- 30 Jan, 2017 6 commits
-
-
Mika Kuoppala authored
Move the invariant parts of context desc setup from execlist init to context creation. This is advantageous when we need to create different templates based on the context parametrization, ie. for svm capable contexts. v2: s/create/default, remove engine->ctx_desc_template Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by:
Mika Kuoppala <mika.kuoppala@intel.com> Reviewed-by:
Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by:
-
Ander Conselvan de Oliveira authored
The gamma tables in Geminilake were changed. There is no split-gamma mode. Instead, there is a dedicated degamma table that is enabled whenever pipe CSC is enabled. The dedicated gamma table has 16 bit precision but doesn't support separate channels. Since that doesn't match the per-channel format of the degamma LUT property, for now only a linear table is loaded and the property ignored. v2: Remove empty line. (Ville) Reuse broadwell code. (Ville) v3: Don't write PIPE_CSC_MODE. (Ville) Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Reviewed-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170127090230.20302-1-ander.conselvan.de.oliveira@intel.com
-
Ander Conselvan de Oliveira authored
Split the logic for progamming each LUT out of broadwell_load_luts(), so we can reuse part of it for geminilake. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
-
Ander Conselvan de Oliveira authored
In Geminilake, the bits for enabling pipe csc, pipe gamma and plane gamma moved to a new register. So update the plane update functions to set the right bits. Pipe CSC is kept disabled though, since enabling that also enables the dedicated degamma table, and that is not properly programmed yet, leading to a black screen. v2: Use plane_id. (Ville) Remove unnecessary variable. (Ville) Keep registers in offset order. (Ville) Don't set plane gamma disable twice. (Ander) Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: -
Ander Conselvan de Oliveira authored
The plane gamma tables are never programmed, so just disable it, like it is done for the primary plane. Signed-off-by:
-
Ander Conselvan de Oliveira authored
Apply workarounds to Geminilake, and annotate those that are applied unconditionally when they apply to GLK based on the workaround database. v2: Fix commit message typos. (David) v3: Rebase. Cc: David Weinehall <david.weinehall@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@gmail.com> Signed-off-by:
David Weinehall <david.weinehall@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1485422218-9102-1-git-send-email-ander.conselvan.de.oliveira@intel.com
-
- 27 Jan, 2017 7 commits
-
-
Chris Wilson authored
Now that the user can opt-out of implicit fencing, we need to give them back control over the fencing. We employ sync_file to wrap our drm_i915_gem_request and provide an fd that userspace can merge with other sync_file fds and pass back to the kernel to wait upon before future execution. Testcase: igt/gem_exec_fence Signed-off-by:
Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by:
Chad Versace <chadversary@chromium.org> Link: http://patchwork.freedesktop.org/patch/msgid/20170127094008.27489-2-chris@chris-wilson.co.uk
-
Chris Wilson authored
Userspace is faced with a dilemma. The kernel requires implicit fencing to manage resource usage (we always must wait for the GPU to finish before releasing its PTE) and for third parties. However, userspace may wish to avoid this serialisation if it is either using explicit fencing between parties and wants more fine-grained access to buffers (e.g. it may partition the buffer between uses and track fences on ranges rather than the implicit fences tracking the whole object). It follows that userspace needs a mechanism to avoid the kernel's serialisation on its implicit fences before execbuf execution. The next question is whether this is an object, execbuf or context flag. Hybrid users (such as using explicit EGL_ANDROID_native_sync fencing on shared winsys buffers, but implicit fencing on internal surfaces) require a per-object level flag. Given that this flag need to be only set once for the lifetime of the object, this reduces the convenience of having an execbuf or context level flag (and avoids having multiple pieces of uABI controlling the same feature). Incorrect use of this flag will result in rendering corruption and GPU hangs - but will not result in use-after-free or similar resource tracking issues. Serious caveat: write ordering is not strictly correct after setting this flag on a render target on multiple engines. This affects all subsequent GEM operations (execbuf, set-domain, pread) and shared dma-buf operations. A fix is possible - but costly (both in terms of further ABI changes and runtime overhead). Testcase: igt/gem_exec_async Signed-off-by:
