May 2018 Archives by date

• Messages sorted by: [ thread ] [ subject ] [ author ]
• More info on this list...

Starting: Tue May 1 08:33:52 UTC 2018
Ending: Thu May 31 17:29:27 UTC 2018
Messages: 2459

• [PATCH 1/2] drm/i915/guc: Assert we have the doorbell before setting it up   Chris Wilson
• [PATCH 2/2] drm/i915: Move timeline from GTT to ring   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• [PATCH 1/3] drm/i915/guc: Assert we have the doorbell before setting it up   Chris Wilson
• [PATCH 2/3] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 3/3] drm/i915: Move timeline from GTT to ring   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• [PATCH 1/5] drm/i915/guc: Assert we have the doorbell before setting it up   Chris Wilson
• [PATCH 2/5] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 3/5] drm/i915/execlists: Emit i915_trace_request_out for preemption   Chris Wilson
• [PATCH 4/5] drm/i915: Include completed status in HW request tracepoints   Chris Wilson
• [PATCH 5/5] drm/i915: Move timeline from GTT to ring   Chris Wilson
• [PATCH 1/6] drm/i915/guc: Assert we have the doorbell before setting it up   Chris Wilson
• [PATCH 2/6] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 3/6] drm/i915/execlists: Emit i915_trace_request_out for preemption   Chris Wilson
• [PATCH 4/6] drm/i915: Include completed status in HW request tracepoints   Chris Wilson
• [PATCH 5/6] drm/i915: Move timeline from GTT to ring   Chris Wilson
• [PATCH 6/6] drm/i915: Split i915_gem_timeline into individual timelines   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/6] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/5] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/6] drm/i915/guc: Assert we have the doorbell before setting it up   Patchwork
• [PATCH 01/68] drm/i915/gtt: Tidy up duplicate branches in gen8_gmch_probe()   Chris Wilson
• [PATCH 02/68] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 03/68] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 04/68] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 05/68] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 06/68] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 07/68] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 08/68] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 09/68] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 10/68] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 11/68] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 12/68] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 13/68] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/68] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 15/68] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 16/68] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 17/68] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 18/68] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 19/68] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 20/68] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 21/68] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 22/68] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 23/68] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 24/68] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 25/68] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 26/68] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 27/68] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 28/68] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 29/68] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 30/68] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 31/68] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 32/68] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 33/68] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 34/68] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 35/68] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 36/68] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 37/68] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 38/68] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 39/68] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 40/68] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 41/68] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 42/68] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 43/68] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 44/68] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 45/68] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 46/68] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 47/68] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 48/68] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 49/68] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 50/68] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 51/68] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 52/68] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 53/68] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 54/68] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 55/68] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 56/68] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 57/68] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 58/68] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 59/68] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 60/68] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 61/68] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 62/68] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 63/68] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 64/68] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 65/68] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 66/68] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 67/68] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 68/68] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/68] drm/i915/gtt: Tidy up duplicate branches in gen8_gmch_probe()   Patchwork
• ✓ Fi.CI.BAT: success for drm/i915: Discard two CRC's for GEN8+ platforms (rev2)   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/68] drm/i915/gtt: Tidy up duplicate branches in gen8_gmch_probe()   Patchwork
• [PATCH 01/69] drm/i915: Remove assertion of active_rings must be non-empty if active_requests   Chris Wilson
• [PATCH 02/69] drm/i915/gtt: Tidy up duplicate branches in gen8_gmch_probe()   Chris Wilson
• [PATCH 03/69] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 04/69] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 05/69] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 06/69] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 07/69] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 08/69] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 09/69] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 10/69] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 11/69] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 12/69] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 13/69] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 14/69] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 15/69] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 16/69] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 17/69] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 18/69] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 19/69] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 20/69] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 21/69] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 22/69] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 23/69] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 24/69] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 25/69] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 26/69] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 27/69] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 28/69] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 29/69] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 30/69] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 31/69] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 32/69] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 33/69] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 34/69] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 35/69] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 36/69] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 37/69] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 38/69] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 39/69] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 40/69] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 41/69] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 42/69] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 43/69] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 44/69] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 45/69] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 46/69] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 47/69] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 48/69] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 49/69] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 50/69] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 51/69] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 52/69] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 53/69] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 54/69] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 55/69] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 56/69] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 57/69] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 58/69] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 59/69] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 60/69] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 61/69] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 62/69] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 63/69] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 64/69] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 65/69] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 66/69] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 67/69] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 68/69] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 69/69] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.IGT: success for drm/i915: Discard two CRC's for GEN8+ platforms (rev2)   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/69] drm/i915: Remove assertion of active_rings must be non-empty if active_requests   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/69] drm/i915: Remove assertion of active_rings must be non-empty if active_requests   Patchwork
• [PATCH 01/70] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 02/70] drm/i915/selftests: Skip the execlists tests on !execlists machines   Chris Wilson
• [PATCH 03/70] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 04/70] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 05/70] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 06/70] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 07/70] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 08/70] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 09/70] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 10/70] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 11/70] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 12/70] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 13/70] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 14/70] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 15/70] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 16/70] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 17/70] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 18/70] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 19/70] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 20/70] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 21/70] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 22/70] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 23/70] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 24/70] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 25/70] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 26/70] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 27/70] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 28/70] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 29/70] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 30/70] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 31/70] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 32/70] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 33/70] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 34/70] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 35/70] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 36/70] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 37/70] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 38/70] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 39/70] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 40/70] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 41/70] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 42/70] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 43/70] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 44/70] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 45/70] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 46/70] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 47/70] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 48/70] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 49/70] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 50/70] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 51/70] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 52/70] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 53/70] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 54/70] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 55/70] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 56/70] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 57/70] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 58/70] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 59/70] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 60/70] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 61/70] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 62/70] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 63/70] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 64/70] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 65/70] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 66/70] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 67/70] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 68/70] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 69/70] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 70/70] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/70] drm/i915/execlists: Disable submission tasklets when rescheduling   Patchwork
• [PATCH 01/16] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 02/16] drm/i915/selftests: Skip the execlists tests on !execlists machines   Chris Wilson
• [PATCH 03/16] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 04/16] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 05/16] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 06/16] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 07/16] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 08/16] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 09/16] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 10/16] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 11/16] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 12/16] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 13/16] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 14/16] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 15/16] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 16/16] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/16] drm/i915/execlists: Disable submission tasklets when rescheduling   Patchwork
• [PATCH 01/71] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 02/71] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 03/71] drm/i915/selftests: Skip the execlists tests on !execlists machines   Chris Wilson
• [PATCH 04/71] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 05/71] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 06/71] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 07/71] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 08/71] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 09/71] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 10/71] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 11/71] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 12/71] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 13/71] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 14/71] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 15/71] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 16/71] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 17/71] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 18/71] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 19/71] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 20/71] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 21/71] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 22/71] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 23/71] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 24/71] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 25/71] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 26/71] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 27/71] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 28/71] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 29/71] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 30/71] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 31/71] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 32/71] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 33/71] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 34/71] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 35/71] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 36/71] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 37/71] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 38/71] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 39/71] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 40/71] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 41/71] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 42/71] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 43/71] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 44/71] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 45/71] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 46/71] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 47/71] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 48/71] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 49/71] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 50/71] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 51/71] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 52/71] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 53/71] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 54/71] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 55/71] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 56/71] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 57/71] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 58/71] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 59/71] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 60/71] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 61/71] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 62/71] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 63/71] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 64/71] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 65/71] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 66/71] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 67/71] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 68/71] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 69/71] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 70/71] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 71/71] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/71] drm/i915/dp: Silence static checkers for n_entries   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/16] drm/i915/execlists: Disable submission tasklets when rescheduling   Patchwork
