Skip to contentSkip to conversation

init: calibrate: don't print out bogomips value on boot

PATCH In review Will Deacon 2 participants 3 messages Oct 9, 2013 – Oct 11, 2013

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 init: calibrate: don't print out bogomips value on boot

BogoMIPs is a confusing concept to the ill-informed, so allow architectures to print it only if they find it worthwhile. The delay calibration code should stick to lpj and avoid trying to draw any correlation with BogoMIPs, which may be a fixed value derived from a timer frequency independent of the CPU clock speed.

---

This is mostly cosmetic, but we've killed bogomips from the cpuinfo on ARM and arm64, so this would be the final nail on the coffin...

 init/calibrate.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
init/calibrate.c +3 -5
index 520702d..eabc2ca 100644
@@ -135,13 +135,13 @@ static unsigned long calibrate_delay_direct(void)
 		if ((measured_times[max] - estimate) <
 				(estimate - measured_times[min])) {
 			printk(KERN_NOTICE "calibrate_delay_direct() dropping "
-					"min bogoMips estimate %d = %lu\n",
+					"min delay estimate %d = %lu\n",
 				min, measured_times[min]);
 			measured_times[min] = 0;
 			min = max;
 		} else {
 			printk(KERN_NOTICE "calibrate_delay_direct() dropping "
-					"max bogoMips estimate %d = %lu\n",
+					"max delay estimate %d = %lu\n",
 				max, measured_times[max]);
 			measured_times[max] = 0;
 			max = min;
@@ -295,9 +295,7 @@ void calibrate_delay(void)
 	}
 	per_cpu(cpu_loops_per_jiffy, this_cpu) = lpj;
 	if (!printed)
-		pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			lpj/(500000/HZ),
-			(lpj/(5000/HZ)) % 100, lpj);
+		pr_cont("lpj=%lu\n", lpj);
 
 	loops_per_jiffy = lpj;
 	printed = true;
Signature · 1 line
1.8.2.2
#1 lore raw

init/calibrate.c void calibrate_delay(void) · line 301

Quoted 11 lines · On Wed, 9 Oct 2013 15:57:20 +0100 Will Deacon <[email protected]> wrote:
init/calibrate.c +1 -3
@@ -295,9 +295,7 @@ void calibrate_delay(void)
 	}
 	per_cpu(cpu_loops_per_jiffy, this_cpu) = lpj;
 	if (!printed)
-		pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			lpj/(500000/HZ),
-			(lpj/(5000/HZ)) % 100, lpj);
+		pr_cont("lpj=%lu\n", lpj);
 
 	loops_per_jiffy = lpj;
 	printed = true;

Kidding? BogoMIPS is an industry-standard metric which was established before many of us were born. It even has its own wikipedia page!

You have no heart.

#2 lore raw
Quoted 11 lines · Andrew Morton
Quoted 9 lines · On Wed, 9 Oct 2013 15:57:20 +0100 Will Deacon <[email protected]> wrote:
init/calibrate.c +1 -3
@@ -295,9 +295,7 @@ void calibrate_delay(void)
 	}
 	per_cpu(cpu_loops_per_jiffy, this_cpu) = lpj;
 	if (!printed)
-		pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n",
-			lpj/(500000/HZ),
-			(lpj/(5000/HZ)) % 100, lpj);
+		pr_cont("lpj=%lu\n", lpj);
 
Quoted 5 lines

init/calibrate.c void calibrate_delay(void) · line 301

Quoted 2 lines
init/calibrate.c +0 -0
 	loops_per_jiffy = lpj;
 	printed = true;

Kidding? BogoMIPS is an industry-standard metric which was established before many of us were born. It even has its own wikipedia page!

Crikey, have you looked at that page? ARM is ominously described as not enough data (yet)' when discussing its BogoMIPS rating'. This is all more food for the marketing machines :(

Quoted 1 line

You have no heart.

Maybe I should stop worrying and learn to love the BogoMIPS.

Will