Re: 2.4.5 VM

Re: 2.4.5 VM

[Martin.Knoblauch]

Hi,

 first of all, I am not complaining, or calling things buggy. I know
that what I am running is "work in progress" and that one gets what one
deserves :-) 2.4.x has been stable for me and given me no severe problem
besides the changed pcmcia/cardbus support somewhere in 2.4.4-acx

 Just let me add my observation. The VM behaviour of 2.4.5 (started with
some 2.4.4-ac kernel) is definitely less than an improvement for *my*
setup. I am running a Thinkpad570 with 128 MB memory and about the same
amount of swap (I know, against reccomendation).

 Under the new VM behaviour, I easily get in a situation where the
system feels very sluggish. At that point in time, about 70-80% of
memory are Cache, the rest is Used and some small amount of free. Swap
is usually less than half filled and paging activity is about zero (some
sporadic page out). Typical case is a kernel build plus a Netscape
session. No unusal behaviour showing up in "top" - just sluggish system
response.

 My gut feeling is that the Cache is pressing to hard against process
memory. This may be great for some setups, but it is not good for others
(like mine).

 What would be great (maybe someone is already working on it) are some
tuning measures to tweak the cacheing behaviour.

 Just my 2 (Euro-)cents.

Martin
-- 
------------------------------------------------------------------
Martin Knoblauch         |    email:  Martin.Knoblauch@TeraPort.de
TeraPort GmbH            |    Phone:  +49-89-510857-309
C+ITS                    |    Fax:    +49-89-510857-111
http://www.teraport.de   |    Mobile: +49-170-4904759

Re: 2.4.5 VM

[Alan Cox]

>  Just let me add my observation. The VM behaviour of 2.4.5 (started with
> some 2.4.4-ac kernel) is definitely less than an improvement for *my*

2.4.5 and 2.4.4-ac are unrelated VM setups. 2.4.5-ac is probably much better
for a lot of loads as it has the 2.4.5 general behaviour but gets the aging
a lot more sane

Re: 2.4.5 VM

[Martin.Knoblauch]

[-- Attachment #1: Type: text/plain, Size: 1027 bytes --]

Alan Cox wrote:
> 
> >  Just let me add my observation. The VM behaviour of 2.4.5 (started with
> > some 2.4.4-ac kernel) is definitely less than an improvement for *my*
> 
> 2.4.5 and 2.4.4-ac are unrelated VM setups. 2.4.5-ac is probably much better
> for a lot of loads as it has the 2.4.5 general behaviour but gets the aging
> a lot more sane

 OK. Maybe my wording was implying a nonexisting relation. In any case,
the behaviour I described started with some of the later (-ac12)
versions of 2.4.4-ac and continues with 2.4.5-ac.

 On a side question: does Linux support swap-files in addition to
sawp-partitions? Even if that has a performance penalty, when the system
is swapping performance is dead anyway.

Martin
-- 
------------------------------------------------------------------
Martin Knoblauch         |    email:  Martin.Knoblauch@TeraPort.de
TeraPort GmbH            |    Phone:  +49-89-510857-309
C+ITS                    |    Fax:    +49-89-510857-111
http://www.teraport.de   |    Mobile: +49-170-4904759

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Card for Martin.Knoblauch --]
[-- Type: text/x-vcard; charset=us-ascii; name="Martin.Knoblauch.vcf", Size: 375 bytes --]

begin:vcard 
n:Knoblauch;Martin
tel;cell:+49-170-4904759
tel;fax:+49-89-510857-111
tel;work:+49-89-510857-309
x-mozilla-html:FALSE
url:http://www.teraport.de
org:TeraPort GmbH;C+ITS
adr:;;Garmischer Straße 4;München;Bayern;D-80339;Germany
version:2.1
email;internet:Martin.Knoblauch@TeraPort.de
title:Senior System Engineer
x-mozilla-cpt:;-7008
fn:Martin Knoblauch
end:vcard

Re: 2.4.5 VM

[Christian Bornträger]

>  On a side question: does Linux support swap-files in addition to
> sawp-partitions? Even if that has a performance penalty, when the system
> is swapping performance is dead anyway.


