Greetings all,
I performed the following exercise with CML 1.2.4 and kernel 2.4.4-pre6:
Using a clean separate 2.4.4-pre6 tree, I ran CML2 1.2.4 make xconfig
starting with a .config from a CML1 2.4.4-pre6 tree.
I saved the configuration without changing anything, and then compared
the original .config from the CML1 tree to the config.out file in the CML2 tree.
Here are the additional options set to y by the CML2 1.2.4 configurator:
CONFIG_CPU_32=y
CONFIG_CPU_HAS_LLSC=y
CONFIG_DRAM_VIRTUAL_BASE=y
CONFIG_GSC_DINO=y
CONFIG_MATHEMU=y
CONFIG_SA1100_BRUTUS=y
It seems clear from reading the entries in Configure.help that
MATHEMU and SA1100_BRUTUS are mistakes.
I don't actually see any symbol CONFIG_DRAM_VIRTUAL_BASE, but
there is a CONFIG_ETRAX_DRAM_VIRTUAL_BASE in 2.4.3-ac14,
so this would also seem to be a mistake.
CONFIG_GSC_DINO although setable, doesn't have any help, and seemed to have
gone the way of the DINOsaurs in 2.4.3-ac14, until I found that it moved
from arch/parisc/config.in to arch/parisc/config.common. Hmm, sneaky those
parisc folks. In any case, it looks like this option is parisc-specific, so having
it set is not necessary.
Don't know about CONFIG_CPU_32 and CONFIG_CPU_HAS_LLSC.
Steven
_______________________________________________
kbuild-devel mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/kbuild-devel
