On Sun, 2023-03-26 at 12:20 +0700, Bui Quang Minh wrote:
+static void apic_mem_write(void *opaque, hwaddr addr, uint64_t val,
+ unsigned size)
+{
+ int index = (addr >> 4) & 0xff;
+
+ if (size < 4) {
+ return;
+ }
+
+ if (addr > 0xfff || !index) {
+ /* MSI and MMIO APIC are at the same memory location,
+ * but actually not on the global bus: MSI is on PCI bus
+ * APIC is connected directly to the CPU.
+ * Mapping them on the global bus happens to work because
+ * MSI registers are reserved in APIC MMIO and vice versa.
*/
+ MSIMessage msi = { .address = addr, .data = val };
+ apic_send_msi(&msi);
+ return;
+ }
I know you're just moving this bit around, but note that it means we
*can't* implement the 15-bit MSI trick as things stand, because those
extra 7 bits end up in bits 4-11 of the address, and that means the
'addr > 0xfff' check isn't correct any more.
However, that's only relevant in X2APIC mode... and there's no MMIO
access to registers in X2APIC mode. So the check could perhaps become
something like...
DeviceState *apic = cpu_get_current_apic();
if (!apic || is_x2apic_mode(apic) || addr > 0xfff || !index) {
/* MSI and MMIO APIC are at the same memory location,
* but actually not on the global bus: MSI is on PCI bus
* APIC is connected directly to the CPU.
* Mapping them on the global bus happens to work because
* MSI registers are reserved in xAPIC MMIO and vice versa.
* In X2APIC mode, there is no MMIO and bits 4-11 of the
* address *might* be used to encode the extended dest ID.
*/
MSIMessage msi = ...