Hi,
On Mon, Mar 07, 2022 at 02:42:43PM +0000, Sebastian Ene wrote:
> This patch adds support for stolen time by sharing a memory region
> with the guest which will be used by the hypervisor to store the stolen
> time information. Reserve a 64kb MMIO memory region after the RTC peripheral
> to be used by pvtime. The exact format of the structure stored by the
> hypervisor is described in the ARM DEN0057A document.
>
> Signed-off-by: Sebastian Ene <sebastianene@xxxxxxxxxx>
> ---
> Makefile | 1 +
> arm/aarch64/arm-cpu.c | 2 +-
> arm/aarch64/include/kvm/kvm-cpu-arch.h | 2 +
> arm/aarch64/pvtime.c | 108 +++++++++++++++++++++++++
> arm/include/arm-common/kvm-arch.h | 6 +-
> arm/kvm-cpu.c | 1 +
> include/kvm/kvm-config.h | 1 +
> 7 files changed, 119 insertions(+), 2 deletions(-)
> create mode 100644 arm/aarch64/pvtime.c
>
> diff --git a/Makefile b/Makefile
> index f251147..e9121dc 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -182,6 +182,7 @@ ifeq ($(ARCH), arm64)
> OBJS += arm/aarch64/arm-cpu.o
> OBJS += arm/aarch64/kvm-cpu.o
> OBJS += arm/aarch64/kvm.o
> + OBJS += arm/aarch64/pvtime.o
> ARCH_INCLUDE := $(HDRS_ARM_COMMON)
> ARCH_INCLUDE += -Iarm/aarch64/include
>
> diff --git a/arm/aarch64/arm-cpu.c b/arm/aarch64/arm-cpu.c
> index d7572b7..7e4a3c1 100644
> --- a/arm/aarch64/arm-cpu.c
> +++ b/arm/aarch64/arm-cpu.c
> @@ -22,7 +22,7 @@ static void generate_fdt_nodes(void *fdt, struct kvm *kvm)
> static int arm_cpu__vcpu_init(struct kvm_cpu *vcpu)
> {
> vcpu->generate_fdt_nodes = generate_fdt_nodes;
> - return 0;
> + return kvm_cpu__setup_pvtime(vcpu);
> }
>
> static struct kvm_arm_target target_generic_v8 = {
> diff --git a/arm/aarch64/include/kvm/kvm-cpu-arch.h b/arm/aarch64/include/kvm/kvm-cpu-arch.h
> index 8dfb82e..35996dc 100644
> --- a/arm/aarch64/include/kvm/kvm-cpu-arch.h
> +++ b/arm/aarch64/include/kvm/kvm-cpu-arch.h
> @@ -19,5 +19,7 @@
>
> void kvm_cpu__select_features(struct kvm *kvm, struct kvm_vcpu_init *init);
> int kvm_cpu__configure_features(struct kvm_cpu *vcpu);
> +int kvm_cpu__setup_pvtime(struct kvm_cpu *vcpu);
> +int kvm_cpu__teardown_pvtime(struct kvm *kvm);
>
> #endif /* KVM__KVM_CPU_ARCH_H */
> diff --git a/arm/aarch64/pvtime.c b/arm/aarch64/pvtime.c
> new file mode 100644
> index 0000000..4db2e9f
> --- /dev/null
> +++ b/arm/aarch64/pvtime.c
> @@ -0,0 +1,108 @@
> +#include "kvm/kvm.h"
> +#include "kvm/kvm-cpu.h"
> +#include "kvm/util.h"
> +
> +#include <linux/byteorder.h>
> +#include <linux/types.h>
> +
> +#define ARM_PVTIME_STRUCT_SIZE (64)
> +
> +struct pvtime_data_priv {
> + bool is_failed_cfg;
> + void *usr_mem;
> +};
> +
> +static struct pvtime_data_priv pvtime_data = {
> + .is_failed_cfg = true,
