* [PATCH v2 00/12] Add riscv kvm accel support
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang
This series adds both riscv32 and riscv64 kvm support, and implements
migration based on riscv.
Because of RISC-V KVM has been merged into the Linux master, so this
series are changed from RFC to patch.
Several steps to use this:
1. Build emulation
$ ./configure --target-list=riscv64-softmmu
$ make -j$(nproc)
2. Build kernel
3. Build QEMU VM
Cross built in riscv toolchain.
$ PKG_CONFIG_LIBDIR=<toolchain pkgconfig path>
$ export PKG_CONFIG_SYSROOT_DIR=<toolchain sysroot path>
$ ./configure --target-list=riscv64-softmmu --enable-kvm \
--cross-prefix=riscv64-linux-gnu- --disable-libiscsi --disable-glusterfs \
--disable-libusb --disable-usb-redir --audio-drv-list= --disable-opengl \
--disable-libxml2
$ make -j$(nproc)
4. Start emulation
$ ./qemu-system-riscv64 -M virt -m 4096M -cpu rv64,x-h=true -nographic \
-name guest=riscv-hyp,debug-threads=on \
-smp 4 \
-bios ./fw_jump.bin \
-kernel ./Image \
-drive file=./hyp.img,format=raw,id=hd0 \
-device virtio-blk-device,drive=hd0 \
-append "root=/dev/vda rw console=ttyS0 earlycon=sbi"
5. Start kvm-acceled QEMU VM in emulation
$ ./qemu-system-riscv64 -M virt,accel=kvm -m 1024M -cpu host -nographic \
-name guest=riscv-guset \
-smp 2 \
-bios none \
-kernel ./Image \
-drive file=./guest.img,format=raw,id=hd0 \
-device virtio-blk-device,drive=hd0 \
-append "root=/dev/vda rw console=ttyS0 earlycon=sbi"
Changes since patch v1
- Rebase on recent commit a216e7cf119c91ffdf5931834a1a030ebea40d70
- Sync-up headers with Linux-5.16-rc4.
- Fixbug in kvm_arch_init_vcpu.
- Create a macro for get and put regs csr.
- Start kernel directly when kvm_enabled.
- Use riscv_cpu_set_irq to inject KVM interrupts.
- Use the Semihosting Console API for RISC-V kvm handle sbi.
- Update vmstate_riscv_cpu version id.
Placing kvm_timer into a subsection.
Changes since RFC v6
- Rebase on recent commit 8627edfb3f1fca24a96a0954148885c3241c10f8
- Sync-up headers with Linux-5.16-rc1
Changes since RFC v5
- Rebase on QEMU v6.1.0-rc1 and kvm-riscv linux v19.
- Move kvm interrupt setting to riscv_cpu_update_mip().
- Replace __u64 with uint64_t.
Changes since RFC v4
- Rebase on QEMU v6.0.0-rc2 and kvm-riscv linux v17.
- Remove time scaling support as software solution is incomplete.
Because it will cause unacceptable performance degradation. and
We will post a better solution.
- Revise according to Alistair's review comments.
- Remove compile time XLEN checks in kvm_riscv_reg_id
- Surround TYPE_RISCV_CPU_HOST definition by CONFIG_KVM and share
it between RV32 and RV64.
- Add kvm-stub.c for reduce unnecessary compilation checks.
- Add riscv_setup_direct_kernel() to direct boot kernel for KVM.
Changes since RFC v3
- Rebase on QEMU v5.2.0-rc2 and kvm-riscv linux v15.
- Add time scaling support(New patches 13, 14 and 15).
- Fix the bug that guest vm can't reboot.
Changes since RFC v2
- Fix checkpatch error at target/riscv/sbi_ecall_interface.h.
- Add riscv migration support.
Changes since RFC v1
- Add separate SBI ecall interface header.
- Add riscv32 kvm accel support.
Yifei Jiang (12):
update-linux-headers: Add asm-riscv/kvm.h
target/riscv: Add target/riscv/kvm.c to place the public kvm interface
target/riscv: Implement function kvm_arch_init_vcpu
target/riscv: Implement kvm_arch_get_registers
target/riscv: Implement kvm_arch_put_registers
target/riscv: Support start kernel directly by KVM
target/riscv: Support setting external interrupt by KVM
target/riscv: Handle KVM_EXIT_RISCV_SBI exit
target/riscv: Add host cpu type
target/riscv: Add kvm_riscv_get/put_regs_timer
target/riscv: Implement virtual time adjusting with vm state changing
target/riscv: Support virtual time context synchronization
hw/riscv/boot.c | 11 +
hw/riscv/virt.c | 54 ++-
include/hw/riscv/boot.h | 1 +
linux-headers/asm-riscv/kvm.h | 128 +++++++
linux-headers/linux/kvm.h | 8 +
meson.build | 2 +
target/riscv/cpu.c | 29 +-
target/riscv/cpu.h | 11 +
target/riscv/kvm-stub.c | 30 ++
target/riscv/kvm.c | 528 +++++++++++++++++++++++++++++
target/riscv/kvm_riscv.h | 25 ++
target/riscv/machine.c | 37 +-
target/riscv/meson.build | 1 +
target/riscv/sbi_ecall_interface.h | 72 ++++
14 files changed, 918 insertions(+), 19 deletions(-)
create mode 100644 linux-headers/asm-riscv/kvm.h
create mode 100644 target/riscv/kvm-stub.c
create mode 100644 target/riscv/kvm.c
create mode 100644 target/riscv/kvm_riscv.h
create mode 100644 target/riscv/sbi_ecall_interface.h
--
2.19.1
^ permalink raw reply [flat|nested] 56+ messages in thread
* [PATCH v2 00/12] Add riscv kvm accel support
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang
This series adds both riscv32 and riscv64 kvm support, and implements
migration based on riscv.
Because of RISC-V KVM has been merged into the Linux master, so this
series are changed from RFC to patch.
Several steps to use this:
1. Build emulation
$ ./configure --target-list=riscv64-softmmu
$ make -j$(nproc)
2. Build kernel
3. Build QEMU VM
Cross built in riscv toolchain.
$ PKG_CONFIG_LIBDIR=<toolchain pkgconfig path>
$ export PKG_CONFIG_SYSROOT_DIR=<toolchain sysroot path>
$ ./configure --target-list=riscv64-softmmu --enable-kvm \
--cross-prefix=riscv64-linux-gnu- --disable-libiscsi --disable-glusterfs \
--disable-libusb --disable-usb-redir --audio-drv-list= --disable-opengl \
--disable-libxml2
$ make -j$(nproc)
4. Start emulation
$ ./qemu-system-riscv64 -M virt -m 4096M -cpu rv64,x-h=true -nographic \
-name guest=riscv-hyp,debug-threads=on \
-smp 4 \
-bios ./fw_jump.bin \
-kernel ./Image \
-drive file=./hyp.img,format=raw,id=hd0 \
-device virtio-blk-device,drive=hd0 \
-append "root=/dev/vda rw console=ttyS0 earlycon=sbi"
5. Start kvm-acceled QEMU VM in emulation
$ ./qemu-system-riscv64 -M virt,accel=kvm -m 1024M -cpu host -nographic \
-name guest=riscv-guset \
-smp 2 \
-bios none \
-kernel ./Image \
-drive file=./guest.img,format=raw,id=hd0 \
-device virtio-blk-device,drive=hd0 \
-append "root=/dev/vda rw console=ttyS0 earlycon=sbi"
Changes since patch v1
- Rebase on recent commit a216e7cf119c91ffdf5931834a1a030ebea40d70
- Sync-up headers with Linux-5.16-rc4.
- Fixbug in kvm_arch_init_vcpu.
- Create a macro for get and put regs csr.
- Start kernel directly when kvm_enabled.
- Use riscv_cpu_set_irq to inject KVM interrupts.
- Use the Semihosting Console API for RISC-V kvm handle sbi.
- Update vmstate_riscv_cpu version id.
Placing kvm_timer into a subsection.
Changes since RFC v6
- Rebase on recent commit 8627edfb3f1fca24a96a0954148885c3241c10f8
- Sync-up headers with Linux-5.16-rc1
Changes since RFC v5
- Rebase on QEMU v6.1.0-rc1 and kvm-riscv linux v19.
- Move kvm interrupt setting to riscv_cpu_update_mip().
- Replace __u64 with uint64_t.
Changes since RFC v4
- Rebase on QEMU v6.0.0-rc2 and kvm-riscv linux v17.
- Remove time scaling support as software solution is incomplete.
Because it will cause unacceptable performance degradation. and
We will post a better solution.
- Revise according to Alistair's review comments.
- Remove compile time XLEN checks in kvm_riscv_reg_id
- Surround TYPE_RISCV_CPU_HOST definition by CONFIG_KVM and share
it between RV32 and RV64.
- Add kvm-stub.c for reduce unnecessary compilation checks.
- Add riscv_setup_direct_kernel() to direct boot kernel for KVM.
Changes since RFC v3
- Rebase on QEMU v5.2.0-rc2 and kvm-riscv linux v15.
- Add time scaling support(New patches 13, 14 and 15).
- Fix the bug that guest vm can't reboot.
Changes since RFC v2
- Fix checkpatch error at target/riscv/sbi_ecall_interface.h.
- Add riscv migration support.
Changes since RFC v1
- Add separate SBI ecall interface header.
- Add riscv32 kvm accel support.
Yifei Jiang (12):
update-linux-headers: Add asm-riscv/kvm.h
target/riscv: Add target/riscv/kvm.c to place the public kvm interface
target/riscv: Implement function kvm_arch_init_vcpu
target/riscv: Implement kvm_arch_get_registers
target/riscv: Implement kvm_arch_put_registers
target/riscv: Support start kernel directly by KVM
target/riscv: Support setting external interrupt by KVM
target/riscv: Handle KVM_EXIT_RISCV_SBI exit
target/riscv: Add host cpu type
target/riscv: Add kvm_riscv_get/put_regs_timer
target/riscv: Implement virtual time adjusting with vm state changing
target/riscv: Support virtual time context synchronization
hw/riscv/boot.c | 11 +
hw/riscv/virt.c | 54 ++-
include/hw/riscv/boot.h | 1 +
linux-headers/asm-riscv/kvm.h | 128 +++++++
linux-headers/linux/kvm.h | 8 +
meson.build | 2 +
target/riscv/cpu.c | 29 +-
target/riscv/cpu.h | 11 +
target/riscv/kvm-stub.c | 30 ++
target/riscv/kvm.c | 528 +++++++++++++++++++++++++++++
target/riscv/kvm_riscv.h | 25 ++
target/riscv/machine.c | 37 +-
target/riscv/meson.build | 1 +
target/riscv/sbi_ecall_interface.h | 72 ++++
14 files changed, 918 insertions(+), 19 deletions(-)
create mode 100644 linux-headers/asm-riscv/kvm.h
create mode 100644 target/riscv/kvm-stub.c
create mode 100644 target/riscv/kvm.c
create mode 100644 target/riscv/kvm_riscv.h
create mode 100644 target/riscv/sbi_ecall_interface.h
--
2.19.1
^ permalink raw reply [flat|nested] 56+ messages in thread
* [PATCH v2 01/12] update-linux-headers: Add asm-riscv/kvm.h
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Add asm-riscv/kvm.h for RISC-V KVM, and update linux/kvm.h
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
| 128 ++++++++++++++++++++++++++++++++++
| 8 +++
2 files changed, 136 insertions(+)
create mode 100644 linux-headers/asm-riscv/kvm.h
--git a/linux-headers/asm-riscv/kvm.h b/linux-headers/asm-riscv/kvm.h
new file mode 100644
index 0000000000..f808ad1ce5
--- /dev/null
+++ b/linux-headers/asm-riscv/kvm.h
@@ -0,0 +1,128 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ * Anup Patel <anup.patel@wdc.com>
+ */
+
+#ifndef __LINUX_KVM_RISCV_H
+#define __LINUX_KVM_RISCV_H
+
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+
+#define __KVM_HAVE_READONLY_MEM
+
+#define KVM_COALESCED_MMIO_PAGE_OFFSET 1
+
+#define KVM_INTERRUPT_SET -1U
+#define KVM_INTERRUPT_UNSET -2U
+
+/* for KVM_GET_REGS and KVM_SET_REGS */
+struct kvm_regs {
+};
+
+/* for KVM_GET_FPU and KVM_SET_FPU */
+struct kvm_fpu {
+};
+
+/* KVM Debug exit structure */
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
+/* definition of registers in kvm_run */
+struct kvm_sync_regs {
+};
+
+/* for KVM_GET_SREGS and KVM_SET_SREGS */
+struct kvm_sregs {
+};
+
+/* CONFIG registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_config {
+ unsigned long isa;
+};
+
+/* CORE registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_core {
+ struct user_regs_struct regs;
+ unsigned long mode;
+};
+
+/* Possible privilege modes for kvm_riscv_core */
+#define KVM_RISCV_MODE_S 1
+#define KVM_RISCV_MODE_U 0
+
+/* CSR registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_csr {
+ unsigned long sstatus;
+ unsigned long sie;
+ unsigned long stvec;
+ unsigned long sscratch;
+ unsigned long sepc;
+ unsigned long scause;
+ unsigned long stval;
+ unsigned long sip;
+ unsigned long satp;
+ unsigned long scounteren;
+};
+
+/* TIMER registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_timer {
+ __u64 frequency;
+ __u64 time;
+ __u64 compare;
+ __u64 state;
+};
+
+/* Possible states for kvm_riscv_timer */
+#define KVM_RISCV_TIMER_STATE_OFF 0
+#define KVM_RISCV_TIMER_STATE_ON 1
+
+#define KVM_REG_SIZE(id) \
+ (1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
+
+/* If you need to interpret the index values, here is the key: */
+#define KVM_REG_RISCV_TYPE_MASK 0x00000000FF000000
+#define KVM_REG_RISCV_TYPE_SHIFT 24
+
+/* Config registers are mapped as type 1 */
+#define KVM_REG_RISCV_CONFIG (0x01 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CONFIG_REG(name) \
+ (offsetof(struct kvm_riscv_config, name) / sizeof(unsigned long))
+
+/* Core registers are mapped as type 2 */
+#define KVM_REG_RISCV_CORE (0x02 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CORE_REG(name) \
+ (offsetof(struct kvm_riscv_core, name) / sizeof(unsigned long))
+
+/* Control and status registers are mapped as type 3 */
+#define KVM_REG_RISCV_CSR (0x03 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CSR_REG(name) \
+ (offsetof(struct kvm_riscv_csr, name) / sizeof(unsigned long))
+
+/* Timer registers are mapped as type 4 */
+#define KVM_REG_RISCV_TIMER (0x04 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_TIMER_REG(name) \
+ (offsetof(struct kvm_riscv_timer, name) / sizeof(__u64))
+
+/* F extension registers are mapped as type 5 */
+#define KVM_REG_RISCV_FP_F (0x05 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_FP_F_REG(name) \
+ (offsetof(struct __riscv_f_ext_state, name) / sizeof(__u32))
+
+/* D extension registers are mapped as type 6 */
+#define KVM_REG_RISCV_FP_D (0x06 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_FP_D_REG(name) \
+ (offsetof(struct __riscv_d_ext_state, name) / sizeof(__u64))
+
+#endif
+
+#endif /* __LINUX_KVM_RISCV_H */
--git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h
index bcaf66cc4d..5e290c3c3e 100644
--- a/linux-headers/linux/kvm.h
+++ b/linux-headers/linux/kvm.h
@@ -269,6 +269,7 @@ struct kvm_xen_exit {
#define KVM_EXIT_AP_RESET_HOLD 32
#define KVM_EXIT_X86_BUS_LOCK 33
#define KVM_EXIT_XEN 34
+#define KVM_EXIT_RISCV_SBI 35
/* For KVM_EXIT_INTERNAL_ERROR */
/* Emulate instruction failed. */
@@ -469,6 +470,13 @@ struct kvm_run {
} msr;
/* KVM_EXIT_XEN */
struct kvm_xen_exit xen;
+ /* KVM_EXIT_RISCV_SBI */
+ struct {
+ unsigned long extension_id;
+ unsigned long function_id;
+ unsigned long args[6];
+ unsigned long ret[2];
+ } riscv_sbi;
/* Fix the size of the union. */
char padding[256];
};
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 01/12] update-linux-headers: Add asm-riscv/kvm.h
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Add asm-riscv/kvm.h for RISC-V KVM, and update linux/kvm.h
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
| 128 ++++++++++++++++++++++++++++++++++
| 8 +++
2 files changed, 136 insertions(+)
create mode 100644 linux-headers/asm-riscv/kvm.h
--git a/linux-headers/asm-riscv/kvm.h b/linux-headers/asm-riscv/kvm.h
new file mode 100644
index 0000000000..f808ad1ce5
--- /dev/null
+++ b/linux-headers/asm-riscv/kvm.h
@@ -0,0 +1,128 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+/*
+ * Copyright (C) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ * Anup Patel <anup.patel@wdc.com>
+ */
+
+#ifndef __LINUX_KVM_RISCV_H
+#define __LINUX_KVM_RISCV_H
+
+#ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+
+#define __KVM_HAVE_READONLY_MEM
+
+#define KVM_COALESCED_MMIO_PAGE_OFFSET 1
+
+#define KVM_INTERRUPT_SET -1U
+#define KVM_INTERRUPT_UNSET -2U
+
+/* for KVM_GET_REGS and KVM_SET_REGS */
+struct kvm_regs {
+};
+
+/* for KVM_GET_FPU and KVM_SET_FPU */
+struct kvm_fpu {
+};
+
+/* KVM Debug exit structure */
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
+/* definition of registers in kvm_run */
+struct kvm_sync_regs {
+};
+
+/* for KVM_GET_SREGS and KVM_SET_SREGS */
+struct kvm_sregs {
+};
+
+/* CONFIG registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_config {
+ unsigned long isa;
+};
+
+/* CORE registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_core {
+ struct user_regs_struct regs;
+ unsigned long mode;
+};
+
+/* Possible privilege modes for kvm_riscv_core */
+#define KVM_RISCV_MODE_S 1
+#define KVM_RISCV_MODE_U 0
+
+/* CSR registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_csr {
+ unsigned long sstatus;
+ unsigned long sie;
+ unsigned long stvec;
+ unsigned long sscratch;
+ unsigned long sepc;
+ unsigned long scause;
+ unsigned long stval;
+ unsigned long sip;
+ unsigned long satp;
+ unsigned long scounteren;
+};
+
+/* TIMER registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
+struct kvm_riscv_timer {
+ __u64 frequency;
+ __u64 time;
+ __u64 compare;
+ __u64 state;
+};
+
+/* Possible states for kvm_riscv_timer */
+#define KVM_RISCV_TIMER_STATE_OFF 0
+#define KVM_RISCV_TIMER_STATE_ON 1
+
+#define KVM_REG_SIZE(id) \
+ (1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT))
+
+/* If you need to interpret the index values, here is the key: */
+#define KVM_REG_RISCV_TYPE_MASK 0x00000000FF000000
+#define KVM_REG_RISCV_TYPE_SHIFT 24
+
+/* Config registers are mapped as type 1 */
+#define KVM_REG_RISCV_CONFIG (0x01 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CONFIG_REG(name) \
+ (offsetof(struct kvm_riscv_config, name) / sizeof(unsigned long))
+
+/* Core registers are mapped as type 2 */
+#define KVM_REG_RISCV_CORE (0x02 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CORE_REG(name) \
+ (offsetof(struct kvm_riscv_core, name) / sizeof(unsigned long))
+
+/* Control and status registers are mapped as type 3 */
+#define KVM_REG_RISCV_CSR (0x03 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_CSR_REG(name) \
+ (offsetof(struct kvm_riscv_csr, name) / sizeof(unsigned long))
+
+/* Timer registers are mapped as type 4 */
+#define KVM_REG_RISCV_TIMER (0x04 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_TIMER_REG(name) \
+ (offsetof(struct kvm_riscv_timer, name) / sizeof(__u64))
+
+/* F extension registers are mapped as type 5 */
+#define KVM_REG_RISCV_FP_F (0x05 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_FP_F_REG(name) \
+ (offsetof(struct __riscv_f_ext_state, name) / sizeof(__u32))
+
+/* D extension registers are mapped as type 6 */
+#define KVM_REG_RISCV_FP_D (0x06 << KVM_REG_RISCV_TYPE_SHIFT)
+#define KVM_REG_RISCV_FP_D_REG(name) \
+ (offsetof(struct __riscv_d_ext_state, name) / sizeof(__u64))
+
+#endif
+
+#endif /* __LINUX_KVM_RISCV_H */
--git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h
index bcaf66cc4d..5e290c3c3e 100644
--- a/linux-headers/linux/kvm.h
+++ b/linux-headers/linux/kvm.h
@@ -269,6 +269,7 @@ struct kvm_xen_exit {
#define KVM_EXIT_AP_RESET_HOLD 32
#define KVM_EXIT_X86_BUS_LOCK 33
#define KVM_EXIT_XEN 34
+#define KVM_EXIT_RISCV_SBI 35
/* For KVM_EXIT_INTERNAL_ERROR */
/* Emulate instruction failed. */
@@ -469,6 +470,13 @@ struct kvm_run {
} msr;
/* KVM_EXIT_XEN */
struct kvm_xen_exit xen;
+ /* KVM_EXIT_RISCV_SBI */
+ struct {
+ unsigned long extension_id;
+ unsigned long function_id;
+ unsigned long args[6];
+ unsigned long ret[2];
+ } riscv_sbi;
/* Fix the size of the union. */
char padding[256];
};
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 02/12] target/riscv: Add target/riscv/kvm.c to place the public kvm interface
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Add target/riscv/kvm.c to place kvm_arch_* function needed by
kvm/kvm-all.c. Meanwhile, add kvm support in meson.build file.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
meson.build | 2 +
target/riscv/kvm.c | 133 +++++++++++++++++++++++++++++++++++++++
target/riscv/meson.build | 1 +
3 files changed, 136 insertions(+)
create mode 100644 target/riscv/kvm.c
diff --git a/meson.build b/meson.build
index 96de1a6ef9..ae35e76ea4 100644
--- a/meson.build
+++ b/meson.build
@@ -77,6 +77,8 @@ elif cpu in ['ppc', 'ppc64']
kvm_targets = ['ppc-softmmu', 'ppc64-softmmu']
elif cpu in ['mips', 'mips64']
kvm_targets = ['mips-softmmu', 'mipsel-softmmu', 'mips64-softmmu', 'mips64el-softmmu']
+elif cpu in ['riscv']
+ kvm_targets = ['riscv32-softmmu', 'riscv64-softmmu']
else
kvm_targets = []
endif
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
new file mode 100644
index 0000000000..687dd4b621
--- /dev/null
+++ b/target/riscv/kvm.c
@@ -0,0 +1,133 @@
+/*
+ * RISC-V implementation of KVM hooks
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include <sys/ioctl.h>
+
+#include <linux/kvm.h>
+
+#include "qemu-common.h"
+#include "qemu/timer.h"
+#include "qemu/error-report.h"
+#include "qemu/main-loop.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
+#include "sysemu/kvm_int.h"
+#include "cpu.h"
+#include "trace.h"
+#include "hw/pci/pci.h"
+#include "exec/memattrs.h"
+#include "exec/address-spaces.h"
+#include "hw/boards.h"
+#include "hw/irq.h"
+#include "qemu/log.h"
+#include "hw/loader.h"
+
+const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
+ KVM_CAP_LAST_INFO
+};
+
+int kvm_arch_get_registers(CPUState *cs)
+{
+ return 0;
+}
+
+int kvm_arch_put_registers(CPUState *cs, int level)
+{
+ return 0;
+}
+
+int kvm_arch_release_virq_post(int virq)
+{
+ return 0;
+}
+
+int kvm_arch_fixup_msi_route(struct kvm_irq_routing_entry *route,
+ uint64_t address, uint32_t data, PCIDevice *dev)
+{
+ return 0;
+}
+
+int kvm_arch_destroy_vcpu(CPUState *cs)
+{
+ return 0;
+}
+
+unsigned long kvm_arch_vcpu_id(CPUState *cpu)
+{
+ return cpu->cpu_index;
+}
+
+void kvm_arch_init_irq_routing(KVMState *s)
+{
+}
+
+int kvm_arch_init_vcpu(CPUState *cs)
+{
+ return 0;
+}
+
+int kvm_arch_msi_data_to_gsi(uint32_t data)
+{
+ abort();
+}
+
+int kvm_arch_add_msi_route_post(struct kvm_irq_routing_entry *route,
+ int vector, PCIDevice *dev)
+{
+ return 0;
+}
+
+int kvm_arch_init(MachineState *ms, KVMState *s)
+{
+ return 0;
+}
+
+int kvm_arch_irqchip_create(KVMState *s)
+{
+ return 0;
+}
+
+int kvm_arch_process_async_events(CPUState *cs)
+{
+ return 0;
+}
+
+void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
+{
+}
+
+MemTxAttrs kvm_arch_post_run(CPUState *cs, struct kvm_run *run)
+{
+ return MEMTXATTRS_UNSPECIFIED;
+}
+
+bool kvm_arch_stop_on_emulation_error(CPUState *cs)
+{
+ return true;
+}
+
+int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
+{
+ return 0;
+}
+
+bool kvm_arch_cpu_check_are_resettable(void)
+{
+ return true;
+}
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index d5e0bc93ea..2faf08a941 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -19,6 +19,7 @@ riscv_ss.add(files(
'bitmanip_helper.c',
'translate.c',
))
+riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
riscv_softmmu_ss = ss.source_set()
riscv_softmmu_ss.add(files(
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 02/12] target/riscv: Add target/riscv/kvm.c to place the public kvm interface
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Add target/riscv/kvm.c to place kvm_arch_* function needed by
kvm/kvm-all.c. Meanwhile, add kvm support in meson.build file.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
meson.build | 2 +
target/riscv/kvm.c | 133 +++++++++++++++++++++++++++++++++++++++
target/riscv/meson.build | 1 +
3 files changed, 136 insertions(+)
create mode 100644 target/riscv/kvm.c
diff --git a/meson.build b/meson.build
index 96de1a6ef9..ae35e76ea4 100644
--- a/meson.build
+++ b/meson.build
@@ -77,6 +77,8 @@ elif cpu in ['ppc', 'ppc64']
kvm_targets = ['ppc-softmmu', 'ppc64-softmmu']
elif cpu in ['mips', 'mips64']
kvm_targets = ['mips-softmmu', 'mipsel-softmmu', 'mips64-softmmu', 'mips64el-softmmu']
+elif cpu in ['riscv']
+ kvm_targets = ['riscv32-softmmu', 'riscv64-softmmu']
else
kvm_targets = []
endif
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
new file mode 100644
index 0000000000..687dd4b621
--- /dev/null
+++ b/target/riscv/kvm.c
@@ -0,0 +1,133 @@
+/*
+ * RISC-V implementation of KVM hooks
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "qemu/osdep.h"
+#include <sys/ioctl.h>
+
+#include <linux/kvm.h>
+
+#include "qemu-common.h"
+#include "qemu/timer.h"
+#include "qemu/error-report.h"
+#include "qemu/main-loop.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
+#include "sysemu/kvm_int.h"
+#include "cpu.h"
+#include "trace.h"
+#include "hw/pci/pci.h"
+#include "exec/memattrs.h"
+#include "exec/address-spaces.h"
+#include "hw/boards.h"
+#include "hw/irq.h"
