linux-kernel - Re: files/process scaling problem? (was: [PATCH] Export shmem_file_setup and shmem

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <1217809152.23437.401.camel@koto.keithp.com>
Date:	Sun, 03 Aug 2008 17:19:12 -0700
From:	Keith Packard <keithp@...thp.com>
To:	Ingo Oeser <ioe-lkml@...eria.de>
Cc:	keithp@...thp.com, John Stoffel <john@...ffel.org>,
	Christoph Hellwig <hch@...radead.org>,
	Eric Anholt <eric@...olt.net>, linux-kernel@...r.kernel.org
Subject: Re: files/process scaling problem? (was: [PATCH] Export
	shmem_file_setup and shmem_getpage for DRM-GEM)

On Mon, 2008-08-04 at 01:35 +0200, Ingo Oeser wrote:

> What actions on many thousand fds are supported poorly or not at all?
> Are you only concerned about the memory requirements?

I didn't notice that the change in the maximum number of fds per process
from 1024 to 1024*1024 back in February. That makes it possible,
although requiring root privs, to allocate enough fds for this to work.

The other issue is that several important applications (including the X
server) use select instead of poll, and they have a small maximum number
of fds that they support. It seems like this could be worked around by
dup2'ing the shmem fds up a ways.

> Please elaborate or point me to a place where you did already :-)

I should have looked at fs/file.c.

-- 
keith.packard@...el.com

Download attachment "signature.asc" of type "application/pgp-signature" (190 bytes)