linux/drivers/gpu/drm/i915/i915_gem_timeline.h
Chris Wilson 4797948071 drm/i915: Squash repeated awaits on the same fence
Track the latest fence waited upon on each context, and only add a new
asynchronous wait if the new fence is more recent than the recorded
fence for that context. This requires us to filter out unordered
timelines, which are noted by DMA_FENCE_NO_CONTEXT. However, in the
absence of a universal identifier, we have to use our own
i915->mm.unordered_timeline token.

v2: Throw around the debug crutches
v3: Inline the likely case of the pre-allocation cache being full.
v4: Drop the pre-allocation support, we can lose the most recent fence
in case of allocation failure -- it just means we may emit more awaits
than strictly necessary but will not break.
v5: Trim allocation size for leaf nodes, they only need an array of u32
not pointers.
v6: Create mock_timeline to tidy selftest writing
v7: s/intel_timeline_sync_get/intel_timeline_sync_is_later/ (Tvrtko)
v8: Prune the stale sync points when we idle.
v9: Include a small benchmark in the kselftests
v10: Separate the idr implementation into its own compartment. (Tvrkto)
v11: Refactor igt_sync kselftests to avoid deep nesting (Tvrkto)
v12: __sync_leaf_idx() to assert that p->height is 0 when checking leaves
v13: kselftests to investigate struct i915_syncmap itself (Tvrtko)
v14: Foray into ascii art graphs
v15: Take into account that the random lookup/insert does 2 prng calls,
not 1, when benchmarking, and use for_each_set_bit() (Tvrtko)
v16: Improved ascii art

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: http://patchwork.freedesktop.org/patch/msgid/20170503093924.5320-4-chris@chris-wilson.co.uk
2017-05-03 11:08:48 +01:00

117 lines
3.7 KiB
C

/*
* Copyright © 2016 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
*/
#ifndef I915_GEM_TIMELINE_H
#define I915_GEM_TIMELINE_H
#include <linux/list.h>
#include "i915_utils.h"
#include "i915_gem_request.h"
#include "i915_syncmap.h"
struct i915_gem_timeline;
struct intel_timeline {
u64 fence_context;
u32 seqno;
/**
* Count of outstanding requests, from the time they are constructed
* to the moment they are retired. Loosely coupled to hardware.
*/
u32 inflight_seqnos;
spinlock_t lock;
/**
* List of breadcrumbs associated with GPU requests currently
* outstanding.
*/
struct list_head requests;
/* Contains an RCU guarded pointer to the last request. No reference is
* held to the request, users must carefully acquire a reference to
* the request using i915_gem_active_get_request_rcu(), or hold the
* struct_mutex.
*/
struct i915_gem_active last_request;
/**
* We track the most recent seqno that we wait on in every context so
* that we only have to emit a new await and dependency on a more
* recent sync point. As the contexts may be executed out-of-order, we
* have to track each individually and can not rely on an absolute
* global_seqno. When we know that all tracked fences are completed
* (i.e. when the driver is idle), we know that the syncmap is
* redundant and we can discard it without loss of generality.
*/
struct i915_syncmap *sync;
u32 sync_seqno[I915_NUM_ENGINES];
struct i915_gem_timeline *common;
};
struct i915_gem_timeline {
struct list_head link;
struct drm_i915_private *i915;
const char *name;
struct intel_timeline engine[I915_NUM_ENGINES];
};
int i915_gem_timeline_init(struct drm_i915_private *i915,
struct i915_gem_timeline *tl,
const char *name);
int i915_gem_timeline_init__global(struct drm_i915_private *i915);
void i915_gem_timelines_mark_idle(struct drm_i915_private *i915);
void i915_gem_timeline_fini(struct i915_gem_timeline *tl);
static inline int __intel_timeline_sync_set(struct intel_timeline *tl,
u64 context, u32 seqno)
{
return i915_syncmap_set(&tl->sync, context, seqno);
}
static inline int intel_timeline_sync_set(struct intel_timeline *tl,
const struct dma_fence *fence)
{
return __intel_timeline_sync_set(tl, fence->context, fence->seqno);
}
static inline bool __intel_timeline_sync_is_later(struct intel_timeline *tl,
u64 context, u32 seqno)
{
return i915_syncmap_is_later(&tl->sync, context, seqno);
}
static inline bool intel_timeline_sync_is_later(struct intel_timeline *tl,
const struct dma_fence *fence)
{
return __intel_timeline_sync_is_later(tl, fence->context, fence->seqno);
}
#endif