• [PATCH 1/8] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 2/8] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 3/8] drm/i915/selftests: Skip the execlists tests on !execlists machines   Chris Wilson
• [PATCH 4/8] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 5/8] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 6/8] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 7/8] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 8/8] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/8] drm/i915/dp: Silence static checkers for n_entries   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/71] drm/i915/dp: Silence static checkers for n_entries   Patchwork
• [PATCH 01/71] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 02/71] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 03/71] drm/i915/selftests: Skip the execlists tests on !execlists machines   Chris Wilson
• [PATCH 04/71] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 05/71] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 06/71] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 07/71] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 08/71] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 09/71] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 10/71] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 11/71] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 12/71] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 13/71] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 14/71] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 15/71] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 16/71] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 17/71] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 18/71] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 19/71] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 20/71] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 21/71] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 22/71] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 23/71] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 24/71] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 25/71] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 26/71] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 27/71] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 28/71] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 29/71] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 30/71] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 31/71] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 32/71] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 33/71] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 34/71] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 35/71] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 36/71] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 37/71] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 38/71] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 39/71] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 40/71] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 41/71] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 42/71] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 43/71] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 44/71] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 45/71] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 46/71] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 47/71] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 48/71] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 49/71] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 50/71] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 51/71] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 52/71] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 53/71] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 54/71] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 55/71] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 56/71] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 57/71] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 58/71] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 59/71] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 60/71] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 61/71] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 62/71] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 63/71] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 64/71] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 65/71] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 66/71] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 67/71] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 68/71] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 69/71] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 70/71] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 71/71] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/71] drm/i915/dp: Silence static checkers for n_entries   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/71] drm/i915/dp: Silence static checkers for n_entries   Patchwork
• [PATCH 01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 02/74] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 03/74] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 04/74] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 05/74] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 06/74] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 07/74] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 08/74] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 09/74] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 10/74] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 11/74] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 12/74] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 13/74] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 14/74] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 15/74] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 16/74] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 17/74] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 18/74] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 19/74] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 20/74] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 21/74] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 22/74] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 23/74] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 24/74] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 25/74] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 26/74] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 27/74] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 28/74] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 29/74] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 30/74] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 31/74] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 32/74] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 33/74] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 34/74] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 35/74] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 36/74] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 37/74] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 38/74] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 39/74] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 40/74] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 41/74] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 42/74] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 43/74] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 44/74] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 45/74] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 46/74] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 47/74] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 48/74] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 49/74] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 50/74] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 51/74] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 52/74] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 53/74] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 54/74] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 55/74] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 56/74] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 57/74] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 58/74] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 59/74] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 60/74] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 61/74] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 62/74] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 63/74] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 64/74] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 65/74] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 66/74] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 67/74] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 68/74] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 69/74] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 70/74] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 71/74] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 72/74] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 73/74] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 74/74] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Patchwork
• [PATCH 01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 02/74] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 03/74] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 04/74] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 05/74] drm/i915/execlists: Disable submission tasklets when rescheduling   Chris Wilson
• [PATCH 06/74] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 07/74] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 08/74] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 09/74] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 10/74] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 11/74] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 12/74] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 13/74] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 14/74] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 15/74] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 16/74] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 17/74] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 18/74] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 19/74] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 20/74] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 21/74] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 22/74] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 23/74] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 24/74] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 25/74] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 26/74] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 27/74] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 28/74] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 29/74] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 30/74] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 31/74] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 32/74] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 33/74] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 34/74] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 35/74] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 36/74] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 37/74] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 38/74] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 39/74] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 40/74] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 41/74] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 42/74] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 43/74] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 44/74] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 45/74] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 46/74] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 47/74] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 48/74] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 49/74] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 50/74] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 51/74] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 52/74] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 53/74] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 54/74] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 55/74] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 56/74] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 57/74] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 58/74] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 59/74] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 60/74] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 61/74] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 62/74] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 63/74] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 64/74] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 65/74] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 66/74] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 67/74] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 68/74] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 69/74] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 70/74] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 71/74] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 72/74] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 73/74] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 74/74] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/74] drm/syncobj: Use dma_fence_wait for the simple wait case   Patchwork
• [PATCH 1/9] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 2/9] drm/i915: Don't request a bug report for unsafe module parameters   Chris Wilson
• [PATCH 3/9] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 4/9] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 5/9] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 6/9] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 7/9] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 8/9] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 9/9] drm/i915/execlists: Direct submit onto idle engine   Chris Wilson
• [PATCH 1/9] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 2/9] drm/i915: Don't request a bug report for unsafe module parameters   Chris Wilson
• [PATCH 3/9] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 4/9] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 5/9] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 6/9] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 7/9] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 8/9] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 9/9] drm/i915/execlists: Direct submit onto idle engine   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• [PATCH 1/9] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 2/9] drm/i915: Don't request a bug report for unsafe module parameters   Chris Wilson
• [PATCH 3/9] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 4/9] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 5/9] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 6/9] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 7/9] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 8/9] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 9/9] drm/i915/execlists: Direct submit onto idle engine   Chris Wilson
• [PATCH 1/9] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 2/9] drm/i915: Don't request a bug report for unsafe module parameters   Chris Wilson
• [PATCH 3/9] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 4/9] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 5/9] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 6/9] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 7/9] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 8/9] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 9/9] drm/i915/execlists: Direct submit onto idle engine   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/9] drm/i915: Speed up idle detection by kicking the tasklets   Patchwork
• [PATCH 1/6] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 2/6] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 3/6] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 4/6] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 5/6] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 6/6] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 1/7] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 2/7] drm/i915: Combine set-wedged protection and tasklet kicking   Chris Wilson
• [PATCH 3/7] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 4/7] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 5/7] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 6/7] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 7/7] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/6] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/6] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• [PATCH 1/7] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 2/7] drm/i915: Disable tasklet scheduling across initial scheduling   Chris Wilson
• [PATCH 3/7] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 4/7] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 5/7] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 6/7] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 7/7] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• [PATCH 1/7] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 2/7] drm/i915: Disable tasklet scheduling across initial scheduling   Chris Wilson
