[v4,08/14] RISC-V: KVM: Add SBI PMU extension support

SBI PMU extension allows KVM guests to configure/start/stop/query about
the PMU counters in virtualized enviornment as well.

In order to allow that, KVM implements the entire SBI PMU extension.

Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Atish Patra <atishp@rivosinc.com>
---
 arch/riscv/kvm/Makefile       |  2 +-
 arch/riscv/kvm/vcpu_sbi.c     | 11 +++++
 arch/riscv/kvm/vcpu_sbi_pmu.c | 85 +++++++++++++++++++++++++++++++++++
 3 files changed, 97 insertions(+), 1 deletion(-)
 create mode 100644 arch/riscv/kvm/vcpu_sbi_pmu.c

Hey Atish,

On Wed, Feb 01, 2023 at 03:12:44PM -0800, Atish Patra wrote:
> SBI PMU extension allows KVM guests to configure/start/stop/query about
> the PMU counters in virtualized enviornment as well.
> 
> In order to allow that, KVM implements the entire SBI PMU extension.
> 
> Reviewed-by: Anup Patel <anup@brainfault.org>
> Signed-off-by: Atish Patra <atishp@rivosinc.com>

Could complaints from CI for you:

+      1 ../arch/riscv/kvm/vcpu_sbi_pmu.c:73:15: warning: no previous prototype for 'kvm_sbi_ext_pmu_probe' [-Wmissing-prototypes]
+      1 ../arch/riscv/kvm/vcpu_sbi_pmu.c:73:15: warning: symbol 'kvm_sbi_ext_pmu_probe' was not declared. Should it be static?
+      1 ../arch/riscv/kvm/vcpu_sbi_pmu.c:80:37: warning: symbol 'vcpu_sbi_ext_pmu' was not declared. Should it be static?

Thanks,
Conor.