-
Jani Nikula authored
This reverts commit 2eebe4f2, because I screwed up and applied it to the wrong branch. Signed-off-by:
Jani Nikula <jani.nikula@intel.com>
-
Andrey Ryabinin authored
page_flip_completed() dereferences 'work' variable after executing queue_work(). This is not safe as the 'work' item might be already freed by queued work: BUG: KASAN: use-after-free in page_flip_completed+0x3ff/0x490 at addr ffff8803dc010f90 Call Trace: __asan_report_load8_noabort+0x59/0x80 page_flip_completed+0x3ff/0x490 intel_finish_page_flip_mmio+0xe3/0x130 intel_pipe_handle_vblank+0x2d/0x40 gen8_irq_handler+0x4a7/0xed0 __handle_irq_event_percpu+0xf6/0x860 handle_irq_event_percpu+0x6b/0x160 handle_irq_event+0xc7/0x1b0 handle_edge_irq+0x1f4/0xa50 handle_irq+0x41/0x70 do_IRQ+0x9a/0x200 common_interrupt+0x89/0x89 Freed: kfree+0x113/0x4d0 intel_unpin_work_fn+0x29a/0x3b0 process_one_work+0x79e/0x1b70 worker_thread+0x611/0x1460 kthread+0x241/0x3a0 ret_from_fork+0x27/0x40 Move queue_work() after trace_i915_flip_complete() to fix this. Fixes: e5510fac ("drm/i915: add tracepoints for flip requests & completions") Signed-off-by:Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: <stable@vger.kernel.org> # v2.6.36+ Reviewed-by:
Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20170126143211.24013-1-aryabinin@virtuozzo.com
-
Paulo Zanoni authored
The i915_stolen_to_physical() function has 'unsigned long' as its return type but it returns the 'base' variable, which is of type 'u32'. The only place where this function is called assigns the returned value to dev_priv->mm.stolen_base, which is of type 'phys_addr_t'. The return value is actually a physical address and everything else in the stolen memory code seems to be using phys_addr_t, so fix i915_stolen_to_physical() to use phys_addr_t. v2: Add missing blank lines after declarations (Chris, checkpatch.pl). Reviewed-by:
Paulo Zanoni <paulo.r.zanoni@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1485461947-16030-1-git-send-email-paulo.r.zanoni@intel.com
-
Jani Nikula authored
The function is not that big, but it's also not used for anything performance critical. Make it a normal function. As a side effect, this apparently makes sparse smarter about what it's doing, and gets rid of the warning: ./include/drm/drm_color_mgmt.h:53:28: warning: shift too big (4294967295) for type unsigned long ./include/drm/drm_color_mgmt.h:53:28: warning: cast truncates bits from constant value (8000000000000000 becomes 0) Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by:
Lionel Landwerlin <lionel.g.landwerlin@intel.com> Signed-off-by:
-
Paulo Zanoni authored
So don't forget to reserve its stolen memory bits. Cc: Ingo Molnar <mingo@kernel.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Cc: x86@kernel.org Reviewed-by:
Ingo Molnar <mingo@kernel.org> Signed-off-by:
-
- 26 Jan, 2017 7 commits
-
-
Chris Wilson authored
As the previous punit i/o may have failed, the contents of the PDATA are undefined. Always clear it to 0 prior to sending the command. Signed-off-by:
-
Chris Wilson authored
The write to the punit may fail, so propagate the error code back to its callers. Of particular interest are the RPS writes, so add appropriate user error codes and logging. v2: Add DEBUG for failed frequency changes during RPS. Signed-off-by:
-
Manasi Navare authored