Yes.
A possible solution could be:

> dd if=/dev/zero of=/swap bs=1M count=<whatever you like in MB>
> mkswap /swap
> swapon /swap

Works fine for me.

Re: 2.4.5 VM

[Alan Cox]

>  On a side question: does Linux support swap-files in addition to
> sawp-partitions? Even if that has a performance penalty, when the system

since before 1.0 I believe 8)

Re: 2.4.5 VM

[Jonathan Morton]

> On a side question: does Linux support swap-files in addition to
>sawp-partitions? Even if that has a performance penalty, when the system
>is swapping performance is dead anyway.

Yes.  Simply use mkswap and swapon/off on a regular file instead of a
partition device.  I don't notice any significant performance penalty (a
swapfile on a SCSI disk is faster than a swap-partition on an IDE disk),
although you'd be advised to attempt to keep the file unfragmented.

--------------------------------------------------------------
from:     Jonathan "Chromatix" Morton
mail:     chromi@cyberspace.org  (not for attachments)
big-mail: chromatix@penguinpowered.com
uni-mail: j.d.morton@lancaster.ac.uk

The key to knowledge is not to rely on people to teach you it.

Get VNC Server for Macintosh from http://www.chromatix.uklinux.net/vnc/

-----BEGIN GEEK CODE BLOCK-----
Version 3.12
GCS$/E/S dpu(!) s:- a20 C+++ UL++ P L+++ E W+ N- o? K? w--- O-- M++$ V? PS
PE- Y+ PGP++ t- 5- X- R !tv b++ DI+++ D G e+ h+ r++ y+(*)
-----END GEEK CODE BLOCK-----

Re: 2.4.5 VM

[Miquel Colom Piza]

This is my first email to the list. I'm not subscribed but I've read it
for years.

I don't agree with those claiming that 2.4.xx is bad or still beta.

We the administrators have the responsability to test early kernels and
send  good bug reports so the developers can solve the bugs. That's the
way we can contribute to the community.

But it's really risky to use these kernels on MAIN 24x7  production
servers.

This has been true for 1.2.x  2.0.x  (I think that was the best linux
kernel series) 2.2.x and 2.4.x and will be for 2.6.x also

Given we know that the support  from open source developers is clearly
better than commercial contract supports, I don't see the reason to
complain about the work of those wonderfull hackers spending their spare
time coding for all of us.

(I'm not subscribed to the list, Please CC me).

Re: 2.4.5 VM

[Ken Brownfield]

I'd be forced to agree.  I have 2.4.x in limited production, and with 
the exception of the HP/APIC fatal issues that have a "noapic" 
work-around, I have had no problem at all with any of the 2.4.x kernels 
I've used.

Open software by definition will never reach the kind of monolithic 
stability that years of code freeze requires.  Linux (especially 2.4.x) 
offers too much in return, and I can always run a 2.2.x kernel.  I would 
say that the stability of the kernel has been *above* my expectations, 
frankly, considering all that's changed.

It's definitely our responsibility as admins to test these kernels.  I 
was running 2.4.0-test1 the second it was released, and the one problem 
I've found has been reported and investigated (it's apparently a tough 
one).

As far as VM, I've never had the severe issues that some are reporting.  
This doesn't mean it's not a problem, but it definitely indicates that 
it's not a global showstopper.  For VM-intense applications, I roll out 
a 2.2.19 kernel as a preventative measure while I wait for the VM code 
to be tweaked.  I guess I would have expected these complaints during 
the -test phase.  Not to mention that the distributions seem to have 
rolled out 2.4.x just fine.

To wit:

box1 1 ~> uptime
  10:27am  up 168 days,  2:43,  3 users,  load average: 2.45, 2.30, 2.32
box1 2 ~> uname -a
Linux box1.mynetwork.tld 2.4.0-test6 #1 SMP Sat Aug 19 04:26:58 PDT 2000 
i686 unknown

Not true production, but totally representative of my experiences FWIW.  
IMHO, YMMV, etc.
--
Ken.