> ---
>  arch/riscv/kvm/Makefile       |  2 +-
>  arch/riscv/kvm/vcpu_sbi.c     | 11 +++++
>  arch/riscv/kvm/vcpu_sbi_pmu.c | 85 +++++++++++++++++++++++++++++++++++
>  3 files changed, 97 insertions(+), 1 deletion(-)
>  create mode 100644 arch/riscv/kvm/vcpu_sbi_pmu.c
> 
> diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile
> index 5de1053..278e97c 100644
> --- a/arch/riscv/kvm/Makefile
> +++ b/arch/riscv/kvm/Makefile
> @@ -25,4 +25,4 @@ kvm-y += vcpu_sbi_base.o
>  kvm-y += vcpu_sbi_replace.o
>  kvm-y += vcpu_sbi_hsm.o
>  kvm-y += vcpu_timer.o
> -kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o
> +kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o vcpu_sbi_pmu.o
> diff --git a/arch/riscv/kvm/vcpu_sbi.c b/arch/riscv/kvm/vcpu_sbi.c
> index fe2897e..15fde15 100644
> --- a/arch/riscv/kvm/vcpu_sbi.c
> +++ b/arch/riscv/kvm/vcpu_sbi.c
> @@ -20,6 +20,16 @@ static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_v01 = {
>  };
>  #endif
>  
> +#ifdef CONFIG_RISCV_PMU_SBI
> +extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu;
> +#else
> +static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> +	.extid_start = -1UL,
> +	.extid_end = -1UL,
> +	.handler = NULL,
> +};
> +#endif
> +
>  static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
>  	&vcpu_sbi_ext_v01,
>  	&vcpu_sbi_ext_base,
> @@ -28,6 +38,7 @@ static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
>  	&vcpu_sbi_ext_rfence,
>  	&vcpu_sbi_ext_srst,
>  	&vcpu_sbi_ext_hsm,
> +	&vcpu_sbi_ext_pmu,
>  	&vcpu_sbi_ext_experimental,
>  	&vcpu_sbi_ext_vendor,
>  };
> diff --git a/arch/riscv/kvm/vcpu_sbi_pmu.c b/arch/riscv/kvm/vcpu_sbi_pmu.c
> new file mode 100644
> index 0000000..e028b0a
> --- /dev/null
> +++ b/arch/riscv/kvm/vcpu_sbi_pmu.c
> @@ -0,0 +1,85 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (c) 2023 Rivos Inc
> + *
> + * Authors:
> + *     Atish Patra <atishp@rivosinc.com>
> + */
> +
> +#include <linux/errno.h>
> +#include <linux/err.h>
> +#include <linux/kvm_host.h>
> +#include <asm/csr.h>
> +#include <asm/sbi.h>
> +#include <asm/kvm_vcpu_sbi.h>
> +
> +static int kvm_sbi_ext_pmu_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
> +				   struct kvm_vcpu_sbi_return *retdata)
> +{
> +	int ret = 0;
> +	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
> +	struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> +	unsigned long funcid = cp->a6;
> +	uint64_t temp;
> +
> +	/* Return not supported if PMU is not initialized */
> +	if (!kvpmu->init_done)
> +		return -EINVAL;
> +
> +	switch (funcid) {
> +	case SBI_EXT_PMU_NUM_COUNTERS:
> +		ret = kvm_riscv_vcpu_pmu_num_ctrs(vcpu, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_GET_INFO:
> +		ret = kvm_riscv_vcpu_pmu_ctr_info(vcpu, cp->a0, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_CFG_MATCH:
> +#if defined(CONFIG_32BIT)
> +		temp = ((uint64_t)cp->a5 << 32) | cp->a4;
> +#else
> +		temp = cp->a4;
> +#endif
> +		/*
> +		 * This can fail if perf core framework fails to create an event.
> +		 * Forward the error to the user space because its an error happened
> +		 * within host kernel. The other option would be convert this to
> +		 * an SBI error and forward to the guest.
> +		 */
> +		ret = kvm_riscv_vcpu_pmu_ctr_cfg_match(vcpu, cp->a0, cp->a1,
> +						       cp->a2, cp->a3, temp, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_START:
> +#if defined(CONFIG_32BIT)
> +		temp = ((uint64_t)cp->a4 << 32) | cp->a3;
> +#else
> +		temp = cp->a3;
> +#endif
> +		ret = kvm_riscv_vcpu_pmu_ctr_start(vcpu, cp->a0, cp->a1, cp->a2,
> +						   temp, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_STOP:
> +		ret = kvm_riscv_vcpu_pmu_ctr_stop(vcpu, cp->a0, cp->a1, cp->a2, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_FW_READ:
> +		ret = kvm_riscv_vcpu_pmu_ctr_read(vcpu, cp->a0, retdata);
> +		break;
> +	default:
> +		retdata->err_val = SBI_ERR_NOT_SUPPORTED;
> +	}
> +
> +	return ret;
> +}
> +
> +unsigned long kvm_sbi_ext_pmu_probe(struct kvm_vcpu *vcpu)
> +{
> +	struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> +
> +	return kvpmu->init_done;
> +}
> +
> +const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> +	.extid_start = SBI_EXT_PMU,
> +	.extid_end = SBI_EXT_PMU,
> +	.handler = kvm_sbi_ext_pmu_handler,
> +	.probe = kvm_sbi_ext_pmu_probe,
> +};
> -- 
> 2.25.1
> 
>

