Skip to contentSkip to conversation

Some questions about kernel behavior towards INIT interrupt.

DISCUSSION Francois Wellenreiter 1 participant 1 message Mar 15, 2004

lore
Share & tools
lore
#0 lore raw

Hi all,

I'd like to submit a problem I met using ITP tool on an SMP Itanium2
machine with kernel 2.6.4.
I've noticed that SAL called "ia64_monarch_init_handler" function for EACH processor when generating an INIT interrupt.

In fact, considering the registration done to the SAL, this function may be called only once for monarch processor and the function "ia64_init_slave_stack" should be called for the others (using the rendez-vous mechanism).

Am I wrong ? If I'm right, then had someone already encountered this behavior ?
Thanks for help,

Francois WELLENREITER