target/i386/kvm: Add Hyper-V direct tlb flush support
v3 cover target/i386/kvm: Add Hyper-V direct tlb flush support
From: Tianyu Lan <[email protected]>
This patchset is to enable Hyper-V direct tlbflush feature. The feature is to let L0 hypervisor to handle tlb flush hypercall for L1 hypervisor.
Tianyu Lan (2):
linux headers: update against Linux 5.4-rc2 target/i386/kvm: Add Hyper-V direct tlb flush support
docs/hyperv.txt | 10 ++++ include/standard-headers/asm-x86/bootparam.h | 2 + include/standard-headers/asm-x86/kvm_para.h | 1 + include/standard-headers/linux/ethtool.h | 24 ++++++++++ include/standard-headers/linux/pci_regs.h | 19 +++++++- include/standard-headers/linux/virtio_ids.h | 2 + include/standard-headers/linux/virtio_pmem.h | 6 +-- linux-headers/asm-arm/kvm.h | 16 ++++++- linux-headers/asm-arm/unistd-common.h | 2 + linux-headers/asm-arm64/kvm.h | 21 +++++++- linux-headers/asm-generic/mman-common.h | 18 ++++--- linux-headers/asm-generic/mman.h | 10 ++-- linux-headers/asm-generic/unistd.h | 10 +++- linux-headers/asm-mips/mman.h | 3 ++ linux-headers/asm-mips/unistd_n32.h | 2 + linux-headers/asm-mips/unistd_n64.h | 2 + linux-headers/asm-mips/unistd_o32.h | 2 + linux-headers/asm-powerpc/mman.h | 6 +-- linux-headers/asm-powerpc/unistd_32.h | 2 + linux-headers/asm-powerpc/unistd_64.h | 2 + linux-headers/asm-s390/kvm.h | 6 +++ linux-headers/asm-s390/unistd_32.h | 2 + linux-headers/asm-s390/unistd_64.h | 2 + linux-headers/asm-x86/kvm.h | 28 ++++++++--- linux-headers/asm-x86/unistd.h | 2 +- linux-headers/asm-x86/unistd_32.h | 2 + linux-headers/asm-x86/unistd_64.h | 2 + linux-headers/asm-x86/unistd_x32.h | 2 + linux-headers/linux/kvm.h | 12 ++++- linux-headers/linux/psp-sev.h | 5 +- linux-headers/linux/vfio.h | 71 ++++++++++++++++++++-------- target/i386/cpu.c | 2 + target/i386/cpu.h | 1 + target/i386/kvm.c | 24 ++++++++++ 34 files changed, 262 insertions(+), 59 deletions(-)
Signature · 1 line
2.14.5
PATCH v3 1/2 linux headers: update against Linux 5.4-rc2
From: Tianyu Lan <[email protected]>
commit da0c9ea146cbe92b832f1b0f694840ea8eb33cce
--- include/standard-headers/asm-x86/bootparam.h | 2 + include/standard-headers/asm-x86/kvm_para.h | 1 + include/standard-headers/linux/ethtool.h | 24 ++++++++++ include/standard-headers/linux/pci_regs.h | 19 +++++++- include/standard-headers/linux/virtio_ids.h | 2 + include/standard-headers/linux/virtio_pmem.h | 6 +-- linux-headers/asm-arm/kvm.h | 16 ++++++- linux-headers/asm-arm/unistd-common.h | 2 + linux-headers/asm-arm64/kvm.h | 21 +++++++- linux-headers/asm-generic/mman-common.h | 18 ++++--- linux-headers/asm-generic/mman.h | 10 ++-- linux-headers/asm-generic/unistd.h | 10 +++- linux-headers/asm-mips/mman.h | 3 ++ linux-headers/asm-mips/unistd_n32.h | 2 + linux-headers/asm-mips/unistd_n64.h | 2 + linux-headers/asm-mips/unistd_o32.h | 2 + linux-headers/asm-powerpc/mman.h | 6 +-- linux-headers/asm-powerpc/unistd_32.h | 2 + linux-headers/asm-powerpc/unistd_64.h | 2 + linux-headers/asm-s390/kvm.h | 6 +++ linux-headers/asm-s390/unistd_32.h | 2 + linux-headers/asm-s390/unistd_64.h | 2 + linux-headers/asm-x86/kvm.h | 28 ++++++++--- linux-headers/asm-x86/unistd.h | 2 +- linux-headers/asm-x86/unistd_32.h | 2 + linux-headers/asm-x86/unistd_64.h | 2 + linux-headers/asm-x86/unistd_x32.h | 2 + linux-headers/linux/kvm.h | 12 ++++- linux-headers/linux/psp-sev.h | 5 +- linux-headers/linux/vfio.h | 71 ++++++++++++++++++++-------- 30 files changed, 225 insertions(+), 59 deletions(-)
include/standard-headers/asm-x86/bootparam.h +2 -0
@@ -29,6 +29,8 @@#define XLF_EFI_HANDOVER_32 (1<<2)#define XLF_EFI_HANDOVER_64 (1<<3)#define XLF_EFI_KEXEC (1<<4) +#define XLF_5LEVEL (1<<5) +#define XLF_5LEVEL_ENABLED (1<<6)#endif /* _ASM_X86_BOOTPARAM_H */
include/standard-headers/asm-x86/kvm_para.h +1 -0
@@ -30,6 +30,7 @@#define KVM_FEATURE_ASYNC_PF_VMEXIT 10#define KVM_FEATURE_PV_SEND_IPI 11#define KVM_FEATURE_POLL_CONTROL 12 +#define KVM_FEATURE_PV_SCHED_YIELD 13#define KVM_HINTS_REALTIME 0
include/standard-headers/linux/ethtool.h +24 -0
@@ -259,10 +259,32 @@ struct ethtool_tunable {#define ETHTOOL_PHY_FAST_LINK_DOWN_ON 0#define ETHTOOL_PHY_FAST_LINK_DOWN_OFF 0xff +/* Energy Detect Power Down (EDPD) is a feature supported by some PHYs, where +* the PHY's RX & TX blocks are put into a low-power mode when there is no +* link detected (typically cable is un-plugged). For RX, only a minimal +* link-detection is available, and for TX the PHY wakes up to send link pulses +* to avoid any lock-ups in case the peer PHY may also be running in EDPD mode. +* +* Some PHYs may support configuration of the wake-up interval for TX pulses, +* and some PHYs may support only disabling TX pulses entirely. For the latter +* a special value is required (ETHTOOL_PHY_EDPD_NO_TX) so that this can be +* configured from userspace (should the user want it). +* +* The interval units for TX wake-up are in milliseconds, since this should +* cover a reasonable range of intervals: +* - from 1 millisecond, which does not sound like much of a power-saver +* - to ~65 seconds which is quite a lot to wait for a link to come up when +* plugging a cable +*/ +#define ETHTOOL_PHY_EDPD_DFLT_TX_MSECS 0xffff +#define ETHTOOL_PHY_EDPD_NO_TX 0xfffe +#define ETHTOOL_PHY_EDPD_DISABLE 0 +enum phy_tunable_id {ETHTOOL_PHY_ID_UNSPEC ,ETHTOOL_PHY_DOWNSHIFT ,ETHTOOL_PHY_FAST_LINK_DOWN , +ETHTOOL_PHY_EDPD ,/* * Add your fresh new phy tunable attribute above and remember to update * phy_tunable_strings[] in net/core/ethtool.c
@@ -1483,6 +1505,8 @@ enum ethtool_link_mode_bit_indices { ETHTOOL_LINK_MODE_200000baseLR4_ER4_FR4_Full_BIT =64 , ETHTOOL_LINK_MODE_200000baseDR4_Full_BIT =65 , ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT =66 , + ETHTOOL_LINK_MODE_100baseT1_Full_BIT =67 , + ETHTOOL_LINK_MODE_1000baseT1_Full_BIT =68 ,/* must be last entry */ __ETHTOOL_LINK_MODE_MASK_NBITS
include/standard-headers/linux/pci_regs.h +18 -1