• [PATCH 3/7] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 4/7] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 5/7] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 6/7] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 7/7] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/7] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• [PATCH 01/80] drm/i915: Flush submission tasklet after bumping priority   Chris Wilson
• [PATCH 02/80] drm/i915: Disable tasklet scheduling across initial scheduling   Chris Wilson
• [PATCH 03/80] drm/i915/execlists: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 04/80] drm/i915/guc: Make submission tasklet hardirq safe   Chris Wilson
• [PATCH 05/80] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 06/80] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 07/80] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 08/80] drm/i915: Don't request a bug report for unsafe module parameters   Chris Wilson
• [PATCH 09/80] drm/syncobj: Use dma_fence_wait for the simple wait case   Chris Wilson
• [PATCH 10/80] drm/i915/selftests: Refactor common flush_test()   Chris Wilson
• [PATCH 11/80] drm/i915/selftests: Flush GPU activity before completing live_contexts   Chris Wilson
• [PATCH 12/80] drm/i915/dp: Silence static checkers for n_entries   Chris Wilson
• [PATCH 13/80] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 14/80] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 15/80] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 16/80] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 17/80] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 18/80] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 19/80] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 20/80] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 21/80] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 22/80] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 23/80] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 24/80] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 25/80] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 26/80] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 27/80] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 28/80] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 29/80] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 30/80] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 31/80] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 32/80] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 33/80] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 34/80] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 35/80] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 36/80] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 37/80] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 38/80] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 39/80] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 40/80] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 41/80] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 42/80] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 43/80] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 44/80] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 45/80] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 46/80] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 47/80] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 48/80] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 49/80] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 50/80] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 51/80] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 52/80] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 53/80] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 54/80] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 55/80] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 56/80] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 57/80] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 58/80] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 59/80] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 60/80] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 61/80] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 62/80] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 63/80] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 64/80] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 65/80] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 66/80] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 67/80] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 68/80] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 69/80] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 70/80] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 71/80] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 72/80] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 73/80] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 74/80] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 75/80] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 76/80] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 77/80] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 78/80] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 79/80] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• [PATCH 80/80] preempt   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/80] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/80] drm/i915: Flush submission tasklet after bumping priority   Patchwork
• [PATCH 1/5] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 2/5] drm/i915: Combine tasklet_kill and tasklet_disable   Chris Wilson
• [PATCH 3/5] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 4/5] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 5/5] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915: Wrap tasklet_struct for abuse   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/5] drm/i915: Wrap tasklet_struct for abuse   Patchwork
• [PATCH 01/74] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/74] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 03/74] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 04/74] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 05/74] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 06/74] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 07/74] drm/i915/execlists: Use rmb() to order CSB reads   Chris Wilson
• [PATCH 08/74] Revert "drm/i915/cnl: Use mmio access to context status buffer"   Chris Wilson
• [PATCH 09/74] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 10/74] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 11/74] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 12/74] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 13/74] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 14/74] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 15/74] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 16/74] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 17/74] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 18/74] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 19/74] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 20/74] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 21/74] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 22/74] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 23/74] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 24/74] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 25/74] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 26/74] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 27/74] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 28/74] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 29/74] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 30/74] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 31/74] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 32/74] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 33/74] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 34/74] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 35/74] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 36/74] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 37/74] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 38/74] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 39/74] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 40/74] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 41/74] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 42/74] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 43/74] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 44/74] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 45/74] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 46/74] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 47/74] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 48/74] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 49/74] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 50/74] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 51/74] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 52/74] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 53/74] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 54/74] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 55/74] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 56/74] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 57/74] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 58/74] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 59/74] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 60/74] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 61/74] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 62/74] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 63/74] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 64/74] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 65/74] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 66/74] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 67/74] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 68/74] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 69/74] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 70/74] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 71/74] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 72/74] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 73/74] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• [PATCH 74/74] preempt   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/74] drm/i915: Remove tasklet flush before disable   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/74] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/77] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/77] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 03/77] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/77] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 05/77] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 06/77] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 07/77] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 08/77] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 09/77] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 10/77] drm/i915/execlists: Use rmb() to order CSB reads   Chris Wilson
• [PATCH 11/77] Revert "drm/i915/cnl: Use mmio access to context status buffer"   Chris Wilson
• [PATCH 12/77] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 13/77] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 14/77] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 15/77] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 16/77] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 17/77] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 18/77] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 19/77] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 20/77] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 21/77] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/77] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/77] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/77] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 25/77] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 26/77] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 27/77] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 28/77] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 29/77] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 30/77] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 31/77] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 32/77] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 33/77] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 34/77] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 35/77] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 36/77] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 37/77] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 38/77] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 39/77] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 40/77] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 41/77] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 42/77] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 43/77] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 44/77] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 45/77] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 46/77] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 47/77] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 48/77] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 49/77] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 50/77] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 51/77] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 52/77] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 53/77] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 54/77] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 55/77] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 56/77] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 57/77] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 58/77] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 59/77] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 60/77] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 61/77] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 62/77] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 63/77] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 64/77] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 65/77] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 66/77] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 67/77] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 68/77] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 69/77] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 70/77] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 71/77] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 72/77] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 73/77] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 74/77] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 75/77] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 76/77] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• [PATCH 77/77] preempt   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/77] drm/i915: Remove tasklet flush before disable   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/77] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/28] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/28] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 03/28] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/28] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 05/28] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 06/28] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 07/28] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 08/28] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 09/28] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 10/28] drm/i915/execlists: Use rmb() to order CSB reads   Chris Wilson
• [PATCH 11/28] Revert "drm/i915/cnl: Use mmio access to context status buffer"   Chris Wilson
• [PATCH 12/28] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 13/28] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 14/28] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 15/28] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 16/28] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 17/28] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 18/28] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 19/28] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 20/28] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 21/28] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/28] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/28] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/28] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 25/28] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 26/28] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 27/28] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 28/28] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/28] drm/i915: Remove tasklet flush before disable   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/28] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH] drm/i915/selftests: scrub 64K   Matthew Auld
• ✓ Fi.CI.BAT: success for drm/i915/selftests: scrub 64K   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915/selftests: scrub 64K   Patchwork
• [PATCH] drm/i915/icl: Read the correct Gen11 interrupt registers   Oscar Mateo
• ✓ Fi.CI.BAT: success for drm/i915/icl: Read the correct Gen11 interrupt registers   Patchwork
• ✗ Fi.CI.IGT: failure for drm/i915/icl: Read the correct Gen11 interrupt registers   Patchwork
• [PATCH 01/59] drm/i915/execlists: Relax CSB force-mmio for VT-d   Chris Wilson