+#include "qemu/log.h"
+#include "hw/loader.h"
+
+const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
+ KVM_CAP_LAST_INFO
+};
+
+int kvm_arch_get_registers(CPUState *cs)
+{
+ return 0;
+}
+
+int kvm_arch_put_registers(CPUState *cs, int level)
+{
+ return 0;
+}
+
+int kvm_arch_release_virq_post(int virq)
+{
+ return 0;
+}
+
+int kvm_arch_fixup_msi_route(struct kvm_irq_routing_entry *route,
+ uint64_t address, uint32_t data, PCIDevice *dev)
+{
+ return 0;
+}
+
+int kvm_arch_destroy_vcpu(CPUState *cs)
+{
+ return 0;
+}
+
+unsigned long kvm_arch_vcpu_id(CPUState *cpu)
+{
+ return cpu->cpu_index;
+}
+
+void kvm_arch_init_irq_routing(KVMState *s)
+{
+}
+
+int kvm_arch_init_vcpu(CPUState *cs)
+{
+ return 0;
+}
+
+int kvm_arch_msi_data_to_gsi(uint32_t data)
+{
+ abort();
+}
+
+int kvm_arch_add_msi_route_post(struct kvm_irq_routing_entry *route,
+ int vector, PCIDevice *dev)
+{
+ return 0;
+}
+
+int kvm_arch_init(MachineState *ms, KVMState *s)
+{
+ return 0;
+}
+
+int kvm_arch_irqchip_create(KVMState *s)
+{
+ return 0;
+}
+
+int kvm_arch_process_async_events(CPUState *cs)
+{
+ return 0;
+}
+
+void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
+{
+}
+
+MemTxAttrs kvm_arch_post_run(CPUState *cs, struct kvm_run *run)
+{
+ return MEMTXATTRS_UNSPECIFIED;
+}
+
+bool kvm_arch_stop_on_emulation_error(CPUState *cs)
+{
+ return true;
+}
+
+int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
+{
+ return 0;
+}
+
+bool kvm_arch_cpu_check_are_resettable(void)
+{
+ return true;
+}
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index d5e0bc93ea..2faf08a941 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -19,6 +19,7 @@ riscv_ss.add(files(
'bitmanip_helper.c',
'translate.c',
))
+riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
riscv_softmmu_ss = ss.source_set()
riscv_softmmu_ss.add(files(
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 03/12] target/riscv: Implement function kvm_arch_init_vcpu
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get isa info from kvm while kvm init.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 32 +++++++++++++++++++++++++++++++-
1 file changed, 31 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 687dd4b621..ccf3753048 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -38,6 +38,23 @@
#include "qemu/log.h"
#include "hw/loader.h"
+static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
+{
+ uint64_t id = KVM_REG_RISCV | type | idx;
+
+ switch (riscv_cpu_mxl(env)) {
+ case MXL_RV32:
+ id |= KVM_REG_SIZE_U32;
+ break;
+ case MXL_RV64:
+ id |= KVM_REG_SIZE_U64;
+ break;
+ default:
+ g_assert_not_reached();
+ }
+ return id;
+}
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
@@ -79,7 +96,20 @@ void kvm_arch_init_irq_routing(KVMState *s)
int kvm_arch_init_vcpu(CPUState *cs)
{
- return 0;
+ int ret = 0;
+ target_ulong isa;
+ RISCVCPU *cpu = RISCV_CPU(cs);
+ CPURISCVState *env = &cpu->env;
+ uint64_t id;
+
+ id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
+ ret = kvm_get_one_reg(cs, id, &isa);
+ if (ret) {
+ return ret;
+ }
+ env->misa_ext = isa;
+
+ return ret;
}
int kvm_arch_msi_data_to_gsi(uint32_t data)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 03/12] target/riscv: Implement function kvm_arch_init_vcpu
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get isa info from kvm while kvm init.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 32 +++++++++++++++++++++++++++++++-
1 file changed, 31 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 687dd4b621..ccf3753048 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -38,6 +38,23 @@
#include "qemu/log.h"
#include "hw/loader.h"
+static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
+{
+ uint64_t id = KVM_REG_RISCV | type | idx;
+
+ switch (riscv_cpu_mxl(env)) {
+ case MXL_RV32:
+ id |= KVM_REG_SIZE_U32;
+ break;
+ case MXL_RV64:
+ id |= KVM_REG_SIZE_U64;
+ break;
+ default:
+ g_assert_not_reached();
+ }
+ return id;
+}
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
@@ -79,7 +96,20 @@ void kvm_arch_init_irq_routing(KVMState *s)
int kvm_arch_init_vcpu(CPUState *cs)
{
- return 0;
+ int ret = 0;
+ target_ulong isa;
+ RISCVCPU *cpu = RISCV_CPU(cs);
+ CPURISCVState *env = &cpu->env;
+ uint64_t id;
+
+ id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
+ ret = kvm_get_one_reg(cs, id, &isa);
+ if (ret) {
+ return ret;
+ }
+ env->misa_ext = isa;
+
+ return ret;
}
int kvm_arch_msi_data_to_gsi(uint32_t data)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 04/12] target/riscv: Implement kvm_arch_get_registers
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get GPR CSR and FP registers from kvm by KVM_GET_ONE_REG ioctl.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 112 ++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 111 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index ccf3753048..6d4df0ef6d 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -55,13 +55,123 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
return id;
}
+#define RISCV_CORE_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, \
+ KVM_REG_RISCV_CORE_REG(name))
+
+#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
+ KVM_REG_RISCV_CSR_REG(name))
+
+#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
+
+#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
+
+#define KVM_RISCV_GET_CSR(cs, env, csr, reg) \
+ do { \
+ int ret = kvm_get_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
+ if (ret) { \
+ return ret; \
+ } \
+ } while(0)
+
+static int kvm_riscv_get_regs_core(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ target_ulong reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ ret = kvm_get_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
+ if (ret) {
+ return ret;
+ }
+ env->pc = reg;
+
+ for (i = 1; i < 32; i++) {
+ uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
+ ret = kvm_get_one_reg(cs, id, ®);
+ if (ret) {
+ return ret;
+ }
+ env->gpr[i] = reg;
+ }
+
+ return ret;
+}
+
+static int kvm_riscv_get_regs_csr(CPUState *cs)
+{
+ int ret = 0;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ KVM_RISCV_GET_CSR(cs, env, sstatus, env->mstatus);
+ KVM_RISCV_GET_CSR(cs, env, sie, env->mie);
+ KVM_RISCV_GET_CSR(cs, env, stvec, env->stvec);
+ KVM_RISCV_GET_CSR(cs, env, sscratch, env->sscratch);
+ KVM_RISCV_GET_CSR(cs, env, sepc, env->sepc);
+ KVM_RISCV_GET_CSR(cs, env, scause, env->scause);
+ KVM_RISCV_GET_CSR(cs, env, stval, env->stval);
+ KVM_RISCV_GET_CSR(cs, env, sip, env->mip);
+ KVM_RISCV_GET_CSR(cs, env, satp, env->satp);
+ return ret;
+}
+
+static int kvm_riscv_get_regs_fp(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (riscv_has_ext(env, RVD)) {
+ uint64_t reg;
+ for (i = 0; i < 32; i++) {
+ ret = kvm_get_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ env->fpr[i] = reg;
+ }
+ return ret;
+ }
+
+ if (riscv_has_ext(env, RVF)) {
+ uint32_t reg;
+ for (i = 0; i < 32; i++) {
+ ret = kvm_get_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ env->fpr[i] = reg;
+ }
+ return ret;
+ }
+
+ return ret;
+}
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
int kvm_arch_get_registers(CPUState *cs)
{
- return 0;
+ int ret = 0;
+
+ ret = kvm_riscv_get_regs_core(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_get_regs_csr(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_get_regs_fp(cs);
+ if (ret) {
+ return ret;
+ }
+
+ return ret;
}
int kvm_arch_put_registers(CPUState *cs, int level)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 04/12] target/riscv: Implement kvm_arch_get_registers
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get GPR CSR and FP registers from kvm by KVM_GET_ONE_REG ioctl.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 112 ++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 111 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index ccf3753048..6d4df0ef6d 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -55,13 +55,123 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
return id;
}
+#define RISCV_CORE_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, \
+ KVM_REG_RISCV_CORE_REG(name))
+
+#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
+ KVM_REG_RISCV_CSR_REG(name))
+
+#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
+
+#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
+
+#define KVM_RISCV_GET_CSR(cs, env, csr, reg) \
+ do { \
+ int ret = kvm_get_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
+ if (ret) { \
+ return ret; \
+ } \
+ } while(0)
+
+static int kvm_riscv_get_regs_core(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ target_ulong reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ ret = kvm_get_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
+ if (ret) {
+ return ret;
+ }
+ env->pc = reg;
+
+ for (i = 1; i < 32; i++) {
+ uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
+ ret = kvm_get_one_reg(cs, id, ®);
+ if (ret) {
+ return ret;
+ }
+ env->gpr[i] = reg;
+ }
+
+ return ret;
+}
+
+static int kvm_riscv_get_regs_csr(CPUState *cs)
+{
+ int ret = 0;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ KVM_RISCV_GET_CSR(cs, env, sstatus, env->mstatus);
+ KVM_RISCV_GET_CSR(cs, env, sie, env->mie);
+ KVM_RISCV_GET_CSR(cs, env, stvec, env->stvec);
+ KVM_RISCV_GET_CSR(cs, env, sscratch, env->sscratch);
+ KVM_RISCV_GET_CSR(cs, env, sepc, env->sepc);
+ KVM_RISCV_GET_CSR(cs, env, scause, env->scause);
+ KVM_RISCV_GET_CSR(cs, env, stval, env->stval);
+ KVM_RISCV_GET_CSR(cs, env, sip, env->mip);
+ KVM_RISCV_GET_CSR(cs, env, satp, env->satp);
+ return ret;
+}
+
+static int kvm_riscv_get_regs_fp(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (riscv_has_ext(env, RVD)) {
+ uint64_t reg;
+ for (i = 0; i < 32; i++) {
+ ret = kvm_get_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ env->fpr[i] = reg;
+ }
+ return ret;
+ }
+
+ if (riscv_has_ext(env, RVF)) {
+ uint32_t reg;
+ for (i = 0; i < 32; i++) {
+ ret = kvm_get_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ env->fpr[i] = reg;
+ }
+ return ret;
+ }
+
+ return ret;
+}
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
int kvm_arch_get_registers(CPUState *cs)
{
- return 0;
+ int ret = 0;
+
+ ret = kvm_riscv_get_regs_core(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_get_regs_csr(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_get_regs_fp(cs);
+ if (ret) {
+ return ret;
+ }
+
+ return ret;
}
int kvm_arch_put_registers(CPUState *cs, int level)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 05/12] target/riscv: Implement kvm_arch_put_registers
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Put GPR CSR and FP registers to kvm by KVM_SET_ONE_REG ioctl
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 104 ++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 103 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 6d4df0ef6d..e695b91dc7 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -73,6 +73,14 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
} \
} while(0)
+#define KVM_RISCV_SET_CSR(cs, env, csr, reg) \
+ do { \
+ int ret = kvm_set_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
+ if (ret) { \
+ return ret; \
+ } \
+ } while(0)
+
static int kvm_riscv_get_regs_core(CPUState *cs)
{
int ret = 0;
@@ -98,6 +106,31 @@ static int kvm_riscv_get_regs_core(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_core(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ target_ulong reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ reg = env->pc;
+ ret = kvm_set_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
+ if (ret) {
+ return ret;
+ }
+
+ for (i = 1; i < 32; i++) {
+ uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
+ reg = env->gpr[i];
+ ret = kvm_set_one_reg(cs, id, ®);
+ if (ret) {
+ return ret;
+ }
+ }
+
+ return ret;
+}
+
static int kvm_riscv_get_regs_csr(CPUState *cs)
{
int ret = 0;
@@ -115,6 +148,24 @@ static int kvm_riscv_get_regs_csr(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_csr(CPUState *cs)
+{
+ int ret = 0;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ KVM_RISCV_SET_CSR(cs, env, sstatus, env->mstatus);
+ KVM_RISCV_SET_CSR(cs, env, sie, env->mie);
+ KVM_RISCV_SET_CSR(cs, env, stvec, env->stvec);
+ KVM_RISCV_SET_CSR(cs, env, sscratch, env->sscratch);
+ KVM_RISCV_SET_CSR(cs, env, sepc, env->sepc);
+ KVM_RISCV_SET_CSR(cs, env, scause, env->scause);
+ KVM_RISCV_SET_CSR(cs, env, stval, env->stval);
+ KVM_RISCV_SET_CSR(cs, env, sip, env->mip);
+ KVM_RISCV_SET_CSR(cs, env, satp, env->satp);
+
+ return ret;
+}
+
static int kvm_riscv_get_regs_fp(CPUState *cs)
{
int ret = 0;
@@ -148,6 +199,40 @@ static int kvm_riscv_get_regs_fp(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_fp(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (riscv_has_ext(env, RVD)) {
+ uint64_t reg;
+ for (i = 0; i < 32; i++) {
+ reg = env->fpr[i];
+ ret = kvm_set_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ }
+ return ret;
+ }
+
+ if (riscv_has_ext(env, RVF)) {
+ uint32_t reg;
+ for (i = 0; i < 32; i++) {
+ reg = env->fpr[i];
+ ret = kvm_set_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ }
+ return ret;
+ }
+
+ return ret;
+}
+
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
@@ -176,7 +261,24 @@ int kvm_arch_get_registers(CPUState *cs)
int kvm_arch_put_registers(CPUState *cs, int level)
{
- return 0;
+ int ret = 0;
+
+ ret = kvm_riscv_put_regs_core(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_put_regs_csr(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_put_regs_fp(cs);
+ if (ret) {
+ return ret;
+ }
+
+ return ret;
}
int kvm_arch_release_virq_post(int virq)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 05/12] target/riscv: Implement kvm_arch_put_registers
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Put GPR CSR and FP registers to kvm by KVM_SET_ONE_REG ioctl
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/kvm.c | 104 ++++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 103 insertions(+), 1 deletion(-)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 6d4df0ef6d..e695b91dc7 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -73,6 +73,14 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
} \
} while(0)
+#define KVM_RISCV_SET_CSR(cs, env, csr, reg) \
+ do { \
+ int ret = kvm_set_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
+ if (ret) { \
+ return ret; \
+ } \
+ } while(0)
+
static int kvm_riscv_get_regs_core(CPUState *cs)
{
int ret = 0;
@@ -98,6 +106,31 @@ static int kvm_riscv_get_regs_core(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_core(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ target_ulong reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ reg = env->pc;
+ ret = kvm_set_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
+ if (ret) {
+ return ret;
+ }
+
+ for (i = 1; i < 32; i++) {
+ uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
+ reg = env->gpr[i];
+ ret = kvm_set_one_reg(cs, id, ®);
+ if (ret) {
+ return ret;
+ }
+ }
+
+ return ret;
+}
+
static int kvm_riscv_get_regs_csr(CPUState *cs)
{
int ret = 0;
@@ -115,6 +148,24 @@ static int kvm_riscv_get_regs_csr(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_csr(CPUState *cs)
+{
+ int ret = 0;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ KVM_RISCV_SET_CSR(cs, env, sstatus, env->mstatus);
+ KVM_RISCV_SET_CSR(cs, env, sie, env->mie);
+ KVM_RISCV_SET_CSR(cs, env, stvec, env->stvec);
+ KVM_RISCV_SET_CSR(cs, env, sscratch, env->sscratch);
+ KVM_RISCV_SET_CSR(cs, env, sepc, env->sepc);
+ KVM_RISCV_SET_CSR(cs, env, scause, env->scause);
+ KVM_RISCV_SET_CSR(cs, env, stval, env->stval);
+ KVM_RISCV_SET_CSR(cs, env, sip, env->mip);
+ KVM_RISCV_SET_CSR(cs, env, satp, env->satp);
+
+ return ret;
+}
+
static int kvm_riscv_get_regs_fp(CPUState *cs)
{
int ret = 0;
@@ -148,6 +199,40 @@ static int kvm_riscv_get_regs_fp(CPUState *cs)
return ret;
}
+static int kvm_riscv_put_regs_fp(CPUState *cs)
+{
+ int ret = 0;
+ int i;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (riscv_has_ext(env, RVD)) {
+ uint64_t reg;
+ for (i = 0; i < 32; i++) {
+ reg = env->fpr[i];
+ ret = kvm_set_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ }
+ return ret;
+ }
+
+ if (riscv_has_ext(env, RVF)) {
+ uint32_t reg;
+ for (i = 0; i < 32; i++) {
+ reg = env->fpr[i];
+ ret = kvm_set_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
+ if (ret) {
+ return ret;
+ }
+ }
+ return ret;
+ }
+
+ return ret;
+}
+
+
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
@@ -176,7 +261,24 @@ int kvm_arch_get_registers(CPUState *cs)
int kvm_arch_put_registers(CPUState *cs, int level)
{
- return 0;
+ int ret = 0;
+
+ ret = kvm_riscv_put_regs_core(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_put_regs_csr(cs);
+ if (ret) {
+ return ret;
+ }
+
+ ret = kvm_riscv_put_regs_fp(cs);
+ if (ret) {
+ return ret;
+ }
+
+ return ret;
}
int kvm_arch_release_virq_post(int virq)
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get kernel and fdt start address in virt.c, and pass them to KVM
when cpu reset. In addition, add kvm_riscv.h to place riscv specific
interface.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
hw/riscv/boot.c | 11 ++++++++
hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
include/hw/riscv/boot.h | 1 +
target/riscv/cpu.c | 8 ++++++
target/riscv/cpu.h | 3 +++
target/riscv/kvm-stub.c | 25 +++++++++++++++++++
target/riscv/kvm.c | 14 +++++++++++
target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
target/riscv/meson.build | 2 +-
9 files changed, 125 insertions(+), 17 deletions(-)
create mode 100644 target/riscv/kvm-stub.c
create mode 100644 target/riscv/kvm_riscv.h
diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
index 519fa455a1..00df6d7810 100644
--- a/hw/riscv/boot.c
+++ b/hw/riscv/boot.c
@@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState *machine, RISCVHartArrayState *harts
return;
}
+
+void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
+{
+ CPUState *cs;
+
+ for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
+ RISCVCPU *riscv_cpu = RISCV_CPU(cs);
+ riscv_cpu->env.kernel_addr = kernel_addr;
+ riscv_cpu->env.fdt_addr = fdt_addr;
+ }
+}
diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index 3af074148e..c8bcd9d9e5 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -38,6 +38,7 @@
#include "chardev/char.h"
#include "sysemu/device_tree.h"
#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
#include "hw/pci/pci.h"
#include "hw/pci-host/gpex.h"
#include "hw/display/ramfb.h"
@@ -801,23 +802,25 @@ static void virt_machine_init(MachineState *machine)
hart_count, &error_abort);
sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
- /* Per-socket CLINT */
- riscv_aclint_swi_create(
- memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
- base_hartid, hart_count, false);
- riscv_aclint_mtimer_create(
- memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
- RISCV_ACLINT_SWI_SIZE,
- RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
- RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
- RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
-
- /* Per-socket ACLINT SSWI */
- if (s->have_aclint) {
+ if (!kvm_enabled()) {
+ /* Per-socket CLINT */
riscv_aclint_swi_create(
- memmap[VIRT_ACLINT_SSWI].base +
- i * memmap[VIRT_ACLINT_SSWI].size,
- base_hartid, hart_count, true);
+ memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
+ base_hartid, hart_count, false);
+ riscv_aclint_mtimer_create(
+ memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
+ RISCV_ACLINT_SWI_SIZE,
+ RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
+ RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
+ RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
+
+ /* Per-socket ACLINT SSWI */
+ if (s->have_aclint) {
+ riscv_aclint_swi_create(
+ memmap[VIRT_ACLINT_SSWI].base +
+ i * memmap[VIRT_ACLINT_SSWI].size,
+ base_hartid, hart_count, true);
+ }
}
/* Per-socket PLIC hart topology configuration string */
@@ -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base,
mask_rom);
+ /*
+ * Only direct boot kernel is currently supported for KVM VM,
+ * so the "-bios" parameter is ignored and treated like "-bios none"
+ * when KVM is enabled.
+ */
+ if (kvm_enabled()) {
+ g_free(machine->firmware);
+ machine->firmware = g_strdup("none");
+ }
+
if (riscv_is_32bit(&s->soc[0])) {
firmware_end_addr = riscv_find_and_load_firmware(machine,
RISCV32_BIOS_BIN, start_addr, NULL);
@@ -941,6 +954,15 @@ static void virt_machine_init(MachineState *machine)
virt_memmap[VIRT_MROM].size, kernel_entry,
fdt_load_addr, machine->fdt);
+ /*
+ * Only direct boot kernel is currently supported for KVM VM,
+ * So here setup kernel start address and fdt address.
+ * TODO:Support firmware loading and integrate to TCG start
+ */
+ if (kvm_enabled()) {
+ riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
+ }
+
/* SiFive Test MMIO device */
sifive_test_create(memmap[VIRT_TEST].base);
diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
index baff11dd8a..5834c234aa 100644
--- a/include/hw/riscv/boot.h
+++ b/include/hw/riscv/boot.h
@@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
hwaddr rom_size,
uint32_t reset_vec_size,
uint64_t kernel_entry);
+void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
#endif /* RISCV_BOOT_H */
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index f812998123..1c944872a3 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -29,6 +29,8 @@
#include "hw/qdev-properties.h"
#include "migration/vmstate.h"
#include "fpu/softfloat-helpers.h"
+#include "sysemu/kvm.h"
+#include "kvm_riscv.h"
/* RISC-V CPU definitions */
@@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
cs->exception_index = RISCV_EXCP_NONE;
env->load_res = -1;
set_default_nan_mode(1, &env->fp_status);
+
+#ifndef CONFIG_USER_ONLY
+ if (kvm_enabled()) {
+ kvm_riscv_reset_vcpu(cpu);
+ }
+#endif
}
static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 0760c0af93..2807eb1bcb 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -255,6 +255,9 @@ struct CPURISCVState {
/* Fields from here on are preserved across CPU reset. */
QEMUTimer *timer; /* Internal timer */
+
+ hwaddr kernel_addr;
+ hwaddr fdt_addr;
};
OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
new file mode 100644
index 0000000000..39b96fe3f4
--- /dev/null
+++ b/target/riscv/kvm-stub.c
@@ -0,0 +1,25 @@
+/*
+ * QEMU KVM RISC-V specific function stubs
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "kvm_riscv.h"
+
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
+{
+ abort();
+}
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index e695b91dc7..db6d8a5b6e 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -37,6 +37,7 @@
#include "hw/irq.h"
#include "qemu/log.h"
#include "hw/loader.h"
+#include "kvm_riscv.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -369,6 +370,19 @@ int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
return 0;
}
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
+{
+ CPURISCVState *env = &cpu->env;
+
+ if (!kvm_enabled()) {
+ return;
+ }
+ env->pc = cpu->env.kernel_addr;
+ env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
+ env->gpr[11] = cpu->env.fdt_addr; /* a1 */
+ env->satp = 0;
+}
+
bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
new file mode 100644
index 0000000000..f38c82bf59
--- /dev/null
+++ b/target/riscv/kvm_riscv.h
@@ -0,0 +1,24 @@
+/*
+ * QEMU KVM support -- RISC-V specific functions.