@@ -528,6 +528,7 @@#define PCI_EXP_LNKCAP_SLS_5_0GB 0x00000002/* LNKCAP2 SLS Vector bit 1 */ #define PCI_EXP_LNKCAP_SLS_8_0GB 0x00000003/* LNKCAP2 SLS Vector bit 2 */ #define PCI_EXP_LNKCAP_SLS_16_0GB 0x00000004/* LNKCAP2 SLS Vector bit 3 */ +#define PCI_EXP_LNKCAP_SLS_32_0GB 0x00000005/* LNKCAP2 SLS Vector bit 4 */ #define PCI_EXP_LNKCAP_MLW 0x000003f0/* Maximum Link Width */ #define PCI_EXP_LNKCAP_ASPMS 0x00000c00/* ASPM Support */ #define PCI_EXP_LNKCAP_L0SEL 0x00007000/* L0s Exit Latency */
@@ -556,6 +557,7 @@#define PCI_EXP_LNKSTA_CLS_5_0GB 0x0002/* Current Link Speed 5.0GT/s */ #define PCI_EXP_LNKSTA_CLS_8_0GB 0x0003/* Current Link Speed 8.0GT/s */ #define PCI_EXP_LNKSTA_CLS_16_0GB 0x0004/* Current Link Speed 16.0GT/s */ +#define PCI_EXP_LNKSTA_CLS_32_0GB 0x0005/* Current Link Speed 32.0GT/s */ #define PCI_EXP_LNKSTA_NLW 0x03f0/* Negotiated Link Width */ #define PCI_EXP_LNKSTA_NLW_X1 0x0010/* Current Link Width x1 */ #define PCI_EXP_LNKSTA_NLW_X2 0x0020/* Current Link Width x2 */
@@ -589,6 +591,7 @@#define PCI_EXP_SLTCTL_CCIE 0x0010/* Command Completed Interrupt Enable */ #define PCI_EXP_SLTCTL_HPIE 0x0020/* Hot-Plug Interrupt Enable */ #define PCI_EXP_SLTCTL_AIC 0x00c0/* Attention Indicator Control */ +#define PCI_EXP_SLTCTL_ATTN_IND_SHIFT 6/* Attention Indicator shift */ #define PCI_EXP_SLTCTL_ATTN_IND_ON 0x0040/* Attention Indicator on */ #define PCI_EXP_SLTCTL_ATTN_IND_BLINK 0x0080/* Attention Indicator blinking */ #define PCI_EXP_SLTCTL_ATTN_IND_OFF 0x00c0/* Attention Indicator off */
@@ -661,6 +664,7 @@#define PCI_EXP_LNKCAP2_SLS_5_0GB 0x00000004/* Supported Speed 5GT/s */ #define PCI_EXP_LNKCAP2_SLS_8_0GB 0x00000008/* Supported Speed 8GT/s */ #define PCI_EXP_LNKCAP2_SLS_16_0GB 0x00000010/* Supported Speed 16GT/s */ +#define PCI_EXP_LNKCAP2_SLS_32_0GB 0x00000020/* Supported Speed 32GT/s */ #define PCI_EXP_LNKCAP2_CROSSLINK 0x00000100/* Crosslink supported */ #define PCI_EXP_LNKCTL2 48/* Link Control 2 */ #define PCI_EXP_LNKCTL2_TLS 0x000f
@@ -668,6 +672,7 @@#define PCI_EXP_LNKCTL2_TLS_5_0GT 0x0002/* Supported Speed 5GT/s */ #define PCI_EXP_LNKCTL2_TLS_8_0GT 0x0003/* Supported Speed 8GT/s */ #define PCI_EXP_LNKCTL2_TLS_16_0GT 0x0004/* Supported Speed 16GT/s */ +#define PCI_EXP_LNKCTL2_TLS_32_0GT 0x0005/* Supported Speed 32GT/s */ #define PCI_EXP_LNKSTA2 50/* Link Status 2 */ #define PCI_CAP_EXP_ENDPOINT_SIZEOF_V2 52/* v2 endpoints with link end here */ #define PCI_EXP_SLTCAP2 52/* Slot Capabilities 2 */
@@ -709,7 +714,9 @@#define PCI_EXT_CAP_ID_DPC 0x1D/* Downstream Port Containment */ #define PCI_EXT_CAP_ID_L1SS 0x1E/* L1 PM Substates */ #define PCI_EXT_CAP_ID_PTM 0x1F/* Precision Time Measurement */ -#define PCI_EXT_CAP_ID_MAX PCI_EXT_CAP_ID_PTM +#define PCI_EXT_CAP_ID_DLF 0x25/* Data Link Feature */ +#define PCI_EXT_CAP_ID_PL_16GT 0x26/* Physical Layer 16.0 GT/s */ +#define PCI_EXT_CAP_ID_MAX PCI_EXT_CAP_ID_PL_16GT#define PCI_EXT_CAP_DSN_SIZEOF 12#define PCI_EXT_CAP_MCAST_ENDPOINT_SIZEOF 40
@@ -1049,4 +1056,14 @@#define PCI_L1SS_CTL1_LTR_L12_TH_SCALE 0xe0000000/* LTR_L1.2_THRESHOLD_Scale */ #define PCI_L1SS_CTL2 0x0c/* Control 2 Register */ +/* Data Link Feature */ +#define PCI_DLF_CAP 0x04/* Capabilities Register */ +#define PCI_DLF_EXCHANGE_ENABLE 0x80000000/* Data Link Feature Exchange Enable */ + +/* Physical Layer 16.0 GT/s */ +#define PCI_PL_16GT_LE_CTRL 0x20/* Lane Equalization Control Register */ +#define PCI_PL_16GT_LE_CTRL_DSP_TX_PRESET_MASK 0x0000000F +#define PCI_PL_16GT_LE_CTRL_USP_TX_PRESET_MASK 0x000000F0 +#define PCI_PL_16GT_LE_CTRL_USP_TX_PRESET_SHIFT 4 +#endif /* LINUX_PCI_REGS_H */
include/standard-headers/linux/virtio_ids.h +2 -0
@@ -43,6 +43,8 @@#define VIRTIO_ID_INPUT 18/* virtio input */ #define VIRTIO_ID_VSOCK 19/* virtio vsock transport */ #define VIRTIO_ID_CRYPTO 20/* virtio crypto */ +#define VIRTIO_ID_IOMMU 23/* virtio IOMMU */ +#define VIRTIO_ID_FS 26/* virtio filesystem */ #define VIRTIO_ID_PMEM 27/* virtio pmem */ #endif /* _LINUX_VIRTIO_IDS_H */
include/standard-headers/linux/virtio_pmem.h +3 -3
@@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause */ +/* SPDX-License-Identifier: (GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause */ /* * Definitions for virtio-pmem devices. *
@@ -7,8 +7,8 @@* Author(s): Pankaj Gupta <[email protected]> */ -#ifndef _UAPI_LINUX_VIRTIO_PMEM_H -#define _UAPI_LINUX_VIRTIO_PMEM_H +#ifndef _LINUX_VIRTIO_PMEM_H +#define _LINUX_VIRTIO_PMEM_H#include "standard-headers/linux/types.h" #include "standard-headers/linux/virtio_ids.h"
linux-headers/asm-arm/kvm.h +15 -1
@@ -214,6 +214,18 @@ struct kvm_vcpu_events {#define KVM_REG_ARM_FW_REG (r) (KVM_REG_ARM | KVM_REG_SIZE_U64 | \ KVM_REG_ARM_FW | ((r) & 0xffff))#define KVM_REG_ARM_PSCI_VERSION KVM_REG_ARM_FW_REG(0) +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1 KVM_REG_ARM_FW_REG(1) +/* Higher values mean better protection. */ +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL 0 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL 1 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED 2 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2 KVM_REG_ARM_FW_REG(2) +/* Higher values mean better protection. */ +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL 0 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN 1 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL 2 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED 3 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED (1U << 4)/* Device Control API: ARM VGIC */ #define KVM_DEV_ARM_VGIC_GRP_ADDR 0
@@ -254,8 +266,10 @@ struct kvm_vcpu_events {#define KVM_DEV_ARM_ITS_CTRL_RESET 4/* KVM_IRQ_LINE irq field index values */ +#define KVM_ARM_IRQ_VCPU2_SHIFT 28 +#define KVM_ARM_IRQ_VCPU2_MASK 0xf#define KVM_ARM_IRQ_TYPE_SHIFT 24 -#define KVM_ARM_IRQ_TYPE_MASK 0xff +#define KVM_ARM_IRQ_TYPE_MASK 0xf#define KVM_ARM_IRQ_VCPU_SHIFT 16#define KVM_ARM_IRQ_VCPU_MASK 0xff#define KVM_ARM_IRQ_NUM_SHIFT 0
linux-headers/asm-arm/unistd-common.h +2 -0
