Michael, > Not sure what's miss match FWR, but if you can't boot > the need to specify /proc/bus/pci/ is going to be the least of your problems.
Are you sure that we will never face a situation were mthca driver is not loaded and we need to burn a new HCA FWR ? What we expect the user to do in this case ? Send the HCA to Mellanox ? > > Please understand, I'd like people to start using mthca0 for device name rather than defaulting to lspci as the first resort. I think that having " mstflint -d mthca0 ... " is really good and user friendly. BUT please notice that , Plenty of the mstflint uses are done when a customer buy / install new IB lab equipment. New users that does not know IB yet knows lspci . Lspci is easy to find info and very convenient for scripts writing. So , Can you explain what's wrong with " mstflint -d <lspci info> .." and why you don't want user to use it ? Moshe ____________________________________________________________ Moshe Katzir | +972-9971-8639 (o) | +972-52-860-6042 (m) Voltaire - The Grid Backbone www.voltaire.com -----Original Message----- From: Michael S. Tsirkin [mailto:[EMAIL PROTECTED] Sent: Tuesday, October 03, 2006 9:45 AM To: Moshe Kazir Cc: Tseng-Hui (Frank) Lin; [EMAIL PROTECTED]; [email protected] Subject: Re: FW: Mstflint - not working on ppc64 andwhendriver is notloaded on AMD Quoting r. Moshe Kazir <[EMAIL PROTECTED]>: > A work-around that enable the use of mstflint only when the driver is > loaded is not sufficient I think I somewhat understand the mmap related kernel bug thing, (although I'd like to see this discussed on lkml) but I still don't understand where the "driver is loaded" thing comes from, and I'd like to. > What you plan to do when you have system error -> > - boot fail when IB started, > - driver loading fail as result of driver error / miss match FWR, etc. Not sure what's miss match FWR, but if you can't boot the need to specify /proc/bus/pci/ is going to be the least of your problems. Please understand, I'd like people to start using mthca0 for device name rather than defaulting to lspci as the first resort. -- MST _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
