From ea812ca1b06113597adcd8e70c0f84a413d97544 Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Tue, 29 Jun 2010 18:38:00 +0000 Subject: x86: Align skb w/ start of cacheline on newer core 2/Xeon Arch x86 architectures can handle unaligned accesses in hardware, and it has been shown that unaligned DMA accesses can be expensive on Nehalem architectures. As such we should overwrite NET_IP_ALIGN to resolve this issue. Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org Signed-off-by: Alexander Duyck Signed-off-by: Jeff Kirsher Acked-by: H. Peter Anvin Signed-off-by: David S. Miller --- arch/x86/include/asm/system.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h index b8fe48ee2ed9..b4293fc8b798 100644 --- a/arch/x86/include/asm/system.h +++ b/arch/x86/include/asm/system.h @@ -457,4 +457,13 @@ static inline void rdtsc_barrier(void) alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC); } +#ifdef CONFIG_MCORE2 +/* + * We handle most unaligned accesses in hardware. On the other hand + * unaligned DMA can be quite expensive on some Nehalem processors. + * + * Based on this we disable the IP header alignment in network drivers. + */ +#define NET_IP_ALIGN 0 +#endif #endif /* _ASM_X86_SYSTEM_H */ -- cgit v1.2.3 From 7475271004b66e9c22e1bb28f240a38c5d6fe76e Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 1 Jul 2010 13:28:27 +0000 Subject: x86: Drop CONFIG_MCORE2 check around setting of NET_IP_ALIGN This patch removes the CONFIG_MCORE2 check from around NET_IP_ALIGN. It is based on a suggestion from Andi Kleen. The assumption is that there are not any x86 cores where unaligned access is really slow, and this change would allow for a performance improvement to still exist on configurations that are not necessarily optimized for Core 2. Cc: Andi Kleen Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org Signed-off-by: Alexander Duyck Signed-off-by: Jeff Kirsher Acked-by: H. Peter Anvin Signed-off-by: David S. Miller --- arch/x86/include/asm/system.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h index b4293fc8b798..1db9bd2281d7 100644 --- a/arch/x86/include/asm/system.h +++ b/arch/x86/include/asm/system.h @@ -457,7 +457,6 @@ static inline void rdtsc_barrier(void) alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC); } -#ifdef CONFIG_MCORE2 /* * We handle most unaligned accesses in hardware. On the other hand * unaligned DMA can be quite expensive on some Nehalem processors. @@ -465,5 +464,4 @@ static inline void rdtsc_barrier(void) * Based on this we disable the IP header alignment in network drivers. */ #define NET_IP_ALIGN 0 -#endif #endif /* _ASM_X86_SYSTEM_H */ -- cgit v1.2.3 From c516ac583973196162b1ba7e4d597d6f6892dac0 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Mon, 5 Jul 2010 23:03:18 +0800 Subject: x86: Add i8042 pre-detection hook to x86_platform_ops Some x86 platforms like Intel MID platforms don't have i8042 controllers, and i8042 driver's probe to some legacy IO ports may hang the MID processor. With this hook, i8042 driver can runtime check and skip the probe when the pretection fail which also saves some probe time [ hpa note: this is currently a compile-time check, which breaks the i386 allyesconfig build. This patch series thus does fix a regression. ] Signed-off-by: Feng Tang LKML-Reference: <1278342202-10973-2-git-send-email-feng.tang@intel.com> Acked-by: Dmitry Torokhov Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/x86_init.h | 2 ++ arch/x86/kernel/x86_init.c | 4 +++- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index 519b54327d75..baa579c8e038 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -142,6 +142,7 @@ struct x86_cpuinit_ops { * @set_wallclock: set time back to HW clock * @is_untracked_pat_range exclude from PAT logic * @nmi_init enable NMI on cpus + * @i8042_detect pre-detect if i8042 controller exists */ struct x86_platform_ops { unsigned long (*calibrate_tsc)(void); @@ -150,6 +151,7 @@ struct x86_platform_ops { void (*iommu_shutdown)(void); bool (*is_untracked_pat_range)(u64 start, u64 end); void (*nmi_init)(void); + int (*i8042_detect)(void); }; extern struct x86_init_ops x86_init; diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index ebfb8e4c9f22..cd6da6bf3eca 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -86,6 +86,7 @@ struct x86_cpuinit_ops x86_cpuinit __cpuinitdata = { }; static void default_nmi_init(void) { }; +static int default_i8042_detect(void) { return 1; }; struct x86_platform_ops x86_platform = { .calibrate_tsc = native_calibrate_tsc, @@ -93,7 +94,8 @@ struct x86_platform_ops x86_platform = { .set_wallclock = mach_set_rtc_mmss, .iommu_shutdown = iommu_shutdown_noop, .is_untracked_pat_range = is_ISA_range, - .nmi_init = default_nmi_init + .nmi_init = default_nmi_init, + .i8042_detect = default_i8042_detect }; EXPORT_SYMBOL_GPL(x86_platform); -- cgit v1.2.3 From 18f19aa62a267f2f759e278018f1032adf4c3774 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 14 May 2010 12:38:24 +0100 Subject: xen: Add support for HVM hypercalls. