[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: kernel package



On 05/14/2007 05:03 PM, Jay Estabrook wrote:
Oliver Falk wrote:
On 05/11/2007 02:01 PM, David Woodhouse wrote:

Once upon a time the folks at HP were threatening to send me SMP Alpha
boxen -- not sure if I could still pull that off...
Anyone from HP here? Jay, do you have any good contacts?

I've been trying, but unfortunately, no one has stepped forward... :-(

I'll keep trying, but don't hold your breath.

OK. Good to know you're on it.

+Source50: kernel-%{kversion}-alpha.config
+Source50: kernel-%{kversion}-alpha-smp.config
I don't have a smp config at the moment. So we need to put this out,
else up will no build clean.

SMP config is really just UP config with SMP turned on; another option,
NR_CPUS should be in the range 2-32, others should default OK, unless
things have changed drastically. I've been building against the latest
FC5 kernel, ie 2.6.20-1.2316, and haven't had anything unusual turn up.

CONFIG_SMP=y
CONFIG_NR_CPUS=16

Is what I saw recently... Will put a up and smp config in my tree now. If it builds fine, someone will be able to test it - I'm sure :-)

 # Exec shield
+%ifnarch alpha alphaev5 alphaev56 alphaev6 alphaev6
 %patch810 -p1
+%endif
Even more compile errors. :-)

One reason we seem to have to put in the multiple Alpha
arches is, that if ":--target=..." is set for some reason,
without all the arches listed, build will fail. If there
were some way to override and reset to base-arch==alpha if
one of the others shows up would be fine. We really do NOT
want to build the GENERIC kernel for anything BUT the base
architecture, of course. Other more specific kernels, like
TITAN or LX164 or the like, may themselves ask "--mcpu=ev67"
or "--mcpu=ev56" or some such.

ok.

-of


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]