On RISC-V, there's no possibility to parameterise IPIs resp. enrich them
with some userdata. Great. We need to do it in software...
We differentiate between two IPI types in software: IPI_CAUSE_MGMT, and
IPI_CAUSE_GUEST.
When a guest sends an IPI to a HART within the cell, the HV will trap,
and we forward the IPI call to the SBI. Before, we set the cause to
IPI_CAUSE_GUEST.
Follow the same logic for management IPIs accordingly.
Signed-off-by: Ralf Ramsauer <ralf.ramsa...@oth-regensburg.de>
---
hypervisor/arch/riscv/control.c | 33 +++++++++++++++++++++++++++++++++
1 file changed, 33 insertions(+)
diff --git a/hypervisor/arch/riscv/control.c b/hypervisor/arch/riscv/control.c
index b848994b..7d4c31be 100644
--- a/hypervisor/arch/riscv/control.c
+++ b/hypervisor/arch/riscv/control.c
@@ -11,6 +11,8 @@
*/
#include <jailhouse/control.h>
+#include <jailhouse/printk.h>
+#include <asm/sbi.h>
int arch_cell_create(struct cell *cell)
{
@@ -68,4 +70,35 @@ void arch_park_cpu(unsigned int cpu_id)
void arch_send_event(struct public_per_cpu *target_data)
{
+ struct sbiret result;
+
+ /*
+ * For sending an event to a remote HART, we need two conditions to be
+ * set:
+ * 1. We have to grab the CPU's control lock. Having the control lock
+ * ensures that the CPU is currently not processing any other IPI.
+ * If the CPU is processing an IPI, then we have to busy wait for
+ * completion.
+ * 2. target_data->ipi_cause must be NONE at the moment of grabbing
+ * the lock. This ensures that no other CPU enqueued an IPI
+ * before.
+ */
+retry:
+ spin_lock(&target_data->control_lock);
+ if (target_data->ipi_cause != IPI_CAUSE_NONE) {
+ spin_unlock(&target_data->control_lock);
+ goto retry;
+ }
+
+ target_data->ipi_cause = IPI_CAUSE_MGMT;
+ memory_barrier();
+ spin_unlock(&target_data->control_lock);
+
+ result = sbi_send_ipi(1UL << (target_data->phys_id % BITS_PER_LONG),
+ target_data->phys_id / BITS_PER_LONG);
+ if (result.error != SBI_SUCCESS) {
+ printk("IPI send to HART %lu failed: %ld\n",
+ target_data->phys_id, result.error);
+ panic_stop();
+ }
}
--
2.40.1
--
You received this message because you are subscribed to the Google Groups
"Jailhouse" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to jailhouse-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/jailhouse-dev/20230519204033.643200-42-ralf.ramsauer%40oth-regensburg.de.
