drm/i915: Set adjustment to zero on Up/Down interrupts if freq is already max/min
When we reach the user's RPS limits, stop requesting an adjustment. Even
though we will clamp the requested frequency later, we rely on interrupt
masking to disable further adjustments in the same direction. Even
though it is unlikely (one scenario is a bug in the driver, another is
careful manipulation through the uAPI) if we keep exponentially
increasing the adjustment value, it will wrap and cause a negative
adjustment.
Chris Wilson [Thu, 19 Jan 2017 11:37:49 +0000 (11:37 +0000)]
drm/i915: Release temporary load-detect state upon switching
After we call drm_atomic_commit() on the load-detect state, we can free
our local reference. Upon restore, we only apply and free the previous state.
Fixes: 0853695c3ba4 ("drm: Add reference counting to drm_atomic_state") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.10-rc1+ Link: http://patchwork.freedesktop.org/patch/msgid/20170119113749.2517-1-chris@chris-wilson.co.uk Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Chris Wilson [Mon, 16 Jan 2017 15:21:30 +0000 (15:21 +0000)]
drm/i915: Remove i915_vma_create from VMA API
With the introduce of i915_vma_instance() for obtaining the VMA
singleton for a (obj, vm, view) tuple, we can remove the
i915_vma_create() in favour of a single entry point. We do incur a
lookup onto an empty tree, but the i915_vma_create() were being called
infrequently and during initialisation, so the small overhead is
negligible.
v2: Drop the i915_ prefix from the now static vma_create() function
Chris Wilson [Mon, 16 Jan 2017 15:21:28 +0000 (15:21 +0000)]
drm/i915: Rename some warts in the VMA API
Whilst writing testcases to exercise the VMA API, some oddities came to
light, such as i915_gem_obj_lookup_or_create(). Joonas suggested
i915_vma_instance() as a neat replacement, so rename them, move them to
i915_vma.c and add some kerneldoc as a sugary bonus.
Chris Wilson [Mon, 16 Jan 2017 15:21:27 +0000 (15:21 +0000)]
drm/i915: Track pinned vma in intel_plane_state
With atomic plane states we are able to track an allocation right from
preparation, during use and through to the final free after being
swapped out for a new plane. We can couple the VMA we pin for the
framebuffer (and its rotation) to this lifetime and avoid all the clumsy
lookups in between.
v2: Remove residual vma on plane cleanup (Chris)
v3: Add a description for the vma destruction in
intel_plane_destroy_state (Maarten)
Anusha Srivatsa [Wed, 18 Jan 2017 16:05:58 +0000 (08:05 -0800)]
drm/i915/get_params: Add HuC status to getparams
This patch will allow for getparams to return the status of the HuC.
As the HuC has to be validated by the GuC this patch uses the validated
status to show when the HuC is loaded and ready for use. You cannot use
the loaded status as with the GuC as the HuC is verified after it is
loaded and is not usable until it is verified.
v2: removed the forewakes as the registers are already force-woken.
(T.Ursulin)
v3: rebased on top of drm-tip. Removed any reference to intel_huc.h
v4: rebased. Rename I915_PARAM_HAS_HUC to I915_PARAM_HUC_STATUS.
Remove intel_is_huc_valid() since it is used only in one place.
Put the case of I915_PARAM_HAS_HUC() in the right place.
v5: rebased. Add a comment to specify that I915_READ(reg)
does not read garbage value. The register HUC_STATUS2 is force
woken and no rpm is needed.
Anusha Srivatsa [Wed, 18 Jan 2017 16:05:57 +0000 (08:05 -0800)]
drm/i915/huc: Support HuC authentication
The HuC authentication is done by host2guc call. The HuC RSA keys
are sent to GuC for authentication.
v2: rebased on top of drm-tip. Changed name format and upped
version 1.7.
v3: changed wait_for_atomic to wait_for
v4: rebased. Rename intel_huc_auh() to intel_guc_auth_huc()
and place the prototype in intel_guc.h,correct the comments.
v5: rebased. Moved intel_guc_auth_huc from i915_guc_submission.c
to intel_uc.c.Update dev to dev_priv in intel_guc_auth_huc().
Renamed HOST2GUC_ACTION_AUTHENTICATE_HUC TO INTEL_GUC_ACTION_
AUTHENTICATE_HUC
v6: rebased. Add newline on DRM_ERRORs that already dont have one.
v7: rebased. Replace wait_for with intel_wait_for_register() since
the latter employs sleep optimisations for quick responses- as pointed
out by Chris Wilson.
v8: rebased. Cleanup the intel_guc_auth_huc() by removing checks
already performed in earlier functions. Make comments more descriptive.
v9: rebased. Changed the bias for pinning the HuC object. Move
intel_guc_auth_huc() to intel_huc.c. Change DRM_DEBUGs to DRM_ERRORs
in intel_guc_auth_huc(). Add return status to DRM_ERRORs.
v10: Remove message not required for the user..
Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Tested-by: Xiang Haihao <haihao.xiang@intel.com> Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by: Alex Dai <yu.dai@intel.com> Signed-off-by: Peter Antoine <peter.antoine@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484755558-1234-5-git-send-email-anusha.srivatsa@intel.com
Anusha Srivatsa [Wed, 18 Jan 2017 16:05:56 +0000 (08:05 -0800)]
drm/i915/huc: Add debugfs for HuC loading status check
Add debugfs entry for HuC loading status check.
v2: rebased on top of drm-tip.
Cc: Michal wajdeczko <michal.wajdeczko@intel.com> Tested-by: Xiang Haihao <haihao.xiang@intel.com> Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by: Alex Dai <yu.dai@intel.com> Signed-off-by: Peter Antoine <peter.antoine@intel.com> Reviewed-by: Jeff McGee <jeff.mcgee@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484755558-1234-4-git-send-email-anusha.srivatsa@intel.com
Anusha Srivatsa [Wed, 18 Jan 2017 16:05:53 +0000 (08:05 -0800)]
drm/i915/huc: Add HuC fw loading support
The HuC loading process is similar to GuC. The intel_uc_fw_fetch()
is used for both cases.
