commit 5724dbd167
Author: Damien Lespiau <damien.lespiau@intel.com>
Date: Tue Jan 20 12:51:52 2015 +0000
drm/i915: Rename plane_config to initial_plane_config
to make it clear that this code is all special-purpose for the initial
plane takeover.
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
psr.active is being unset out of the if so this here is useless and
duplicated.
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This is a very similar bug in the load detect code fixed in
commit 9128b040eb
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue Mar 3 17:31:21 2015 +0100
drm/i915: Fix modeset state confusion in the load detect code
But this time around it was the initial fb code that forgot to update
the plane->crtc pointer. Otherwise it's the exact same bug, with the
exact same restrains (any set_config call/ioctl that doesn't disable
the pipe papers over the bug for free, so fairly hard to hit in normal
testing). So if you want the full explanation just go read that one
over there - it's rather long ...
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Josh Boyer <jwboyer@fedoraproject.org>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reported-and-tested-by: Josh Boyer <jwboyer@fedoraproject.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This flag was being mostly used as a meta flag in some
cases and not covering other cases.
One of the risks is that it was masking some frontbuffer
trackings without disabling PSR.
So, better to kill this at once and avoid umbrella parameters.
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Drop unused out: label to appease gcc.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This is a very similar bug in the load detect code fixed in
commit 9128b040eb
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue Mar 3 17:31:21 2015 +0100
drm/i915: Fix modeset state confusion in the load detect code
But this time around it was the initial fb code that forgot to update
the plane->crtc pointer. Otherwise it's the exact same bug, with the
exact same restrains (any set_config call/ioctl that doesn't disable
the pipe papers over the bug for free, so fairly hard to hit in normal
testing). So if you want the full explanation just go read that one
over there - it's rather long ...
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Josh Boyer <jwboyer@fedoraproject.org>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reported-and-tested-by: Josh Boyer <jwboyer@fedoraproject.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[Jani: backported to drm-intel-fixes for v4.0-rc]
Reference: http://mid.gmane.org/CA+5PVA7ChbtJrknqws1qvZcbrg1CW2pQAFkSMURWWgyASRyGXg@mail.gmail.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Right now, we get a warning when taking over the firmware fb:
[drm:drm_atomic_plane_check] FB set but no CRTC
with the following backtrace:
[<ffffffffa010339d>] drm_atomic_check_only+0x35d/0x510 [drm]
[<ffffffffa0103567>] drm_atomic_commit+0x17/0x60 [drm]
[<ffffffffa00a6ccd>] drm_atomic_helper_plane_set_property+0x8d/0xd0 [drm_kms_helper]
[<ffffffffa00f1fed>] drm_mode_plane_set_obj_prop+0x2d/0x90 [drm]
[<ffffffffa00a8a1b>] restore_fbdev_mode+0x6b/0xf0 [drm_kms_helper]
[<ffffffffa00aa969>] drm_fb_helper_restore_fbdev_mode_unlocked+0x29/0x80 [drm_kms_helper]
[<ffffffffa00aa9e2>] drm_fb_helper_set_par+0x22/0x50 [drm_kms_helper]
[<ffffffffa050a71a>] intel_fbdev_set_par+0x1a/0x60 [i915]
[<ffffffff813ad444>] fbcon_init+0x4f4/0x580
That's because we update the plane state with the fb from the firmware, but we
never associate the plane to that CRTC.
We don't quite have the full DRM take over from HW state just yet, so
fake enough of the plane atomic state to pass the checks.
v2: Fix the state on which we set the CRTC in the case we're sharing the
initial fb with another pipe. (Matt)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[Jani: backported to drm-intel-fixes for v4.0-rc]
Reference: http://mid.gmane.org/CA+5PVA7yXH=U757w8V=Zj2U1URG4nYNav20NpjtQ4svVueyPNw@mail.gmail.com
Reference: http://lkml.kernel.org/r/CA+55aFweWR=nDzc2Y=rCtL_H8JfdprQiCimN5dwc+TgyD4Bjsg@mail.gmail.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
If we retire requests last, we may use a later seqno and so clear
the requests lists without clearing the active list, leading to
confusion. Hence we should retire requests first for consistency with
the early return. The order used to be important as the lifecycle for
the object on the active list was determined by request->seqno. However,
the requests themselves are now reference counted removing the
constraint from the order of retirement.
Fixes regression from
commit 1b5a433a4d
Author: John Harrison <John.C.Harrison@Intel.com>
Date: Mon Nov 24 18:49:42 2014 +0000
drm/i915: Convert 'i915_seqno_passed' calls into 'i915_gem_request_completed
'
and a
WARNING: CPU: 0 PID: 1383 at drivers/gpu/drm/i915/i915_gem_evict.c:279 i915_gem_evict_vm+0x10c/0x140()
WARN_ON(!list_empty(&vm->active_list))
Identified by updating WATCH_LISTS:
[drm:i915_verify_lists] *ERROR* blitter ring: active list not empty, but no requests
WARNING: CPU: 0 PID: 681 at drivers/gpu/drm/i915/i915_gem.c:2751 i915_gem_retire_requests_ring+0x149/0x230()
WARN_ON(i915_verify_lists(ring->dev))
Note that this is only a problem in evict_vm where the following happens
after a retire_request has cleaned out all requests, but not all active
bo:
- intel_ring_idle called from i915_gpu_idle notices that no requests are
outstanding and immediately returns.
