[Qemu-devel] [PATCH v3] ppc: make idle_timer a per-cpu variable

[Qemu-devel] [PATCH v3] ppc: make idle_timer a per-cpu variable

[Shivaprasad G Bhat]

The current code is broken for more than vcpu as
each thread would overwrite and there were memory leaks.

Make it part of PowerPCCPU so that every thread has a
separate one. Avoid using the timer_new_ns which is
not the preferred way to create timers.

Signed-off-by: Shivaprasad G Bhat <sbhat@linux.ibm.com>
---
 v2: https://lists.gnu.org/archive/html/qemu-devel/2019-07/msg04023.html
 Changes from v2:
   v2 just looked at avoiding the memory leak.
   This patch incorporates all of Greg's suggestions.

 target/ppc/cpu.h |    1 +
 target/ppc/kvm.c |   31 ++++++++++++++++---------------
 2 files changed, 17 insertions(+), 15 deletions(-)

diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index c9beba2a5c..521086d91a 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -1190,6 +1190,7 @@ struct PowerPCCPU {
     void *machine_data;
     int32_t node_id; /* NUMA node this CPU belongs to */
     PPCHash64Options *hash64_opts;
+    QEMUTimer idle_timer;
 
     /* Fields related to migration compatibility hacks */
     bool pre_2_8_migration;
diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c
index 8a06d3171e..6e1b96bb0a 100644
--- a/target/ppc/kvm.c
+++ b/target/ppc/kvm.c
@@ -87,18 +87,6 @@ static int cap_large_decr;
 
 static uint32_t debug_inst_opcode;
 
-/*
- * XXX We have a race condition where we actually have a level triggered
- *     interrupt, but the infrastructure can't expose that yet, so the guest
- *     takes but ignores it, goes to sleep and never gets notified that there's
- *     still an interrupt pending.
- *
- *     As a quick workaround, let's just wake up again 20 ms after we injected
- *     an interrupt. That way we can assure that we're always reinjecting
- *     interrupts in case the guest swallowed them.
- */
-static QEMUTimer *idle_timer;
-
 static void kvm_kick_cpu(void *opaque)
 {
     PowerPCCPU *cpu = opaque;
@@ -491,7 +479,7 @@ int kvm_arch_init_vcpu(CPUState *cs)
         return ret;
     }
 
-    idle_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, kvm_kick_cpu, cpu);
+    timer_init_ns(&cpu->idle_timer, QEMU_CLOCK_VIRTUAL, kvm_kick_cpu, cpu);
 
     switch (cenv->mmu_model) {
     case POWERPC_MMU_BOOKE206:
@@ -523,6 +511,10 @@ int kvm_arch_init_vcpu(CPUState *cs)
 
 int kvm_arch_destroy_vcpu(CPUState *cs)
 {
+    PowerPCCPU *cpu = POWERPC_CPU(cs);
+
+    timer_deinit(&cpu->idle_timer);
+
     return 0;
 }
 
@@ -1379,8 +1371,17 @@ void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
             printf("cpu %d fail inject %x\n", cs->cpu_index, irq);
         }
 
-        /* Always wake up soon in case the interrupt was level based */
-        timer_mod(idle_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
+        /*
+         * XXX We have a race condition where we actually have a level
+         *     triggered interrupt, but the infrastructure can't expose that
+         *     yet, so the guest takes but ignores it, goes to sleep and
+         *     never gets notified that there's still an interrupt pending.
+         *
+         *     As a quick workaround, let's just wake up again 20 ms after
+         *     we injected an interrupt. That way we can assure that we're
+         *     always reinjecting interrupts in case the guest swallowed them.
+         */
+        timer_mod(&cpu->idle_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
                        (NANOSECONDS_PER_SECOND / 50));
     }
 

Re: [Qemu-devel] [Qemu-ppc] [PATCH v3] ppc: make idle_timer a per-cpu variable

[Greg Kurz]

On Thu, 18 Jul 2019 10:21:28 -0500
Shivaprasad G Bhat <sbhat@linux.ibm.com> wrote:

