If we restrict ourselves to only using a cacheline for each timeline's
HWSP (we could go smaller, but want to avoid needless polluting
cachelines on different engines between different contexts), then we can
suballocate a single 4k page into 64 different timeline HWSP. By
treating each fresh allocation as a slab of 64 entries, we can keep it
around for the next 64 allocation attempts until we need to refresh the
slab cache.
John Harrison noted the issue of fragmentation leading to the same worst
case performance of one page per timeline as before, which can be
mitigated by adopting a freelist.
v2: Keep all partially allocated HWSP on a freelist
This is still without migration, so it is possible for the system to end
up with each timeline in its own page, but we ensure that no new
allocation would needless allocate a fresh page!
v3: Throw a selftest at the allocator to try and catch invalid cacheline
reuse.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128181812.22804-4-chris@chris-wilson.co.uk
As realised by commit 9e3d6223d2 ("math64, timers: Fix 32bit
mul_u64_u32_shr() and friends"), GCC does not always generate ideal code
for performing a 32b x 32b multiply returning a 64b result (i.e. where
we idiomatically use u64 result = (u64)x * (u32)x).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20170913105154.2910-2-chris@chris-wilson.co.uk
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
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
Some pieces of code are independent of hardware but are very tricky to
exercise through the normal userspace ABI or via debugfs hooks. Being
able to create mock unit tests and execute them through CI is vital.
Start by adding a central point where we can execute unit tests and
a parameter to enable them. This is disabled by default as the
expectation is that these tests will occasionally explode.
To facilitate integration with igt, any parameter beginning with
i915.igt__ is interpreted as a subtest executable independently via
igt/drv_selftest.
Two classes of selftests are recognised: mock unit tests and integration
tests. Mock unit tests are run as soon as the module is loaded, before
the device is probed. At that point there is no driver instantiated and
all hw interactions must be "mocked". This is very useful for writing
universal tests to exercise code not typically run on a broad range of
architectures. Alternatively, you can hook into the live selftests and
run when the device has been instantiated - hw interactions are real.
v2: Add a macro for compiling conditional code for mock objects inside
real objects.
v3: Differentiate between mock unit tests and late integration test.
v4: List the tests in natural order, use igt to sort after modparam.
v5: s/late/live/
v6: s/unsigned long/unsigned int/
v7: Use igt_ prefixes for long helpers.
v8: Deobfuscate macros overriding functions, stop using -I$(src)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170213171558.20942-1-chris@chris-wilson.co.uk