@@ -388,5 +388,7 @@#define __NR_fsconfig (__NR_SYSCALL_BASE + 431)#define __NR_fsmount (__NR_SYSCALL_BASE + 432)#define __NR_fspick (__NR_SYSCALL_BASE + 433) +#define __NR_pidfd_open (__NR_SYSCALL_BASE + 434) +#define __NR_clone3 (__NR_SYSCALL_BASE + 435)#endif /* _ASM_ARM_UNISTD_COMMON_H */
linux-headers/asm-arm64/kvm.h +20 -1
@@ -229,6 +229,16 @@ struct kvm_vcpu_events {#define KVM_REG_ARM_FW_REG (r) (KVM_REG_ARM64 | KVM_REG_SIZE_U64 | \ KVM_REG_ARM_FW | ((r) & 0xffff))#define KVM_REG_ARM_PSCI_VERSION KVM_REG_ARM_FW_REG(0) +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1 KVM_REG_ARM_FW_REG(1) +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL 0 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL 1 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED 2 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2 KVM_REG_ARM_FW_REG(2) +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL 0 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN 1 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL 2 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED 3 +#define KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED (1U << 4)/* SVE registers */ #define KVM_REG_ARM64_SVE (0x15 << KVM_REG_ARM_COPROC_SHIFT)
@@ -260,6 +270,13 @@ struct kvm_vcpu_events { KVM_REG_SIZE_U256 | \ ((i) & (KVM_ARM64_SVE_MAX_SLICES - 1))) +/* +* Register values for KVM_REG_ARM64_SVE_ZREG(), KVM_REG_ARM64_SVE_PREG() and +* KVM_REG_ARM64_SVE_FFR() are represented in memory in an endianness- +* invariant layout which differs from the layout used for the FPSIMD +* V-registers on big-endian systems: see sigcontext.h for more explanation. +*/ +#define KVM_ARM64_SVE_VQ_MIN __SVE_VQ_MIN#define KVM_ARM64_SVE_VQ_MAX __SVE_VQ_MAX
@@ -308,8 +325,10 @@ struct kvm_vcpu_events {#define KVM_ARM_VCPU_TIMER_IRQ_PTIMER 1/* KVM_IRQ_LINE irq field index values */ +#define KVM_ARM_IRQ_VCPU2_SHIFT 28 +#define KVM_ARM_IRQ_VCPU2_MASK 0xf#define KVM_ARM_IRQ_TYPE_SHIFT 24 -#define KVM_ARM_IRQ_TYPE_MASK 0xff +#define KVM_ARM_IRQ_TYPE_MASK 0xf#define KVM_ARM_IRQ_VCPU_SHIFT 16#define KVM_ARM_IRQ_VCPU_MASK 0xff#define KVM_ARM_IRQ_NUM_SHIFT 0
linux-headers/asm-generic/mman-common.h +12 -6
@@ -19,15 +19,18 @@#define MAP_TYPE 0x0f/* Mask for type of mapping */ #define MAP_FIXED 0x10/* Interpret addr exactly */ #define MAP_ANONYMOUS 0x20/* don't use a file */ -#ifdef CONFIG_MMAP_ALLOW_UNINITIALIZED -# define MAP_UNINITIALIZED 0x4000000/* For anonymous mmap, memory could be uninitialized */ -#else -# define MAP_UNINITIALIZED 0x0/* Don't support this flag */ -#endif -/* 0x0100 - 0x80000 flags are defined in asm-generic/mman.h */ +/* 0x0100 - 0x4000 flags are defined in asm-generic/mman.h */ +#define MAP_POPULATE 0x008000/* populate (prefault) pagetables */ +#define MAP_NONBLOCK 0x010000/* do not block on IO */ +#define MAP_STACK 0x020000/* give out an address that is best suited for process/thread stacks */ +#define MAP_HUGETLB 0x040000/* create a huge page mapping */ +#define MAP_SYNC 0x080000/* perform synchronous page faults for the mapping */ #define MAP_FIXED_NOREPLACE 0x100000/* MAP_FIXED which doesn't unmap underlying mapping */ +#define MAP_UNINITIALIZED 0x4000000/* For anonymous mmap, memory could be +* uninitialized */ +/* * Flags for mlock */
@@ -64,6 +67,9 @@#define MADV_WIPEONFORK 18/* Zero memory on fork, child only */ #define MADV_KEEPONFORK 19/* Undo MADV_WIPEONFORK */ +#define MADV_COLD 20/* deactivate these pages */ +#define MADV_PAGEOUT 21/* reclaim these pages */ +/* compatibility flags */ #define MAP_FILE 0
linux-headers/asm-generic/mman.h +4 -6
@@ -9,13 +9,11 @@#define MAP_EXECUTABLE 0x1000/* mark it as an executable */ #define MAP_LOCKED 0x2000/* pages are locked */ #define MAP_NORESERVE 0x4000/* don't check for reservations */ -#define MAP_POPULATE 0x8000/* populate (prefault) pagetables */ -#define MAP_NONBLOCK 0x10000/* do not block on IO */ -#define MAP_STACK 0x20000/* give out an address that is best suited for process/thread stacks */ -#define MAP_HUGETLB 0x40000/* create a huge page mapping */ -#define MAP_SYNC 0x80000/* perform synchronous page faults for the mapping */ -/* Bits [26:31] are reserved, see mman-common.h for MAP_HUGETLB usage */ +/* +* Bits [26:31] are reserved, see asm-generic/hugetlb_encode.h +* for MAP_HUGETLB usage +*/ #define MCL_CURRENT 1/* lock all current mappings */ #define MCL_FUTURE 2/* lock all future mappings */
linux-headers/asm-generic/unistd.h +8 -2
@@ -569,7 +569,7 @@ __SYSCALL(__NR_semget, sys_semget) __SC_COMP(__NR_semctl, sys_semctl, compat_sys_semctl)#if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG !=32 #define __NR_semtimedop 192 -__SC_COMP(__NR_semtimedop, sys_semtimedop, sys_semtimedop_time32) +__SC_3264(__NR_semtimedop, sys_semtimedop_time32, sys_semtimedop)#endif #define __NR_semop 193__SYSCALL (__NR_semop, sys_semop)
@@ -844,9 +844,15 @@ __SYSCALL(__NR_fsconfig, sys_fsconfig) __SYSCALL(__NR_fsmount, sys_fsmount)#define __NR_fspick 433 __SYSCALL(__NR_fspick, sys_fspick) +#define __NR_pidfd_open 434 +__SYSCALL(__NR_pidfd_open, sys_pidfd_open) +#ifdef __ARCH_WANT_SYS_CLONE3 +#define __NR_clone3 435 +__SYSCALL(__NR_clone3, sys_clone3) +#endif #undef __NR_syscalls -#define __NR_syscalls 434 +#define __NR_syscalls 436/* * 32 bit systems traditionally used different
linux-headers/asm-mips/mman.h +3 -0
@@ -95,6 +95,9 @@#define MADV_WIPEONFORK 18/* Zero memory on fork, child only */ #define MADV_KEEPONFORK 19/* Undo MADV_WIPEONFORK */ +#define MADV_COLD 20/* deactivate these pages */ +#define MADV_PAGEOUT 21/* reclaim these pages */ +/* compatibility flags */ #define MAP_FILE 0
linux-headers/asm-mips/unistd_n32.h +2 -0
@@ -363,6 +363,8 @@#define __NR_fsconfig (__NR_Linux + 431)#define __NR_fsmount (__NR_Linux + 432)#define __NR_fspick (__NR_Linux + 433) +#define __NR_pidfd_open (__NR_Linux + 434) +#define __NR_clone3 (__NR_Linux + 435)#endif /* _ASM_MIPS_UNISTD_N32_H */
linux-headers/asm-mips/unistd_n64.h +2 -0
