Message ID | 20180111125444.6498-1-pbonzini@redhat.com |
---|---|
State | New |
Headers | show |
Series | cpus: unify qemu_*_wait_io_event | expand |
Paolo Bonzini <pbonzini@redhat.com> writes: > Except for round-robin TCG, every other accelerator is using more or > less the same code around qemu_wait_io_event_common. The exception > is HAX, which also has to eat the dummy APC that is queued by > qemu_cpu_kick_thread. > > We can add the SleepEx call to qemu_wait_io_event under "if > (!tcg_enabled())", since that is the condition that is used in > qemu_cpu_kick_thread, and unify the function for KVM, HAX, HVF and > multi-threaded TCG. Single-threaded TCG code can also be simplified > since it is only used in the round-robin, sleep-if-all-CPUs-idle case. > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> I had trouble applying this patch but the change seems sane to me Acked-by: Alex Bennée <alex.bennee@linaro.org> > --- > cpus.c | 49 +++++++++++++++++-------------------------------- > 1 file changed, 17 insertions(+), 32 deletions(-) > > diff --git a/cpus.c b/cpus.c > index 440b9291f5..3b6c9879ec 100644 > --- a/cpus.c > +++ b/cpus.c > @@ -911,7 +911,8 @@ static void kick_tcg_thread(void *opaque) > > static void start_tcg_kick_timer(void) > { > - if (!mttcg_enabled && !tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { > + assert(!mttcg_enabled); > + if (!tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { > tcg_kick_vcpu_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, > kick_tcg_thread, NULL); > timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick()); > @@ -920,6 +921,7 @@ static void start_tcg_kick_timer(void) > > static void stop_tcg_kick_timer(void) > { > + assert(!mttcg_enabled); > if (tcg_kick_vcpu_timer) { > timer_del(tcg_kick_vcpu_timer); > tcg_kick_vcpu_timer = NULL; > @@ -1154,18 +1156,9 @@ static void qemu_wait_io_event_common(CPUState *cpu) > process_queued_cpu_work(cpu); > } > > -static bool qemu_tcg_should_sleep(CPUState *cpu) > +static void qemu_tcg_rr_wait_io_event(CPUState *cpu) > { > - if (mttcg_enabled) { > - return cpu_thread_is_idle(cpu); > - } else { > - return all_cpu_threads_idle(); > - } > -} > - > -static void qemu_tcg_wait_io_event(CPUState *cpu) > -{ > - while (qemu_tcg_should_sleep(cpu)) { > + while (all_cpu_threads_idle()) { > stop_tcg_kick_timer(); > qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); > } > @@ -1175,20 +1168,18 @@ static void qemu_tcg_wait_io_event(CPUState *cpu) > qemu_wait_io_event_common(cpu); > } > > -static void qemu_kvm_wait_io_event(CPUState *cpu) > +static void qemu_wait_io_event(CPUState *cpu) > { > while (cpu_thread_is_idle(cpu)) { > qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); > } > > - qemu_wait_io_event_common(cpu); > -} > - > -static void qemu_hvf_wait_io_event(CPUState *cpu) > -{ > - while (cpu_thread_is_idle(cpu)) { > - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); > +#ifdef _WIN32 > + /* Eat dummy APC queued by qemu_cpu_kick_thread. */ > + if (!tcg_enabled()) { > + SleepEx(0, TRUE); > } > +#endif > qemu_wait_io_event_common(cpu); > } > > @@ -1224,7 +1215,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg) > cpu_handle_guest_debug(cpu); > } > } > - qemu_kvm_wait_io_event(cpu); > + qemu_wait_io_event(cpu); > } while (!cpu->unplug || cpu_can_run(cpu)); > > qemu_kvm_destroy_vcpu(cpu); > @@ -1270,7 +1261,7 @@ static void *qemu_dummy_cpu_thread_fn(void *arg) > exit(1); > } > qemu_mutex_lock_iothread(); > - qemu_wait_io_event_common(cpu); > + qemu_wait_io_event(cpu); > } > > return NULL; > @@ -1487,7 +1478,7 @@ static void *qemu_tcg_rr_cpu_thread_fn(void *arg) > atomic_mb_set(&cpu->exit_request, 