HuC loading needs to be before GuC loading. The WOPCM setting must
be done early before loading any of them.
v2: rebased on-top of drm-intel-nightly.
removed if(HAS_GUC()) before the guc call. (D.Gordon)
update huc_version number of format.
v3: rebased to drm-intel-nightly, changed the file name format to
match the one in the huc package.
Changed dev->dev_private to to_i915()
v4: moved function back to where it was.
change wait_for_atomic to wait_for.
v5: rebased. Changed the year in the copyright message to reflect
the right year.Correct the comments,remove the unwanted WARN message,
replace drm_gem_object_unreference() with i915_gem_object_put().Make the
prototypes in intel_huc.h non-extern.
v6: rebased. Update the file construction done by HuC. It is similar to
GuC.Adopted the approach used in-
https://patchwork.freedesktop.org/patch/104355/ <Tvrtko Ursulin>
v7: Change dev to dev_priv in macro definition.
Corrected comments.
v8: rebased on top of drm-tip. Updated functions intel_huc_load(),
intel_huc_init() and intel_uc_fw_fetch() to accept dev_priv instead of
dev. Moved contents of intel_huc.h to intel_uc.h.
v9: change SKL_FW_ to SKL_HUC_FW_. Add intel_ prefix to guc_wopcm_size().
Remove unwanted checks in intel_uc.h. Rename huc_fw in struct intel_huc to
simply fw to avoid redundency.
v10: rebased. Correct comments. Make intel_huc_fini() accept dev_priv
instead of dev like intel_huc_init() and intel_huc_load().Move definition
to i915_guc_reg.h from intel_uc.h. Clean DMA_CTRL bits after HuC DMA
transfer in huc_ucode_xfer() instead of guc_ucode_xfer(). Add suitable
WARNs to give extra info.
v11: rebased. Add proper bias for HuC and make sure there are
asserts on failure by using guc_ggtt_offset_vma(). Introduce
intel_huc.c and remove intel_huc_loader.c since it has functions that
do more than just loading.Correct year in copyright.
v12: remove invalidates that are not required anymore.
Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Tested-by: Xiang Haihao <haihao.xiang@intel.com> Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by: Alex Dai <yu.dai@intel.com> Signed-off-by: Peter Antoine <peter.antoine@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484755558-1234-1-git-send-email-anusha.srivatsa@intel.com
Clint Taylor [Wed, 18 Jan 2017 21:38:43 +0000 (13:38 -0800)]
drm/i915: prevent crash with .disable_display parameter
The .disable_display parameter was causing a fatal crash when fbdev
was dereferenced during driver init.
V1: protection in i915_drv.c
V2: Moved protection to intel_fbdev.c
Fixes: 43cee314345a ("drm/i915/fbdev: Limit the global async-domain synchronization")
Testcase: igt/drv_module_reload/basic-no-display Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484775523-29428-1-git-send-email-clinton.a.taylor@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Lukas Wunner <lukas@wunner.de> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: <stable@vger.kernel.org> # v4.8+ Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
drm/i915: Avoid drm_atomic_state_put(NULL) on error paths
The error paths in hsw_trans_edp_pipe_A_crc_wa() and
intel_prepare_reset() would potentially call drm_atomic_state_put with a
NULL state, which would lead to a NULL pointer dereference.
Found by coverity.
v2: Improve the error paths. (Chris)
Fixes: 0853695c3ba4 ("drm: Add reference counting to drm_atomic_state") Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.10-rc1+ Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484742868-9551-1-git-send-email-ander.conselvan.de.oliveira@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 15 Jan 2017 12:58:25 +0000 (12:58 +0000)]
drm/i915: Avoid drm_atomic_state_put(NULL) in intel_display_resume
intel_display_resume() may be called without an atomic state to restore,
i.e. dev_priv->modeset_reset_restore state is NULL. One such case is
following a lid open/close event and the forced modeset in
intel_lid_notify().
Reported-by: Stefan Seyfried <stefan.seyfried@googlemail.com> Tested-by: Stefan Seyfried <stefan.seyfried@googlemail.com> Fixes: 0853695c3ba4 ("drm: Add reference counting to drm_atomic_state") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: <drm-intel-fixes@lists.freedesktop.org> # v4.10-rc1+ Link: http://patchwork.freedesktop.org/patch/msgid/20170115125825.18597-1-chris@chris-wilson.co.uk Reviewed-by: Ander Conselvan de Oliveira <conselvan2@gmail.com>
Chris Wilson [Tue, 17 Jan 2017 15:59:06 +0000 (17:59 +0200)]
drm/i915: Detect a failed GPU reset+recovery
If we can't recover the GPU after the reset, mark it as wedged to cancel
the outstanding tasks and to prevent new users from trying to use the
broken GPU.
v2: Check the same ring is hung again before declaring the reset broken.
v3: use engine_stalled (Mika)
Mika Kuoppala [Tue, 17 Jan 2017 15:59:03 +0000 (17:59 +0200)]
drm/i915: Cleanup request skip decision
Since we now only skip banned contexts, preventing the skip of default
contexts is no longer sensible. For a similar argument as before
'commit 7ec73b7e36d0 ("drm/i915: Only skip requests once a context is banned")'
we end up with an inconsistent API if we only mark future execbufs from
the default ctx as banned but fail to mark those currently executing
as failed.
Chris Wilson [Tue, 17 Jan 2017 15:59:01 +0000 (17:59 +0200)]
drm/i915: Move engine reset preparation to i915_gem_reset_prepare()
Now that we have prepare/finish routines for the GEM reset, move the
disabling of the engine->irq_tasklet into them to reduce repetition. The
device irq enable/disable is split out to ensure it is run first and
last always (even if the GPU reset fails).
