aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* drm/i915/execlists: Microoptimise execlists_cancel_port_request()Chris Wilson2017-09-25
| | | | | | | | | Just rearrange the code slightly to trim the number of iterations required. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170925124929.16974-1-chris@chris-wilson.co.uk Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
* drm/i915: Don't rmw PIPESTAT enable bitsVille Syrjälä2017-09-25
| | | | | | | | | | | | | | | | | | | | i830 seems to occasionally forget the PIPESTAT enable bits when we read the register. These aren't the only registers on i830 that have problems with RMW, as reading the double buffered plane registers returns the latched value rather than the last written value. So something similar is perhaps going on with PIPESTAT. This corruption results on vblank interrupts occasionally turning off on their own, which leads to vblank timeouts and generally a stuck display subsystem. So let's not RMW the pipestat enable bits, and instead use the cached copy we have around. Cc: Imre Deak <imre.deak@intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170914151731.5034-1-ville.syrjala@linux.intel.com Reviewed-by: Imre Deak <imre.deak@intel.com>
* drm/i915: Speed up DMC firmware loadingDavid Weinehall2017-09-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we're doing: 1. acquire lock 2. write word to hardware 3. release lock 4. repeat from 1 to load the DMC firmware. Due to the cost of acquiring/releasing a lock, and the size of the DMC firmware, this slows down DMC loading a lot. This patch simply acquires the lock, writes the entire firmware, then releases the lock. Testing shows resume speedups in the order of 10ms on platforms with DMC firmware (GEN9+). v2: Per feedback from Chris & Ville there's no need to do the whole forcewake dance, so lose that bit (Chris, Ville) v3: Actually send the new version of the patch... v4: Don't acquire the uncore lock. Disable preempt. (Chris) Signed-off-by: David Weinehall <david.weinehall@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170905131050.11655-1-david.weinehall@linux.intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915/dp: Remove useless debug about TPS3 supportDhinakaran Pandiyan2017-09-25
| | | | | | | | | | We already print training pattern used during link training and also print if the source or sink does not support TPS3 for HBR2 link rates, see intel_dp_training_pattern(). Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918222141.4674-5-dhinakaran.pandiyan@intel.com
* drm/i915/dp: Fix buffer size for sink_irq_esi readPandiyan, Dhinakaran2017-09-25
| | | | | | | | | | The buffer size defined is 16 bytes whereas only 14 bytes are read. Add a macro to avoid this discrepancy. Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: James Ausmus <james.ausmus@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918222141.4674-3-dhinakaran.pandiyan@intel.com
* drm/i915/mst: Print active mst links after updatePandiyan, Dhinakaran2017-09-25
| | | | | | | | | | | | | Both mst_disable_dp and mst_post_disable_dp print number of active links before the variable has been updated. Move the print statement in mst_disable_dp after the decrement so that the printed values indicate the disabing of a mst connector. Also, add some text to clarify what we are printing. Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: James Ausmus <james.ausmus@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918222141.4674-2-dhinakaran.pandiyan@intel.com
* drm/i915/mst: Debug log connector name in destroy_connector()Pandiyan, Dhinakaran2017-09-25
| | | | | | | | | | | Print connector name in destroy_connect() and this doesn't add any extra lines to dmesg. The debug macro has been moved before the unregister call so that we don't lose the connector name and id. Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: James Ausmus <james.ausmus@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918222141.4674-1-dhinakaran.pandiyan@intel.com
* drm/i915/lrc: Skip no-op per-bb buffer on gen9Chris Wilson2017-09-25
| | | | | | | | | | | | | | Since we inherited the context image setup from gen8 which needed a per-bb workaround (for GPGPU), we are submitting an empty per-bb buffer on gen9. Now that we can skip adding the buffer to the context image, remove the dangling per-bb. This slightly improves execution latency, most notably on an idle engine. References: https://bugs.freedesktop.org/show_bug.cgi?id=87725 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170921135444.27330-2-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
* drm/i915/lrc: Only enable per-context and per-bb buffers if setChris Wilson2017-09-25
| | | | | | | | | | | | | The per-context and per-batch workaround buffers are optional, yet we tell the GPU to execute them even if they contain no instructions. Doing so incurs the dispatch latency, which we can avoid if we don't ask the GPU to execute the no-op buffers. Allow ourselves to skip setup of empty buffer, and then to only enable non-empty buffers in the context image. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170921135444.27330-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