On Friday, June 1, 2001, at 10:15 AM, Miquel Colom Piza wrote:

> This is my first email to the list. I'm not subscribed but I've read it
> for years.
>
> I don't agree with those claiming that 2.4.xx is bad or still beta.
>
> We the administrators have the responsability to test early kernels and
> send  good bug reports so the developers can solve the bugs. That's the
> way we can contribute to the community.
>
> But it's really risky to use these kernels on MAIN 24x7  production
> servers.
>
> This has been true for 1.2.x  2.0.x  (I think that was the best linux
> kernel series) 2.2.x and 2.4.x and will be for 2.6.x also
>
> Given we know that the support  from open source developers is clearly
> better than commercial contract supports, I don't see the reason to
> complain about the work of those wonderfull hackers spending their spare
> time coding for all of us.
>
> (I'm not subscribed to the list, Please CC me).
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" 
> in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: 2.4.5 VM

[Marcin Kowalski]

Relating to Huge Dcache and InodeCache Entries and < 2xMem Swap.
I have a server with > 1.1gig of RAM which I have limited to 1gig (due to
stability - BUG at HIGHMEM.c: 155 crashes)...

The size of the Swap space is 620mb... my memory usage is below :

Mem:        900008     892748       7260          0      14796     171796
-/+ buffers/cache:     706156     193852
Swap:       641016       1792     639224

The extract out of slabinfo is:
inode_cache       903876 930840    480 116355 116355    1 :  124   62
bdev_cache          1160   1357     64   23   23    1 :  252  126
sigqueue             261    261    132    9    9    1 :  252  126
dentry_cache      775520 934560    128 31152 31152    1 :  252  126

As you can see the usage is crazy....bearing in mind a number of things.
The system is running hardware raid5, reiserfs and massive amount of files
> 500000 in >2000 directories..

It is a 2x933mhz PIII + 1.1gig of ram NEtservers

What I really want to know is DOES THIS REALLY Matter. IE If memory is
needed by and application are these entries simply purged then.. In which
case there is no problem????

Regards
and THanks
MarCin


 -------------------------------------
#    Marcin Kowalski                # 
      On_Linux Developer.
       ->Datrix Solutions.<-
	
	Tel. 770-6146
#	Cel. 082-400-7603           #
-------------------------------------

Re: 2.4.5 VM

[Marcelo Tosatti]

On Fri, 1 Jun 2001, Marcin Kowalski wrote:

> Relating to Huge Dcache and InodeCache Entries and < 2xMem Swap.
> I have a server with > 1.1gig of RAM which I have limited to 1gig (due to
> stability - BUG at HIGHMEM.c: 155 crashes)...
> 
> The size of the Swap space is 620mb... my memory usage is below :
> 
> Mem:        900008     892748       7260          0      14796     171796
> -/+ buffers/cache:     706156     193852
> Swap:       641016       1792     639224
> 
> The extract out of slabinfo is:
> inode_cache       903876 930840    480 116355 116355    1 :  124   62
> bdev_cache          1160   1357     64   23   23    1 :  252  126
> sigqueue             261    261    132    9    9    1 :  252  126
> dentry_cache      775520 934560    128 31152 31152    1 :  252  126
> 
> As you can see the usage is crazy....bearing in mind a number of things.
> The system is running hardware raid5, reiserfs and massive amount of files
> > 500000 in >2000 directories..
> 
> It is a 2x933mhz PIII + 1.1gig of ram NEtservers
> 
> What I really want to know is DOES THIS REALLY Matter. IE If memory is
> needed by and application are these entries simply purged then.. In which
> case there is no problem????

Could you check if they are purged heavily enough for you case when you
start a big app? 

2.4.5 VM

[Trever L. Adams]

In my opinion 2.4.x is NOT ready for primetime.  The VM has been getting 
worse since 2.4.0, I believe.  Definitely since and including 2.4.3.  I 
cannot even edit a few images in gimp where the entire working set used 
to fit entirely in memory.  The system now locks in some loop (SAK still 
works).

FILE CACHING IS BROKEN.  I don't care who says what, by the time swap is 
half filled, it is time to start throwing away simple caches.  Not wait 
until there is no more memory free and then lock in an infinite loop.

My system has 128 Meg of Swap and RAM.

