This structure derives from host1x_client. DRM-specific fields are moved
from host1x_client to this structure, so that host1x_client can remain
agnostic of DRM.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Rename the host1x_to_drm_bo() macro to host1x_to_tegra_bo() for
consistency and fixup various stylistic issues.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Other drivers use the tegra- prefix in their names, so add it to this
driver's name as well for consistency.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Rework the address table code for the host1x firewall. The previous
implementation allocated a bitfield but didn't check for a valid pointer
so it could potentially crash. Instead, embed a static bitmap within the
gr2d structure to avoid the allocation and use the Linux bitmap API to
reduce code complexity.
Don't annotate the driver's .remove() function __exit. Even if built in
the driver can be unloaded via sysfs, so .remove() needs to stick around
after initialization. Also remove the explicit initialization of the
driver's .owner field to THIS_MODULE because that's now handled by the
driver core.
Furthermore make an error message more consistent with other subdrivers,
index the syncpts array for better readability, remove a gratuituous
newline and reorder some variable declarations to make the code easier
to read.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The same code sequence is used in various places to validate a register
access in the command stream. This can be refactored into a separate
function.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The value stored in this field is a pointer to a command buffer, not an
ID. Avoid some confusion by reflecting that in the field's name.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Arguments on subsequent lines should be aligned with the first argument.
This one occurrence went unnoticed during code review.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The device can be unbound from the driver via sysfs, so regardless of
whether the driver is builtin or a module, its .remove() function needs
to stick around.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The structure represents a context associated with a particular process
that has opened the Tegra DRM device and requested a channel. This is a
very DRM-specific notion and has nothing to do with host1x. Rename the
structure to more clearly mark the boundaries between the two.
Signed-off-by: Thierry Reding <treding@nvidia.com>
This structure extends drm_file with Tegra DRM specific fields and has
nothing to do with host1x. Rename the structure to more clearly mark the
boundaries between host1x and Tegra DRM.
While at it, move the structure definition out of the header. It's never
used outside of the drm.c source file, so it can be defined within that.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The host1x and Tegra DRM drivers are currently tightly coupled. Renaming
the structure marks the boundary more clearly.
Signed-off-by: Thierry Reding <treding@nvidia.com>
The num_relocs count are passed to the kernel per job, not per gather.
For multi-gather jobs, we would previously fail if there were relocs in
other gathers aside from the first one.
Fix this by simply moving the check until all gathers have been
consumed.
Signed-off-by: Erik Faye-Lund <kusmabite@gmail.com>
Reviewed-by: Arto Merilainen <amerilainen@nvidia.com>
Acked-By: Terje Bergstrom <tbergstrom@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Fix a typo (iotcl -> ioctl) in the debug message when an unknown IOCTL
is encountered.
Acked-by: David Airlie <airlied@linux.ie>
Signed-off-by: Thierry Reding <treding@nvidia.com>
When userspace removes the active framebuffer using DRM_IOCTL_MODE_RMFB,
or explicitly disables the CRTC (by calling drmModeSetCrtc(..., NULL)
for example), a NULL framebuffer will be passed to the .set_config()
implementation of a CRTC. The drm_crtc_helper_set_config() helper will
decide to disable a CRTC when that happens.
To do so, it calls drm_crtc_helper_disable(), which in turn will iterate
over all encoders and decouple them from their connectors and finally
call drm_helper_disable_unused_functions() to clean up and call the
.disable() or .dpms() implementation for each encoder. However, at no
point during this sequence does it track the DPMS mode of a connector,
so it will usually remain on after this.
When a connector is enabled again, drm_helper_connector_dpms() will not
notice that the DPMS mode actually changed and won't do anything, which
causes the connector to stay disabled indefinitely.
To prevent this from happening, explicitly set the connector's DPMS mode
to off when the CRTC is disabled. That way it reflects the correct state
and can be enabled again.
This solves an issue observed when terminating an X server running on
the xf86-video-modesetting driver. Without this patch, the connector
would not be enabled properly and the screen would stay dark.
Acked-by: David Airlie <airlied@linux.ie>
Signed-off-by: Thierry Reding <treding@nvidia.com>
The conflict in intel_drv.h tripped me up a bit since a patch in dinq
moves all the functions around, but another one in drm-next removes a
single function. So I'ev figured backing this into a backmerge would
be good.
i915_dma.c is just adjacent lines changed, nothing nefarious there.