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef QEMU_KVM_RISCV_H
+#define QEMU_KVM_RISCV_H
+
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
+
+#endif
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index 2faf08a941..fe41cc5805 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -19,7 +19,7 @@ riscv_ss.add(files(
'bitmanip_helper.c',
'translate.c',
))
-riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
+riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c'))
riscv_softmmu_ss = ss.source_set()
riscv_softmmu_ss.add(files(
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
Get kernel and fdt start address in virt.c, and pass them to KVM
when cpu reset. In addition, add kvm_riscv.h to place riscv specific
interface.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
hw/riscv/boot.c | 11 ++++++++
hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
include/hw/riscv/boot.h | 1 +
target/riscv/cpu.c | 8 ++++++
target/riscv/cpu.h | 3 +++
target/riscv/kvm-stub.c | 25 +++++++++++++++++++
target/riscv/kvm.c | 14 +++++++++++
target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
target/riscv/meson.build | 2 +-
9 files changed, 125 insertions(+), 17 deletions(-)
create mode 100644 target/riscv/kvm-stub.c
create mode 100644 target/riscv/kvm_riscv.h
diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
index 519fa455a1..00df6d7810 100644
--- a/hw/riscv/boot.c
+++ b/hw/riscv/boot.c
@@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState *machine, RISCVHartArrayState *harts
return;
}
+
+void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
+{
+ CPUState *cs;
+
+ for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
+ RISCVCPU *riscv_cpu = RISCV_CPU(cs);
+ riscv_cpu->env.kernel_addr = kernel_addr;
+ riscv_cpu->env.fdt_addr = fdt_addr;
+ }
+}
diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
index 3af074148e..c8bcd9d9e5 100644
--- a/hw/riscv/virt.c
+++ b/hw/riscv/virt.c
@@ -38,6 +38,7 @@
#include "chardev/char.h"
#include "sysemu/device_tree.h"
#include "sysemu/sysemu.h"
+#include "sysemu/kvm.h"
#include "hw/pci/pci.h"
#include "hw/pci-host/gpex.h"
#include "hw/display/ramfb.h"
@@ -801,23 +802,25 @@ static void virt_machine_init(MachineState *machine)
hart_count, &error_abort);
sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
- /* Per-socket CLINT */
- riscv_aclint_swi_create(
- memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
- base_hartid, hart_count, false);
- riscv_aclint_mtimer_create(
- memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
- RISCV_ACLINT_SWI_SIZE,
- RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
- RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
- RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
-
- /* Per-socket ACLINT SSWI */
- if (s->have_aclint) {
+ if (!kvm_enabled()) {
+ /* Per-socket CLINT */
riscv_aclint_swi_create(
- memmap[VIRT_ACLINT_SSWI].base +
- i * memmap[VIRT_ACLINT_SSWI].size,
- base_hartid, hart_count, true);
+ memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
+ base_hartid, hart_count, false);
+ riscv_aclint_mtimer_create(
+ memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
+ RISCV_ACLINT_SWI_SIZE,
+ RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
+ RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
+ RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
+
+ /* Per-socket ACLINT SSWI */
+ if (s->have_aclint) {
+ riscv_aclint_swi_create(
+ memmap[VIRT_ACLINT_SSWI].base +
+ i * memmap[VIRT_ACLINT_SSWI].size,
+ base_hartid, hart_count, true);
+ }
}
/* Per-socket PLIC hart topology configuration string */
@@ -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base,
mask_rom);
+ /*
+ * Only direct boot kernel is currently supported for KVM VM,
+ * so the "-bios" parameter is ignored and treated like "-bios none"
+ * when KVM is enabled.
+ */
+ if (kvm_enabled()) {
+ g_free(machine->firmware);
+ machine->firmware = g_strdup("none");
+ }
+
if (riscv_is_32bit(&s->soc[0])) {
firmware_end_addr = riscv_find_and_load_firmware(machine,
RISCV32_BIOS_BIN, start_addr, NULL);
@@ -941,6 +954,15 @@ static void virt_machine_init(MachineState *machine)
virt_memmap[VIRT_MROM].size, kernel_entry,
fdt_load_addr, machine->fdt);
+ /*
+ * Only direct boot kernel is currently supported for KVM VM,
+ * So here setup kernel start address and fdt address.
+ * TODO:Support firmware loading and integrate to TCG start
+ */
+ if (kvm_enabled()) {
+ riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
+ }
+
/* SiFive Test MMIO device */
sifive_test_create(memmap[VIRT_TEST].base);
diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
index baff11dd8a..5834c234aa 100644
--- a/include/hw/riscv/boot.h
+++ b/include/hw/riscv/boot.h
@@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
hwaddr rom_size,
uint32_t reset_vec_size,
uint64_t kernel_entry);
+void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
#endif /* RISCV_BOOT_H */
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index f812998123..1c944872a3 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -29,6 +29,8 @@
#include "hw/qdev-properties.h"
#include "migration/vmstate.h"
#include "fpu/softfloat-helpers.h"
+#include "sysemu/kvm.h"
+#include "kvm_riscv.h"
/* RISC-V CPU definitions */
@@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
cs->exception_index = RISCV_EXCP_NONE;
env->load_res = -1;
set_default_nan_mode(1, &env->fp_status);
+
+#ifndef CONFIG_USER_ONLY
+ if (kvm_enabled()) {
+ kvm_riscv_reset_vcpu(cpu);
+ }
+#endif
}
static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 0760c0af93..2807eb1bcb 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -255,6 +255,9 @@ struct CPURISCVState {
/* Fields from here on are preserved across CPU reset. */
QEMUTimer *timer; /* Internal timer */
+
+ hwaddr kernel_addr;
+ hwaddr fdt_addr;
};
OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
new file mode 100644
index 0000000000..39b96fe3f4
--- /dev/null
+++ b/target/riscv/kvm-stub.c
@@ -0,0 +1,25 @@
+/*
+ * QEMU KVM RISC-V specific function stubs
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+#include "qemu/osdep.h"
+#include "cpu.h"
+#include "kvm_riscv.h"
+
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
+{
+ abort();
+}
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index e695b91dc7..db6d8a5b6e 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -37,6 +37,7 @@
#include "hw/irq.h"
#include "qemu/log.h"
#include "hw/loader.h"
+#include "kvm_riscv.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -369,6 +370,19 @@ int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
return 0;
}
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
+{
+ CPURISCVState *env = &cpu->env;
+
+ if (!kvm_enabled()) {
+ return;
+ }
+ env->pc = cpu->env.kernel_addr;
+ env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
+ env->gpr[11] = cpu->env.fdt_addr; /* a1 */
+ env->satp = 0;
+}
+
bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
new file mode 100644
index 0000000000..f38c82bf59
--- /dev/null
+++ b/target/riscv/kvm_riscv.h
@@ -0,0 +1,24 @@
+/*
+ * QEMU KVM support -- RISC-V specific functions.
+ *
+ * Copyright (c) 2020 Huawei Technologies Co., Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2 or later, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef QEMU_KVM_RISCV_H
+#define QEMU_KVM_RISCV_H
+
+void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
+
+#endif
diff --git a/target/riscv/meson.build b/target/riscv/meson.build
index 2faf08a941..fe41cc5805 100644
--- a/target/riscv/meson.build
+++ b/target/riscv/meson.build
@@ -19,7 +19,7 @@ riscv_ss.add(files(
'bitmanip_helper.c',
'translate.c',
))
-riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
+riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c'))
riscv_softmmu_ss = ss.source_set()
riscv_softmmu_ss.add(files(
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
When KVM is enabled, set the S-mode external interrupt through
kvm_riscv_set_irq function.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 6 +++++-
target/riscv/kvm-stub.c | 5 +++++
target/riscv/kvm.c | 17 +++++++++++++++++
target/riscv/kvm_riscv.h | 1 +
4 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 1c944872a3..71a7ac6831 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq, int level)
case IRQ_S_EXT:
case IRQ_VS_EXT:
case IRQ_M_EXT:
- riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
+ if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
+ kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
+ } else {
+ riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
+ }
break;
default:
g_assert_not_reached();
diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
index 39b96fe3f4..4e8fc31a21 100644
--- a/target/riscv/kvm-stub.c
+++ b/target/riscv/kvm-stub.c
@@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
{
abort();
}
+
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
+{
+ abort();
+}
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index db6d8a5b6e..0027f11f45 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
env->satp = 0;
}
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
+{
+ int ret;
+ unsigned virq = level ? KVM_INTERRUPT_SET : KVM_INTERRUPT_UNSET;
+
+ if (irq != IRQ_S_EXT) {
+ perror("kvm riscv set irq != IRQ_S_EXT\n");
+ abort();
+ }
+
+ ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
+ if (ret < 0) {
+ perror("Set irq failed");
+ abort();
+ }
+}
+
bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
index f38c82bf59..ed281bdce0 100644
--- a/target/riscv/kvm_riscv.h
+++ b/target/riscv/kvm_riscv.h
@@ -20,5 +20,6 @@
#define QEMU_KVM_RISCV_H
void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
#endif
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
When KVM is enabled, set the S-mode external interrupt through
kvm_riscv_set_irq function.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
---
target/riscv/cpu.c | 6 +++++-
target/riscv/kvm-stub.c | 5 +++++
target/riscv/kvm.c | 17 +++++++++++++++++
target/riscv/kvm_riscv.h | 1 +
4 files changed, 28 insertions(+), 1 deletion(-)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 1c944872a3..71a7ac6831 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq, int level)
case IRQ_S_EXT:
case IRQ_VS_EXT:
case IRQ_M_EXT:
- riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
+ if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
+ kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
+ } else {
+ riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
+ }
break;
default:
g_assert_not_reached();
diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
index 39b96fe3f4..4e8fc31a21 100644
--- a/target/riscv/kvm-stub.c
+++ b/target/riscv/kvm-stub.c
@@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
{
abort();
}
+
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
+{
+ abort();
+}
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index db6d8a5b6e..0027f11f45 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
env->satp = 0;
}
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
+{
+ int ret;
+ unsigned virq = level ? KVM_INTERRUPT_SET : KVM_INTERRUPT_UNSET;
+
+ if (irq != IRQ_S_EXT) {
+ perror("kvm riscv set irq != IRQ_S_EXT\n");
+ abort();
+ }
+
+ ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
+ if (ret < 0) {
+ perror("Set irq failed");
+ abort();
+ }
+}
+
bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
index f38c82bf59..ed281bdce0 100644
--- a/target/riscv/kvm_riscv.h
+++ b/target/riscv/kvm_riscv.h
@@ -20,5 +20,6 @@
#define QEMU_KVM_RISCV_H
void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
+void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
#endif
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 08/12] target/riscv: Handle KVM_EXIT_RISCV_SBI exit
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Use char-fe to handle console sbi call, which implement early
console io while apply 'earlycon=sbi' into kernel parameters.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/kvm.c | 38 +++++++++++++++-
target/riscv/sbi_ecall_interface.h | 72 ++++++++++++++++++++++++++++++
2 files changed, 109 insertions(+), 1 deletion(-)
create mode 100644 target/riscv/sbi_ecall_interface.h
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 0027f11f45..171a32adf9 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -38,6 +38,8 @@
#include "qemu/log.h"
#include "hw/loader.h"
#include "kvm_riscv.h"
+#include "sbi_ecall_interface.h"
+#include "semihosting/console.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -365,9 +367,43 @@ bool kvm_arch_stop_on_emulation_error(CPUState *cs)
return true;
}
+static int kvm_riscv_handle_sbi(CPUState *cs, struct kvm_run *run)
+{
+ int ret = 0;
+ unsigned char ch;
+ switch (run->riscv_sbi.extension_id) {
+ case SBI_EXT_0_1_CONSOLE_PUTCHAR:
+ ch = run->riscv_sbi.args[0];
+ qemu_semihosting_log_out((const char *)&ch, sizeof(ch));
+ break;
+ case SBI_EXT_0_1_CONSOLE_GETCHAR:
+ run->riscv_sbi.args[0] =
+ (unsigned long)qemu_semihosting_console_inc(cs->env_ptr);
+ break;
+ default:
+ qemu_log_mask(LOG_UNIMP,
+ "%s: un-handled SBI EXIT, specific reasons is %lu\n",
+ __func__, run->riscv_sbi.extension_id);
+ ret = -1;
+ break;
+ }
+ return ret;
+}
+
int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
{
- return 0;
+ int ret = 0;
+ switch (run->exit_reason) {
+ case KVM_EXIT_RISCV_SBI:
+ ret = kvm_riscv_handle_sbi(cs, run);
+ break;
+ default:
+ qemu_log_mask(LOG_UNIMP, "%s: un-handled exit reason %d\n",
+ __func__, run->exit_reason);
+ ret = -1;
+ break;
+ }
+ return ret;
}
void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
diff --git a/target/riscv/sbi_ecall_interface.h b/target/riscv/sbi_ecall_interface.h
new file mode 100644
index 0000000000..fb1a3fa8f2
--- /dev/null
+++ b/target/riscv/sbi_ecall_interface.h
@@ -0,0 +1,72 @@
+/*
+ * SPDX-License-Identifier: BSD-2-Clause
+ *
+ * Copyright (c) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ * Anup Patel <anup.patel@wdc.com>
+ */
+
+#ifndef __SBI_ECALL_INTERFACE_H__
+#define __SBI_ECALL_INTERFACE_H__
+
+/* clang-format off */
+
+/* SBI Extension IDs */
+#define SBI_EXT_0_1_SET_TIMER 0x0
+#define SBI_EXT_0_1_CONSOLE_PUTCHAR 0x1
+#define SBI_EXT_0_1_CONSOLE_GETCHAR 0x2
+#define SBI_EXT_0_1_CLEAR_IPI 0x3
+#define SBI_EXT_0_1_SEND_IPI 0x4
+#define SBI_EXT_0_1_REMOTE_FENCE_I 0x5
+#define SBI_EXT_0_1_REMOTE_SFENCE_VMA 0x6
+#define SBI_EXT_0_1_REMOTE_SFENCE_VMA_ASID 0x7
+#define SBI_EXT_0_1_SHUTDOWN 0x8
+#define SBI_EXT_BASE 0x10
+#define SBI_EXT_TIME 0x54494D45
+#define SBI_EXT_IPI 0x735049
+#define SBI_EXT_RFENCE 0x52464E43
+#define SBI_EXT_HSM 0x48534D
+
+/* SBI function IDs for BASE extension*/
+#define SBI_EXT_BASE_GET_SPEC_VERSION 0x0
+#define SBI_EXT_BASE_GET_IMP_ID 0x1
+#define SBI_EXT_BASE_GET_IMP_VERSION 0x2
+#define SBI_EXT_BASE_PROBE_EXT 0x3
+#define SBI_EXT_BASE_GET_MVENDORID 0x4
+#define SBI_EXT_BASE_GET_MARCHID 0x5
+#define SBI_EXT_BASE_GET_MIMPID 0x6
+
+/* SBI function IDs for TIME extension*/
+#define SBI_EXT_TIME_SET_TIMER 0x0
+
+/* SBI function IDs for IPI extension*/
+#define SBI_EXT_IPI_SEND_IPI 0x0
+
+/* SBI function IDs for RFENCE extension*/
+#define SBI_EXT_RFENCE_REMOTE_FENCE_I 0x0
+#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA 0x1
+#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID 0x2
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA 0x3
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA_VMID 0x4
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA 0x5
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID 0x6
+
+/* SBI function IDs for HSM extension */
+#define SBI_EXT_HSM_HART_START 0x0
+#define SBI_EXT_HSM_HART_STOP 0x1
+#define SBI_EXT_HSM_HART_GET_STATUS 0x2
+
+#define SBI_HSM_HART_STATUS_STARTED 0x0
+#define SBI_HSM_HART_STATUS_STOPPED 0x1
+#define SBI_HSM_HART_STATUS_START_PENDING 0x2
+#define SBI_HSM_HART_STATUS_STOP_PENDING 0x3
+
+#define SBI_SPEC_VERSION_MAJOR_OFFSET 24
+#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f
+#define SBI_SPEC_VERSION_MINOR_MASK 0xffffff
+#define SBI_EXT_VENDOR_START 0x09000000
+#define SBI_EXT_VENDOR_END 0x09FFFFFF
+/* clang-format on */
+
+#endif
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 08/12] target/riscv: Handle KVM_EXIT_RISCV_SBI exit
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Use char-fe to handle console sbi call, which implement early
console io while apply 'earlycon=sbi' into kernel parameters.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/kvm.c | 38 +++++++++++++++-
target/riscv/sbi_ecall_interface.h | 72 ++++++++++++++++++++++++++++++
2 files changed, 109 insertions(+), 1 deletion(-)
create mode 100644 target/riscv/sbi_ecall_interface.h
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 0027f11f45..171a32adf9 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -38,6 +38,8 @@
#include "qemu/log.h"
#include "hw/loader.h"
#include "kvm_riscv.h"
+#include "sbi_ecall_interface.h"
+#include "semihosting/console.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -365,9 +367,43 @@ bool kvm_arch_stop_on_emulation_error(CPUState *cs)
return true;
}
+static int kvm_riscv_handle_sbi(CPUState *cs, struct kvm_run *run)
+{
+ int ret = 0;
+ unsigned char ch;
+ switch (run->riscv_sbi.extension_id) {
+ case SBI_EXT_0_1_CONSOLE_PUTCHAR:
+ ch = run->riscv_sbi.args[0];
+ qemu_semihosting_log_out((const char *)&ch, sizeof(ch));
+ break;
+ case SBI_EXT_0_1_CONSOLE_GETCHAR:
+ run->riscv_sbi.args[0] =
+ (unsigned long)qemu_semihosting_console_inc(cs->env_ptr);
+ break;
+ default:
+ qemu_log_mask(LOG_UNIMP,
+ "%s: un-handled SBI EXIT, specific reasons is %lu\n",
+ __func__, run->riscv_sbi.extension_id);
+ ret = -1;
+ break;
+ }
+ return ret;
+}
+
int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
{
- return 0;
+ int ret = 0;
+ switch (run->exit_reason) {
+ case KVM_EXIT_RISCV_SBI:
+ ret = kvm_riscv_handle_sbi(cs, run);
+ break;
+ default:
+ qemu_log_mask(LOG_UNIMP, "%s: un-handled exit reason %d\n",
+ __func__, run->exit_reason);
+ ret = -1;
+ break;
+ }
+ return ret;
}
void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
diff --git a/target/riscv/sbi_ecall_interface.h b/target/riscv/sbi_ecall_interface.h
new file mode 100644
index 0000000000..fb1a3fa8f2
--- /dev/null
+++ b/target/riscv/sbi_ecall_interface.h
@@ -0,0 +1,72 @@
+/*
+ * SPDX-License-Identifier: BSD-2-Clause
+ *
+ * Copyright (c) 2019 Western Digital Corporation or its affiliates.
+ *
+ * Authors:
+ * Anup Patel <anup.patel@wdc.com>
+ */
+
+#ifndef __SBI_ECALL_INTERFACE_H__
+#define __SBI_ECALL_INTERFACE_H__
+
+/* clang-format off */
+
+/* SBI Extension IDs */
+#define SBI_EXT_0_1_SET_TIMER 0x0
+#define SBI_EXT_0_1_CONSOLE_PUTCHAR 0x1
+#define SBI_EXT_0_1_CONSOLE_GETCHAR 0x2
+#define SBI_EXT_0_1_CLEAR_IPI 0x3
+#define SBI_EXT_0_1_SEND_IPI 0x4
+#define SBI_EXT_0_1_REMOTE_FENCE_I 0x5
+#define SBI_EXT_0_1_REMOTE_SFENCE_VMA 0x6
+#define SBI_EXT_0_1_REMOTE_SFENCE_VMA_ASID 0x7
+#define SBI_EXT_0_1_SHUTDOWN 0x8
+#define SBI_EXT_BASE 0x10
+#define SBI_EXT_TIME 0x54494D45
+#define SBI_EXT_IPI 0x735049
+#define SBI_EXT_RFENCE 0x52464E43
+#define SBI_EXT_HSM 0x48534D
+
+/* SBI function IDs for BASE extension*/
+#define SBI_EXT_BASE_GET_SPEC_VERSION 0x0
+#define SBI_EXT_BASE_GET_IMP_ID 0x1
+#define SBI_EXT_BASE_GET_IMP_VERSION 0x2
+#define SBI_EXT_BASE_PROBE_EXT 0x3
+#define SBI_EXT_BASE_GET_MVENDORID 0x4
+#define SBI_EXT_BASE_GET_MARCHID 0x5
+#define SBI_EXT_BASE_GET_MIMPID 0x6
+
+/* SBI function IDs for TIME extension*/
+#define SBI_EXT_TIME_SET_TIMER 0x0
+
+/* SBI function IDs for IPI extension*/
+#define SBI_EXT_IPI_SEND_IPI 0x0
+
+/* SBI function IDs for RFENCE extension*/
+#define SBI_EXT_RFENCE_REMOTE_FENCE_I 0x0
+#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA 0x1
+#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID 0x2
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA 0x3
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA_VMID 0x4
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA 0x5
+#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID 0x6
+
+/* SBI function IDs for HSM extension */
+#define SBI_EXT_HSM_HART_START 0x0
+#define SBI_EXT_HSM_HART_STOP 0x1
+#define SBI_EXT_HSM_HART_GET_STATUS 0x2
+
+#define SBI_HSM_HART_STATUS_STARTED 0x0
+#define SBI_HSM_HART_STATUS_STOPPED 0x1
+#define SBI_HSM_HART_STATUS_START_PENDING 0x2
+#define SBI_HSM_HART_STATUS_STOP_PENDING 0x3
+
+#define SBI_SPEC_VERSION_MAJOR_OFFSET 24
+#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f
+#define SBI_SPEC_VERSION_MINOR_MASK 0xffffff
+#define SBI_EXT_VENDOR_START 0x09000000
+#define SBI_EXT_VENDOR_END 0x09FFFFFF
+/* clang-format on */
+
+#endif
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 09/12] target/riscv: Add host cpu type
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
'host' type cpu is set isa to RV32 or RV64 simply, more isa info
will obtain from KVM in kvm_arch_init_vcpu()
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
target/riscv/cpu.c | 15 +++++++++++++++
target/riscv/cpu.h | 1 +
2 files changed, 16 insertions(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 71a7ac6831..a137fd3d2b 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -213,6 +213,18 @@ static void rv32_imafcu_nommu_cpu_init(Object *obj)
}
#endif
+#if defined(CONFIG_KVM)
+static void riscv_host_cpu_init(Object *obj)
+{
+ CPURISCVState *env = &RISCV_CPU(obj)->env;
+#if defined(TARGET_RISCV32)
+ set_misa(env, MXL_RV32, 0);
+#elif defined(TARGET_RISCV64)
+ set_misa(env, MXL_RV64, 0);
+#endif
+}
+#endif
+
static ObjectClass *riscv_cpu_class_by_name(const char *cpu_model)
{
ObjectClass *oc;
@@ -814,6 +826,9 @@ static const TypeInfo riscv_cpu_type_infos[] = {
.class_init = riscv_cpu_class_init,
},
DEFINE_CPU(TYPE_RISCV_CPU_ANY, riscv_any_cpu_init),
+#if defined(CONFIG_KVM)
+ DEFINE_CPU(TYPE_RISCV_CPU_HOST, riscv_host_cpu_init),
+#endif
#if defined(TARGET_RISCV32)
DEFINE_CPU(TYPE_RISCV_CPU_BASE32, rv32_base_cpu_init),
DEFINE_CPU(TYPE_RISCV_CPU_IBEX, rv32_ibex_cpu_init),
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 2807eb1bcb..e7dba35acb 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -45,6 +45,7 @@
#define TYPE_RISCV_CPU_SIFIVE_E51 RISCV_CPU_TYPE_NAME("sifive-e51")
#define TYPE_RISCV_CPU_SIFIVE_U34 RISCV_CPU_TYPE_NAME("sifive-u34")
#define TYPE_RISCV_CPU_SIFIVE_U54 RISCV_CPU_TYPE_NAME("sifive-u54")
+#define TYPE_RISCV_CPU_HOST RISCV_CPU_TYPE_NAME("host")
#if defined(TARGET_RISCV32)
# define TYPE_RISCV_CPU_BASE TYPE_RISCV_CPU_BASE32
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 09/12] target/riscv: Add host cpu type
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li, Alistair Francis
'host' type cpu is set isa to RV32 or RV64 simply, more isa info
will obtain from KVM in kvm_arch_init_vcpu()
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Anup Patel <anup.patel@wdc.com>
---
target/riscv/cpu.c | 15 +++++++++++++++
target/riscv/cpu.h | 1 +
2 files changed, 16 insertions(+)
diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
index 71a7ac6831..a137fd3d2b 100644
--- a/target/riscv/cpu.c
+++ b/target/riscv/cpu.c
@@ -213,6 +213,18 @@ static void rv32_imafcu_nommu_cpu_init(Object *obj)
}
#endif
+#if defined(CONFIG_KVM)
+static void riscv_host_cpu_init(Object *obj)
+{
+ CPURISCVState *env = &RISCV_CPU(obj)->env;
+#if defined(TARGET_RISCV32)
+ set_misa(env, MXL_RV32, 0);
+#elif defined(TARGET_RISCV64)
+ set_misa(env, MXL_RV64, 0);
+#endif
+}
+#endif
+
static ObjectClass *riscv_cpu_class_by_name(const char *cpu_model)
{
ObjectClass *oc;
@@ -814,6 +826,9 @@ static const TypeInfo riscv_cpu_type_infos[] = {
.class_init = riscv_cpu_class_init,
},
DEFINE_CPU(TYPE_RISCV_CPU_ANY, riscv_any_cpu_init),
+#if defined(CONFIG_KVM)
+ DEFINE_CPU(TYPE_RISCV_CPU_HOST, riscv_host_cpu_init),
+#endif
#if defined(TARGET_RISCV32)
DEFINE_CPU(TYPE_RISCV_CPU_BASE32, rv32_base_cpu_init),
DEFINE_CPU(TYPE_RISCV_CPU_IBEX, rv32_ibex_cpu_init),
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 2807eb1bcb..e7dba35acb 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -45,6 +45,7 @@
#define TYPE_RISCV_CPU_SIFIVE_E51 RISCV_CPU_TYPE_NAME("sifive-e51")
#define TYPE_RISCV_CPU_SIFIVE_U34 RISCV_CPU_TYPE_NAME("sifive-u34")
#define TYPE_RISCV_CPU_SIFIVE_U54 RISCV_CPU_TYPE_NAME("sifive-u54")
+#define TYPE_RISCV_CPU_HOST RISCV_CPU_TYPE_NAME("host")
#if defined(TARGET_RISCV32)
# define TYPE_RISCV_CPU_BASE TYPE_RISCV_CPU_BASE32
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Add kvm_riscv_get/put_regs_timer to synchronize virtual time context
from KVM.