Chris Wilson [Mon, 16 Jan 2017 13:06:22 +0000 (13:06 +0000)]
drm/i915/psr: Fix compiler warnings for hsw_psr_disable()
drivers/gpu/drm/i915/intel_psr.c:634 hsw_psr_disable() warn: if statement not indented
drivers/gpu/drm/i915/intel_psr.c:644 hsw_psr_disable() warn: if statement not indented
Fixes: 3fcb0ca1d8db ("drm/i915/psr: fix blank screen issue for psr2") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Cc: Patil Deepti <deepti.patil@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170116130622.20369-2-chris@chris-wilson.co.uk Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Anusha Srivatsa [Sat, 14 Jan 2017 01:17:05 +0000 (17:17 -0800)]
drm/i915/huc: Unified css_header struct for GuC and HuC
HuC firmware css header has almost exactly same definition as GuC
firmware except for the sw_version. Also, add a new member fw_type
into intel_uc_fw to indicate what kind of fw it is. So, the loader
will pull right sw_version from header.
v2: rebased on-top of drm-intel-nightly
v3: rebased. Rename device_id to guc_branch_client_version,
make guc_sw_version a union. <Jeff Mcgee>. Put UC_FW_TYPE_GUC
and UC_FW_TYPE_HUC into an enum.
v4: rebased on top of drm-tip.Update dev to dev_priv in
intel_uc_fw_fetch.
v5: rebased. Add INTEL_ prefix to an enum. Add fw_type declaration
from patch 1.Combine two different unions for huc and guc version,
reserved etc into one union with two structs.
v6: rebased. Change fw_type to enum.
v7: rebased. Rename the enum fw_type to intel_uc_fw_type.
Cc: Michal Wajdeczko <michal.wajdeczko.@intel.com> Tested-by: Xiang Haihao <haihao.xiang@intel.com> Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by: Alex Dai <yu.dai@intel.com> Signed-off-by: Peter Antoine <peter.antoine@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484356631-16139-3-git-send-email-anusha.srivatsa@intel.com
Anusha Srivatsa [Sat, 14 Jan 2017 01:17:04 +0000 (17:17 -0800)]
drm/i915/guc: Make the GuC fw loading helper functions general
Rename some of the GuC fw loading code to make them more general. We
will utilise them for HuC loading as well.
s/intel_guc_fw/intel_uc_fw/g
s/GUC_FIRMWARE/INTEL_UC_FIRMWARE/g
Struct intel_guc_fw is renamed to intel_uc_fw. Prefix of tts members,
such as 'guc' or 'guc_fw' either is renamed to 'uc' or removed for
same purpose.
v2: rebased on top of nightly.
reapplied the search/replace as upstream code as changed.
v3: removed G from messages in shared fw fetch function.
v4: rebased.Updated dev to dev_priv in intel_guc_setup(), guc_fw_getch()
and intel_guc_init().
v5: rebased. Remove uint32_t fw_type to patch 2. Add INTEL_ prefix for
fields in enum intel_uc_fw_status. Remove uc_dev field since its never
used.Rename uc_fw to just fw and guc_fw to fw to avoid redundency.
v6: rebased. Remove sections of code that were commented and no longer
required.
v7: rebased. Remove uc_fw_ prefix from path and obj fields
in intel_uc_fw struct as suggested by Michal.
v8: rebased. Add declaration of intel_guc_wopcm_size() in
this patch instead of patch 3.
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Signed-off-by: Anusha Srivatsa <anusha.srivatsa@intel.com> Signed-off-by: Alex Dai <yu.dai@intel.com> Signed-off-by: Peter Antoine <peter.antoine@intel.com> Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484356631-16139-2-git-send-email-anusha.srivatsa@intel.com
Michal Wajdeczko [Fri, 13 Jan 2017 17:41:57 +0000 (17:41 +0000)]
drm/i915/guc: Move GuC log related functions into dedicated file
Functions supporting GuC logging capabilities were spread across
many files, with unnecessary exposures and mixed with unrelated
code. Dedicate file will make maintenance of all GuC functions
easier as more functions are coming to support GuC submissions.
Ville Syrjälä [Mon, 7 Nov 2016 20:20:54 +0000 (22:20 +0200)]
drm/i915: Ignore bogus plane coordinates on SKL when the plane is not visible
When the plane is invisible we may have all sorts of bogus stuff
in the coordinates, which we must ignore or else we might fail the
plane update. This started to happen on SKL when I moved the plane
offset computation to happen in the check phase. Previously we
happily ignored it all since we never called the update_plane hook
with an invisible plane.
Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com> Cc: drm-intel-fixes@lists.freedesktop.org Fixes: b63a16f6cd89 ("drm/i915: Compute display surface offset in the plane check hook for SKL+") Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98258
Testcase: igt/pm_rpm/legacy-planes
Testcase: igt/pm_rpm/universal-planes Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1478550057-24864-3-git-send-email-ville.syrjala@linux.intel.com
Zhenyu Wang [Tue, 17 Jan 2017 14:06:11 +0000 (22:06 +0800)]
drm/i915: Fix a typo in vgt_balloon_space()
Commit 625d988acc28 ("drm/i915: Extract reserving space in the GTT to a
helper") introduces this typo which can cause a driver loading failure
in Linux GVT-g guest.
Fixes: 625d988acc28 ("drm/i915: Extract reserving space in the GTT to a helper") Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Zhiyuan Lv <zhiyuan.lv@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com> Signed-off-by: Zhi Wang <zhi.a.wang@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484661972-9366-1-git-send-email-zhi.a.wang@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 29 Dec 2016 14:40:37 +0000 (14:40 +0000)]
drm/i915: Skip switch to kernel context if already done
Some engines are never user or already sitting idle in the kernel
context and for those we can skip flushing the current context for
i915_gem_switch_to_kernel_context(). We used to perform this
optimisation but that was removed for convenience of converting over to
multiple timelines and handling the pending request queues.
From the perspective of writing selftests, reducing the number of
background operations on the engines makes defining assertions easier.
