diff options
| author | Chris Wilson <[email protected]> | 2018-01-22 13:55:41 +0000 |
|---|---|---|
| committer | Chris Wilson <[email protected]> | 2018-01-22 18:27:04 +0000 |
| commit | c1beabcf143c72ecdfe76bf50aa6e385a6192d08 (patch) | |
| tree | 3eed763f4d93143cee7314f218ade1dd0fbee7b5 /tools/perf/scripts/python | |
| parent | 10bde236eff901407bdbcad2e605edaf7ba23bdb (diff) | |
drm/i915: Increase render/media power gating hysteresis for gen9+
On gen9+, after an idle period the HW will disable the entire power well
to conserve power (by preventing current leakage). It takes around a 100
microseconds to bring the power well back online afterwards. With the
current hysteresis value of 25us (really 25 * 1280ns), we do not have
sufficient time to respond to an interrupt and schedule the next execution
before the HW powers itself down. (At present, we prevent this by
grabbing the forcewake for prolonged periods of time, but that overkill
fixed in the next patch.) The minimum we want to set the power gating
hysteresis to is the length of time it takes us to service the GPU, which
across a broad spectrum of machines is about 250us.
(Note this also brings guc latency into the same ballpark as execlists.)
v2: Include some notes on where I plucked the numbers from.
Testcase: igt/gem_exec_nop/sequential
Signed-off-by: Chris Wilson <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Michal Wajdeczko <[email protected]>
Cc: Sagar Arun Kamble <[email protected]>
Cc: Michel Thierry <[email protected]>
Cc: Michal Winiarski <[email protected]>
Reviewed-by: Sagar Arun Kamble <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Diffstat (limited to 'tools/perf/scripts/python')
0 files changed, 0 insertions, 0 deletions