0); > } > > - qemu_tcg_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); > + qemu_tcg_rr_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); > deal_with_unplugged_cpus(); > } > > @@ -1518,13 +1509,7 @@ static void *qemu_hax_cpu_thread_fn(void *arg) > } > } > > - while (cpu_thread_is_idle(cpu)) { > - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); > - } > -#ifdef _WIN32 > - SleepEx(0, TRUE); > -#endif > - qemu_wait_io_event_common(cpu); > + qemu_wait_io_event(cpu); > } > return NULL; > } > @@ -1561,7 +1546,7 @@ static void *qemu_hvf_cpu_thread_fn(void *arg) > cpu_handle_guest_debug(cpu); > } > } > - qemu_hvf_wait_io_event(cpu); > + qemu_wait_io_event(cpu); > } while (!cpu->unplug || cpu_can_run(cpu)); > > hvf_vcpu_destroy(cpu); > @@ -1640,7 +1625,7 @@ static void *qemu_tcg_cpu_thread_fn(void *arg) > } > > atomic_mb_set(&cpu->exit_request, 0); > - qemu_tcg_wait_io_event(cpu); > + qemu_wait_io_event(cpu); > } > > return NULL; -- Alex Bennée
On 04/04/2018 12:34, Alex Bennée wrote: > > Paolo Bonzini <pbonzini@redhat.com> writes: > >> Except for round-robin TCG, every other accelerator is using more or >> less the same code around qemu_wait_io_event_common. The exception >> is HAX, which also has to eat the dummy APC that is queued by >> qemu_cpu_kick_thread. >> >> We can add the SleepEx call to qemu_wait_io_event under "if >> (!tcg_enabled())", since that is the condition that is used in >> qemu_cpu_kick_thread, and unify the function for KVM, HAX, HVF and >> multi-threaded TCG. Single-threaded TCG code can also be simplified >> since it is only used in the round-robin, sleep-if-all-CPUs-idle case. >> >> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > > I had trouble applying this patch but the change seems sane to me > > Acked-by: Alex Bennée <alex.bennee@linaro.org> Isn't this commit db08b687cdd5319286665aabd34f82665630416f? Paolo >> --- >> cpus.c | 49 +++++++++++++++++-------------------------------- >> 1 file changed, 17 insertions(+), 32 deletions(-) >> >> diff --git a/cpus.c b/cpus.c >> index 440b9291f5..3b6c9879ec 100644 >> --- a/cpus.c >> +++ b/cpus.c >> @@ -911,7 +911,8 @@ static void kick_tcg_thread(void *opaque) >> >> static void start_tcg_kick_timer(void) >> { >> - if (!mttcg_enabled && !tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { >> + assert(!mttcg_enabled); >> + if (!tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { >> tcg_kick_vcpu_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, >> kick_tcg_thread, NULL); >> timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick()); >> @@ -920,6 +921,7 @@ static void start_tcg_kick_timer(void) >> >> static void stop_tcg_kick_timer(void) >> { >> + assert(!mttcg_enabled); >> if (tcg_kick_vcpu_timer) { >> timer_del(tcg_kick_vcpu_timer); >> tcg_kick_vcpu_timer = NULL; >> @@ -1154,18 +1156,9 @@ static void qemu_wait_io_event_common(CPUState *cpu) >> process_queued_cpu_work(cpu); >> } >> >> -static bool qemu_tcg_should_sleep(CPUState *cpu) >> +static void qemu_tcg_rr_wait_io_event(CPUState *cpu) >> { >> - if (mttcg_enabled) { >> - return cpu_thread_is_idle(cpu); >> - } else { >> - return all_cpu_threads_idle(); >> - } >> -} >> - >> -static void qemu_tcg_wait_io_event(CPUState *cpu) >> -{ >> - while (qemu_tcg_should_sleep(cpu)) { >> + while (all_cpu_threads_idle()) { >> stop_tcg_kick_timer(); >> qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >> } >> @@ -1175,20 +1168,18 @@ static void qemu_tcg_wait_io_event(CPUState *cpu) >> qemu_wait_io_event_common(cpu); >> } >> >> -static void qemu_kvm_wait_io_event(CPUState *cpu) >> +static void qemu_wait_io_event(CPUState *cpu) >> { >> while (cpu_thread_is_idle(cpu)) { >> qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >> } >> >> - qemu_wait_io_event_common(cpu); >> -} >> - >> -static