To set register of RISCV_TIMER_REG(state) will occur a error from KVM
on kvm_timer_state == 0. It's better to adapt in KVM, but it doesn't matter
that adaping in QEMU.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/cpu.h | 7 +++++
target/riscv/kvm.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 79 insertions(+)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index e7dba35acb..c892a2c8b7 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -259,6 +259,13 @@ struct CPURISCVState {
hwaddr kernel_addr;
hwaddr fdt_addr;
+
+ /* kvm timer */
+ bool kvm_timer_dirty;
+ uint64_t kvm_timer_time;
+ uint64_t kvm_timer_compare;
+ uint64_t kvm_timer_state;
+ uint64_t kvm_timer_frequency;
};
OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 171a32adf9..802c076b22 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -64,6 +64,9 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
KVM_REG_RISCV_CSR_REG(name))
+#define RISCV_TIMER_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_TIMER, \
+ KVM_REG_RISCV_TIMER_REG(name))
+
#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
@@ -235,6 +238,75 @@ static int kvm_riscv_put_regs_fp(CPUState *cs)
return ret;
}
+static void kvm_riscv_get_regs_timer(CPUState *cs)
+{
+ int ret;
+ uint64_t reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (env->kvm_timer_dirty) {
+ return;
+ }
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_time = reg;
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_compare = reg;
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_state = reg;
+
+ env->kvm_timer_dirty = true;
+}
+
+static void kvm_riscv_put_regs_timer(CPUState *cs)
+{
+ int ret;
+ uint64_t reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (!env->kvm_timer_dirty) {
+ return;
+ }
+
+ reg = env->kvm_timer_time;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
+ if (ret) {
+ abort();
+ }
+
+ reg = env->kvm_timer_compare;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
+ if (ret) {
+ abort();
+ }
+
+ /*
+ * To set register of RISCV_TIMER_REG(state) will occur a error from KVM
+ * on env->kvm_timer_state == 0, It's better to adapt in KVM, but it
+ * doesn't matter that adaping in QEMU now.
+ * TODO If KVM changes, adapt here.
+ */
+ if (env->kvm_timer_state) {
+ reg = env->kvm_timer_state;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
+ if (ret) {
+ abort();
+ }
+ }
+
+ env->kvm_timer_dirty = false;
+}
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Add kvm_riscv_get/put_regs_timer to synchronize virtual time context
from KVM.
To set register of RISCV_TIMER_REG(state) will occur a error from KVM
on kvm_timer_state == 0. It's better to adapt in KVM, but it doesn't matter
that adaping in QEMU.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/cpu.h | 7 +++++
target/riscv/kvm.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 79 insertions(+)
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index e7dba35acb..c892a2c8b7 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -259,6 +259,13 @@ struct CPURISCVState {
hwaddr kernel_addr;
hwaddr fdt_addr;
+
+ /* kvm timer */
+ bool kvm_timer_dirty;
+ uint64_t kvm_timer_time;
+ uint64_t kvm_timer_compare;
+ uint64_t kvm_timer_state;
+ uint64_t kvm_timer_frequency;
};
OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 171a32adf9..802c076b22 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -64,6 +64,9 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
KVM_REG_RISCV_CSR_REG(name))
+#define RISCV_TIMER_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_TIMER, \
+ KVM_REG_RISCV_TIMER_REG(name))
+
#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
@@ -235,6 +238,75 @@ static int kvm_riscv_put_regs_fp(CPUState *cs)
return ret;
}
+static void kvm_riscv_get_regs_timer(CPUState *cs)
+{
+ int ret;
+ uint64_t reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (env->kvm_timer_dirty) {
+ return;
+ }
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_time = reg;
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_compare = reg;
+
+ ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
+ if (ret) {
+ abort();
+ }
+ env->kvm_timer_state = reg;
+
+ env->kvm_timer_dirty = true;
+}
+
+static void kvm_riscv_put_regs_timer(CPUState *cs)
+{
+ int ret;
+ uint64_t reg;
+ CPURISCVState *env = &RISCV_CPU(cs)->env;
+
+ if (!env->kvm_timer_dirty) {
+ return;
+ }
+
+ reg = env->kvm_timer_time;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
+ if (ret) {
+ abort();
+ }
+
+ reg = env->kvm_timer_compare;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
+ if (ret) {
+ abort();
+ }
+
+ /*
+ * To set register of RISCV_TIMER_REG(state) will occur a error from KVM
+ * on env->kvm_timer_state == 0, It's better to adapt in KVM, but it
+ * doesn't matter that adaping in QEMU now.
+ * TODO If KVM changes, adapt here.
+ */
+ if (env->kvm_timer_state) {
+ reg = env->kvm_timer_state;
+ ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
+ if (ret) {
+ abort();
+ }
+ }
+
+ env->kvm_timer_dirty = false;
+}
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 11/12] target/riscv: Implement virtual time adjusting with vm state changing
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
We hope that virtual time adjusts with vm state changing. When a vm
is stopped, guest virtual time should stop counting and kvm_timer
should be stopped. When the vm is resumed, guest virtual time should
continue to count and kvm_timer should be restored.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/kvm.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 802c076b22..be95dbc3f3 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -40,6 +40,7 @@
#include "kvm_riscv.h"
#include "sbi_ecall_interface.h"
#include "semihosting/console.h"
+#include "sysemu/runstate.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -377,6 +378,17 @@ unsigned long kvm_arch_vcpu_id(CPUState *cpu)
return cpu->cpu_index;
}
+static void kvm_riscv_vm_state_change(void *opaque, bool running, RunState state)
+{
+ CPUState *cs = opaque;
+
+ if (running) {
+ kvm_riscv_put_regs_timer(cs);
+ } else {
+ kvm_riscv_get_regs_timer(cs);
+ }
+}
+
void kvm_arch_init_irq_routing(KVMState *s)
{
}
@@ -389,6 +401,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
CPURISCVState *env = &cpu->env;
uint64_t id;
+ qemu_add_vm_change_state_handler(kvm_riscv_vm_state_change, cs);
+
id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
ret = kvm_get_one_reg(cs, id, &isa);
if (ret) {
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 11/12] target/riscv: Implement virtual time adjusting with vm state changing
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
We hope that virtual time adjusts with vm state changing. When a vm
is stopped, guest virtual time should stop counting and kvm_timer
should be stopped. When the vm is resumed, guest virtual time should
continue to count and kvm_timer should be restored.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/kvm.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
index 802c076b22..be95dbc3f3 100644
--- a/target/riscv/kvm.c
+++ b/target/riscv/kvm.c
@@ -40,6 +40,7 @@
#include "kvm_riscv.h"
#include "sbi_ecall_interface.h"
#include "semihosting/console.h"
+#include "sysemu/runstate.h"
static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
{
@@ -377,6 +378,17 @@ unsigned long kvm_arch_vcpu_id(CPUState *cpu)
return cpu->cpu_index;
}
+static void kvm_riscv_vm_state_change(void *opaque, bool running, RunState state)
+{
+ CPUState *cs = opaque;
+
+ if (running) {
+ kvm_riscv_put_regs_timer(cs);
+ } else {
+ kvm_riscv_get_regs_timer(cs);
+ }
+}
+
void kvm_arch_init_irq_routing(KVMState *s)
{
}
@@ -389,6 +401,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
CPURISCVState *env = &cpu->env;
uint64_t id;
+ qemu_add_vm_change_state_handler(kvm_riscv_vm_state_change, cs);
+
id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
ret = kvm_get_one_reg(cs, id, &isa);
if (ret) {
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-10 10:07 ` Yifei Jiang via
-1 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Add virtual time context description to vmstate_kvmtimer. After cpu being
loaded, virtual time context is updated to KVM.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/machine.c | 37 +++++++++++++++++++++++++++++++++++--
1 file changed, 35 insertions(+), 2 deletions(-)
diff --git a/target/riscv/machine.c b/target/riscv/machine.c
index ad8248ebfd..f46443c316 100644
--- a/target/riscv/machine.c
+++ b/target/riscv/machine.c
@@ -164,10 +164,42 @@ static const VMStateDescription vmstate_pointermasking = {
}
};
+static bool kvmtimer_needed(void *opaque)
+{
+ return kvm_enabled();
+}
+
+
+static const VMStateDescription vmstate_kvmtimer = {
+ .name = "cpu/kvmtimer",
+ .version_id = 1,
+ .minimum_version_id = 1,
+ .needed = kvmtimer_needed,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
+ VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
+ VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
+
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+static int cpu_post_load(void *opaque, int version_id)
+{
+ RISCVCPU *cpu = opaque;
+ CPURISCVState *env = &cpu->env;
+
+ if (kvm_enabled()) {
+ env->kvm_timer_dirty = true;
+ }
+ return 0;
+}
+
const VMStateDescription vmstate_riscv_cpu = {
.name = "cpu",
- .version_id = 3,
- .minimum_version_id = 3,
+ .version_id = 4,
+ .minimum_version_id = 4,
+ .post_load = cpu_post_load,
.fields = (VMStateField[]) {
VMSTATE_UINTTL_ARRAY(env.gpr, RISCVCPU, 32),
VMSTATE_UINT64_ARRAY(env.fpr, RISCVCPU, 32),
@@ -218,6 +250,7 @@ const VMStateDescription vmstate_riscv_cpu = {
&vmstate_hyper,
&vmstate_vector,
&vmstate_pointermasking,
+ &vmstate_kvmtimer,
NULL
}
};
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
@ 2021-12-10 10:07 ` Yifei Jiang via
0 siblings, 0 replies; 56+ messages in thread
From: Yifei Jiang via @ 2021-12-10 10:07 UTC (permalink / raw)
To: qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Yifei Jiang, Mingwang Li
Add virtual time context description to vmstate_kvmtimer. After cpu being
loaded, virtual time context is updated to KVM.
Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
Signed-off-by: Mingwang Li <limingwang@huawei.com>
---
target/riscv/machine.c | 37 +++++++++++++++++++++++++++++++++++--
1 file changed, 35 insertions(+), 2 deletions(-)
diff --git a/target/riscv/machine.c b/target/riscv/machine.c
index ad8248ebfd..f46443c316 100644
--- a/target/riscv/machine.c
+++ b/target/riscv/machine.c
@@ -164,10 +164,42 @@ static const VMStateDescription vmstate_pointermasking = {
}
};
+static bool kvmtimer_needed(void *opaque)
+{
+ return kvm_enabled();
+}
+
+
+static const VMStateDescription vmstate_kvmtimer = {
+ .name = "cpu/kvmtimer",
+ .version_id = 1,
+ .minimum_version_id = 1,
+ .needed = kvmtimer_needed,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
+ VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
+ VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
+
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+static int cpu_post_load(void *opaque, int version_id)
+{
+ RISCVCPU *cpu = opaque;
+ CPURISCVState *env = &cpu->env;
+
+ if (kvm_enabled()) {
+ env->kvm_timer_dirty = true;
+ }
+ return 0;
+}
+
const VMStateDescription vmstate_riscv_cpu = {
.name = "cpu",
- .version_id = 3,
- .minimum_version_id = 3,
+ .version_id = 4,
+ .minimum_version_id = 4,
+ .post_load = cpu_post_load,
.fields = (VMStateField[]) {
VMSTATE_UINTTL_ARRAY(env.gpr, RISCVCPU, 32),
VMSTATE_UINT64_ARRAY(env.fpr, RISCVCPU, 32),
@@ -218,6 +250,7 @@ const VMStateDescription vmstate_riscv_cpu = {
&vmstate_hyper,
&vmstate_vector,
&vmstate_pointermasking,
+ &vmstate_kvmtimer,
NULL
}
};
--
2.19.1
^ permalink raw reply related [flat|nested] 56+ messages in thread
* Re: [PATCH v2 03/12] target/riscv: Implement function kvm_arch_init_vcpu
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:15 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:15 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get isa info from kvm while kvm init.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 32 +++++++++++++++++++++++++++++++-
> 1 file changed, 31 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 687dd4b621..ccf3753048 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -38,6 +38,23 @@
> #include "qemu/log.h"
> #include "hw/loader.h"
>
> +static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> +{
> + uint64_t id = KVM_REG_RISCV | type | idx;
> +
> + switch (riscv_cpu_mxl(env)) {
> + case MXL_RV32:
> + id |= KVM_REG_SIZE_U32;
> + break;
> + case MXL_RV64:
> + id |= KVM_REG_SIZE_U64;
> + break;
> + default:
> + g_assert_not_reached();
> + }
> + return id;
> +}
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
> @@ -79,7 +96,20 @@ void kvm_arch_init_irq_routing(KVMState *s)
>
> int kvm_arch_init_vcpu(CPUState *cs)
> {
> - return 0;
> + int ret = 0;
> + target_ulong isa;
> + RISCVCPU *cpu = RISCV_CPU(cs);
> + CPURISCVState *env = &cpu->env;
> + uint64_t id;
> +
> + id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
> + ret = kvm_get_one_reg(cs, id, &isa);
> + if (ret) {
> + return ret;
> + }
> + env->misa_ext = isa;
> +
> + return ret;
> }
>
> int kvm_arch_msi_data_to_gsi(uint32_t data)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 03/12] target/riscv: Implement function kvm_arch_init_vcpu
@ 2021-12-13 4:15 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:15 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get isa info from kvm while kvm init.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 32 +++++++++++++++++++++++++++++++-
> 1 file changed, 31 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 687dd4b621..ccf3753048 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -38,6 +38,23 @@
> #include "qemu/log.h"
> #include "hw/loader.h"
>
> +static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> +{
> + uint64_t id = KVM_REG_RISCV | type | idx;
> +
> + switch (riscv_cpu_mxl(env)) {
> + case MXL_RV32:
> + id |= KVM_REG_SIZE_U32;
> + break;
> + case MXL_RV64:
> + id |= KVM_REG_SIZE_U64;
> + break;
> + default:
> + g_assert_not_reached();
> + }
> + return id;
> +}
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
> @@ -79,7 +96,20 @@ void kvm_arch_init_irq_routing(KVMState *s)
>
> int kvm_arch_init_vcpu(CPUState *cs)
> {
> - return 0;
> + int ret = 0;
> + target_ulong isa;
> + RISCVCPU *cpu = RISCV_CPU(cs);
> + CPURISCVState *env = &cpu->env;
> + uint64_t id;
> +
> + id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
> + ret = kvm_get_one_reg(cs, id, &isa);
> + if (ret) {
> + return ret;
> + }
> + env->misa_ext = isa;
> +
> + return ret;
> }
>
> int kvm_arch_msi_data_to_gsi(uint32_t data)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 04/12] target/riscv: Implement kvm_arch_get_registers
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:16 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:16 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get GPR CSR and FP registers from kvm by KVM_GET_ONE_REG ioctl.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 112 ++++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 111 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index ccf3753048..6d4df0ef6d 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -55,13 +55,123 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> return id;
> }
>
> +#define RISCV_CORE_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, \
> + KVM_REG_RISCV_CORE_REG(name))
> +
> +#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
> + KVM_REG_RISCV_CSR_REG(name))
> +
> +#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
> +
> +#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
> +
> +#define KVM_RISCV_GET_CSR(cs, env, csr, reg) \
> + do { \
> + int ret = kvm_get_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
> + if (ret) { \
> + return ret; \
> + } \
> + } while(0)
> +
> +static int kvm_riscv_get_regs_core(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + target_ulong reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + ret = kvm_get_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
> + if (ret) {
> + return ret;
> + }
> + env->pc = reg;
> +
> + for (i = 1; i < 32; i++) {
> + uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
> + ret = kvm_get_one_reg(cs, id, ®);
> + if (ret) {
> + return ret;
> + }
> + env->gpr[i] = reg;
> + }
> +
> + return ret;
> +}
> +
> +static int kvm_riscv_get_regs_csr(CPUState *cs)
> +{
> + int ret = 0;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + KVM_RISCV_GET_CSR(cs, env, sstatus, env->mstatus);
> + KVM_RISCV_GET_CSR(cs, env, sie, env->mie);
> + KVM_RISCV_GET_CSR(cs, env, stvec, env->stvec);
> + KVM_RISCV_GET_CSR(cs, env, sscratch, env->sscratch);
> + KVM_RISCV_GET_CSR(cs, env, sepc, env->sepc);
> + KVM_RISCV_GET_CSR(cs, env, scause, env->scause);
> + KVM_RISCV_GET_CSR(cs, env, stval, env->stval);
> + KVM_RISCV_GET_CSR(cs, env, sip, env->mip);
> + KVM_RISCV_GET_CSR(cs, env, satp, env->satp);
> + return ret;
> +}
> +
> +static int kvm_riscv_get_regs_fp(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (riscv_has_ext(env, RVD)) {
> + uint64_t reg;
> + for (i = 0; i < 32; i++) {
> + ret = kvm_get_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + env->fpr[i] = reg;
> + }
> + return ret;
> + }
> +
> + if (riscv_has_ext(env, RVF)) {
> + uint32_t reg;
> + for (i = 0; i < 32; i++) {
> + ret = kvm_get_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + env->fpr[i] = reg;
> + }
> + return ret;
> + }
> +
> + return ret;
> +}
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
>
> int kvm_arch_get_registers(CPUState *cs)
> {
> - return 0;
> + int ret = 0;
> +
> + ret = kvm_riscv_get_regs_core(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_get_regs_csr(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_get_regs_fp(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + return ret;
> }
>
> int kvm_arch_put_registers(CPUState *cs, int level)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 04/12] target/riscv: Implement kvm_arch_get_registers
@ 2021-12-13 4:16 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:16 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get GPR CSR and FP registers from kvm by KVM_GET_ONE_REG ioctl.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 112 ++++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 111 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index ccf3753048..6d4df0ef6d 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -55,13 +55,123 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> return id;
> }
>
> +#define RISCV_CORE_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, \
> + KVM_REG_RISCV_CORE_REG(name))
> +
> +#define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
> + KVM_REG_RISCV_CSR_REG(name))
> +
> +#define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
> +
> +#define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
> +
> +#define KVM_RISCV_GET_CSR(cs, env, csr, reg) \
> + do { \
> + int ret = kvm_get_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
> + if (ret) { \
> + return ret; \
> + } \
> + } while(0)
> +
> +static int kvm_riscv_get_regs_core(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + target_ulong reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + ret = kvm_get_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
> + if (ret) {
> + return ret;
> + }
> + env->pc = reg;
> +
> + for (i = 1; i < 32; i++) {
> + uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
> + ret = kvm_get_one_reg(cs, id, ®);
> + if (ret) {
> + return ret;
> + }
> + env->gpr[i] = reg;
> + }
> +
> + return ret;
> +}
> +
> +static int kvm_riscv_get_regs_csr(CPUState *cs)
> +{
> + int ret = 0;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + KVM_RISCV_GET_CSR(cs, env, sstatus, env->mstatus);
> + KVM_RISCV_GET_CSR(cs, env, sie, env->mie);
> + KVM_RISCV_GET_CSR(cs, env, stvec, env->stvec);
> + KVM_RISCV_GET_CSR(cs, env, sscratch, env->sscratch);
> + KVM_RISCV_GET_CSR(cs, env, sepc, env->sepc);
> + KVM_RISCV_GET_CSR(cs, env, scause, env->scause);
> + KVM_RISCV_GET_CSR(cs, env, stval, env->stval);
> + KVM_RISCV_GET_CSR(cs, env, sip, env->mip);
> + KVM_RISCV_GET_CSR(cs, env, satp, env->satp);
> + return ret;
> +}
> +
> +static int kvm_riscv_get_regs_fp(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (riscv_has_ext(env, RVD)) {
> + uint64_t reg;
> + for (i = 0; i < 32; i++) {
> + ret = kvm_get_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + env->fpr[i] = reg;
> + }
> + return ret;
> + }
> +
> + if (riscv_has_ext(env, RVF)) {
> + uint32_t reg;
> + for (i = 0; i < 32; i++) {
> + ret = kvm_get_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + env->fpr[i] = reg;
> + }
> + return ret;
> + }
> +
> + return ret;
> +}
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
>
> int kvm_arch_get_registers(CPUState *cs)
> {
> - return 0;
> + int ret = 0;
> +
> + ret = kvm_riscv_get_regs_core(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_get_regs_csr(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_get_regs_fp(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + return ret;
> }
>
> int kvm_arch_put_registers(CPUState *cs, int level)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 05/12] target/riscv: Implement kvm_arch_put_registers
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:17 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:17 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Put GPR CSR and FP registers to kvm by KVM_SET_ONE_REG ioctl
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 104 ++++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 103 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 6d4df0ef6d..e695b91dc7 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -73,6 +73,14 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> } \
> } while(0)
>
> +#define KVM_RISCV_SET_CSR(cs, env, csr, reg) \
> + do { \
> + int ret = kvm_set_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
> + if (ret) { \
> + return ret; \
> + } \
> + } while(0)
> +
> static int kvm_riscv_get_regs_core(CPUState *cs)
> {
> int ret = 0;
> @@ -98,6 +106,31 @@ static int kvm_riscv_get_regs_core(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_core(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + target_ulong reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + reg = env->pc;
> + ret = kvm_set_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
> + if (ret) {
> + return ret;
> + }
> +
> + for (i = 1; i < 32; i++) {
> + uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
> + reg = env->gpr[i];
> + ret = kvm_set_one_reg(cs, id, ®);
> + if (ret) {
> + return ret;
> + }
> + }
> +
> + return ret;
> +}
> +
> static int kvm_riscv_get_regs_csr(CPUState *cs)
> {
> int ret = 0;
> @@ -115,6 +148,24 @@ static int kvm_riscv_get_regs_csr(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_csr(CPUState *cs)
> +{
> + int ret = 0;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + KVM_RISCV_SET_CSR(cs, env, sstatus, env->mstatus);
> + KVM_RISCV_SET_CSR(cs, env, sie, env->mie);
> + KVM_RISCV_SET_CSR(cs, env, stvec, env->stvec);
> + KVM_RISCV_SET_CSR(cs, env, sscratch, env->sscratch);
> + KVM_RISCV_SET_CSR(cs, env, sepc, env->sepc);
> + KVM_RISCV_SET_CSR(cs, env, scause, env->scause);
> + KVM_RISCV_SET_CSR(cs, env, stval, env->stval);
> + KVM_RISCV_SET_CSR(cs, env, sip, env->mip);
> + KVM_RISCV_SET_CSR(cs, env, satp, env->satp);
> +
> + return ret;
> +}
> +
> static int kvm_riscv_get_regs_fp(CPUState *cs)
> {
> int ret = 0;
> @@ -148,6 +199,40 @@ static int kvm_riscv_get_regs_fp(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_fp(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (riscv_has_ext(env, RVD)) {
> + uint64_t reg;
> + for (i = 0; i < 32; i++) {
> + reg = env->fpr[i];
> + ret = kvm_set_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + }
> + return ret;
> + }
> +
> + if (riscv_has_ext(env, RVF)) {
> + uint32_t reg;
> + for (i = 0; i < 32; i++) {
> + reg = env->fpr[i];
> + ret = kvm_set_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + }
> + return ret;
> + }
> +
> + return ret;
> +}
> +
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
> @@ -176,7 +261,24 @@ int kvm_arch_get_registers(CPUState *cs)
>
> int kvm_arch_put_registers(CPUState *cs, int level)
> {
> - return 0;
> + int ret = 0;
> +
> + ret = kvm_riscv_put_regs_core(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_put_regs_csr(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_put_regs_fp(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + return ret;
> }
>
> int kvm_arch_release_virq_post(int virq)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 05/12] target/riscv: Implement kvm_arch_put_registers
@ 2021-12-13 4:17 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:17 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Put GPR CSR and FP registers to kvm by KVM_SET_ONE_REG ioctl
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 104 ++++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 103 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 6d4df0ef6d..e695b91dc7 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -73,6 +73,14 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> } \
> } while(0)
>
> +#define KVM_RISCV_SET_CSR(cs, env, csr, reg) \
> + do { \
> + int ret = kvm_set_one_reg(cs, RISCV_CSR_REG(env, csr), ®); \
> + if (ret) { \
> + return ret; \
> + } \
> + } while(0)
> +
> static int kvm_riscv_get_regs_core(CPUState *cs)
> {
> int ret = 0;
> @@ -98,6 +106,31 @@ static int kvm_riscv_get_regs_core(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_core(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + target_ulong reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + reg = env->pc;
> + ret = kvm_set_one_reg(cs, RISCV_CORE_REG(env, regs.pc), ®);
> + if (ret) {
> + return ret;
> + }
> +
> + for (i = 1; i < 32; i++) {
> + uint64_t id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CORE, i);
> + reg = env->gpr[i];
> + ret = kvm_set_one_reg(cs, id, ®);
> + if (ret) {
> + return ret;
> + }
> + }
> +
> + return ret;
> +}
> +
> static int kvm_riscv_get_regs_csr(CPUState *cs)
> {
> int ret = 0;
> @@ -115,6 +148,24 @@ static int kvm_riscv_get_regs_csr(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_csr(CPUState *cs)
> +{
> + int ret = 0;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + KVM_RISCV_SET_CSR(cs, env, sstatus, env->mstatus);
> + KVM_RISCV_SET_CSR(cs, env, sie, env->mie);
> + KVM_RISCV_SET_CSR(cs, env, stvec, env->stvec);
> + KVM_RISCV_SET_CSR(cs, env, sscratch, env->sscratch);
> + KVM_RISCV_SET_CSR(cs, env, sepc, env->sepc);
> + KVM_RISCV_SET_CSR(cs, env, scause, env->scause);
> + KVM_RISCV_SET_CSR(cs, env, stval, env->stval);
> + KVM_RISCV_SET_CSR(cs, env, sip, env->mip);
> + KVM_RISCV_SET_CSR(cs, env, satp, env->satp);
> +
> + return ret;
> +}
> +
> static int kvm_riscv_get_regs_fp(CPUState *cs)
> {
> int ret = 0;
> @@ -148,6 +199,40 @@ static int kvm_riscv_get_regs_fp(CPUState *cs)
> return ret;
> }
>
> +static int kvm_riscv_put_regs_fp(CPUState *cs)
> +{
> + int ret = 0;
> + int i;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (riscv_has_ext(env, RVD)) {
> + uint64_t reg;
> + for (i = 0; i < 32; i++) {
> + reg = env->fpr[i];
> + ret = kvm_set_one_reg(cs, RISCV_FP_D_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + }
> + return ret;
> + }
> +
> + if (riscv_has_ext(env, RVF)) {
> + uint32_t reg;
> + for (i = 0; i < 32; i++) {
> + reg = env->fpr[i];
> + ret = kvm_set_one_reg(cs, RISCV_FP_F_REG(env, i), ®);
> + if (ret) {
> + return ret;
> + }
> + }
> + return ret;
> + }
> +
> + return ret;
> +}
> +
> +
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> };
> @@ -176,7 +261,24 @@ int kvm_arch_get_registers(CPUState *cs)
>
> int kvm_arch_put_registers(CPUState *cs, int level)
> {
> - return 0;
> + int ret = 0;
> +
> + ret = kvm_riscv_put_regs_core(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_put_regs_csr(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + ret = kvm_riscv_put_regs_fp(cs);
> + if (ret) {
> + return ret;
> + }
> +
> + return ret;
> }
>
> int kvm_arch_release_virq_post(int virq)
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:21 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:21 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get kernel and fdt start address in virt.c, and pass them to KVM
> when cpu reset. In addition, add kvm_riscv.h to place riscv specific
> interface.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> ---
> hw/riscv/boot.c | 11 ++++++++
> hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
> include/hw/riscv/boot.h | 1 +
> target/riscv/cpu.c | 8 ++++++
> target/riscv/cpu.h | 3 +++
> target/riscv/kvm-stub.c | 25 +++++++++++++++++++
> target/riscv/kvm.c | 14 +++++++++++
> target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
> target/riscv/meson.build | 2 +-
> 9 files changed, 125 insertions(+), 17 deletions(-)
> create mode 100644 target/riscv/kvm-stub.c
> create mode 100644 target/riscv/kvm_riscv.h
>
> diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
> index 519fa455a1..00df6d7810 100644
> --- a/hw/riscv/boot.c
> +++ b/hw/riscv/boot.c
> @@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState *machine, RISCVHartArrayState *harts
>
> return;
> }
> +
> +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
> +{
> + CPUState *cs;
> +
> + for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
> + RISCVCPU *riscv_cpu = RISCV_CPU(cs);
> + riscv_cpu->env.kernel_addr = kernel_addr;
> + riscv_cpu->env.fdt_addr = fdt_addr;
> + }
> +}
> diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
> index 3af074148e..c8bcd9d9e5 100644
> --- a/hw/riscv/virt.c
> +++ b/hw/riscv/virt.c
> @@ -38,6 +38,7 @@
> #include "chardev/char.h"
> #include "sysemu/device_tree.h"
> #include "sysemu/sysemu.h"
> +#include "sysemu/kvm.h"
> #include "hw/pci/pci.h"
> #include "hw/pci-host/gpex.h"
> #include "hw/display/ramfb.h"
> @@ -801,23 +802,25 @@ static void virt_machine_init(MachineState *machine)
> hart_count, &error_abort);
> sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
>
> - /* Per-socket CLINT */
> - riscv_aclint_swi_create(
> - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> - base_hartid, hart_count, false);
> - riscv_aclint_mtimer_create(
> - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
> - RISCV_ACLINT_SWI_SIZE,
> - RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
> - RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
> - RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> -
> - /* Per-socket ACLINT SSWI */
> - if (s->have_aclint) {
> + if (!kvm_enabled()) {
> + /* Per-socket CLINT */
> riscv_aclint_swi_create(
> - memmap[VIRT_ACLINT_SSWI].base +
> - i * memmap[VIRT_ACLINT_SSWI].size,
> - base_hartid, hart_count, true);
> + memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> + base_hartid, hart_count, false);
> + riscv_aclint_mtimer_create(
> + memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
> + RISCV_ACLINT_SWI_SIZE,
> + RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
> + RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
> + RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> +
> + /* Per-socket ACLINT SSWI */
> + if (s->have_aclint) {
> + riscv_aclint_swi_create(
> + memmap[VIRT_ACLINT_SSWI].base +
> + i * memmap[VIRT_ACLINT_SSWI].size,
> + base_hartid, hart_count, true);
> + }
Along with this, we should not generate FDT nodes of ACLINT (or SiFive CLINT)
when KVM is enabled.
