Skip to contentSkip to conversation

__timer_udelay(1) may return immediately

PATCH Mason 1 participant 2 messages Apr 2, 2015 Arm (32-bit)

lore
Share & tools
lore

Mason

author
#0 lore raw

Hello everyone,

I'm using timer-based delays from arch/arm/lib/delay.c

Consider the following configuration:
HZ=100
timer->freq = 1000000

Thus
UDELAY_MULT = 107374
ticks_per_jiffy = 10000

Thus __timer_udelay(1) =>
__timer_const_udelay(107374) =>
__timer_delay(0) => calls get_cycles() twice then returns prematurely

The issue comes from a tiny rounding error as 107374 * ticks_per_jiffy >> UDELAY_SHIFT = 0,9999983 which is rounded down to 0.

The root of the issue is that mathematically, UDELAY_MULT = 2199023 * HZ / 2048 = 107374,169921875 which is rounded down to 107374.

It seems to me that a simple solution would be to round UDELAY_MULT up instead of down.

Thus UDELAY_MULT = 107375
107375 * ticks_per_jiffy >> UDELAY_SHIFT = 1,0000076

We might end up sleeping one cycle more than necessary, but I don't think spinning a bit longer would be a problem?

Patch provided for illustration purposes.

What do you think?

Regards.

arch/arm/include/asm/delay.h +1 -1
index dff714d..873a43e 100644
@@ -10,7 +10,7 @@
  #include <asm/param.h> /* HZ */
  
  #define MAX_UDELAY_MS  2
-#define UDELAY_MULT    ((UL(2199023) * HZ) >> 11)
+#define UDELAY_MULT    (((UL(2199023) * HZ) >> 11) + 1)
  #define UDELAY_SHIFT   30
  
  #ifndef __ASSEMBLY__
Signature · 2 lines
(Note to self: tangentially relevant discussion)
http://thread.gmane.org/gmane.linux.kernel/1858260
author
#1 lore raw
Quoted 43 lines · Mason

I'm using timer-based delays from arch/arm/lib/delay.c

Consider the following configuration:
HZ=100
timer->freq = 1000000

Thus
UDELAY_MULT = 107374
ticks_per_jiffy = 10000

Thus __timer_udelay(1) =>
__timer_const_udelay(107374) =>
__timer_delay(0) => calls get_cycles() twice then returns prematurely

The issue comes from a tiny rounding error as 107374 * ticks_per_jiffy >> UDELAY_SHIFT = 0,9999983 which is rounded down to 0.

The root of the issue is that mathematically, UDELAY_MULT = 2199023 * HZ / 2048 = 107374,169921875 which is rounded down to 107374.

It seems to me that a simple solution would be to round UDELAY_MULT up instead of down.

Thus UDELAY_MULT = 107375
107375 * ticks_per_jiffy >> UDELAY_SHIFT = 1,0000076

We might end up sleeping one cycle more than necessary, but I don't think spinning a bit longer would be a problem?

Patch provided for illustration purposes.

What do you think?

Regards.

arch/arm/include/asm/delay.h +0 -0
index dff714d..873a43e 100644

arch/arm/include/asm/delay.h line 16

Quoted 8 lines
arch/arm/include/asm/delay.h +1 -1
@@ -10,7 +10,7 @@
  #include <asm/param.h> /* HZ */
 
  #define MAX_UDELAY_MS  2
-#define UDELAY_MULT    ((UL(2199023) * HZ) >> 11)
+#define UDELAY_MULT    (((UL(2199023) * HZ) >> 11) + 1)
  #define UDELAY_SHIFT   30
 
  #ifndef __ASSEMBLY__

Come to think of it, a closely related issue is: what to do when the user requests a delay which resolves to a cycle count with a non-zero fractional part? (e.g. delay for 7.2 cycles)

I think we should round up these values (delay for 8 cycles in the example). So forget the first patch, keep the rounded down value for UDELAY_MULT, and round up the cycle count.

arch/arm/lib/delay.c +1 -1
index 5306de3..a9b3c75 100644
@@ -59,7 +59,7 @@ static void __timer_const_udelay(unsigned long xloops)
  {
         unsigned long long loops = xloops;
         loops *= arm_delay_ops.ticks_per_jiffy;
-       __timer_delay(loops >> UDELAY_SHIFT);
+       __timer_delay((loops >> UDELAY_SHIFT) + 1);
  }
  
  static void __timer_udelay(unsigned long usecs)

Also, I was thinking of implementing ndelay() in delay.h

Would it make sense to define

#define NSDELAY_MULT ((UL(281475) * HZ) >> 18) // or perhaps 281474?
and have ndelay(ns) resolve __const_udelay((ns) * NSDELAY_MULT))

Or should I just keep that in platform-specific headers?

Regards.