- i915_gem_retire_requests_ring called from i915_gem_retire_requests also
immediately returns when there's no request, still leaving the bo on the
active list.
- evict_vm hits the WARN_ON(!list_empty(&vm->active_list)) after evicting
all active objects that there's still stuff left that shouldn't be
there.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
The faulting virtual address is >32bits and has been moved
to different registers. Add to error state and output upper
register first, in the same line for easy reconstruction of
the fault address.
v2: correct gen masking (Michel)
v3: s/TBL/TLB (Ville)
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
It should have been negative since it is returned with ERR_PTR().
Introduced in new code commit:
commit 50470bb011
Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon Mar 23 11:10:36 2015 +0000
drm/i915/skl: Support secondary (rotated) frame buffer mapping
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Or users can just spam the log all they want.
Issue introduced in
commit 9a8f0a1290
Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Fri Feb 27 11:15:24 2015 +0000
drm/i915/skl: Allow Y (and Yf) frame buffer creation
References: https://bugs.freedesktop.org/show_bug.cgi?id=89628
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The current code can only support one kgd instance. We have to
support multiple kgd instances in one system. i.e two amdgpu or two
radeon or one amdgpu + one radeon or more than two kgd instances.
Signed-off-by: Xihan Zhang <xihan.zhang@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
Convert the timestamping in the amdkfd driver to use a timespec64 and 64bit
time accessors.
Although the existing code is completely safe beyond y2038 because it deals
with monotonic time, this patch is still needed in order to kill off all uses
of struct timespec.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
fence_wait_timeout() is an exported kernel symbol, so we should rename our
local function to something different.
Signed-off-by: Oded Gabbay <oded.gabbay@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
This series converts of_graph_get_next_endpoint to decrement the refcount of
the passed prev parameter. This allows to add a for_each_endpoint_of_node
helper macro to loop over all endpoints in a device tree node.
The of_graph_get_port_by_id function is added to retrieve a port by its known
port id (contained in the reg property) from the device tree.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=jtB3
-----END PGP SIGNATURE-----
Merge tag 'of-graph-for-4.0' of git://git.pengutronix.de/git/pza/linux into for-next
Pull of-graph helpers from Philipp Zabel:
of: Add of-graph helpers to loop over endpoints and find ports by id
This series converts of_graph_get_next_endpoint to decrement the refcount of
the passed prev parameter. This allows to add a for_each_endpoint_of_node
helper macro to loop over all endpoints in a device tree node.
The of_graph_get_port_by_id function is added to retrieve a port by its known
port id (contained in the reg property) from the device tree.
Tvrtko noticed a new warning on boot:
WARNING: CPU: 1 PID: 353 at include/linux/kref.h:47 drm_framebuffer_reference+0x6c/0x80 [drm]()
Call Trace:
[<ffffffff8161f10c>] dump_stack+0x4f/0x7b
[<ffffffff81052caa>] warn_slowpath_common+0xaa/0xd0
[<ffffffff81052d8a>] warn_slowpath_null+0x1a/0x20
[<ffffffffa00d035c>] drm_framebuffer_reference+0x6c/0x80 [drm]
[<ffffffffa01c0df7>] update_state_fb.isra.54+0x47/0x50 [i915]
[<ffffffffa01ccd5c>] skylake_get_initial_plane_config+0x93c/0x950 [i915]
[<ffffffffa01e8721>] intel_modeset_init+0x1551/0x17c0 [i915]
[<ffffffffa02476e0>] i915_driver_load+0xed0/0x11e0 [i915]
[<ffffffff81627aa1>] ? _raw_spin_unlock_irqrestore+0x51/0x70
[<ffffffffa00ca8b7>] drm_dev_register+0x77/0x110 [drm]
[<ffffffffa00cda3b>] drm_get_pci_dev+0x11b/0x1f0 [drm]
[<ffffffff81098e3d>] ? trace_hardirqs_on+0xd/0x10
[<ffffffff81627aa1>] ? _raw_spin_unlock_irqrestore+0x51/0x70
[<ffffffffa0145276>] i915_pci_probe+0x56/0x60 [i915]
[<ffffffff813ad59c>] pci_device_probe+0x7c/0x100
[<ffffffff81466aad>] driver_probe_device+0x16d/0x380
We cannot take a reference at this point, not before
intel_framebuffer_init() and the underlying drm_framebuffer_init().
Introduced in:
commit 706dc7b549175e47f23e913b7f1e52874a7d0f56
Author: Matt Roper <matthew.d.roper@intel.com>
Date: Tue Feb 3 13:10:04 2015 -0800
drm/i915: Ensure plane->state->fb stays in sync with plane->fb
v2: Don't move update_state_fb(). It was moved around because I
originally put update_state_fb() in intel_alloc_plane_obj() before
finding a better place. (Matt)
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Reported-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From drm-next:
(cherry picked from commit f55548b5af)
Signed-off-by: Dave Airlie <airlied@redhat.com>
Setting a dev_pm_ops resume callback but not a set of hibernation
handler means that pm function will not be called upon hibernation.
