[Intel-gfx] [PATCH v2 1/2] drm/i915: Extend intel_wait_for_register_fw() with fast timeout

[Chris Wilson]

On Fri, Apr 07, 2017 at 04:01:44PM +0000, Michal Wajdeczko wrote:
> In some cases we may want to spend more time in atomic wait than
> hardcoded 2us. Let's add additional fast timeout parameter to allow
> flexible configuration of atomic timeout before switching into heavy wait.
> Add also possibility to return registry value to avoid extra read.
> 
> v2: use explicit fast timeout (Tvrtko/Chris)
>     allow returning register value (Chris)
> 
> Signed-off-by: Michal Wajdeczko <michal.wajdeczko at intel.com>
> Suggested-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
> Cc: Chris Wilson <chris at chris-wilson.co.uk>

For fun this causes:

add/remove: 2/1 grow/shrink: 6/2 up/down: 626/-348 (278)
function                                     old     new   delta
__intel_wait_for_register_fw                   -     429    +429
gen6_hw_domain_reset                           -      57     +57
sandybridge_pcode_write                      494     528     +34
sandybridge_pcode_read                       524     554     +30
gen8_reset_engines                           287     313     +26
gen6_bsd_submit_request                      156     178     +22
init_ring_common                            1368    1382     +14
gmbus_wait_idle                              204     218     +14
gen6_reset_engines                           183     160     -23
intel_guc_reset                              140     109     -31
intel_wait_for_register_fw                   294       -    -294

Hmm. Thinking about it more, using _wait_for() at all here is pointless.

You just want to do something like,

if (fast_timeout_us > 10)
	fast_timeout_us = 10;

So

-               ret = _wait_for(done, fast_timeout_us, 10);
-       else
-               ret = _wait_for_atomic(done, fast_timeout_us, 0);
+       if (fast_timeout_us > 50000)
+               fast_timeout_us = 50000;
+       ret = _wait_for_atomic(done, fast_timeout_us, 0);

gives

add/remove: 2/1 grow/shrink: 6/2 up/down: 515/-348 (167)
function                                     old     new   delta
__intel_wait_for_register_fw                   -     318    +318
gen6_hw_domain_reset                           -      57     +57
sandybridge_pcode_write                      494     528     +34
sandybridge_pcode_read                       524     554     +30
gen8_reset_engines                           287     313     +26
gen6_bsd_submit_request                      156     178     +22
init_ring_common                            1368    1382     +14
gmbus_wait_idle                              204     218     +14
gen6_reset_engines                           183     160     -23
intel_guc_reset                              140     109     -31
intel_wait_for_register_fw                   294       -    -294

which is more reasonable, and fits in with the current warnings. Happy?
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre