Skip to contentSkip to conversation

omap3: pm: cpufreq: populate l3 opp1 again

PATCH In review Nishanth Menon 2 participants 2 messages Jan 28, 2010 – Feb 4, 2010

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 omap3: pm: cpufreq: populate l3 opp1 again

We had removed the frequency for OPP1 L3 when we used to use frequency to enable/disable frequencies. It is better to populate the same instead of confusing future readers of the code. The OPP1 remains disabled as explained in the discussion.

Discussion: http://marc.info/?t=126453821900001&r=1&w=2

Cc Andrew Murray <[email protected]>
Cc Kevin Hilman <[email protected]>
Cc Romit Dasgupta <[email protected]>

---

Ref:
v1: http://marc.info/?t=126465729200002&r=1&w=2

Adding Signed-off-by Benoit to attribute the comment message which I "borrowed" from his excellent explanation in the mail thread and put as comment.

 arch/arm/mach-omap2/cpufreq34xx.c |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)
arch/arm/mach-omap2/cpufreq34xx.c +9 -2
index 07873e8..9882451 100644
@@ -42,8 +42,15 @@ static struct omap_opp_def __initdata omap34xx_mpu_rate_table[] = {
 };
 
 static struct omap_opp_def __initdata omap34xx_l3_rate_table[] = {
-	/* OPP1 */
-	OMAP_OPP_DEF(false, 0, 975000),
+	/*
+	 * OPP1 - 41.5 MHz is disabled because: The voltage for that OPP is
+	 * almost the same than the one at 83MHz thus providing very little
+	 * gain for the power point of view. In term of energy it will even
+	 * increase the consumption due to the very negative performance
+	 * impact that frequency will do to the MPU and the whole system in
+	 * general.
+	 */
+	OMAP_OPP_DEF(false, 41500000, 975000),
 	/* OPP2 */
 	OMAP_OPP_DEF(true, 83000000, 1050000),
 	/* OPP3 */
Signature · 1 line
1.6.3.3
#1 lore raw
Quoted 13 lines · Nishanth Menon

We had removed the frequency for OPP1 L3 when we used to use frequency to enable/disable frequencies. It is better to populate the same instead of confusing future readers of the code. The OPP1 remains disabled as explained in the discussion.

Discussion: http://marc.info/?t=126453821900001&r=1&w=2

Cc Andrew Murray <[email protected]>
Cc Kevin Hilman <[email protected]>
Cc Romit Dasgupta <[email protected]>
Quoted 8 lines

---

Ref:
v1: http://marc.info/?t=126465729200002&r=1&w=2

Adding Signed-off-by Benoit to attribute the comment message which I "borrowed" from his excellent explanation in the mail thread and put as comment.

Thanks, pulled into pm-wip-opp branch.

Kevin

Quoted 27 lines
 arch/arm/mach-omap2/cpufreq34xx.c |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)
arch/arm/mach-omap2/cpufreq34xx.c +9 -2
index 07873e8..9882451 100644
@@ -42,8 +42,15 @@ static struct omap_opp_def __initdata omap34xx_mpu_rate_table[] = {
 };
 
 static struct omap_opp_def __initdata omap34xx_l3_rate_table[] = {
-	/* OPP1 */
-	OMAP_OPP_DEF(false, 0, 975000),
+	/*
+	 * OPP1 - 41.5 MHz is disabled because: The voltage for that OPP is
+	 * almost the same than the one at 83MHz thus providing very little
+	 * gain for the power point of view. In term of energy it will even
+	 * increase the consumption due to the very negative performance
+	 * impact that frequency will do to the MPU and the whole system in
+	 * general.
+	 */
+	OMAP_OPP_DEF(false, 41500000, 975000),
 	/* OPP2 */
 	OMAP_OPP_DEF(true, 83000000, 1050000),
 	/* OPP3 */
Signature · 1 line
1.6.3.3