• [PATCH 02/59] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/59] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 04/59] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 05/59] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 06/59] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 07/59] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 08/59] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/59] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 10/59] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 11/59] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 12/59] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 13/59] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 14/59] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 15/59] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 16/59] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 17/59] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 18/59] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 19/59] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 20/59] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 21/59] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 22/59] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/59] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/59] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 25/59] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 26/59] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 27/59] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 28/59] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 29/59] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 30/59] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 31/59] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 32/59] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 33/59] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 34/59] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 35/59] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 36/59] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 37/59] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 38/59] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 39/59] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 40/59] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 41/59] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 42/59] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 43/59] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 44/59] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 45/59] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 46/59] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 47/59] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 48/59] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 49/59] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 50/59] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 51/59] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 52/59] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 53/59] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 54/59] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 55/59] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 56/59] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 57/59] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 58/59] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 59/59] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/59] drm/i915/execlists: Relax CSB force-mmio for VT-d   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/59] drm/i915/execlists: Relax CSB force-mmio for VT-d   Patchwork
• [PATCH 1/2] evict-stack   Chris Wilson
• [PATCH 2/2] cache-size   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] evict-stack   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] evict-stack   Patchwork
• [PATCH] evict-stack   Chris Wilson
• [PATCH 1/5] evict-stack   Chris Wilson
• [PATCH 2/5] cache-size   Chris Wilson
• [PATCH 3/5] mm-once   Chris Wilson
• [PATCH 4/5] pin-high   Chris Wilson
• [PATCH 5/5] pin-high-ring   Chris Wilson
• ✓ Fi.CI.BAT: success for evict-stack   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/5] evict-stack   Patchwork
• ✓ Fi.CI.IGT: success for evict-stack   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/5] evict-stack   Patchwork
• [PATCH 1/2] evict-stack   Chris Wilson
• [PATCH 2/2] cache-size   Chris Wilson
• [PATCH 1/6] evict-stack   Chris Wilson
• [PATCH 2/6] cache-size   Chris Wilson
• [PATCH 3/6] mm-once   Chris Wilson
• [PATCH 4/6] pin-high   Chris Wilson
• [PATCH 5/6] pin-high-ring   Chris Wilson
• [PATCH 6/6] purge   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] evict-stack   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/6] evict-stack   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/2] evict-stack   Patchwork
• [PATCH 01/84] evict-stack   Chris Wilson
• [PATCH 02/84] cache-size   Chris Wilson
• [PATCH 03/84] mm-once   Chris Wilson
• [PATCH 04/84] pin-high   Chris Wilson
• [PATCH 05/84] pin-high-ring   Chris Wilson
• [PATCH 06/84] purge   Chris Wilson
• [PATCH 07/84] drm/i915: Mark up nested spinlocks   Chris Wilson
• [PATCH 08/84] drm/i915/execlists: Relax CSB force-mmio for VT-d   Chris Wilson
• [PATCH 09/84] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 10/84] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 11/84] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 12/84] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 13/84] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 14/84] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 15/84] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 16/84] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 17/84] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 18/84] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 19/84] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 20/84] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 21/84] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 22/84] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 23/84] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 24/84] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 25/84] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 26/84] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 27/84] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 28/84] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 29/84] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 30/84] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 31/84] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 32/84] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 33/84] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 34/84] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 35/84] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 36/84] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 37/84] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 38/84] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 39/84] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 40/84] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 41/84] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 42/84] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 43/84] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 44/84] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 45/84] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 46/84] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 47/84] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 48/84] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 49/84] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 50/84] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 51/84] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 52/84] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 53/84] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 54/84] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 55/84] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 56/84] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 57/84] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 58/84] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 59/84] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 60/84] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 61/84] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 62/84] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 63/84] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 64/84] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 65/84] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 66/84] drm/i915: Reduce spinlock hold time during notify_ring() interrupt   Chris Wilson
• [PATCH 67/84] drm/i915: Move the irq_counter inside the spinlock   Chris Wilson
• [PATCH 68/84] drm/i915: Only signal from interrupt when requested   Chris Wilson
• [PATCH 69/84] drm/i915/execlists: Switch to rb_root_cached   Chris Wilson
• [PATCH 70/84] drm/i915: Priority boost for new clients   Chris Wilson
• [PATCH 71/84] drm/i915: Priority boost switching to an idle ring   Chris Wilson
• [PATCH 72/84] drm/i915: Refactor export_fence() after i915_vma_move_to_active()   Chris Wilson
• [PATCH 73/84] drm/i915: Start returning an error from i915_vma_move_to_active()   Chris Wilson
• [PATCH 74/84] drm/i915: Track vma activity per fence.context, not per engine   Chris Wilson
• [PATCH 75/84] drm/i915: Track the last-active inside the i915_vma   Chris Wilson
• [PATCH 76/84] drm/i915: Introduce the i915_user_extension_method   Chris Wilson
• [PATCH 77/84] drm/i915: Extend CREATE_CONTEXT to allow inheritance ala clone()   Chris Wilson
• [PATCH 78/84] drm/i915: Allow contexts to share a single timeline across all engines   Chris Wilson
• [PATCH 79/84] drm/i915: Fix I915_EXEC_RING_MASK   Chris Wilson
• [PATCH 80/84] drm/i915: Re-arrange execbuf so context is known before engine   Chris Wilson
• [PATCH 81/84] drm/i915: Allow a context to define its set of engines   Chris Wilson
• [PATCH 82/84] drm/i915/execlists: Flush the tasklet before unpinning   Chris Wilson
• [PATCH 83/84] RFC drm/i915: Load balancing across a virtual engine   Chris Wilson
• [PATCH 84/84] preempt   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/84] evict-stack   Patchwork
• [PATCH 1/7] evict-stack   Chris Wilson
• [PATCH 2/7] cache-size   Chris Wilson
• [PATCH 3/7] mm-once   Chris Wilson
• [PATCH 4/7] pin-high   Chris Wilson
• [PATCH 5/7] pin-high-ring   Chris Wilson
• [PATCH 6/7] purge   Chris Wilson
• [PATCH 7/7] shrink-stat   Chris Wilson
• ✓ Fi.CI.IGT: success for series starting with [1/6] evict-stack   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/7] evict-stack   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/84] evict-stack   Patchwork
• [PATCH] cache-size   Chris Wilson
• ✓ Fi.CI.BAT: success for cache-size   Patchwork
• ✓ Fi.CI.IGT: success for cache-size   Patchwork
• [PATCH 1/7] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 2/7] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 3/7] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 4/7] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 5/7] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 6/7] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 7/7] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/7] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/7] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/59] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/59] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/59] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/59] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/59] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 06/59] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 07/59] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 08/59] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 09/59] drm/i915: Mark up nested spinlocks   Chris Wilson
• [PATCH 10/59] drm/i915/execlists: Relax CSB force-mmio for VT-d   Chris Wilson
• [PATCH 11/59] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 12/59] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 13/59] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 14/59] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 15/59] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 16/59] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 17/59] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 18/59] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 19/59] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 20/59] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 21/59] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 22/59] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 23/59] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 24/59] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 25/59] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 26/59] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 27/59] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 28/59] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 29/59] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 30/59] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 31/59] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 32/59] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 33/59] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 34/59] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 35/59] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 36/59] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 37/59] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 38/59] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 39/59] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 40/59] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 41/59] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 42/59] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 43/59] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 44/59] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 45/59] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 46/59] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 47/59] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 48/59] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 49/59] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 50/59] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 51/59] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 52/59] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 53/59] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 54/59] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 55/59] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 56/59] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 57/59] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 58/59] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 59/59] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 01/67] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/67] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/67] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/67] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/67] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 06/67] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 07/67] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 08/67] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 09/67] drm/i915: Mark up nested spinlocks   Chris Wilson
• [PATCH 10/67] drm/i915/execlists: Relax CSB force-mmio for VT-d   Chris Wilson
• [PATCH 11/67] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 12/67] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 13/67] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 14/67] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 15/67] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 16/67] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 17/67] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 18/67] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 19/67] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 20/67] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 21/67] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 22/67] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 23/67] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 24/67] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 25/67] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 26/67] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 27/67] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 28/67] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 29/67] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 30/67] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 31/67] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 32/67] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 33/67] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 34/67] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 35/67] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 36/67] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 37/67] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 38/67] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 39/67] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 40/67] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 41/67] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 42/67] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 43/67] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 44/67] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 45/67] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 46/67] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 47/67] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 48/67] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 49/67] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 50/67] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 51/67] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 52/67] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 53/67] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 54/67] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 55/67] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 56/67] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 57/67] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 58/67] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 59/67] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 60/67] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 61/67] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 62/67] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 63/67] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 64/67] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 65/67] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 66/67] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 67/67] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/59] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/67] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/59] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/67] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/67] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/67] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/67] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/67] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/67] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 06/67] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 07/67] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 08/67] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 09/67] drm/i915: Mark up nested spinlocks   Chris Wilson
• [PATCH 10/67] drm/i915/execlists: Relax CSB force-mmio for VT-d   Chris Wilson