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Sheng Yang Signed-off-by: Stefano Stabellini --- arch/x86/include/asm/xen/hypercall.h | 6 +++ include/xen/hvm.h | 24 +++++++++ include/xen/interface/hvm/hvm_op.h | 35 +++++++++++++ include/xen/interface/hvm/params.h | 95 ++++++++++++++++++++++++++++++++++++ 4 files changed, 160 insertions(+) create mode 100644 include/xen/hvm.h create mode 100644 include/xen/interface/hvm/hvm_op.h create mode 100644 include/xen/interface/hvm/params.h (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index 9c371e4a9fa6..7fda040a76cd 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -417,6 +417,12 @@ HYPERVISOR_nmi_op(unsigned long op, unsigned long arg) return _hypercall2(int, nmi_op, op, arg); } +static inline unsigned long __must_check +HYPERVISOR_hvm_op(int op, void *arg) +{ + return _hypercall2(unsigned long, hvm_op, op, arg); +} + static inline void MULTI_fpu_taskswitch(struct multicall_entry *mcl, int set) { diff --git a/include/xen/hvm.h b/include/xen/hvm.h new file mode 100644 index 000000000000..5dfe8fb86e67 --- /dev/null +++ b/include/xen/hvm.h @@ -0,0 +1,24 @@ +/* Simple wrappers around HVM functions */ +#ifndef XEN_HVM_H__ +#define XEN_HVM_H__ + +#include + +static inline int hvm_get_parameter(int idx, uint64_t *value) +{ + struct xen_hvm_param xhv; + int r; + + xhv.domid = DOMID_SELF; + xhv.index = idx; + r = HYPERVISOR_hvm_op(HVMOP_get_param, &xhv); + if (r < 0) { + printk(KERN_ERR "Cannot get hvm parameter %d: %d!\n", + idx, r); + return r; + } + *value = xhv.value; + return r; +} + +#endif /* XEN_HVM_H__ */ diff --git a/include/xen/interface/hvm/hvm_op.h b/include/xen/interface/hvm/hvm_op.h new file mode 100644 index 000000000000..73c8c7eba48d --- /dev/null +++ b/include/xen/interface/hvm/hvm_op.h @@ -0,0 +1,35 @@ +/* + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + +#ifndef __XEN_PUBLIC_HVM_HVM_OP_H__ +#define __XEN_PUBLIC_HVM_HVM_OP_H__ + +/* Get/set subcommands: the second argument of the hypercall is a + * pointer to a xen_hvm_param struct. */ +#define HVMOP_set_param 0 +#define HVMOP_get_param 1 +struct xen_hvm_param { + domid_t domid; /* IN */ + uint32_t index; /* IN */ + uint64_t value; /* IN/OUT */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_hvm_param); + +#endif /* __XEN_PUBLIC_HVM_HVM_OP_H__ */ diff --git a/include/xen/interface/hvm/params.h b/include/xen/interface/hvm/params.h new file mode 100644 index 000000000000..1888d8c157e6 --- /dev/null +++ b/include/xen/interface/hvm/params.h @@ -0,0 +1,95 @@ +/* + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + */ + +#ifndef __XEN_PUBLIC_HVM_PARAMS_H__ +#define __XEN_PUBLIC_HVM_PARAMS_H__ + +#include "hvm_op.h" + +/* + * Parameter space for HVMOP_{set,get}_param. + */ + +/* + * How should CPU0 event-channel notifications be delivered? + * val[63:56] == 0: val[55:0] is a delivery GSI (Global System Interrupt). + * val[63:56] == 1: val[55:0] is a delivery PCI INTx line, as follows: + * Domain = val[47:32], Bus = val[31:16], + * DevFn = val[15: 8], IntX = val[ 1: 0] + * val[63:56] == 2: val[7:0] is a vector number. + * If val == 0 then CPU0 event-channel notifications are not delivered. + */ +#define HVM_PARAM_CALLBACK_IRQ 0 + +#define HVM_PARAM_STORE_PFN 1 +#define HVM_PARAM_STORE_EVTCHN 2 + +#define HVM_PARAM_PAE_ENABLED 4 + +#define HVM_PARAM_IOREQ_PFN 5 + +#define HVM_PARAM_BUFIOREQ_PFN 6 + +/* + * Set mode for virtual timers (currently x86 only): + * delay_for_missed_ticks (default): + * Do not advance a vcpu's time beyond the correct delivery time for + * interrupts that have been missed due to preemption. Deliver missed + * interrupts when the vcpu is rescheduled and advance the vcpu's virtual + * time stepwise for each one. + * no_delay_for_missed_ticks: + * As above, missed interrupts are delivered, but guest time always