@@ -339,6 +339,8 @@#define __NR_fsconfig (__NR_Linux + 431)#define __NR_fsmount (__NR_Linux + 432)#define __NR_fspick (__NR_Linux + 433) +#define __NR_pidfd_open (__NR_Linux + 434) +#define __NR_clone3 (__NR_Linux + 435)#endif /* _ASM_MIPS_UNISTD_N64_H */
linux-headers/asm-mips/unistd_o32.h +2 -0
@@ -409,6 +409,8 @@#define __NR_fsconfig (__NR_Linux + 431)#define __NR_fsmount (__NR_Linux + 432)#define __NR_fspick (__NR_Linux + 433) +#define __NR_pidfd_open (__NR_Linux + 434) +#define __NR_clone3 (__NR_Linux + 435)#endif /* _ASM_MIPS_UNISTD_O32_H */
linux-headers/asm-powerpc/mman.h +1 -5
@@ -21,15 +21,11 @@#define MAP_DENYWRITE 0x0800/* ETXTBSY */ #define MAP_EXECUTABLE 0x1000/* mark it as an executable */ +#define MCL_CURRENT 0x2000/* lock all currently mapped pages */ #define MCL_FUTURE 0x4000/* lock all additions to address space */ #define MCL_ONFAULT 0x8000/* lock all pages that are faulted in */ -#define MAP_POPULATE 0x8000/* populate (prefault) pagetables */ -#define MAP_NONBLOCK 0x10000/* do not block on IO */ -#define MAP_STACK 0x20000/* give out an address that is best suited for process/thread stacks */ -#define MAP_HUGETLB 0x40000/* create a huge page mapping */ -/* Override any generic PKEY permission defines */ #define PKEY_DISABLE_EXECUTE 0x4#undef PKEY_ACCESS_MASK
linux-headers/asm-powerpc/unistd_32.h +2 -0
@@ -416,6 +416,8 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_POWERPC_UNISTD_32_H */
linux-headers/asm-powerpc/unistd_64.h +2 -0
@@ -388,6 +388,8 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_POWERPC_UNISTD_64_H */
linux-headers/asm-s390/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {#define KVM_SYNC_GSCB (1UL << 9)#define KVM_SYNC_BPBC (1UL << 10)#define KVM_SYNC_ETOKEN (1UL << 11) + +#define KVM_SYNC_S390_VALID_FIELDS \ + (KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \ + KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \ + KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN) +/* length and alignment of the sdnx as a power of two */ #define SDNXC 8#define SDNXL (1UL << SDNXC)
linux-headers/asm-s390/unistd_32.h +2 -0
@@ -406,5 +406,7 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_S390_UNISTD_32_H */
linux-headers/asm-s390/unistd_64.h +2 -0
@@ -354,5 +354,7 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_S390_UNISTD_64_H */
linux-headers/asm-x86/kvm.h +21 -7
@@ -378,23 +378,24 @@ struct kvm_sync_regs {struct kvm_vcpu_events events; }; -#define KVM_X86_QUIRK_LINT0_REENABLED (1 << 0) -#define KVM_X86_QUIRK_CD_NW_CLEARED (1 << 1) -#define KVM_X86_QUIRK_LAPIC_MMIO_HOLE (1 << 2) -#define KVM_X86_QUIRK_OUT_7E_INC_RIP (1 << 3) +#define KVM_X86_QUIRK_LINT0_REENABLED (1 << 0) +#define KVM_X86_QUIRK_CD_NW_CLEARED (1 << 1) +#define KVM_X86_QUIRK_LAPIC_MMIO_HOLE (1 << 2) +#define KVM_X86_QUIRK_OUT_7E_INC_RIP (1 << 3) +#define KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT (1 << 4)#define KVM_STATE_NESTED_FORMAT_VMX 0 -#define KVM_STATE_NESTED_FORMAT_SVM 1 +#define KVM_STATE_NESTED_FORMAT_SVM 1/* unused */ #define KVM_STATE_NESTED_GUEST_MODE 0x00000001#define KVM_STATE_NESTED_RUN_PENDING 0x00000002#define KVM_STATE_NESTED_EVMCS 0x00000004 -#define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 -#define KVM_STATE_NESTED_SMM_GUEST_MODE 0x00000001#define KVM_STATE_NESTED_SMM_VMXON 0x00000002 +#define KVM_STATE_NESTED_VMX_VMCS_SIZE 0x1000 +struct kvm_vmx_nested_state_data {__u8 vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE ];__u8 shadow_vmcs12[KVM_STATE_NESTED_VMX_VMCS_SIZE ];
@@ -432,4 +433,17 @@ struct kvm_nested_state { } data; }; +/* for KVM_CAP_PMU_EVENT_FILTER */ +struct kvm_pmu_event_filter { +__u32 action; +__u32 nevents; +__u32 fixed_counter_bitmap; +__u32 flags; +__u32 pad[4 ]; +__u64 events[0 ]; +}; + +#define KVM_PMU_EVENT_ALLOW 0 +#define KVM_PMU_EVENT_DENY 1 +#endif /* _ASM_X86_KVM_H */
linux-headers/asm-x86/unistd.h +1 -1
@@ -3,7 +3,7 @@#define _ASM_X86_UNISTD_H/* x32 syscall flag bit */ -#define __X32_SYSCALL_BIT 0x40000000 +#define __X32_SYSCALL_BIT 0x40000000UL# ifdef __i386__# include <asm/unistd_32.h>
linux-headers/asm-x86/unistd_32.h +2 -0
@@ -424,5 +424,7 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_X86_UNISTD_32_H */
linux-headers/asm-x86/unistd_64.h +2 -0
@@ -346,5 +346,7 @@#define __NR_fsconfig 431#define __NR_fsmount 432#define __NR_fspick 433 +#define __NR_pidfd_open 434 +#define __NR_clone3 435#endif /* _ASM_X86_UNISTD_64_H */
linux-headers/asm-x86/unistd_x32.h +2 -0
@@ -299,6 +299,8 @@#define __NR_fsconfig (__X32_SYSCALL_BIT + 431)#define __NR_fsmount (__X32_SYSCALL_BIT + 432)#define __NR_fspick (__X32_SYSCALL_BIT + 433) +#define __NR_pidfd_open (__X32_SYSCALL_BIT + 434) +#define __NR_clone3 (__X32_SYSCALL_BIT + 435)#define __NR_rt_sigaction (__X32_SYSCALL_BIT + 512)#define __NR_rt_sigreturn (__X32_SYSCALL_BIT + 513)#define __NR_ioctl (__X32_SYSCALL_BIT + 514)
linux-headers/linux/kvm.h +10 -2
@@ -116,7 +116,7 @@ struct kvm_irq_level {* ACPI gsi notion of irq. * For IA-64 (APIC model) IOAPIC0: irq 0-23; IOAPIC1: irq 24-47.. * For X86 (standard AT mode) PIC0/1: irq 0-15. IOAPIC0: 0-23.. -* For ARM: See Documentation/virtual/kvm/api.txt +* For ARM: See Documentation/virt/kvm/api.txt */ union {__u32 irq;
@@ -243,6 +243,8 @@ struct kvm_hyperv_exit {#define KVM_INTERNAL_ERROR_SIMUL_EX 2/* Encounter unexpected vm-exit due to delivery event. */ #define KVM_INTERNAL_ERROR_DELIVERY_EV 3 +/* Encounter unexpected vm-exit reason */ +#define KVM_INTERNAL_ERROR_UNEXPECTED_EXIT_REASON 4/* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */ struct kvm_run {
@@ -995,6 +997,9 @@ struct kvm_ppc_resize_hpt {#define KVM_CAP_ARM_SVE 170#define KVM_CAP_ARM_PTRAUTH_ADDRESS 171#define KVM_CAP_ARM_PTRAUTH_GENERIC 172 +#define KVM_CAP_PMU_EVENT_FILTER 173 +#define KVM_CAP_ARM_IRQ_LINE_LAYOUT_2 174 +#define KVM_CAP_HYPERV_DIRECT_TLBFLUSH 175#ifdef KVM_CAP_IRQ_ROUTING