• [PATCH 11/67] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 12/67] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 13/67] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 14/67] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 15/67] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 16/67] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 17/67] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 18/67] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 19/67] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• [PATCH 20/67] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 21/67] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 22/67] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 23/67] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 24/67] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 25/67] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 26/67] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 27/67] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 28/67] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 29/67] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 30/67] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 31/67] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 32/67] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 33/67] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 34/67] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 35/67] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 36/67] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 37/67] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 38/67] drm/i915: Disable preemption and sleeping while using the punit sideband   Chris Wilson
• [PATCH 39/67] drm/i915: Lift acquiring the vlv punit magic to a common sb-get   Chris Wilson
• [PATCH 40/67] drm/i915: Lift sideband locking for vlv_punit_(read|write)   Chris Wilson
• [PATCH 41/67] drm/i915: Reduce RPS update frequency on Valleyview/Cherryview   Chris Wilson
• [PATCH 42/67] Revert "drm/i915: Avoid tweaking evaluation thresholds on Baytrail v3"   Chris Wilson
• [PATCH 43/67] drm/i915: Replace pcu_lock with sb_lock   Chris Wilson
• [PATCH 44/67] drm/i915: Separate sideband declarations to intel_sideband.h   Chris Wilson
• [PATCH 45/67] drm/i915: Merge sbi read/write into a single accessor   Chris Wilson
• [PATCH 46/67] drm/i915: Merge sandybridge_pcode_(read|write)   Chris Wilson
• [PATCH 47/67] drm/i915: Move sandybride pcode access to intel_sideband.c   Chris Wilson
• [PATCH 48/67] drm/i915: Mark up Ironlake ips with rpm wakerefs   Chris Wilson
• [PATCH 49/67] drm/i915: Record logical context support in driver caps   Chris Wilson
• [PATCH 50/67] drm/i915: Generalize i915_gem_sanitize() to reset contexts   Chris Wilson
• [PATCH 51/67] drm/i915: Enable render context support for Ironlake (gen5)   Chris Wilson
• [PATCH 52/67] drm/i915: Enable render context support for gen4 (Broadwater to Cantiga)   Chris Wilson
• [PATCH 53/67] drm/i915: Split GT powermanagement functions to intel_gt_pm.c   Chris Wilson
• [PATCH 54/67] drm/i915: Move rps worker to intel_gt_pm.c   Chris Wilson
• [PATCH 55/67] drm/i915: Move all the RPS irq handlers to intel_gt_pm   Chris Wilson
• [PATCH 56/67] drm/i915: Track HAS_RPS alongside HAS_RC6 in the device info   Chris Wilson
• [PATCH 57/67] drm/i915: Remove defunct intel_suspend_gt_powersave()   Chris Wilson
• [PATCH 58/67] drm/i915: Reorder GT interface code   Chris Wilson
• [PATCH 59/67] drm/i915: Split control of rps and rc6   Chris Wilson
• [PATCH 60/67] drm/i915: Enabling rc6 and rps have different requirements, so separate them   Chris Wilson
• [PATCH 61/67] drm/i915: Simplify rc6/rps enabling   Chris Wilson
• [PATCH 62/67] drm/i915: Refactor frequency bounds computation   Chris Wilson
• [PATCH 63/67] drm/i915: Rename rps min/max frequencies   Chris Wilson
• [PATCH 64/67] drm/i915: Pull IPS into GT power management   Chris Wilson
• [PATCH 65/67] drm/i915,intel_ips: Enable GPU wait-boosting with IPS   Chris Wilson
• [PATCH 66/67] drm/i915: Remove unwarranted clamping for hsw/bdw   Chris Wilson
• [PATCH 67/67] drm/i915: Support per-context user requests for GPU frequency control   Chris Wilson
• [PATCH 01/10] drm/i915: Mark up nested spinlocks   Chris Wilson
• [PATCH 02/10] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/10] drm/i915: Wrap tasklet_struct for abuse   Chris Wilson
• [PATCH 04/10] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 05/10] drm/i915/execlists: Direct submit onto idle engines   Chris Wilson
• [PATCH 06/10] drm/i915/execlists: Direct submission from irq handler   Chris Wilson
• [PATCH 07/10] drm/i915: Rearrange gen8_cs_irq_handler   Chris Wilson
• [PATCH 08/10] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/10] drm/i915: Speed up idle detection by kicking the tasklets   Chris Wilson
• [PATCH 10/10] drm/i915: Detect if we missed kicking the execlists tasklet   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/67] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/10] drm/i915: Mark up nested spinlocks   Patchwork
• [PATCH] drm/i915: Warn when connection_mutex is not locked in intel_modeset_checks.   Maarten Lankhorst
• ✗ Fi.CI.IGT: failure for series starting with [01/67] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/10] drm/i915: Mark up nested spinlocks   Patchwork
• ✓ Fi.CI.BAT: success for drm/i915: Warn when connection_mutex is not locked in intel_modeset_checks.   Patchwork
• [PATCH 1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Yunwei Zhang
• [PATCH 2/3] drm/i915/icl: Enable WaProgramMgsrForCorrectSliceSpecificMmioReads   Yunwei Zhang
• [PATCH 3/3] drm/i915: Implement WaProgramMgsrForL3BankSpecificMmioReads   Yunwei Zhang
• ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Patchwork
• ✗ Fi.CI.IGT: failure for drm/i915: Warn when connection_mutex is not locked in intel_modeset_checks.   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Patchwork
• [PATCH 1/7] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 2/7] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 3/7] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 4/7] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 5/7] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 6/7] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 7/7] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 1/8] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 2/8] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 3/8] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 4/8] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 5/8] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 6/8] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 7/8] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 8/8] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH] ddb allocation algorithm optimization   Mahesh Kumar
• [PATCH 01/25] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/25] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/25] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/25] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/25] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/25] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/25] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/25] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/25] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/25] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/25] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/25] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/25] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/25] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/25] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/25] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/25] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/25] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/25] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/25] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/25] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/25] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/25] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/25] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/25] post-reset-csb   Chris Wilson
• [PATCH 01/26] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/26] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/26] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/26] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/26] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/26] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/26] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/26] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/26] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/26] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/26] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/26] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/26] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/26] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/26] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/26] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/26] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/26] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/26] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/26] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/26] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/26] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/26] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/26] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/26] post-reset-csb   Chris Wilson
• [PATCH 26/26] process-csb-from-irq   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/7] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Matthew Auld
• [PATCH 2/4] mm: check for proper migrate type during isolation   Matthew Auld
• [PATCH 3/4] mm: add find_alloc_contig_pages() interface   Matthew Auld
• [PATCH 4/4] drm/i915: support 1G pages for the 48b PPGTT   Matthew Auld
• ✓ Fi.CI.BAT: success for series starting with [1/8] drm/i915: Shrink search list for active timelines   Patchwork
• [PATCH 01/27] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/27] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/27] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/27] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/27] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/27] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/27] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/27] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/27] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/27] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/27] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/27] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/27] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/27] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/27] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/27] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/27] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/27] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/27] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/27] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/27] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/27] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/27] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/27] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/27] post-reset-csb   Chris Wilson
• [PATCH 26/27] process-csb-from-irq   Chris Wilson
• [PATCH 27/27] direct-submit   Chris Wilson
• ✗ Fi.CI.BAT: failure for ddb allocation algorithm optimization   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/25] drm/i915: Shrink search list for active timelines   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/26] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Patchwork
• [PATCH] ddb allocation algorithm optimization   Mahesh Kumar
• ✗ Fi.CI.BAT: failure for series starting with [01/27] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.BAT: success for ddb allocation algorithm optimization (rev2)   Patchwork
• [PATCH 01/27] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/27] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/27] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/27] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/27] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/27] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/27] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/27] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/27] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/27] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/27] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/27] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/27] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/27] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/27] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/27] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/27] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/27] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/27] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/27] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/27] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/27] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/27] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/27] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/27] post-reset-csb   Chris Wilson
• [PATCH 26/27] process-csb-from-irq   Chris Wilson
• [PATCH 27/27] direct-submit   Chris Wilson
• [PATCH 01/27] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/27] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/27] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/27] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/27] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/27] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/27] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/27] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/27] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/27] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/27] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/27] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/27] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/27] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/27] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/27] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/27] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/27] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/27] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/27] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/27] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/27] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/27] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/27] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/27] post-reset-csb   Chris Wilson
• [PATCH 26/27] process-csb-from-irq   Chris Wilson
• [PATCH 27/27] unify csb register pointers   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/27] drm/i915: Shrink search list for active timelines   Patchwork
• [PATCH 01/28] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/28] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/28] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/28] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/28] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/28] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/28] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/28] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/28] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/28] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/28] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/28] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/28] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/28] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/28] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/28] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/28] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/28] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/28] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/28] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/28] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/28] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/28] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/28] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/28] post-reset-csb   Chris Wilson
• [PATCH 26/28] process-csb-from-irq   Chris Wilson
• [PATCH 27/28] unify csb register pointers   Chris Wilson
• [PATCH 28/28] direct-submit   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/27] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/28] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/7] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH] ddb allocation algorithm optimization   Mahesh Kumar
• ✓ Fi.CI.IGT: success for series starting with [1/8] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.BAT: success for ddb allocation algorithm optimization (rev3)   Patchwork
• ✗ Fi.CI.IGT: failure for ddb allocation algorithm optimization   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [01/25] drm/i915: Shrink search list for active timelines   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Patchwork
• [PATCH 01/25] drm/i915: Shrink search list for active timelines   Chris Wilson
• [PATCH 02/25] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 03/25] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 04/25] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 05/25] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 06/25] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 07/25] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/25] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/25] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/25] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/25] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/25] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/25] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/25] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/25] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/25] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/25] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/25] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/25] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/25] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/25] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/25] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/25] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/25] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/25] post-reset-csb   Chris Wilson
• ✓ Fi.CI.IGT: success for ddb allocation algorithm optimization (rev2)   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/25] drm/i915: Shrink search list for active timelines   Patchwork
