summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/intel_lrc.c
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'drm-next-2018-12-14' of git://anongit.freedesktop.org/drm/drmLinus Torvalds2018-12-251-281/+95
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull drm updates from Dave Airlie: "Core: - shared fencing staging removal - drop transactional atomic helpers and move helpers to new location - DP/MST atomic cleanup - Leasing cleanups and drop EXPORT_SYMBOL - Convert drivers to atomic helpers and generic fbdev. - removed deprecated obj_ref/unref in favour of get/put - Improve dumb callback documentation - MODESET_LOCK_BEGIN/END helpers panels: - CDTech panels, Banana Pi Panel, DLC1010GIG, - Olimex LCD-O-LinuXino, Samsung S6D16D0, Truly NT35597 WQXGA, - Himax HX8357D, simulated RTSM AEMv8. - GPD Win2 panel - AUO G101EVN010 vgem: - render node support ttm: - move global init out of drivers - fix LRU handling for ghost objects - Support for simultaneous submissions to multiple engines scheduler: - timeout/fault handling changes to help GPU recovery - helpers for hw with preemption support i915: - Scaler/Watermark fixes - DP MST + powerwell fixes - PSR fixes - Break long get/put shmemfs pages - Icelake fixes - Icelake DSI video mode enablement - Engine workaround improvements amdgpu: - freesync support - GPU reset enabled on CI, VI, SOC15 dGPUs - ABM support in DC - KFD support for vega12/polaris12 - SDMA paging queue on vega - More amdkfd code sharing - DCC scanout on GFX9 - DC kerneldoc - Updated SMU firmware for GFX8 chips - XGMI PSP + hive reset support - GPU reset - DC trace support - Powerplay updates for newer Polaris - Cursor plane update fast path - kfd dma-buf support virtio-gpu: - add EDID support vmwgfx: - pageflip with damage support nouveau: - Initial Turing TU104/TU106 modesetting support msm: - a2xx gpu support for apq8060 and imx5 - a2xx gpummu support - mdp4 display support for apq8060 - DPU fixes and cleanups - enhanced profiling support - debug object naming interface - get_iova/page pinning decoupling tegra: - Tegra194 host1x, VIC and display support enabled - Audio over HDMI for Tegra186 and Tegra194 exynos: - DMA/IOMMU refactoring - plane alpha + blend mode support - Color format fixes for mixer driver rcar-du: - R8A7744 and R8A77470 support - R8A77965 LVDS support imx: - fbdev emulation fix - multi-tiled scalling fixes - SPDX identifiers rockchip - dw_hdmi support - dw-mipi-dsi + dual dsi support - mailbox read size fix qxl: - fix cursor pinning vc4: - YUV support (scaling + cursor) v3d: - enable TFU (Texture Formatting Unit) mali-dp: - add support for linear tiled formats sun4i: - Display Engine 3 support - H6 DE3 mixer 0 support - H6 display engine support - dw-hdmi support - H6 HDMI phy support - implicit fence waiting - BGRX8888 support meson: - Overlay plane support - implicit fence waiting - HDMI 1.4 4k modes bridge: - i2c fixes for sii902x" * tag 'drm-next-2018-12-14' of git://anongit.freedesktop.org/drm/drm: (1403 commits) drm/amd/display: Add fast path for cursor plane updates drm/amdgpu: Enable GPU recovery by default for CI drm/amd/display: Fix duplicating scaling/underscan connector state drm/amd/display: Fix unintialized max_bpc state values Revert "drm/amd/display: Set RMX_ASPECT as default" drm/amdgpu: Fix stub function name drm/msm/dpu: Fix clock issue after bind failure drm/msm/dpu: Clean up dpu_media_info.h static inline functions drm/msm/dpu: Further cleanups for static inline functions drm/msm/dpu: Cleanup the debugfs functions drm/msm/dpu: Remove dpu_irq and unused functions drm/msm: Make irq_postinstall optional drm/msm/dpu: Cleanup callers of dpu_hw_blk_init drm/msm/dpu: Remove unused functions drm/msm/dpu: Remove dpu_crtc_is_enabled() drm/msm/dpu: Remove dpu_crtc_get_mixer_height drm/msm/dpu: Remove dpu_dbg drm/msm: dpu: Remove crtc_lock drm/msm: dpu: Remove vblank_requested flag from dpu_crtc drm/msm: dpu: Separate crtc assignment from vblank enable ...
| * drm/i915: Allocate a common scratch pageChris Wilson2018-12-041-12/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we allocate a scratch page for each engine, but since we only ever write into it for post-sync operations, it is not exposed to userspace nor do we care for coherency. As we then do not care about its contents, we can use one page for all, reducing our allocations and avoid complications by not assuming per-engine isolation. For later use, it simplifies engine initialisation (by removing the allocation that required struct_mutex!) and means that we can always rely on there being a scratch page. v2: Check that we allocated a large enough scratch for I830 w/a Fixes: 06e562e7f515 ("drm/i915/ringbuffer: Delay after EMIT_INVALIDATE for gen4/gen5") # v4.18.20 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108850 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181204141522.13640-1-chris@chris-wilson.co.uk Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: <stable@vger.kernel.org> # v4.18.20+
| * drm/i915: Fuse per-context workaround handling with the common frameworkTvrtko Ursulin2018-12-041-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Convert the per context workaround handling code to run against the newly introduced common workaround framework and fuse the two to use the existing smarter list add helper, the one which does the sorted insert and merges registers where possible. This completes migration of all four classes of workarounds onto the common framework. Existing macros are kept untouched for smaller code churn. v2: * Rename to list name ctx_wa_list and move from dev_priv to engine. v3: * API rename and parameters tweaking. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20181203133357.10341-1-tvrtko.ursulin@linux.intel.com
| * drm/i915: Move register white-listing to the common workaround frameworkTvrtko Ursulin2018-12-041-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of having a separate list of white-listed registers we can trivially move this to the common workarounds framework. This brings us one step closer to the goal of driving all workaround classes using the same code. v2: * Use GEM_DEBUG_WARN_ON for the sanity check. (Chris Wilson) v3: * API rename. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20181203125014.3219-6-tvrtko.ursulin@linux.intel.com
| * drm/i915: Introduce per-engine workaroundsTvrtko Ursulin2018-12-041-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We stopped re-applying the GT workarounds after engine reset since commit 59b449d5c82a ("drm/i915: Split out functions for different kinds of workarounds"). Issue with this is that some of the GT workarounds live in the MMIO space which gets lost during engine resets. So far the registers in 0x2xxx and 0xbxxx address range have been identified to be affected. This losing of applied workarounds has obvious negative effects and can even lead to hard system hangs (see the linked Bugzilla). Rather than just restoring this re-application, because we have also observed that it is not safe to just re-write all GT workarounds after engine resets (GPU might be live and weird hardware states can happen), we introduce a new class of per-engine workarounds and move only the affected GT workarounds over. Using the framework introduced in the previous patch, we therefore after engine reset, re-apply only the workarounds living in the affected MMIO address ranges. v2: * Move Wa_1406609255:icl to engine workarounds as well. * Rename API. (Chris Wilson) * Drop redundant IS_KABYLAKE. (Chris Wilson) * Re-order engine wa/ init so latest platforms are first. (Rodrigo Vivi) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Bugzilla: https://bugzilla.freedesktop.org/show_bug.cgi?id=107945 Fixes: 59b449d5c82a ("drm/i915: Split out functions for different kinds of workarounds") Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: intel-gfx@lists.freedesktop.org Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20181203133341.10258-1-tvrtko.ursulin@linux.intel.com