Conflicts:
drivers/gpu/drm/i915/i915_dma.c
drivers/gpu/drm/i915/intel_drv.h
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
It's not really needed, rather just adds another place to hold
intermediate values that could go wrong, and it's not clear that the
training pattern set or training lane set should be written at this
point at all.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The user of these counters was killed in
commit d79cdc8312
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu Aug 8 15:41:32 2013 +0200
drm: no-op out GET_STATS ioctl
so clean up the leftovers as well.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
We can get the PCI vendor and device IDs via dev->pdev. So we can drop
the duplicated information.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The only user of ctx_count is the via driver, and we can replace that
use with list_is_singular().
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
irq_enabled is only ever 0 or 1, so make it a bool.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
drm_vblank_init() is too ugly. Make it a bit easier on the eye by
collecting all the per-crtc vblank counters, timestamps etc. to
a structure and just allocate an array of those.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
vblank_enabled is only ever 0 or 1, so make it a bool.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
vblank_disable_allowed is only ever 0 or 1, so make it a bool.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
drm_helper_probe_single_connector_modes() can be used to implement
->fill_modes(), not ->probe().
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
If the firmware is not builtin and userspace is not yet running, we can
stall the boot process for a minute whilst the firmware loader times
out. This is contrary to expectations of providing a builtin EDID!
In the process, we can rearrange the code to make the error handling
more resilient and prevent gcc warning about unitialised variables along
the error paths.
v2: Load builtins first, fix gcc second (Jani) and cosmetics (Ville).
v3: Verify that we do not read beyond the end of the fwdata (Ville)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Analog to drm_dev_register(), we now provide drm_dev_unregister() which
does the reverse. drm_dev_put() is still in place and combines the calls
to drm_dev_unregister() and drm_dev_free() so buses don't have to change.
*_get() and *_put() are used for reference-counting in the kernel.
However, drm_dev_put() definitely does not do any kind of ref-counting.
Hence, use the more appropriate *_register(), *_unregister(), *_alloc()
and *_free() names.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
The error paths in DRM bus drivers currently leak memory as they don't
correctly revert drm_dev_alloc(). Introduce drm_dev_free() to free DRM
devices which haven't been registered, yet.
We must be careful not to introduce any side-effects with cleanups done in
drm_dev_free(). drm_ht_remove(), drm_ctxbitmap_cleanup() and
drm_gem_destroy() are all fine in that regard.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Try to keep all functions that handle DRM file_operations in drm_fops.c
so internal helpers can be marked static later.
This makes the split between the 3 core files more obvious:
- drm_stub.c: DRM device allocation/destruction and management
- drm_fops.c: DRM file_operations (except for ioctl)
- drm_drv.c: Global DRM init + ioctl handling
Well, ioctl handling is still spread throughout hundreds of source files,
but at least the others are clearly defined this way.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
All bus drivers do device setup themselves. This requires us to adjust all
of them if we introduce new core features. Thus, merge all these into a
uniform drm_dev_register() helper.
Note that this removes the drm_lastclose() error path for AGP as it is
horribly broken. Moreover, no bus driver called this in any other error
path either. Instead, we use the recently introduced AGP cleanup helpers.
We also keep a DRIVER_MODESET condition around pci_set_drvdata() to keep
semantics.
[airlied: keep passing flags through so drivers don't oops on load]
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Instead of managing device allocation+initialization in each bus-driver,
we should do that in a central place. drm_fill_in_dev() already does most
of it, but also requires the global drm lock for partial AGP device
registration.
Split both apart so we have a clean device initialization/allocation
phase, and a registration phase.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
All drivers embed gem-objects into their own buffer objects. There is no
reason to keep drm_gem_object_alloc(), gem->driver_private and
->gem_init_object() anymore.
New drivers are highly encouraged to do the same. There is no benefit in
allocating gem-objects separately.
Cc: Dave Airlie <airlied@gmail.com>
Cc: Alex Deucher <alexdeucher@gmail.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Ben Skeggs <skeggsb@gmail.com>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
There is no reason to keep the gem object separately allocated. nouveau is
the last user of gem_obj->driver_private, so if we embed it, we can get
rid of 8bytes per gem-object.