* drm/i915: Make execlist port count variableMika Kuoppala2017-09-25
| | | | | | | | | | | | | | As we emulate execlists on top of the GuC workqueue, it is not restricted to just 2 ports and we can increase that number arbitrarily to trade-off queue depth (i.e. scheduling latency) against pipeline bubbles. v2: rebase. better commit msg (Chris) v3: rebase Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-5-mika.kuoppala@intel.com
* drm/i915: Add execlist_port_completeMika Kuoppala2017-09-25
| | | | | | | | | | | | | When first execlist entry is processed, we move the port (contents). Introduce function for this as execlist and guc use this common operation. v2: rebase. s/GEM_DEBUG_BUG/GEM_BUG (Chris) v3: rebase Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-4-mika.kuoppala@intel.com
* drm/i915: Wrap port cancellation into a functionMika Kuoppala2017-09-25
| | | | | | | | | | | | | | | | On reset and wedged path, we want to release the requests that are tied to ports and then mark the ports to be unset. Introduce a function for this. v2: rebase v3: drop local, keep GEM_BUG_ON (Michał, Chris) v4: rebase Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-3-mika.kuoppala@intel.com
* drm/i915: Move execlist initialization into intel_engine_cs.cMika Kuoppala2017-09-25
| | | | | | | | | | | | Move execlist init into a common engine setup. As it is common to both guc and hw execlists. v2: rebase with csb changes v3: rebase Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-2-mika.kuoppala@intel.com
* drm/i915: Make own struct for execlist itemsMika Kuoppala2017-09-25
| | | | | | | | | | | | | | | | | | | | Engine's execlist related items have been increasing to a point where a separate struct is warranted. Carve execlist specific items to a dedicated struct to add clarity. v2: add kerneldoc and fix whitespace (Joonas, Chris) v3: csb_mmio changes, rebase v4: s/\b(el|execlist)\b/execlists/ (Joonas) Suggested-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> (v3) Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v3) Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170922124307.10914-1-mika.kuoppala@intel.com
* drm/i915/bios: ignore HDMI on port AJani Nikula2017-09-25
| | | | | | | | | | | | | | | | The hardware state readout oopses after several warnings when trying to use HDMI on port A, if such a combination is configured in VBT. Filter the combo out already at the VBT parsing phase. v2: also ignore DVI (Ville) Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102889 Cc: stable@vger.kernel.org Cc: Imre Deak <imre.deak@intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Tested-by: Daniel Drake <dan@reactivated.net> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170921141920.18172-1-jani.nikula@intel.com
* drm/i915: Make i915_spin_request() staticChris Wilson2017-09-22
| | | | | | | | | | | | | | No users now outside of i915_wait_request(), so we can make it private to i915_gem_request.c, and assume the caller knows the seqno. In the process, also remove i915_gem_request_started() as that was only ever used by i915_spin_request(). Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michal Winiarski <michal.winiarski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170922120333.25535-1-chris@chris-wilson.co.uk Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
* drm/i915/cnl: Add Gen10 LRC sizeOscar Mateo2017-09-22
| | | | | | | | | | | | | | The total size of the context has decreased with the removal of the URB_ATOMIC section. BSpec indicates 16750 DWORDs (17 pages), plus one page for PPHWSP, and I'm throwing an extra page for precaution. Signed-off-by: Oscar Mateo <oscar.mateo@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Ben Widawsky <ben@bwidawsk.net> Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1506035989-14295-1-git-send-email-oscar.mateo@intel.com
* drm/i915: Rename global i915 to i915_modparamsMichal Wajdeczko2017-09-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Our global struct with params is named exactly the same way as new preferred name for the drm_i915_private function parameter. To avoid such name reuse lets use different name for the global. v5: pure rename v6: fix Credits-to: Coccinelle @@ identifier n; @@ ( - i915.n + i915_modparams.n ) Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Ville Syrjala <ville.syrjala@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170919193846.38060-1-michal.wajdeczko@intel.com
* drm/i915: Confirm request->global_seqno after spin completionChris Wilson2017-09-22
| | | | | | | | | | | | | After we see our target seqno has been completed by the hw, we need to confirm that it still matches the request (as it may have been preempted before the spin completes). If the request no longer matches the target seqno, we need to restart the wait to reacquire that seqno. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Michal Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170921210903.18337-1-chris@chris-wilson.co.uk Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