• [PATCH 01/24] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/24] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/24] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/24] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/24] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/24] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 07/24] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 08/24] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/24] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 10/24] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 11/24] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 12/24] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 13/24] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 14/24] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 15/24] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 16/24] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 17/24] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 18/24] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 19/24] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 20/24] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 21/24] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 22/24] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/24] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/24] sanitize-engines   Chris Wilson
• [PATCH 01/24] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/24] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/24] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/24] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/24] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/24] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 07/24] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 08/24] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/24] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 10/24] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 11/24] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 12/24] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 13/24] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 14/24] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 15/24] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 16/24] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 17/24] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 18/24] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 19/24] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 20/24] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 21/24] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 22/24] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/24] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/24] sanitize-engines   Chris Wilson
• ✗ Fi.CI.IGT: failure for series starting with [01/27] drm/i915: Shrink search list for active timelines   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/24] drm/i915: Remove tasklet flush before disable   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/28] drm/i915: Shrink search list for active timelines   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/24] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/29] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/29] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/29] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/29] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/29] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/29] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 07/29] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 08/29] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/29] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 10/29] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 11/29] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 12/29] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 13/29] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 14/29] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 15/29] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 16/29] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 17/29] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 18/29] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 19/29] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 20/29] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 21/29] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 22/29] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 23/29] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 24/29] sanitize-engines   Chris Wilson
• [PATCH 25/29] post-reset-csb   Chris Wilson
• [PATCH 26/29] process-csb-from-irq   Chris Wilson
• [PATCH 27/29] unify csb register pointers   Chris Wilson
• [PATCH 28/29] direct-submit   Chris Wilson
• [PATCH 29/29] gtirr   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/29] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.IGT: success for ddb allocation algorithm optimization (rev3)   Patchwork
• [PATCH 1/7] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 2/7] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 3/7] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 4/7] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 5/7] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 6/7] drm/i915/execlists: Split out CSB processing   Chris Wilson
• [PATCH 7/7] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Matthew Auld
• [PATCH 2/4] mm: check for proper migrate type during isolation   Matthew Auld
• [PATCH 3/4] mm: add find_alloc_contig_pages() interface   Matthew Auld
• [PATCH 4/4] drm/i915: support 1G pages for the 48b PPGTT   Matthew Auld
• ✓ Fi.CI.BAT: success for series starting with [1/7] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/29] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/34] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/34] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/34] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/34] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/34] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/34] drm/i915/execlists: Split out CSB processing   Chris Wilson
• [PATCH 07/34] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/34] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/34] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/34] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 11/34] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 12/34] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 13/34] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 14/34] drm/i915: Track the purgeable objects on a separate eviction list   Chris Wilson
• [PATCH 15/34] drm/i915: Refactor unsettting obj->mm.pages   Chris Wilson
• [PATCH 16/34] drm/i915: Report all objects with allocated pages to the shrinker   Chris Wilson
• [PATCH 17/34] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 18/34] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 19/34] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 20/34] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 21/34] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 22/34] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 23/34] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 24/34] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 25/34] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 26/34] sanitize-engines   Chris Wilson
• [PATCH 27/34] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 28/34] post-reset-csb   Chris Wilson
• [PATCH 29/34] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 30/34] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 31/34] unify csb register pointers   Chris Wilson
• [PATCH 32/34] process-csb-from-irq   Chris Wilson
• [PATCH 33/34] direct-submit   Chris Wilson
• [PATCH 34/34] gtirr   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/34] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/27] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/27] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/27] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/27] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/27] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/27] drm/i915/execlists: Split out CSB processing   Chris Wilson
• [PATCH 07/27] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/27] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/27] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 10/27] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 11/27] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 12/27] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 13/27] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/27] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 15/27] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 16/27] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 17/27] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 18/27] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 19/27] sanitize-engines   Chris Wilson
• [PATCH 20/27] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 21/27] post-reset-csb   Chris Wilson
• [PATCH 22/27] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 23/27] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 24/27] unify csb register pointers   Chris Wilson
• [PATCH 25/27] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 26/27] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 27/27] gtirr   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/27] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH 01/19] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/19] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/19] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/19] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/19] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/19] drm/i915/execlists: Split out CSB processing   Chris Wilson
• [PATCH 07/19] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/19] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/19] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 10/19] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 11/19] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 12/19] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 13/19] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/19] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 15/19] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 16/19] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 17/19] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 18/19] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 19/19] sanitize-engines   Chris Wilson
• [PATCH 01/27] drm/i915: Remove tasklet flush before disable   Chris Wilson
• [PATCH 02/27] drm/i915: Only sync tasklets once for recursive reset preparation   Chris Wilson
• [PATCH 03/27] drm/i915/execlists: Refactor out complete_preempt_context()   Chris Wilson
• [PATCH 04/27] drm/i915: Move engine reset prepare/finish to backends   Chris Wilson
• [PATCH 05/27] drm/i915: Split execlists/guc reset preparations   Chris Wilson
• [PATCH 06/27] drm/i915/execlists: Split out CSB processing   Chris Wilson
• [PATCH 07/27] drm/i915/execlists: Flush pending preemption events during reset   Chris Wilson
• [PATCH 08/27] drm/i915: Stop parking the signaler around reset   Chris Wilson
• [PATCH 09/27] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 10/27] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 11/27] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 12/27] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 13/27] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/27] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 15/27] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 16/27] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 17/27] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 18/27] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 19/27] sanitize-engines   Chris Wilson
• [PATCH 20/27] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 21/27] post-reset-csb   Chris Wilson
• [PATCH 22/27] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 23/27] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 24/27] unify csb register pointers   Chris Wilson
• [PATCH 25/27] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 26/27] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 27/27] gtirr   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/19] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/27] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/7] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/4] mm: change type of free_contig_range(nr_pages) to unsigned long   Patchwork
• [PATCH 01/24] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 02/24] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 03/24] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 04/24] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 05/24] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 06/24] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 07/24] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 08/24] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 09/24] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 10/24] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 11/24] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 12/24] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 13/24] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 14/24] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 15/24] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 16/24] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 17/24] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 18/24] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 19/24] drm/i915: Combine gt irq ack/handlers   Chris Wilson
• [PATCH 20/24] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 21/24] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 22/24] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 23/24] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 24/24] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• ✗ Fi.CI.IGT: failure for series starting with [01/19] drm/i915: Remove tasklet flush before disable   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/24] drm/i915: Move request->ctx aside   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/27] drm/i915: Remove tasklet flush before disable   Patchwork
• [PATCH] no-bcs   Chris Wilson
• ✗ Fi.CI.BAT: failure for no-bcs   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/24] drm/i915: Move request->ctx aside   Patchwork
• [PATCH] drm/i915/icl: Don't update enabled dbuf slices struct until updated in hw   Mahesh Kumar
• ✓ Fi.CI.BAT: success for drm/i915/icl: Don't update enabled dbuf slices struct until updated in hw   Patchwork
• [PATCH 01/19] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 02/19] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 03/19] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 04/19] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 05/19] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 06/19] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 07/19] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 08/19] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 09/19] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 10/19] drm/i915/execlists: HWACK checking superseded checking port[0].count   Chris Wilson
• [PATCH 11/19] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 12/19] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 13/19] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 14/19] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 15/19] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 16/19] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 17/19] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 18/19] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 19/19] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/19] drm/i915/selftests: Wait longer for the old active request   Patchwork
• [PATCH 1/2] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 2/2] HAX guc please   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• ✗ Fi.CI.IGT: failure for drm/i915/icl: Don't update enabled dbuf slices struct until updated in hw   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/19] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• [PATCH 01/21] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 02/21] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 03/21] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 04/21] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 05/21] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 06/21] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 07/21] drm/i915: Flush the RING stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 08/21] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 09/21] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 10/21] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 11/21] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 12/21] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 13/21] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 14/21] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 15/21] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 16/21] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 17/21] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 18/21] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 19/21] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 20/21] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 21/21] drm/i915: Combine gt irq ack/handlers   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/21] drm/i915: Move request->ctx aside   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/21] drm/i915: Move request->ctx aside   Patchwork
• [PATCH 01/17] drm/i915: Move request->ctx aside   Chris Wilson
• [PATCH 02/17] drm/i915: Move fiddling with engine->last_retired_context   Chris Wilson
• [PATCH 03/17] drm/i915: Store a pointer to intel_context in i915_request   Chris Wilson
• [PATCH 04/17] drm/i915: Pull the context->pin_count dec into the common intel_context_unpin   Chris Wilson
• [PATCH 05/17] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 06/17] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 07/17] drm/i915: Flush the RING stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 08/17] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 09/17] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 10/17] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 11/17] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 12/17] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 13/17] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 14/17] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 15/17] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 16/17] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 17/17] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/17] drm/i915: Move request->ctx aside   Patchwork
• [PATCH 01/18] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 02/18] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 03/18] drm/i915: Flush the RING stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 04/18] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 05/18] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 06/18] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 07/18] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 08/18] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/18] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 10/18] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 11/18] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 12/18] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 13/18] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 14/18] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 15/18] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 16/18] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 17/18] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 18/18] drm/i915: Combine gt irq ack/handlers   Chris Wilson