The implementation follows the radeon driver. bo->gem is only valid, iff
the bo was created via the gem helpers _and_ iff the user holds a valid
gem reference. That is, as the gem object holds a reference to the
nouveau_bo. If you use nouveau_ref() to gain a bo reference, you are not
guaranteed to also hold a gem reference. The gem object might get
destroyed after the last user drops the gem-ref via
drm_gem_object_unreference(). Use drm_gem_object_reference() to gain a
gem-reference.
For debugging, we can use bo->gem.filp != NULL to test whether a gem-bo is
valid. However, this shouldn't be used for real functionality to avoid
gem-internal dependencies.
Note that the implementation follows the previous style. However, we no
longer can check for bo->gem != NULL to test for a valid gem object. This
wasn't done before, so we should be safe now.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
When booting with i915.fastboot=1, we always take tha code path and end
up undoing what we're trying to do with adjusted_mode.
Hopefully, as the fastboot hardware readout code is using adjusted_mode
as well, it should be equivalent.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Instead of it just being on the mailing list, let's put Jesse's
explanation next to the code in question.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
During system boot up, by default, the power gate for render, media and
display well still power gated. Normally, BIOS will turn off the power
gate. In the BIOS-less system, the driver need to turn off the power
gate very early during driver load.
v2: Move this to intel_uncore_sanitize to allow it to get call during
resume path. (Daniel)
v3: Remove redundant write 0 to DPIO_CTL, and use DPIO_RESET instead of
just 0x1 (Ville)
Add turn of power gate for display 2d/render well/media well.
v4: Remove toggle cmnreset in intel_uncore_sanitize. Cmnreset should
toggle after CRI clock source has been selected. Jesse DPIO reset patch
which toggle the cmnreset in intel_modeset_init_hw() should handle it.
(Ville)
Signed-off-by: Chon Ming Lee <chon.ming.lee@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
DPIO needs to have common reset de-asserted on soft resets like boot and
S3. In some cases, the BIOS will have done this for us, but it should
be safe to do at runtime as well, as long as we do it when the pipes are
otherwise off.
v2: update bit name to match docs better (Ville)
reset after CRI clock select (Ville)
References: https://bugs.freedesktop.org/show_bug.cgi?id=69166
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
for igt test case.
v2: remove trailing spaces and fix conflicts
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@gmail.com>
[danvet:
- make it comipile
- s/IS_HASWELL/HAS_PSR/]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
After applying wait-boost we often find ourselves stuck at higher clocks
than required. The current threshold value requires the GPU to be
continuously and completely idle for 313ms before it is dropped by one
bin. Conversely, we require the GPU to be busy for an average of 90% over
a 84ms period before we upclock. So the current thresholds almost never
downclock the GPU, and respond very slowly to sudden demands for more
power. It is easy to observe that we currently lock into the wrong bin
and both underperform in benchmarks and consume more power than optimal
(just by repeating the task and measuring the different results).
An alternative approach, as discussed in the bspec, is to use a
continuous threshold for upclocking, and an average value for downclocking.
This is good for quickly detecting and reacting to state changes within a
frame, however it fails with the common throttling method of waiting
upon the outstanding frame - at least it is difficult to choose a
threshold that works well at 15,000fps and at 60fps. So continue to use
average busy/idle loads to determine frequency change.
v2: Use 3 power zones to keep frequencies low in steady-state mostly
idle (e.g. scrolling, interactive 2D drawing), and frequencies high
for demanding games. In between those end-states, we use a
fast-reclocking algorithm to converge more quickly on the desired bin.
v3: Bug fixes - make sure we reset adj after switching power zones.
v4: Tune - drop the continuous busy thresholds as it prevents us from
choosing the right frequency for glxgears style swap benchmarks. Instead
the goal is to be able to find the right clocks irrespective of the
wait-boost.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: Stéphane Marchesin <stephane.marchesin@gmail.com>
Cc: Owen Taylor <otaylor@redhat.com>
Cc: "Meng, Mengmeng" <mengmeng.meng@intel.com>
Cc: "Zhuang, Lena" <lena.zhuang@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
If we encounter a situation where the CPU blocks waiting for results
from the GPU, give the GPU a kick to boost its the frequency.