Fix this by using SIMPLE_DEV_PM_OPS, which appropriately assigns the
suspend and hibernation handlers and move omap_dmm_resume under
CONFIG_PM_SLEEP to avoid build warnings.
Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
[tomi valkeinen: add missing 'static']
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Setting a dev_pm_ops suspend/resume pair but not a set of hibernation
functions means those pm functions will not be called upon hibernation.
Fix this by using SIMPLE_DEV_PM_OPS, which appropriately assigns the
suspend and hibernation handlers and move
omap_drm_suspend/omap_drm_resume under CONFIG_PM_SLEEP to avoid build
warnings.
Signed-off-by: Grygorii Strashko <Grygorii.Strashko@linaro.org>
[tomi.valkeinen@ti.com: fix conflict, clean up description]
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
We store the fb being page-flipped to 'old_fb' field, but we don't
increase the ref count of the fb when doing that. While I am not
sure if it can cause problem in practice, it's still safer to keep a ref
when storing a pointer to a fb.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The omapdrm DMM code sometimes crashes with:
WARNING: CPU: 0 PID: 1235 at lib/list_debug.c:36 __list_add+0x8c/0xbc()
list_add double add: new=e9265368, prev=e90139c4, next=e9265368.
This is caused by the code calling release_engine() twice for the same
engine.
dmm_txn_commit(wait=true) call is supposed to wait until the DMM
transaction has been finished. And it does that, but it does not wait
for the irq handler to finish.
What happens is that the irq handler is triggered, and it either wakes
up the thread that called dmm_txn_commit(), or that thread never even
slept because the transaction was finished in the HW very quickly. That
thread then continues executing, even if the irq handler is not yet
finished, and a new transaction may be initiated. If that transaction is
async (i.e. wait=false), a 'async' flag is set to true. The original irq
handler, which has yet not finished, then sees the transaction as
'async', even if it was supposed to be 'sync'.
When that happens, the irq handler does an extra release_engine() call
because it thinks it need to release the engine, leading to the crash.
This patch fixes the issue by using completion to ensure that the irq
handler has finished before a dmm_txn_commit(wait=true) may continue.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omap_gem_objects are added to dev->obj_list in omap_gem_new, and removed
in omap_gem_free_object. Unfortunately there's no locking for
dev->obj_list, which eventually leads to a crash:
WARNING: CPU: 1 PID: 1123 at lib/list_debug.c:59 __list_del_entry+0xa4/0xe0()
list_del corruption. prev->next should be e9281344, but was ea722b84
Add a spinlock to protect dev->obj_list.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
spin_is_locked(x) returns always 0 on uniprocessor, triggering BUG() in
omapdrm.
Change it to use assert_spin_locked() to fix the issue.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
We need to ignore DIGIT SYNC LOST error when enabling/disabling TV
output. The code does that, but it ignores the DIGI SYNC LOST when
enabling any output. Normally this does no harm, but it could make us
miss DIGIT SYNC LOST on some rare occasions.
Fix the code to only ignore DIGIT SYNC LOST when enabling/disabling TV.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omapdrm tries to avoid error floods by unregistering the error irq when
an error happens, and then registering the error irq again later.
However, the code is racy, as it sometimes tries to unregister the error
irq when it's already unregistered, leading to WARN().
Also, the code only registers the error irq again when something is done
on that particular output, i.e. if only TV is used to flip the buffers,
and LCD is showing a same buffer, an error on LCD will cause the LCD
error irq to be unregistered and never registered again.
To fix this, let's keep the error irqs always enabled and trust the
DRM_ERROR_RATELIMITED to limit the flood.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omapdrm uses normal DRM_ERROR() print when the HW reports an error. As
we sometimes may get a flood of errors, let's rather use
DRM_ERROR_RATELIMITED().
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
When not using proper hotplug detection, DRM polls periodically the
connectors to find out if a cable is connected. This polling can happen
at any time, even very late in the suspend process.
This causes a problem with omapdrm, when the poll happens during the
suspend process after GPIOs have been disabled, leading to a crash in
gpio_get().
This patch fixes the issue by adding suspend and resume hooks to
omapdrm, in which we disable and enable, respectively, the polling.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omapdrm has dummy functions for platform_device's
suspend/resume/shutdown. The functions don't do anything, and those
platform device functions are deprecated, so remove them from omapdrm.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The tiler irq handler uses engine->async value, but the code that sets
engine->async and enables the interrupt does not have a barrier. This
may cause the irq handler to see the old value of engine->async, causing
memory corruption.
Reported-by: Harinarayan Bhatta <harinarayan@ti.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omap_plane_pre_apply() sets the plane's output channel too late, only
after the plane has already been otherwise configured and enabled. This
causes problems, as at the configuration stage we need to make decisions
based on the output channel.
This may lead to bad plane settings or failing to setup the plane.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
On OMAP5 it is not possible to use TILER buffer with CPU when caching or
write-combining is used. Doing so leads to errors from the memory
manager.
However, on OMAP4, write-combining works fine.
This patch adds platform specific data for the TILER, and a function
tiler_get_cpu_cache_flags() which can be used to get the caching mode to
be used.
Note that without write-combining the use of the TILER buffer with CPU
is unusably slow. It's still good to have it operational for testing
purposes.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omapdrm doesn't check if the pitch of the framebuffer and the color
format's bits-per-pixel are compatible. omapdss requires that the stride
of a buffer is an integer number of pixels
For example, when using modetest with a display that has x resolution of
1280, and using packed 24 RGB mode (3 bytes per pixel), modetest
allocates a buffer with a byte stride of 4 * 1280 = 5120. But 5120 / 3 =
1706.666... pixels, which causes wrong colors and a tilt on the screen.
Add a check into omapdrm to return an error if the user tries to use
such a combination.
Note: this is not a HW requirement at least for non-rotation use cases,
but a SW driver requirement. In the future we should study if also
rotation use cases are fine with any stride size, and if so, change the
driver to allow these strides.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
When an error happens in omap_framebuffer_create(),
omap_framebuffer_create() calls omap_framebuffer_destroy() if the fb
struct has been allocated. However, that crashes, as
omap_framebuffer_destroy(), which calls drm_framebuffer_cleanup(),
should only be called after drm_framebuffer_init()
Fix this by just calling kfree() for the allocated fb when an error
happens.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omapdrm should work fine even if fbdev is missing. The current driver
crashes in that case, though, as it is missing checks for the fbdev.
Add the checks so that we don't free fbdev or restore fbdev mode when
there's no fbdev.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
unpin_worker() calls omap_framebuffer_unpin() without any locks, which
looks very suspicious. However, both pin and unpin are always called via
the driver's private workqueue, so the access is synchronized that way.
Add a comment to make this clear.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
omap_framebuffer_pin() and omap_framebuffer_unpin() are currently
broken, as they cannot be called multiple times (i.e. pin, pin, unpin,
unpin), which is what happens in certain cases. This issue causes the
driver to possibly use 0 as an address for a displayed buffer, leading
to OCP error from DSS.
This patch fixes the issue by adding a simple pin_count, used to track
the number of pins.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Clear omap_obj's paddr when unmapping the memory, so that it's easier to
catch bad use of the paddr.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The DRM documentation says:
"If a page flip is already pending, the page_flip operation must return
-EBUSY."
Currently omapdrm returns -EINVAL instead. Fix omapdrm by returning
-EBUSY.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
OMAP DSS hardware supports changing the output port to which an overlay
manager's video stream goes. For example, DPI video stream can come from
any of the four overlay managers on OMAP5.
However, as it's difficult to manage the change in the driver, the
omapdss driver does not support that at the moment, and has a hardcoded
overlay manager per output.
omapdrm, on the other hand, uses the hardware features to find out which
overlay manager to use for an output, which causes problems. For
example, on OMAP5, omapdrm tries to use DIGIT overlay manager for DPI
output, instead of the LCD3 required by the omapdss driver.
This patch changes the omapdrm to use the omapdss driver's hardcoded
overlay managers, which fixes the issue.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The ps8622_attach and ps8522_driver symbols are never used outside this
file, so they should be static.
Signed-off-by: Thierry Reding <treding@nvidia.com>
As part of allocation of the drm_i915_private variable, drrs capability
enum is initialized to DRRS_NOT_SUPPORTED. Hence need not initialize at
each connector init.
Moreover initializing this enum at connector init will reset
the successful DRRS initialization of previous connector, as we have
the DRRS support for only one panel at a time.
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The logical place for clearing the RPS latched interrupt bits is when
resetting the RPS interrupts, so move the corresponding part from the RPS
disable function to the reset function. During resetting we already
cleared the IIR bits, so the only thing missing there was clearing pm_iir.
Note that we call gen6_disable_rps_interrupts() also during driver load
and resume time via intel_uncore_sanitize() when i915 interrupts are
still not installed. If there are any pending RPS bits at this point
(which after this patch wouldn't be cleared) they will be cleared by the
reset code via the interrupt preinstall hooks.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When disabling RPS interrupts there is a race where we disable RPS
inerrupts while the interrupt handler is running and the handler has
already latched the pending RPS interrupt from the master IIR register.
Afterwards the disabling path clears the PM IIR bits, making the state
of pending interrupts inconsistent from the interrupt handler's point of
view. This triggers the following warning: "The master control interrupt
lied (PM)!".
To fix this make sure that any running interrupt handler (which may
have already latched the master IIR) finishes before clearing the IIR
bits.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=87347
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Originally it was impossible to be dropping the last refcount in this
function since there was always one around still from the idr. But in
commit 83f45fc360
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed Aug 6 09:10:18 2014 +0200
drm: Don't grab an fb reference for the idr
we've switched to weak references, broke that assumption but forgot to
fix it up.