The intel_dp_autotest_video_pattern() function gets invoked through the compliance test handler on a HPD short pulse if the test type is set to DP_TEST_VIDEO_PATTERN. This performs the DPCD registers reads to read the requested test pattern, video pattern resolution, frame rate and bits per color value. The results of this analysis are handed off to userspace so that the userspace app can set the video pattern mode appropriately for the test result/response. When the test is requested with specific BPC value, we read the BPC value from the DPCD register. If this BPC value in intel_dp structure has a non-zero value and we're on a display port connector, then we use the value to calculate the bpp for the pipe. Also in this case if its a 18bpp video pattern request, then we force the dithering on pipe to be disabled since it causes CRC mismatches. The compliance_test_active flag is set at the end of the individual test handling functions. This is so that the kernel-side operations can be completed without the risk of interruption from the userspace app that is polling on that flag. v5: * Remove test_result variable * Populate the compliance test data at the end of the function (Jani Nikula) v4: *Return TEST_NAK on read failures and invalid values (Jani Nikula) * Address CRC mismatch errors v3: * Use the updated properly shifted bit definitions (Jani Nikula) * Force dithering to be disabled on 18bpp compliance test request (Manasi Navare) v2: * Updated the DPCD Register reads based on proper defines in header (Jani Nikula) * Squahsed the patch that forced the pipe bpp to compliance test bpp (Jani Nikula) Signed-off-by:
Manasi Navare <manasi.d.navare@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Ville Syrjala <ville.syrjala@linux.intel.com> Signed-off-by:
-
Manasi Navare authored
v4: * Remove redundant single bit defs (Jani Nikula) v3: * Fix the conventions in bit definitions (Jani Nikula) v2: * Add all the other DP Complianec TEST register defs (Jani Nikula) Cc: dri-devel@lists.freedesktop.org Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Ville Syrjala <ville.syrjala@linux.intel.com> Signed-off-by:
-
Manasi Navare authored
This patch addresses a few issues from the original patch for DP Compliance EDID test support submitted by Todd Previte<todd.previte@gmail.com> Video Mode requested in the EDID test handler for the EDID Read test (CTS 4.2.2.3) should be set to PREFERRED as per the CTS spec. v2: * Added read debugfs data from test_data.edid if its EDID test (Jani NIkula) Signed-off-by:
-
Manasi Navare authored
This patch adds support to handle automated DP compliance link training test requests. This patch has been tested with Unigraf DPR-120 DP Compliance device for testing Link Training Compliance. After we get a short pulse Compliance test request, test request values are read and hotplug uevent is sent in order to trigger another modeset during which the pipe is configured and link is retrained and enabled for link parameters requested by the test. v5: * Only modify the compliance structure after all validation is done (Jani Nikula) * Remove the variable test_result (Jani Nikula) v4: * Return TEST_NAK for read failures and invalid values (Jani Nikula) * Conver the test link BW to link rate before storing (Jani Nikula) v3: * Validate the test link rate and lane count as soon as the request comes (Jani Nikula) v2: * Validate the test lane count before using it in intel_dp_compute_config (Jani Nikula) Signed-off-by:
-
Chris Wilson authored
Remove WaGsvDisableTurbo and WaRsUseTimeoutMode as these were only for pre-production Broxton devices, and this code is now defunct. Signed-off-by:
-
- 24 Jan, 2017 14 commits
-
-
Rodrigo Vivi authored
Along with GLK it was introduced the .is_lp and IS_GEN9_LP. So, following the same simplification standard we can put Skylake and Kabylake under the same bucket for most of the things. So let's add the IS_GEN9_BC for "Big Core" (non Atom based platforms). The i915_drv.c was let out of this patch on purpose because that is really a decision per platform, just like other cases where IS_KABYLAKE is different from IS_SKYLAKE. v2: fix conflict with IS_LP and 3 new cases for this big core bucket: - intel_ddi.c: intel_ddi_get_link_dpll - intel_fbc.c: find_compression_threshold - i915_gem_gtt.c: gtt_write_workarounds Cc: Anusha Srivatsa <anusha.srivatsa@intel.com> Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Signed-off-by:Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by:
Ander Conselvan de Oliveira <conselvan2@gmail.com> Acked-by:
-
Rodrigo Vivi authored
According to wa_database this Wa persist on KBL as it was on SKL. Cc: Tim Gore <tim.gore@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Signed-off-by:
-
Chris Wilson authored
