[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1460111501.1800.28.camel@perches.com>
Date: Fri, 08 Apr 2016 03:31:41 -0700
From: Joe Perches <joe@...ches.com>
To: Yuyang Du <yuyang.du@...el.com>, peterz@...radead.org,
mingo@...nel.org, linux-kernel@...r.kernel.org
Cc: bsegall@...gle.com, pjt@...gle.com, morten.rasmussen@....com,
vincent.guittot@...aro.org, dietmar.eggemann@....com
Subject: Re: [PATCH] sched/fair: Optimize sum computation with a lookup table
On Fri, 2016-04-08 at 10:07 +0800, Yuyang Du wrote:
> __compute_runnable_contrib() uses a loop to compute sum, whereas a
> table lookup can do it faster in a constant time.
Perhaps this becomes rather fragile code overly dependent on the
current #define values of LOAD_AVG_MAX_N and LOAD_AVG_PERIOD.
Perhaps this comment just above the definitions of LOAD_AVG_MAX_N
and LOAD_AVG_PERIOD should be updated to include this new table:
* Note: The tables runnable_avg_yN_inv and runnable_avg_yN_sum are
* dependent on this value.
Perhaps the __ prefix for __accumulated_sum_N32 is odd as both of
the runnable_avg_yN_ tables are not __ prefixed.
Powered by blists - more mailing lists