Since we still force-disable planes it's only possible to hit this
when racing multiple rmfb with fbdev restoring or similar evil things.
As long as userspace is nice it's impossible to hit the BUG_ON.
But the BUG_ON would most likely be hit from fbdev code, which usually
invovles the console_lock besides all modeset locks. So very likely
we'd never get the bug reports if this was hit in the wild, hence
better be safe than sorry and backport.
Spotted by Matt Roper while reviewing other patches.
[airlied: pull this back into 4.0 - the oops happens there]
Cc: stable@vger.kernel.org
Cc: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
drm-intel-next-2015-03-13-rebased:
- EU count report param for gen9+ (Jeff McGee)
- piles of pll/wm/... fixes for chv, finally out of preliminary hw support
(Ville, Vijay)
- gen9 rps support from Akash
- more work to move towards atomic from Matt, Ander and others
- runtime pm support for skl (Damien)
- edp1.4 intermediate link clock support (Sonika)
- use frontbuffer tracking for fbc (Paulo)
- remove ilk rc6 (John Harrison)
- a bunch of smaller things and fixes all over
Includes backmerge because git rerere couldn't keep up any more.
* tag 'drm-intel-next-2015-03-13-merge' of git://anongit.freedesktop.org/drm-intel: (366 commits)
drm/i915: Make sure the primary plane is enabled before reading out the fb state
drm/i915: Update DRIVER_DATE to 20150313
drm/i915: Fix vmap_batch page iterator overrun
drm/i915: Export total subslice and EU counts
drm/i915: redefine WARN_ON_ONCE to include the condition
drm/i915/skl: Implement WaDisableHBR2
drm/i915: Remove the preliminary_hw_support shackles from CHV
drm/i915: Read CHV_PLL_DW8 from the correct offset
drm/i915: Rewrite IVB FDI bifurcation conflict checks
drm/i915: Rewrite some some of the FDI lane checks
drm/i915/skl: Enable the RPS interrupts programming
drm/i915/skl: Enabling processing of Turbo interrupts
drm/i915/skl: Updated the i915_frequency_info debugfs function
drm/i915: Simplify the way BC bifurcation state consistency is kept
drm/i915/skl: Updated the act_freq_mhz_show sysfs function
drm/i915/skl: Updated the gen9_enable_rps function
drm/i915/skl: Updated the gen6_rps_limits function
drm/i915/skl: Restructured the gen6_set_rps_thresholds function
drm/i915/skl: Updated the gen6_set_rps function
drm/i915/skl: Updated the gen6_init_rps_frequencies function
...
Before, we would set the property, but also return -EINVAL because of a
broken fallthrough.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Active was here, and we allowed users to set it, but not to get it as
well.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Just as we provide crtc->mode pre-populated with the requested mode,
move adjusted_mode into hwmode before we call the crtc's mode_set,
making sure to restore it on failure.
Allows drivers which thoughtlessly discard adjusted_mode in their
mode_set hooks (e.g. Exynos) to use hwmode directly, and also provides
some neat symmetry with crtc->mode.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since we're now using mode == NULL to represent disabled, it's not
wholly surprising that we'd want to compare NULL modes.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
mode is always NULL at this point in the function, so make our intention
clear.
Signed-off-by: Daniel Stone <daniels@collabora.com>
[danvet: Stop clearing mode too to enlist gcc in tracking
uninitialized usage. And remove a space while at it.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When doing a native or i2c aux write the sink will indicate the number
of bytes written even if it the nacks the transfer. When we receive a
nack we just return an error upwards, but it might still be interesting
to see how many bytes made it before the nack. So include that information
in the debug messages.
v2: Also print the message size (Jani)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
v2: Pass in rotation info to sprite plane updates as well.
v3: Use helper to determine 90/270 rotation. (Michel Thierry)
v4: Rebased for fb modifiers and atomic changes.
For: VIZ-4546
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> (v3)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Need to do this in order to support 90/270 rotated display.
v2: Pass in drm_plane instead of plane index to intel_obj_display_address.
v3:
* Renamed intel_obj_display_address to intel_plane_obj_offset.
(Chris Wilson)
* Simplified rotation check to bitwise AND. (Chris Wilson)
v4:
* Extracted 90/270 rotation check into a helper function. (Michel Thierry)
v5:
* Rebased for ggtt view changes.
For: VIZ-4545
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
90/270 rotated scanout needs a rotated GTT view of the framebuffer.
This is put in a separate VMA with a dedicated ggtt view and wired such that
it is created when a framebuffer is pinned to a 90/270 rotated plane.
Rotation is only possible with Yb/Yf buffers and error is propagated to
user space in case of a mismatch.
Special rotated page view is constructed at the VMA creation time by
borrowing the DMA addresses from obj->pages.
v2:
* Do not bother with pages for rotated sg list, just populate the DMA
addresses. (Daniel Vetter)
* Checkpatch cleanup.
v3:
* Rebased on top of new plane handling (create rotated mapping when
setting the rotation property).
* Unpin rotated VMA on unpinning from display plane.