> }
>
> /* Per-socket PLIC hart topology configuration string */
> @@ -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
> memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base,
> mask_rom);
>
> + /*
> + * Only direct boot kernel is currently supported for KVM VM,
> + * so the "-bios" parameter is ignored and treated like "-bios none"
> + * when KVM is enabled.
> + */
> + if (kvm_enabled()) {
> + g_free(machine->firmware);
> + machine->firmware = g_strdup("none");
> + }
> +
> if (riscv_is_32bit(&s->soc[0])) {
> firmware_end_addr = riscv_find_and_load_firmware(machine,
> RISCV32_BIOS_BIN, start_addr, NULL);
> @@ -941,6 +954,15 @@ static void virt_machine_init(MachineState *machine)
> virt_memmap[VIRT_MROM].size, kernel_entry,
> fdt_load_addr, machine->fdt);
>
> + /*
> + * Only direct boot kernel is currently supported for KVM VM,
> + * So here setup kernel start address and fdt address.
> + * TODO:Support firmware loading and integrate to TCG start
> + */
> + if (kvm_enabled()) {
> + riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
> + }
> +
> /* SiFive Test MMIO device */
> sifive_test_create(memmap[VIRT_TEST].base);
>
> diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
> index baff11dd8a..5834c234aa 100644
> --- a/include/hw/riscv/boot.h
> +++ b/include/hw/riscv/boot.h
> @@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
> hwaddr rom_size,
> uint32_t reset_vec_size,
> uint64_t kernel_entry);
> +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
>
> #endif /* RISCV_BOOT_H */
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index f812998123..1c944872a3 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -29,6 +29,8 @@
> #include "hw/qdev-properties.h"
> #include "migration/vmstate.h"
> #include "fpu/softfloat-helpers.h"
> +#include "sysemu/kvm.h"
> +#include "kvm_riscv.h"
>
> /* RISC-V CPU definitions */
>
> @@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
> cs->exception_index = RISCV_EXCP_NONE;
> env->load_res = -1;
> set_default_nan_mode(1, &env->fp_status);
> +
> +#ifndef CONFIG_USER_ONLY
> + if (kvm_enabled()) {
> + kvm_riscv_reset_vcpu(cpu);
> + }
> +#endif
> }
>
> static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info)
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index 0760c0af93..2807eb1bcb 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -255,6 +255,9 @@ struct CPURISCVState {
>
> /* Fields from here on are preserved across CPU reset. */
> QEMUTimer *timer; /* Internal timer */
> +
> + hwaddr kernel_addr;
> + hwaddr fdt_addr;
> };
>
> OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
> diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
> new file mode 100644
> index 0000000000..39b96fe3f4
> --- /dev/null
> +++ b/target/riscv/kvm-stub.c
> @@ -0,0 +1,25 @@
> +/*
> + * QEMU KVM RISC-V specific function stubs
> + *
> + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +#include "qemu/osdep.h"
> +#include "cpu.h"
> +#include "kvm_riscv.h"
> +
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> +{
> + abort();
> +}
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index e695b91dc7..db6d8a5b6e 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -37,6 +37,7 @@
> #include "hw/irq.h"
> #include "qemu/log.h"
> #include "hw/loader.h"
> +#include "kvm_riscv.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -369,6 +370,19 @@ int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> return 0;
> }
>
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> +{
> + CPURISCVState *env = &cpu->env;
> +
> + if (!kvm_enabled()) {
> + return;
> + }
> + env->pc = cpu->env.kernel_addr;
> + env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
> + env->gpr[11] = cpu->env.fdt_addr; /* a1 */
> + env->satp = 0;
> +}
> +
> bool kvm_arch_cpu_check_are_resettable(void)
> {
> return true;
> diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
> new file mode 100644
> index 0000000000..f38c82bf59
> --- /dev/null
> +++ b/target/riscv/kvm_riscv.h
> @@ -0,0 +1,24 @@
> +/*
> + * QEMU KVM support -- RISC-V specific functions.
> + *
> + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef QEMU_KVM_RISCV_H
> +#define QEMU_KVM_RISCV_H
> +
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> +
> +#endif
> diff --git a/target/riscv/meson.build b/target/riscv/meson.build
> index 2faf08a941..fe41cc5805 100644
> --- a/target/riscv/meson.build
> +++ b/target/riscv/meson.build
> @@ -19,7 +19,7 @@ riscv_ss.add(files(
> 'bitmanip_helper.c',
> 'translate.c',
> ))
> -riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
> +riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c'))
>
> riscv_softmmu_ss = ss.source_set()
> riscv_softmmu_ss.add(files(
> --
> 2.19.1
>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
@ 2021-12-13 4:21 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:21 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Get kernel and fdt start address in virt.c, and pass them to KVM
> when cpu reset. In addition, add kvm_riscv.h to place riscv specific
> interface.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> ---
> hw/riscv/boot.c | 11 ++++++++
> hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
> include/hw/riscv/boot.h | 1 +
> target/riscv/cpu.c | 8 ++++++
> target/riscv/cpu.h | 3 +++
> target/riscv/kvm-stub.c | 25 +++++++++++++++++++
> target/riscv/kvm.c | 14 +++++++++++
> target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
> target/riscv/meson.build | 2 +-
> 9 files changed, 125 insertions(+), 17 deletions(-)
> create mode 100644 target/riscv/kvm-stub.c
> create mode 100644 target/riscv/kvm_riscv.h
>
> diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c
> index 519fa455a1..00df6d7810 100644
> --- a/hw/riscv/boot.c
> +++ b/hw/riscv/boot.c
> @@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState *machine, RISCVHartArrayState *harts
>
> return;
> }
> +
> +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr)
> +{
> + CPUState *cs;
> +
> + for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
> + RISCVCPU *riscv_cpu = RISCV_CPU(cs);
> + riscv_cpu->env.kernel_addr = kernel_addr;
> + riscv_cpu->env.fdt_addr = fdt_addr;
> + }
> +}
> diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c
> index 3af074148e..c8bcd9d9e5 100644
> --- a/hw/riscv/virt.c
> +++ b/hw/riscv/virt.c
> @@ -38,6 +38,7 @@
> #include "chardev/char.h"
> #include "sysemu/device_tree.h"
> #include "sysemu/sysemu.h"
> +#include "sysemu/kvm.h"
> #include "hw/pci/pci.h"
> #include "hw/pci-host/gpex.h"
> #include "hw/display/ramfb.h"
> @@ -801,23 +802,25 @@ static void virt_machine_init(MachineState *machine)
> hart_count, &error_abort);
> sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
>
> - /* Per-socket CLINT */
> - riscv_aclint_swi_create(
> - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> - base_hartid, hart_count, false);
> - riscv_aclint_mtimer_create(
> - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
> - RISCV_ACLINT_SWI_SIZE,
> - RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
> - RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
> - RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> -
> - /* Per-socket ACLINT SSWI */
> - if (s->have_aclint) {
> + if (!kvm_enabled()) {
> + /* Per-socket CLINT */
> riscv_aclint_swi_create(
> - memmap[VIRT_ACLINT_SSWI].base +
> - i * memmap[VIRT_ACLINT_SSWI].size,
> - base_hartid, hart_count, true);
> + memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> + base_hartid, hart_count, false);
> + riscv_aclint_mtimer_create(
> + memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size +
> + RISCV_ACLINT_SWI_SIZE,
> + RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid, hart_count,
> + RISCV_ACLINT_DEFAULT_MTIMECMP, RISCV_ACLINT_DEFAULT_MTIME,
> + RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> +
> + /* Per-socket ACLINT SSWI */
> + if (s->have_aclint) {
> + riscv_aclint_swi_create(
> + memmap[VIRT_ACLINT_SSWI].base +
> + i * memmap[VIRT_ACLINT_SSWI].size,
> + base_hartid, hart_count, true);
> + }
Along with this, we should not generate FDT nodes of ACLINT (or SiFive CLINT)
when KVM is enabled.
> }
>
> /* Per-socket PLIC hart topology configuration string */
> @@ -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
> memory_region_add_subregion(system_memory, memmap[VIRT_MROM].base,
> mask_rom);
>
> + /*
> + * Only direct boot kernel is currently supported for KVM VM,
> + * so the "-bios" parameter is ignored and treated like "-bios none"
> + * when KVM is enabled.
> + */
> + if (kvm_enabled()) {
> + g_free(machine->firmware);
> + machine->firmware = g_strdup("none");
> + }
> +
> if (riscv_is_32bit(&s->soc[0])) {
> firmware_end_addr = riscv_find_and_load_firmware(machine,
> RISCV32_BIOS_BIN, start_addr, NULL);
> @@ -941,6 +954,15 @@ static void virt_machine_init(MachineState *machine)
> virt_memmap[VIRT_MROM].size, kernel_entry,
> fdt_load_addr, machine->fdt);
>
> + /*
> + * Only direct boot kernel is currently supported for KVM VM,
> + * So here setup kernel start address and fdt address.
> + * TODO:Support firmware loading and integrate to TCG start
> + */
> + if (kvm_enabled()) {
> + riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
> + }
> +
> /* SiFive Test MMIO device */
> sifive_test_create(memmap[VIRT_TEST].base);
>
> diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h
> index baff11dd8a..5834c234aa 100644
> --- a/include/hw/riscv/boot.h
> +++ b/include/hw/riscv/boot.h
> @@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState *machine, hwaddr rom_base,
> hwaddr rom_size,
> uint32_t reset_vec_size,
> uint64_t kernel_entry);
> +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
>
> #endif /* RISCV_BOOT_H */
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index f812998123..1c944872a3 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -29,6 +29,8 @@
> #include "hw/qdev-properties.h"
> #include "migration/vmstate.h"
> #include "fpu/softfloat-helpers.h"
> +#include "sysemu/kvm.h"
> +#include "kvm_riscv.h"
>
> /* RISC-V CPU definitions */
>
> @@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
> cs->exception_index = RISCV_EXCP_NONE;
> env->load_res = -1;
> set_default_nan_mode(1, &env->fp_status);
> +
> +#ifndef CONFIG_USER_ONLY
> + if (kvm_enabled()) {
> + kvm_riscv_reset_vcpu(cpu);
> + }
> +#endif
> }
>
> static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info)
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index 0760c0af93..2807eb1bcb 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -255,6 +255,9 @@ struct CPURISCVState {
>
> /* Fields from here on are preserved across CPU reset. */
> QEMUTimer *timer; /* Internal timer */
> +
> + hwaddr kernel_addr;
> + hwaddr fdt_addr;
> };
>
> OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
> diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
> new file mode 100644
> index 0000000000..39b96fe3f4
> --- /dev/null
> +++ b/target/riscv/kvm-stub.c
> @@ -0,0 +1,25 @@
> +/*
> + * QEMU KVM RISC-V specific function stubs
> + *
> + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +#include "qemu/osdep.h"
> +#include "cpu.h"
> +#include "kvm_riscv.h"
> +
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> +{
> + abort();
> +}
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index e695b91dc7..db6d8a5b6e 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -37,6 +37,7 @@
> #include "hw/irq.h"
> #include "qemu/log.h"
> #include "hw/loader.h"
> +#include "kvm_riscv.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -369,6 +370,19 @@ int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> return 0;
> }
>
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> +{
> + CPURISCVState *env = &cpu->env;
> +
> + if (!kvm_enabled()) {
> + return;
> + }
> + env->pc = cpu->env.kernel_addr;
> + env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
> + env->gpr[11] = cpu->env.fdt_addr; /* a1 */
> + env->satp = 0;
> +}
> +
> bool kvm_arch_cpu_check_are_resettable(void)
> {
> return true;
> diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
> new file mode 100644
> index 0000000000..f38c82bf59
> --- /dev/null
> +++ b/target/riscv/kvm_riscv.h
> @@ -0,0 +1,24 @@
> +/*
> + * QEMU KVM support -- RISC-V specific functions.
> + *
> + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program. If not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef QEMU_KVM_RISCV_H
> +#define QEMU_KVM_RISCV_H
> +
> +void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> +
> +#endif
> diff --git a/target/riscv/meson.build b/target/riscv/meson.build
> index 2faf08a941..fe41cc5805 100644
> --- a/target/riscv/meson.build
> +++ b/target/riscv/meson.build
> @@ -19,7 +19,7 @@ riscv_ss.add(files(
> 'bitmanip_helper.c',
> 'translate.c',
> ))
> -riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
> +riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false: files('kvm-stub.c'))
>
> riscv_softmmu_ss = ss.source_set()
> riscv_softmmu_ss.add(files(
> --
> 2.19.1
>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:32 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:32 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> When KVM is enabled, set the S-mode external interrupt through
> kvm_riscv_set_irq function.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> ---
> target/riscv/cpu.c | 6 +++++-
> target/riscv/kvm-stub.c | 5 +++++
> target/riscv/kvm.c | 17 +++++++++++++++++
> target/riscv/kvm_riscv.h | 1 +
> 4 files changed, 28 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index 1c944872a3..71a7ac6831 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq, int level)
> case IRQ_S_EXT:
> case IRQ_VS_EXT:
> case IRQ_M_EXT:
> - riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> + if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
> + kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
> + } else {
> + riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> + }
This does not look right.
I suggest the following:
if (kvm_enabled()) {
kvm_riscv_set_irq(cpu, irq, level);
} else {
riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
}
> break;
> default:
> g_assert_not_reached();
> diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
> index 39b96fe3f4..4e8fc31a21 100644
> --- a/target/riscv/kvm-stub.c
> +++ b/target/riscv/kvm-stub.c
> @@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> {
> abort();
> }
> +
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
> +{
> + abort();
> +}
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index db6d8a5b6e..0027f11f45 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> env->satp = 0;
> }
>
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
> +{
> + int ret;
> + unsigned virq = level ? KVM_INTERRUPT_SET : KVM_INTERRUPT_UNSET;
> +
> + if (irq != IRQ_S_EXT) {
> + perror("kvm riscv set irq != IRQ_S_EXT\n");
> + abort();
> + }
> +
> + ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
> + if (ret < 0) {
> + perror("Set irq failed");
> + abort();
> + }
> +}
> +
> bool kvm_arch_cpu_check_are_resettable(void)
> {
> return true;
> diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
> index f38c82bf59..ed281bdce0 100644
> --- a/target/riscv/kvm_riscv.h
> +++ b/target/riscv/kvm_riscv.h
> @@ -20,5 +20,6 @@
> #define QEMU_KVM_RISCV_H
>
> void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
>
> #endif
> --
> 2.19.1
>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
@ 2021-12-13 4:32 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:32 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> When KVM is enabled, set the S-mode external interrupt through
> kvm_riscv_set_irq function.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> ---
> target/riscv/cpu.c | 6 +++++-
> target/riscv/kvm-stub.c | 5 +++++
> target/riscv/kvm.c | 17 +++++++++++++++++
> target/riscv/kvm_riscv.h | 1 +
> 4 files changed, 28 insertions(+), 1 deletion(-)
>
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index 1c944872a3..71a7ac6831 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq, int level)
> case IRQ_S_EXT:
> case IRQ_VS_EXT:
> case IRQ_M_EXT:
> - riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> + if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
> + kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
> + } else {
> + riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> + }
This does not look right.
I suggest the following:
if (kvm_enabled()) {
kvm_riscv_set_irq(cpu, irq, level);
} else {
riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
}
> break;
> default:
> g_assert_not_reached();
> diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c
> index 39b96fe3f4..4e8fc31a21 100644
> --- a/target/riscv/kvm-stub.c
> +++ b/target/riscv/kvm-stub.c
> @@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> {
> abort();
> }
> +
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
> +{
> + abort();
> +}
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index db6d8a5b6e..0027f11f45 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> env->satp = 0;
> }
>
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level)
> +{
> + int ret;
> + unsigned virq = level ? KVM_INTERRUPT_SET : KVM_INTERRUPT_UNSET;
> +
> + if (irq != IRQ_S_EXT) {
> + perror("kvm riscv set irq != IRQ_S_EXT\n");
> + abort();
> + }
> +
> + ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
> + if (ret < 0) {
> + perror("Set irq failed");
> + abort();
> + }
> +}
> +
> bool kvm_arch_cpu_check_are_resettable(void)
> {
> return true;
> diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h
> index f38c82bf59..ed281bdce0 100644
> --- a/target/riscv/kvm_riscv.h
> +++ b/target/riscv/kvm_riscv.h
> @@ -20,5 +20,6 @@
> #define QEMU_KVM_RISCV_H
>
> void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
>
> #endif
> --
> 2.19.1
>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 08/12] target/riscv: Handle KVM_EXIT_RISCV_SBI exit
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:37 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:37 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Use char-fe to handle console sbi call, which implement early
> console io while apply 'earlycon=sbi' into kernel parameters.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 38 +++++++++++++++-
> target/riscv/sbi_ecall_interface.h | 72 ++++++++++++++++++++++++++++++
> 2 files changed, 109 insertions(+), 1 deletion(-)
> create mode 100644 target/riscv/sbi_ecall_interface.h
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 0027f11f45..171a32adf9 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -38,6 +38,8 @@
> #include "qemu/log.h"
> #include "hw/loader.h"
> #include "kvm_riscv.h"
> +#include "sbi_ecall_interface.h"
> +#include "semihosting/console.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -365,9 +367,43 @@ bool kvm_arch_stop_on_emulation_error(CPUState *cs)
> return true;
> }
>
> +static int kvm_riscv_handle_sbi(CPUState *cs, struct kvm_run *run)
> +{
> + int ret = 0;
> + unsigned char ch;
> + switch (run->riscv_sbi.extension_id) {
> + case SBI_EXT_0_1_CONSOLE_PUTCHAR:
> + ch = run->riscv_sbi.args[0];
> + qemu_semihosting_log_out((const char *)&ch, sizeof(ch));
> + break;
> + case SBI_EXT_0_1_CONSOLE_GETCHAR:
> + run->riscv_sbi.args[0] =
> + (unsigned long)qemu_semihosting_console_inc(cs->env_ptr);
> + break;
> + default:
> + qemu_log_mask(LOG_UNIMP,
> + "%s: un-handled SBI EXIT, specific reasons is %lu\n",
> + __func__, run->riscv_sbi.extension_id);
> + ret = -1;
> + break;
> + }
> + return ret;
> +}
> +
> int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> {
> - return 0;
> + int ret = 0;
> + switch (run->exit_reason) {
> + case KVM_EXIT_RISCV_SBI:
> + ret = kvm_riscv_handle_sbi(cs, run);
> + break;
> + default:
> + qemu_log_mask(LOG_UNIMP, "%s: un-handled exit reason %d\n",
> + __func__, run->exit_reason);
> + ret = -1;
> + break;
> + }
> + return ret;
> }
>
> void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> diff --git a/target/riscv/sbi_ecall_interface.h b/target/riscv/sbi_ecall_interface.h
> new file mode 100644
> index 0000000000..fb1a3fa8f2
> --- /dev/null
> +++ b/target/riscv/sbi_ecall_interface.h
> @@ -0,0 +1,72 @@
> +/*
> + * SPDX-License-Identifier: BSD-2-Clause
> + *
> + * Copyright (c) 2019 Western Digital Corporation or its affiliates.
