Skip to content

make mdelay, udelay and ssleep calls smaller

Denis Vlasenko lkml 1 version New

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest0 of 3 patches 0 repliesNew

Patches in v1 No applies recorded yet

None of the 3 patches the cover letter announces are in this archive.

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH 0/3] make mdelay, udelay and ssleep calls smaller

Hello Andrew,

This version is basically unchanged except for rediff against 2.6.18. I was running previous version for several weeks on 2.6.17.x kernels.

Russell King was giving me some feedback. I think that my reply satisfied him - I got no further comments:

Quoted 14 lines · Denis Vlasenko
Quoted 1 line · Russell King

Please keep a "const" version in ARM.  Thanks.

Why do you want to keep it?
I mean, without it udelay(n) will become slower by the time needed for one extra multiply. So we will have maybe udelay(n) ==> udelay(n+0.1).

Quoted 2 lines

Since the multiply is pure overhead, it's better
to get rid of it.

You do not need to optimize delay for speed. If you really want to, then you can do:

Quoted 7 lines

#define udelay(n) do {} while(0)

Makes sense? No it does not. That's what I'm trying to say. You WANT to pause for thousands of cycles. Why you are trying to shave off ~10 cycles at the very same time?

Hope it makes things clearer.

On with patch explanation:

Currently, magic in include/linux/delay.h
inlines mdelay and ssleep, and various arches
do the same to udelay.

This is pointless. We are going to perform a delay of 1000+ CPU cycles anyway, no need to optimize away a few cycles.

This patchset converts magic calls to these inlines/macros into pure and simple function calls, with no additional math done or hidden arguments pushed to stack
at the callsite.

Reduces allyesconfig kernel by about 7 kb.

--
vda