| * drm/i915: Complete the fences as they are cancelled due to wedgingChris Wilson2018-12-041-2/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We inspect the requests under the assumption that they will be marked as completed when they are removed from the queue. Currently however, in the process of wedging the requests will be removed from the queue before they are completed, so rearrange the code to complete the fences before the locks are dropped. <1>[ 354.473346] BUG: unable to handle kernel NULL pointer dereference at 0000000000000250 <6>[ 354.473363] PGD 0 P4D 0 <4>[ 354.473370] Oops: 0000 [#1] PREEMPT SMP PTI <4>[ 354.473380] CPU: 0 PID: 4470 Comm: gem_eio Tainted: G U 4.20.0-rc4-CI-CI_DRM_5216+ #1 <4>[ 354.473393] Hardware name: Intel Corporation NUC7CJYH/NUC7JYB, BIOS JYGLKCPX.86A.0027.2018.0125.1347 01/25/2018 <4>[ 354.473480] RIP: 0010:__i915_schedule+0x311/0x5e0 [i915] <4>[ 354.473490] Code: 49 89 44 24 20 4d 89 4c 24 28 4d 89 29 44 39 b3 a0 04 00 00 7d 3a 41 8b 44 24 78 85 c0 74 13 48 8b 93 78 04 00 00 48 83 e2 fc <39> 82 50 02 00 00 79 1e 44 89 b3 a0 04 00 00 48 8d bb d0 03 00 00 <4>[ 354.473515] RSP: 0018:ffffc900001bba90 EFLAGS: 00010046 <4>[ 354.473524] RAX: 0000000000000003 RBX: ffff8882624c8008 RCX: f34a737800000000 <4>[ 354.473535] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff8882624c8048 <4>[ 354.473545] RBP: ffffc900001bbab0 R08: 000000005963f1f1 R09: 0000000000000000 <4>[ 354.473556] R10: ffffc900001bba10 R11: ffff8882624c8060 R12: ffff88824fdd7b98 <4>[ 354.473567] R13: ffff88824fdd7bb8 R14: 0000000000000001 R15: ffff88824fdd7750 <4>[ 354.473578] FS: 00007f44b4b5b980(0000) GS:ffff888277e00000(0000) knlGS:0000000000000000 <4>[ 354.473590] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 <4>[ 354.473599] CR2: 0000000000000250 CR3: 000000026976e000 CR4: 0000000000340ef0 <4>[ 354.473611] Call Trace: <4>[ 354.473622] ? lock_acquire+0xa6/0x1c0 <4>[ 354.473677] ? i915_schedule_bump_priority+0x57/0xd0 [i915] <4>[ 354.473736] i915_schedule_bump_priority+0x72/0xd0 [i915] <4>[ 354.473792] i915_request_wait+0x4db/0x840 [i915] <4>[ 354.473804] ? get_pwq.isra.4+0x2c/0x50 <4>[ 354.473813] ? ___preempt_schedule+0x16/0x18 <4>[ 354.473824] ? wake_up_q+0x70/0x70 <4>[ 354.473831] ? wake_up_q+0x70/0x70 <4>[ 354.473882] ? gen6_rps_boost+0x118/0x120 [i915] <4>[ 354.473936] i915_gem_object_wait_fence+0x8a/0x110 [i915] <4>[ 354.473991] i915_gem_object_wait+0x113/0x500 [i915] <4>[ 354.474047] i915_gem_wait_ioctl+0x11c/0x2f0 [i915] <4>[ 354.474101] ? i915_gem_unset_wedged+0x210/0x210 [i915] <4>[ 354.474113] drm_ioctl_kernel+0x81/0xf0 <4>[ 354.474123] drm_ioctl+0x2de/0x390 <4>[ 354.474175] ? i915_gem_unset_wedged+0x210/0x210 [i915] <4>[ 354.474187] ? finish_task_switch+0x95/0x260 <4>[ 354.474197] ? lock_acquire+0xa6/0x1c0 <4>[ 354.474207] do_vfs_ioctl+0xa0/0x6e0 <4>[ 354.474217] ? __fget+0xfc/0x1e0 <4>[ 354.474225] ksys_ioctl+0x35/0x60 <4>[ 354.474233] __x64_sys_ioctl+0x11/0x20 <4>[ 354.474241] do_syscall_64+0x55/0x190 <4>[ 354.474251] entry_SYSCALL_64_after_hwframe+0x49/0xbe <4>[ 354.474260] RIP: 0033:0x7f44b3de65d7 <4>[ 354.474267] Code: b3 66 90 48 8b 05 b1 48 2d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 48 2d 00 f7 d8 64 89 01 48 <4>[ 354.474293] RSP: 002b:00007fff974948e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 <4>[ 354.474305] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f44b3de65d7 <4>[ 354.474316] RDX: 00007fff97494940 RSI: 00000000c010646c RDI: 0000000000000007 <4>[ 354.474327] RBP: 00007fff97494940 R08: 0000000000000000 R09: 00007f44b40bbc40 <4>[ 354.474337] R10: 0000000000000000 R11: 0000000000000246 R12: 00000000c010646c <4>[ 354.474348] R13: 0000000000000007 R14: 0000000000000000 R15: 0000000000000000 v2: Avoid floating requests. v3: Can't call dma_fence_signal() under the timeline lock! v4: Can't call dma_fence_signal() from inside another fence either. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181203113701.12106-2-chris@chris-wilson.co.uk
| * drm/i915/vgpu: Disallow loading on old vGPU hostsChris Wilson2018-12-031-24/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since commit fd8526e50902 ("drm/i915/execlists: Trust the CSB") we actually broke the force-mmio mode for our execlists implementation. No one noticed, so ergo no one is actually using an old vGPU host (where we required the older method) and so can simply remove the broken support. v2: csb_read can go as well (Mika) Reported-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Fixes: fd8526e50902 ("drm/i915/execlists: Trust the CSB") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181130125954.11924-1-chris@chris-wilson.co.uk
| * drm/i915: Mark up early pre-production KabylakesChris Wilson2018-11-281-12/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Mark A0 as the one and only pre-production variant of Kabylake and remove its couple of workarounds, consigning them to the annals of history. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181128135325.10641-1-chris@chris-wilson.co.uk
| * drm/i915: Set MI_INVALIDATE_BSD for all video-decode enginesChris Wilson2018-11-081-1/+1
| | | | | | | | | | | | | | | | | | | | | | We have multiple instances of VCS but only remember to invalidate the BSD caches on the first, ignoring the stale caches of any other engine. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181108140039.12254-1-chris@chris-wilson.co.uk
| * drm/i915/execlists: Force write serialisation into context image vs executionChris Wilson2018-11-081-1/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Ensure that the writes into the context image are completed prior to the register mmio to trigger execution. Although previously we were assured by the SDM that all writes are flushed before an uncached memory transaction (our mmio write to submit the context to HW for execution), we have empirical evidence to believe that this is not actually the case. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108656 References: https://bugs.freedesktop.org/show_bug.cgi?id=108315 References: https://bugs.freedesktop.org/show_bug.cgi?id=106887 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181108081740.25615-1-chris@chris-wilson.co.uk Cc: stable@vger.kernel.org