> + *
> + * Authors:
> + * Anup Patel <anup.patel@wdc.com>
> + */
> +
> +#ifndef __SBI_ECALL_INTERFACE_H__
> +#define __SBI_ECALL_INTERFACE_H__
> +
> +/* clang-format off */
> +
> +/* SBI Extension IDs */
> +#define SBI_EXT_0_1_SET_TIMER 0x0
> +#define SBI_EXT_0_1_CONSOLE_PUTCHAR 0x1
> +#define SBI_EXT_0_1_CONSOLE_GETCHAR 0x2
> +#define SBI_EXT_0_1_CLEAR_IPI 0x3
> +#define SBI_EXT_0_1_SEND_IPI 0x4
> +#define SBI_EXT_0_1_REMOTE_FENCE_I 0x5
> +#define SBI_EXT_0_1_REMOTE_SFENCE_VMA 0x6
> +#define SBI_EXT_0_1_REMOTE_SFENCE_VMA_ASID 0x7
> +#define SBI_EXT_0_1_SHUTDOWN 0x8
> +#define SBI_EXT_BASE 0x10
> +#define SBI_EXT_TIME 0x54494D45
> +#define SBI_EXT_IPI 0x735049
> +#define SBI_EXT_RFENCE 0x52464E43
> +#define SBI_EXT_HSM 0x48534D
> +
> +/* SBI function IDs for BASE extension*/
> +#define SBI_EXT_BASE_GET_SPEC_VERSION 0x0
> +#define SBI_EXT_BASE_GET_IMP_ID 0x1
> +#define SBI_EXT_BASE_GET_IMP_VERSION 0x2
> +#define SBI_EXT_BASE_PROBE_EXT 0x3
> +#define SBI_EXT_BASE_GET_MVENDORID 0x4
> +#define SBI_EXT_BASE_GET_MARCHID 0x5
> +#define SBI_EXT_BASE_GET_MIMPID 0x6
> +
> +/* SBI function IDs for TIME extension*/
> +#define SBI_EXT_TIME_SET_TIMER 0x0
> +
> +/* SBI function IDs for IPI extension*/
> +#define SBI_EXT_IPI_SEND_IPI 0x0
> +
> +/* SBI function IDs for RFENCE extension*/
> +#define SBI_EXT_RFENCE_REMOTE_FENCE_I 0x0
> +#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA 0x1
> +#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID 0x2
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA 0x3
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA_VMID 0x4
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA 0x5
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID 0x6
> +
> +/* SBI function IDs for HSM extension */
> +#define SBI_EXT_HSM_HART_START 0x0
> +#define SBI_EXT_HSM_HART_STOP 0x1
> +#define SBI_EXT_HSM_HART_GET_STATUS 0x2
> +
> +#define SBI_HSM_HART_STATUS_STARTED 0x0
> +#define SBI_HSM_HART_STATUS_STOPPED 0x1
> +#define SBI_HSM_HART_STATUS_START_PENDING 0x2
> +#define SBI_HSM_HART_STATUS_STOP_PENDING 0x3
> +
> +#define SBI_SPEC_VERSION_MAJOR_OFFSET 24
> +#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f
> +#define SBI_SPEC_VERSION_MINOR_MASK 0xffffff
> +#define SBI_EXT_VENDOR_START 0x09000000
> +#define SBI_EXT_VENDOR_END 0x09FFFFFF
> +/* clang-format on */
> +
> +#endif
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 08/12] target/riscv: Handle KVM_EXIT_RISCV_SBI exit
@ 2021-12-13 4:37 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:37 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Use char-fe to handle console sbi call, which implement early
> console io while apply 'earlycon=sbi' into kernel parameters.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 38 +++++++++++++++-
> target/riscv/sbi_ecall_interface.h | 72 ++++++++++++++++++++++++++++++
> 2 files changed, 109 insertions(+), 1 deletion(-)
> create mode 100644 target/riscv/sbi_ecall_interface.h
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 0027f11f45..171a32adf9 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -38,6 +38,8 @@
> #include "qemu/log.h"
> #include "hw/loader.h"
> #include "kvm_riscv.h"
> +#include "sbi_ecall_interface.h"
> +#include "semihosting/console.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -365,9 +367,43 @@ bool kvm_arch_stop_on_emulation_error(CPUState *cs)
> return true;
> }
>
> +static int kvm_riscv_handle_sbi(CPUState *cs, struct kvm_run *run)
> +{
> + int ret = 0;
> + unsigned char ch;
> + switch (run->riscv_sbi.extension_id) {
> + case SBI_EXT_0_1_CONSOLE_PUTCHAR:
> + ch = run->riscv_sbi.args[0];
> + qemu_semihosting_log_out((const char *)&ch, sizeof(ch));
> + break;
> + case SBI_EXT_0_1_CONSOLE_GETCHAR:
> + run->riscv_sbi.args[0] =
> + (unsigned long)qemu_semihosting_console_inc(cs->env_ptr);
> + break;
> + default:
> + qemu_log_mask(LOG_UNIMP,
> + "%s: un-handled SBI EXIT, specific reasons is %lu\n",
> + __func__, run->riscv_sbi.extension_id);
> + ret = -1;
> + break;
> + }
> + return ret;
> +}
> +
> int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> {
> - return 0;
> + int ret = 0;
> + switch (run->exit_reason) {
> + case KVM_EXIT_RISCV_SBI:
> + ret = kvm_riscv_handle_sbi(cs, run);
> + break;
> + default:
> + qemu_log_mask(LOG_UNIMP, "%s: un-handled exit reason %d\n",
> + __func__, run->exit_reason);
> + ret = -1;
> + break;
> + }
> + return ret;
> }
>
> void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> diff --git a/target/riscv/sbi_ecall_interface.h b/target/riscv/sbi_ecall_interface.h
> new file mode 100644
> index 0000000000..fb1a3fa8f2
> --- /dev/null
> +++ b/target/riscv/sbi_ecall_interface.h
> @@ -0,0 +1,72 @@
> +/*
> + * SPDX-License-Identifier: BSD-2-Clause
> + *
> + * Copyright (c) 2019 Western Digital Corporation or its affiliates.
> + *
> + * Authors:
> + * Anup Patel <anup.patel@wdc.com>
> + */
> +
> +#ifndef __SBI_ECALL_INTERFACE_H__
> +#define __SBI_ECALL_INTERFACE_H__
> +
> +/* clang-format off */
> +
> +/* SBI Extension IDs */
> +#define SBI_EXT_0_1_SET_TIMER 0x0
> +#define SBI_EXT_0_1_CONSOLE_PUTCHAR 0x1
> +#define SBI_EXT_0_1_CONSOLE_GETCHAR 0x2
> +#define SBI_EXT_0_1_CLEAR_IPI 0x3
> +#define SBI_EXT_0_1_SEND_IPI 0x4
> +#define SBI_EXT_0_1_REMOTE_FENCE_I 0x5
> +#define SBI_EXT_0_1_REMOTE_SFENCE_VMA 0x6
> +#define SBI_EXT_0_1_REMOTE_SFENCE_VMA_ASID 0x7
> +#define SBI_EXT_0_1_SHUTDOWN 0x8
> +#define SBI_EXT_BASE 0x10
> +#define SBI_EXT_TIME 0x54494D45
> +#define SBI_EXT_IPI 0x735049
> +#define SBI_EXT_RFENCE 0x52464E43
> +#define SBI_EXT_HSM 0x48534D
> +
> +/* SBI function IDs for BASE extension*/
> +#define SBI_EXT_BASE_GET_SPEC_VERSION 0x0
> +#define SBI_EXT_BASE_GET_IMP_ID 0x1
> +#define SBI_EXT_BASE_GET_IMP_VERSION 0x2
> +#define SBI_EXT_BASE_PROBE_EXT 0x3
> +#define SBI_EXT_BASE_GET_MVENDORID 0x4
> +#define SBI_EXT_BASE_GET_MARCHID 0x5
> +#define SBI_EXT_BASE_GET_MIMPID 0x6
> +
> +/* SBI function IDs for TIME extension*/
> +#define SBI_EXT_TIME_SET_TIMER 0x0
> +
> +/* SBI function IDs for IPI extension*/
> +#define SBI_EXT_IPI_SEND_IPI 0x0
> +
> +/* SBI function IDs for RFENCE extension*/
> +#define SBI_EXT_RFENCE_REMOTE_FENCE_I 0x0
> +#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA 0x1
> +#define SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID 0x2
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA 0x3
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_GVMA_VMID 0x4
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA 0x5
> +#define SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID 0x6
> +
> +/* SBI function IDs for HSM extension */
> +#define SBI_EXT_HSM_HART_START 0x0
> +#define SBI_EXT_HSM_HART_STOP 0x1
> +#define SBI_EXT_HSM_HART_GET_STATUS 0x2
> +
> +#define SBI_HSM_HART_STATUS_STARTED 0x0
> +#define SBI_HSM_HART_STATUS_STOPPED 0x1
> +#define SBI_HSM_HART_STATUS_START_PENDING 0x2
> +#define SBI_HSM_HART_STATUS_STOP_PENDING 0x3
> +
> +#define SBI_SPEC_VERSION_MAJOR_OFFSET 24
> +#define SBI_SPEC_VERSION_MAJOR_MASK 0x7f
> +#define SBI_SPEC_VERSION_MINOR_MASK 0xffffff
> +#define SBI_EXT_VENDOR_START 0x09000000
> +#define SBI_EXT_VENDOR_END 0x09FFFFFF
> +/* clang-format on */
> +
> +#endif
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 11/12] target/riscv: Implement virtual time adjusting with vm state changing
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:39 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:39 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:38 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> We hope that virtual time adjusts with vm state changing. When a vm
> is stopped, guest virtual time should stop counting and kvm_timer
> should be stopped. When the vm is resumed, guest virtual time should
> continue to count and kvm_timer should be restored.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 14 ++++++++++++++
> 1 file changed, 14 insertions(+)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 802c076b22..be95dbc3f3 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -40,6 +40,7 @@
> #include "kvm_riscv.h"
> #include "sbi_ecall_interface.h"
> #include "semihosting/console.h"
> +#include "sysemu/runstate.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -377,6 +378,17 @@ unsigned long kvm_arch_vcpu_id(CPUState *cpu)
> return cpu->cpu_index;
> }
>
> +static void kvm_riscv_vm_state_change(void *opaque, bool running, RunState state)
> +{
> + CPUState *cs = opaque;
> +
> + if (running) {
> + kvm_riscv_put_regs_timer(cs);
> + } else {
> + kvm_riscv_get_regs_timer(cs);
> + }
> +}
> +
> void kvm_arch_init_irq_routing(KVMState *s)
> {
> }
> @@ -389,6 +401,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
> CPURISCVState *env = &cpu->env;
> uint64_t id;
>
> + qemu_add_vm_change_state_handler(kvm_riscv_vm_state_change, cs);
> +
> id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
> ret = kvm_get_one_reg(cs, id, &isa);
> if (ret) {
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 11/12] target/riscv: Implement virtual time adjusting with vm state changing
@ 2021-12-13 4:39 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:39 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:38 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> We hope that virtual time adjusts with vm state changing. When a vm
> is stopped, guest virtual time should stop counting and kvm_timer
> should be stopped. When the vm is resumed, guest virtual time should
> continue to count and kvm_timer should be restored.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
Looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
> ---
> target/riscv/kvm.c | 14 ++++++++++++++
> 1 file changed, 14 insertions(+)
>
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 802c076b22..be95dbc3f3 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -40,6 +40,7 @@
> #include "kvm_riscv.h"
> #include "sbi_ecall_interface.h"
> #include "semihosting/console.h"
> +#include "sysemu/runstate.h"
>
> static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx)
> {
> @@ -377,6 +378,17 @@ unsigned long kvm_arch_vcpu_id(CPUState *cpu)
> return cpu->cpu_index;
> }
>
> +static void kvm_riscv_vm_state_change(void *opaque, bool running, RunState state)
> +{
> + CPUState *cs = opaque;
> +
> + if (running) {
> + kvm_riscv_put_regs_timer(cs);
> + } else {
> + kvm_riscv_get_regs_timer(cs);
> + }
> +}
> +
> void kvm_arch_init_irq_routing(KVMState *s)
> {
> }
> @@ -389,6 +401,8 @@ int kvm_arch_init_vcpu(CPUState *cs)
> CPURISCVState *env = &cpu->env;
> uint64_t id;
>
> + qemu_add_vm_change_state_handler(kvm_riscv_vm_state_change, cs);
> +
> id = kvm_riscv_reg_id(env, KVM_REG_RISCV_CONFIG, KVM_REG_RISCV_CONFIG_REG(isa));
> ret = kvm_get_one_reg(cs, id, &isa);
> if (ret) {
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 4:42 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:42 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:38 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Add virtual time context description to vmstate_kvmtimer. After cpu being
> loaded, virtual time context is updated to KVM.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> ---
> target/riscv/machine.c | 37 +++++++++++++++++++++++++++++++++++--
> 1 file changed, 35 insertions(+), 2 deletions(-)
>
> diff --git a/target/riscv/machine.c b/target/riscv/machine.c
> index ad8248ebfd..f46443c316 100644
> --- a/target/riscv/machine.c
> +++ b/target/riscv/machine.c
> @@ -164,10 +164,42 @@ static const VMStateDescription vmstate_pointermasking = {
> }
> };
>
> +static bool kvmtimer_needed(void *opaque)
> +{
> + return kvm_enabled();
> +}
> +
> +
Remove extra newline from here.
> +static const VMStateDescription vmstate_kvmtimer = {
> + .name = "cpu/kvmtimer",
> + .version_id = 1,
> + .minimum_version_id = 1,
> + .needed = kvmtimer_needed,
> + .fields = (VMStateField[]) {
> + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> +
> + VMSTATE_END_OF_LIST()
> + }
> +};
> +
> +static int cpu_post_load(void *opaque, int version_id)
> +{
> + RISCVCPU *cpu = opaque;
> + CPURISCVState *env = &cpu->env;
> +
> + if (kvm_enabled()) {
> + env->kvm_timer_dirty = true;
> + }
> + return 0;
> +}
> +
> const VMStateDescription vmstate_riscv_cpu = {
> .name = "cpu",
> - .version_id = 3,
> - .minimum_version_id = 3,
> + .version_id = 4,
> + .minimum_version_id = 4,
> + .post_load = cpu_post_load,
> .fields = (VMStateField[]) {
> VMSTATE_UINTTL_ARRAY(env.gpr, RISCVCPU, 32),
> VMSTATE_UINT64_ARRAY(env.fpr, RISCVCPU, 32),
> @@ -218,6 +250,7 @@ const VMStateDescription vmstate_riscv_cpu = {
> &vmstate_hyper,
> &vmstate_vector,
> &vmstate_pointermasking,
> + &vmstate_kvmtimer,
> NULL
> }
> };
> --
> 2.19.1
>
>
> --
> kvm-riscv mailing list
> kvm-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/kvm-riscv
Otherwise, it looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
@ 2021-12-13 4:42 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 4:42 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:38 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Add virtual time context description to vmstate_kvmtimer. After cpu being
> loaded, virtual time context is updated to KVM.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> ---
> target/riscv/machine.c | 37 +++++++++++++++++++++++++++++++++++--
> 1 file changed, 35 insertions(+), 2 deletions(-)
>
> diff --git a/target/riscv/machine.c b/target/riscv/machine.c
> index ad8248ebfd..f46443c316 100644
> --- a/target/riscv/machine.c
> +++ b/target/riscv/machine.c
> @@ -164,10 +164,42 @@ static const VMStateDescription vmstate_pointermasking = {
> }
> };
>
> +static bool kvmtimer_needed(void *opaque)
> +{
> + return kvm_enabled();
> +}
> +
> +
Remove extra newline from here.
> +static const VMStateDescription vmstate_kvmtimer = {
> + .name = "cpu/kvmtimer",
> + .version_id = 1,
> + .minimum_version_id = 1,
> + .needed = kvmtimer_needed,
> + .fields = (VMStateField[]) {
> + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> +
> + VMSTATE_END_OF_LIST()
> + }
> +};
> +
> +static int cpu_post_load(void *opaque, int version_id)
> +{
> + RISCVCPU *cpu = opaque;
> + CPURISCVState *env = &cpu->env;
> +
> + if (kvm_enabled()) {
> + env->kvm_timer_dirty = true;
> + }
> + return 0;
> +}
> +
> const VMStateDescription vmstate_riscv_cpu = {
> .name = "cpu",
> - .version_id = 3,
> - .minimum_version_id = 3,
> + .version_id = 4,
> + .minimum_version_id = 4,
> + .post_load = cpu_post_load,
> .fields = (VMStateField[]) {
> VMSTATE_UINTTL_ARRAY(env.gpr, RISCVCPU, 32),
> VMSTATE_UINT64_ARRAY(env.fpr, RISCVCPU, 32),
> @@ -218,6 +250,7 @@ const VMStateDescription vmstate_riscv_cpu = {
> &vmstate_hyper,
> &vmstate_vector,
> &vmstate_pointermasking,
> + &vmstate_kvmtimer,
> NULL
> }
> };
> --
> 2.19.1
>
>
> --
> kvm-riscv mailing list
> kvm-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/kvm-riscv
Otherwise, it looks good to me.
Reviewed-by: Anup Patel <anup.patel@wdc.com>
Regards,
Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 5:05 ` Anup Patel
-1 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 5:05 UTC (permalink / raw)
To: Yifei Jiang
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, fanliang, Wubin (H),
wanghaibin.wang, wanbo13, Mingwang Li
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Add kvm_riscv_get/put_regs_timer to synchronize virtual time context
> from KVM.
>
> To set register of RISCV_TIMER_REG(state) will occur a error from KVM
> on kvm_timer_state == 0. It's better to adapt in KVM, but it doesn't matter
> that adaping in QEMU.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> ---
> target/riscv/cpu.h | 7 +++++
> target/riscv/kvm.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 79 insertions(+)
>
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index e7dba35acb..c892a2c8b7 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -259,6 +259,13 @@ struct CPURISCVState {
>
> hwaddr kernel_addr;
> hwaddr fdt_addr;
> +
> + /* kvm timer */
> + bool kvm_timer_dirty;
> + uint64_t kvm_timer_time;
> + uint64_t kvm_timer_compare;
> + uint64_t kvm_timer_state;
> + uint64_t kvm_timer_frequency;
> };
>
> OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 171a32adf9..802c076b22 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -64,6 +64,9 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> #define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
> KVM_REG_RISCV_CSR_REG(name))
>
> +#define RISCV_TIMER_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_TIMER, \
> + KVM_REG_RISCV_TIMER_REG(name))
> +
> #define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
>
> #define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
> @@ -235,6 +238,75 @@ static int kvm_riscv_put_regs_fp(CPUState *cs)
> return ret;
> }
>
> +static void kvm_riscv_get_regs_timer(CPUState *cs)
> +{
> + int ret;
> + uint64_t reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (env->kvm_timer_dirty) {
> + return;
> + }
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_time = reg;
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_compare = reg;
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_state = reg;
Please read the timer frequency here.
> +
> + env->kvm_timer_dirty = true;
> +}
> +
> +static void kvm_riscv_put_regs_timer(CPUState *cs)
> +{
> + int ret;
> + uint64_t reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (!env->kvm_timer_dirty) {
> + return;
> + }
Over here, we should get the timer frequency and abort() with an
error message if it does not match env->kvm_timer_frequency
For now, migration will not work between Hosts with different
timer frequency.
Regards,
Anup
> +
> + reg = env->kvm_timer_time;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
> + if (ret) {
> + abort();
> + }
> +
> + reg = env->kvm_timer_compare;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
> + if (ret) {
> + abort();
> + }
> +
> + /*
> + * To set register of RISCV_TIMER_REG(state) will occur a error from KVM
> + * on env->kvm_timer_state == 0, It's better to adapt in KVM, but it
> + * doesn't matter that adaping in QEMU now.
> + * TODO If KVM changes, adapt here.
> + */
> + if (env->kvm_timer_state) {
> + reg = env->kvm_timer_state;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> + if (ret) {
> + abort();
> + }
> + }
> +
> + env->kvm_timer_dirty = false;
> +}
>
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
@ 2021-12-13 5:05 ` Anup Patel
0 siblings, 0 replies; 56+ messages in thread
From: Anup Patel @ 2021-12-13 5:05 UTC (permalink / raw)
To: Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
>
> Add kvm_riscv_get/put_regs_timer to synchronize virtual time context
> from KVM.
>
> To set register of RISCV_TIMER_REG(state) will occur a error from KVM
> on kvm_timer_state == 0. It's better to adapt in KVM, but it doesn't matter
> that adaping in QEMU.
>
> Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> Signed-off-by: Mingwang Li <limingwang@huawei.com>
> ---
> target/riscv/cpu.h | 7 +++++
> target/riscv/kvm.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 79 insertions(+)
>
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index e7dba35acb..c892a2c8b7 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -259,6 +259,13 @@ struct CPURISCVState {
>
> hwaddr kernel_addr;
> hwaddr fdt_addr;
> +
> + /* kvm timer */
> + bool kvm_timer_dirty;
> + uint64_t kvm_timer_time;
> + uint64_t kvm_timer_compare;
> + uint64_t kvm_timer_state;
> + uint64_t kvm_timer_frequency;
> };
>
> OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass,
> diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c
> index 171a32adf9..802c076b22 100644
> --- a/target/riscv/kvm.c
> +++ b/target/riscv/kvm.c
> @@ -64,6 +64,9 @@ static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type, uint64_t idx
> #define RISCV_CSR_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_CSR, \
> KVM_REG_RISCV_CSR_REG(name))
>
> +#define RISCV_TIMER_REG(env, name) kvm_riscv_reg_id(env, KVM_REG_RISCV_TIMER, \
> + KVM_REG_RISCV_TIMER_REG(name))
> +
> #define RISCV_FP_F_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_F, idx)
>
> #define RISCV_FP_D_REG(env, idx) kvm_riscv_reg_id(env, KVM_REG_RISCV_FP_D, idx)
> @@ -235,6 +238,75 @@ static int kvm_riscv_put_regs_fp(CPUState *cs)
> return ret;
> }
>
> +static void kvm_riscv_get_regs_timer(CPUState *cs)
> +{
> + int ret;
> + uint64_t reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (env->kvm_timer_dirty) {
> + return;
> + }
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_time = reg;
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_compare = reg;
> +
> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> + if (ret) {
> + abort();
> + }
> + env->kvm_timer_state = reg;
Please read the timer frequency here.
> +
> + env->kvm_timer_dirty = true;
> +}
> +
> +static void kvm_riscv_put_regs_timer(CPUState *cs)
> +{
> + int ret;
> + uint64_t reg;
> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> +
> + if (!env->kvm_timer_dirty) {
> + return;
> + }
Over here, we should get the timer frequency and abort() with an
error message if it does not match env->kvm_timer_frequency
For now, migration will not work between Hosts with different
timer frequency.
Regards,
Anup
> +
> + reg = env->kvm_timer_time;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, time), ®);
> + if (ret) {
> + abort();
> + }
> +
> + reg = env->kvm_timer_compare;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, compare), ®);
> + if (ret) {
> + abort();
> + }
> +
> + /*
> + * To set register of RISCV_TIMER_REG(state) will occur a error from KVM
> + * on env->kvm_timer_state == 0, It's better to adapt in KVM, but it
> + * doesn't matter that adaping in QEMU now.
> + * TODO If KVM changes, adapt here.
> + */
> + if (env->kvm_timer_state) {
> + reg = env->kvm_timer_state;
> + ret = kvm_set_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> + if (ret) {
> + abort();
> + }
> + }
> +
> + env->kvm_timer_dirty = false;
> +}
>
> const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
> KVM_CAP_LAST_INFO
> --
> 2.19.1
>
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
2021-12-13 5:05 ` Anup Patel
@ 2021-12-13 15:19 ` Richard Henderson
-1 siblings, 0 replies; 56+ messages in thread
From: Richard Henderson @ 2021-12-13 15:19 UTC (permalink / raw)
To: Anup Patel, Yifei Jiang
Cc: Bin Meng, open list:RISC-V, Mingwang Li, KVM General,
libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, Alistair Francis,
fanliang, Wubin (H)
On 12/12/21 9:05 PM, Anup Patel wrote:
>> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
>> + if (ret) {
>> + abort();
>> + }
>> + env->kvm_timer_state = reg;
>
> Please read the timer frequency here.
Yep.
>> +
>> + env->kvm_timer_dirty = true;
>> +}
>> +
>> +static void kvm_riscv_put_regs_timer(CPUState *cs)
>> +{
>> + int ret;
>> + uint64_t reg;
>> + CPURISCVState *env = &RISCV_CPU(cs)->env;
>> +
>> + if (!env->kvm_timer_dirty) {
>> + return;
>> + }
>
> Over here, we should get the timer frequency and abort() with an
> error message if it does not match env->kvm_timer_frequency
>
> For now, migration will not work between Hosts with different
> timer frequency.
You shouldn't have to do this every "put", only on migration, at which point you can
actually signal a migration error rather than aborting directly.
r~
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
@ 2021-12-13 15:19 ` Richard Henderson
0 siblings, 0 replies; 56+ messages in thread
From: Richard Henderson @ 2021-12-13 15:19 UTC (permalink / raw)
To: Anup Patel, Yifei Jiang
Cc: Alistair Francis, Bin Meng, open list:RISC-V, Mingwang Li,
KVM General, libvir-list, Anup Patel, QEMU Developers, wanbo13,
Palmer Dabbelt, kvm-riscv, wanghaibin.wang, fanliang, Wubin (H)
On 12/12/21 9:05 PM, Anup Patel wrote:
>> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
>> + if (ret) {
>> + abort();
>> + }
>> + env->kvm_timer_state = reg;
>
> Please read the timer frequency here.
Yep.
>> +
>> + env->kvm_timer_dirty = true;
>> +}
>> +
>> +static void kvm_riscv_put_regs_timer(CPUState *cs)
>> +{
>> + int ret;
>> + uint64_t reg;
>> + CPURISCVState *env = &RISCV_CPU(cs)->env;
>> +
>> + if (!env->kvm_timer_dirty) {
>> + return;
>> + }
>
> Over here, we should get the timer frequency and abort() with an
> error message if it does not match env->kvm_timer_frequency
>
> For now, migration will not work between Hosts with different
> timer frequency.