• ✓ Fi.CI.IGT: success for series starting with [01/17] drm/i915: Move request->ctx aside   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/18] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• [PATCH 01/25] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 02/25] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 03/25] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 04/25] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 05/25] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 06/25] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/25] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/25] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 09/25] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 10/25] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 11/25] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 12/25] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 13/25] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 14/25] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 15/25] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 16/25] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 17/25] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 18/25] drm/i915: Combine gt irq ack/handlers   Chris Wilson
• [PATCH 19/25] drm/i915/execlists: Force preemption via reset on timeout   Chris Wilson
• [PATCH 20/25] drm/i915/execlists: Try preempt-reset from hardirq timer context   Chris Wilson
• [PATCH 21/25] drm/i915/preemption: Select timeout when scheduling   Chris Wilson
• [PATCH 22/25] drm/i915: Use a preemption timeout to enforce interactivity   Chris Wilson
• [PATCH 23/25] drm/i915: Allow user control over preempt timeout on their important context   Chris Wilson
• [PATCH 24/25] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 25/25] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/25] drm/i915/selftests: Wait longer for the old active request   Patchwork
• [PATCH 01/18] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 02/18] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 03/18] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 04/18] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 05/18] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 06/18] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/18] drm/i915: Correctly handle error path in i915_gem_init_hw   Chris Wilson
• [PATCH 08/18] drm/i915: Move i915_gem_fini to i915_gem.c   Chris Wilson
• [PATCH 09/18] drm/i915: Introduce i915_gem_fini_hw for symmetry with i915_gem_init_hw   Chris Wilson
• [PATCH 10/18] drm/i915: Add i915_gem_fini_hw to i915_gem_suspend   Chris Wilson
• [PATCH 11/18] drm/i915: Add i915_gem_fini_hw to i915_reset   Chris Wilson
• [PATCH 12/18] drm/i915: Park before resetting the submission backend   Chris Wilson
• [PATCH 13/18] drm/i915/guc: Restore symmetric doorbell cleanup   Chris Wilson
• [PATCH 14/18] drm/i915/uc: Fully sanitize uC within intel_uc_fini_hw   Chris Wilson
• [PATCH 15/18] drm/i915/uc: Use correct error code for GuC initialization failure   Chris Wilson
• [PATCH 16/18] drm/i915/uc: Use helper functions to detect fw load status   Chris Wilson
• [PATCH 17/18] drm/i915/uc: Trivial s/dev_priv/i915 in intel_uc.c   Chris Wilson
• [PATCH 18/18] HAX: Enable GuC for CI   Chris Wilson
• ✓ Fi.CI.IGT: success for series starting with [01/18] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• [PATCH 1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Yunwei Zhang
• [PATCH 2/3] drm/i915/icl: Enable WaProgramMgsrForCorrectSliceSpecificMmioReads   Yunwei Zhang
• [PATCH 3/3] drm/i915: Implement WaProgramMgsrForL3BankSpecificMmioReads   Yunwei Zhang
• ✓ Fi.CI.BAT: success for series starting with [01/18] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/25] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Patchwork
• [PATCH 01/12] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 02/12] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 03/12] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 04/12] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 05/12] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 06/12] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/12] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/12] flush-once   Chris Wilson
• [PATCH 09/12] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/12] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 11/12] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 12/12] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 01/19] drm/i915/selftests: Wait longer for the old active request   Chris Wilson
• [PATCH 02/19] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 03/19] drm/i915: Make intel_engine_dump irqsafe   Chris Wilson
• [PATCH 04/19] drm/i915/execlists: Handle copying default context state for atomic reset   Chris Wilson
• [PATCH 05/19] drm/i915: Allow init_breadcrumbs to be used from irq context   Chris Wilson
• [PATCH 06/19] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/19] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/19] flush-once   Chris Wilson
• [PATCH 09/19] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/19] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 11/19] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 12/19] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 13/19] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 14/19] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 15/19] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 16/19] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 17/19] drm/i915/execlists: Process the CSB directly from inside the irq handler   Chris Wilson
• [PATCH 18/19] drm/i915/execlists: Direct submission (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 19/19] drm/i915: Combine gt irq ack/handlers   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/12] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/19] drm/i915/selftests: Wait longer for the old active request   Patchwork
• [PATCH] drm/i915/icl: fix icl_unmap/map_plls_to_ports   Mahesh Kumar
• ✓ Fi.CI.BAT: success for drm/i915/icl: fix icl_unmap/map_plls_to_ports   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/18] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/3] drm/i915/cnl: Implement WaProgramMgsrForCorrectSliceSpecificMmioReads   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/12] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [01/19] drm/i915/selftests: Wait longer for the old active request   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915/icl: fix icl_unmap/map_plls_to_ports   Patchwork
• [PATCH] drm/i915/breadcrumbs: Tighten irqoff to the inner irq_lock   Chris Wilson
• ✗ Fi.CI.BAT: failure for drm/i915/breadcrumbs: Tighten irqoff to the inner irq_lock   Patchwork
• [PATCH] drm/i915/breadcrumbs: Tighten irqoff to the inner irq_lock   Chris Wilson
• ✗ Fi.CI.BAT: failure for drm/i915/breadcrumbs: Tighten irqoff to the inner irq_lock (rev2)   Patchwork
• [PATCH 01/19] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/19] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/19] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/19] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/19] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/19] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/19] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/19] flush-once   Chris Wilson
• [PATCH 09/19] drm/i915: Remove USES_GUC_SUBMISSION() pointer chasing from gen8_cs_irq_handler   Chris Wilson
• [PATCH 10/19] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 11/19] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 12/19] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 13/19] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 14/19] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 15/19] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 16/19] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 17/19] drm/i915/execlists: Direct submission from process (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 18/19] retire-req   Chris Wilson
• [PATCH 19/19] prev-rq   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/19] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/16] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/16] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/16] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/16] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/16] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/16] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/16] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/16] flush-once   Chris Wilson
• [PATCH 09/16] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 10/16] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 11/16] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 12/16] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 13/16] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 14/16] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 15/16] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 16/16] drm/i915/execlists: Direct submission from process (avoid tasklet/ksoftirqd)   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/16] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/10] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/10] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/10] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/10] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/10] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/10] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/10] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/10] flush-once   Chris Wilson
• [PATCH 09/10] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 10/10] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 01/17] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/17] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/17] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/17] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/17] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/17] drm/i915/guc: Keep guc submission permanently engaged   Chris Wilson
• [PATCH 07/17] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/17] flush-once   Chris Wilson
• [PATCH 09/17] drm/i915/execlists: Double check rpm wakeref   Chris Wilson
• [PATCH 10/17] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 11/17] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 12/17] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 13/17] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 14/17] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 15/17] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 16/17] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 17/17] drm/i915/execlists: Direct submission from process (avoid tasklet/ksoftirqd)   Chris Wilson
• ✗ Fi.CI.IGT: failure for series starting with [01/16] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/10] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/17] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/10] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 1/9] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 2/9] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 3/9] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 4/9] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 5/9] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 6/9] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 7/9] flush-once   Chris Wilson
• [PATCH 8/9] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 9/9] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 01/15] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/15] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/15] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/15] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/15] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/15] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 07/15] flush-once   Chris Wilson
• [PATCH 08/15] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 09/15] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 10/15] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 11/15] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 12/15] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 13/15] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 14/15] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 15/15] drm/i915/execlists: Direct submission from process (avoid tasklet/ksoftirqd)   Chris Wilson
• ✗ Fi.CI.IGT: failure for series starting with [01/17] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/15] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 02/15] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 03/15] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 04/15] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 05/15] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 06/15] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 07/15] flush-once   Chris Wilson
• [PATCH 08/15] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 09/15] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 10/15] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 11/15] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 12/15] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 13/15] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 14/15] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 15/15] drm/i915/execlists: Direct submission from process (avoid tasklet/ksoftirqd)   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/9] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/15] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [01/15] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/9] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/15] drm/mm: Reject over-sized allocation requests early   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/15] drm/mm: Reject over-sized allocation requests early   Patchwork
• [PATCH 01/10] drm/i915/gtt: Allow pagedirectory allocations to fail   Chris Wilson
• [PATCH 02/10] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 03/10] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 04/10] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 05/10] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 06/10] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 07/10] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/10] flush-once   Chris Wilson
• [PATCH 09/10] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 10/10] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 01/16] drm/i915/gtt: Allow pagedirectory allocations to fail   Chris Wilson
• [PATCH 02/16] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 03/16] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 04/16] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 05/16] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 06/16] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 07/16] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/16] flush-once   Chris Wilson
• [PATCH 09/16] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 10/16] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 11/16] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 12/16] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 13/16] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 14/16] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 15/16] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 16/16] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 1/6] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/6] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/6] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/6] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 5/6] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 6/6] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/10] drm/i915/gtt: Allow pagedirectory allocations to fail   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/16] drm/i915/gtt: Allow pagedirectory allocations to fail   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/6] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/6] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/6] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/6] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/6] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 5/6] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 6/6] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/6] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/4] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/4] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/4] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/4] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/4] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Special case kernel_context switch request   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/10] drm/i915/gtt: Allow pagedirectory allocations to fail   Patchwork
• [PATCH 1/3] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/3] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/3] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 1/4] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/4] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/4] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/4] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/3] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/4] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 1/3] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/3] drm/i915: Special case kernel_context switch request   Chris Wilson
• [PATCH 3/3] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/3] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Special case kernel_context switch request   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Look for an active kernel context before switching   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/8] drm-tip: 2018y-05m-23d-12h-29m-52s UTC integration manifest   Chris Wilson
• [PATCH 2/8] drm/i915/selftests: Pin the mock kernel context   Chris Wilson
• [PATCH 3/8] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 4/8] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 5/8] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 6/8] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 7/8] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 8/8] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [HAX] drm/i915/psr: Enable PSR   Dhinakaran Pandiyan
• ✗ Fi.CI.BAT: failure for series starting with [1/8] drm-tip: 2018y-05m-23d-12h-29m-52s UTC integration manifest   Patchwork