| * drm/i915: GEM_WARN_ON considered harmfulTvrtko Ursulin2018-10-181-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GEM_WARN_ON currently has dangerous semantics where it is completely compiled out on !GEM_DEBUG builds. This can leave users who expect it to be more like a WARN_ON, just without a warning in non-debug builds, in complete ignorance. Another gotcha with it is that it cannot be used as a statement. Which is again different from a standard kernel WARN_ON. This patch fixes both problems by making it behave as one would expect. It can now be used both as an expression and as statement, and also the condition evaluates properly in all builds - code under the conditional will therefore not unexpectedly disappear. To satisfy call sites which really want the code under the conditional to completely disappear, we add GEM_DEBUG_WARN_ON and convert some of the callers to it. This one can also be used as both expression and statement. >From the above it follows GEM_DEBUG_WARN_ON should be used in situations where we are certain the condition will be hit during development, but at a place in code where error can be handled to the benefit of not crashing the machine. GEM_WARN_ON on the other hand should be used where condition may happen in production and we just want to distinguish the level of debugging output emitted between the production and debug build. v2: * Dropped BUG_ON hunk. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.auld@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Tomasz Lis <tomasz.lis@intel.com> Reviewed-by: Tomasz Lis <tomasz.lis@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181012063142.16080-1-tvrtko.ursulin@linux.intel.com
| * drm/i915/execlists: Flush the CS events before unpinningChris Wilson2018-10-031-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Inside the execlists submission tasklet, we often make the mistake of assuming that everything beneath the request is available for use. However, the submission and the request live on two separate timelines, and the request contents may be freed from an early retirement before we have had a chance to run the submission tasklet (think ksoftirqd). To safeguard ourselves against any mistakes, flush the tasklet before we unpin the context if execlists still has a reference to this context. v2: Pull hw_context->active tracking into schedule_in and schedule_out. References: 60367132a214 ("drm/i915: Avoid use-after-free of ctx in request tracepoints") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181003110941.27886-1-chris@chris-wilson.co.uk
| * drm/i915: Pull scheduling under standalone lockChris Wilson2018-10-011-263/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, the backend scheduling code abuses struct_mutex into order to have a global lock to manipulate a temporary list (without widespread allocation) and to protect against list modifications. This is an extraneous coupling to struct_mutex and further can not extend beyond the local device. Pull all the code that needs to be under the one true lock into i915_scheduler.c, and make it so. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181001144755.7978-2-chris@chris-wilson.co.uk
| * drm/i915: Priority boost for new clientsChris Wilson2018-10-011-6/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Taken from an idea used for FQ_CODEL, we give the first request of a new request flows a small priority boost. These flows are likely to correspond with short, interactive tasks and so be more latency sensitive than the longer free running queues. As soon as the client has more than one request in the queue, further requests are not boosted and it settles down into ordinary steady state behaviour. Such small kicks dramatically help combat the starvation issue, by allowing each client the opportunity to run even when the system is under heavy throughput load (within the constraints of the user selected priority). v2: Mark the preempted request as the start of a new flow, to prevent a single client being continually gazumped by its peers. Testcase: igt/benchmarks/rrul Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181001144755.7978-1-chris@chris-wilson.co.uk
| * drm/i915: Combine multiple internal plists into the same i915_priolist bucketChris Wilson2018-10-011-27/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As we are about to allow ourselves to slightly bump the user priority into a few different sublevels, packthose internal priority lists into the same i915_priolist to keep the rbtree compact and avoid having to allocate the default user priority even after the internal bumping. The downside to having an requests[] rather than a node per active list, is that we then have to walk over the empty higher priority lists. To compensate, we track the active buckets and use a small bitmap to skip over any inactive ones. v2: Use MASK of internal levels to simplify our usage. v3: Prevent overflow when SHIFT is zero. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181001123204.23982-4-chris@chris-wilson.co.uk
| * drm/i915/execlists: Avoid kicking priority on the current contextChris Wilson2018-09-271-11/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the request is currently on the HW (in port 0), then we do not need to kick the submission tasklet to evaluate whether we should be preempting itself in order to execute it again. In the case that was annoying me: execlists_schedule: rq(18:211173).prio=0 -> 2 need_preempt: last(18:211174).prio=0, queue.prio=2 We are bumping the priority of the first of a pair of requests running in the current context. Then when evaluating preempt, we would see that that our priority request is higher than the last executing request in ELSP0 and so trigger preemption, not realising that our intended request was already executing. v2: As we assume state of the execlists->port[] that is only valid while we hold the timeline lock we have to repeat some earlier tests that on the validity of the node. v3: Wrap guc submission under the timeline.lock as is now the way of all things. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180925083205.2229-2-chris@chris-wilson.co.uk
| * drm/i915: Remove i915.enable_ppgtt overrideChris Wilson2018-09-271-8/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that we are confident in providing full-ppgtt where supported, remove the ability to override the context isolation. v2: Remove faked aliasing-ppgtt for testing as it no longer is accepted. v3: s/USES/HAS/ to match usage and reject attempts to load the module on old GVT-g setups that do not provide support for full-ppgtt. v4: Insulate ABI ppGTT values from our internal enum (later plans involve moving ppGTT depth out of the enum, thus potentially breaking ABI unless we document the current values). Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Acked-by: Zhi Wang <zhi.a.wang@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180926201222.5643-1-chris@chris-wilson.co.uk
| * drm/i915: Log HWS seqno consistentlyTvrtko Ursulin2018-09-271-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | We mix hexa- and decimal which is confusing when reading the logs. So make the single odd one out instance decimal for consistency. v2: * Do the intel_ringbuffer.c as well. (Chris Wilson) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180926145033.16318-1-tvrtko.ursulin@linux.intel.com
| * drm/i915: Match code to comment and enforce ppgtt for execlistsChris Wilson2018-09-251-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | Our execlist dispatch code requires a ppGTT so make sure we enforce that option in intel_sanitize_enable_ppgtt(). The comment already tries to explain that execlists requires ppgtt, but was written when gen8 may have also taken the legacy path; so rewrite the code to match the comment by using HAS_EXECLISTS() feature instead of the gen. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180922141804.21183-1-chris@chris-wilson.co.uk