Chris Wilson [Sat, 14 Jan 2017 16:23:33 +0000 (16:23 +0000)]
drm/i915: Construct a request even if the GPU is currently hung
As we now have the ability to directly reset the GPU from the waiter
(and so do not need to drop the lock in order to let the reset proceed)
and also do not lose requests over a reset, we can now simply queue the
request to occur after the reset rather than roundtripping to userspace
(or worse failing with EIO).
Chris Wilson [Sun, 15 Jan 2017 13:47:46 +0000 (13:47 +0000)]
drm/i915: Catch attempting to use the aliasing_gtt's drm_mm
The aliasing_gtt is just that, an alias of the global GTT. We do not
populate it directly, instead we always use the global GTT. Catch any
attempt to incorrectly allocate ranges from the aliasing_gtt.
Chris Wilson [Sat, 14 Jan 2017 10:51:12 +0000 (10:51 +0000)]
drm/i915: Use __printf markup to silence compiler
drivers/gpu/drm/i915/i915_gpu_error.c: In function ‘i915_error_vprintf’:
drivers/gpu/drm/i915/i915_gpu_error.c:137:3: warning:
function might be possible candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]
len = vsnprintf(NULL, 0, f, tmp);
^~~
drivers/gpu/drm/i915/i915_gpu_error.c:144:2: warning:
function might be possible candidate for ‘gnu_printf’ format attribute [-Wsuggest-attribute=format]
len = vsnprintf(e->buf + e->bytes, e->size - e->bytes, f, args);
^~~
Since commit 058d88c4330f ("drm/i915: Track pinned VMA"), there is only
one user of i915_ggtt_view_normal rodate. Just treat NULL as no special
view in pin_to_display() like everywhere else.
It is only being used to clear a struct and set the type, after which it
is overwritten. Since we no longer check the unset bits of the union,
skipping the clear is permissible.
Chris Wilson [Sat, 14 Jan 2017 00:28:25 +0000 (00:28 +0000)]
drm/i915: Convert i915_ggtt_view to use an anonymous union
Reading the ggtt_views is much more pleasant without the extra
characters from specifying the union (i.e. ggtt_view.partial rather than
ggtt_view.params.partial). To make this work inside i915_vma_compare()
with only a single memcmp requires us to ensure that there are no
uninitialised bytes within each branch of the union (we make sure the
structs are packed) and we need to store the size of each branch.
v4: Rewrite changelog and add comments explaining the assert.
Chris Wilson [Sat, 14 Jan 2017 00:28:23 +0000 (00:28 +0000)]
drm/i915: Compact memcmp in i915_vma_compare()
In preparation for the next patch to convert to using an anonymous union
and leaving the excess bytes in the union uninitialised, we first need
to make sure we do not compare using those uninitialised bytes. We also
want to preserve the compactness of the code, avoiding a second call to
memcmp or introducing a switch, so we take advantage of using the type
as an encoded size (as well as a unique identifier for each type of view).
v2: Add the rationale for why we encode size into ggtt_view.type as a
comment before the memcmp()
v3: Use a switch to also assert that no two i915_ggtt_view_type have the same
value.
Chris Wilson [Sat, 14 Jan 2017 00:28:22 +0000 (00:28 +0000)]
drm/i915: Mark the ggtt_view structs as packed
In the next few patches, we will depend upon there being no
uninitialised bits inside the ggtt_view. To ensure this we add the
__packed attribute and double check with a build bug that gcc hasn't
expanded the struct to include some padding bytes.
Reports live state of PSR2 form PSR2_STATUS register.
bit field 31:28 gives the live state of PSR2.
It can be used to check if system is in deep sleep,
selective update or selective update standby.
During video play back, we can use this to check
if system is entering SU mode or not.
when system is in idle state, DEEP_SLEEP(8) must be entered.
When video playback is happening, system must be in
SLEEP(3 / selective update) or SU_STANDBY( 6 / selective update standby)
v2: (Rodrigo)
- Remove EDP_PSR2_STATUS_TG_ON=a ,instead use ARRAY_SIZE
Psr2 is enabled only for y cordinate panels.Once GTC (global time code)
is implemented,this restriction is removed so that psr2
can work on panels without y cordinate support.
v2: (Rodrigo)
- Move the check to intel_psr_match_conditions
Program EDP_PSR_DEBUG_CTL (PSR_MASK) to enable system
to go to deep sleep while in psr2.PSR2_STATUS bit 31:28
should report value 8 , if system enters deep sleep state.
Also, EDP_FRAMES_BEFORE_SU_ENTRY is set 1 , if not set,
flickering is observed on psr2 panel.
v2: (Ilia Mirkin)
- Remove duplicate bit definition 25:27
As per bpsec, CHICKEN_TRANS_EDP bit 12 ,15 must be programmed in
psr2 enable sequence.
bit 12 : Program Transcoder EDP VSC DIP header with a valid setting for
PSR2 and Set CHICKEN_TRANS_EDP(0x420cc) bit 12 for programmable
header packet.
bit 15 : Set CHICKEN_TRANS_EDP(0x420cc) bit 15 if Y coordinate is supported
v2: (Rodrigo)
- move CHICKEN_TRANS_EDP bit set logic right after setup_vsc
v3:(Rodrigo)
- initialize chicken_trans to CHICKEN_TRANS_BIT12 instead of 0
Chris Wilson [Wed, 11 Jan 2017 21:09:26 +0000 (21:09 +0000)]
drm/i915: Assert that we have allocated the drm_mm_node upon pinning
We currently check after the slow path that the vma is bound correctly,
but we don't currently check after the fast path. This is important in
case we accidentally take the fast path and leave the vma misplaced.
As per edp1.4 spec , alpm is required for psr2 operation as it's
used for all psr2 main link power down management and alpm enable
bit must be set for psr2 operation.
Screen freeze observed if AUX_FRAME_SYNC is not disabled
on psr2 exit.AUX_FRAME_SYNC needed for psr2 is enabled during
psr2 entry. It must be disabled on psr2 exit.