• ✗ Fi.CI.BAT: failure for drm/i915/psr: Enable PSR   Patchwork
• [PATCH 1/7] drm/i915/selftests: Pin the mock kernel context   Chris Wilson
• [PATCH 2/7] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 3/7] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 4/7] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 5/7] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 6/7] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 7/7] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 01/22] drm/i915/selftests: Pin the mock kernel context   Chris Wilson
• [PATCH 02/22] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 03/22] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 04/22] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 05/22] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 06/22] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 07/22] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 08/22] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 09/22] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 10/22] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 11/22] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 12/22] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 13/22] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/22] drm/i915/execlists: Wait for ELSP submission on restart   Chris Wilson
• [PATCH 15/22] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 16/22] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 17/22] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 18/22] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 19/22] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 20/22] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 21/22] prev-rq   Chris Wilson
• [PATCH 22/22] retire-req   Chris Wilson
• [PATCH 01/23] drm/i915/selftests: Pin the mock kernel context   Chris Wilson
• [PATCH 02/23] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 03/23] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 04/23] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 05/23] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 06/23] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 07/23] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 08/23] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 09/23] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 10/23] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 11/23] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 12/23] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 13/23] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/23] drm/i915/execlists: Wait for ELSP submission on restart   Chris Wilson
• [PATCH 15/23] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 16/23] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 17/23] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 18/23] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 19/23] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 20/23] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 21/23] prev-rq   Chris Wilson
• [PATCH 22/23] retire-req   Chris Wilson
• [PATCH 23/23] drm/i915: Wait for engines to idle before retiring   Chris Wilson
• [PATCH 01/24] drm/i915/selftests: Pin the mock kernel context   Chris Wilson
• [PATCH 02/24] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 03/24] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 04/24] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 05/24] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 06/24] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 07/24] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 08/24] drm/mm: Reject over-sized allocation requests early   Chris Wilson
• [PATCH 09/24] drm/mm: Add a search-by-address variant to only inspect a single hole   Chris Wilson
• [PATCH 10/24] drm/i915: Limit searching for PIN_HIGH   Chris Wilson
• [PATCH 11/24] drm/i915: Pin the ring high   Chris Wilson
• [PATCH 12/24] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 13/24] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 14/24] drm/i915/execlists: Wait for ELSP submission on restart   Chris Wilson
• [PATCH 15/24] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 16/24] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 17/24] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 18/24] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 19/24] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 20/24] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 21/24] prev-rq   Chris Wilson
• [PATCH 22/24] drm/i915: Wait for engines to idle before retiring   Chris Wilson
• [PATCH 23/24] drm/i915: Move engine request retirement to intel_engine_cs   Chris Wilson
• [PATCH 24/24] retire-req   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/7] drm/i915/selftests: Pin the mock kernel context   Patchwork
• [PATCH 1/4] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 2/4] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 3/4] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/4] huh   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/22] drm/i915/selftests: Pin the mock kernel context   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [01/23] drm/i915/selftests: Pin the mock kernel context   Patchwork
• [PATCH] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH] drm/i915: Look for an active kernel context before switching   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/24] drm/i915/selftests: Pin the mock kernel context   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/4] drm/i915: Look for an active kernel context before switching   Patchwork
• ✓ Fi.CI.BAT: success for drm/i915: Look for an active kernel context before switching (rev2)   Patchwork
• ✗ Fi.CI.IGT: failure for drm/i915: Look for an active kernel context before switching (rev2)   Patchwork
• [PATCH 1/4] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 2/4] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 3/4] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 4/4] huh   Chris Wilson
• [PATCH 1/8] drm/i915: Look for an active kernel context before switching   Chris Wilson
• [PATCH 2/8] fixup   Chris Wilson
• [PATCH 3/8] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 4/8] RFC drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 5/8] huh   Chris Wilson
• [PATCH 6/8] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 7/8] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 8/8] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 1/5] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/5] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 3/5] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 4/5] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 5/5] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/4] drm/i915: Look for an active kernel context before switching   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/8] drm/i915: Look for an active kernel context before switching   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 01/18] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 02/18] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 03/18] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 04/18] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 05/18] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 06/18] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 07/18] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/18] drm/i915/execlists: Wait for ELSP submission on restart   Chris Wilson
• [PATCH 09/18] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 10/18] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 11/18] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 12/18] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 13/18] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 14/18] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 15/18] prev-rq   Chris Wilson
• [PATCH 16/18] drm/i915: Wait for engines to idle before retiring   Chris Wilson
• [PATCH 17/18] drm/i915: Move engine request retirement to intel_engine_cs   Chris Wilson
• [PATCH 18/18] retire-req   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [01/18] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 01/18] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 02/18] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 03/18] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 04/18] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 05/18] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 06/18] drm/i915: Flush the ring stop bit after clearing RING_HEAD in reset   Chris Wilson
• [PATCH 07/18] drm/i915: Be irqsafe inside reset   Chris Wilson
• [PATCH 08/18] drm/i915/execlists: Wait for ELSP submission on restart   Chris Wilson
• [PATCH 09/18] drm/i915/execlists: Reset the CSB head tracking on reset/sanitization   Chris Wilson
• [PATCH 10/18] drm/i915/execlists: Pull submit after dequeue under timeline lock   Chris Wilson
• [PATCH 11/18] drm/i915/execlists: Pull CSB reset under the timeline.lock   Chris Wilson
• [PATCH 12/18] drm/i915/execlists: Process one CSB interrupt at a time   Chris Wilson
• [PATCH 13/18] drm/i915/execlists: Unify CSB access pointers   Chris Wilson
• [PATCH 14/18] drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)   Chris Wilson
• [PATCH 15/18] drm/i915: Move rate-limiting request retire to after submission   Chris Wilson
• [PATCH 16/18] drm/i915: Wait for engines to idle before retiring   Chris Wilson
• [PATCH 17/18] drm/i915: Move engine request retirement to intel_engine_cs   Chris Wilson
• [PATCH 18/18] drm/i915: Hold request reference for submission until retirement   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [01/18] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/5] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [01/18] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 1/2] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/2] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 1/5] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/5] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 3/5] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 4/5] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 5/5] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 1/7] drm/i915: Prepare GEM for suspend earlier   Chris Wilson
• [PATCH 2/7] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 3/7] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 4/7] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 5/7] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 6/7] flush-writes-before-suspend   Chris Wilson
• [PATCH 7/7] flush-writes-before-freeze   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915: Prepare GEM for suspend earlier   Patchwork
• [PATCH] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• ✗ Fi.CI.BAT: failure for series starting with [1/7] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.BAT: failure for drm/i915: Switch to kernel context before idling at runtime   Patchwork
• [PATCH 1/2] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 2/2] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/5] drm/i915: Prepare GEM for suspend earlier   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/2] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• [PATCH] drm/i915: Initialize CRC collection   Mika Kahola
• ✓ Fi.CI.BAT: success for drm/i915: Initialize CRC collection   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915: Initialize CRC collection   Patchwork
• [PATCH] mei: discard messages from not connected client during power down.   Martin Peres
• [PATCH] mei: discard messages from not connected client during power down.   Greg KH
• ✓ Fi.CI.BAT: success for mei: discard messages from not connected client during power down.   Patchwork
• [PATCH] mei: discard messages from not connected client during power down.   Martin Peres
• [PATCH 1/5] drm/i915: Remove stale asserts from i915_gem_find_active_request()   Chris Wilson
• [PATCH 2/5] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 3/5] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 4/5] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 5/5] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• ✓ Fi.CI.IGT: success for mei: discard messages from not connected client during power down.   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/5] drm/i915: Remove stale asserts from i915_gem_find_active_request()   Patchwork
• ✓ Fi.CI.BAT: success for drm/i915: Initialize CRC collection   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/5] drm/i915: Remove stale asserts from i915_gem_find_active_request()   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915: Initialize CRC collection   Patchwork
• [HAX] drm/i915/psr: Debug log vbt defaults   Dhinakaran Pandiyan
• ✓ Fi.CI.BAT: success for drm/i915/psr: Debug log vbt defaults   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915/psr: Debug log vbt defaults   Patchwork
• [PATCH v9 01/12] drm/i915/guc: Keep guc submission permanently engaged   Michal Wajdeczko
• [PATCH v9 02/12] drm/i915: Park before resetting the submission backend   Michal Wajdeczko
• [PATCH v9 03/12] drm/i915: Correctly handle error path in i915_gem_init_hw   Michal Wajdeczko
• [PATCH v9 04/12] drm/i915: Move i915_gem_fini to i915_gem.c   Michal Wajdeczko
• [PATCH v9 05/12] drm/i915: Introduce i915_gem_fini_hw for symmetry with i915_gem_init_hw   Michal Wajdeczko
• [PATCH v9 06/12] drm/i915: Add i915_gem_fini_hw to i915_gem_suspend   Michal Wajdeczko
• [PATCH v9 07/12] drm/i915: Add i915_gem_fini_hw to i915_reset   Michal Wajdeczko
• [PATCH v9 08/12] drm/i915/guc: Restore symmetric doorbell cleanup   Michal Wajdeczko
• [PATCH v9 09/12] drm/i915/uc: Fully sanitize uC within intel_uc_fini_hw   Michal Wajdeczko
• [PATCH v9 10/12] drm/i915/uc: Use correct error code for GuC initialization failure   Michal Wajdeczko
• [PATCH v9 11/12] drm/i915/uc: Use helper functions to detect fw load status   Michal Wajdeczko
• [PATCH v9 12/12] HAX: Enable GuC for CI   Michal Wajdeczko
• [PATCH v9 01/11] drm/i915/guc: Keep guc submission permanently engaged   Michal Wajdeczko
• [PATCH v9 02/11] drm/i915: Park before resetting the submission backend   Michal Wajdeczko
• [PATCH v9 03/11] drm/i915: Correctly handle error path in i915_gem_init_hw   Michal Wajdeczko
• [PATCH v9 04/11] drm/i915: Move i915_gem_fini to i915_gem.c   Michal Wajdeczko
• [PATCH v9 05/11] drm/i915: Introduce i915_gem_fini_hw for symmetry with i915_gem_init_hw   Michal Wajdeczko
• [PATCH v9 06/11] drm/i915: Add i915_gem_fini_hw to i915_gem_suspend   Michal Wajdeczko
• [PATCH v9 07/11] drm/i915: Add i915_gem_fini_hw to i915_reset   Michal Wajdeczko
• [PATCH v9 08/11] drm/i915/guc: Restore symmetric doorbell cleanup   Michal Wajdeczko
• [PATCH v9 09/11] drm/i915/uc: Fully sanitize uC within intel_uc_fini_hw   Michal Wajdeczko
• [PATCH v9 10/11] drm/i915/uc: Use correct error code for GuC initialization failure   Michal Wajdeczko
• [PATCH v9 11/11] drm/i915/uc: Use helper functions to detect fw load status   Michal Wajdeczko
• [PATCH v2] drm/i915: Initialize CRC collection   Mika Kahola
• ✓ Fi.CI.BAT: success for series starting with [v9,01/12] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [v9,01/11] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [v9,01/12] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• ✓ Fi.CI.BAT: success for drm/i915: Initialize CRC collection (rev2)   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [v9,01/11] drm/i915/guc: Keep guc submission permanently engaged   Patchwork
• ✓ Fi.CI.IGT: success for drm/i915: Initialize CRC collection (rev2)   Patchwork
• [PATCH] pmu underrun test   Tvrtko Ursulin
• ✓ Fi.CI.BAT: success for pmu underrun test   Patchwork
• ✗ Fi.CI.IGT: failure for pmu underrun test   Patchwork
• [PATCH] pmu underrun test   Tvrtko Ursulin
• [PATCH 1/3] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 2/3] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 3/3] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 1/4] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 2/4] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 3/4] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 4/4] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• ✗ Fi.CI.BAT: failure for pmu underrun test (rev2)   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✓ Fi.CI.BAT: success for series starting with [1/4] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/3] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✓ Fi.CI.IGT: success for series starting with [1/4] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• [PATCH 1/3] drm/i915: Correctly handle error path in i915_gem_init_hw   Michal Wajdeczko
• [PATCH 2/3] drm/i915: Move i915_gem_fini to i915_gem.c   Michal Wajdeczko
• [PATCH 3/3] drm/i915: Introduce i915_gem_fini_hw for symmetry with i915_gem_init_hw   Michal Wajdeczko
• [CI 1/4] drm/i915: Correctly handle error path in i915_gem_init_hw   Michal Wajdeczko
• [CI 2/4] drm/i915: Move i915_gem_fini to i915_gem.c   Michal Wajdeczko
• [CI 3/4] drm/i915: Introduce i915_gem_fini_hw for symmetry with i915_gem_init_hw   Michal Wajdeczko
• [CI 4/4] HAX: Enable GuC for CI   Michal Wajdeczko
• [PATCH 1/6] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 2/6] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 3/6] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 4/6] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 5/6] drm/i915/gtt: Avoid calling non-existent allocate_va_range   Chris Wilson
• [PATCH 6/6] drm/i915/gtt: Enable full-ppgtt by default for HSW   Chris Wilson
• [PATCH] pmu underrun test   Tvrtko Ursulin
• ✓ Fi.CI.BAT: success for series starting with [1/3] drm/i915: Correctly handle error path in i915_gem_init_hw   Patchwork
• [PATCH 1/8] drm/i915: Switch to kernel context before idling at runtime   Chris Wilson
• [PATCH 2/8] drm/i915: "Race-to-idle" after switching to the kernel context   Chris Wilson
• [PATCH 3/8] drm/i915: After reset on sanitization, reset the engine backends   Chris Wilson
• [PATCH 4/8] drm/i915: Only sanitize GEM from late suspend   Chris Wilson
• [PATCH 5/8] flush-writes-before-suspend   Chris Wilson
• [PATCH 6/8] flush-writes-before-freeze   Chris Wilson
• [PATCH 7/8] drm/i915/gtt: Avoid calling non-existent allocate_va_range   Chris Wilson
• [PATCH 8/8] drm/i915/gtt: Enable full-ppgtt by default for HSW   Chris Wilson
• ✓ Fi.CI.BAT: success for series starting with [CI,1/4] drm/i915: Correctly handle error path in i915_gem_init_hw   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/6] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [1/3] drm/i915: Correctly handle error path in i915_gem_init_hw   Patchwork
• ✗ Fi.CI.IGT: failure for series starting with [CI,1/4] drm/i915: Correctly handle error path in i915_gem_init_hw   Patchwork
• ✗ Fi.CI.BAT: failure for pmu underrun test (rev3)   Patchwork
• ✗ Fi.CI.BAT: failure for series starting with [1/8] drm/i915: Switch to kernel context before idling at runtime   Patchwork
• ✓ Fi.CI.BAT: success for pmu underrun test (rev3)   Patchwork
• ✗ Fi.CI.IGT: failure for pmu underrun test (rev3)   Patchwork