Fences are required to support being released from under an atomic context. The drm_atomic_state struct may take a mutex when being released and so we cannot drop a reference to the drm_atomic_state from the fence release path directly, and so we need to defer that unreference to a worker. [ 326.576697] WARNING: CPU: 2 PID: 366 at kernel/sched/core.c:7737 __might_sleep+0x5d/0x80 [ 326.576816] do not call blocking ops when !TASK_RUNNING; state=1 set at [<ffffffffc0359549>] intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.576818] Modules linked in: rfcomm fuse snd_hda_codec_hdmi bnep snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer input_leds led_class snd punit_atom_debug btusb btrtl btbcm btintel intel_rapl bluetooth i915 drm_kms_helper syscopyarea sysfillrect iwlwifi sysimgblt soundcore fb_sys_fops mei_txe cfg80211 drm pwm_lpss_platform pwm_lpss pinctrl_cherryview fjes acpi_pad parport_pc ppdev parport autofs4 [ 326.576899] CPU: 2 PID: 366 Comm: i915/signal:0 Tainted: G U 4.10.0-rc3-patser+ #5030 [ 326.576902] Hardware name: /NUC5PPYB, BIOS PYBSWCEL.86A.0031.2015.0601.1712 06/01/2015 [ 326.576905] Call Trace: [ 326.576920] dump_stack+0x4d/0x6d [ 326.576926] __warn+0xc0/0xe0 [ 326.576931] warn_slowpath_fmt+0x5a/0x80 [ 326.577004] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577075] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577079] __might_sleep+0x5d/0x80 [ 326.577087] mutex_lock+0x1b/0x40 [ 326.577133] drm_property_free_blob+0x1e/0x80 [drm] [ 326.577167] ? drm_property_destroy+0xe0/0xe0 [drm] [ 326.577200] drm_mode_object_unreference+0x5c/0x70 [drm] [ 326.577233] drm_property_unreference_blob+0xe/0x10 [drm] [ 326.577260] __drm_atomic_helper_crtc_destroy_state+0x14/0x40 [drm_kms_helper] [ 326.577278] drm_atomic_helper_crtc_destroy_state+0x10/0x20 [drm_kms_helper] [ 326.577352] intel_crtc_destroy_state+0x9/0x10 [i915] [ 326.577388] drm_atomic_state_default_clear+0xea/0x1d0 [drm] [ 326.577462] intel_atomic_state_clear+0xd/0x20 [i915] [ 326.577497] drm_atomic_state_clear+0x1a/0x30 [drm] [ 326.577532] __drm_atomic_state_free+0x13/0x60 [drm] [ 326.577607] intel_atomic_commit_ready+0x6f/0x78 [i915] [ 326.577670] i915_sw_fence_release+0x3a/0x50 [i915] [ 326.577733] dma_i915_sw_fence_wake+0x39/0x80 [i915] [ 326.577741] dma_fence_signal+0xda/0x120 [ 326.577812] ? intel_breadcrumbs_signaler+0x59/0x270 [i915] [ 326.577884] intel_breadcrumbs_signaler+0xb1/0x270 [i915] [ 326.577889] kthread+0x127/0x130 [ 326.577961] ? intel_engine_remove_wait+0x1a0/0x1a0 [i915] [ 326.577964] ? kthread_stop+0x120/0x120 [ 326.577970] ret_from_fork+0x22/0x30 Fixes: c004a90b ("drm/i915: Restore nonblocking awaits for modesetting") Reported-by:
Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Signed-off-by:
-
Chris Wilson authored
When introduced, I thought that reducing client latency from the signaler was the priority. Since its inception the signaler has become responsible for keeping the execlists full, via the dma-fence. As this is very important to minimise overall execution time, signal the dma-fence first and then signal any waiting clients. Signed-off-by:
Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170124110009.28947-8-chris@chris-wilson.co.uk
-
Chris Wilson authored
If the second ELSP port is available, schedule the execlists tasklet to see if the new request can occupy it. Signed-off-by:
-
Chris Wilson authored
Only the first request added to the execlist queue can be submitted. If this request is not the first request on the queue, it means that there are already higher priority requests waiting upon the tasklet and kicking it will make no difference. This is more relevant for a later patch, where we more eagerly try and kick the tasklet to handle the submission of new requests. Signed-off-by:
-
Chris Wilson authored
If the CSB head/tail pointers are unchanged, we can skip the update of the CSB register afterwards. Signed-off-by:
-
Chris Wilson authored
Mark when we run the execlist tasklet following the interrupt, so we don't probe a potentially uninitialised register when submitting the contexts multiple times before the hardware responds. v2: Use a shared engine->irq_posted v3: Always use locked bitops to be sure of atomicity wrt to other bits in the mask. Signed-off-by:
-
Chris Wilson authored
