load average fix for lasat boards (timer interrupt handling fix)
PATCH 1/1 load average fix for lasat boards (timer interrupt handling fix)
Hi Florian, Ralf,
it seems the method of calling the new time code changed
or it was wrong all along, This patch fixes the problem of the ksoftirqd running all the time on Lasat systems, can you apply Ralf?
/Brian
Index: arch/mips/lasat/setup.c
===================================================================
RCS file: /cvs/linux/arch/mips/lasat/setup.c,v
retrieving revision 1.1.2.5
mips/lasat/setup.c +2 -6
@@ -109,14 +109,10 @@change_cp0_status (ST0_IM ,IE_IRQ0 |IE_IRQ5 ); } +#define MIPS_CPU_TIMER_IRQ 7 asmlinkage void lasat_timer_interrupt(struct pt_regs *regs) { -int cpu =smp_processor_id (); - -timer_interrupt (7 ,NULL , regs); - -if (softirq_pending (cpu)) -do_softirq (); +ll_timer_interrupt (MIPS_CPU_TIMER_IRQ , regs); }void __init lasat_setup (void )
Quoted 5 lines · Brian Murphy
Hi Florian, Ralf,
it seems the method of calling the new time code changed
or it was wrong all along, This patch fixes the problem of the ksoftirqd running all the time on Lasat systems, can you apply Ralf?
Works like a charm - Ralf already applied and i am running the CVS from an hour after that happily ...
Flo
Signature · 2 lines
Florian Lohoff [email protected] +49-5201-669912 Heisenberg may have been here.