> + .usr_mem = NULL
> +};
> +
> +static int pvtime__alloc_region(struct kvm *kvm)
> +{
> + char *mem;
> + int ret = 0;
> +
> + mem = mmap(NULL, ARM_PVTIME_BASE, PROT_RW,
> + MAP_ANON_NORESERVE, -1, 0);
> + if (mem == MAP_FAILED)
> + return -errno;
> +
> + ret = kvm__register_ram(kvm, ARM_PVTIME_BASE,
> + ARM_PVTIME_BASE, mem);
> + if (ret) {
> + munmap(mem, ARM_PVTIME_BASE);
> + return ret;
> + }
> +
> + pvtime_data.usr_mem = mem;
> + return ret;
> +}
> +
> +static int pvtime__teardown_region(struct kvm *kvm)
> +{
> + if (pvtime_data.usr_mem == NULL)
> + return 0;
> +
> + kvm__destroy_mem(kvm, ARM_PVTIME_BASE,
> + ARM_PVTIME_BASE, pvtime_data.usr_mem);
> + munmap(pvtime_data.usr_mem, ARM_PVTIME_BASE);
> + pvtime_data.usr_mem = NULL;
> + return 0;
> +}
> +
> +int kvm_cpu__setup_pvtime(struct kvm_cpu *vcpu)
> +{
> + int ret;
> + bool has_stolen_time;
> + u64 pvtime_guest_addr = ARM_PVTIME_BASE + vcpu->cpu_id *
> + ARM_PVTIME_STRUCT_SIZE;
> + struct kvm_config *kvm_cfg = NULL;
> + struct kvm_device_attr pvtime_attr = (struct kvm_device_attr) {
> + .group = KVM_ARM_VCPU_PVTIME_CTRL,
> + .addr = KVM_ARM_VCPU_PVTIME_IPA
> + };
> +
> + kvm_cfg = &vcpu->kvm->cfg;
> + if (kvm_cfg->no_pvtime)
> + return 0;
> +
> + if (!pvtime_data.is_failed_cfg)
> + return -ENOTSUP;
> +
> + has_stolen_time = kvm__supports_extension(vcpu->kvm,
> + KVM_CAP_STEAL_TIME);
> + if (!has_stolen_time) {
> + kvm_cfg->no_pvtime = true;
> + return 0;
> + }
> +
> + ret = ioctl(vcpu->vcpu_fd, KVM_HAS_DEVICE_ATTR, &pvtime_attr);
> + if (ret) {
> + perror("KVM_HAS_DEVICE_ATTR failed\n");
> + goto out_err;
> + }
> +
> + if (!pvtime_data.usr_mem) {
> + ret = pvtime__alloc_region(vcpu->kvm);
> + if (ret) {
> + perror("Failed allocating pvtime region\n");
> + goto out_err;
> + }
> + }
> +
> + pvtime_attr.addr = (u64)&pvtime_guest_addr;
> + ret = ioctl(vcpu->vcpu_fd, KVM_SET_DEVICE_ATTR, &pvtime_attr);
> + if (!ret)
> + return 0;
> +
> + perror("KVM_SET_DEVICE_ATTR failed\n");
> + pvtime__teardown_region(vcpu->kvm);
> +out_err:
> + pvtime_data.is_failed_cfg = false;
kvm_cpu__init() calls kvm_cpu__arch_init()->kvm_cpu__setup_pvtime() for each
VCPU from the main thread (so sequentually), not from the VCPU threads. If this
function returns an error, kvm_cpu__arch_init() calls die(), which means that
kvmtool will terminate without calling kvm_cpu__setup_pvtime() for the other
VCPUs.
What I'm trying to say is that the field is_failed_cfg is not useful, because if
one VCPU fails initialization, then kvmtool will not attempt to initialize the
rest of the VCPUs.
If you drop is_failed_cfg you can also drop the pvtime_data_priv struct and use
a static user_mem variable (up to you).