Psr1 and psr2 are mutually exclusive,ie when psr2 is enabled,
psr1 should be disabled.When psr2 is exited , bit 31 of reg
PSR2_CTL must be set to 0 but currently bit 31 of SRD_CTL
(psr1 control register)is set to 0.
Also ,PSR2_IDLE state is looked up from SRD_STATUS(psr1 register)
instead of PSR2_STATUS register, which has wrong data, resulting
in blankscreen.
hsw_enable_source is split into hsw_enable_source_psr1 and
hsw_enable_source_psr2 for easier code review and maintenance,
as suggested by rodrigo and jim.
v2: (Rodrigo)
- Rename hsw_enable_source_psr* to intel_enable_source_psr*
v3: (Rodrigo)
- In hsw_psr_disable ,
1) for psr active case, handle psr2 followed by psr1.
2) psr inactive case, handle psr2 followed by psr1
v4:(Rodrigo)
- move psr2 restriction(32X20) to match_conditions function
returning false and fully blocking PSR to a new patch before
this one.
Chris Wilson [Thu, 12 Jan 2017 16:45:59 +0000 (16:45 +0000)]
drm/i915: Fix up kerneldoc parameters for i915_gem_gtt_*()
Parameter: good.
Parameter - bad.
One day I'll learn the syntax.
Fixes: 625d988acc28 ("drm/i915: Extract reserving space in the GTT to a helper") Fixes: e007b19d7ba7 ("drm/i915: Use the MRU stack search after evicting") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170112164559.27232-1-chris@chris-wilson.co.uk Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Chris Wilson [Thu, 12 Jan 2017 13:04:31 +0000 (13:04 +0000)]
drm/i915: Declare i915_gem_object_create_internal() as taking phys_addr_t size
The internal object is a collection of struct pages and so is
intrinsically linked to the available physical memory on the machine,
and not an arbitrary type from the uabi. Use phys_addr_t so the link
between size and memory consumption is clear, and then double check that
we don't overflow the maximum object size.
v2: Also assert that size is not zero - a mistake I made a few times
while writing selftests.
Chris Wilson [Thu, 12 Jan 2017 11:00:49 +0000 (11:00 +0000)]
drm/i915: Invalidate the guc ggtt TLB upon insertion
Move the GuC invalidation of its ggtt TLB to where we perform the ggtt
modification rather than proliferate it into all the callers of the
insert (which may or may not in fact have to do the insertion).
v2: Just do the guc invalidate unconditionally, (afaict) it has no impact
without the guc loaded on gen8+
v3: Conditionally invalidate the guc - just in case that register has
not been validated for other modes.
drm/i915: Remove useless casts to intel_plane_state
The visible member used to be in intel_plane_state->visible,
but has been moved to drm_plane_state->visible. In the conversion
some casts were left in that are now useless.
to_intel_plane_state(x)->base.visible is the same as x->visible,
so use the latter to clear up the code a little.
Michel Thierry [Thu, 12 Jan 2017 04:18:09 +0000 (20:18 -0800)]
drm/i915: Update i915_reset parameter for kerneldoc
Since commit c033666a94b57 ("drm/i915: Store a i915 backpointer from
engine, and use it") i915_reset receives dev_priv, but the kerneldoc
was not updated.
Michel Thierry [Thu, 12 Jan 2017 04:18:08 +0000 (20:18 -0800)]
drm/i915: Keep i915_handle_error kerneldoc parameters together
And before the function description.
Tidy up from commit 14bb2c11796d70b ("drm/i915: Fix a buch of kerneldoc
warnings"), all others kerneldoc blocks look ok.
The WaDisableLSQCROPERFforOCL workaround has the side effect of
disabling an L3SQ optimization that has huge performance implications
and is unlikely to be necessary for the correct functioning of usual
graphic workloads. Userspace is free to re-enable the workaround on
demand, and is generally in a better position to determine whether the
workaround is necessary than the DRM is (e.g. only during the
execution of compute kernels that rely on both L3 fences and HDC R/W
requests).
The same workaround seems to apply to BDW (at least to production
stepping G1) and SKL as well (the internal workaround database claims
that it does for all steppings, while the BSpec workaround table only
mentions pre-production steppings), but the DRM doesn't do anything
beyond whitelisting the L3SQCREG4 register so userspace can enable it
when it sees fit. Do the same on KBL platforms.
Improves performance of the GFXBench4 gl_manhattan31 benchmark by 60%,
and gl_4 (AKA car chase) by 14% on a KBL GT2 running Mesa master --
This is followed by a regression of 35% and 10% respectively for the
same benchmarks and platform caused by my recent patch series
switching userspace to use the dataport constant cache instead of the
sampler to implement uniform pull constant loads, which caused us to
hit more heavily the L3 cache (and on platforms other than KBL had the
opposite effect of improving performance of the same two benchmarks).
The overall effect on KBL of this change combined with the recent
userspace change is respectively 4.6% and 2.6%. SynMark2 OglShMapPcf
was affected by the constant cache changes (though it improved as it
did on other platforms rather than regressing), but is not
significantly affected by this patch (with statistical significance of
5% and sample size 20).
v2: Drop some more code to avoid unused variable warning.
Fixes: 738fa1b3123f ("drm/i915/kbl: Add WaDisableLSQCROPERFforOCL")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99256 Signed-off-by: Francisco Jerez <currojerez@riseup.net> Cc: Matthew Auld <matthew.william.auld@gmail.com> Cc: Eero Tamminen <eero.t.tamminen@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: beignet@lists.freedesktop.org Cc: <stable@vger.kernel.org> # v4.7+ Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
[Removed double Fixes tag] Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1484217894-20505-1-git-send-email-mika.kuoppala@intel.com
Michał Winiarski [Wed, 11 Jan 2017 15:17:39 +0000 (16:17 +0100)]
drm/i915/guc: Make sure vma containing firmware is GuC mappable
Since commit 4741da925fa3 ("drm/i915/guc: Assert that all GGTT offsets used
by the GuC are mappable"), we're asserting that GuC firmware is in the
GuC mappable range.