* drm/i915: Check waiter->seqno carefully in case of preemptionChris Wilson2017-09-22
| | | | | | | | | | | | | | If preemption occurs at precisely the right moment, we may decide that the wait is complete even though the wait's request is no longer executing (having been preempted). We handle this situation by double checking that request following deciding whether the wait is complete. Reported-by: Michał Winiarski <michal.winiarski@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918162734.21294-2-chris@chris-wilson.co.uk Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
* drm/i915: Only wake the waiter from the interrupt if passedChris Wilson2017-09-22
| | | | | | | | | | | | | | As we now check if the seqno is complete in order to signal the fence, we can also decide not to wake up the first_waiter until it is ready (since it is waiting on the same seqno). The only caveat is that if we need the engine->irq_seqno_barrier to enforce some coherency between an interrupt and the seqno read, we have to always wake the waiter in order to perform that heavyweight barrier. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918162734.21294-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
* drm/i915: Document the split in internal and public execbuf flagsChris Wilson2017-09-21
| | | | | | | | | | | | | Since we reuse the same field for the user passing in their control flags, and for the kernel to track a couple of bits of state, document and check that those do not overlap. Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170921110135.15990-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
* drm/i915/kbl: Change a KBL pci id to GT2 from GT1.5Anuj Phogat2017-09-21
| | | | | | | | | | | See Mesa commit 9c588ff Cc: Matt Turner <mattst88@gmail.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170920203126.1323-1-anuj.phogat@gmail.com
* drm/i915: remove redundant variable hw_checkColin Ian King2017-09-20
| | | | | | | | | | | | | | | hw_check is being assigned and updated but is no longer being read, hence it is redundant and can be removed. Detected by clang scan-build: "warning: Value stored to 'hw_check' during its initialization is never read" Fixes: f6d1973db2d2 ("drm/i915: Move modeset state verifier calls") Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/20170914162154.11304-1-colin.king@canonical.com
* drm/i915: Reorganize .disable hooks for pre-DDI DPVille Syrjälä2017-09-20
| | | | | | | | | | | | | | | | | | | | | | | | | Most of our DP encoder hooks are split into per-platform variants. .disable() an exception, and thus it's a bit messy. Let's split it up as well. We'll leave the common parts in a helper called by each platform specific hook. Now each platform has mostly its own hooks. Some hooks are still shared between vlv and chv, and between g4x and ilk. None of the remaining shared hooks have any platform checks in them however so duplicating them doesn't seem particularly useful. There is a subtle change on VLV/CHV where we now disable PSR before audio, whereas before we disabled PSR after audio. That should be totally fine, and PSR is disabled by default anyway. Jani also pointed out to me that PSR + audio doesn't seem like a particularly realistic combination. v2: Drop the PSR HAS_DDI check here (Rodrigo) Pimp up the commit message a bit based on a chat with Jani Cc: Jani Nikula <jani.nikula@intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170920151251.5961-1-ville.syrjala@linux.intel.com
* drm/i915: Drop useless HAS_PSR() checkVille Syrjälä2017-09-20
| | | | | | | | | | | It is safe to call intel_psr_disable() on a platform without PSR. We don't have such a check when calling intel_psr_enable() either. v2: Don't drop the HAS_DDI check quite yet (Rodrigo) Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170920151236.5864-1-ville.syrjala@linux.intel.com
* drm/i915: Stop engines before resetMika Kuoppala2017-09-20
| | | | | | | | | | | | | | | | | | | | | | | On kbl evidence indicates that even if the hardware happily tells us to proceed with reset, it really isn't ready. Resetting a freely running batchbuffer after we have ack for readiness, still can cause a system hang. We also have similar experiences on older gens. So now attempt to stop engines before proceeding for reset, on all gens where we have a gpu reset. This has shown to improve reset reliability and reduce the risk of losing the machine. v2: Add fixme for wa (Joonas) Testcase: igt/prime_busy/hang-* # kbl Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Acked-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170919144128.25506-1-mika.kuoppala@intel.com
* drm/i915: always update ELD connector type after get modesJani Nikula2017-09-20