> + return ret;
> +}
> +
> +int kvm_cpu__teardown_pvtime(struct kvm *kvm)
> +{
> + return pvtime__teardown_region(kvm);
> +}
> diff --git a/arm/include/arm-common/kvm-arch.h b/arm/include/arm-common/kvm-arch.h
> index c645ac0..43b1f77 100644
> --- a/arm/include/arm-common/kvm-arch.h
> +++ b/arm/include/arm-common/kvm-arch.h
> @@ -15,7 +15,8 @@
> * | PCI |////| plat | | | | |
> * | I/O |////| MMIO: | Flash | virtio | GIC | PCI | DRAM
> * | space |////| UART, | | MMIO | | (AXI) |
> - * | |////| RTC | | | | |
> + * | |////| RTC, | | | | |
> + * | |////| PVTIME| | | | |
> * +-------+----+-------+-------+--------+-----+---------+---......
> */
>
> @@ -34,6 +35,9 @@
> #define ARM_RTC_MMIO_BASE (ARM_UART_MMIO_BASE + ARM_UART_MMIO_SIZE)
> #define ARM_RTC_MMIO_SIZE 0x10000
>
> +#define ARM_PVTIME_BASE (ARM_RTC_MMIO_BASE + ARM_RTC_MMIO_SIZE)
> +#define ARM_PVTIME_SIZE SZ_64K
> +
> #define KVM_FLASH_MMIO_BASE (ARM_MMIO_AREA + 0x1000000)
> #define KVM_FLASH_MAX_SIZE 0x1000000
>
> diff --git a/arm/kvm-cpu.c b/arm/kvm-cpu.c
> index 84ac1e9..00660d6 100644
> --- a/arm/kvm-cpu.c
> +++ b/arm/kvm-cpu.c
> @@ -144,6 +144,7 @@ void kvm_cpu__arch_nmi(struct kvm_cpu *cpu)
>
> void kvm_cpu__delete(struct kvm_cpu *vcpu)
> {
> + kvm_cpu__teardown_pvtime(vcpu->kvm);
This causes compilation for aarch32 to fail:
arm/kvm-cpu.c: In function ‘kvm_cpu__delete’:
arm/kvm-cpu.c:147:2: error: implicit declaration of function ‘kvm_cpu__teardown_pvtime’ [-Werror=implicit-function-declaration]
147 | kvm_cpu__teardown_pvtime(vcpu->kvm);
| ^~~~~~~~~~~~~~~~~~~~~~~~
arm/kvm-cpu.c:147:2: error: nested extern declaration of ‘kvm_cpu__teardown_pvtime’ [-Werror=nested-externs]
cc1: all warnings being treated as errors
make: *** [Makefile:482: arm/kvm-cpu.o] Error 1
The reason for that is that there is no stub for kvm_cpu__teardown_pvtime() for
aarch32. This fixes the compilation error for me:
diff --git a/arm/aarch32/include/kvm/kvm-cpu-arch.h b/arm/aarch32/include/kvm/kvm-cpu-arch.h
index 780e0e2f0934..ae77a136d0ce 100644
--- a/arm/aarch32/include/kvm/kvm-cpu-arch.h
+++ b/arm/aarch32/include/kvm/kvm-cpu-arch.h
@@ -19,5 +19,9 @@ static inline int kvm_cpu__configure_features(struct kvm_cpu *vcpu)
{
return 0;
}
+static inline int kvm_cpu__teardown_pvtime(struct kvm *kvm)
+{
+ return 0;
+}
#endif /* KVM__KVM_CPU_ARCH_H */
Thanks,
Alex
> free(vcpu);
> }
>
> diff --git a/include/kvm/kvm-config.h b/include/kvm/kvm-config.h
> index 6a5720c..48adf27 100644
> --- a/include/kvm/kvm-config.h
> +++ b/include/kvm/kvm-config.h
> @@ -62,6 +62,7 @@ struct kvm_config {
> bool no_dhcp;
> bool ioport_debug;
> bool mmio_debug;
> + bool no_pvtime;
> };
>
> #endif
> --
> 2.35.1.616.g0bdcbb4464-goog
>
_______________________________________________
kvmarm mailing list
kvmarm@xxxxxxxxxxxxxxxxxxxxx
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