| * drm/i915/execlists: Assert the queue is non-empty on unsubmittingChris Wilson2018-09-251-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the sequence <0>[ 531.960431] drv_self-4806 7.... 527402570us : intel_gpu_reset: engine_mask=1, ret=0, retry=0 <0>[ 531.960431] drv_self-4806 7.... 527402571us : execlists_reset: rcs0 request global=115de, current=71133 <0>[ 531.960431] drv_self-4806 7d..1 527402571us : execlists_cancel_port_requests: rcs0:port0 global=71134 (fence 826b:198), (current 71133) <0>[ 531.960431] drv_self-4806 7d..1 527402572us : execlists_cancel_port_requests: rcs0:port1 global=71135 (fence 826c:53), (current 71133) <0>[ 531.960431] drv_self-4806 7d..1 527402572us : __i915_request_unsubmit: rcs0 fence 826c:53 <- global=71135, current 71133 <0>[ 531.960431] drv_self-4806 7d..1 527402579us : __i915_request_unsubmit: rcs0 fence 826b:198 <- global=71134, current 71133 <0>[ 531.960431] drv_self-4806 7.... 527402613us : intel_engine_cancel_stop_cs: rcs0 <0>[ 531.960431] drv_self-4806 7.... 527402624us : execlists_reset_finish: rcs0 we are missing the execlists_submission_tasklet() invocation before the execlists_reset_fini() implying that either the queue is empty, or we failed to schedule and run the tasklet on finish. Add an assert so we are sure that on unsubmitting the incomplete request after reset, the queue is indeed populated. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180919195544.1511-12-chris@chris-wilson.co.uk
* | drm/i915: Allocate a common scratch pageChris Wilson2018-12-121-13/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we allocate a scratch page for each engine, but since we only ever write into it for post-sync operations, it is not exposed to userspace nor do we care for coherency. As we then do not care about its contents, we can use one page for all, reducing our allocations and avoid complications by not assuming per-engine isolation. For later use, it simplifies engine initialisation (by removing the allocation that required struct_mutex!) and means that we can always rely on there being a scratch page. v2: Check that we allocated a large enough scratch for I830 w/a Fixes: 06e562e7f515 ("drm/i915/ringbuffer: Delay after EMIT_INVALIDATE for gen4/gen5") # v4.18.20 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108850 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181204141522.13640-1-chris@chris-wilson.co.uk Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: <stable@vger.kernel.org> # v4.18.20+ (cherry picked from commit 5179749925933575a67f9d8f16d0cc204f98a29f) [Joonas: Use new function in gen9_init_indirectctx_bb too] Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
* | drm/i915/execlists: Apply a full mb before execution for BraswellChris Wilson2018-12-121-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Braswell is really picky about having our writes posted to memory before we execute or else the GPU may see stale values. A wmb() is insufficient as it only ensures the writes are visible to other cores, we need a full mb() to ensure the writes are in memory and visible to the GPU. The most frequent failure in flushing before execution is that we see stale PTE values and execute the wrong pages. References: 987abd5c62f9 ("drm/i915/execlists: Force write serialisation into context image vs execution") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: stable@vger.kernel.org Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181206084431.9805-3-chris@chris-wilson.co.uk (cherry picked from commit 490b8c65b9db45896769e1095e78725775f47b3e) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
* | drm/i915: Introduce per-engine workaroundsTvrtko Ursulin2018-12-051-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We stopped re-applying the GT workarounds after engine reset since commit 59b449d5c82a ("drm/i915: Split out functions for different kinds of workarounds"). Issue with this is that some of the GT workarounds live in the MMIO space which gets lost during engine resets. So far the registers in 0x2xxx and 0xbxxx address range have been identified to be affected. This losing of applied workarounds has obvious negative effects and can even lead to hard system hangs (see the linked Bugzilla). Rather than just restoring this re-application, because we have also observed that it is not safe to just re-write all GT workarounds after engine resets (GPU might be live and weird hardware states can happen), we introduce a new class of per-engine workarounds and move only the affected GT workarounds over. Using the framework introduced in the previous patch, we therefore after engine reset, re-apply only the workarounds living in the affected MMIO address ranges. v2: * Move Wa_1406609255:icl to engine workarounds as well. * Rename API. (Chris Wilson) * Drop redundant IS_KABYLAKE. (Chris Wilson) * Re-order engine wa/ init so latest platforms are first. (Rodrigo Vivi) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Bugzilla: https://bugzilla.freedesktop.org/show_bug.cgi?id=107945 Fixes: 59b449d5c82a ("drm/i915: Split out functions for different kinds of workarounds") Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: intel-gfx@lists.freedesktop.org Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20181203133341.10258-1-tvrtko.ursulin@linux.intel.com (cherry picked from commit 4a15c75c42460252a63d30f03b4766a52945fb47) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
* | drm/i915/execlists: Force write serialisation into context image vs executionChris Wilson2018-11-121-1/+13
|/ | | | | | | | | | | | | | | | | | | | | Ensure that the writes into the context image are completed prior to the register mmio to trigger execution. Although previously we were assured by the SDM that all writes are flushed before an uncached memory transaction (our mmio write to submit the context to HW for execution), we have empirical evidence to believe that this is not actually the case. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108656 References: https://bugs.freedesktop.org/show_bug.cgi?id=108315 References: https://bugs.freedesktop.org/show_bug.cgi?id=106887 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181108081740.25615-1-chris@chris-wilson.co.uk Cc: stable@vger.kernel.org (cherry picked from commit 987abd5c62f92ee4970b45aa077f47949974e615) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
* drm/i915/execlists: Onion unwind for logical_ring_init() failureChris Wilson2018-09-201-7/+11
| | | | | | | | | | | Fix up the error unwind for logical_ring_init() failing by moving the cleanup into the callers who own the various bits of state during initialisation, so we don't forget to free the state allocated by the caller. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180920195948.16448-1-chris@chris-wilson.co.uk
* drm/i915/execlists: Use coherent writes into the context imageChris Wilson2018-09-141-2/+4