void qemu_hvf_wait_io_event(CPUState *cpu) >> -{ >> - while (cpu_thread_is_idle(cpu)) { >> - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >> +#ifdef _WIN32 >> + /* Eat dummy APC queued by qemu_cpu_kick_thread. */ >> + if (!tcg_enabled()) { >> + SleepEx(0, TRUE); >> } >> +#endif >> qemu_wait_io_event_common(cpu); >> } >> >> @@ -1224,7 +1215,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg) >> cpu_handle_guest_debug(cpu); >> } >> } >> - qemu_kvm_wait_io_event(cpu); >> + qemu_wait_io_event(cpu); >> } while (!cpu->unplug || cpu_can_run(cpu)); >> >> qemu_kvm_destroy_vcpu(cpu); >> @@ -1270,7 +1261,7 @@ static void *qemu_dummy_cpu_thread_fn(void *arg) >> exit(1); >> } >> qemu_mutex_lock_iothread(); >> - qemu_wait_io_event_common(cpu); >> + qemu_wait_io_event(cpu); >> } >> >> return NULL; >> @@ -1487,7 +1478,7 @@ static void *qemu_tcg_rr_cpu_thread_fn(void *arg) >> atomic_mb_set(&cpu->exit_request, 0); >> } >> >> - qemu_tcg_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); >> + qemu_tcg_rr_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); >> deal_with_unplugged_cpus(); >> } >> >> @@ -1518,13 +1509,7 @@ static void *qemu_hax_cpu_thread_fn(void *arg) >> } >> } >> >> - while (cpu_thread_is_idle(cpu)) { >> - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >> - } >> -#ifdef _WIN32 >> - SleepEx(0, TRUE); >> -#endif >> - qemu_wait_io_event_common(cpu); >> + qemu_wait_io_event(cpu); >> } >> return NULL; >> } >> @@ -1561,7 +1546,7 @@ static void *qemu_hvf_cpu_thread_fn(void *arg) >> cpu_handle_guest_debug(cpu); >> } >> } >> - qemu_hvf_wait_io_event(cpu); >> + qemu_wait_io_event(cpu); >> } while (!cpu->unplug || cpu_can_run(cpu)); >> >> hvf_vcpu_destroy(cpu); >> @@ -1640,7 +1625,7 @@ static void *qemu_tcg_cpu_thread_fn(void *arg) >> } >> >> atomic_mb_set(&cpu->exit_request, 0); >> - qemu_tcg_wait_io_event(cpu); >> + qemu_wait_io_event(cpu); >> } >> >> return NULL; > > > -- > Alex Bennée >
Paolo Bonzini <pbonzini@redhat.com> writes: > On 04/04/2018 12:34, Alex Bennée wrote: >> >> Paolo Bonzini <pbonzini@redhat.com> writes: >> >>> Except for round-robin TCG, every other accelerator is using more or >>> less the same code around qemu_wait_io_event_common. The exception >>> is HAX, which also has to eat the dummy APC that is queued by >>> qemu_cpu_kick_thread. >>> >>> We can add the SleepEx call to qemu_wait_io_event under "if >>> (!tcg_enabled())", since that is the condition that is used in >>> qemu_cpu_kick_thread, and unify the function for KVM, HAX, HVF and >>> multi-threaded TCG. Single-threaded TCG code can also be simplified >>> since it is only used in the round-robin, sleep-if-all-CPUs-idle case. >>> >>> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> >> >> I had trouble applying this patch but the change seems sane to me >> >> Acked-by: Alex Bennée <alex.bennee@linaro.org> > > Isn't this commit db08b687cdd5319286665aabd34f82665630416f? Doh sorry, that makes even more sense. My check-if-merged script failed me... > > Paolo > >>> --- >>> cpus.c | 49 +++++++++++++++++-------------------------------- >>> 1 file changed, 17 insertions(+), 32 deletions(-) >>> >>> diff --git a/cpus.c b/cpus.c >>> index 440b9291f5..3b6c9879ec 100644 >>> --- a/cpus.c >>> +++ b/cpus.c >>> @@ -911,7 +911,8 @@ static void kick_tcg_thread(void *opaque) >>> >>> static void start_tcg_kick_timer(void) >>> { >>> - if (!mttcg_enabled && !tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { >>> + assert(!mttcg_enabled); >>> + if (!tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { >>> tcg_kick_vcpu_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, >>> kick_tcg_thread, NULL); >>> timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick()); >>> @@ -920,6 +921,7 @@ static void start_tcg_kick_timer(void) >>> >>> static void stop_tcg_kick_timer(void) >>> { >>> + assert(!mttcg_enabled); >>> if (tcg_kick_vcpu_timer) { >>> timer_del(tcg_kick_vcpu_timer); >>> tcg_kick_vcpu_timer = NULL; >>> @@ -1154,18 +1156,9 @@ static void qemu_wait_io_event_common(CPUState *cpu) >>> process_queued_cpu_work(cpu); >>> } >>> >>> -static bool qemu_tcg_should_sleep(CPUState *cpu) >>> +static void qemu_tcg_rr_wait_io_event(CPUState *cpu) >>> { >>> - if (mttcg_enabled) { >>> - return cpu_thread_is_idle(cpu); >>> - } else { >>> - return all_cpu_threads_idle(); >>> - } >>> -} >>> - >>> -static void qemu_tcg_wait_io_event(CPUState *cpu) >>> -{ >>> - while (qemu_tcg_should_sleep(cpu)) { >>> + while (all_cpu_threads_idle()) { >>> stop_tcg_kick_timer(); >>> qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >>> } >>> @@ -1175,20 +1168,18 @@ static void qemu_tcg_wait_io_event(CPUState *cpu) >>> qemu_wait_io_event_common(cpu); >>> } >>> >>> -static void qemu_kvm_wait_io_event(CPUState *cpu) >>> +static void qemu_wait_io_event(CPUState *cpu) >>> { >>> while (cpu_thread_is_idle(cpu)) { >>> qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >>> } >>> >>> - qemu_wait_io_event_common(cpu); >>> -} >>> - >>> -static void qemu_hvf_wait_io_event(CPUState *cpu) >>> -{ >>> - while (cpu_thread_is_idle(cpu)) { >>> - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >>> +#ifdef _WIN32 >>> + /* Eat dummy APC queued by qemu_cpu_kick_thread. */ >>> + if (!tcg_enabled()) { >>> + SleepEx(0, TRUE); >>> } >>> +#endif >>> qemu_wait_io_event_common(cpu); >>> } >>> >>> @@ -1224,7 +1215,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg) >>> cpu_handle_guest_debug(cpu); >>> } >>> } >>> - qemu_kvm_wait_io_event(cpu); >>> + qemu_wait_io_event(cpu); >>> } while (!cpu->unplug || cpu_can_run(cpu)); >>> >>> qemu_kvm_destroy_vcpu(cpu); >>> @@ -1270,7 +1261,7 @@ static void *qemu_dummy_cpu_thread_fn(void *arg) >>> exit(1); >>> } >>> qemu_mutex_lock_iothread(); >>> - qemu_wait_io_event_common(cpu); >>> + qemu_wait_io_event(cpu); >>> } >>> >>> return NULL; >>> @@ -1487,7 +1478,7 @@ static void *qemu_tcg_rr_cpu_thread_fn(void *arg) >>> atomic_mb_set(&cpu->exit_request, 0); >>> } >>> >>> - qemu_tcg_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); >>> + qemu_tcg_rr_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); >>> deal_with_unplugged_cpus(); >>> } >>> >>> @@ -1518,13 +1509,7 @@ static void *qemu_hax_cpu_thread_fn(void *arg) >>> } >>> } >>> >>> - while (cpu_thread_is_idle(cpu)) { >>> - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); >>> - } >>> -#ifdef _WIN32 >>> - SleepEx(0, TRUE); >>> -#endif >>> - qemu_wait_io_event_common(cpu); >>> + qemu_wait_io_event(cpu); >>> } >>> return NULL; >>> } >>> @@ -1561,7 +1546,7 @@ static void *qemu_hvf_cpu_thread_fn(void *arg) >>> cpu_handle_guest_debug(cpu); >>> } >>> } >>> - qemu_hvf_wait_io_event(cpu); >>> + qemu_wait_io_event(cpu); >>> } while (!cpu->unplug || cpu_can_run(cpu)); >>> >>> hvf_vcpu_destroy(cpu); >>> @@ -1640,7 +1625,7 @@ static void *qemu_tcg_cpu_thread_fn(void *arg) >>> } >>> >>> atomic_mb_set(&cpu->exit_request, 0); >>> - qemu_tcg_wait_io_event(cpu); >>> + qemu_wait_io_event(cpu); >>> } >>> >>> return NULL; >> >> >> -- >> Alex Bennée >> -- Alex Bennée
