Do not take the BQL before dispatching MMIO requests of KVM VCPUs.
Instead, call the unlocked version of address_space_rw. This enables
completely BQL-free MMIO handling in KVM mode for upcoming devices with
fine-grained locking.
Signed-off-by: Paolo Bonzini <pbonz...@redhat.com>
---
kvm-all.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/kvm-all.c b/kvm-all.c
index 2848e5b..fa0cfed 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -1813,12 +1813,11 @@ int kvm_cpu_exec(CPUState *cpu)
break;
case KVM_EXIT_MMIO:
DPRINTF("handle_mmio\n");
- qemu_mutex_lock_iothread();
- cpu_physical_memory_rw(run->mmio.phys_addr,
- run->mmio.data,
- run->mmio.len,
- run->mmio.is_write);
- qemu_mutex_unlock_iothread();
+ address_space_rw_unlocked(&address_space_memory,
+ run->mmio.phys_addr,
+ run->mmio.data,
+ run->mmio.len,
+ run->mmio.is_write);
ret = 0;
break;
case KVM_EXIT_IRQ_WINDOW_OPEN:
--
2.3.0