* Simplify rotation check using bitwise AND. (Chris Wilson)
v4:
* Fix unpinning of optional rotated mapping so it is really considered
to be optional.
v5:
* Rebased for fb modifier changes.
* Rebased for atomic commit.
* Only pin needed view for display. (Ville Syrjälä, Daniel Vetter)
v6:
* Rebased after preparatory work has been extracted out. (Daniel Vetter)
v7:
* Slightly simplified tiling geometry calculation.
* Moved rotated GGTT view implementation into i915_gem_gtt.c (Daniel Vetter)
v8:
* Do not use i915_gem_obj_size to get object size since that actually
returns the size of an VMA which may not exist.
* Rebased for ggtt view changes.
v9:
* Rebased after code review changes on the preceding patches.
* Tidy function definitions. (Joonas Lahtinen)
For: VIZ-4726
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> (v4)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
For now only default implementation defaulting to normal view.
v2: Some code review cleanups. (Joonas Lahtinen)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> (v2)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Plane state carries the rotation information which is needed for determining
the appropriate GGTT view type.
This just adds the parameter with the actual usage coming in future patches.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
To support frame buffer rotation we need to be able to pass on the information
on what kind of GGTT view is required for display.
This patch just adds the parameter and makes all the callers default to the
normal view.
v2: Rebased for ggtt view changes.
v3: Don't limit PIN_MAPPABLE to normal views just yet. (Joonas Lahtinen)
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> (v3)
[danvet: s/BUG/WARN/ in the patch hunk because. At least where the
BUG_ON isn't fatal right away.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
It will be used in a later patch and also convert all height parameters
from int to unsigned int.
v2: Rebased for fb modifiers.
v3: Fixed v2 rebase.
v4:
* Height should be unsigned int.
* Make it take pixel_format for consistency and simplicity.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> (v1)
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> (v4)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
msleep() can sleep for way too long, so switch wait_for() to use
usleep_range() instead. Following a totally unscientific method
I just picked the range as W-2W.
This cuts the i915 init time on my BSW to almost half:
- initcall i915_init+0x0/0xa8 [i915] returned 0 after 419977 usecs
+ initcall i915_init+0x0/0xa8 [i915] returned 0 after 238419 usecs
Note that I didn't perform any other benchmarks on this so far.
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The next commit will need functions to be reordered. Do it separately to
help review.
This only moves functions without any change to the code.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Planes are destroyed after framebuffers, which has the side effect of
disabling all planes. There is thus no need to disable planes explicitly
when destroying them.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The function will convert the Q16 source coordinates to integers, avoid
converting integers to Q16 first and perform the opposite conversion.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Move the set_enabled function to avoid the forward declaration. While at
it prefix it with omap_crtc_ like most other functions in the file, and
fix the comment stating in which contexts the function is called.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The full_update field is always set to true before calling
omap_crtc_appy(), resulting in its value always being true in the single
location where it is tested, in omap_crtc_pre_apply(). Remove it.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
All the manual update display code implements eventually ends up to just
calls to omap_connector_flush(), currently implemented as an empty TODO
stub. Remove it, the code can always be revived and implemented later if
interest in manual update displays becomes a reality.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The vblank interrupt is used by the driver as a completion signal when
applying new settings.
A race condition exist between enabling the vblank interrupt and
applying new settings to the hardware by setting the GO bit. If a vblank
interrupt occurs in-between, the driver will incorrectly consider the
new settings to be applied. Fix this by enabling the interrupt after
setting the GO bit.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Whether to reset plane properties at disable time isn't well-defined in
DRM, but resetting only part of them is probably as bad as it can get.
Make the behaviour coherent by resetting the zorder property in addition
to the rotation property.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The planes don't care about DPMS states, don't propagate it
unnecessarily to the plane functions.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Remove the CRTC private planes by switching to the universal plane API.
This results in a merge of the CRTC private plane created by the driver
(omap_crtc->plane) and the CRTC primary plane created by the DRM core
(crtc->primary).
Reference counting of the framebuffers in the update plane operation is
thus simplified as no reference needs to be stored in the private plane
anymore.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
The ilace variable is unused and the replication variable is assigned to
false and just passed to a function. Remove them.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Create a omap_modeset_create_crtc() function to avoid duplicating plane
and CRTC creation code.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Write the PLANE_SURF register instead of PLANE_CTL to arm the double
buffer regisrter update.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Replace the RMW access with explicit initialization of the entire plane
control register, as was done for primary planes in:
commit f45651bae2
Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date: Fri Aug 8 21:51:10 2014 +0300
drm/i915: Eliminate rmw from .update_primary_plane()
The automagic primary plane disable is still doing RMWs, but that will
require more work to untangle, so leave it alone for now.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Store the colorkey in intel_plane and kill off all the RMW stuff
handling it.
This is just an intermediate step and eventually the colorkey needs to
be converted into some properties.
v2: Actually update the hardware state in the set_colorkey ioctl (Daniel)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Determining whether we'll need to wait for vblanks is something we
should determine during the atomic 'check' phase, not the 'commit'
phase. Note that we only set these bits in the branch of 'check' where
intel_crtc->active is true so that we don't try to wait on a disabled
CRTC.
