linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Rafael J. Wysocki" <rjw@rjwysocki.net>
To: Linux PM <linux-pm@vger.kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
	Doug Smythies <dsmythies@telus.net>
Subject: [PATCH 16/16] cpufreq: intel_pstate: Eliminate intel_pstate_get_min_max()
Date: Tue, 28 Mar 2017 00:24:26 +0200	[thread overview]
Message-ID: <51413518.j7MmntgXCY@aspire.rjw.lan> (raw)
In-Reply-To: <6409323.DYHvh3CYlO@aspire.rjw.lan>

From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Some computations in intel_pstate_get_min_max() are not necessary
and one of its two callers doesn't even use the full result.

First off, the fixed-point value of cpu->max_perf represents a
non-negative number between 0 and 1 inclusive and cpu->min_perf
cannot be greater than cpu->max_perf.  It is not necessary to check
those conditions every time the numbers in question are used.

Moreover, since intel_pstate_max_within_limits() only needs the
upper boundary, it doesn't make sense to compute the lower one in
there and returning min and max from intel_pstate_get_min_max()
via pointers doesn't look particularly nice.

For the above reasons, drop intel_pstate_get_min_max(), add a helper
to get the base P-state for min/max computations and carry out them
directly in the previous callers of intel_pstate_get_min_max().

Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---
 drivers/cpufreq/intel_pstate.c |   40 ++++++++++++++--------------------------
 1 file changed, 14 insertions(+), 26 deletions(-)

Index: linux-pm/drivers/cpufreq/intel_pstate.c
===================================================================
--- linux-pm.orig/drivers/cpufreq/intel_pstate.c
+++ linux-pm/drivers/cpufreq/intel_pstate.c
@@ -1496,26 +1496,10 @@ static int knl_get_turbo_pstate(void)
 	return ret;
 }
 
-static void intel_pstate_get_min_max(struct cpudata *cpu, int *min, int *max)
+static int intel_pstate_get_base_pstate(struct cpudata *cpu)
 {
-	int max_perf = cpu->pstate.turbo_pstate;
-	int max_perf_adj;
-	int min_perf;
-
-	if (global.no_turbo || global.turbo_disabled)
-		max_perf = cpu->pstate.max_pstate;
-
-	/*
-	 * performance can be limited by user through sysfs, by cpufreq
-	 * policy, or by cpu specific default values determined through
-	 * experimentation.
-	 */
-	max_perf_adj = fp_ext_toint(max_perf * cpu->max_perf);
-	*max = clamp_t(int, max_perf_adj,
-			cpu->pstate.min_pstate, cpu->pstate.turbo_pstate);
-
-	min_perf = fp_ext_toint(max_perf * cpu->min_perf);
-	*min = clamp_t(int, min_perf, cpu->pstate.min_pstate, max_perf);
+	return global.no_turbo || global.turbo_disabled ?
+			cpu->pstate.max_pstate : cpu->pstate.turbo_pstate;
 }
 
 static void intel_pstate_set_pstate(struct cpudata *cpu, int pstate)
@@ -1538,11 +1522,13 @@ static void intel_pstate_set_min_pstate(
 
 static void intel_pstate_max_within_limits(struct cpudata *cpu)
 {
-	int min_pstate, max_pstate;
+	int pstate;
 
 	update_turbo_state();
-	intel_pstate_get_min_max(cpu, &min_pstate, &max_pstate);
-	intel_pstate_set_pstate(cpu, max_pstate);
+	pstate = intel_pstate_get_base_pstate(cpu);
+	pstate = max(cpu->pstate.min_pstate,
+		     fp_ext_toint(pstate * cpu->max_perf));
+	intel_pstate_set_pstate(cpu, pstate);
 }
 
 static void intel_pstate_get_cpu_pstates(struct cpudata *cpu)
@@ -1704,11 +1690,13 @@ static inline int32_t get_target_pstate_
 
 static int intel_pstate_prepare_request(struct cpudata *cpu, int pstate)
 {
-	int max_perf, min_perf;
+	int max_pstate = intel_pstate_get_base_pstate(cpu);
+	int min_pstate;
 
-	intel_pstate_get_min_max(cpu, &min_perf, &max_perf);
-	pstate = clamp_t(int, pstate, min_perf, max_perf);
-	return pstate;
+	min_pstate = max(cpu->pstate.min_pstate,
+			 fp_ext_toint(max_pstate * cpu->min_perf));
+	max_pstate = max(min_pstate, fp_ext_toint(max_pstate * cpu->max_perf));
+	return clamp_t(int, pstate, min_pstate, max_pstate);
 }
 
 static void intel_pstate_update_pstate(struct cpudata *cpu, int pstate)

      parent reply	other threads:[~2017-03-27 22:39 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-27 22:01 [PATCH 00/16] cpufreq: intel_pstate: Cleanups and optimizations Rafael J. Wysocki
2017-03-27 22:03 ` [PATCH 01/16] cpufreq: intel_pstate: Eliminate struct perf_limits Rafael J. Wysocki
2017-03-27 22:04 ` [PATCH 02/16] cpufreq: intel_pstate: Drop pointless initialization of PID parameters Rafael J. Wysocki
2017-03-27 22:05 ` [PATCH 03/16] cpufreq: intel_pstate: Initialize pid_params statically Rafael J. Wysocki
2017-03-27 22:07 ` [PATCH 04/16] cpufreq: intel_pstate: Fold intel_pstate_reset_all_pid() into the caller Rafael J. Wysocki
2017-03-27 22:09 ` [PATCH 05/16] cpufreq: intel_pstate: Clean up intel_pstate_busy_pid_reset() Rafael J. Wysocki
2017-03-27 22:10 ` [PATCH 06/16] cpufreq: intel_pstate: Set HWP sampling interval once Rafael J. Wysocki
2017-03-27 22:11 ` [PATCH 07/16] cpufreq: intel_pstate: Skip unnecessary PID resets on init Rafael J. Wysocki
2017-03-27 22:13 ` [PATCH 08/16] cpufreq: intel_pstate: Drop driver_registered variable Rafael J. Wysocki
2017-03-27 22:14 ` [PATCH 09/16] cpufreq: intel_pstate: Modify check in intel_pstate_update_status() Rafael J. Wysocki
2017-03-27 22:15 ` [PATCH 10/16] cpufreq: intel_pstate: Use different utilization update callbacks Rafael J. Wysocki
2017-03-27 22:17 ` [PATCH 11/16] cpufreq: intel_pstate: Add update_util callback to pstate_funcs Rafael J. Wysocki
2017-03-27 22:18 ` [PATCH 12/16] cpufreq: intel_pstate: Move cpu_defaults definitions Rafael J. Wysocki
2017-03-27 22:19 ` [PATCH 13/16] cpufreq: intel_pstate: Drop struct cpu_defaults Rafael J. Wysocki
2017-03-27 22:20 ` [PATCH 14/16] cpufreq: intel_pstate: Introduce pid_in_use() Rafael J. Wysocki
2017-03-27 22:22 ` [PATCH 15/16] cpufreq: intel_pstate: Do not walk policy->cpus Rafael J. Wysocki
2017-03-27 22:24 ` Rafael J. Wysocki [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=51413518.j7MmntgXCY@aspire.rjw.lan \
    --to=rjw@rjwysocki.net \
    --cc=dsmythies@telus.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=srinivas.pandruvada@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).