| | | | | | | | | | | | | | | | | | That we use a WB mapping for updating the RING_TAIL register inside the context image even on !llc machines has been a source of consternation for every reader. It appears to work on bsw+, but it may just have been that we have been incredibly bad at detecting the errors. v2: With extra enthusiasm. v3: Drop force of map type for pinned default_state as by the time we pin it, the map type is always WB and doesn't conflict with the earlier use by ce->state. v4: Transfer engine->default_state from MAP_WC to MAP_WB on creation so we do not need the MAP_FORCE littered around the backends Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180914123504.2062-3-chris@chris-wilson.co.uk
* drm/i915/execlists: Delay updating ring register state after resumeChris Wilson2018-09-141-18/+15
| | | | | | | | | | | | | | | | Now that we reload both RING_HEAD and RING_TAIL when rebinding the context, we do not need to scrub those registers immediately on resume. v2: Handle the perma-pinned contexts. v3: Set RING_TAIL on context-pin so that we always have known state in the context image for the ring registers and all parties have similar code (ripe for refactoring). Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180914123504.2062-1-chris@chris-wilson.co.uk
* drm/i915: Reduce context HW ID lifetimeChris Wilson2018-09-051-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Future gen reduce the number of bits we will have available to differentiate between contexts, so reduce the lifetime of the ID assignment from that of the context to its current active cycle (i.e. only while it is pinned for use by the HW, will it have a constant ID). This means that instead of a max of 2k allocated contexts (worst case before fun with bit twiddling), we instead have a limit of 2k in flight contexts (minus a few that have been pinned by the kernel or by perf). To reduce the number of contexts id we require, we allocate a context id on first and mark it as pinned for as long as the GEM context itself is, that is we keep it pinned it while active on each engine. If we exhaust our context id space, then we try to reclaim an id from an idle context. In the extreme case where all context ids are pinned by active contexts, we force the system to idle in order to recover ids. We cannot reduce the scope of an HW-ID to an engine (allowing the same gem_context to have different ids on each engine) as in the future we will need to preassign an id before we know which engine the context is being executed on. v2: Improved commentary (Tvrtko) [I tried at least] References: https://bugs.freedesktop.org/show_bug.cgi?id=107788 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Michel Thierry <michel.thierry@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180904153117.3907-1-chris@chris-wilson.co.uk
* drm/i915/icl: Fix context RPCS programmingTvrtko Ursulin2018-09-041-13/+74
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are two issues with the current RPCS programming for Icelake: Expansion of the slice count bitfield has been missed, as well as the required programming workaround for the subslice count bitfield size limitation. 1) Bitfield width for configuring the active slice count has grown so we need to program the GEN8_R_PWR_CLK_STATE accordingly. Current code was always requesting eight times the number of slices (due writing to a bitfield starting three bits higher than it should). These requests were luckily a) capped by the hardware to the available number of slices, and b) we haven't yet exported the code to ask for reduced slice configurations. Due both of the above there was no impact from this incorrect programming but we should still fix it. 2) Due subslice count bitfield being only three bits wide and furthermore capped to a maximum documented value of four, special programming workaround is needed to enable more than four subslices. With this programming driver has to consider the GT configuration as 2x4x8, while the hardware internally translates this to 1x8x8. A limitation stemming from this is that either a subslice count between one and four can be selected, or a subslice count equaling the total number of subslices in all selected slices. In other words, odd subslice counts greater than four are impossible, as are odd subslice counts greater than a single slice subslice count. This also had no impact in the current code base due breakage from 1) always reqesting more than one slice. While fixing this we also add some asserts to flag up any future bitfield overflows. v2: * Use a local in all branches for clarity. (Lionel) Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Bspec: 12247 Reported-by: tony.ye@intel.com Suggested-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Cc: tony.ye@intel.com Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180903113007.2643-1-tvrtko.ursulin@linux.intel.com
* drm/i915/execlists: Flush tasklet directly from reset-finishChris Wilson2018-08-291-11/+6
| | | | | | | | | | | | | | | | | | | | | | | | On finishing the reset, the intention is to restart the GPU before we relinquish the forcewake taken to handle the reset - the goal being the GPU reloads a context before it is allowed to sleep. For this purpose, we used tasklet_flush() which although it accomplished the goal of restarting the GPU, carried with it a sting in its tail: it cleared the TASKLET_STATE_SCHED bit. This meant that if another CPU queued a new request to this engine, we would clear the flag and later attempt to requeue the tasklet on the local CPU, breaking the per-cpu softirq lists. Remove the dangerous tasklet_kill() and just run the tasklet func directly as we know it is safe to do so (the tasklets are internally locked to allow mixed usage from direct submission). Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Michel Thierry <michel.thierry@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180828152702.27536-1-chris@chris-wilson.co.uk
* drm/i915/execlists: Include reset depth in tracesChris Wilson2018-08-161-2/+4
| | | | | | | | | Show the reset depth (the tasklet disable count) in the GEM_TRACE to indicate when we might not expect tasklets to be flushed. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180815135827.25869-1-chris@chris-wilson.co.uk
* drm/i915: Remove useless error return from intel_init_mocs_engine()Chris Wilson2018-08-151-5/+1
| | | | | | | | | | As the only error is for a programming error in constructing the static tables describing the register values, replace the error code propagation with an assert. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180815184251.5850-1-chris@chris-wilson.co.uk
* drm/i915/icl: account for context save/restore removed bitsPaulo Zanoni2018-08-101-8/+6
| | | | | | | | | | | | | | | | | | | | | | The RS_CTX_ENABLE and CTX_SAVE_INHIBIT bits are not present on ICL anymore, but we still try to set them and then check them with GEM_BUG_ON, resulting in a BUG() call. The bug can be reproduced by igt/drv_selftest/live_hangcheck/others-priority and our CI was able to catch it. It is worth noticing that commit 05f0addd9b10 ("drm/i915/icl: Enhanced execution list support") already tried to avoid the save bits on ICL, but only inside populate_lr_context(). Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Testcase: igt/drv_selftest/live_hangcheck/others-priority Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107399 References: 05f0addd9b10 ("drm/i915/icl: Enhanced execution list support") Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809235852.24516-1-paulo.r.zanoni@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
* drm/i915: kill resource streamer supportLucas De Marchi2018-08-061-6/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use RS years ago, but nothing ever landed or shipped because there was no performance improvement. This removes it from kernel keeping the uapi defines around for compatibility. v2: - re-add the inadvertent removal of CTX_CTRL_INHIBIT_SYN_CTX_SWITCH - don't bother trying to document removed params on uapi header: applications should know that from the query. (from Chris) v3: - disable CTX_CTRL_RS_CTX_ENABLE istead of removing it - reword commit message after Daniele confirmed no performance regression on his machine - reword commit message to make clear RS is being removed due to never been used v4: - move I915_EXEC_RESOURCE_STREAMER to __I915_EXEC_ILLEGAL_FLAGS so the check on ioctl() is made much earlier by i915_gem_check_execbuffer() (suggested by Tvrtko) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Acked-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180803232443.17193-1-lucas.demarchi@intel.com
* drm/i915/execlists: Terminate the context image with BB_ENDChris Wilson2018-08-011-0/+4
| | | | | | | | | | | | | In the aub trace utility, the context images are terminated with a MI_BATCH_BUFFER_END; the simulator is reported as complaining otherwise. Do the same for our protocontext image for completeness, and in passing apply the magic bit for gen10 to mark the end of the context image. Reported-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180730164325.12770-1-chris@chris-wilson.co.uk
* drm/i915: Drop unneed i915 parameter from intel_ring_pin()Chris Wilson2018-07-271-1/+1
| | | | | | | | | As we now have a ring->vma available, we can just lookup our i915 pointer from inside the vm, and so not require the unsightly parameter. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180727155501.18963-1-chris@chris-wilson.co.uk
* drm/i915: Remove unnecessary ggtt_offset_bias from i915_gem_contextJakub Bartmiński2018-07-271-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | | Since ggtt_offset_bias is now stored in ggtt.pin_bias, it is duplicated inside i915_gem_context, and can instead be accessed directly from ggtt. v3: Added a helper function to retrieve the ggtt.pin_bias from the vma. v4: Moved the helper function to the previous patch in the series. Dropped the bias from intel_ring_pin. This introduces a slight functional change since we are always pinning the ring a bit higher if GuC is present even though we don't really need to. v8: Fixed patch not applying on the most recent upstream. Signed-off-by: Jakub Bartmiński <jakub.bartminski@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Link: https://patchwork.freedesktop.org/patch/msgid/20180727141148.30874-4-jakub.bartminski@intel.com
* drm/i915: Remove superfluous GEN8_LR_CONTEXT_ALIGNChris Wilson2018-07-271-1/+1
| | | | | | | | | | As GEN8_LR_CONTEXT_ALIGN is I915_GTT_MIN_ALIGNMENT is it functionally equivalent to 0, and we will not be able to reduce the min-alignment for the GTT, so passing 0 is and will remain equivalent. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180727092947.1953-1-chris@chris-wilson.co.uk
* drm/i915: Eliminate use of PAGE_SIZE as a virtual alignmentChris Wilson2018-07-271-1/+1
| | | | | | | | | | | | Using PAGE_SIZE for virtual offset alignment is superfluous as it is equal to the minimum gtt alignment and so equivalent to 0. It is also the wrong value to use as we stopped using physical page constructs for the virtual GTT, i.e. it would be preferrable to use I915_GTT_PAGE_SIZE and in these cases merely imply I915_GTT_MIN_ALIGNMENT. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180727091855.1879-1-chris@chris-wilson.co.uk
* drm/i915: Pull unpin map into vma releaseChris Wilson2018-07-241-1/+1
| | | | | | | | | | | | | | | | | | A reasonably common operation is to pin the map of the vma alongside the vma itself for the lifetime of the vma, and so release both pins at the same time as destroying the vma. It is common enough to pull into the release function, making that central function more attractive to a couple of other callsites. The continual ulterior motive is to sweep over errors on module load aborting... Testcase: igt/drv_module_reload/basic-reload-inject Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180721125037.20127-1-chris@chris-wilson.co.uk
* drm/i915/execlists: Move the assertion we have the rpm wakeref downChris Wilson2018-07-191-14/+11
| | | | | | | | | | | | | | | | | | There's a race between idling the engine and finishing off the last tasklet (as we may kick the tasklets after declaring an individual engine idle). However, since we do not need to access the device until we try to submit to the ELSP register (processing the CSB just requires normal CPU access to the HWSP, and when idle we should not need to submit!) we can defer the assertion unto that point. The assertion is still useful as it does verify that we do hold the longterm GT wakeref taken from request allocation until request completion. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107274 Fixes: 9512f985c32d ("drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180719075029.28643-1-chris@chris-wilson.co.uk
* drm/i915/guc: Keep guc submission permanently engagedChris Wilson2018-07-191-2/+2
| | | | | | | | | | | | | | | | | | | | | | | We make a decision at module load whether to use the GuC backend or not, but lose that setup across set-wedge. Currently, the guc doesn't override the engine->set_default_submission hook letting execlists sneak back in temporarily on unwedging leading to an unbalanced park/unpark. v2: Remove comment about switching back temporarily to execlists on guc_submission_disable(). We currently only call disable on shutdown, and plan to also call disable before suspend and reset, in which case we will either restore guc submission or mark the driver as wedged, making the reset back to execlists pointless. v3: Move reset.prepare across Fixes: 63572937cebf ("drm/i915/execlists: Flush pending preemption events during reset") Testcase: igt/drv_module_reload/basic-reload-inject Testcase: igt/gem_eio Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michał Winiarski <michal.winiarski@intel.com> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180717202932.1423-1-chris@chris-wilson.co.uk
* drm/i915/selftests: Force a preemption hangChris Wilson2018-07-161-0/+3
| | | | | | | | | | | | | Inject a failure into preemption completion to pretend as if the HW didn't successfully handle preemption and we are forced to do a reset in the middle. v2: Wait for preemption, to force testing with the missed preemption. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180716132154.12539-1-chris@chris-wilson.co.uk
* drm/i915/execlists: Always clear preempt status on cancelling allChris Wilson2018-07-161-4/+2
| | | | | | | | | | | | | | | | On reset/wedging, we cancel all pending replies from the HW and we also want to cancel an outstanding preemption event. Since we use the same function to cancel the pending replies for reset and for a preemption event, we can simply clear the active tracking for all. v2: Keep execlists_user_end() markup for wedging v3: Move assignment to inline to hide the bare assignment. Fixes: 60a943245413 ("drm/i915/execlists: Drop clear_gtiir() on GPU reset") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180716125424.5715-1-chris@chris-wilson.co.uk