| | | | | | | | | | | | | | | | | | | | | | | | | drm_edid_to_eld() initializes the connector ELD to zero, overwriting the ELD connector type initialized in intel_audio_codec_enable(). If userspace does getconnector and thus get_modes after modeset, a subsequent audio component i915_audio_component_get_eld() call will receive an ELD without the connector type properly set. It's fine for HDMI, but screws up audio for DP. Always set the ELD connector type at intel_connector_update_modes() based on the connector type. We can drop the connector type update from intel_audio_codec_enable(). Credits to Joseph Nuzman <jnuzman@gmail.com> for figuring this out. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Joseph Nuzman <jnuzman@gmail.com> Reported-by: Joseph Nuzman <jnuzman@gmail.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101583 Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Tested-by: Joseph Nuzman <jnuzman@gmail.com> Cc: stable@vger.kernel.org # v4.10+, maybe earlier Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170919153813.29808-1-jani.nikula@intel.com
* drm/i915/cfl: Remove alpha support protection.Rodrigo Vivi2017-09-19
| | | | | | | | | | | | | | | We now have Coffee Lake on our CI systems. Coffee Lake is at this point in same stage as Kaby Lake. And it seems that we don't have any risk of bad blank screens or anything like that. So let's remove the protection. Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170907230632.25650-1-rodrigo.vivi@intel.com
* drm/i915/cnp: Display Wa #1179: WaHardHangonHotPlugRodrigo Vivi2017-09-19
| | | | | | | | | | | | | | | | | | | | | | | | | "CNL PCH chance of hang when software accesses south display registers after hotplug is enabled. Workaround: Program 0xC2000 bits 11:8 = 0xF before enabling south display hotplug detection." "Workaround only needs to be applied to pre-production steppings used in graphics capable SKUs, but it is easier to apply to everything, and does not hurt." v2: Moving from clock gating to right before enabling SHOTPLUG_CTL as it should be. v3: Align with SOUTH_CHICKEN1 (DK) and consequently use proper spaces on bits definition since other bits around already use new style. And now that checkpatch is not noise anymore I also fixed the reg read mask to avoid going over 80 chars. Suggested-by: Ben Widawsky <ben@bwidawsk.net> Cc: Ben Widawsky <ben@bwidawsk.net> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170919215703.25947-1-rodrigo.vivi@intel.com
* drm/i915: Shrink cnl_ddi_buf_transVille Syrjälä2017-09-19
| | | | | | | | | | All the values we put into the CNL buf_trans tables fit into 8 bits. So switch over to u8 from the u32 we use currently. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918182604.9519-4-ville.syrjala@linux.intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Jani Nikula <jani.nikula@intel.com>
* drm/i915: Shrink bxt_ddi_buf_transVille Syrjälä2017-09-19
| | | | | | | | | | All the values we put into the BXT buf_trans tables fit into 8 bits. So switch over to u8 from the u32 we use currently. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918182604.9519-3-ville.syrjala@linux.intel.com Reviewed-by: Jani Nikula <jani.nikula@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915: Replace some spaces with tabsVille Syrjälä2017-09-19
| | | | | | | | | | Some comments in intel_ddi.c are indented with spaces instead of tabs. Fix that up. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918182604.9519-2-ville.syrjala@linux.intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Jani Nikula <jani.nikula@intel.com>
* drm/i915/fence: Avoid del_timer_sync() from inside a timerChris Wilson2017-09-19
| | | | | | | | | | | | | | | | | | | | | A fence may be signaled from any context, including from inside a timer. One example is timer_i915_sw_fence_wake() which is used to provide a safety-net when waiting on an external fence. If the external fence is not signaled within a timely fashion, we signal our fence on its behalf, and so we then may process subsequent fences in the chain from within that timer context. Given that dma_i915_sw_fence_wake() may be from inside a timer, we cannot then use del_timer_sync() as that requires the timer lock for itself. To circumvent this, while trying to keep the signal propagation as low latency as possible, move the completion into a worker and use a bit of atomic switheroo to serialise the timer-callback and the dma-callback. Testcase: igt/gem_eio/in-flight-external Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170911084135.22903-3-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
* drm/i915/selftests: Only touch archdata.iommu when it existsChris Wilson2017-09-19
| | | | | | | | | | | | archdata.iommu only exists when CONFIG_IOMMU_API is enabled (and only applies to intel-iommu in our case) so conditionally compile it out when it doesn't exist. Fixes: b5891fb520f7 ("drm/i915/selftests: Disable iommu for the mock device") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918164652.14200-1-chris@chris-wilson.co.uk Reviewed-by: Matthew Auld <matthew.auld@intel.com>
