Hi, I have three machines that all run opensolaris. My intention is to compile the source of a kernel and to distribute the compilation over these machines to save time.
As I understand it, in dmake you have a host (where you start the compilation) and a number of build servers (where the compilation is distributed). My concern is this: The host is somewhat different than the build servers architecture-wise. Since I'm compiling kernel code I'm afraid that some parts of the kernel code will look at the machine it is currently compiling on and thus create a different result than if only the compilation would only be done on a single machine (the host). Do you follow me? All machines use the sunstudio compiler with onbld. However, the host uses b58 versions and the build servers uses b77. Can that be an issue when using dmake? I hope that dmake is not that rigid that all build numbers must math in order for it to work. Here is the hardware info I got (using /usr/sbin/prtdiag and uname -a on each machine): machine #1 (host) System Configuration: Sun Microsystems sun4u Sun Ultra 5/10 UPA/PCI (UltraSPARC-IIi 440MHz) System clock frequency: 110 MHz Memory size: 256 Megabytes -bash-3.00$ uname -a SunOS <removed name> 5.11 opensol-b58 sun4u sparc SUNW,Ultra-5_10 machine #2 and #3 (build servers) System Configuration: Sun Microsystems sun4u Netra T1 200 (UltraSPARC-IIe 500M Hz) System clock frequency: 100 MHz Memory size: 1024 Megabytes bash-3.2# uname -a SunOS <removed name> 5.11 snv_77 sun4u sparc SUNW,UltraAX-i2 Can anyone see if there might be a conflict - or does the fact that both all are detected as "sun4u" machines make this a non-issue? Regards, Mladen This message posted from opensolaris.org _______________________________________________ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
