Skip to contentSkip to conversation

Does anyone use CONFIG_TINY_PREEMPT_RCU?

DISCUSSION Paul E. McKenney 5 participants 21 messages Nov 13, 2012 – Nov 14, 2012 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

Thanx, Paul

#1 lore raw

2012/11/13 Paul E. McKenney <[email protected]>:

Quoted 10 lines

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I don't use it personally but if you remove it, does that mean that RCU couldn't be preemptible on UP?

#2 lore raw
Quoted 9 lines · Frederic Weisbecker

2012/11/13 Paul E. McKenney <[email protected]>:

Quoted 8 lines

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I don't use it personally but if you remove it, does that mean that RCU couldn't be preemptible on UP?

No, it would mean that on UP you could choose between TINY_RCU and TREE_PREEMPT_RCU, depending on whether you want tiny or preemptible.

- Josh Triplett

#3 lore raw

2012/11/13 Josh Triplett <[email protected]>:

Quoted 10 lines · Frederic Weisbecker
Quoted 9 lines

2012/11/13 Paul E. McKenney <[email protected]>:

Quoted 8 lines

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I don't use it personally but if you remove it, does that mean that RCU couldn't be preemptible on UP?

No, it would mean that on UP you could choose between TINY_RCU and TREE_PREEMPT_RCU, depending on whether you want tiny or preemptible.

Ok. I thought the TREE version wasn't possible anymore on UP when I saw some patches that removed optimizations for nr_online_cpus=1. Hence the confusion.

#4 lore raw
Quoted 17 lines · Frederic Weisbecker

2012/11/13 Josh Triplett <[email protected]>:

Quoted 16 lines · Frederic Weisbecker
Quoted 14 lines

2012/11/13 Paul E. McKenney <[email protected]>:

Quoted 10 lines

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I don't use it personally but if you remove it, does that mean that RCU couldn't be preemptible on UP?

Quoted 6 lines
Quoted 2 lines

No, it would mean that on UP you could choose between TINY_RCU and TREE_PREEMPT_RCU, depending on whether you want tiny or preemptible.

Ok. I thought the TREE version wasn't possible anymore on UP when I saw some patches that removed optimizations for nr_online_cpus=1. Hence the confusion.

Those optimizations are not critically important. That said, yes, I will need to restart testing of TREE_PREEMPT_RCU on !SMP kernels.

Thanx, Paul

#5 lore raw
In reply to Frederic Weisbecker
Quoted 9 lines · Frederic Weisbecker

2012/11/13 Paul E. McKenney <[email protected]>:

Quoted 8 lines

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I don't use it personally but if you remove it, does that mean that RCU couldn't be preemptible on UP?

It would mean that a kernel built with SMP=n and PREEMPT=y would use TREE_PREEMPT_RCU rather than the current TINY_PREEMPT_RCU. So it would work just as well, but use somewhat more memory.

Thanx, Paul

#6 lore raw
In reply to Paul E. McKenney
Quoted 10 lines · Paul E. McKenney

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

Yes, I use TINY_PREEMPT_RCU on my UP machines. It is, in fact, the only option.

Cheers,

Signature · 1 line
Nick Bowler, Elliptic Technologies (http://www.elliptictech.com/)
#7 lore raw
Quoted 9 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

Yes, I use TINY_PREEMPT_RCU on my UP machines. It is, in fact, the only option.

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

Thanx, Paul

#8 lore raw
Quoted 10 lines · Paul E. McKenney
Quoted 9 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

Yes, I use TINY_PREEMPT_RCU on my UP machines. It is, in fact, the only option.

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

To be honest I don't really know what the difference is, other than what the help text says, which is:

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

As a side note, I wonder why any of these RCU implementations are user-seclectable options in the first place? It looks like you will only ever have one choice, since the dependencies all seem mutually exclusive:

  TREE_RCU         depends on !PREEMPT &&  SMP
  TREE_PREEMPT_RCU depends on  PREEMPT &&  SMP
  TINY_RCU         depends on !PREEMPT && !SMP
  TINY_PREEMPT_RCU depends on  PREEMPT && !SMP

Cheers,

Signature · 1 line
Nick Bowler, Elliptic Technologies (http://www.elliptictech.com/)
#9 lore raw
Quoted 20 lines · Nick Bowler
Quoted 10 lines · Paul E. McKenney
Quoted 9 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney

Hello!