Trever Adams

Re: 2.4.5 VM

[Alan Cox]

> My system has 128 Meg of Swap and RAM.

Linus 2.4.0 notes are quite clear that you need at least twice RAM of swap
with 2.4.

Marcelo is working to change that but right now you are running something 
explicitly explained as not going to work as you want

Re: 2.4.5 VM

[Trever L. Adams]

Alan Cox wrote:

>>My system has 128 Meg of Swap and RAM.
>>
> 
> Linus 2.4.0 notes are quite clear that you need at least twice RAM of swap
> with 2.4.
> 
> Marcelo is working to change that but right now you are running something 
> explicitly explained as not going to work as you want
> 
> 

Alan,

Actually I have tried 1x,2x,3x.  In 2.4.0 to 2.4.3 I had some issues but 
never a system freeze of any kind.  With 2.4.4 I had more problems, but 
I was ok.  2.4.5 I now have these freezes.  Maybe I should go back to 
2x, but I still find this behavior crazy.

This still doesn't negate the point of freeing simple caches.

Trever Adams

Re: 2.4.5 VM

[Alan Cox]

> Actually I have tried 1x,2x,3x.  In 2.4.0 to 2.4.3 I had some issues but 
> never a system freeze of any kind.  With 2.4.4 I had more problems, but 
> I was ok.  2.4.5 I now have these freezes.  Maybe I should go back to 
> 2x, but I still find this behavior crazy.
> This still doesn't negate the point of freeing simple caches.

The caches are in part shared. Remember page cache memory and read only
application pages are the same thing - so its not that simple. I found 2.4.5
pretty bad. 2.4.5-ac seems to be better on the whole but I know its definitely
not right yet. Marcelo and Rik are working on that more and more.

Marcelo has a test patch to fix the (documented but annoying) 2x memory
swap rule stuff. The balancing problem is harder but being worked on.

If you can give Rik a summary of your config/what apps run/ps data then it
may be valuable as he can duplicate your exact setup for testing his
vm changes and add it to the test sets.

Alan

Re: 2.4.5 VM

[Billy Harvey]

 > > Actually I have tried 1x,2x,3x.  In 2.4.0 to 2.4.3 I had some issues but 
 > > never a system freeze of any kind.  With 2.4.4 I had more problems, but 
 > > I was ok.  2.4.5 I now have these freezes.  Maybe I should go back to 
 > > 2x, but I still find this behavior crazy.
 > > This still doesn't negate the point of freeing simple caches.
 > 
 > The caches are in part shared. Remember page cache memory and read only
 > application pages are the same thing - so its not that simple. I found 2.4.5
 > pretty bad. 2.4.5-ac seems to be better on the whole but I know its definitely
 > not right yet. Marcelo and Rik are working on that more and more.
 > 
 > Marcelo has a test patch to fix the (documented but annoying) 2x memory
 > swap rule stuff. The balancing problem is harder but being worked on.
 > 
 > If you can give Rik a summary of your config/what apps run/ps data then it
 > may be valuable as he can duplicate your exact setup for testing his
 > vm changes and add it to the test sets.
 > 
 > Alan

As a data point, running 2.4.5-ac1 on a Thinkpad with 512M ram and 1G
swap partition, I used the imagemagick programs to manipulate a couple
of dozen large images that would create intermediate images a little
over 1G in size.

This would totally freeze the system for interactive use.  Left alone,
the system was stable, nothing was killed by the system, and
everything recovered fine, however once the swap started, nothing else
could be initiated or even monitored.  Even the xclock would freeze
solid for several minutes during the program run.  Running the same
thing under 'nice' seemed to make no difference.

Running X, xemacs, etc. in my standard configuraton, top looks at the
moment like (using -ac5 now) during 'idle':

 00:17:13 up 52 min,  6 users,  load average: 2.00, 2.06, 1.98
74 processes: 70 sleeping, 4 running, 0 zombie, 0 stopped
CPU states:   4.0% user,   2.2% system,  93.8% nice,   0.0% idle
Mem:    512728K total,   436120K used,    76608K free,   234380K buffers
Swap:   982792K total,     3540K used,   979252K free,    92332K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
  638 weh       20  19  4344 4340   816 R N  47.2  0.8  23:16 fahclient
  614 weh       19  19 15360  15M   644 R N  47.0  2.9  23:49 setiathome
  531 root       5 -10 22948 4760  2040 S <   1.5  0.9   0:11 XFree86
  582 weh        9   0 43080  42M  3400 S     1.5  8.4   0:20 xemacs
  603 weh       11   0  1072 1072   776 R     0.5  0.2   0:19 top
  588 weh        9   0  4472 4472  1876 S     0.3  0.8   0:07 wish
    1 root       8   0   120   80    68 S     0.0  0.0   0:01 init
...