This should work to reduce user interface stalls and to quickly promote
mesa to high frequencies - but the cost is that our requested frequency
stalls high (as we do not idle for long enough before rc6 to start
reducing frequencies, nor are we aggressive at down clocking an
underused GPU). However, this should be mitigated by rc6 itself powering
off the GPU when idle, and that energy use is dependent upon the workload
of the GPU in addition to its frequency (e.g. the math or sampler
functions only consume power when used). Still, this is likely to
adversely affect light workloads.
In particular, this nearly eliminates the highly noticeable wake-up lag
in animations from idle. For example, expose or workspace transitions.
(However, given the situation where we fail to downclock, our requested
frequency is almost always the maximum, except for Baytrail where we
manually downclock upon idling. This often masks the latency of
upclocking after being idle, so animations are typically smooth - at the
cost of increased power consumption.)
Stéphane raised the concern that this will punish good applications and
reward bad applications - but due to the nature of how mesa performs its
client throttling, I believe all mesa applications will be roughly
equally affected. To address this concern, and to prevent applications
like compositors from permanently boosting the RPS state, we ratelimit the
frequency of the wait-boosts each client recieves.
Unfortunately, this techinique is ineffective with Ironlake - which also
has dynamic render power states and suffers just as dramatically. For
Ironlake, the thermal/power headroom is shared with the CPU through
Intelligent Power Sharing and the intel-ips module. This leaves us with
no GPU boost frequencies available when coming out of idle, and due to
hardware limitations we cannot change the arbitration between the CPU and
GPU quickly enough to be effective.
v2: Limit each client to receiving a single boost for each active period.
Tested by QA to only marginally increase power, and to demonstrably
increase throughput in games. No latency measurements yet.
v3: Cater for front-buffer rendering with manual throttling.
v4: Tidy up.
v5: Sadly the compositor needs frequent boosts as it may never idle, but
due to its picking mechanism (using ReadPixels) may require frequent
waits. Those waits, along with the waits for the vrefresh swap, conspire
to keep the GPU at low frequencies despite the interactive latency. To
overcome this we ditch the one-boost-per-active-period and just ratelimit
the number of wait-boosts each client can receive.
Reported-and-tested-by: Paul Neumann <paul104x@yahoo.de>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68716
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: Stéphane Marchesin <stephane.marchesin@gmail.com>
Cc: Owen Taylor <otaylor@redhat.com>
Cc: "Meng, Mengmeng" <mengmeng.meng@intel.com>
Cc: "Zhuang, Lena" <lena.zhuang@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[danvet: No extern for function prototypes in headers.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When we switched to always using a timeout in conjunction with
wait_seqno, we lost the ability to detect missed interrupts. Since, we
have had issues with interrupts on a number of generations, and they are
required to be delivered in a timely fashion for a smooth UX, it is
important that we do log errors found in the wild and prevent the
display stalling for upwards of 1s every time the seqno interrupt is
missed.
Rather than continue to fix up the timeouts to work around the interface
impedence in wait_event_*(), open code the combination of
wait_event[_interruptible][_timeout], and use the exposed timer to
poll for seqno should we detect a lost interrupt.
v2: In order to satisfy the debug requirement of logging missed
interrupts with the real world requirments of making machines work even
if interrupts are hosed, we revert to polling after detecting a missed
interrupt.
v3: Throw in a debugfs interface to simulate broken hw not reporting
interrupts.
v4: s/EGAIN/EAGAIN/ (Imre)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Imre Deak <imre.deak@intel.com>
[danvet: Don't use the struct typedef in new code.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We missed adding a few cleanup steps for recent additions.
Reviewer: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This patch attempts to clean up the ring/IA scaling programming in the
following ways.
1. Fix the comment about the DDR frequency. The math is 266MHz, not
133MHz. Formula was right, docs are wrong.
2. Mask the DCLK register since I don't know how it is defined on future
platforms.
3. use mult_frac instead of magic math.
This helps for future platform enabling.
v2: Actually use the right patch. The v1 was a mix of things, none of
which was right. Note that due to rounding, we actually get different
values (slightly higher) for the effective ring frequency.
v3: Use 1.25 instead of 1.33 as the original code did. (Jesse)
CC: Jesse Barnes <jbarnes@virtuousgeek.org>
CC: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Thierry Reding