In the next patch, we will use the irq_posted technique for another engine interrupt, rather than use two members for the atomic updates, we can use two bits of one instead. First, we need to update the breadcrumbs to use the new common engine->irq_posted. v2: Use set_bit() rather than __set_bit() to ensure atomicity with respect to other bits in the mask Signed-off-by:
-
Chris Wilson authored
We need to prevent resubmission of the context immediately following an initial resubmit (which does a lite-restore preemption). Currently we do this by disabling all submission whilst the context is still active, but we can improve this by limiting the restriction to only until we receive notification from the context-switch interrupt that the lite-restore preemption is complete. Signed-off-by:
-
Chris Wilson authored
To complement the check in execlists_elsp_ready(), also assert that we don't submit the same context while it has a lite restore still pending. Signed-off-by:
-
Jani Nikula authored
There is no point in setting intel_dp->compliance.test_type, and proceeding with the autotests, if we're about to NAK the request. Some drive-by cleanups while at it. v2: look at the ACK bit, as the result may also contain TEST_EDID_CHECKSUM_WRITE Cc: Manasi Navare <manasi.d.navare@intel.com> Reviewed-by: -
Chris Wilson authored
The GPU may be in an unknown state following resume and module load. The previous occupant may have left contexts loaded, or other dangerous state, which can cause an immediate GPU hang for us. The only save course of action is to reset the GPU prior to using it - similarly to how we reset the GPU prior to unload (before a second user may be affected by our leftover state). We need to reset the GPU very early in our load/resume sequence so that any stale HW pointers are revoked prior to any resource allocations we make (that may conflict). A reset should only be a couple of milliseconds on a slow device, a cost we should easily be able to absorb into our initialisation times. Signed-off-by:
-
Chris Wilson authored
In order to reset the GPU early on in the module load sequence, we need to allocate the basic engine structs (to populate the mmio offsets etc). Currently, the engine initialisation allocates both the base struct and also allocate auxiliary objects, which depend upon state setup quite late in the load sequence. We split off the allocation callback for later and allow ourselves to allocate the engine structs themselves early. v2: Different paint for the unwind following error. Signed-off-by:
-
- 23 Jan, 2017 6 commits
-
-
Chris Wilson authored
This w/a (WaClearTdlStateAckDirtyBits) was only used for preproduction hw, which is no longer in use. Remove the workaround to simplify the code. Signed-off-by:
-
Chris Wilson authored
This w/a (WaSetDisablePixMaskCammingAndRhwoInCommonSliceChicken) was only used for preproduction hw, which is no longer in use. Remove the workaround to simplify the code. Signed-off-by:
-
Chris Wilson authored
This w/a (WaDisableCtxRestoreArbitration) was only used for preproduction hw, which is no longer in use. Remove the workaround to simplify the code. Signed-off-by:
-
Chris Wilson authored
This w/a was only used for preproduction hw, which is no longer in use. Remove the workaround to simplify the code. Signed-off-by:
-
Chris Wilson authored
This w/a (WaEnableForceRestoreInCtxtDescForVCS) was only used for preproduction hw, which is no longer in use. Remove the workaround to simplify the code. Signed-off-by:
-
Chris Wilson authored
Since tweaking i915_vma_compare() we allowed constructors to skip clearing the ggtt_view believing that we didn't access the unused members. That, as it turns out, was not entirely true. In particular, i915_gem_fault() uses ret = remap_io_mapping(area, area->vm_start + (vma->ggtt_view.partial.offset << PAGE_SHIFT), (ggtt->mappable_base + vma->node.start) >> PAGE_SHIFT, min_t(u64, vma->size, area->vm_end - area->vm_start), &ggtt->mappable); i.e. the ggtt_view.partial for both normal and partial views. If we allowed garbage into the normal vma->ggtt_view and then try userspace tried to mmap it, we could explode in an unobvious fashion. Fixes: 7b92c047 ("drm/i915: Eliminate superfluous i915_ggtt_view_rotated") Fixes: 3bf4d575 ("drm/i915: Stop clearing i915_ggtt_view") Reported-by:Matthew Auld <matthew.william.auld@gmail.com> Signed-off-by:
Matthew Auld <matthew.auld@intel.com> Reviewed-by:
-
