Date: Mon, 28 Nov 2005 19:09:59 +0100 From: Claus Guttesen <kometen@gmail.com> To: Vivek Khera <vivek@khera.org> Cc: freebsd-amd64 List <freebsd-amd64@freebsd.org> Subject: increasing PMAP_SHPGPERPROC (was: FreeBSD 5.4 or 6.0 as postgresql-server) Message-ID: <b41c75520511281009j4ce6b29bh@mail.gmail.com>
next in thread | raw e-mail | index | archive | help
> > I'd be interested in your experiences w/6.0. Would you mind testing > > the onboard bge nics at the kind of loads that was causing them to > > puke under 5.4? I've got same system and as of yet not encountered > > issues w/ the onboard bge nics. The machine is also at the latest > > BIOS, however, so that may contribute as well. I just put our new db-server in production and so far everything works fine, including the two build-in bge-nics. When load hit the server the following showed up in /var/log/messages: kernel: collecting pv entries -- suggest increasing PMAP_SHPGPERPROC The default value is found in /usr/src/sys/amd64/conf/NOTES and says: # The value below is the one more than the default. # options PMAP_SHPGPERPROC=3D201 But I can't find any rule of thumb on what value to raise it to, 402, 512, just above it says: # Set the number of PV entries per process. Increasing this can # stop panics related to heavy use of shared memory. However, that can # (combined with large amounts of physical memory) cause panics at # boot time due the kernel running out of VM space. regards Claus
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?b41c75520511281009j4ce6b29bh>