tests/perf_pmu: Improve accuracy by waiting on spinner to start

More than one test assumes that the spinner is running pretty much
immediately after we have create or submitted it.

In actuality there is a variable delay, especially on execlists platforms,
between submission and spin batch starting to run on the hardware.

To enable tests which care about this level of timing to account for this,
we add a new spin batch constructor which provides an output field which
can be polled to determine when the batch actually started running.

This is implemented via MI_STOREDW_IMM from the spin batch, writing into
memory mapped page shared with userspace.

Using this facility from perf_pmu, where applicable, should improve very
occasional test fails across the set and platforms.

v2:
 Chris Wilson:
 * Use caching mapping if available.
 * Handle old gens better.
 * Use gem_can_store_dword.
 * Cache exec obj array in spin_batch_t for easier resubmit.

v3:
 * Forgot I915_EXEC_NO_RELOC. (Chris Wilson)

v4:
 * Mask out all non-engine flags in gem_can_store_dword.
 * Added some debug logging.

v5:
 * Fix relocs and batch munmap. (Chris)
 * Added assert idle spinner batch looks as expected.

v6:
 * Skip accuracy tests when !gem_can_store_dword.

v7:
 * Fix batch recursion reloc address.

v8:
 Chris Wilson:
 * Pull up gem_can_store_dword check before we start submitting.
 * Build spinner batch in a way we can skip store dword when not
   needed so we can run on SandyBridge.

v9:
 * Fix wait on spinner.
 * More tweaks to accuracy test.

v10:
 * Dropped accuracy subtest changes due problems with RT thread and
   tasklet submission.

v11:
 * Use READ_ONCE.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> # IRC

0 files changed, 0 insertions, 0 deletions