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

Yes, I use TINY_PREEMPT_RCU on my UP machines. It is, in fact, the only option.

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

Quoted 6 lines

To be honest I don't really know what the difference is, other than what the help text says, which is:

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory? How much memory does your device have, if I may ask?

Quoted 9 lines

As a side note, I wonder why any of these RCU implementations are user-seclectable options in the first place? It looks like you will only ever have one choice, since the dependencies all seem mutually exclusive:

  TREE_RCU         depends on !PREEMPT &&  SMP
  TREE_PREEMPT_RCU depends on  PREEMPT &&  SMP
  TINY_RCU         depends on !PREEMPT && !SMP
  TINY_PREEMPT_RCU depends on  PREEMPT && !SMP

Inertia on my part. ;-)

Thanx, Paul

#10 lore raw
Quoted 8 lines · Paul E. McKenney
Quoted 7 lines · Nick Bowler
Quoted 2 lines · Paul E. McKenney

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

To be honest I don't really know what the difference is, other than what the help text says, which is:

Quoted 6 lines
Quoted 3 lines

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory?

Well, I have no idea. If I was given the choice between TREE_PREEMPT_RCU and TINY_PREEMPT_RCU, absent any information not in the description of these options, I would choose TINY. The description suggests that the memory savings come at the expense of SMP support, which sounds like a great tradeoff to make for a UP system.

Quoted 1 line

How much memory does your device have, if I may ask?

It's a (pretty old!) desktop. I recently had to upgrade it to two gigabytes due to unbearable thrashing with only one...

Cheers,

Signature · 1 line
Nick Bowler, Elliptic Technologies (http://www.elliptictech.com/)
#11 lore raw
Quoted 22 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney
Quoted 7 lines · Nick Bowler
Quoted 2 lines · Paul E. McKenney

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

To be honest I don't really know what the difference is, other than what the help text says, which is:

Quoted 6 lines
Quoted 3 lines

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory?

Well, I have no idea. If I was given the choice between TREE_PREEMPT_RCU and TINY_PREEMPT_RCU, absent any information not in the description of these options, I would choose TINY. The description suggests that the memory savings come at the expense of SMP support, which sounds like a great tradeoff to make for a UP system.

Quoted 4 lines
Quoted 1 line

How much memory does your device have, if I may ask?

It's a (pretty old!) desktop. I recently had to upgrade it to two gigabytes due to unbearable thrashing with only one...

If you have two gigabytes (or even one gigabyte), you won't notice the few kilobytes of difference between TINY_PREEMPT_RCU and TREE_PREEMPT_RCU.

Thanx, Paul

#12 lore raw
Quoted 23 lines · Paul E. McKenney
Quoted 22 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney
Quoted 7 lines · Nick Bowler
Quoted 2 lines · Paul E. McKenney

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

To be honest I don't really know what the difference is, other than what the help text says, which is:

Quoted 6 lines
Quoted 3 lines

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory?

Well, I have no idea. If I was given the choice between TREE_PREEMPT_RCU and TINY_PREEMPT_RCU, absent any information not in the description of these options, I would choose TINY. The description suggests that the memory savings come at the expense of SMP support, which sounds like a great tradeoff to make for a UP system.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

How much memory does your device have, if I may ask?

It's a (pretty old!) desktop. I recently had to upgrade it to two gigabytes due to unbearable thrashing with only one...

If you have two gigabytes (or even one gigabyte), you won't notice the few kilobytes of difference between TINY_PREEMPT_RCU and TREE_PREEMPT_RCU.

Well then TINY_PREEMPT_RCU doesn't sound all that useful for me! Perhaps the help text could be improved... such as changing the words "greatly reduced" to "marginally reduced" as a first step?

Is there no significant cache impact due to the larger implementation? I don't really have the time or expertise to do measurements in this regard, but if TREE_PREEMPT_RCU was actually a selectable option I could at least choose it to see if anything explodes horribly...

