On Fri, Apr 03, 2009 at 11:23:59AM +0100, Daniel P. Berrange wrote:
On Fri, Apr 03, 2009 at 12:04:05AM +0200, Gerrit Slomma wrote:
The virsh-command freecell hands out bytes but affixes those with kB. The error ist found in virsh.c on line 1663 and following. I have corrected this and altered the output to the method i chose for virt-manager - in the days of 96 GB per socket (Nehalem-EP) no one cares even about a fraction of a Megabyte.
For command line tools like virsh I prefer to have it consistently report in the same units, so if someone wants to script it from the shell they don't have to concern themselves with changing units.
All the other virsh commands report in KB, so the simple fix is to just divide by 1024.
Agreed, let's not make it too hard for tools who would parse virsh output, even though we should not encourage that practice :-)
In checking this I discovered a whole bunch of other fun bugs in the NUMA support :-) The QEMU impl was not returning the correct return code - it used -1 instead of 0. The QEMU impl was also not setting an error if the requested cell was out of range. The libvirtd remote driver was not correctly seeing return value of -1 due to casting it to an unsigned int. virsh was not checking return values correctly either.
Heh !
So I propose the following patch....
ACK, fine by me ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel@veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/