* drm/i915/cnp: Don't touch other PCH clock gating bits.Rodrigo Vivi2017-09-18
| | | | | | | | | | | | | | Don't touch other bits. My bad. I haven't seen any case where those other bits appeard to be set before we touch it, but it is safe to avoid touching other bits we weren't told to touch. Fixes: 0a46ddd57c9e ("drm/i915/cnp: Wa 1181: Fix Backlight issue") Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170908234534.17986-1-rodrigo.vivi@intel.com
* drm/i915: Do not enable DRRS when PSR is enabledRadhakrishna Sripada2017-09-18
| | | | | | | | | | | | | | | | | | | | | Some platforms do not support PSR and DRRS simultaneously. Visual artifacts and flickering were reported on BDW HP Spectre x360 Convertible. Deferring to PSR when both PSR and DRRS are supported by the panel. V2: Minor code-style changes suggested by Rodrigo V3: Add a WARN_ON during PSR init suggested by Dhinakaran Correct debug message,title suggested by Jani Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101111 Cc: Nicholas Stommel <nicholas.stommel@gmail.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Clinton Taylor <clinton.a.taylor@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170914181641.24393-1-radhakrishna.sripada@intel.com
* drm/i915/perf: add support for Coffeelake GT2Lionel Landwerlin2017-09-18
| | | | | | | | Add the test configuration & timestamp frequency for Coffeelake GT2. Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918112124.29541-3-lionel.g.landwerlin@intel.com
* drm/i915/perf: disable clk ratio reports on gen9Lionel Landwerlin2017-09-18
| | | | | | | | | We're doing this on all Gen9 based platforms, let's just check the gen rather than listing every single platforms. Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170918112124.29541-2-lionel.g.landwerlin@intel.com
* drm/i915/selftests: Disable iommu for the mock deviceChris Wilson2017-09-18
| | | | | | | | | | | | | | On some machines, the iommu cannot allocate a domain for the mock device causing the dma_map_sg() to fail, and the selftest to fail with -ENOMEM. For the mock selftests, we are using a fake device and do not care about iommu; so convince intel_iommu to treat us as a dummy device with an identity mapping (and no iommu domain). Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101080 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170914162240.18310-1-chris@chris-wilson.co.uk Tested-by: Elizabeth De La Torre Mena <elizabethx.de.la.torre.mena@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com
* drm/i915: Return the correct score in i915_ppat_get()Zhi Wang2017-09-18
| | | | | | | | | | | | | | | | | | | | The cache attribute of the required entry has to be the same with the existing value. After this requirement is met, the futher comparison should be performed. After this fix, the refined test case can pass. v2: - Refine the tittle and comments. (Rodrigo) Fixes: 4395890a4855 ("drm/i915: Introduce private PAT management") Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Ben Widawsky <benjamin.widawsky@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1505741794-10593-1-git-send-email-zhi.a.wang@intel.com Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* uapi/drm/i915: document field usage of drm_i915_perf_oa_configLionel Landwerlin2017-09-18
| | | | | | | | | Document the expected length of buffers config pointers (tuple of u32 values). Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170918114241.30105-1-lionel.g.landwerlin@intel.com
* drm/i915/guc: Cleanup adding GuC work itemsMichał Winiarski2017-09-18
| | | | | | | | | | | | | | | | | | | | | | We can just operate on the wq_tail directly (in the process descriptor). This allows us to remove the duplicated tail from the client. While I'm here let's also remove the constants kept in the client and document our locking requirements. This causes a small change in one of GuC debugfs files. We're no longer reporting constant values (which I don't think is a problem), but we're also no longer reporting the tail (does anyone care?). v2: Update tail after wqi contents. (Chris) v3: Really update tail after wqi contents. Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Oscar Mateo <oscar.mateo@intel.com> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170918092536.12287-1-michal.winiarski@intel.com Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915/guc: Simplify GuC doorbell logicMichał Winiarski2017-09-18