* drm/i915/execlists: Disable submission tasklet upon wedgingChris Wilson2018-07-161-0/+8
| | | | | | | | | | | | | | | | | If we declare the driver wedged before the GPU truly is, then we may see the GPU complete some CS events following our cancellation. This leaves us quite confused as we deleted all the bookkeeping and thus complain about the inconsistent state. We can just ignore the remaining events and let the GPU idle by not feeding it, and so avoid trying to racily overwrite shared state. We rely on there being a full GPU reset before unwedging, giving us the opportunity to reset the shared state. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107188 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180716080332.32283-4-chris@chris-wilson.co.uk
* drm/i915/execlists: Drop clear_gtiir() on GPU resetChris Wilson2018-07-131-68/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the new CSB processing code, we are not vulnerable to delayed delivery of a pre-reset interrupt as we use the CSB status pointers in the HWSP to decide if we need to parse any CSB events and no longer need to wait for the first post-reset interrupt to be assured that the CSB mmio registers are valid. The new icl code to clear registers has a nasty lock inversion: [ 57.409776] ====================================================== [ 57.409779] WARNING: possible circular locking dependency detected [ 57.409783] 4.18.0-rc4-CI-CI_DII_1137+ #1 Tainted: G U W [ 57.409785] ------------------------------------------------------ [ 57.409788] swapper/6/0 is trying to acquire lock: [ 57.409790] 000000004f304ee5 (&engine->timeline.lock/1){-.-.}, at: execlists_submit_request+0x2b/0x1a0 [i915] [ 57.409841] but task is already holding lock: [ 57.409844] 00000000aad89594 (&(&rq->lock)->rlock#2){-.-.}, at: notify_ring+0x2b2/0x480 [i915] [ 57.409869] which lock already depends on the new lock. [ 57.409872] the existing dependency chain (in reverse order) is: [ 57.409876] -> #2 (&(&rq->lock)->rlock#2){-.-.}: [ 57.409900] notify_ring+0x2b2/0x480 [i915] [ 57.409922] gen8_cs_irq_handler+0x39/0xa0 [i915] [ 57.409943] gen11_irq_handler+0x2f0/0x420 [i915] [ 57.409949] __handle_irq_event_percpu+0x42/0x370 [ 57.409952] handle_irq_event_percpu+0x2b/0x70 [ 57.409956] handle_irq_event+0x2f/0x50 [ 57.409959] handle_edge_irq+0xe7/0x190 [ 57.409964] handle_irq+0x67/0x160 [ 57.409967] do_IRQ+0x5e/0x120 [ 57.409971] ret_from_intr+0x0/0x1d [ 57.409974] _raw_spin_unlock_irqrestore+0x4e/0x60 [ 57.409979] tasklet_action_common.isra.5+0x47/0xb0 [ 57.409982] __do_softirq+0xd9/0x505 [ 57.409985] irq_exit+0xa9/0xc0 [ 57.409988] do_IRQ+0x9a/0x120 [ 57.409991] ret_from_intr+0x0/0x1d [ 57.409995] cpuidle_enter_state+0xac/0x360 [ 57.409999] do_idle+0x1f3/0x250 [ 57.410004] cpu_startup_entry+0x6a/0x70 [ 57.410010] start_secondary+0x19d/0x1f0 [ 57.410015] secondary_startup_64+0xa5/0xb0 [ 57.410018] -> #1 (&(&dev_priv->irq_lock)->rlock){-.-.}: [ 57.410081] clear_gtiir+0x30/0x200 [i915] [ 57.410116] execlists_reset+0x6e/0x2b0 [i915] [ 57.410140] i915_reset_engine+0x111/0x190 [i915] [ 57.410165] i915_handle_error+0x11a/0x4a0 [i915] [ 57.410198] i915_hangcheck_elapsed+0x378/0x530 [i915] [ 57.410204] process_one_work+0x248/0x6c0 [ 57.410207] worker_thread+0x37/0x380 [ 57.410211] kthread+0x119/0x130 [ 57.410215] ret_from_fork+0x3a/0x50 [ 57.410217] -> #0 (&engine->timeline.lock/1){-.-.}: [ 57.410224] _raw_spin_lock_irqsave+0x33/0x50 [ 57.410256] execlists_submit_request+0x2b/0x1a0 [i915] [ 57.410289] submit_notify+0x8d/0x124 [i915] [ 57.410314] __i915_sw_fence_complete+0x81/0x250 [i915] [ 57.410339] dma_i915_sw_fence_wake+0xd/0x20 [i915] [ 57.410344] dma_fence_signal_locked+0x79/0x200 [ 57.410368] notify_ring+0x2ba/0x480 [i915] [ 57.410392] gen8_cs_irq_handler+0x39/0xa0 [i915] [ 57.410416] gen11_irq_handler+0x2f0/0x420 [i915] [ 57.410421] __handle_irq_event_percpu+0x42/0x370 [ 57.410425] handle_irq_event_percpu+0x2b/0x70 [ 57.410428] handle_irq_event+0x2f/0x50 [ 57.410432] handle_edge_irq+0xe7/0x190 [ 57.410436] handle_irq+0x67/0x160 [ 57.410439] do_IRQ+0x5e/0x120 [ 57.410445] ret_from_intr+0x0/0x1d [ 57.410449] cpuidle_enter_state+0xac/0x360 [ 57.410453] do_idle+0x1f3/0x250 [ 57.410456] cpu_startup_entry+0x6a/0x70 [ 57.410460] start_secondary+0x19d/0x1f0 [ 57.410464] secondary_startup_64+0xa5/0xb0 [ 57.410466] other info that might help us debug this: [ 57.410471] Chain exists of: &engine->timeline.lock/1 --> &(&dev_priv->irq_lock)->rlock --> &(&rq->lock)->rlock#2 [ 57.410481] Possible unsafe locking scenario: [ 57.410485] CPU0 CPU1 [ 57.410487] ---- ---- [ 57.410490] lock(&(&rq->lock)->rlock#2); [ 57.410494] lock(&(&dev_priv->irq_lock)->rlock); [ 57.410498] lock(&(&rq->lock)->rlock#2); [ 57.410503] lock(&engine->timeline.lock/1); [ 57.410506] *** DEADLOCK *** [ 57.410511] 4 locks held by swapper/6/0: [ 57.410514] #0: 0000000074575789 (&(&dev_priv->irq_lock)->rlock){-.-.}, at: gen11_irq_handler+0x8a/0x420 [i915] [ 57.410542] #1: 000000009b29b30e (rcu_read_lock){....}, at: notify_ring+0x1a/0x480 [i915] [ 57.410573] #2: 00000000aad89594 (&(&rq->lock)->rlock#2){-.-.}, at: notify_ring+0x2b2/0x480 [i915] [ 57.410601] #3: 000000009b29b30e (rcu_read_lock){....}, at: submit_notify+0x35/0x124 [i915] [ 57.410635] stack backtrace: [ 57.410640] CPU: 6 PID: 0 Comm: swapper/6 Tainted: G U W 4.18.0-rc4-CI-CI_DII_1137+ #1 [ 57.410644] Hardware name: Intel Corporation Ice Lake Client Platform/IceLake U DDR4 SODIMM PD RVP, BIOS ICLSFWR1.R00.2222.A01.1805300339 05/30/2018 [ 57.410650] Call Trace: [ 57.410652] <IRQ> [ 57.410657] dump_stack+0x67/0x9b [ 57.410662] print_circular_bug.isra.16+0x1c8/0x2b0 [ 57.410666] __lock_acquire+0x1897/0x1b50 [ 57.410671] ? lock_acquire+0xa6/0x210 [ 57.410674] lock_acquire+0xa6/0x210 [ 57.410706] ? execlists_submit_request+0x2b/0x1a0 [i915] [ 57.410711] _raw_spin_lock_irqsave+0x33/0x50 [ 57.410741] ? execlists_submit_request+0x2b/0x1a0 [i915] [ 57.410769] execlists_submit_request+0x2b/0x1a0 [i915] [ 57.410774] ? _raw_spin_unlock_irqrestore+0x39/0x60 [ 57.410804] submit_notify+0x8d/0x124 [i915] [ 57.410828] __i915_sw_fence_complete+0x81/0x250 [i915] [ 57.410854] dma_i915_sw_fence_wake+0xd/0x20 [i915] [ 57.410858] dma_fence_signal_locked+0x79/0x200 [ 57.410882] notify_ring+0x2ba/0x480 [i915] [ 57.410907] gen8_cs_irq_handler+0x39/0xa0 [i915] [ 57.410933] gen11_irq_handler+0x2f0/0x420 [i915] [ 57.410938] __handle_irq_event_percpu+0x42/0x370 [ 57.410943] handle_irq_event_percpu+0x2b/0x70 [ 57.410947] handle_irq_event+0x2f/0x50 [ 57.410951] handle_edge_irq+0xe7/0x190 [ 57.410955] handle_irq+0x67/0x160 [ 57.410958] do_IRQ+0x5e/0x120 [ 57.410962] common_interrupt+0xf/0xf [ 57.410965] </IRQ> [ 57.410969] RIP: 0010:cpuidle_enter_state+0xac/0x360 [ 57.410972] Code: 44 00 00 31 ff e8 84 93 91 ff 45 84 f6 74 12 9c 58 f6 c4 02 0f 85 31 02 00 00 31 ff e8 7d 30 98 ff e8 e8 0e 94 ff fb 4c 29 fb <48> ba cf f7 53 e3 a5 9b c4 20 48 89 d8 48 c1 fb 3f 48 f7 ea b8 ff [ 57.411015] RSP: 0018:ffffc90000133e90 EFLAGS: 00000216 ORIG_RAX: ffffffffffffffdd [ 57.411023] RAX: ffff8804ae748040 RBX: 000000000002a97d RCX: 0000000000000000 [ 57.411029] RDX: 0000000000000046 RSI: ffffffff82141263 RDI: ffffffff820f05a7 [ 57.411035] RBP: 0000000000000001 R08: 0000000000000001 R09: 0000000000000000 [ 57.411041] R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff8229f078 [ 57.411045] R13: ffff8804ab2adfa8 R14: 0000000000000000 R15: 0000000d5de092e3 [ 57.411052] do_idle+0x1f3/0x250 [ 57.411055] cpu_startup_entry+0x6a/0x70 [ 57.411059] start_secondary+0x19d/0x1f0 [ 57.411064] secondary_startup_64+0xa5/0xb0 The easiest remedy is to remove the defunct code. Fixes: ff047a87cfac ("drm/i915/icl: Correctly clear lost ctx-switch interrupts across reset for Gen11") References: fd8526e50902 ("drm/i915/execlists: Trust the CSB") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michel Thierry <michel.thierry@intel.com> Cc: Oscar Mateo <oscar.mateo@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Reviewed-by: Michel Thierry <michel.thierry@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180713203529.1973-3-chris@chris-wilson.co.uk