Cheers,

Signature · 1 line
Nick Bowler, Elliptic Technologies (http://www.elliptictech.com/)
#13 lore raw
Quoted 32 lines · Nick Bowler
Quoted 23 lines · Paul E. McKenney
Quoted 22 lines · Nick Bowler
Quoted 8 lines · Paul E. McKenney
Quoted 7 lines · Nick Bowler
Quoted 2 lines · Paul E. McKenney

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT builds. Would that work for you?

To be honest I don't really know what the difference is, other than what the help text says, which is:

Quoted 6 lines
Quoted 3 lines

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory?

Well, I have no idea. If I was given the choice between TREE_PREEMPT_RCU and TINY_PREEMPT_RCU, absent any information not in the description of these options, I would choose TINY. The description suggests that the memory savings come at the expense of SMP support, which sounds like a great tradeoff to make for a UP system.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

How much memory does your device have, if I may ask?

It's a (pretty old!) desktop. I recently had to upgrade it to two gigabytes due to unbearable thrashing with only one...

If you have two gigabytes (or even one gigabyte), you won't notice the few kilobytes of difference between TINY_PREEMPT_RCU and TREE_PREEMPT_RCU.

Quoted 8 lines

Well then TINY_PREEMPT_RCU doesn't sound all that useful for me! Perhaps the help text could be improved... such as changing the words "greatly reduced" to "marginally reduced" as a first step?

Is there no significant cache impact due to the larger implementation? I don't really have the time or expertise to do measurements in this regard, but if TREE_PREEMPT_RCU was actually a selectable option I could at least choose it to see if anything explodes horribly...

Indeed, if I do this, the first step would be to re-introduce the choice between CONFIG_TREE_PREEMPT_RCU and CONFIG_TINY_PREEMPT_RCU so that you could check for horrible explosions. If the past is any guide, there would be at least a few. ;-)

Thanx, Paul

A message that is not in the archive (sent off-list or not yet received): CAGChsmOBB1yFNP5xSa06v+CYn0A=AjcZTyAzL8npWRdNYYLBZA@mail.gmail.com

#15 lore raw
Quoted 33 lines · Darren Hart
Quoted 30 lines · Paul E. McKenney
Quoted 4 lines · Nick Bowler
Quoted 3 lines · Paul E. McKenney
Quoted 2 lines · Nick Bowler
Quoted 1 line · Paul E. McKenney

Suppose that TREE_PREEMPT_RCU was available for !SMP && PREEMPT

builds.

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 1 line

Would that work for you?

To be honest I don't really know what the difference is, other than

what

Quoted 19 lines
Quoted 8 lines
Quoted 5 lines

the help text says, which is:

[TINY_PREEMPT_RCU] greatly reduces the memory footprint of RCU.

"Greatly reduced memory footprint" sounds pretty useful...

OK, so from your viewpoint, the only possible benefit is smaller memory?

Well, I have no idea. If I was given the choice between TREE_PREEMPT_RCU and TINY_PREEMPT_RCU, absent any information not in the description of these options, I would choose TINY. The description suggests that the memory savings come at the expense of SMP support, which sounds like a great tradeoff to make for a UP system.

Quoted 1 line

How much memory does your device have, if I may ask?

It's a (pretty old!) desktop. I recently had to upgrade it to two gigabytes due to unbearable thrashing with only one...

Quoted 8 lines
Quoted 4 lines

If you have two gigabytes (or even one gigabyte), you won't notice the few kilobytes of difference between TINY_PREEMPT_RCU and TREE_PREEMPT_RCU.

In that case I don't have any reason to be concerned about it. It seems then as though "tiny" is meant to reduce memory usage on systems with a great deal of memory where the structures add up to consume significant amounts of memory?

The "tiny" is for extremely small systems. The difference in memory consumption is otherwise not all that big a deal.

CONFIG_TINY_PREEMPT_RCU:

   text	   data	    bss	    dec	    hex	filename
   1079	    173	      0	   1252	    4e4	/tmp/b/kernel/rcupdate.o
   3223	    105	      4	   3332	    d04	/tmp/b/kernel/rcutiny.o

CONFIG_TREE_PREEMPT_RCU, NR_CPUS=2:

text data bss dec hex filename
1376 197 0 1573 625 /tmp/b/kernel/rcupdate.o
16545 2030 12 18587 489b /tmp/b/kernel/rcutree.o

So you get about 15Kbytes by using CONFIG_TINY_PREEMPT_RCU instead of CONFIG_TREE_PREEMPT_RCU. Actually you get a bit less than that, because CONFIG_TREE_PREEMPT_RCU's data would be a bit smaller for NR_CPUS=1.