Billy

Re: 2.4.5 VM

[Marcin Kowalski]

Hi

I found this post of interest. I have 1.1 Gig of RAM but only 800mb of
Swap as I expect NOT to use that much memory... Could this be the cause of
the machines VERY erratic behaviour??? Kernel Panics, HUGE INOde and
Dcache.... ??

Regards
MarCin


-------------------------------------
#    Marcin Kowalski                # 
      On_Linux Developer.
       ->Datrix Solutions.<-
	
	Tel. 770-6146
#	Cel. 082-400-7603           #
-------------------------------------

On Thu, 31 May 2001 alan@lxorguk.ukuu.org.uk wrote:

> > My system has 128 Meg of Swap and RAM.
> 
> Linus 2.4.0 notes are quite clear that you need at least twice RAM of swap
> with 2.4.
> 
> Marcelo is working to change that but right now you are running something 
> explicitly explained as not going to work as you want
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: 2.4.5 VM

[David Rees]

I don't know myself, (it sounds like other bigmem problems), but setting up a
2GB swap file is easy enough to test.  :-)

-Dave

On Fri, Jun 01, 2001 at 10:29:39AM +0200, Marcin Kowalski wrote:
> 
> I found this post of interest. I have 1.1 Gig of RAM but only 800mb of
> Swap as I expect NOT to use that much memory... Could this be the cause of
> the machines VERY erratic behaviour??? Kernel Panics, HUGE INOde and
> Dcache.... ??
> 
> On Thu, 31 May 2001 alan@lxorguk.ukuu.org.uk wrote:
> 
> > > My system has 128 Meg of Swap and RAM.
> > 
> > Linus 2.4.0 notes are quite clear that you need at least twice RAM of swap
> > with 2.4.
> > 
> > Marcelo is working to change that but right now you are running something 
> > explicitly explained as not going to work as you want

Re: 2.4.5 VM

[Russell Leighton]

I have a 2.4.5-ac3 box with 1G RAM and 2.6G Swap....first time
developers hit apache/php/zendcache after reboot and it swapped to a stop.

I stop/restarted apache and it seems very happy...can't goto production like this tho :(

Alan Cox wrote:

> > My system has 128 Meg of Swap and RAM.
>
> Linus 2.4.0 notes are quite clear that you need at least twice RAM of swap
> with 2.4.
>
> Marcelo is working to change that but right now you are running something
> explicitly explained as not going to work as you want
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

--
---------------------------------------------------
Russell Leighton    russell.leighton@247media.com

Programming today is a race between software
engineers striving to build bigger and better
idiot-proof programs, and the Universe trying to
produce bigger and better idiots. So far, the
Universe is winning. - Rich Cook
---------------------------------------------------

Re: 2.4.5 VM

[Christopher Zimmerman]

"Trever L. Adams" wrote:

> In my opinion 2.4.x is NOT ready for primetime.  The VM has been getting
> worse since 2.4.0, I believe.  Definitely since and including 2.4.3.  I
> cannot even edit a few images in gimp where the entire working set used
> to fit entirely in memory.  The system now locks in some loop (SAK still
> works).
>
> FILE CACHING IS BROKEN.  I don't care who says what, by the time swap is
> half filled, it is time to start throwing away simple caches.  Not wait
> until there is no more memory free and then lock in an infinite loop.
>
> My system has 128 Meg of Swap and RAM.
>
> Trever Adams
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

I've found that with the latest kernel release (2.4.5) VM performance has
been greatly improved.  kswapd and bdflush no longer use 200% of my cpu
cycles when simply doing a dd bs=1024 count=8388608 if=/dev/zero
of=test.file.  All of my test systems remain responsive with about 180% cpu
available.  These systems are running software RAID and 3ware IDE raid with
2GB of memory and 4GB swap.  Have you tried 2.4.5?

