[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 09/19] main-loop: use qemu_mutex_lock_iothread consis
From: |
Paolo Bonzini |
Subject: |
[Qemu-devel] [PULL 09/19] main-loop: use qemu_mutex_lock_iothread consistently |
Date: |
Mon, 6 Jul 2015 14:34:58 +0200 |
The next patch will require the BQL to be always taken with
qemu_mutex_lock_iothread(), while right now this isn't the case.
Outside TCG mode this is not a problem. In TCG mode, we need to be
careful and avoid the "prod out of compiled code" step if already
in a VCPU thread. This is easily done with a check on current_cpu,
i.e. qemu_in_vcpu_thread().
Hopefully, multithreaded TCG will get rid of the whole logic to kick
VCPUs whenever an I/O event occurs!
Cc: Frederic Konrad <address@hidden>
Message-Id: <address@hidden>
Signed-off-by: Paolo Bonzini <address@hidden>
---
cpus.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/cpus.c b/cpus.c
index 4f0e54d..c09fbef 100644
--- a/cpus.c
+++ b/cpus.c
@@ -954,7 +954,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg)
CPUState *cpu = arg;
int r;
- qemu_mutex_lock(&qemu_global_mutex);
+ qemu_mutex_lock_iothread();
qemu_thread_get_self(cpu->thread);
cpu->thread_id = qemu_get_thread_id();
cpu->can_do_io = 1;
@@ -1034,10 +1034,10 @@ static void *qemu_tcg_cpu_thread_fn(void *arg)
{
CPUState *cpu = arg;
+ qemu_mutex_lock_iothread();
qemu_tcg_init_cpu_signals();
qemu_thread_get_self(cpu->thread);
- qemu_mutex_lock(&qemu_global_mutex);
CPU_FOREACH(cpu) {
cpu->thread_id = qemu_get_thread_id();
cpu->created = true;
@@ -1149,7 +1149,11 @@ bool qemu_in_vcpu_thread(void)
void qemu_mutex_lock_iothread(void)
{
atomic_inc(&iothread_requesting_mutex);
- if (!tcg_enabled() || !first_cpu || !first_cpu->thread) {
+ /* In the simple case there is no need to bump the VCPU thread out of
+ * TCG code execution.
+ */
+ if (!tcg_enabled() || qemu_in_vcpu_thread() ||
+ !first_cpu || !first_cpu->thread) {
qemu_mutex_lock(&qemu_global_mutex);
atomic_dec(&iothread_requesting_mutex);
} else {
--
2.4.3
- [Qemu-devel] [PULL 00/19] Multithread, multiarch, multicompiler, multi^WKVM changes for 2015-07-06, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 02/19] cutils: allow compilation with icc, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 01/19] qemu-common: add VEC_OR macro, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 04/19] cpu-defs: Move CPU_TEMP_BUF_NLONGS to tcg, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 05/19] include/exec: Move standard exceptions to cpu-all.h, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 07/19] cpu-defs: Move out TB_JMP defines, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 03/19] memory_mapping: Rework cpu related includes, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 06/19] include/exec: Move tb hash functions out, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 08/19] Fix irq route entries exceeding KVM_MAX_IRQ_ROUTES, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 09/19] main-loop: use qemu_mutex_lock_iothread consistently,
Paolo Bonzini <=
- [Qemu-devel] [PULL 11/19] memory: Add global-locking property to memory regions, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 12/19] exec: pull qemu_flush_coalesced_mmio_buffer() into address_space_rw/ld*/st*, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 13/19] memory: let address_space_rw/ld*/st* run outside the BQL, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 10/19] main-loop: introduce qemu_mutex_iothread_locked, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 15/19] kvm: Switch to unlocked PIO, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 14/19] kvm: First step to push iothread lock out of inner run loop, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 16/19] acpi: mark PMTIMER as unlocked, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 19/19] exec: skip MMIO regions correctly in cpu_physical_memory_write_rom_internal, Paolo Bonzini, 2015/07/06
- [Qemu-devel] [PULL 18/19] Stop including qemu-common.h in memory.h, Paolo Bonzini, 2015/07/06