diff mbox

[v9,3/6] target-arm: kvm - support for single step

Message ID 1447345251-22625-4-git-send-email-alex.bennee@linaro.org (mailing list archive)
State New, archived
Headers show

Commit Message

Alex Bennée Nov. 12, 2015, 4:20 p.m. UTC
This adds support for single-step. There isn't much to do on the QEMU
side as after we set-up the request for single step via the debug ioctl
it is all handled within the kernel.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

---
v2
  - convert to using HSR_EC
v3
  - use internals.h definitions
---
 target-arm/kvm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

Comments

Peter Maydell Nov. 20, 2015, 3:30 p.m. UTC | #1
On 12 November 2015 at 16:20, Alex Bennée <alex.bennee@linaro.org> wrote:
> This adds support for single-step. There isn't much to do on the QEMU
> side as after we set-up the request for single step via the debug ioctl
> it is all handled within the kernel.
>
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>
> ---
> v2
>   - convert to using HSR_EC
> v3
>   - use internals.h definitions
> ---
>  target-arm/kvm.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
>
> diff --git a/target-arm/kvm.c b/target-arm/kvm.c
> index 50f70ef..d505a7e 100644
> --- a/target-arm/kvm.c
> +++ b/target-arm/kvm.c
> @@ -535,6 +535,13 @@ static int kvm_handle_debug(CPUState *cs, struct kvm_run *run)
>      kvm_cpu_synchronize_state(cs);
>
>      switch (hsr_ec) {
> +    case EC_SOFTWARESTEP:
> +        if (cs->singlestep_enabled) {
> +            return true;
> +        } else {
> +            error_report("Came out of SINGLE STEP when not enabled");
> +        }
> +        break;
>      case EC_AA64_BKPT:
>          if (kvm_find_sw_breakpoint(cs, env->pc)) {
>              return true;
> @@ -595,6 +602,9 @@ int kvm_arch_on_sigbus(int code, void *addr)
>
>  void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
>  {
> +    if (cs->singlestep_enabled) {
> +        dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_SINGLESTEP;
> +    }

Doesn't kvm_update_guest_debug() already set these bits, or am
I misreading it?

>      if (kvm_sw_breakpoints_active(cs)) {
>          dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP;
>      }
> --
> 2.6.3

thanks
-- PMM
Alex Bennée Dec. 8, 2015, 11:49 a.m. UTC | #2
Peter Maydell <peter.maydell@linaro.org> writes:

> On 12 November 2015 at 16:20, Alex Bennée <alex.bennee@linaro.org> wrote:
>> This adds support for single-step. There isn't much to do on the QEMU
>> side as after we set-up the request for single step via the debug ioctl
>> it is all handled within the kernel.
>>
>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>>
>> ---
>> v2
>>   - convert to using HSR_EC
>> v3
>>   - use internals.h definitions
>> ---
>>  target-arm/kvm.c | 10 ++++++++++
>>  1 file changed, 10 insertions(+)
>>
>> diff --git a/target-arm/kvm.c b/target-arm/kvm.c
>> index 50f70ef..d505a7e 100644
>> --- a/target-arm/kvm.c
>> +++ b/target-arm/kvm.c
>> @@ -535,6 +535,13 @@ static int kvm_handle_debug(CPUState *cs, struct kvm_run *run)
>>      kvm_cpu_synchronize_state(cs);
>>
>>      switch (hsr_ec) {
>> +    case EC_SOFTWARESTEP:
>> +        if (cs->singlestep_enabled) {
>> +            return true;
>> +        } else {
>> +            error_report("Came out of SINGLE STEP when not enabled");
>> +        }
>> +        break;
>>      case EC_AA64_BKPT:
>>          if (kvm_find_sw_breakpoint(cs, env->pc)) {
>>              return true;
>> @@ -595,6 +602,9 @@ int kvm_arch_on_sigbus(int code, void *addr)
>>
>>  void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
>>  {
>> +    if (cs->singlestep_enabled) {
>> +        dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_SINGLESTEP;
>> +    }
>
> Doesn't kvm_update_guest_debug() already set these bits, or am
> I misreading it?

Yeah. This raises an interesting problem about what to do when we don't
have the capability. I could suppress those bits in the update function
but that seems a bit hacky.

Looking at the GDB capability code there doesn't seem to report
breakpoint capability short of just failing when you try to set one.

>
>>      if (kvm_sw_breakpoints_active(cs)) {
>>          dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP;
>>      }
>> --
>> 2.6.3
>
> thanks
> -- PMM


--
Alex Bennée
diff mbox

Patch

diff --git a/target-arm/kvm.c b/target-arm/kvm.c
index 50f70ef..d505a7e 100644
--- a/target-arm/kvm.c
+++ b/target-arm/kvm.c
@@ -535,6 +535,13 @@  static int kvm_handle_debug(CPUState *cs, struct kvm_run *run)
     kvm_cpu_synchronize_state(cs);
 
     switch (hsr_ec) {
+    case EC_SOFTWARESTEP:
+        if (cs->singlestep_enabled) {
+            return true;
+        } else {
+            error_report("Came out of SINGLE STEP when not enabled");
+        }
+        break;
     case EC_AA64_BKPT:
         if (kvm_find_sw_breakpoint(cs, env->pc)) {
             return true;
@@ -595,6 +602,9 @@  int kvm_arch_on_sigbus(int code, void *addr)
 
 void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
 {
+    if (cs->singlestep_enabled) {
+        dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_SINGLESTEP;
+    }
     if (kvm_sw_breakpoints_active(cs)) {
         dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP;
     }