> The current code is broken for more than vcpu as
> each thread would overwrite and there were memory leaks.
> 
> Make it part of PowerPCCPU so that every thread has a
> separate one. Avoid using the timer_new_ns which is
> not the preferred way to create timers.
> 
> Signed-off-by: Shivaprasad G Bhat <sbhat@linux.ibm.com>
> ---
>  v2: https://lists.gnu.org/archive/html/qemu-devel/2019-07/msg04023.html
>  Changes from v2:
>    v2 just looked at avoiding the memory leak.
>    This patch incorporates all of Greg's suggestions.
> 
>  target/ppc/cpu.h |    1 +
>  target/ppc/kvm.c |   31 ++++++++++++++++---------------
>  2 files changed, 17 insertions(+), 15 deletions(-)
> 
> diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
> index c9beba2a5c..521086d91a 100644
> --- a/target/ppc/cpu.h
> +++ b/target/ppc/cpu.h
> @@ -1190,6 +1190,7 @@ struct PowerPCCPU {
>      void *machine_data;
>      int32_t node_id; /* NUMA node this CPU belongs to */
>      PPCHash64Options *hash64_opts;
> +    QEMUTimer idle_timer;
>  
>      /* Fields related to migration compatibility hacks */
>      bool pre_2_8_migration;
> diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c
> index 8a06d3171e..6e1b96bb0a 100644
> --- a/target/ppc/kvm.c
> +++ b/target/ppc/kvm.c
> @@ -87,18 +87,6 @@ static int cap_large_decr;
>  
>  static uint32_t debug_inst_opcode;
>  
> -/*
> - * XXX We have a race condition where we actually have a level triggered
> - *     interrupt, but the infrastructure can't expose that yet, so the guest
> - *     takes but ignores it, goes to sleep and never gets notified that there's
> - *     still an interrupt pending.
> - *
> - *     As a quick workaround, let's just wake up again 20 ms after we injected
> - *     an interrupt. That way we can assure that we're always reinjecting
> - *     interrupts in case the guest swallowed them.
> - */
> -static QEMUTimer *idle_timer;
> -
>  static void kvm_kick_cpu(void *opaque)
>  {
>      PowerPCCPU *cpu = opaque;
> @@ -491,7 +479,7 @@ int kvm_arch_init_vcpu(CPUState *cs)
>          return ret;
>      }
>  
> -    idle_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, kvm_kick_cpu, cpu);
> +    timer_init_ns(&cpu->idle_timer, QEMU_CLOCK_VIRTUAL, kvm_kick_cpu, cpu);
>  
>      switch (cenv->mmu_model) {
>      case POWERPC_MMU_BOOKE206:
> @@ -523,6 +511,10 @@ int kvm_arch_init_vcpu(CPUState *cs)
>  
>  int kvm_arch_destroy_vcpu(CPUState *cs)
>  {
> +    PowerPCCPU *cpu = POWERPC_CPU(cs);
> +
> +    timer_deinit(&cpu->idle_timer);

As stated in the timer.h header file, timer_del() should always be called
before timer_deinit().

With that fixed:

Reviewed-by: Greg Kurz <groug@kaod.org>

> +
>      return 0;
>  }
>  
> @@ -1379,8 +1371,17 @@ void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
>              printf("cpu %d fail inject %x\n", cs->cpu_index, irq);
>          }
>  
> -        /* Always wake up soon in case the interrupt was level based */
> -        timer_mod(idle_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
> +        /*
> +         * XXX We have a race condition where we actually have a level
> +         *     triggered interrupt, but the infrastructure can't expose that
> +         *     yet, so the guest takes but ignores it, goes to sleep and
> +         *     never gets notified that there's still an interrupt pending.
> +         *
> +         *     As a quick workaround, let's just wake up again 20 ms after
> +         *     we injected an interrupt. That way we can assure that we're
> +         *     always reinjecting interrupts in case the guest swallowed them.
> +         */
> +        timer_mod(&cpu->idle_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
>                         (NANOSECONDS_PER_SECOND / 50));
>      }
>  
> 
> 

Re: [Qemu-devel] [Qemu-ppc] [PATCH v3] ppc: make idle_timer a per-cpu variable

[Shivaprasad G Bhat]

On 07/18/2019 09:47 PM, Greg Kurz wrote:
> On Thu, 18 Jul 2019 10:21:28 -0500
> Shivaprasad G Bhat <sbhat@linux.ibm.com> wrote:
>
> +    PowerPCCPU *cpu = POWERPC_CPU(cs);
> +
> +    timer_deinit(&cpu->idle_timer);
> As stated in the timer.h header file, timer_del() should always be called
> before timer_deinit().
>
> With that fixed:
>
> Reviewed-by: Greg Kurz <groug@kaod.org>

Thanks. Fixed, and posted the v4.

Regards,
Shivaprasad