* drm/i915/execlists: Switch to rb_root_cachedChris Wilson2018-07-111-21/+11
| | | | | | | | | | | The kernel recently gained an augmented rbtree with the purpose of cacheing the leftmost element of the rbtree, a frequent optimisation to avoid calls to rb_first() which is also employed by the execlists->queue. Switch from our open-coded cache to the library. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180629075348.27358-9-chris@chris-wilson.co.uk
* drm/i915: Only reset hangcheck at the start of an activity cycleChris Wilson2018-07-101-1/+0
| | | | | | | | | | | | | | | | Across a reset, the seqno (and thus hangcheck) should restart and the hangcheck naturally progress, for when it does not, we want to declare an emergency. Currently, we only detect if reset and reinit fails, but we do not detect if the call to reinit succeeds but the HW is fried - as we are resetting hangcheck on initialisation the engine. Remove that and rely on the natural progress to reset the hangcheck timer. References: e21b141376f9 ("drm/i915: Mark the hangcheck as idle when unparking the engines") References: 1fd00c0faeec ("drm/i915: Declare the driver wedged if hangcheck makes no progress") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180709130208.11730-2-chris@chris-wilson.co.uk
* drm/i915/execlists: Direct submission of new requests (avoid tasklet/ksoftirqd)Chris Wilson2018-06-281-36/+66
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Back in commit 27af5eea54d1 ("drm/i915: Move execlists irq handler to a bottom half"), we came to the conclusion that running our CSB processing and ELSP submission from inside the irq handler was a bad idea. A really bad idea as we could impose nearly 1s latency on other users of the system, on average! Deferring our work to a tasklet allowed us to do the processing with irqs enabled, reducing the impact to an average of about 50us. We have since eradicated the use of forcewaked mmio from inside the CSB processing and ELSP submission, bringing the impact down to around 5us (on Kabylake); an order of magnitude better than our measurements 2 years ago on Broadwell and only about 2x worse on average than the gem_syslatency on an unladen system. In this iteration of the tasklet-vs-direct submission debate, we seek a compromise where by we submit new requests immediately to the HW but defer processing the CS interrupt onto a tasklet. We gain the advantage of low-latency and ksoftirqd avoidance when waking up the HW, while avoiding the system-wide starvation of our CS irq-storms. Comparing the impact on the maximum latency observed (that is the time stolen from an RT process) over a 120s interval, repeated several times (using gem_syslatency, similar to RT's cyclictest) while the system is fully laden with i915 nops, we see that direct submission an actually improve the worse case. Maximum latency in microseconds of a third party RT thread (gem_syslatency -t 120 -f 2) x Always using tasklets (a couple of >1000us outliers removed) + Only using tasklets from CS irq, direct submission of requests +------------------------------------------------------------------------+ | + | | + | | + | | + + | | + + + | | + + + + x x x | | +++ + + + x x x x x x | | +++ + ++ + + *x x x x x x | | +++ + ++ + * *x x * x x x | | + +++ + ++ * * +*xxx * x x xx | | * +++ + ++++* *x+**xx+ * x x xxxx x | | **x++++*++**+*x*x****x+ * +x xx xxxx x x | |x* ******+***************++*+***xxxxxx* xx*x xxx + x+| | |__________MA___________| | | |______M__A________| | +------------------------------------------------------------------------+ N Min Max Median Avg Stddev x 118 91 186 124 125.28814 16.279137 + 120 92 187 109 112.00833 13.458617 Difference at 95.0% confidence -13.2798 +/- 3.79219 -10.5994% +/- 3.02677% (Student's t, pooled s = 14.9237) However the mean latency is adversely affected: Mean latency in microseconds of a third party RT thread (gem_syslatency -t 120 -f 1) x Always using tasklets + Only using tasklets from CS irq, direct submission of requests +------------------------------------------------------------------------+ | xxxxxx + ++ | | xxxxxx + ++ | | xxxxxx + +++ ++ | | xxxxxxx +++++ ++ | | xxxxxxx +++++ ++ | | xxxxxxx +++++ +++ | | xxxxxxx + ++++++++++ | | xxxxxxxx ++ ++++++++++ | | xxxxxxxx ++ ++++++++++ | | xxxxxxxxxx +++++++++++++++ | | xxxxxxxxxxx x +++++++++++++++ | |x xxxxxxxxxxxxx x + + ++++++++++++++++++ +| | |__A__| | | |____A___| | +------------------------------------------------------------------------+ N Min Max Median Avg Stddev x 120 3.506 3.727 3.631 3.6321417 0.02773109 + 120 3.834 4.149 4.039 4.0375167 0.041221676 Difference at 95.0% confidence 0.405375 +/- 0.00888913 11.1608% +/- 0.244735% (Student's t, pooled s = 0.03513) However, since the mean latency corresponds to the amount of irqsoff processing we have to do for a CS interrupt, we only need to speed that up to benefit not just system latency but our own throughput. v2: Remember to defer submissions when under reset. v4: Only use direct submission for new requests v5: Be aware that with mixing direct tasklet evaluation and deferred tasklets, we may end up idling before running the deferred tasklet. v6: Remove the redudant likely() from tasklet_is_enabled(), restrict the annotation to reset_in_progress(). v7: Take the full timeline.lock when enabling perf_pmu stats as the tasklet is no longer a valid guard. A consequence is that the stats are now only valid for engines also using the timeline.lock to process state. Testcase: igt/gem_exec_latency/*rthog* References: 27af5eea54d1 ("drm/i915: Move execlists irq handler to a bottom half") Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-9-chris@chris-wilson.co.uk
* drm/i915/execlists: Trust the CSBChris Wilson2018-06-281-25/+4
| | | | | | | | | | | | | | | | | Now that we use the CSB stored in the CPU friendly HWSP, we do not need to track interrupts for when the mmio CSB registers are valid and can just check where we read up to last from the cached HWSP. This means we can forgo the atomic bit tracking from interrupt, and in the next patch it means we can check the CSB at any time. v2: Change the splitting inside reset_prepare, we only want to lose testing the interrupt in this patch, the next patch requires the change in locking Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-8-chris@chris-wilson.co.uk
generated by cgit v1.2.3 (git 2.25.1) at 2024-08-12 18:04:59 +0000