commit 025a451a927564b086664c8b4379c2ea698af0b6
author Madhukar Pappireddy <madhukar.pappireddy@arm.com> Mon Oct 14 22:09:19 2024 -0500
committer Madhukar Pappireddy <madhukar.pappireddy@arm.com> Tue Nov 05 17:17:59 2024 -0600
tree 9736d62f7cd8b1139646c20c04ee9227780bce85
parent c3fd9756a53ecd13af27b529ed24dca6003cda1f

fix: simplify secure interrupt handling

Now that hafnium supports queueing of virtual interrupts for each
execution context of an SP, there is no need to disallow high priority
interrupts from preempting a vCPU while it is handling a virtual
interrupt.

Hence, remove the actions of masking and unmasking physical interrupts
before/after vCPU is run.

Arm recommends the priority for G1 non-secure interrupts to be in the
range 0x80 to 0xff when programming from secure world. Currently, the
SRI priority is incorrect. This patch also fixes the priority value.

Change-Id: Icdc36f3aa09fab83cb58214e14e5a9bd174bc28b
Signed-off-by: Madhukar Pappireddy <madhukar.pappireddy@arm.com>

src/api.c

diff --git a/src/api.c b/src/api.c
index db7c538..79278ce 100644
--- a/src/api.c
+++ b/src/api.c
@@ -3176,15 +3176,10 @@
 	if (api_ffa_is_managed_exit_ongoing(current_locked)) {
 		struct interrupts *interrupts = &current->interrupts;
 
-		/*
-		 * Per FF-A v1.1 EAC0 section 8.3.1.2.1 rule 6, SPMC can signal
-		 * a secure interrupt to a SP that is performing managed exit.
-		 * We have taken a implementation defined choice to not allow
-		 * Managed exit while a SP is processing a secure interrupt.
-		 */
 		CHECK(current->scheduling_mode != SPMC_MODE);
 
-		plat_interrupts_set_priority_mask(current->priority_mask);
+		plat_interrupts_set_priority_mask(
+			current->prev_interrupt_priority);
 		/*
 		 * A SP may be signaled a managed exit but actually not trap
 		 * the virtual interrupt, probably because it has virtual