@@ -1085,7 +1090,7 @@ struct kvm_xen_hvm_config {* * KVM_IRQFD_FLAG_RESAMPLE indicates resamplefd is valid and specifies * the irqfd to operate in resampling mode for level triggered interrupt -* emulation. See Documentation/virtual/kvm/api.txt. +* emulation. See Documentation/virt/kvm/api.txt. */ #define KVM_IRQFD_FLAG_RESAMPLE (1 << 1)
@@ -1141,6 +1146,7 @@ struct kvm_dirty_tlb {#define KVM_REG_S390 0x5000000000000000ULL#define KVM_REG_ARM64 0x6000000000000000ULL#define KVM_REG_MIPS 0x7000000000000000ULL +#define KVM_REG_RISCV 0x8000000000000000ULL#define KVM_REG_SIZE_SHIFT 52#define KVM_REG_SIZE_MASK 0x00f0000000000000ULL
@@ -1329,6 +1335,8 @@ struct kvm_s390_ucas_mapping {#define KVM_PPC_GET_RMMU_INFO _IOW(KVMIO, 0xb0, struct kvm_ppc_rmmu_info)/* Available with KVM_CAP_PPC_GET_CPU_CHAR */ #define KVM_PPC_GET_CPU_CHAR _IOR(KVMIO, 0xb1, struct kvm_ppc_cpu_char) +/* Available with KVM_CAP_PMU_EVENT_FILTER */ +#define KVM_SET_PMU_EVENT_FILTER _IOW(KVMIO, 0xb2, struct kvm_pmu_event_filter)/* ioctl for vm fd */ #define KVM_CREATE_DEVICE _IOWR(KVMIO, 0xe0, struct kvm_create_device)
linux-headers/linux/psp-sev.h +1 -4
@@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */ /* * Userspace interface for AMD Secure Encrypted Virtualization (SEV) * platform management commands.
@@ -7,10 +8,6 @@* Author: Brijesh Singh <[email protected]> * * SEV API specification is available at: https://developer.amd.com/sev/ -* -* This program is free software; you can redistribute it and/or modify -* it under the terms of the GNU General Public License version 2 as -* published by the Free Software Foundation. */ #ifndef __PSP_SEV_USER_H__
linux-headers/linux/vfio.h +51 -20
98 lines not on this page · raw message
Signature · 1 line
2.14.5
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 116 lines · On 16/10/19 15:07, [email protected] wrote:
From: Tianyu Lan <[email protected]>
Hyper-V direct tlb flush targets KVM on Hyper-V guest. Enable direct TLB flush for its guests meaning that TLB flush hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM in Level 1. Due to the different ABI for hypercall parameters between Hyper-V and KVM, KVM capabilities should be hidden when enable Hyper-V direct tlb flush otherwise KVM hypercalls may be intercepted by Hyper-V. Add new parameter "hv-direct-tlbflush". Check
expose_kvmand Hyper-V tlb flush capability status before enabling the feature.Signed-off-by Tianyu Lan <[email protected]>---
Change sicne v2:
- Update new feature description and name.
- Change failure print log.Change since v1:
- Add direct tlb flush's Hyper-V property and usehv_cpuid_check_and_set()to check the dependency of tlbflush feature.
- Make new feature work with Hyper-V passthrough mode.--- docs/hyperv.txt | 10 ++++++++++ target/i386/cpu.c | 2 ++ target/i386/cpu.h | 1 + target/i386/kvm.c | 24 ++++++++++++++++++++++++ 4 files changed, 37 insertions(+)docs/hyperv.txt +10
-0@@ -184,6 +184,16 @@ enabled. Requires: hv-vpindex, hv-synic, hv-time, hv-stimer +3.18. hv-direct-tlbflush +======================= +Enable direct TLB flush for KVM when it is running as a nested +hypervisor on top Hyper-V. When enabled, TLB flush hypercalls from L2 +guests are being passed through to L0 (Hyper-V) for handling. Due to ABI +differences between Hyper-V and KVM hypercalls, L2 guests will not be +able to issue KVM hypercalls (as those could be mishanled by L0 +Hyper-V), this requires KVM hypervisor signature to be hidden. + +Requires: hv-tlbflush, -kvm 4. Development features ========================target/i386/cpu.c +2
-0@@ -6156,6 +6156,8 @@ static Property x86_cpu_properties[] = { HYPERV_FEAT_IPI, 0),DEFINE_PROP_BIT64 ("hv-stimer-direct" ,X86CPU , hyperv_features,HYPERV_FEAT_STIMER_DIRECT ,0 ), +DEFINE_PROP_BIT64 ("hv-direct-tlbflush" ,X86CPU , hyperv_features, +HYPERV_FEAT_DIRECT_TLBFLUSH ,0 ),DEFINE_PROP_BOOL ("hv-passthrough" ,X86CPU , hyperv_passthrough, false), DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true),target/i386/cpu.h +1
-0@@ -907,6 +907,7 @@ typedef uint64_t FeatureWordArray[FEATURE_WORDS];#define HYPERV_FEAT_EVMCS 12#define HYPERV_FEAT_IPI 13#define HYPERV_FEAT_STIMER_DIRECT 14 +#define HYPERV_FEAT_DIRECT_TLBFLUSH 15#ifndef HYPERV_SPINLOCK_NEVER_RETRY #define HYPERV_SPINLOCK_NEVER_RETRY 0xFFFFFFFFtarget/i386/kvm.c +17
-0@@ -900,6 +900,10 @@ static struct { }, .dependencies = BIT(HYPERV_FEAT_STIMER) }, + [HYPERV_FEAT_DIRECT_TLBFLUSH] = { + .desc = "direct paravirtualized TLB flush (hv-direct-tlbflush)", + .dependencies =BIT (HYPERV_FEAT_TLBFLUSH ) + }, };static struct kvm_cpuid2 *try_get_hv_cpuid (CPUState *cs,int max)@@ -1224,6 +1228,7 @@ static int hyperv_handle_properties(CPUState *cs, r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_EVMCS ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_IPI ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_STIMER_DIRECT ); + r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_DIRECT_TLBFLUSH );/* Additional dependencies not covered by kvm_hyperv_properties[] */ if (hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNIC) &&@@ -1243,6 +1248,25 @@ static int hyperv_handle_properties(CPUState *cs, goto free; } +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) || + cpu->hyperv_passthrough) { +if (!cpu->expose_kvm) { + r =kvm_vcpu_enable_cap (cs,KVM_CAP_HYPERV_DIRECT_TLBFLUSH ,0 ,0 ); +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) && r) { +fprintf (stderr, +"Hyper-V %s is not supported by kernel\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH ].desc); +return -ENOSYS ; + } + }else if (!cpu->hyperv_passthrough) { + fprintf(stderr,
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1272
Quoted 10 lines
target/i386/kvm.c +7
-0+"Hyper-V %s requires KVM hypervisor signature " +"to be hidden (-kvm).\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH].desc); +return -ENOSYS ; + } + } +if (cpu->hyperv_passthrough) {/* We already copied all feature words from KVM as is */ r = cpuid->nent;
Queued, thanks. Patch 1 is not needed anymore.