Except we're not pinning the object with bias, which means it's possible
to trigger this assert. Let's add a proper bias.
Fixes: 4741da925fa3 ("drm/i915/guc: Assert that all GGTT offsets used by the GuC are mappable") Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Tested-by: Tomi Sarvela <tomi.p.sarvela@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/20170111151739.28965-1-michal.winiarski@intel.com
Libin Yang [Thu, 12 Jan 2017 08:04:53 +0000 (16:04 +0800)]
ALSA: hda - add DP MST audio support
This patch adds the DP MST audio support on i915 platform and
it will enable dyn_pcm_assign feature.
DP MST supports several device entry on the same port and each
device entry can map to one pcm stream. For example, on i915,
there are 3 pins, and each pin has 3 device entries. This means
there should be 3x3 pcms. However, there is only 3 pipe lines in
i915. This means 3 pcms are actived at most at the same moment.
We will create 5 pcms (pin number + dev entry num - 1) in this case.
For the details, please refer commit a76056f2e57e
("ALSA: hda - hdmi dynamically bind PCM to pin when monitor hotplug")
Each device entry is a virtual pin. It is described by pin_nid and dev_id
in struct hdmi_spec_per_pin.
Zhenyu Wang [Mon, 9 Jan 2017 13:14:53 +0000 (21:14 +0800)]
drm/i915: check ppgtt validity when init reg state
Check if ppgtt is valid for context when init reg state. For gvt
context which has no i915 allocated ppgtt, failed to check that
would cause kernel null ptr reference error.
v2: remove !48bit ppgtt case as we'll always update before submit (Chris)
Chris Wilson [Wed, 11 Jan 2017 18:15:16 +0000 (18:15 +0000)]
drm/i915: Detect vma reserved for execbuf in evict-for-node
The vma->exec_list is still the only means we have for both reserving an
object in execbuf, and for constructing the eviction list. So during the
construction of the eviction list, we must treat anything already on the
exec_list as being pinned.
Yes, this sharing of two semantically different lists will be fixed! But
in the meantime, we have the issue that this is tripping up CI since we
started using i915_gem_gtt_reserve_node() + i915_gem_evict_for_node()
from the regular execbuf reservation path in commit 606fec956c0e
("drm/i915: Prefer random replacement before eviction search"):
Chris Wilson [Wed, 11 Jan 2017 14:08:58 +0000 (14:08 +0000)]
drm/i915: Add a sanity check that no request is submitted in the middle
It is an error to start a new request on the same timeline (ringbuffer)
as the current one before the current is submitted. If there are two
requests emitting to the ringbuffer at the same time, the operation is
undefined. We can catch this by checking for the timeline having a later
seqno than ours when we come to submit our request.
Currently we have this check at the end of __i915_add_request, but
having an early check as well isolates a failure in the caller versus a
failure in sealing the request (i.e. from inside __i915_add_request
itself). For example, CI is currently tripping over this late assertion
on ctg/ilk:
Chris Wilson [Wed, 11 Jan 2017 11:23:12 +0000 (11:23 +0000)]
drm/i915: Prefer random replacement before eviction search
Performing an eviction search can be very, very slow especially for a
range restricted replacement. For example, a workload like
gem_concurrent_blit will populate the entire GTT and then cause aperture
thrashing. Since the GTT is a mix of active and inactive tiny objects,
we have to search through almost 400k objects before finding anything
inside the mappable region, and as this search is required before every
operation performance falls off a cliff.
Instead of performing the full search, we do a trial replacement of the
node at a random location fitting the specified restrictions. We lose
the strict LRU property of the GTT in exchange for avoiding the slow
search (several orders of runtime improvement for gem_concurrent_blit
4KiB-global-gtt, e.g. from 5000s to 20s). The loss of LRU replacement is
(later) mitigated firstly by only doing replacement if we find no
freespace and secondly by execbuf doing a PIN_NONBLOCK search first before
it starts thrashing (i.e. the random replacement will only occur from the
already inactive set of objects).
v2: Ascii-art, and check preconditionst
v3: Rephrase final sentence in comment to explain why we don't bother
with if (i915_is_ggtt(vm)) for preferring random replacement.
Chris Wilson [Wed, 11 Jan 2017 11:23:10 +0000 (11:23 +0000)]
drm/i915: Use the MRU stack search after evicting
When we evict from the GTT to make room for an object, the hole we
create is put onto the MRU stack inside the drm_mm range manager. On the
next search pass, we can speed up a PIN_HIGH allocation by referencing
that stack for the new hole.
v2: Pull together the 3 identical implements (ahem, a couple were
outdated) into a common routine for allocating a node and evicting as
necessary.
v3: Detect invalid calls to i915_gem_gtt_insert()
v4: kerneldoc
Function hsw_psr_setup handles vsc header setup for psr1 and
skl_psr_setup_vsc handles vsc header setup for psr2.
Setup VSC header in function skl_psr_setup_vsc for psr2 support,
as per edp 1.4 spec, table 6-11:VSC SDP HEADER Extension for psr2
operation.
v2: (Jani)
- Initialize variables to 0
- intel_dp_get_y_cord_status and intel_dp_get_y_cord_status made static
- Correct indentation for continuation lines
- Change DP_PSR_Y_COORDINATE to DP_PSR2_SU_Y_COORDINATE_REQUIRED
- Change DPRX_FEATURE_ENUMERATION_LIST to DP_DPRX_*
- Change VSC_SDP_EXT_FOR_COLORIMETRY_SUPPORTED to DP_VSC_*
PSR2 vsc revision number hb2( as per table 6-11)is updated to
4 or 5 based on Y cordinate and Colorimetry Format as below
04h = 3D stereo + PSR/PSR2 + Y-coordinate.