But people who -really- care about memory measured in kilobytes usually go all the way to CONFIG_TINY_RCU:

CONFIG_TINY_RCU:

   text	   data	    bss	    dec	    hex	filename
    858	    173	      0	   1031	    407	/tmp/b/kernel/rcupdate.o
   1788	     47	      0	   1835	    72b	/tmp/b/kernel/rcutiny.o

This is about 2.5K better than CONFIG_TINY_PREEMPT_RCU. Also, !PREEMPT reduces the size of the rest of the kernel as well.

Hence my belief that CONFIG_TINY_PREEMPT_RCU isn't really helping much here.

Thanx, Paul

#16 lore raw
In reply to Paul E. McKenney

Hi Paul

Quoted 8 lines

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I am using TINY_PREEMPT_RCU on the i.mx35 platform. I am currently on 3.4-rt since the 3.6 compile failed due to for my config? Some strange device tree stuff which i haven't had the time to figure out why.

Best regards
Tim

#17 lore raw
Quoted 7 lines · Tim Sander

Hi Paul

Quoted 6 lines

I know of people using TINY_RCU, TREE_RCU, and TREE_PREEMPT_RCU, but I have not heard of anyone using TINY_PREEMPT_RCU for whom TREE_PREEMPT_RCU was not a viable option (in contrast, the people running Linux on tiny-memmory systems typically use TINY_RCU). Of course, if no one really needs it, the proper thing to do is to remove it.

Quoted 5 lines
Quoted 2 lines

So, if you need TINY_PREEMPT_RCU, please let me know. Otherwise, I will remove it, probably in the 3.9 timeframe.

I am using TINY_PREEMPT_RCU on the i.mx35 platform. I am currently on 3.4-rt since the 3.6 compile failed due to for my config? Some strange device tree stuff which i haven't had the time to figure out why.

OK, how much memory does your device have?

Thanx, Paul

#18 lore raw

Hi Paul

Sorry english somehow i tried to shoehorn more than one sentence into one sentence in the last mail.

Quoted 1 line

OK, how much memory does your device have?

It's 128Mb of memory.

Best regards
Tim

#19 lore raw
Quoted 6 lines · Tim Sander

Hi Paul

Sorry english somehow i tried to shoehorn more than one sentence into one sentence in the last mail.

Quoted 1 line

OK, how much memory does your device have?

It's 128Mb of memory.

That is a bit on the smaller side. But still quite large compared to the 15K difference between CONFIG_TREE_PREEMPT_RCU and CONFIG_TINY_PREEMPT_RCU.

Thanx, Paul

#20 lore raw
Quoted 11 lines · Paul E. McKenney
Quoted 6 lines · Tim Sander

Hi Paul

Sorry english somehow i tried to shoehorn more than one sentence into one sentence in the last mail.

Quoted 1 line

OK, how much memory does your device have?

It's 128Mb of memory.

That is a bit on the smaller side. But still quite large compared to the 15K difference between CONFIG_TREE_PREEMPT_RCU and CONFIG_TINY_PREEMPT_RCU.

I'd also expect CONFIG_TINY_PREEMPT_RCU to have marginally more performance on a UP system, but probably not enough to matter.

- Josh Triplett

#21 lore raw
Quoted 7 lines · Josh Triplett
Quoted 6 lines · Paul E. McKenney
Quoted 5 lines · Tim Sander

Hi Paul

Sorry english somehow i tried to shoehorn more than one sentence into one sentence in the last mail.

Quoted 1 line

OK, how much memory does your device have?

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

It's 128Mb of memory.

That is a bit on the smaller side. But still quite large compared to the 15K difference between CONFIG_TREE_PREEMPT_RCU and CONFIG_TINY_PREEMPT_RCU.

I'd also expect CONFIG_TINY_PREEMPT_RCU to have marginally more performance on a UP system, but probably not enough to matter.

Agreed, I would expect this to be down in the noise.

Thanx, Paul