On Wed, Feb 01, 2023 at 03:12:44PM -0800, Atish Patra wrote:
> SBI PMU extension allows KVM guests to configure/start/stop/query about
> the PMU counters in virtualized enviornment as well.
> 
> In order to allow that, KVM implements the entire SBI PMU extension.
> 
> Reviewed-by: Anup Patel <anup@brainfault.org>
> Signed-off-by: Atish Patra <atishp@rivosinc.com>
> ---
>  arch/riscv/kvm/Makefile       |  2 +-
>  arch/riscv/kvm/vcpu_sbi.c     | 11 +++++
>  arch/riscv/kvm/vcpu_sbi_pmu.c | 85 +++++++++++++++++++++++++++++++++++
>  3 files changed, 97 insertions(+), 1 deletion(-)
>  create mode 100644 arch/riscv/kvm/vcpu_sbi_pmu.c
> 
> diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile
> index 5de1053..278e97c 100644
> --- a/arch/riscv/kvm/Makefile
> +++ b/arch/riscv/kvm/Makefile
> @@ -25,4 +25,4 @@ kvm-y += vcpu_sbi_base.o
>  kvm-y += vcpu_sbi_replace.o
>  kvm-y += vcpu_sbi_hsm.o
>  kvm-y += vcpu_timer.o
> -kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o
> +kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o vcpu_sbi_pmu.o
> diff --git a/arch/riscv/kvm/vcpu_sbi.c b/arch/riscv/kvm/vcpu_sbi.c
> index fe2897e..15fde15 100644
> --- a/arch/riscv/kvm/vcpu_sbi.c
> +++ b/arch/riscv/kvm/vcpu_sbi.c
> @@ -20,6 +20,16 @@ static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_v01 = {
>  };
>  #endif
>  
> +#ifdef CONFIG_RISCV_PMU_SBI
> +extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu;
> +#else
> +static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> +	.extid_start = -1UL,
> +	.extid_end = -1UL,
> +	.handler = NULL,
> +};
> +#endif
> +
>  static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
>  	&vcpu_sbi_ext_v01,
>  	&vcpu_sbi_ext_base,
> @@ -28,6 +38,7 @@ static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
>  	&vcpu_sbi_ext_rfence,
>  	&vcpu_sbi_ext_srst,
>  	&vcpu_sbi_ext_hsm,
> +	&vcpu_sbi_ext_pmu,
>  	&vcpu_sbi_ext_experimental,
>  	&vcpu_sbi_ext_vendor,
>  };
> diff --git a/arch/riscv/kvm/vcpu_sbi_pmu.c b/arch/riscv/kvm/vcpu_sbi_pmu.c
> new file mode 100644
> index 0000000..e028b0a
> --- /dev/null
> +++ b/arch/riscv/kvm/vcpu_sbi_pmu.c
> @@ -0,0 +1,85 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (c) 2023 Rivos Inc
> + *
> + * Authors:
> + *     Atish Patra <atishp@rivosinc.com>
> + */
> +
> +#include <linux/errno.h>
> +#include <linux/err.h>
> +#include <linux/kvm_host.h>
> +#include <asm/csr.h>
> +#include <asm/sbi.h>
> +#include <asm/kvm_vcpu_sbi.h>
> +
> +static int kvm_sbi_ext_pmu_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
> +				   struct kvm_vcpu_sbi_return *retdata)
> +{
> +	int ret = 0;
> +	struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
> +	struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> +	unsigned long funcid = cp->a6;
> +	uint64_t temp;
> +
> +	/* Return not supported if PMU is not initialized */
> +	if (!kvpmu->init_done)
> +		return -EINVAL;

Shouldn't this be the following?

  if (!kvpmu->init_done)
     retdata->err_val = SBI_ERR_NOT_SUPPORTED;
     return 0;
  }

> +
> +	switch (funcid) {
> +	case SBI_EXT_PMU_NUM_COUNTERS:
> +		ret = kvm_riscv_vcpu_pmu_num_ctrs(vcpu, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_GET_INFO:
> +		ret = kvm_riscv_vcpu_pmu_ctr_info(vcpu, cp->a0, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_CFG_MATCH:
> +#if defined(CONFIG_32BIT)
> +		temp = ((uint64_t)cp->a5 << 32) | cp->a4;
> +#else
> +		temp = cp->a4;
> +#endif
> +		/*
> +		 * This can fail if perf core framework fails to create an event.
> +		 * Forward the error to the user space because its an error happened

"Forward the error to userspace because it's an error which happened within
 the host kernel."
                                        
> +		 * within host kernel. The other option would be convert this to
                                                                ^ to
> +		 * an SBI error and forward to the guest.
> +		 */
> +		ret = kvm_riscv_vcpu_pmu_ctr_cfg_match(vcpu, cp->a0, cp->a1,
> +						       cp->a2, cp->a3, temp, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_START:
> +#if defined(CONFIG_32BIT)
> +		temp = ((uint64_t)cp->a4 << 32) | cp->a3;
> +#else
> +		temp = cp->a3;
> +#endif
> +		ret = kvm_riscv_vcpu_pmu_ctr_start(vcpu, cp->a0, cp->a1, cp->a2,
> +						   temp, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_STOP:
> +		ret = kvm_riscv_vcpu_pmu_ctr_stop(vcpu, cp->a0, cp->a1, cp->a2, retdata);
> +		break;
> +	case SBI_EXT_PMU_COUNTER_FW_READ:
> +		ret = kvm_riscv_vcpu_pmu_ctr_read(vcpu, cp->a0, retdata);
> +		break;
> +	default:
> +		retdata->err_val = SBI_ERR_NOT_SUPPORTED;
> +	}
> +
> +	return ret;
> +}
> +
> +unsigned long kvm_sbi_ext_pmu_probe(struct kvm_vcpu *vcpu)
> +{
> +	struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> +
> +	return kvpmu->init_done;
> +}
> +
> +const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> +	.extid_start = SBI_EXT_PMU,
> +	.extid_end = SBI_EXT_PMU,
> +	.handler = kvm_sbi_ext_pmu_handler,
> +	.probe = kvm_sbi_ext_pmu_probe,
> +};
> -- 
> 2.25.1
>

