I had already rebooted the guests after adding "mem=758m" to the zipl. I looked in /proc/cmdline to see if it took.
On 10/16/07 2:57 PM, "Bob Neill" <[EMAIL PROTECTED]> wrote: > I'll be the first to admit that I'm not a Linux expert, but I believe the > > DCSS doesn't have to be at the top of the guest's memory. As long as you > r > Linux guest can address the DCSS you should be OK. I'm guessing that add > ing > the mem=529m to your parameter file and not rebooting the linux guest m > ay be > the reason you couldn't read the DCSS. When you changed your guest's size > > and rebooted you picked up the mem=529m parameter. > > On Tue, 16 Oct 2007 10:33:47 -0500, RPN01 <[EMAIL PROTECTED]> wrote: > >> I hate it when I think of an answer my own questions as I type them, and > >> then hit send anyway. >> >> Changing the size of the guest from 512m to 528m and restarting solved t > he >> problem. The segment does indeed have to match the top of the guest¹s >> memory. >> >> -- >> .~. Robert P. Nix Mayo Foundation >> /V\ RO-OE-5-55 200 First Street SW >> / ( ) \ 507-284-0844 Rochester, MN 55905 >> ^^-^^ ----- >> "In theory, theory and practice are the same, but ³Join the story. > .. >> Ride Ural.² >> in practice, theory and practice are different." >> >> >> ------ Forwarded Message >> I¹m in the process of installing Omegamon XE for z/VM and Linux, and I > ¹ve >> gotten to the point of defining the DCSS to the linux guest serving as t > he >> monitor collection agent. >> >> I defined the Linux guest having 512m of memory. I defined the perfout d > css >> with ³DEFSEG 21000-210FF SN², which, I think, should be from 528M > 529M. I >> added ³mem=768m² to the zipl parameter line and reran mkinitrd and > zipl. >> >> I did the ³modprobe dcssblk² to install the driver. I then did ³ec > ho perfout >>> /sys/devices/dcssblk/add², and the response is as follows: >> >> tvmpzl01:/sys/devices/dcssblk # echo perfout > add >> -bash: echo: write error: Numerical result out of range >> >> Obviously, I¹ve missed something. Does the segment need to be immediat > ely >> adjacent to the top of the guest¹s virtual memory? Or can the definiti > on >> leave some space between the two, as above? What other problems can be >> encountered? The 2.6 device drivers manual documents the dcssblk driver, > but >> includes no indication of what errors can be encountered or what message > s >> might be generated. >> >> Any help, or nudges in the right direction, would be appreciated. >> >>
