Jens Axboe wrote:
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
[...]
+void arch_send_call_function_single_ipi(int cpu)
{
[...]
+void arch_send_call_function_ipi(cpumask_t mask)
{
[...]
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
+void arch_send_call_function_ipi(cpumask_t mask)
+{
[...]
+void arch_send_call_function_single_ipi(int cpu)
+{
+ xen_send_IPI_mask(cpumask_of_cpu(cpu), XEN_CALL_FUNCTION_SINGLE_VECTOR);
Xen vs native is a runtime choice, so you can't just statically define
arch_send_call_function(_single)_ipi for each. You need to define the
arch_* functions once, make them call via smp_ops, and call into the
respective native and Xen versions of the ipi functions.
Aside from that it looks good, with a very appealing code size reduction.
J
--
To unsubscribe from this list: send the line "unsubscribe linux-arch" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