Thanks,
drew

On Thu, Feb 2, 2023 at 9:29 AM Andrew Jones <ajones@ventanamicro.com> wrote:
>
> On Wed, Feb 01, 2023 at 03:12:44PM -0800, Atish Patra wrote:
> > SBI PMU extension allows KVM guests to configure/start/stop/query about
> > the PMU counters in virtualized enviornment as well.
> >
> > In order to allow that, KVM implements the entire SBI PMU extension.
> >
> > Reviewed-by: Anup Patel <anup@brainfault.org>
> > Signed-off-by: Atish Patra <atishp@rivosinc.com>
> > ---
> >  arch/riscv/kvm/Makefile       |  2 +-
> >  arch/riscv/kvm/vcpu_sbi.c     | 11 +++++
> >  arch/riscv/kvm/vcpu_sbi_pmu.c | 85 +++++++++++++++++++++++++++++++++++
> >  3 files changed, 97 insertions(+), 1 deletion(-)
> >  create mode 100644 arch/riscv/kvm/vcpu_sbi_pmu.c
> >
> > diff --git a/arch/riscv/kvm/Makefile b/arch/riscv/kvm/Makefile
> > index 5de1053..278e97c 100644
> > --- a/arch/riscv/kvm/Makefile
> > +++ b/arch/riscv/kvm/Makefile
> > @@ -25,4 +25,4 @@ kvm-y += vcpu_sbi_base.o
> >  kvm-y += vcpu_sbi_replace.o
> >  kvm-y += vcpu_sbi_hsm.o
> >  kvm-y += vcpu_timer.o
> > -kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o
> > +kvm-$(CONFIG_RISCV_PMU_SBI) += vcpu_pmu.o vcpu_sbi_pmu.o
> > diff --git a/arch/riscv/kvm/vcpu_sbi.c b/arch/riscv/kvm/vcpu_sbi.c
> > index fe2897e..15fde15 100644
> > --- a/arch/riscv/kvm/vcpu_sbi.c
> > +++ b/arch/riscv/kvm/vcpu_sbi.c
> > @@ -20,6 +20,16 @@ static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_v01 = {
> >  };
> >  #endif
> >
> > +#ifdef CONFIG_RISCV_PMU_SBI
> > +extern const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu;
> > +#else
> > +static const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> > +     .extid_start = -1UL,
> > +     .extid_end = -1UL,
> > +     .handler = NULL,
> > +};
> > +#endif
> > +
> >  static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
> >       &vcpu_sbi_ext_v01,
> >       &vcpu_sbi_ext_base,
> > @@ -28,6 +38,7 @@ static const struct kvm_vcpu_sbi_extension *sbi_ext[] = {
> >       &vcpu_sbi_ext_rfence,
> >       &vcpu_sbi_ext_srst,
> >       &vcpu_sbi_ext_hsm,
> > +     &vcpu_sbi_ext_pmu,
> >       &vcpu_sbi_ext_experimental,
> >       &vcpu_sbi_ext_vendor,
> >  };
> > diff --git a/arch/riscv/kvm/vcpu_sbi_pmu.c b/arch/riscv/kvm/vcpu_sbi_pmu.c
> > new file mode 100644
> > index 0000000..e028b0a
> > --- /dev/null
> > +++ b/arch/riscv/kvm/vcpu_sbi_pmu.c
> > @@ -0,0 +1,85 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright (c) 2023 Rivos Inc
> > + *
> > + * Authors:
> > + *     Atish Patra <atishp@rivosinc.com>
> > + */
> > +
> > +#include <linux/errno.h>
> > +#include <linux/err.h>
> > +#include <linux/kvm_host.h>
> > +#include <asm/csr.h>
> > +#include <asm/sbi.h>
> > +#include <asm/kvm_vcpu_sbi.h>
> > +
> > +static int kvm_sbi_ext_pmu_handler(struct kvm_vcpu *vcpu, struct kvm_run *run,
> > +                                struct kvm_vcpu_sbi_return *retdata)
> > +{
> > +     int ret = 0;
> > +     struct kvm_cpu_context *cp = &vcpu->arch.guest_context;
> > +     struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> > +     unsigned long funcid = cp->a6;
> > +     uint64_t temp;
> > +
> > +     /* Return not supported if PMU is not initialized */
> > +     if (!kvpmu->init_done)
> > +             return -EINVAL;
>
> Shouldn't this be the following?
>
>   if (!kvpmu->init_done)
>      retdata->err_val = SBI_ERR_NOT_SUPPORTED;
>      return 0;
>   }
>