Paolo
Quoted 1 line · Paolo Bonzini
On 16/10/19 15:07, [email protected] wrote:
Somehow this patch never got through to me so I'll reply here.
Quoted 90 lines
Quoted 90 lines
From: Tianyu Lan <[email protected]>
Hyper-V direct tlb flush targets KVM on Hyper-V guest. Enable direct TLB flush for its guests meaning that TLB flush hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM in Level 1. Due to the different ABI for hypercall parameters between Hyper-V and KVM, KVM capabilities should be hidden when enable Hyper-V direct tlb flush otherwise KVM hypercalls may be intercepted by Hyper-V. Add new parameter "hv-direct-tlbflush". Check
expose_kvmand Hyper-V tlb flush capability status before enabling the feature.Signed-off-by Tianyu Lan <[email protected]>---
Change sicne v2:
- Update new feature description and name.
- Change failure print log.Change since v1:
- Add direct tlb flush's Hyper-V property and usehv_cpuid_check_and_set()to check the dependency of tlbflush feature.
- Make new feature work with Hyper-V passthrough mode.--- docs/hyperv.txt | 10 ++++++++++ target/i386/cpu.c | 2 ++ target/i386/cpu.h | 1 + target/i386/kvm.c | 24 ++++++++++++++++++++++++ 4 files changed, 37 insertions(+)docs/hyperv.txt +10
-0@@ -184,6 +184,16 @@ enabled. Requires: hv-vpindex, hv-synic, hv-time, hv-stimer +3.18. hv-direct-tlbflush +======================= +Enable direct TLB flush for KVM when it is running as a nested +hypervisor on top Hyper-V. When enabled, TLB flush hypercalls from L2 +guests are being passed through to L0 (Hyper-V) for handling. Due to ABI +differences between Hyper-V and KVM hypercalls, L2 guests will not be +able to issue KVM hypercalls (as those could be mishanled by L0 +Hyper-V), this requires KVM hypervisor signature to be hidden. + +Requires: hv-tlbflush, -kvm 4. Development features ========================target/i386/cpu.c +2
-0@@ -6156,6 +6156,8 @@ static Property x86_cpu_properties[] = { HYPERV_FEAT_IPI, 0),DEFINE_PROP_BIT64 ("hv-stimer-direct" ,X86CPU , hyperv_features,HYPERV_FEAT_STIMER_DIRECT ,0 ), +DEFINE_PROP_BIT64 ("hv-direct-tlbflush" ,X86CPU , hyperv_features, +HYPERV_FEAT_DIRECT_TLBFLUSH ,0 ),DEFINE_PROP_BOOL ("hv-passthrough" ,X86CPU , hyperv_passthrough, false), DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true),target/i386/cpu.h +1
-0@@ -907,6 +907,7 @@ typedef uint64_t FeatureWordArray[FEATURE_WORDS];#define HYPERV_FEAT_EVMCS 12#define HYPERV_FEAT_IPI 13#define HYPERV_FEAT_STIMER_DIRECT 14 +#define HYPERV_FEAT_DIRECT_TLBFLUSH 15#ifndef HYPERV_SPINLOCK_NEVER_RETRY #define HYPERV_SPINLOCK_NEVER_RETRY 0xFFFFFFFFtarget/i386/kvm.c +4
-0@@ -900,6 +900,10 @@ static struct { }, .dependencies = BIT(HYPERV_FEAT_STIMER) }, + [HYPERV_FEAT_DIRECT_TLBFLUSH] = { + .desc = "direct paravirtualized TLB flush (hv-direct-tlbflush)", + .dependencies =BIT (HYPERV_FEAT_TLBFLUSH ) + }, };
Quoted 6 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1231
Quoted 6 lines
target/i386/kvm.c +1
-0static struct kvm_cpuid2 *try_get_hv_cpuid (CPUState *cs,int max)@@ -1224,6 +1228,7 @@ static int hyperv_handle_properties(CPUState *cs, r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_EVMCS ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_IPI ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_STIMER_DIRECT ); + r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_DIRECT_TLBFLUSH );
AFAICS this will turn HYPERV_FEAT_DIRECT_TLBFLUSH on if hyperv_passthrough is on, so ...
Quoted 9 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1252
Quoted 9 lines
target/i386/kvm.c +2
-0/* Additional dependencies not covered by kvm_hyperv_properties[] */ if (hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNIC) &&@@ -1243,6 +1248,25 @@ static int hyperv_handle_properties(CPUState *cs, goto free; } +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) || + cpu->hyperv_passthrough) {
... the test for ->hyperv_passthrough is redundant, and ...
Quoted 3 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1255
Quoted 3 lines
target/i386/kvm.c +3
-0+if (!cpu->expose_kvm) { + r =kvm_vcpu_enable_cap (cs,KVM_CAP_HYPERV_DIRECT_TLBFLUSH ,0 ,0 ); +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) && r) {
... , more importantly, this will abort QEMU if HYPERV_FEAT_DIRECT_TLBFLUSH wasn't requested explicitly, but was activated by ->hyperv_passthrough, and setting the capability failed. I think the meaning of hyperv_passthrough is "enable all hyperv features supported by the KVM", so in this case it looks more correct to just clear the feature bit and go ahead.
Quoted 4 lines
Quoted 4 lines
target/i386/kvm.c +4
-0+fprintf (stderr, +"Hyper-V %s is not supported by kernel\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH ].desc); +return -ENOSYS ;
Quoted 8 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1267
Quoted 8 lines
target/i386/kvm.c +8
-0+ } + }else if (!cpu->hyperv_passthrough) { +fprintf (stderr, +"Hyper-V %s requires KVM hypervisor signature " +"to be hidden (-kvm).\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH ].desc); +return -ENOSYS ; + }
You reach here if ->expose_kvm && ->hyperv_passthrough, and no capability is activated, and you go ahead with the feature bit set. This doesn't look right either.
So in general it should probably look like
if (hyperv_feat_enabled(HYPERV_FEAT_DIRECT_TLBFLUSH)) {
if (kvm_vcpu_enable_cap(KVM_CAP_HYPERV_DIRECT_TLBFLUSH)) {
if (!cpu->hyperv_passthrough) {
... report feature unsupported by kernel ...
return -ENOSYS;
}
cpu->hyperv_features &= ~BIT(HYPERV_FEAT_DIRECT_TLBFLUSH);
} else if (cpu->expose_kvm) {
... report conflict ...
return -ENOSYS;
}
}[Yes, hyperv_passthrough hurts, but you've been warned ;)]
Thanks,
Roman.
Paolo Bonzini
maintainerQuoted 160 lines · Roman Kagan
Quoted 1 line · Paolo Bonzini
On 16/10/19 15:07, [email protected] wrote:
Somehow this patch never got through to me so I'll reply here.