The whole 'wait for vblank after update' flag should go away in the
future, once we start handling watermarks in a proper atomic manner.
This regression has been introduced in
commit 2fdd7def16dd7580f297827930126c16b152ec11
Author: Matt Roper <matthew.d.roper@intel.com>
Date: Wed Mar 4 10:49:04 2015 -0800
drm/i915: Don't clobber plane state on internal disables
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Root-cause-analysis-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89550
Testcase: igt/pm_rpm/legacy-planes
Testcase: igt/pm_rpm/legacy-planes-dpms
Testcase: igt/pm_rpm/universal-planes
Testcase: igt/pm_rpm/universal-planes-dpms
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Tested-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Prepare chv_find_best_dpll to be used for BXT too, where we want to
consider the error between target and calculated frequency too when
choosing a better PLL configuration.
No functional change.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Factor out the logic to decide whether the newly calculated dividers are
better than the best found so far. Do this for clarity and to prepare
for the upcoming BXT helper needing the same.
No functional change.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
intel_plane->obj is not used anymore so kill it. Also don't pass both
the fb and obj to the sprite .update_plane() hook, as just passing the fb
is enough.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The problem is we're going to switch to a new context, which could be
the default context. The plan was to use restore inhibit, which would be
fine, except if we are using dynamic page tables (which we will). If we
use dynamic page tables and we don't load new page tables, the previous
page tables might go away, and future operations will fault.
CTXA runs.
switch to default, restore inhibit
CTXA dies and has its address space taken away.
Run CTXB, tries to save using the context A's address space - this
fails.
The general solution is to make sure every context has it's own state,
and its own address space. For cases when we must restore inhibit, first
thing we do is load a valid address space. I thought this would be
enough, but apparently there are references within the context itself
which will refer to the old address space - therefore, we also must
reinitialize.
v2: to->ppgtt is only valid in full ppgtt.
v3: Rebased.
v4: Make post PDP update clearer.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This patch was formerly known as, "Force pd restore when PDEs change,
gen6-7." I had to change the name because it is needed for GEN8 too.
The real issue this is trying to solve is when a new object is mapped
into the current address space. The GPU does not snoop the new mapping
so we must do the gen specific action to reload the page tables.
GEN8 and GEN7 do differ in the way they load page tables for the RCS.
GEN8 does so with the context restore, while GEN7 requires the proper
load commands in the command streamer. Non-render is similar for both.
Caveat for GEN7
The docs say you cannot change the PDEs of a currently running context.
We never map new PDEs of a running context, and expect them to be
present - so I think this is okay. (We can unmap, but this should also
be okay since we only unmap unreferenced objects that the GPU shouldn't
be tryingto va->pa xlate.) The MI_SET_CONTEXT command does have a flag
to signal that even if the context is the same, force a reload. It's
unclear exactly what this does, but I have a hunch it's the right thing
to do.
The logic assumes that we always emit a context switch after mapping new
PDEs, and before we submit a batch. This is the case today, and has been
the case since the inception of hardware contexts. A note in the comment
let's the user know.
It's not just for gen8. If the current context has mappings change, we
need a context reload to switch
v2: Rebased after ppgtt clean up patches. Split the warning for aliasing
and true ppgtt options. And do not break aliasing ppgtt, where to->ppgtt
is always null.
v3: Invalidate PPGTT TLBs inside alloc_va_range.
v4: Rename ppgtt_invalidate_tlbs to mark_tlbs_dirty and move
pd_dirty_rings from i915_address_space to i915_hw_ppgtt. Fixes when
neither ctx->ppgtt and aliasing_ppgtt exist.
v5: Removed references to teardown_va_range.
v6: Updated needs_pd_load_pre/post.
v7: Fix pd_dirty_rings check in needs_pd_load_post, and update/move
comment about updated PDEs to object_pin/bind (Mika).
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Instead of implementing the full tracking + dynamic allocation, this
patch does a bit less than half of the work, by tracking and warning on
unexpected conditions. The tracking itself follows which PTEs within a
page table are currently being used for objects. The next patch will
modify this to actually allocate the page tables only when necessary.
With the current patch there isn't much in the way of making a gen
agnostic range allocation function. However, in the next patch we'll add
more specificity which makes having separate functions a bit easier to
manage.
One important change introduced here is that DMA mappings are
created/destroyed at the same page directories/tables are
allocated/deallocated.
Notice that aliasing PPGTT is not managed here. The patch which actually
begins dynamic allocation/teardown explains the reasoning for this.
v2: s/pdp.page_directory/pdp.page_directories
Make a scratch page allocation helper
v3: Rebase and expand commit message.
v4: Allocate required pagetables only when it is needed, _bind_to_vm
instead of bind_vma (Daniel).
v5: Rebased to remove the unnecessary noise in the diff, also:
- PDE mask is GEN agnostic, renamed GEN6_PDE_MASK to I915_PDE_MASK.
- Removed unnecessary checks in gen6_alloc_va_range.