05h = -3D stereo- + PSR/PSR2 + Y-coordinate + Pixel Encoding/Colorimetry
Format indication. A DP Source device is allowed to indicate the pixel
encoding/colorimetry format to the DP Sink device with VSC SDP only when
the DP Sink device supports it (
i.e.,VSC_SDP_EXTENSION_FOR_COLORIMETRY_SUPPORTED bit in the
DPRX_FEATURE_ENUMERATION_LIST register (DPCD Address 02210h, bit 3;
is set to 1).
v2: (Jani)
- Change DP_PSR_Y_COORDINATE to DP_PSR2_SU_Y_COORDINATE_REQUIRED.
- Add DP_PSR2_SU_GRANULARITY_REQUIRED.
- Change DPRX_FEATURE_ENUMERATION_LIST to DP_DPRX.
- Add GTC_CAP and AV_SYNC_CAP, other bits in DPRX_FEATURE_ENUMERATION_LIST.
v3: (Jani)
- Add support for bits 7:4 and 1 as per DP v1.4 for
DPRX_FEATURE_ENUMERATION_LIST.
Chris Wilson [Tue, 10 Jan 2017 14:47:34 +0000 (14:47 +0000)]
drm/i915: Replace 4096 with PAGE_SIZE or I915_GTT_PAGE_SIZE
Start converting over from the byte count to its semantic macro, either
we want to allocate the size of a physical page in main memory or we
want the size of a virtual page in the GTT. 4096 could mean either, but
PAGE_SIZE and I915_GTT_PAGE_SIZE are explicit and should help improve
code comprehension and future changes. In the future, we may want to use
variable GTT page sizes and so have the challenge of knowing which
hardcoded values were used to represent a physical page vs the virtual
page.
v2: Look for a few more 4096s to convert, discover IS_ALIGNED().
v3: 4096ul paranoia, make fence alignment a distinct value of 4096, keep
bdw stolen w/a as 4096 until we know better.
v4: Add asserts that i915_vma_insert() start/end are aligned to GTT page
sizes.
Chris Wilson [Tue, 10 Jan 2017 17:22:46 +0000 (17:22 +0000)]
drm/i915: Rename i915_gem_engine_cleanup() to engine_set_wedged()
It has been some time since i915_gem_engine_cleanup was only called from
the module unload path, and now it is only called when the GPU is
wedged. Mika complained that the name is confusing, especially in light
of the existence of i915_gem_cleanup_engines().
Chris Wilson [Tue, 10 Jan 2017 17:22:45 +0000 (17:22 +0000)]
drm/i915: Mark all incomplete requests as -EIO when wedged
Similarly to a normal reset, after we mark the GPU as wedged (completely
fubar and no more requests can be executed), set the error status on all
the in flight requests.
Chris Wilson [Tue, 10 Jan 2017 17:22:44 +0000 (17:22 +0000)]
drm/i915: Set an error status for a resubmitted request
Let userspace know if its request was resubmitted due to it being
executed at the time of a global reset. In this case, the reset was for
a guilty request on another engine, and this request was an innocent
victim that will be re-executed upon restarting. However, since it was
running at the time of the reset, we can not guarantee that it suffered
no ill-effects from the reset (e.g. some context state may be lost, or
some self-modifying fragment shaders will be restarted from the final
state not their initial state), to let userspace know that it has been
corrupted set a special value on the fence->error, -EAGAIN.
If the request does hang on resubmission, the error will be overwritten
with -EIO.
Chris Wilson [Tue, 10 Jan 2017 17:22:43 +0000 (17:22 +0000)]
drm/i915: Set guilty-flag on fence after detecting a hang
The struct dma_fence carries a status field exposed to userspace by
sync_file. This is inspected after the fence is signaled and can convey
whether or not the request completed successfully, or in our case if we
detected a hang during the request (signaled via -EIO in
SYNC_IOC_FILE_INFO).
Chris Wilson [Tue, 10 Jan 2017 17:22:42 +0000 (17:22 +0000)]
drm/i915: Consolidate reset_request()
Always reset the requests of the guilty context, including the hung
request that we tell the hardware to skip. This should help if the
reprogram fails entirely, but more importantly makes the guilty path
more uniform (and simplifies the subsequent patch to tweak the cancelled
requests).
Tomeu Vizoso [Tue, 10 Jan 2017 13:43:05 +0000 (14:43 +0100)]
drm/i915: Put "cooked" vlank counters in frame CRC lines
Use drm_accurate_vblank_count so we have the full 32 bit to represent
the frame counter and userspace has a simpler way of knowing when the
counter wraps around.
Tomeu Vizoso [Tue, 10 Jan 2017 13:43:04 +0000 (14:43 +0100)]
drm/i915: Use new CRC debugfs API
The core provides now an ABI to userspace for generation of frame CRCs,
so implement the ->set_crc_source() callback and reuse as much code as
possible with the previous ABI implementation.
When handling the pageflip interrupt, we skip 1 or 2 frames depending on
the HW because they contain wrong values. For the legacy ABI for
generating frame CRCs, this was done in userspace but now that we have a
generic ABI it's better if it's not exposed by the kernel.
v2:
- Leave the legacy implementation in place as the ABI implementation
in the core is incompatible with it.
v3:
- Use the "cooked" vblank counter so we have a whole 32 bits.
- Make sure we don't mess with the state of the legacy CRC capture
ABI implementation.
v4:
- Keep use of get_vblank_counter as in the legacy code, will be
changed in a followup commit.
v5:
- Skip first frame or two as it's known that they contain wrong
data.
- A few fixes suggested by Emil Velikov.
v6:
- Rework programming of the HW registers to preserve previous
behavior.
v7:
- Address whitespace issue.
- Added a comment on why in the implementation of the new ABI we
skip the 1st or 2nd frames.
v9:
- Add stub for intel_crtc_set_crc_source.
v12:
- Rebased.
- Remove stub for intel_crtc_set_crc_source and instead set the
callback to NULL (Jani Nikula).
