All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Seth, Rohit" <rohit.seth@intel.com>
To: linux-ia64@vger.kernel.org
Subject: RE: [Linux-ia64] Kernel Shared Memory Max (shmmax)
Date: Fri, 18 Oct 2002 16:24:55 +0000	[thread overview]
Message-ID: <marc-linux-ia64-105590709805196@msgid-missing> (raw)
In-Reply-To: <marc-linux-ia64-105590701905286@msgid-missing>

The restriction is coming out of default parameters set in Linux (file
include/linux/shm.h).
These values could be changed through /proc/sys/kernel/shm* interface.  And
if a site permanentlly wants to have higher values for this parameters then
that can be achieved by something like rc.local

Have not seen any distribution increasing the default values for IA-64 yet.

rohit


> -----Original Message-----
> From: miyoshi@hpc.bs1.fc.nec.co.jp 
> [mailto:miyoshi@hpc.bs1.fc.nec.co.jp] 
> Sent: Thursday, October 17, 2002 6:07 PM
> To: rohit.seth@intel.com; dcooper@atcc.necsys.com
> Cc: ia64-list@redhat.com; linux-ia64@linuxia64.org; 
> jfredin@atcc.necsys.com
> Subject: Re: [Linux-ia64] Kernel Shared Memory Max (shmmax)
> 
> 
> 
> Hi,
> 
> I suppose '8G' described below means default shmall, that is 
> 524288 * 16K. And I think there is no restrictions around 8G on IA-64.
> 
> Regards,
> 
> > > And based on that the total amount of shared memory 
> segment that can 
> > > be created in IA-64 platform (with 16K page size) is 8G.
> > 
> > Does this still hold true in newer kernel releases (since 
> 03/2002)?  
> > Is 8G the
> > max for IA64 kernel?
> > 
> > Thanks,
> > Donny
> > 
> > On Thu March 14 2002 5:13 pm, Seth, Rohit wrote:
> > > shmmax is the size of largest shared memory (in bytes) 
> segment that 
> > > the kernel allows.  System wide value of total shared 
> memory value 
> > > is given by shmall parameter.  Currently the default value for 
> > > shmmax is  (extrememly
> > > low) 32MB.  And based on that the total amount of shared 
> memory segment
> > > that can be created in IA-64 platform (with 16K page 
> size) is 8G. There is
> > > no direct relationship between shmmax value and the 
> amount of physical
> > > memory available in the system.  Though you will never be 
> able to create a
> > > shared memory segment bigger then the total memory (Free 
> RAM+Free SWAP)
> > > available on the machine (unless you change the 
> sysctl_overcommit_memory).
> > > Besides that, this low default value of 32MB just adds 
> overhead for big
> > > memory systems.  It will be nice to change this default 
> value to 2G at
> > > least for IA-64 machines.
> > >
> > > You can also change this value of shm* (shmmax etc.) by writing 
> > > proper values in /proc/sys/kernel/shm*
> > >
> > > rohit
> > >
> > > -----Original Message-----
> > > From: Donny Cooper [mailto:dcooper@atcc.necsys.com]
> > > Sent: Thursday, March 14, 2002 2:40 PM
> > > To: ia64-list@redhat.com; linux-ia64@linuxia64.org
> > > Subject: [Linux-ia64] Kernel Shared Memory Max (shmmax)
> > >
> > >
> > >
> > > Does the maximum possible shared memory (shmmax) increase 
> with the 
> > > Enterprise kernel?
> > >
> > > Below is from doc/sysctl/kernel.txt document.
> > > Is this document valid for IA-64 kernel?
> > >
> > > It seems you can set this value to pretty much anything 
> (even larger 
> > > than the available memory), so I'm assuming the value gets 
> > > intelligently truncated to the max supported.  How can I 
> check how 
> > > much shared mem is available, `free` or `top` just shows what's 
> > > used, right?
> > >
> > > ...doc/sysctl/kernel.txt 
> > > ===============================
> > >
> > > shmmax:
> > >
> > > This value can be used to query and set the run time limit on the 
> > > maximum shared memory segment size that can be created. Shared 
> > > memory segments up to 1Gb are now supported in the kernel.  This 
> > > value defaults to SHMMAX.
> > >
> > > ===============================
> > >
> > >
> > > Thanks,
> > > -----------------------------------
> > > Donny Cooper
> > > NEC Systems, Inc.
> > > Advanced Technical Computing Center
> > > dcooper@atcc.necsys.com
> > > -----------------------------------
> > >
> > >
> > > _______________________________________________
> > > Linux-IA64 mailing list
> > > Linux-IA64@linuxia64.org 
> > > http://lists.linuxia64.org/lists/listinfo/linux-ia64
> > >
> > >
> > >
> > > _______________________________________________
> > > ia64-list mailing list
> > > ia64-list@redhat.com
> > 
> > --
> > ================> > 	     Donny Cooper
> >       NEC Solutions (America), Inc.
> > Advanced Technical Computing Center
> > 	 ph: +1-281-465-1506
> >      Donny.Cooper@NECsam.com
> > ================> > 
> > 
> > 
> > _______________________________________________
> > ia64-list mailing list
> > ia64-list@redhat.com 
> > https://listman.redhat.com/mailman/listinfo/ia64-list
> > 
> 


  parent reply	other threads:[~2002-10-18 16:24 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-03-14 22:40 [Linux-ia64] Kernel Shared Memory Max (shmmax) Donny Cooper
2002-03-14 23:13 ` Seth, Rohit
2002-03-14 23:32 ` David Mosberger
2002-03-14 23:46 ` Seth, Rohit
2002-10-18  1:06 ` miyoshi
2002-10-18 16:24 ` Seth, Rohit [this message]
2002-10-21 16:42 ` Seth, Rohit

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=marc-linux-ia64-105590709805196@msgid-missing \
    --to=rohit.seth@intel.com \
    --cc=linux-ia64@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.