diff --git a/cpus.c b/cpus.c index 440b9291f5..3b6c9879ec 100644 --- a/cpus.c +++ b/cpus.c @@ -911,7 +911,8 @@ static void kick_tcg_thread(void *opaque) static void start_tcg_kick_timer(void) { - if (!mttcg_enabled && !tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { + assert(!mttcg_enabled); + if (!tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) { tcg_kick_vcpu_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, kick_tcg_thread, NULL); timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick()); @@ -920,6 +921,7 @@ static void start_tcg_kick_timer(void) static void stop_tcg_kick_timer(void) { + assert(!mttcg_enabled); if (tcg_kick_vcpu_timer) { timer_del(tcg_kick_vcpu_timer); tcg_kick_vcpu_timer = NULL; @@ -1154,18 +1156,9 @@ static void qemu_wait_io_event_common(CPUState *cpu) process_queued_cpu_work(cpu); } -static bool qemu_tcg_should_sleep(CPUState *cpu) +static void qemu_tcg_rr_wait_io_event(CPUState *cpu) { - if (mttcg_enabled) { - return cpu_thread_is_idle(cpu); - } else { - return all_cpu_threads_idle(); - } -} - -static void qemu_tcg_wait_io_event(CPUState *cpu) -{ - while (qemu_tcg_should_sleep(cpu)) { + while (all_cpu_threads_idle()) { stop_tcg_kick_timer(); qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); } @@ -1175,20 +1168,18 @@ static void qemu_tcg_wait_io_event(CPUState *cpu) qemu_wait_io_event_common(cpu); } -static void qemu_kvm_wait_io_event(CPUState *cpu) +static void qemu_wait_io_event(CPUState *cpu) { while (cpu_thread_is_idle(cpu)) { qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); } - qemu_wait_io_event_common(cpu); -} - -static void qemu_hvf_wait_io_event(CPUState *cpu) -{ - while (cpu_thread_is_idle(cpu)) { - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); +#ifdef _WIN32 + /* Eat dummy APC queued by qemu_cpu_kick_thread. */ + if (!tcg_enabled()) { + SleepEx(0, TRUE); } +#endif qemu_wait_io_event_common(cpu); } @@ -1224,7 +1215,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg) cpu_handle_guest_debug(cpu); } } - qemu_kvm_wait_io_event(cpu); + qemu_wait_io_event(cpu); } while (!cpu->unplug || cpu_can_run(cpu)); qemu_kvm_destroy_vcpu(cpu); @@ -1270,7 +1261,7 @@ static void *qemu_dummy_cpu_thread_fn(void *arg) exit(1); } qemu_mutex_lock_iothread(); - qemu_wait_io_event_common(cpu); + qemu_wait_io_event(cpu); } return NULL; @@ -1487,7 +1478,7 @@ static void *qemu_tcg_rr_cpu_thread_fn(void *arg) atomic_mb_set(&cpu->exit_request, 0); } - qemu_tcg_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); + qemu_tcg_rr_wait_io_event(cpu ? cpu : QTAILQ_FIRST(&cpus)); deal_with_unplugged_cpus(); } @@ -1518,13 +1509,7 @@ static void *qemu_hax_cpu_thread_fn(void *arg) } } - while (cpu_thread_is_idle(cpu)) { - qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex); - } -#ifdef _WIN32 - SleepEx(0, TRUE); -#endif - qemu_wait_io_event_common(cpu); + qemu_wait_io_event(cpu); } return NULL; } @@ -1561,7 +1546,7 @@ static void *qemu_hvf_cpu_thread_fn(void *arg) cpu_handle_guest_debug(cpu); } } - qemu_hvf_wait_io_event(cpu); + qemu_wait_io_event(cpu); } while (!cpu->unplug || cpu_can_run(cpu)); hvf_vcpu_destroy(cpu); @@ -1640,7 +1625,7 @@ static void *qemu_tcg_cpu_thread_fn(void *arg) } atomic_mb_set(&cpu->exit_request, 0); - qemu_tcg_wait_io_event(cpu); + qemu_wait_io_event(cpu); } return NULL;
Except for round-robin TCG, every other accelerator is using more or less the same code around qemu_wait_io_event_common. The exception is HAX, which also has to eat the dummy APC that is queued by qemu_cpu_kick_thread. We can add the SleepEx call to qemu_wait_io_event under "if (!tcg_enabled())", since that is the condition that is used in qemu_cpu_kick_thread, and unify the function for KVM, HAX, HVF and multi-threaded TCG. Single-threaded TCG code can also be simplified since it is only used in the round-robin, sleep-if-all-CPUs-idle case. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> --- cpus.c | 49 +++++++++++++++++-------------------------------- 1 file changed, 17 insertions(+), 32 deletions(-)