| | | | | | | | | | | | | | | | | | | | | | | | All we're really doing is incrementing a simple counter in a doorbell_info struct. We can do without extra variables and a separate counter kept in guc_client. Since it's gone, we're also removing its debugfs. The only functional change here, is that we're no longer treating 0 as a special value. GuC doesn't seem to care, why should we? v2: Restore desc->tail update. v3: Drop the retry loop, assert that doorbell cookie doesn't change behind our back. v4: WARN rather than BUG, use xchg. (Chris) Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Oscar Mateo <oscar.mateo@intel.com> Suggested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170914105125.3031-1-michal.winiarski@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915/guc: Submit GuC workitems containing coalesced requestsMichał Winiarski2017-09-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | To create an upper bound on number of GuC workitems, we need to change the way that requests are being submitted. Rather than submitting each request as an individual workitem, we can do coalescing in a similar way we're handlig execlist submission ports. We also need to stop pretending that we're doing "lite-restore" in GuC submission (we would create a workitem each time we hit this condition). This allows us to completely remove the reservation, replacing it with a compile time check. v2: Also coalesce when replaying on reset (Daniele) v3: Consistent wq_resv - per-request (Daniele) v4: Squash removing wq_resv v5: Reflect i915_guc_submit argument changes in doc v6: Rebase on top of execlists reset/restart fix (Chris,Michał) References: https://bugs.freedesktop.org/show_bug.cgi?id=101873 Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Jeff McGee <jeff.mcgee@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Oscar Mateo <oscar.mateo@intel.com> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170914083216.10192-2-michal.winiarski@intel.com Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915/guc: Remove obsolete comments and remove unused variableMichał Winiarski2017-09-18
| | | | | | | | | | | | | | | | | | Originally removed in: c1adab970348 ("drm/i915/guc: Remove failed doorbell stat from debugfs") f1448a62a103 ("drm/i915/guc: Remove last submission result from debugfs") Were accidentally restored in: 925344ccc91d ("BackMerge tag 'v4.12-rc5' into drm-next") We can also remove unused variable and replace it with a WARN. Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Signed-off-by: Michał Winiarski <michal.winiarski@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20170914083216.10192-1-michal.winiarski@intel.com Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915/execlists: Unwind incomplete requests on resetsChris Wilson2017-09-18
| | | | | | | | | | | | | | | | | | | | | | | Given the mechanism to unwind and replay requests (designed to support preemption), we have an alternative to the current method of resubmitting the ELSP upon reset. Resubmitting ELSP turns out to be more complicated than expected, due to having to handle lost context-switch interrupts and so guessing what ELSP we need to resubmit later. Instead, by unwinding the requests and clearing the ELSP tracking entirely, we can then just dequeue the first pair of ready requests after resetting, using the normal submission procedure. Currently, the unwound requests have maximum priority and so are guaranteed to be resubmitted upon resume. If we are lucky, we may be able to coalesce a new request on top! Suggested-by: Michał Winiarski <michal.winiarski@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170916204414.32762-4-chris@chris-wilson.co.uk Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
* drm/i915/execlists: Split insert_request()Chris Wilson2017-09-18
| | | | | | | | | | | | In the next patch we will want to reinsert a request not at the end of the priority queue, but at the front. Here we split insert_request() into two, the first function retrieves the priority list (for reuse for unsubmit later) and a wrapper function to insert at the end of that list and to schedule the tasklet if we were first. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170916204414.32762-3-chris@chris-wilson.co.uk
* drm/i915/execlists: Move insert_request()Chris Wilson2017-09-18
| | | | | | | | | Move insert_request() earlier to avoid a forward declaration in a later patch. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170916204414.32762-2-chris@chris-wilson.co.uk
* drm/i915/execlists: Kick start request processing after a resetChris Wilson2017-09-18
| | | | | | | | | | | | | | | | | | | | | | | During a reset, we may skip over completed requests and lost context-switch interrupts. Following the reset, we may then may end up with no active requests in the ELSP (and so do not resubmit to restart the engine), but have a queue of requests ready for execution. This is unlikely, it requires the last request to complete after the hang is detected, but not impossible. The outcome of this is that the engine stalls, possibly leading to full ring and indefinite wait under struct_mutex, eventually leading to a full driver hang. Alternatively, we can solve this by unsubmitting the incomplete requests and just kickstarting the tasklet. Michał has patches for that, which I initially disliked due to the extra complexity, but the complexity of this "simple" restart is growing... Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170916204414.32762-1-chris@chris-wilson.co.uk Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
generated by cgit v1.2.2 (git 2.25.0) at 2025-08-19 08:51:28 -0400