You shouldn't have to do this every "put", only on migration, at which point you can
actually signal a migration error rather than aborting directly.
r~
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
2021-12-10 10:07 ` Yifei Jiang via
@ 2021-12-13 15:21 ` Richard Henderson
-1 siblings, 0 replies; 56+ messages in thread
From: Richard Henderson @ 2021-12-13 15:21 UTC (permalink / raw)
To: Yifei Jiang, qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, fanliang, wu.wubin, wanghaibin.wang,
wanbo13, Mingwang Li
On 12/10/21 2:07 AM, Yifei Jiang via wrote:
> +static bool kvmtimer_needed(void *opaque)
> +{
> + return kvm_enabled();
> +}
> +
> +
> +static const VMStateDescription vmstate_kvmtimer = {
> + .name = "cpu/kvmtimer",
> + .version_id = 1,
> + .minimum_version_id = 1,
> + .needed = kvmtimer_needed,
> + .fields = (VMStateField[]) {
> + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> +
> + VMSTATE_END_OF_LIST()
> + }
> +};
> +
> +static int cpu_post_load(void *opaque, int version_id)
> +{
> + RISCVCPU *cpu = opaque;
> + CPURISCVState *env = &cpu->env;
> +
> + if (kvm_enabled()) {
> + env->kvm_timer_dirty = true;
> + }
> + return 0;
> +}
This post-load belongs on the vmstate_kvmtimer struct, so that you need not re-check
kvm_enabled().
> const VMStateDescription vmstate_riscv_cpu = {
> .name = "cpu",
> - .version_id = 3,
> - .minimum_version_id = 3,
> + .version_id = 4,
> + .minimum_version_id = 4,
> + .post_load = cpu_post_load,
No need for version change.
r~
^ permalink raw reply [flat|nested] 56+ messages in thread
* Re: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
@ 2021-12-13 15:21 ` Richard Henderson
0 siblings, 0 replies; 56+ messages in thread
From: Richard Henderson @ 2021-12-13 15:21 UTC (permalink / raw)
To: Yifei Jiang, qemu-devel, qemu-riscv
Cc: bin.meng, Mingwang Li, kvm, libvir-list, anup.patel, wanbo13,
Alistair.Francis, kvm-riscv, wanghaibin.wang, palmer, fanliang,
wu.wubin
On 12/10/21 2:07 AM, Yifei Jiang via wrote:
> +static bool kvmtimer_needed(void *opaque)
> +{
> + return kvm_enabled();
> +}
> +
> +
> +static const VMStateDescription vmstate_kvmtimer = {
> + .name = "cpu/kvmtimer",
> + .version_id = 1,
> + .minimum_version_id = 1,
> + .needed = kvmtimer_needed,
> + .fields = (VMStateField[]) {
> + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> +
> + VMSTATE_END_OF_LIST()
> + }
> +};
> +
> +static int cpu_post_load(void *opaque, int version_id)
> +{
> + RISCVCPU *cpu = opaque;
> + CPURISCVState *env = &cpu->env;
> +
> + if (kvm_enabled()) {
> + env->kvm_timer_dirty = true;
> + }
> + return 0;
> +}
This post-load belongs on the vmstate_kvmtimer struct, so that you need not re-check
kvm_enabled().
> const VMStateDescription vmstate_riscv_cpu = {
> .name = "cpu",
> - .version_id = 3,
> - .minimum_version_id = 3,
> + .version_id = 4,
> + .minimum_version_id = 4,
> + .post_load = cpu_post_load,
No need for version change.
r~
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
2021-12-13 4:21 ` Anup Patel
@ 2021-12-20 12:59 ` Jiangyifei via
-1 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei @ 2021-12-20 12:59 UTC (permalink / raw)
To: Anup Patel
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, Fanliang (EulerOS), Wubin (H), Wanghaibin (D),
wanbo (G), limingwang (A)
> -----Original Message-----
> From: Anup Patel [mailto:anup@brainfault.org]
> Sent: Monday, December 13, 2021 12:21 PM
> To: Jiangyifei <jiangyifei@huawei.com>
> Cc: QEMU Developers <qemu-devel@nongnu.org>; open list:RISC-V
> <qemu-riscv@nongnu.org>; kvm-riscv@lists.infradead.org; KVM General
> <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; Palmer Dabbelt <palmer@dabbelt.com>; Alistair
> Francis <Alistair.Francis@wdc.com>; Bin Meng <bin.meng@windriver.com>;
> Fanliang (EulerOS) <fanliang@huawei.com>; Wubin (H)
> <wu.wubin@huawei.com>; Wanghaibin (D) <wanghaibin.wang@huawei.com>;
> wanbo (G) <wanbo13@huawei.com>; limingwang (A)
> <limingwang@huawei.com>
> Subject: Re: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
>
> On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
> >
> > Get kernel and fdt start address in virt.c, and pass them to KVM when
> > cpu reset. In addition, add kvm_riscv.h to place riscv specific
> > interface.
> >
> > Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> > Signed-off-by: Mingwang Li <limingwang@huawei.com>
> > Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> > ---
> > hw/riscv/boot.c | 11 ++++++++
> > hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
> > include/hw/riscv/boot.h | 1 +
> > target/riscv/cpu.c | 8 ++++++
> > target/riscv/cpu.h | 3 +++
> > target/riscv/kvm-stub.c | 25 +++++++++++++++++++
> > target/riscv/kvm.c | 14 +++++++++++
> > target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
> > target/riscv/meson.build | 2 +-
> > 9 files changed, 125 insertions(+), 17 deletions(-) create mode
> > 100644 target/riscv/kvm-stub.c create mode 100644
> > target/riscv/kvm_riscv.h
> >
> > diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c index
> > 519fa455a1..00df6d7810 100644
> > --- a/hw/riscv/boot.c
> > +++ b/hw/riscv/boot.c
> > @@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState
> > *machine, RISCVHartArrayState *harts
> >
> > return;
> > }
> > +
> > +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr) {
> > + CPUState *cs;
> > +
> > + for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
> > + RISCVCPU *riscv_cpu = RISCV_CPU(cs);
> > + riscv_cpu->env.kernel_addr = kernel_addr;
> > + riscv_cpu->env.fdt_addr = fdt_addr;
> > + }
> > +}
> > diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c index
> > 3af074148e..c8bcd9d9e5 100644
> > --- a/hw/riscv/virt.c
> > +++ b/hw/riscv/virt.c
> > @@ -38,6 +38,7 @@
> > #include "chardev/char.h"
> > #include "sysemu/device_tree.h"
> > #include "sysemu/sysemu.h"
> > +#include "sysemu/kvm.h"
> > #include "hw/pci/pci.h"
> > #include "hw/pci-host/gpex.h"
> > #include "hw/display/ramfb.h"
> > @@ -801,23 +802,25 @@ static void virt_machine_init(MachineState
> *machine)
> > hart_count, &error_abort);
> > sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
> >
> > - /* Per-socket CLINT */
> > - riscv_aclint_swi_create(
> > - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> > - base_hartid, hart_count, false);
> > - riscv_aclint_mtimer_create(
> > - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size
> +
> > - RISCV_ACLINT_SWI_SIZE,
> > - RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid,
> hart_count,
> > - RISCV_ACLINT_DEFAULT_MTIMECMP,
> RISCV_ACLINT_DEFAULT_MTIME,
> > - RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> > -
> > - /* Per-socket ACLINT SSWI */
> > - if (s->have_aclint) {
> > + if (!kvm_enabled()) {
> > + /* Per-socket CLINT */
> > riscv_aclint_swi_create(
> > - memmap[VIRT_ACLINT_SSWI].base +
> > - i * memmap[VIRT_ACLINT_SSWI].size,
> > - base_hartid, hart_count, true);
> > + memmap[VIRT_CLINT].base + i *
> memmap[VIRT_CLINT].size,
> > + base_hartid, hart_count, false);
> > + riscv_aclint_mtimer_create(
> > + memmap[VIRT_CLINT].base + i *
> memmap[VIRT_CLINT].size +
> > + RISCV_ACLINT_SWI_SIZE,
> > + RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid,
> hart_count,
> > + RISCV_ACLINT_DEFAULT_MTIMECMP,
> RISCV_ACLINT_DEFAULT_MTIME,
> > + RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> > +
> > + /* Per-socket ACLINT SSWI */
> > + if (s->have_aclint) {
> > + riscv_aclint_swi_create(
> > + memmap[VIRT_ACLINT_SSWI].base +
> > + i * memmap[VIRT_ACLINT_SSWI].size,
> > + base_hartid, hart_count, true);
> > + }
>
> Along with this, we should not generate FDT nodes of ACLINT (or SiFive CLINT)
> when KVM is enabled.
>
OK, and I will remove M-mode PLIC contexts when kvm is enabled in next series.
Yifei
> > }
> >
> > /* Per-socket PLIC hart topology configuration string */ @@
> > -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
> > memory_region_add_subregion(system_memory,
> memmap[VIRT_MROM].base,
> > mask_rom);
> >
> > + /*
> > + * Only direct boot kernel is currently supported for KVM VM,
> > + * so the "-bios" parameter is ignored and treated like "-bios none"
> > + * when KVM is enabled.
> > + */
> > + if (kvm_enabled()) {
> > + g_free(machine->firmware);
> > + machine->firmware = g_strdup("none");
> > + }
> > +
> > if (riscv_is_32bit(&s->soc[0])) {
> > firmware_end_addr = riscv_find_and_load_firmware(machine,
> > RISCV32_BIOS_BIN,
> start_addr,
> > NULL); @@ -941,6 +954,15 @@ static void virt_machine_init(MachineState
> *machine)
> > virt_memmap[VIRT_MROM].size,
> kernel_entry,
> > fdt_load_addr, machine->fdt);
> >
> > + /*
> > + * Only direct boot kernel is currently supported for KVM VM,
> > + * So here setup kernel start address and fdt address.
> > + * TODO:Support firmware loading and integrate to TCG start
> > + */
> > + if (kvm_enabled()) {
> > + riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
> > + }
> > +
> > /* SiFive Test MMIO device */
> > sifive_test_create(memmap[VIRT_TEST].base);
> >
> > diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h index
> > baff11dd8a..5834c234aa 100644
> > --- a/include/hw/riscv/boot.h
> > +++ b/include/hw/riscv/boot.h
> > @@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState
> *machine, hwaddr rom_base,
> > hwaddr rom_size,
> > uint32_t reset_vec_size,
> > uint64_t kernel_entry);
> > +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
> >
> > #endif /* RISCV_BOOT_H */
> > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index
> > f812998123..1c944872a3 100644
> > --- a/target/riscv/cpu.c
> > +++ b/target/riscv/cpu.c
> > @@ -29,6 +29,8 @@
> > #include "hw/qdev-properties.h"
> > #include "migration/vmstate.h"
> > #include "fpu/softfloat-helpers.h"
> > +#include "sysemu/kvm.h"
> > +#include "kvm_riscv.h"
> >
> > /* RISC-V CPU definitions */
> >
> > @@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
> > cs->exception_index = RISCV_EXCP_NONE;
> > env->load_res = -1;
> > set_default_nan_mode(1, &env->fp_status);
> > +
> > +#ifndef CONFIG_USER_ONLY
> > + if (kvm_enabled()) {
> > + kvm_riscv_reset_vcpu(cpu);
> > + }
> > +#endif
> > }
> >
> > static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info
> > *info) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index
> > 0760c0af93..2807eb1bcb 100644
> > --- a/target/riscv/cpu.h
> > +++ b/target/riscv/cpu.h
> > @@ -255,6 +255,9 @@ struct CPURISCVState {
> >
> > /* Fields from here on are preserved across CPU reset. */
> > QEMUTimer *timer; /* Internal timer */
> > +
> > + hwaddr kernel_addr;
> > + hwaddr fdt_addr;
> > };
> >
> > OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass, diff --git
> > a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c new file mode
> > 100644 index 0000000000..39b96fe3f4
> > --- /dev/null
> > +++ b/target/riscv/kvm-stub.c
> > @@ -0,0 +1,25 @@
> > +/*
> > + * QEMU KVM RISC-V specific function stubs
> > + *
> > + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> > + *
> > + * This program is free software; you can redistribute it and/or
> > +modify it
> > + * under the terms and conditions of the GNU General Public License,
> > + * version 2 or later, as published by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope it will be useful, but
> > +WITHOUT
> > + * ANY WARRANTY; without even the implied warranty of
> MERCHANTABILITY
> > +or
> > + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> > +License for
> > + * more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > +along with
> > + * this program. If not, see <http://www.gnu.org/licenses/>.
> > + */
> > +#include "qemu/osdep.h"
> > +#include "cpu.h"
> > +#include "kvm_riscv.h"
> > +
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > + abort();
> > +}
> > diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c index
> > e695b91dc7..db6d8a5b6e 100644
> > --- a/target/riscv/kvm.c
> > +++ b/target/riscv/kvm.c
> > @@ -37,6 +37,7 @@
> > #include "hw/irq.h"
> > #include "qemu/log.h"
> > #include "hw/loader.h"
> > +#include "kvm_riscv.h"
> >
> > static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
> > uint64_t idx) { @@ -369,6 +370,19 @@ int
> > kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> > return 0;
> > }
> >
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > + CPURISCVState *env = &cpu->env;
> > +
> > + if (!kvm_enabled()) {
> > + return;
> > + }
> > + env->pc = cpu->env.kernel_addr;
> > + env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
> > + env->gpr[11] = cpu->env.fdt_addr; /* a1 */
> > + env->satp = 0;
> > +}
> > +
> > bool kvm_arch_cpu_check_are_resettable(void)
> > {
> > return true;
> > diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h new
> > file mode 100644 index 0000000000..f38c82bf59
> > --- /dev/null
> > +++ b/target/riscv/kvm_riscv.h
> > @@ -0,0 +1,24 @@
> > +/*
> > + * QEMU KVM support -- RISC-V specific functions.
> > + *
> > + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> > + *
> > + * This program is free software; you can redistribute it and/or
> > +modify it
> > + * under the terms and conditions of the GNU General Public License,
> > + * version 2 or later, as published by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope it will be useful, but
> > +WITHOUT
> > + * ANY WARRANTY; without even the implied warranty of
> MERCHANTABILITY
> > +or
> > + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> > +License for
> > + * more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > +along with
> > + * this program. If not, see <http://www.gnu.org/licenses/>.
> > + */
> > +
> > +#ifndef QEMU_KVM_RISCV_H
> > +#define QEMU_KVM_RISCV_H
> > +
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> > +
> > +#endif
> > diff --git a/target/riscv/meson.build b/target/riscv/meson.build index
> > 2faf08a941..fe41cc5805 100644
> > --- a/target/riscv/meson.build
> > +++ b/target/riscv/meson.build
> > @@ -19,7 +19,7 @@ riscv_ss.add(files(
> > 'bitmanip_helper.c',
> > 'translate.c',
> > ))
> > -riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
> > +riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false:
> > +files('kvm-stub.c'))
> >
> > riscv_softmmu_ss = ss.source_set()
> > riscv_softmmu_ss.add(files(
> > --
> > 2.19.1
> >
>
> Regards,
> Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
@ 2021-12-20 12:59 ` Jiangyifei via
0 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei via @ 2021-12-20 12:59 UTC (permalink / raw)
To: Anup Patel
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, Fanliang (EulerOS), Wubin (H), Wanghaibin (D),
wanbo (G), limingwang (A)
> -----Original Message-----
> From: Anup Patel [mailto:anup@brainfault.org]
> Sent: Monday, December 13, 2021 12:21 PM
> To: Jiangyifei <jiangyifei@huawei.com>
> Cc: QEMU Developers <qemu-devel@nongnu.org>; open list:RISC-V
> <qemu-riscv@nongnu.org>; kvm-riscv@lists.infradead.org; KVM General
> <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; Palmer Dabbelt <palmer@dabbelt.com>; Alistair
> Francis <Alistair.Francis@wdc.com>; Bin Meng <bin.meng@windriver.com>;
> Fanliang (EulerOS) <fanliang@huawei.com>; Wubin (H)
> <wu.wubin@huawei.com>; Wanghaibin (D) <wanghaibin.wang@huawei.com>;
> wanbo (G) <wanbo13@huawei.com>; limingwang (A)
> <limingwang@huawei.com>
> Subject: Re: [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM
>
> On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
> >
> > Get kernel and fdt start address in virt.c, and pass them to KVM when
> > cpu reset. In addition, add kvm_riscv.h to place riscv specific
> > interface.
> >
> > Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> > Signed-off-by: Mingwang Li <limingwang@huawei.com>
> > Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> > ---
> > hw/riscv/boot.c | 11 ++++++++
> > hw/riscv/virt.c | 54 ++++++++++++++++++++++++++++------------
> > include/hw/riscv/boot.h | 1 +
> > target/riscv/cpu.c | 8 ++++++
> > target/riscv/cpu.h | 3 +++
> > target/riscv/kvm-stub.c | 25 +++++++++++++++++++
> > target/riscv/kvm.c | 14 +++++++++++
> > target/riscv/kvm_riscv.h | 24 ++++++++++++++++++
> > target/riscv/meson.build | 2 +-
> > 9 files changed, 125 insertions(+), 17 deletions(-) create mode
> > 100644 target/riscv/kvm-stub.c create mode 100644
> > target/riscv/kvm_riscv.h
> >
> > diff --git a/hw/riscv/boot.c b/hw/riscv/boot.c index
> > 519fa455a1..00df6d7810 100644
> > --- a/hw/riscv/boot.c
> > +++ b/hw/riscv/boot.c
> > @@ -317,3 +317,14 @@ void riscv_setup_rom_reset_vec(MachineState
> > *machine, RISCVHartArrayState *harts
> >
> > return;
> > }
> > +
> > +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr) {
> > + CPUState *cs;
> > +
> > + for (cs = first_cpu; cs; cs = CPU_NEXT(cs)) {
> > + RISCVCPU *riscv_cpu = RISCV_CPU(cs);
> > + riscv_cpu->env.kernel_addr = kernel_addr;
> > + riscv_cpu->env.fdt_addr = fdt_addr;
> > + }
> > +}
> > diff --git a/hw/riscv/virt.c b/hw/riscv/virt.c index
> > 3af074148e..c8bcd9d9e5 100644
> > --- a/hw/riscv/virt.c
> > +++ b/hw/riscv/virt.c
> > @@ -38,6 +38,7 @@
> > #include "chardev/char.h"
> > #include "sysemu/device_tree.h"
> > #include "sysemu/sysemu.h"
> > +#include "sysemu/kvm.h"
> > #include "hw/pci/pci.h"
> > #include "hw/pci-host/gpex.h"
> > #include "hw/display/ramfb.h"
> > @@ -801,23 +802,25 @@ static void virt_machine_init(MachineState
> *machine)
> > hart_count, &error_abort);
> > sysbus_realize(SYS_BUS_DEVICE(&s->soc[i]), &error_abort);
> >
> > - /* Per-socket CLINT */
> > - riscv_aclint_swi_create(
> > - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size,
> > - base_hartid, hart_count, false);
> > - riscv_aclint_mtimer_create(
> > - memmap[VIRT_CLINT].base + i * memmap[VIRT_CLINT].size
> +
> > - RISCV_ACLINT_SWI_SIZE,
> > - RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid,
> hart_count,
> > - RISCV_ACLINT_DEFAULT_MTIMECMP,
> RISCV_ACLINT_DEFAULT_MTIME,
> > - RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> > -
> > - /* Per-socket ACLINT SSWI */
> > - if (s->have_aclint) {
> > + if (!kvm_enabled()) {
> > + /* Per-socket CLINT */
> > riscv_aclint_swi_create(
> > - memmap[VIRT_ACLINT_SSWI].base +
> > - i * memmap[VIRT_ACLINT_SSWI].size,
> > - base_hartid, hart_count, true);
> > + memmap[VIRT_CLINT].base + i *
> memmap[VIRT_CLINT].size,
> > + base_hartid, hart_count, false);
> > + riscv_aclint_mtimer_create(
> > + memmap[VIRT_CLINT].base + i *
> memmap[VIRT_CLINT].size +
> > + RISCV_ACLINT_SWI_SIZE,
> > + RISCV_ACLINT_DEFAULT_MTIMER_SIZE, base_hartid,
> hart_count,
> > + RISCV_ACLINT_DEFAULT_MTIMECMP,
> RISCV_ACLINT_DEFAULT_MTIME,
> > + RISCV_ACLINT_DEFAULT_TIMEBASE_FREQ, true);
> > +
> > + /* Per-socket ACLINT SSWI */
> > + if (s->have_aclint) {
> > + riscv_aclint_swi_create(
> > + memmap[VIRT_ACLINT_SSWI].base +
> > + i * memmap[VIRT_ACLINT_SSWI].size,
> > + base_hartid, hart_count, true);
> > + }
>
> Along with this, we should not generate FDT nodes of ACLINT (or SiFive CLINT)
> when KVM is enabled.
>
OK, and I will remove M-mode PLIC contexts when kvm is enabled in next series.
Yifei
> > }
> >
> > /* Per-socket PLIC hart topology configuration string */ @@
> > -884,6 +887,16 @@ static void virt_machine_init(MachineState *machine)
> > memory_region_add_subregion(system_memory,
> memmap[VIRT_MROM].base,
> > mask_rom);
> >
> > + /*
> > + * Only direct boot kernel is currently supported for KVM VM,
> > + * so the "-bios" parameter is ignored and treated like "-bios none"
> > + * when KVM is enabled.
> > + */
> > + if (kvm_enabled()) {
> > + g_free(machine->firmware);
> > + machine->firmware = g_strdup("none");
> > + }
> > +
> > if (riscv_is_32bit(&s->soc[0])) {
> > firmware_end_addr = riscv_find_and_load_firmware(machine,
> > RISCV32_BIOS_BIN,
> start_addr,
> > NULL); @@ -941,6 +954,15 @@ static void virt_machine_init(MachineState
> *machine)
> > virt_memmap[VIRT_MROM].size,
> kernel_entry,
> > fdt_load_addr, machine->fdt);
> >
> > + /*
> > + * Only direct boot kernel is currently supported for KVM VM,
> > + * So here setup kernel start address and fdt address.
> > + * TODO:Support firmware loading and integrate to TCG start
> > + */
> > + if (kvm_enabled()) {
> > + riscv_setup_direct_kernel(kernel_entry, fdt_load_addr);
> > + }
> > +
> > /* SiFive Test MMIO device */
> > sifive_test_create(memmap[VIRT_TEST].base);
> >
> > diff --git a/include/hw/riscv/boot.h b/include/hw/riscv/boot.h index
> > baff11dd8a..5834c234aa 100644
> > --- a/include/hw/riscv/boot.h
> > +++ b/include/hw/riscv/boot.h
> > @@ -58,5 +58,6 @@ void riscv_rom_copy_firmware_info(MachineState
> *machine, hwaddr rom_base,
> > hwaddr rom_size,
> > uint32_t reset_vec_size,
> > uint64_t kernel_entry);
> > +void riscv_setup_direct_kernel(hwaddr kernel_addr, hwaddr fdt_addr);
> >
> > #endif /* RISCV_BOOT_H */
> > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index
> > f812998123..1c944872a3 100644
> > --- a/target/riscv/cpu.c
> > +++ b/target/riscv/cpu.c
> > @@ -29,6 +29,8 @@
> > #include "hw/qdev-properties.h"
> > #include "migration/vmstate.h"
> > #include "fpu/softfloat-helpers.h"
> > +#include "sysemu/kvm.h"
> > +#include "kvm_riscv.h"
> >
> > /* RISC-V CPU definitions */
> >
> > @@ -380,6 +382,12 @@ static void riscv_cpu_reset(DeviceState *dev)
> > cs->exception_index = RISCV_EXCP_NONE;
> > env->load_res = -1;
> > set_default_nan_mode(1, &env->fp_status);
> > +
> > +#ifndef CONFIG_USER_ONLY
> > + if (kvm_enabled()) {
> > + kvm_riscv_reset_vcpu(cpu);
> > + }
> > +#endif
> > }
> >
> > static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info
> > *info) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index
> > 0760c0af93..2807eb1bcb 100644
> > --- a/target/riscv/cpu.h
> > +++ b/target/riscv/cpu.h
> > @@ -255,6 +255,9 @@ struct CPURISCVState {
> >
> > /* Fields from here on are preserved across CPU reset. */
> > QEMUTimer *timer; /* Internal timer */
> > +
> > + hwaddr kernel_addr;
> > + hwaddr fdt_addr;
> > };
> >
> > OBJECT_DECLARE_TYPE(RISCVCPU, RISCVCPUClass, diff --git
> > a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c new file mode
> > 100644 index 0000000000..39b96fe3f4
> > --- /dev/null
> > +++ b/target/riscv/kvm-stub.c
> > @@ -0,0 +1,25 @@
> > +/*
> > + * QEMU KVM RISC-V specific function stubs
> > + *
> > + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> > + *
> > + * This program is free software; you can redistribute it and/or
> > +modify it
> > + * under the terms and conditions of the GNU General Public License,
> > + * version 2 or later, as published by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope it will be useful, but
> > +WITHOUT
> > + * ANY WARRANTY; without even the implied warranty of
> MERCHANTABILITY
> > +or
> > + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> > +License for
> > + * more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > +along with
> > + * this program. If not, see <http://www.gnu.org/licenses/>.