-zim

Christopher Zimmerman
AltaVista Company

Re: 2.4.5 VM

[Trever L. Adams]

Christopher Zimmerman wrote:

> 
> I've found that with the latest kernel release (2.4.5) VM performance has
> been greatly improved.  kswapd and bdflush no longer use 200% of my cpu
> cycles when simply doing a dd bs=1024 count=8388608 if=/dev/zero
> of=test.file.  All of my test systems remain responsive with about 180% cpu
> available.  These systems are running software RAID and 3ware IDE raid with
> 2GB of memory and 4GB swap.  Have you tried 2.4.5?
> 
> -zim
> 
> Christopher Zimmerman
> AltaVista Company
> 


I have found that 2.4.5 is great, until it decides to stop freeing unused pages

(simple file cache).  Then it goes to hell in a handbasket at the speed of light.


Yes, I have tried it, that is what I wrote about.

Trever

Re: 2.4.5 VM

[Christopher Zimmerman]

Christopher Zimmerman wrote:

> "Trever L. Adams" wrote:
>
> > In my opinion 2.4.x is NOT ready for primetime.  The VM has been getting
> > worse since 2.4.0, I believe.  Definitely since and including 2.4.3.  I
> > cannot even edit a few images in gimp where the entire working set used
> > to fit entirely in memory.  The system now locks in some loop (SAK still
> > works).
> >
> > FILE CACHING IS BROKEN.  I don't care who says what, by the time swap is
> > half filled, it is time to start throwing away simple caches.  Not wait
> > until there is no more memory free and then lock in an infinite loop.
> >
> > My system has 128 Meg of Swap and RAM.
> >
> > Trever Adams
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at  http://www.tux.org/lkml/
>
> I've found that with the latest kernel release (2.4.5) VM performance has
> been greatly improved.  kswapd and bdflush no longer use 200% of my cpu
> cycles when simply doing a dd bs=1024 count=8388608 if=/dev/zero
> of=test.file.  All of my test systems remain responsive with about 180% cpu
> available.  These systems are running software RAID and 3ware IDE raid with
> 2GB of memory and 4GB swap.  Have you tried 2.4.5?
>
> -zim
>
> Christopher Zimmerman
> AltaVista Company
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: 2.4.5 VM

[Christopher Zimmerman]

Actually I take everything back.  I've been testing on linux-2.4.5-xfs and seen
major improvements.

-zim

Christopher Zimmerman wrote:

> Christopher Zimmerman wrote:
>
> > "Trever L. Adams" wrote:
> >
> > > In my opinion 2.4.x is NOT ready for primetime.  The VM has been getting
> > > worse since 2.4.0, I believe.  Definitely since and including 2.4.3.  I
> > > cannot even edit a few images in gimp where the entire working set used
> > > to fit entirely in memory.  The system now locks in some loop (SAK still
> > > works).
> > >
> > > FILE CACHING IS BROKEN.  I don't care who says what, by the time swap is
> > > half filled, it is time to start throwing away simple caches.  Not wait
> > > until there is no more memory free and then lock in an infinite loop.
> > >
> > > My system has 128 Meg of Swap and RAM.
> > >
> > > Trever Adams
> > >
> > > -
> > > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > > the body of a message to majordomo@vger.kernel.org
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > > Please read the FAQ at  http://www.tux.org/lkml/
> >
> > I've found that with the latest kernel release (2.4.5) VM performance has
> > been greatly improved.  kswapd and bdflush no longer use 200% of my cpu
> > cycles when simply doing a dd bs=1024 count=8388608 if=/dev/zero
> > of=test.file.  All of my test systems remain responsive with about 180% cpu
> > available.  These systems are running software RAID and 3ware IDE raid with
> > 2GB of memory and 4GB swap.  Have you tried 2.4.5?
> >
> > -zim
> >
> > Christopher Zimmerman
> > AltaVista Company
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at  http://www.tux.org/lkml/