Quoted 90 lines
Quoted 90 lines
From: Tianyu Lan <[email protected]>
Hyper-V direct tlb flush targets KVM on Hyper-V guest. Enable direct TLB flush for its guests meaning that TLB flush hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM in Level 1. Due to the different ABI for hypercall parameters between Hyper-V and KVM, KVM capabilities should be hidden when enable Hyper-V direct tlb flush otherwise KVM hypercalls may be intercepted by Hyper-V. Add new parameter "hv-direct-tlbflush". Check
expose_kvmand Hyper-V tlb flush capability status before enabling the feature.Signed-off-by Tianyu Lan <[email protected]>---
Change sicne v2:
- Update new feature description and name.
- Change failure print log.Change since v1:
- Add direct tlb flush's Hyper-V property and usehv_cpuid_check_and_set()to check the dependency of tlbflush feature.
- Make new feature work with Hyper-V passthrough mode.--- docs/hyperv.txt | 10 ++++++++++ target/i386/cpu.c | 2 ++ target/i386/cpu.h | 1 + target/i386/kvm.c | 24 ++++++++++++++++++++++++ 4 files changed, 37 insertions(+)docs/hyperv.txt +10
-0@@ -184,6 +184,16 @@ enabled. Requires: hv-vpindex, hv-synic, hv-time, hv-stimer +3.18. hv-direct-tlbflush +======================= +Enable direct TLB flush for KVM when it is running as a nested +hypervisor on top Hyper-V. When enabled, TLB flush hypercalls from L2 +guests are being passed through to L0 (Hyper-V) for handling. Due to ABI +differences between Hyper-V and KVM hypercalls, L2 guests will not be +able to issue KVM hypercalls (as those could be mishanled by L0 +Hyper-V), this requires KVM hypervisor signature to be hidden. + +Requires: hv-tlbflush, -kvm 4. Development features ========================target/i386/cpu.c +2
-0@@ -6156,6 +6156,8 @@ static Property x86_cpu_properties[] = { HYPERV_FEAT_IPI, 0),DEFINE_PROP_BIT64 ("hv-stimer-direct" ,X86CPU , hyperv_features,HYPERV_FEAT_STIMER_DIRECT ,0 ), +DEFINE_PROP_BIT64 ("hv-direct-tlbflush" ,X86CPU , hyperv_features, +HYPERV_FEAT_DIRECT_TLBFLUSH ,0 ),DEFINE_PROP_BOOL ("hv-passthrough" ,X86CPU , hyperv_passthrough, false), DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true),target/i386/cpu.h +1
-0@@ -907,6 +907,7 @@ typedef uint64_t FeatureWordArray[FEATURE_WORDS];#define HYPERV_FEAT_EVMCS 12#define HYPERV_FEAT_IPI 13#define HYPERV_FEAT_STIMER_DIRECT 14 +#define HYPERV_FEAT_DIRECT_TLBFLUSH 15#ifndef HYPERV_SPINLOCK_NEVER_RETRY #define HYPERV_SPINLOCK_NEVER_RETRY 0xFFFFFFFFtarget/i386/kvm.c +4
-0@@ -900,6 +900,10 @@ static struct { }, .dependencies = BIT(HYPERV_FEAT_STIMER) }, + [HYPERV_FEAT_DIRECT_TLBFLUSH] = { + .desc = "direct paravirtualized TLB flush (hv-direct-tlbflush)", + .dependencies =BIT (HYPERV_FEAT_TLBFLUSH ) + }, };Quoted 6 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1231
Quoted 6 lines
target/i386/kvm.c +1
-0static struct kvm_cpuid2 *try_get_hv_cpuid (CPUState *cs,int max)@@ -1224,6 +1228,7 @@ static int hyperv_handle_properties(CPUState *cs, r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_EVMCS ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_IPI ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_STIMER_DIRECT ); + r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_DIRECT_TLBFLUSH );AFAICS this will turn
HYPERV_FEAT_DIRECT_TLBFLUSHon ifhyperv_passthroughis on, so ...Quoted 9 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1252
Quoted 9 lines
target/i386/kvm.c +2
-0/* Additional dependencies not covered by kvm_hyperv_properties[] */ if (hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNIC) &&@@ -1243,6 +1248,25 @@ static int hyperv_handle_properties(CPUState *cs, goto free; } +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) || + cpu->hyperv_passthrough) {... the test for ->hyperv_passthrough is redundant, and ...
Quoted 3 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1255
Quoted 3 lines
target/i386/kvm.c +3
-0+if (!cpu->expose_kvm) { + r =kvm_vcpu_enable_cap (cs,KVM_CAP_HYPERV_DIRECT_TLBFLUSH ,0 ,0 ); +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) && r) {... , more importantly, this will abort QEMU if
HYPERV_FEAT_DIRECT_TLBFLUSHwasn't requested explicitly, but was activated by ->hyperv_passthrough, and setting the capability failed. I think the meaning ofhyperv_passthroughis "enable all hyperv features supported by the KVM", so in this case it looks more correct to just clear the feature bit and go ahead.Quoted 4 lines
Quoted 4 lines
target/i386/kvm.c +4
-0+fprintf (stderr, +"Hyper-V %s is not supported by kernel\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH ].desc); +return -ENOSYS ;Quoted 8 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState *cs, · line 1267
Quoted 8 lines
target/i386/kvm.c +8
-0+ } + }else if (!cpu->hyperv_passthrough) { +fprintf (stderr, +"Hyper-V %s requires KVM hypervisor signature " +"to be hidden (-kvm).\n" , + kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH ].desc); +return -ENOSYS ; + }You reach here if ->expose_kvm && ->hyperv_passthrough, and no capability is activated, and you go ahead with the feature bit set. This doesn't look right either.
So in general it should probably look like
if (hyperv_feat_enabled(HYPERV_FEAT_DIRECT_TLBFLUSH)) { if (kvm_vcpu_enable_cap(KVM_CAP_HYPERV_DIRECT_TLBFLUSH)) { if (!cpu->hyperv_passthrough) { ... report feature unsupported by kernel ... return -ENOSYS; } cpu->hyperv_features &= ~BIT(HYPERV_FEAT_DIRECT_TLBFLUSH); } else if (cpu->expose_kvm) { ... report conflict ... return -ENOSYS; } }
Quoted 1 line
[Yes,
hyperv_passthroughhurts, but you've been warned ;)]
Unqueued, thanks. :)
Paolo
Hi Roman:
Thanks for your review.
Quoted 118 lines
From: Roman Kagan <[email protected]>
Sent: Wednesday, October 23, 2019 4:14 AM
To: Paolo Bonzini <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; vkuznets <[email protected]>; Tianyu Lan <[email protected]>; [email protected]; [email protected]
Subject: Re: [PATCH V3 2/2] target/i386/kvm: Add Hyper-V direct tlb flush supportQuoted 1 line · Paolo Bonzini
On 16/10/19 15:07, [email protected] wrote:
Somehow this patch never got through to me so I'll reply here.
Quoted 38 lines
Quoted 38 lines
From: Tianyu Lan <[email protected]>
Hyper-V direct tlb flush targets KVM on Hyper-V guest. Enable direct TLB flush for its guests meaning that TLB flush hypercalls are handled by Level 0 hypervisor (Hyper-V) bypassing KVM in Level 1. Due to the different ABI for hypercall parameters between Hyper-V and KVM, KVM capabilities should be hidden when enable Hyper-V direct tlb flush otherwise KVM hypercalls may be intercepted by Hyper-V. Add new parameter "hv-direct-tlbflush". Check
expose_kvmand Hyper-V tlb flush capability status before enabling the feature.Signed-off-by Tianyu Lan <[email protected]>---
Change sicne v2:
- Update new feature description and name.
- Change failure print log.Change since v1:
- Add direct tlb flush's Hyper-V property and usehv_cpuid_check_and_set()to check the dependency of tlbflush feature.