Daniel Vetter [Tue, 10 Jan 2017 16:03:46 +0000 (17:03 +0100)]
Merge remote-tracking branch 'airlied/drm-next' into drm-intel-next-queued
Pull in latest drm-next from Dave Airlie to get at all the drm-misc
goodies, specifically:
- dma_fence error state handling rework (Chris needs that for error
recovery)
- crc support locking changes (Tomeu's i915 crc patches need that).
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Chris Wilson [Tue, 10 Jan 2017 12:10:45 +0000 (12:10 +0000)]
drm/i915: Split out i915_gem_object_set_tiling()
Expose an interface for changing the tiling and stride on an object,
that includes the complexity of checking for conflicting bindings and
fence registers.
Chris Wilson [Tue, 10 Jan 2017 09:56:33 +0000 (09:56 +0000)]
drm/i915: Clip the partial view against the object not vma
The VMA is later clipped against the vm_area_struct before insertion of
the faulting PTE so we are free to create the partial view as we desire.
If we use the object as the extents rather than the area, this partial
can then be used for other areas.
Michel Thierry [Mon, 9 Jan 2017 14:51:35 +0000 (16:51 +0200)]
drm/i915/glk: Convert a few more IS_BROXTON() to IS_GEN9_LP()
Commit cc3f90f0633c ("drm/i915/glk: Reuse broxton code for geminilake")
missed a few of occurences of IS_BROXTON() that should have been
coverted to IS_GEN9_LP().
v2: Cite the right commit. (Ander)
Fixes: cc3f90f0633c ("drm/i915/glk: Reuse broxton code for geminilake") Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: intel-gfx@lists.freedesktop.org Signed-off-by: Michel Thierry <michel.thierry@intel.com> Signed-off-by: Tomasz Lis <tomasz.lis@intel.com> Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Reviewed-by: Tomasz Lis <tomasz.lis@intel.com> (v1) Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1483973495-15138-1-git-send-email-ander.conselvan.de.oliveira@intel.com
Chris Wilson [Mon, 9 Jan 2017 16:16:13 +0000 (16:16 +0000)]
drm/i915: Move ggtt fence/alignment to i915_gem_tiling.c
Rename i915_gem_get_ggtt_size() and i915_gem_get_ggtt_alignment() to
i915_gem_fence_size() and i915_gem_fence_alignment() respectively to
better match usage. Similarly move the pair of functions into
i915_gem_tiling.c next to the fence restrictions.
Chris Wilson [Mon, 9 Jan 2017 16:16:12 +0000 (16:16 +0000)]
drm/i915: Remove the rounding down of the gen4+ fence region
Restricting the fence to the end of the previous tile-row breaks access
to the final portion of the object. On gen2/3 we employed lazy fencing
to pad out the fence with scratch page to provide access to the tail,
and now we also pad out the object on gen4+ we can apply the same fix.
Chris Wilson [Mon, 9 Jan 2017 16:16:11 +0000 (16:16 +0000)]
drm/i915: Store required fence size/alignment for GGTT vma
The fence size/alignment is a combination of the vma size plus object
tiling parameters. Those parameters are rarely changed, making the fence
size/alignemnt roughly constant for the lifetime of the VMA. We can
simplify subsequent calculations by precalculating the size/alignment
required for GGTT vma taking fencing into account (with an update if we
do change the tiling or stride).
Chris Wilson [Mon, 9 Jan 2017 16:16:10 +0000 (16:16 +0000)]
drm/i915: Replace WARNs in fence register writes with extensive asserts
All of these conditions are prechecked by i915_tiling_ok() before we
allow setting the tiling/stride on the object and so we should never
fail asserting those conditions before writing the register.
Chris Wilson [Mon, 9 Jan 2017 16:16:09 +0000 (16:16 +0000)]
drm/i915: Align GGTT sizes to a fence tile row
Ensure the view occupies the full tile row so that reads/writes into the
VMA do not escape (via fenced detiling) into neighbouring objects - we
will pad the object with scratch pages to satisfy the fence. This
applies the lazy-tiling we employed on gen2/3 to gen4+.
Dave Airlie [Mon, 9 Jan 2017 22:06:56 +0000 (08:06 +1000)]
Merge tag 'drm-misc-next-2017-01-09' of git://anongit.freedesktop.org/git/drm-misc into drm-next
Back to regular -misc pulls with reasonable sizes:
- dma_fence error clarification (Chris)
- drm_crtc_from_index helper (Shawn), pile more patches on the m-l to roll
this out to drivers
- mmu-less support for fbdev helpers from Benjamin
- piles of kerneldoc work
- some polish for crc support from Tomeu and Benjamin
- odd misc stuff all over
* tag 'drm-misc-next-2017-01-09' of git://anongit.freedesktop.org/git/drm-misc: (48 commits)
dma-fence: Introduce drm_fence_set_error() helper
dma-fence: Wrap querying the fence->status
dma-fence: Clear fence->status during dma_fence_init()
drm: fix compilations issues introduced by "drm: allow to use mmuless SoC"
drm: Change the return type of the unload hook to void
drm: add more document for drm_crtc_from_index()
drm: remove useless parameters from drm_pick_cmdline_mode function
drm: crc: Call wake_up_interruptible() each time there is a new CRC entry
drm: allow to use mmuless SoC
drm: compile drm_vm.c only when needed
fbmem: add a default get_fb_unmapped_area function
drm: crc: Wait for a frame before returning from open()
drm: Move locking into drm_debugfs_crtc_crc_add
drm/imx: imx-tve: Remove unused variable
Revert "drm: nouveau: fix build when LEDS_CLASS=m"
drm: Add kernel-doc for drm_crtc_commit_get/put
drm/atomic: Fix outdated comment.
drm: reference count event->completion
gpu: drm: mgag200: mgag200_main:- Handle error from pci_iomap
drm: Document deprecated load/unload hook
...
projects / linux-beck.git / log