> > + */
> > +#include "qemu/osdep.h"
> > +#include "cpu.h"
> > +#include "kvm_riscv.h"
> > +
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > + abort();
> > +}
> > diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c index
> > e695b91dc7..db6d8a5b6e 100644
> > --- a/target/riscv/kvm.c
> > +++ b/target/riscv/kvm.c
> > @@ -37,6 +37,7 @@
> > #include "hw/irq.h"
> > #include "qemu/log.h"
> > #include "hw/loader.h"
> > +#include "kvm_riscv.h"
> >
> > static uint64_t kvm_riscv_reg_id(CPURISCVState *env, uint64_t type,
> > uint64_t idx) { @@ -369,6 +370,19 @@ int
> > kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
> > return 0;
> > }
> >
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > + CPURISCVState *env = &cpu->env;
> > +
> > + if (!kvm_enabled()) {
> > + return;
> > + }
> > + env->pc = cpu->env.kernel_addr;
> > + env->gpr[10] = kvm_arch_vcpu_id(CPU(cpu)); /* a0 */
> > + env->gpr[11] = cpu->env.fdt_addr; /* a1 */
> > + env->satp = 0;
> > +}
> > +
> > bool kvm_arch_cpu_check_are_resettable(void)
> > {
> > return true;
> > diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h new
> > file mode 100644 index 0000000000..f38c82bf59
> > --- /dev/null
> > +++ b/target/riscv/kvm_riscv.h
> > @@ -0,0 +1,24 @@
> > +/*
> > + * QEMU KVM support -- RISC-V specific functions.
> > + *
> > + * Copyright (c) 2020 Huawei Technologies Co., Ltd
> > + *
> > + * This program is free software; you can redistribute it and/or
> > +modify it
> > + * under the terms and conditions of the GNU General Public License,
> > + * version 2 or later, as published by the Free Software Foundation.
> > + *
> > + * This program is distributed in the hope it will be useful, but
> > +WITHOUT
> > + * ANY WARRANTY; without even the implied warranty of
> MERCHANTABILITY
> > +or
> > + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
> > +License for
> > + * more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> > +along with
> > + * this program. If not, see <http://www.gnu.org/licenses/>.
> > + */
> > +
> > +#ifndef QEMU_KVM_RISCV_H
> > +#define QEMU_KVM_RISCV_H
> > +
> > +void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> > +
> > +#endif
> > diff --git a/target/riscv/meson.build b/target/riscv/meson.build index
> > 2faf08a941..fe41cc5805 100644
> > --- a/target/riscv/meson.build
> > +++ b/target/riscv/meson.build
> > @@ -19,7 +19,7 @@ riscv_ss.add(files(
> > 'bitmanip_helper.c',
> > 'translate.c',
> > ))
> > -riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'))
> > +riscv_ss.add(when: 'CONFIG_KVM', if_true: files('kvm.c'), if_false:
> > +files('kvm-stub.c'))
> >
> > riscv_softmmu_ss = ss.source_set()
> > riscv_softmmu_ss.add(files(
> > --
> > 2.19.1
> >
>
> Regards,
> Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
2021-12-13 4:32 ` Anup Patel
@ 2021-12-20 13:01 ` Jiangyifei via
-1 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei @ 2021-12-20 13:01 UTC (permalink / raw)
To: Anup Patel
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, Fanliang (EulerOS), Wubin (H), Wanghaibin (D),
wanbo (G), limingwang (A)
> -----Original Message-----
> From: Anup Patel [mailto:anup@brainfault.org]
> Sent: Monday, December 13, 2021 12:33 PM
> To: Jiangyifei <jiangyifei@huawei.com>
> Cc: QEMU Developers <qemu-devel@nongnu.org>; open list:RISC-V
> <qemu-riscv@nongnu.org>; kvm-riscv@lists.infradead.org; KVM General
> <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; Palmer Dabbelt <palmer@dabbelt.com>; Alistair
> Francis <Alistair.Francis@wdc.com>; Bin Meng <bin.meng@windriver.com>;
> Fanliang (EulerOS) <fanliang@huawei.com>; Wubin (H)
> <wu.wubin@huawei.com>; Wanghaibin (D) <wanghaibin.wang@huawei.com>;
> wanbo (G) <wanbo13@huawei.com>; limingwang (A)
> <limingwang@huawei.com>
> Subject: Re: [PATCH v2 07/12] target/riscv: Support setting external interrupt
> by KVM
>
> On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
> >
> > When KVM is enabled, set the S-mode external interrupt through
> > kvm_riscv_set_irq function.
> >
> > Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> > Signed-off-by: Mingwang Li <limingwang@huawei.com>
> > Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> > ---
> > target/riscv/cpu.c | 6 +++++-
> > target/riscv/kvm-stub.c | 5 +++++
> > target/riscv/kvm.c | 17 +++++++++++++++++
> > target/riscv/kvm_riscv.h | 1 +
> > 4 files changed, 28 insertions(+), 1 deletion(-)
> >
> > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index
> > 1c944872a3..71a7ac6831 100644
> > --- a/target/riscv/cpu.c
> > +++ b/target/riscv/cpu.c
> > @@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq,
> int level)
> > case IRQ_S_EXT:
> > case IRQ_VS_EXT:
> > case IRQ_M_EXT:
> > - riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> > + if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
> > + kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
> > + } else {
> > + riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> > + }
>
> This does not look right.
>
> I suggest the following:
>
> if (kvm_enabled()) {
> kvm_riscv_set_irq(cpu, irq, level);
> } else {
> riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level)); }
>
The M-mode PLIC context is not deleted. Therefore, IRQ_M_EXT needs to
be converted to IRQ_S_EXT. The M-mode PLIC contexts will be removed in
the next series, so it will be modified as suggested.
Yifei
> > break;
> > default:
> > g_assert_not_reached();
> > diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c index
> > 39b96fe3f4..4e8fc31a21 100644
> > --- a/target/riscv/kvm-stub.c
> > +++ b/target/riscv/kvm-stub.c
> > @@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > abort();
> > }
> > +
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level) {
> > + abort();
> > +}
> > diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c index
> > db6d8a5b6e..0027f11f45 100644
> > --- a/target/riscv/kvm.c
> > +++ b/target/riscv/kvm.c
> > @@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> > env->satp = 0;
> > }
> >
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level) {
> > + int ret;
> > + unsigned virq = level ? KVM_INTERRUPT_SET :
> KVM_INTERRUPT_UNSET;
> > +
> > + if (irq != IRQ_S_EXT) {
> > + perror("kvm riscv set irq != IRQ_S_EXT\n");
> > + abort();
> > + }
> > +
> > + ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
> > + if (ret < 0) {
> > + perror("Set irq failed");
> > + abort();
> > + }
> > +}
> > +
> > bool kvm_arch_cpu_check_are_resettable(void)
> > {
> > return true;
> > diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h index
> > f38c82bf59..ed281bdce0 100644
> > --- a/target/riscv/kvm_riscv.h
> > +++ b/target/riscv/kvm_riscv.h
> > @@ -20,5 +20,6 @@
> > #define QEMU_KVM_RISCV_H
> >
> > void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
> >
> > #endif
> > --
> > 2.19.1
> >
>
> Regards,
> Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 07/12] target/riscv: Support setting external interrupt by KVM
@ 2021-12-20 13:01 ` Jiangyifei via
0 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei via @ 2021-12-20 13:01 UTC (permalink / raw)
To: Anup Patel
Cc: QEMU Developers, open list:RISC-V, kvm-riscv, KVM General,
libvir-list, Anup Patel, Palmer Dabbelt, Alistair Francis,
Bin Meng, Fanliang (EulerOS), Wubin (H), Wanghaibin (D),
wanbo (G), limingwang (A)
> -----Original Message-----
> From: Anup Patel [mailto:anup@brainfault.org]
> Sent: Monday, December 13, 2021 12:33 PM
> To: Jiangyifei <jiangyifei@huawei.com>
> Cc: QEMU Developers <qemu-devel@nongnu.org>; open list:RISC-V
> <qemu-riscv@nongnu.org>; kvm-riscv@lists.infradead.org; KVM General
> <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; Palmer Dabbelt <palmer@dabbelt.com>; Alistair
> Francis <Alistair.Francis@wdc.com>; Bin Meng <bin.meng@windriver.com>;
> Fanliang (EulerOS) <fanliang@huawei.com>; Wubin (H)
> <wu.wubin@huawei.com>; Wanghaibin (D) <wanghaibin.wang@huawei.com>;
> wanbo (G) <wanbo13@huawei.com>; limingwang (A)
> <limingwang@huawei.com>
> Subject: Re: [PATCH v2 07/12] target/riscv: Support setting external interrupt
> by KVM
>
> On Fri, Dec 10, 2021 at 3:37 PM Yifei Jiang <jiangyifei@huawei.com> wrote:
> >
> > When KVM is enabled, set the S-mode external interrupt through
> > kvm_riscv_set_irq function.
> >
> > Signed-off-by: Yifei Jiang <jiangyifei@huawei.com>
> > Signed-off-by: Mingwang Li <limingwang@huawei.com>
> > Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
> > ---
> > target/riscv/cpu.c | 6 +++++-
> > target/riscv/kvm-stub.c | 5 +++++
> > target/riscv/kvm.c | 17 +++++++++++++++++
> > target/riscv/kvm_riscv.h | 1 +
> > 4 files changed, 28 insertions(+), 1 deletion(-)
> >
> > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index
> > 1c944872a3..71a7ac6831 100644
> > --- a/target/riscv/cpu.c
> > +++ b/target/riscv/cpu.c
> > @@ -603,7 +603,11 @@ static void riscv_cpu_set_irq(void *opaque, int irq,
> int level)
> > case IRQ_S_EXT:
> > case IRQ_VS_EXT:
> > case IRQ_M_EXT:
> > - riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> > + if (kvm_enabled() && (irq & IRQ_M_EXT) ) {
> > + kvm_riscv_set_irq(cpu, IRQ_S_EXT, level);
> > + } else {
> > + riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level));
> > + }
>
> This does not look right.
>
> I suggest the following:
>
> if (kvm_enabled()) {
> kvm_riscv_set_irq(cpu, irq, level);
> } else {
> riscv_cpu_update_mip(cpu, 1 << irq, BOOL_TO_MASK(level)); }
>
The M-mode PLIC context is not deleted. Therefore, IRQ_M_EXT needs to
be converted to IRQ_S_EXT. The M-mode PLIC contexts will be removed in
the next series, so it will be modified as suggested.
Yifei
> > break;
> > default:
> > g_assert_not_reached();
> > diff --git a/target/riscv/kvm-stub.c b/target/riscv/kvm-stub.c index
> > 39b96fe3f4..4e8fc31a21 100644
> > --- a/target/riscv/kvm-stub.c
> > +++ b/target/riscv/kvm-stub.c
> > @@ -23,3 +23,8 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu) {
> > abort();
> > }
> > +
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level) {
> > + abort();
> > +}
> > diff --git a/target/riscv/kvm.c b/target/riscv/kvm.c index
> > db6d8a5b6e..0027f11f45 100644
> > --- a/target/riscv/kvm.c
> > +++ b/target/riscv/kvm.c
> > @@ -383,6 +383,23 @@ void kvm_riscv_reset_vcpu(RISCVCPU *cpu)
> > env->satp = 0;
> > }
> >
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level) {
> > + int ret;
> > + unsigned virq = level ? KVM_INTERRUPT_SET :
> KVM_INTERRUPT_UNSET;
> > +
> > + if (irq != IRQ_S_EXT) {
> > + perror("kvm riscv set irq != IRQ_S_EXT\n");
> > + abort();
> > + }
> > +
> > + ret = kvm_vcpu_ioctl(CPU(cpu), KVM_INTERRUPT, &virq);
> > + if (ret < 0) {
> > + perror("Set irq failed");
> > + abort();
> > + }
> > +}
> > +
> > bool kvm_arch_cpu_check_are_resettable(void)
> > {
> > return true;
> > diff --git a/target/riscv/kvm_riscv.h b/target/riscv/kvm_riscv.h index
> > f38c82bf59..ed281bdce0 100644
> > --- a/target/riscv/kvm_riscv.h
> > +++ b/target/riscv/kvm_riscv.h
> > @@ -20,5 +20,6 @@
> > #define QEMU_KVM_RISCV_H
> >
> > void kvm_riscv_reset_vcpu(RISCVCPU *cpu);
> > +void kvm_riscv_set_irq(RISCVCPU *cpu, int irq, int level);
> >
> > #endif
> > --
> > 2.19.1
> >
>
> Regards,
> Anup
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
2021-12-13 15:19 ` Richard Henderson
@ 2021-12-20 13:02 ` Jiangyifei via
-1 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei @ 2021-12-20 13:02 UTC (permalink / raw)
To: Richard Henderson, Anup Patel
Cc: Bin Meng, open list:RISC-V, limingwang (A),
KVM General, libvir-list, Anup Patel, QEMU Developers, wanbo (G),
Palmer Dabbelt, kvm-riscv, Wanghaibin (D),
Alistair Francis, Fanliang (EulerOS), Wubin (H)
> -----Original Message-----
> From: Richard Henderson [mailto:richard.henderson@linaro.org]
> Sent: Monday, December 13, 2021 11:20 PM
> To: Anup Patel <anup@brainfault.org>; Jiangyifei <jiangyifei@huawei.com>
> Cc: Bin Meng <bin.meng@windriver.com>; open list:RISC-V
> <qemu-riscv@nongnu.org>; limingwang (A) <limingwang@huawei.com>; KVM
> General <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; QEMU Developers <qemu-devel@nongnu.org>;
> wanbo (G) <wanbo13@huawei.com>; Palmer Dabbelt <palmer@dabbelt.com>;
> kvm-riscv@lists.infradead.org; Wanghaibin (D)
> <wanghaibin.wang@huawei.com>; Alistair Francis
> <Alistair.Francis@wdc.com>; Fanliang (EulerOS) <fanliang@huawei.com>;
> Wubin (H) <wu.wubin@huawei.com>
> Subject: Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
>
> On 12/12/21 9:05 PM, Anup Patel wrote:
> >> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> >> + if (ret) {
> >> + abort();
> >> + }
> >> + env->kvm_timer_state = reg;
> >
> > Please read the timer frequency here.
>
> Yep.
>
> >> +
> >> + env->kvm_timer_dirty = true;
> >> +}
> >> +
> >> +static void kvm_riscv_put_regs_timer(CPUState *cs) {
> >> + int ret;
> >> + uint64_t reg;
> >> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> >> +
> >> + if (!env->kvm_timer_dirty) {
> >> + return;
> >> + }
> >
> > Over here, we should get the timer frequency and abort() with an error
> > message if it does not match env->kvm_timer_frequency
> >
> > For now, migration will not work between Hosts with different timer
> > frequency.
>
> You shouldn't have to do this every "put", only on migration, at which point you
> can actually signal a migration error rather than aborting directly.
>
>
> r~
Yes, it will be modified in the next series.
Yifei
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
@ 2021-12-20 13:02 ` Jiangyifei via
0 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei via @ 2021-12-20 13:02 UTC (permalink / raw)
To: Richard Henderson, Anup Patel
Cc: Bin Meng, open list:RISC-V, limingwang (A),
KVM General, libvir-list, Anup Patel, QEMU Developers, wanbo (G),
Palmer Dabbelt, kvm-riscv, Wanghaibin (D),
Alistair Francis, Fanliang (EulerOS), Wubin (H)
> -----Original Message-----
> From: Richard Henderson [mailto:richard.henderson@linaro.org]
> Sent: Monday, December 13, 2021 11:20 PM
> To: Anup Patel <anup@brainfault.org>; Jiangyifei <jiangyifei@huawei.com>
> Cc: Bin Meng <bin.meng@windriver.com>; open list:RISC-V
> <qemu-riscv@nongnu.org>; limingwang (A) <limingwang@huawei.com>; KVM
> General <kvm@vger.kernel.org>; libvir-list@redhat.com; Anup Patel
> <anup.patel@wdc.com>; QEMU Developers <qemu-devel@nongnu.org>;
> wanbo (G) <wanbo13@huawei.com>; Palmer Dabbelt <palmer@dabbelt.com>;
> kvm-riscv@lists.infradead.org; Wanghaibin (D)
> <wanghaibin.wang@huawei.com>; Alistair Francis
> <Alistair.Francis@wdc.com>; Fanliang (EulerOS) <fanliang@huawei.com>;
> Wubin (H) <wu.wubin@huawei.com>
> Subject: Re: [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer
>
> On 12/12/21 9:05 PM, Anup Patel wrote:
> >> + ret = kvm_get_one_reg(cs, RISCV_TIMER_REG(env, state), ®);
> >> + if (ret) {
> >> + abort();
> >> + }
> >> + env->kvm_timer_state = reg;
> >
> > Please read the timer frequency here.
>
> Yep.
>
> >> +
> >> + env->kvm_timer_dirty = true;
> >> +}
> >> +
> >> +static void kvm_riscv_put_regs_timer(CPUState *cs) {
> >> + int ret;
> >> + uint64_t reg;
> >> + CPURISCVState *env = &RISCV_CPU(cs)->env;
> >> +
> >> + if (!env->kvm_timer_dirty) {
> >> + return;
> >> + }
> >
> > Over here, we should get the timer frequency and abort() with an error
> > message if it does not match env->kvm_timer_frequency
> >
> > For now, migration will not work between Hosts with different timer
> > frequency.
>
> You shouldn't have to do this every "put", only on migration, at which point you
> can actually signal a migration error rather than aborting directly.
>
>
> r~
Yes, it will be modified in the next series.
Yifei
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
2021-12-13 15:21 ` Richard Henderson
@ 2021-12-20 13:02 ` Jiangyifei via
-1 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei @ 2021-12-20 13:02 UTC (permalink / raw)
To: Richard Henderson, qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, Fanliang (EulerOS), Wubin (H),
Wanghaibin (D), wanbo (G), limingwang (A)
> -----Original Message-----
> From: Richard Henderson [mailto:richard.henderson@linaro.org]
> Sent: Monday, December 13, 2021 11:22 PM
> To: Jiangyifei <jiangyifei@huawei.com>; qemu-devel@nongnu.org;
> qemu-riscv@nongnu.org
> Cc: kvm-riscv@lists.infradead.org; kvm@vger.kernel.org;
> libvir-list@redhat.com; anup.patel@wdc.com; palmer@dabbelt.com;
> Alistair.Francis@wdc.com; bin.meng@windriver.com; Fanliang (EulerOS)
> <fanliang@huawei.com>; Wubin (H) <wu.wubin@huawei.com>; Wanghaibin (D)
> <wanghaibin.wang@huawei.com>; wanbo (G) <wanbo13@huawei.com>;
> limingwang (A) <limingwang@huawei.com>
> Subject: Re: [PATCH v2 12/12] target/riscv: Support virtual time context
> synchronization
>
> On 12/10/21 2:07 AM, Yifei Jiang via wrote:
> > +static bool kvmtimer_needed(void *opaque) {
> > + return kvm_enabled();
> > +}
> > +
> > +
> > +static const VMStateDescription vmstate_kvmtimer = {
> > + .name = "cpu/kvmtimer",
> > + .version_id = 1,
> > + .minimum_version_id = 1,
> > + .needed = kvmtimer_needed,
> > + .fields = (VMStateField[]) {
> > + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> > + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> > + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> > +
> > + VMSTATE_END_OF_LIST()
> > + }
> > +};
> > +
> > +static int cpu_post_load(void *opaque, int version_id) {
> > + RISCVCPU *cpu = opaque;
> > + CPURISCVState *env = &cpu->env;
> > +
> > + if (kvm_enabled()) {
> > + env->kvm_timer_dirty = true;
> > + }
> > + return 0;
> > +}
>
> This post-load belongs on the vmstate_kvmtimer struct, so that you need not
> re-check kvm_enabled().
>
> > const VMStateDescription vmstate_riscv_cpu = {
> > .name = "cpu",
> > - .version_id = 3,
> > - .minimum_version_id = 3,
> > + .version_id = 4,
> > + .minimum_version_id = 4,
> > + .post_load = cpu_post_load,
>
> No need for version change.
>
>
> r~
Ok, it will be modified in the next series.
Yifei
^ permalink raw reply [flat|nested] 56+ messages in thread
* RE: [PATCH v2 12/12] target/riscv: Support virtual time context synchronization
@ 2021-12-20 13:02 ` Jiangyifei via
0 siblings, 0 replies; 56+ messages in thread
From: Jiangyifei via @ 2021-12-20 13:02 UTC (permalink / raw)
To: Richard Henderson, qemu-devel, qemu-riscv
Cc: kvm-riscv, kvm, libvir-list, anup.patel, palmer,
Alistair.Francis, bin.meng, Fanliang (EulerOS), Wubin (H),
Wanghaibin (D), wanbo (G), limingwang (A)
> -----Original Message-----
> From: Richard Henderson [mailto:richard.henderson@linaro.org]
> Sent: Monday, December 13, 2021 11:22 PM
> To: Jiangyifei <jiangyifei@huawei.com>; qemu-devel@nongnu.org;
> qemu-riscv@nongnu.org
> Cc: kvm-riscv@lists.infradead.org; kvm@vger.kernel.org;
> libvir-list@redhat.com; anup.patel@wdc.com; palmer@dabbelt.com;
> Alistair.Francis@wdc.com; bin.meng@windriver.com; Fanliang (EulerOS)
> <fanliang@huawei.com>; Wubin (H) <wu.wubin@huawei.com>; Wanghaibin (D)
> <wanghaibin.wang@huawei.com>; wanbo (G) <wanbo13@huawei.com>;
> limingwang (A) <limingwang@huawei.com>
> Subject: Re: [PATCH v2 12/12] target/riscv: Support virtual time context
> synchronization
>
> On 12/10/21 2:07 AM, Yifei Jiang via wrote:
> > +static bool kvmtimer_needed(void *opaque) {
> > + return kvm_enabled();
> > +}
> > +
> > +
> > +static const VMStateDescription vmstate_kvmtimer = {
> > + .name = "cpu/kvmtimer",
> > + .version_id = 1,
> > + .minimum_version_id = 1,
> > + .needed = kvmtimer_needed,
> > + .fields = (VMStateField[]) {
> > + VMSTATE_UINT64(env.kvm_timer_time, RISCVCPU),
> > + VMSTATE_UINT64(env.kvm_timer_compare, RISCVCPU),
> > + VMSTATE_UINT64(env.kvm_timer_state, RISCVCPU),
> > +
> > + VMSTATE_END_OF_LIST()
> > + }
> > +};
> > +
> > +static int cpu_post_load(void *opaque, int version_id) {
> > + RISCVCPU *cpu = opaque;
> > + CPURISCVState *env = &cpu->env;
> > +
> > + if (kvm_enabled()) {
> > + env->kvm_timer_dirty = true;
> > + }
> > + return 0;
> > +}
>
> This post-load belongs on the vmstate_kvmtimer struct, so that you need not
> re-check kvm_enabled().
>
> > const VMStateDescription vmstate_riscv_cpu = {
> > .name = "cpu",
> > - .version_id = 3,
> > - .minimum_version_id = 3,
> > + .version_id = 4,
> > + .minimum_version_id = 4,
> > + .post_load = cpu_post_load,
>
> No need for version change.
>
>
> r~
Ok, it will be modified in the next series.
Yifei
^ permalink raw reply [flat|nested] 56+ messages in thread
end of thread, other threads:[~2021-12-20 17:20 UTC | newest]
Thread overview: 56+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-10 10:07 [PATCH v2 00/12] Add riscv kvm accel support Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-10 10:07 ` [PATCH v2 01/12] update-linux-headers: Add asm-riscv/kvm.h Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-10 10:07 ` [PATCH v2 02/12] target/riscv: Add target/riscv/kvm.c to place the public kvm interface Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-10 10:07 ` [PATCH v2 03/12] target/riscv: Implement function kvm_arch_init_vcpu Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:15 ` Anup Patel
2021-12-13 4:15 ` Anup Patel
2021-12-10 10:07 ` [PATCH v2 04/12] target/riscv: Implement kvm_arch_get_registers Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:16 ` Anup Patel
2021-12-13 4:16 ` Anup Patel
2021-12-10 10:07 ` [PATCH v2 05/12] target/riscv: Implement kvm_arch_put_registers Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:17 ` Anup Patel
2021-12-13 4:17 ` Anup Patel
2021-12-10 10:07 ` [PATCH v2 06/12] target/riscv: Support start kernel directly by KVM Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:21 ` Anup Patel
2021-12-13 4:21 ` Anup Patel
2021-12-20 12:59 ` Jiangyifei
2021-12-20 12:59 ` Jiangyifei via
2021-12-10 10:07 ` [PATCH v2 07/12] target/riscv: Support setting external interrupt " Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:32 ` Anup Patel
2021-12-13 4:32 ` Anup Patel
2021-12-20 13:01 ` Jiangyifei
2021-12-20 13:01 ` Jiangyifei via
2021-12-10 10:07 ` [PATCH v2 08/12] target/riscv: Handle KVM_EXIT_RISCV_SBI exit Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:37 ` Anup Patel
2021-12-13 4:37 ` Anup Patel
2021-12-10 10:07 ` [PATCH v2 09/12] target/riscv: Add host cpu type Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-10 10:07 ` [PATCH v2 10/12] target/riscv: Add kvm_riscv_get/put_regs_timer Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 5:05 ` Anup Patel
2021-12-13 5:05 ` Anup Patel
2021-12-13 15:19 ` Richard Henderson
2021-12-13 15:19 ` Richard Henderson
2021-12-20 13:02 ` Jiangyifei
2021-12-20 13:02 ` Jiangyifei via
2021-12-10 10:07 ` [PATCH v2 11/12] target/riscv: Implement virtual time adjusting with vm state changing Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:39 ` Anup Patel
2021-12-13 4:39 ` Anup Patel
2021-12-10 10:07 ` [PATCH v2 12/12] target/riscv: Support virtual time context synchronization Yifei Jiang
2021-12-10 10:07 ` Yifei Jiang via
2021-12-13 4:42 ` Anup Patel
2021-12-13 4:42 ` Anup Patel
2021-12-13 15:21 ` Richard Henderson
2021-12-13 15:21 ` Richard Henderson
2021-12-20 13:02 ` Jiangyifei
2021-12-20 13:02 ` Jiangyifei via
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.