- Make new feature work with Hyper-V passthrough mode.--- docs/hyperv.txt | 10 ++++++++++ target/i386/cpu.c | 2 ++ target/i386/cpu.h | 1 + target/i386/kvm.c | 24 ++++++++++++++++++++++++ 4 files changed, 37 insertions(+)docs/hyperv.txt index renamed from docs/hyperv.txt +0
-0
8fdf25c829..140a5c7e44100644docs/hyperv.txt +0
-0@@ -184,6 +184,16 @@ enabled. Requires: hv-vpindex, hv-synic, hv-time, hv-stimerQuoted 8 lines
docs/hyperv.txt enabled. · line 194
Quoted 8 lines
docs/hyperv.txt +8
-0+3.18. hv-direct-tlbflush +======================= +Enable direct TLB flush for KVM when it is running as a nested +hypervisor on top Hyper-V. When enabled, TLB flush hypercalls from +L2 guests are being passed through to L0 (Hyper-V) for handling. +Due to ABI differences between Hyper-V and KVM hypercalls, L2 +guests will not be able to issue KVM hypercalls (as those could be +mishanled by L0 Hyper-V), this requires KVM hypervisor signature to behidden.
Quoted 18 lines
Quoted 18 lines
docs/hyperv.txt +2
-0+ +Requires: hv-tlbflush, -kvm 4. Development features ========================target/i386/cpu.c index renamed from target/i386/cpu.c +0
-0
44f1bbdcac..7bc7fee512100644target/i386/cpu.c +2
-0@@ -6156,6 +6156,8 @@ static Property x86_cpu_properties[] = { HYPERV_FEAT_IPI, 0),DEFINE_PROP_BIT64 ("hv-stimer-direct" ,X86CPU , hyperv_features,HYPERV_FEAT_STIMER_DIRECT ,0 ), +DEFINE_PROP_BIT64 ("hv-direct-tlbflush" ,X86CPU , hyperv_features, +HYPERV_FEAT_DIRECT_TLBFLUSH ,0 ),DEFINE_PROP_BOOL ("hv-passthrough" ,X86CPU , hyperv_passthrough, false),Quoted 6 lines
target/i386/cpu.h typedef uint64_t
Quoted 6 lines
DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true), diff
--gita/target/i386/cpu.hb/target/i386/cpu.hindexeaa5395aa5..3cb105f7d6100644target/i386/cpu.h +0
-0@@ -907,6 +907,7 @@ typedef uint64_tFeatureWordArray[FEATURE_WORDS];
Quoted 21 lines
Quoted 21 lines
target/i386/cpu.h +1
-0#define HYPERV_FEAT_EVMCS 12#define HYPERV_FEAT_IPI 13#define HYPERV_FEAT_STIMER_DIRECT 14 +#define HYPERV_FEAT_DIRECT_TLBFLUSH 15#ifndef HYPERV_SPINLOCK_NEVER_RETRY #define HYPERV_SPINLOCK_NEVER_RETRY 0xFFFFFFFFtarget/i386/kvm.c index renamed from target/i386/kvm.c +0
-0
11b9c854b5..043b66ab22100644target/i386/kvm.c +4
-0@@ -900,6 +900,10 @@ static struct { }, .dependencies = BIT(HYPERV_FEAT_STIMER) }, + [HYPERV_FEAT_DIRECT_TLBFLUSH] = { + .desc = "direct paravirtualized TLB flush (hv-direct-tlbflush)", + .dependencies =BIT (HYPERV_FEAT_TLBFLUSH ) + }, };Quoted 2 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState
Quoted 2 lines
target/i386/kvm.c +0
-0static struct kvm_cpuid2 *try_get_hv_cpuid (CPUState *cs,int max)@@ -1224,6 +1228,7 @@ static int hyperv_handle_properties(CPUState*cs,
Quoted 6 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState · line 1233
Quoted 6 lines
target/i386/kvm.c +2
-0r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_EVMCS ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_IPI ); r |=hv_cpuid_check_and_set (cs, cpuid,HYPERV_FEAT_STIMER_DIRECT ); + r |=hv_cpuid_check_and_set (cs, cpuid, +HYPERV_FEAT_DIRECT_TLBFLUSH );
Quoted 2 lines
AFAICS this will turn
HYPERV_FEAT_DIRECT_TLBFLUSHon ifhyperv_passthroughis on, so ...
Yes.
Quoted 61 lines
Quoted 3 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState · line 1234
Quoted 3 lines
target/i386/kvm.c +0
-0/* Additional dependencies not covered by kvm_hyperv_properties[] */
Quoted 7 lines
target/i386/kvm.c static int hyperv_handle_properties(CPUState
Quoted 7 lines
target/i386/kvm.c +3
-0if (hyperv_feat_enabled(cpu, HYPERV_FEAT_SYNIC) && @@ -1243,6 +1248,25 @@ static int hyperv_handle_properties(CPUState *cs, goto free; } +if (hyperv_feat_enabled (cpu,HYPERV_FEAT_DIRECT_TLBFLUSH ) || + cpu->hyperv_passthrough) {... the test for ->hyperv_passthrough is redundant, and ...
Quoted 2 lines
target/i386/kvm.c
Quoted 2 lines
target/i386/kvm.c +2
-0+if (!cpu->expose_kvm) { + r = kvm_vcpu_enable_cap(cs,
KVM_CAP_HYPERV_DIRECT_TLBFLUSH, 0, 0);Quoted 2 lines
Quoted 2 lines
+ if (hyperv_feat_enabled(cpu,
+HYPERV_FEAT_DIRECT_TLBFLUSH) && r) {... , more importantly, this will abort QEMU if
HYPERV_FEAT_DIRECT_TLBFLUSHwasn't requested explicitly, but was activated by ->hyperv_passthrough, and setting the capability failed. I think the meaning ofhyperv_passthroughis "enable all hyperv features supported by the KVM", so in this case it looks more correct to just clear the feature bit and go ahead.Quoted 3 lines
target/i386/kvm.c
Quoted 3 lines
target/i386/kvm.c +3
-0+ fprintf(stderr, + "Hyper-V %s is not supported by kernel\n", +kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH].desc);
Quoted 7 lines
target/i386/kvm.c
Quoted 7 lines
target/i386/kvm.c +7
-0+return -ENOSYS ; + } + }else if (!cpu->hyperv_passthrough) { + fprintf(stderr, + "Hyper-V %s requires KVM hypervisor signature " + "to be hidden (-kvm).\n", +kvm_hyperv_properties[HYPERV_FEAT_DIRECT_TLBFLUSH].desc);
Quoted 2 lines
target/i386/kvm.c
Quoted 2 lines
target/i386/kvm.c +2
-0+return -ENOSYS ; + }You reach here if ->expose_kvm && ->hyperv_passthrough, and no capability is activated, and you go ahead with the feature bit set. This doesn't look right either.
So in general it should probably look like
if (hyperv_feat_enabled(HYPERV_FEAT_DIRECT_TLBFLUSH)) { if (kvm_vcpu_enable_cap(KVM_CAP_HYPERV_DIRECT_TLBFLUSH)) { if (!cpu->hyperv_passthrough) { ... report feature unsupported by kernel ... return -ENOSYS; } cpu->hyperv_features &= ~BIT(HYPERV_FEAT_DIRECT_TLBFLUSH); } else if (cpu->expose_kvm) { ... report conflict ... return -ENOSYS; } }
Quoted 1 line
[Yes,
hyperv_passthroughhurts, but you've been warned ;)]
Yes, you are right, I ignore HYPERV_FEAT_DIRECT_TLBFLUSH bit was set in the hv_cpuid_check_and_set((). Will update in the next version. Thanks.
Quoted 3 lines
Thanks,
Roman.