This condition presents an additional sanity check and indicates a bug
in the guest code
where it invokes PMU extension calls when the probe function already
returns NOT_SUPPORTED.

Earlier, I was in two minds whether this should just return not
supported to the guest or report an
error. However, given that it's just a driver in the guest, the error
shouldn't be fatal. It should just
return the sbi error back to the guest.

Thanks for catching this. I have fixed it in v5.

> > +
> > +     switch (funcid) {
> > +     case SBI_EXT_PMU_NUM_COUNTERS:
> > +             ret = kvm_riscv_vcpu_pmu_num_ctrs(vcpu, retdata);
> > +             break;
> > +     case SBI_EXT_PMU_COUNTER_GET_INFO:
> > +             ret = kvm_riscv_vcpu_pmu_ctr_info(vcpu, cp->a0, retdata);
> > +             break;
> > +     case SBI_EXT_PMU_COUNTER_CFG_MATCH:
> > +#if defined(CONFIG_32BIT)
> > +             temp = ((uint64_t)cp->a5 << 32) | cp->a4;
> > +#else
> > +             temp = cp->a4;
> > +#endif
> > +             /*
> > +              * This can fail if perf core framework fails to create an event.
> > +              * Forward the error to the user space because its an error happened
>
> "Forward the error to userspace because it's an error which happened within
>  the host kernel."
>

Fixed.

> > +              * within host kernel. The other option would be convert this to
>                                                                 ^ to
> > +              * an SBI error and forward to the guest.
> > +              */
> > +             ret = kvm_riscv_vcpu_pmu_ctr_cfg_match(vcpu, cp->a0, cp->a1,
> > +                                                    cp->a2, cp->a3, temp, retdata);
> > +             break;
> > +     case SBI_EXT_PMU_COUNTER_START:
> > +#if defined(CONFIG_32BIT)
> > +             temp = ((uint64_t)cp->a4 << 32) | cp->a3;
> > +#else
> > +             temp = cp->a3;
> > +#endif
> > +             ret = kvm_riscv_vcpu_pmu_ctr_start(vcpu, cp->a0, cp->a1, cp->a2,
> > +                                                temp, retdata);
> > +             break;
> > +     case SBI_EXT_PMU_COUNTER_STOP:
> > +             ret = kvm_riscv_vcpu_pmu_ctr_stop(vcpu, cp->a0, cp->a1, cp->a2, retdata);
> > +             break;
> > +     case SBI_EXT_PMU_COUNTER_FW_READ:
> > +             ret = kvm_riscv_vcpu_pmu_ctr_read(vcpu, cp->a0, retdata);
> > +             break;
> > +     default:
> > +             retdata->err_val = SBI_ERR_NOT_SUPPORTED;
> > +     }
> > +
> > +     return ret;
> > +}
> > +
> > +unsigned long kvm_sbi_ext_pmu_probe(struct kvm_vcpu *vcpu)
> > +{
> > +     struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu);
> > +
> > +     return kvpmu->init_done;
> > +}
> > +
> > +const struct kvm_vcpu_sbi_extension vcpu_sbi_ext_pmu = {
> > +     .extid_start = SBI_EXT_PMU,
> > +     .extid_end = SBI_EXT_PMU,
> > +     .handler = kvm_sbi_ext_pmu_handler,
> > +     .probe = kvm_sbi_ext_pmu_probe,
> > +};
> > --
> > 2.25.1
> >
>
> Thanks,
> drew