- Changed map/unmap_px_single macros to use dma functions directly and
be part of a static inline function instead.
- Moved drm_device plumbing through page tables operation to its own
patch.
- Moved allocate/teardown_va_range calls until they are fully
implemented (in subsequent patch).
- Merged pt and scratch_pt unmap_and_free path.
- Moved scratch page allocator helper to the patch that will use it.
v6: Reduce complexity by not tearing down pagetables dynamically, the
same can be achieved while freeing empty vms. (Daniel)
v7: s/i915_dma_map_px_single/i915_dma_map_single
s/gen6_write_pdes/gen6_write_pde
Prevent a NULL case when only GGTT is available. (Mika)
v8: Rebased after s/page_tables/page_table/.
v9: Reworked i915_pte_index and i915_pte_count.
Also exercise bitmap allocation here (gen6_alloc_va_range) and fix
incorrect write_page_range in i915_gem_restore_gtt_mappings (Mika).
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v3+)
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
In Gen8, PDPs are saved and restored with legacy contexts (legacy contexts
only exist on the render ring). So change the ordering of LRI vs MI_SET_CONTEXT
for the initialization of the context. Also the only cases in which we
need to manually update the PDPs are when MI_RESTORE_INHIBIT has been
set in MI_SET_CONTEXT (i.e. when the context is not yet initialized or
it is the default context).
Legacy submission is not available post GEN8, so it isn't necessary to
add extra checks for newer generations.
v2: Use new functions to replace the logic right away (Daniel)
v3: Add missing pd load logic.
v4: Add warning in case pd_load_pre & pd_load_post are true, and add
missing trace_switch_mm. Cleaned up pd_load conditions. Add more
information about when is pd_load_post needed. (Mika)
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
No functional changes, but will improve code clarity and removed some
duplicated defines.
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
AUX addresses are 20 bits long. Send out the entire address instead of
just the low 16 bits.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
And remove one bogus * from i915_gem_gtt.c since that's not a
kerneldoc there.
v2: Review from Chris:
- Clarify memory space to better distinguish from address space.
- Add note that shrink doesn't guarantee the freed memory and that
users must fall back to shrink_all.
- Explain how pinning ties in with eviction/shrinker.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Two code changes:
- Extract i915_gem_shrinker_init.
- Inline i915_gem_object_is_purgeable since we open-code it everywhere
else too.
This already has the benefit of pulling all the shrinker code
together, next patch adds a bit of kerneldoc.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Use both up/down manual ei calcuations for symmetry and greater
flexibility for reclocking, instead of faking the down interrupt based
on a fixed integer number of up interrupts.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Deepak S<deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Rewrite commit 31685c258e
Author: Deepak S <deepak.s@linux.intel.com>
Date: Thu Jul 3 17:33:01 2014 -0400
drm/i915/vlv: WA for Turbo and RC6 to work together.
Other than code clarity, the major improvement is to disable the extra
interrupts generated when idle. However, the reclocking remains rather
slow under the new manual regime, in particular it fails to downclock as
quickly as desired. The second major improvement is that for certain
workloads, like games, we need to combine render+media activity counters
as the work of displaying the frame is split across the engines and both
need to be taken into account when deciding the global GPU frequency as
memory cycles are shared.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Deepak S <deepak.s@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Deepak S<deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When we idle, we set the GPU frequency to the hardware minimum (not user
minimum). We introduce a new variable to distinguish between the
different roles, and to allow easy tuning of the idle frequency without
impacting over aspects of RPS. Setting the minimum frequency should be a
safety blanket as the pcu on the GPU should be power gating itself
anyway. However, in order for us to do set the absolute minimum
frequency, we need to relax a few of our assertions that we do not
exceed the user limits.
v2: Add idle_freq
v3: Init idle_freq for vlv and add a bunch of WARNs
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Deepak S <deepak.s@linux.intel.com>
Reviewed-by: Deepak S<deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
If the batch buffer is too large to fit into the aperture and we need a
GTT mapping for relocations, we currently fail. This only applies to a
subset of machines for a subset of environments, quite undesirable. We
can simply check after failing to insert the batch into the GTT as to
whether we only need a mappable binding for relocation and, if so, we can
revert to using a non-mappable binding and an alternate relocation
method. However, using relocate_entry_cpu() is excruciatingly slow for
large buffers on non-LLC as the entire buffer requires clflushing before
and after the relocation handling. Alternatively, we can implement a
third relocation method that only clflushes around the relocation entry.
This is still slower than updating through the GTT, so we prefer using
the GTT where possible, but is orders of magnitude faster as we
typically do not have to then clflush the entire buffer.
An alternative idea of using a temporary WC mapping of the backing store
is promising (it should be faster than using the GTT itself), but
requires fairly extensive arch/x86 support - along the lines of
kmap_atomic_prof_pfn() (which is not universally implemented even for
x86).
Testcase: igt/gem_exec_big #pnv,byt
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88392
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Add a WARN_ONCE for the impossible reloc case and explain in
a short comment why we want to avoid ping-pong.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This makes the interface consistent to old i915_gem_obj_ggtt_pin.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