tracks + * wallclock (i.e., real) time while doing so. + * no_missed_ticks_pending: + * No missed interrupts are held pending. Instead, to ensure ticks are + * delivered at some non-zero rate, if we detect missed ticks then the + * internal tick alarm is not disabled if the VCPU is preempted during the + * next tick period. + * one_missed_tick_pending: + * Missed interrupts are collapsed together and delivered as one 'late tick'. + * Guest time always tracks wallclock (i.e., real) time. + */ +#define HVM_PARAM_TIMER_MODE 10 +#define HVMPTM_delay_for_missed_ticks 0 +#define HVMPTM_no_delay_for_missed_ticks 1 +#define HVMPTM_no_missed_ticks_pending 2 +#define HVMPTM_one_missed_tick_pending 3 + +/* Boolean: Enable virtual HPET (high-precision event timer)? (x86-only) */ +#define HVM_PARAM_HPET_ENABLED 11 + +/* Identity-map page directory used by Intel EPT when CR0.PG=0. */ +#define HVM_PARAM_IDENT_PT 12 + +/* Device Model domain, defaults to 0. */ +#define HVM_PARAM_DM_DOMAIN 13 + +/* ACPI S state: currently support S0 and S3 on x86. */ +#define HVM_PARAM_ACPI_S_STATE 14 + +/* TSS used on Intel when CR0.PE=0. */ +#define HVM_PARAM_VM86_TSS 15 + +/* Boolean: Enable aligning all periodic vpts to reduce interrupts */ +#define HVM_PARAM_VPT_ALIGN 16 + +#define HVM_NR_PARAMS 17 + +#endif /* __XEN_PUBLIC_HVM_PARAMS_H__ */ -- cgit v1.2.3 From bee6ab53e652a414af20392899879b58cd80d033 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Fri, 14 May 2010 12:39:33 +0100 Subject: x86: early PV on HVM features initialization. Initialize basic pv on hvm features adding a new Xen HVM specific hypervisor_x86 structure. Don't try to initialize xen-kbdfront and xen-fbfront when running on HVM because the backends are not available. Signed-off-by: Stefano Stabellini Signed-off-by: Sheng Yang Signed-off-by: Yaozu (Eddie) Dong Signed-off-by: Jeremy Fitzhardinge --- arch/x86/include/asm/hypervisor.h | 1 + arch/x86/kernel/cpu/hypervisor.c | 1 + arch/x86/xen/enlighten.c | 100 ++++++++++++++++++++++++++++++++++++++ drivers/input/xen-kbdfront.c | 2 +- drivers/video/xen-fbfront.c | 2 +- drivers/xen/xenbus/xenbus_probe.c | 21 ++++++-- 6 files changed, 122 insertions(+), 5 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/hypervisor.h b/arch/x86/include/asm/hypervisor.h index 70abda7058c8..ff2546ce7178 100644 --- a/arch/x86/include/asm/hypervisor.h +++ b/arch/x86/include/asm/hypervisor.h @@ -45,5 +45,6 @@ extern const struct hypervisor_x86 *x86_hyper; /* Recognized hypervisors */ extern const struct hypervisor_x86 x86_hyper_vmware; extern const struct hypervisor_x86 x86_hyper_ms_hyperv; +extern const struct hypervisor_x86 x86_hyper_xen_hvm; #endif diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c index dd531cc56a8f..bffd47c10fed 100644 --- a/arch/x86/kernel/cpu/hypervisor.c +++ b/arch/x86/kernel/cpu/hypervisor.c @@ -34,6 +34,7 @@ static const __initconst struct hypervisor_x86 * const hypervisors[] = { &x86_hyper_vmware, &x86_hyper_ms_hyperv, + &x86_hyper_xen_hvm, }; const struct hypervisor_x86 *x86_hyper; diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 65d8d79b46a8..09b36e9d507a 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include @@ -55,7 +56,9 @@ #include #include #include +#include #include +#include #include "xen-ops.h" #include "mmu.h" @@ -76,6 +79,8 @@ struct shared_info xen_dummy_shared_info; void *xen_initial_gdt; +RESERVE_BRK(shared_info_page_brk, PAGE_SIZE); + /* * Point at some empty memory to start with. We map the real shared_info * page as soon as fixmap is up and running. @@ -1206,3 +1211,98 @@ asmlinkage void __init xen_start_kernel(void) x86_64_start_reservations((char *)__pa_symbol(&boot_params)); #endif } + +static uint32_t xen_cpuid_base(void) +{ + uint32_t base, eax, ebx, ecx, edx; + char signature[13]; + + for (base = 0x40000000; base < 0x40010000; base += 0x100) { + cpuid(base, &eax, &ebx, &ecx, &edx); + *(uint32_t *)(signature + 0) = ebx; + *(uint32_t *)(signature + 4) = ecx; + *(uint32_t *)(signature + 8) = edx; + signature[12] = 0; + + if (!strcmp("XenVMMXenVMM", signature) && ((eax - base) >= 2)) + return base; + } + + return 0; +} + +static int init_hvm_pv_info(int *major, int *minor) +{ + uint32_t eax, ebx, ecx, edx, pages, msr, base; + u64 pfn; + + base = xen_cpuid_base(); + cpuid(base + 1, &eax, &ebx, &ecx, &edx); + + *major = eax >> 16; + *minor = eax & 0xffff; + printk(KERN_INFO "Xen version %d.%d.\n", *major, *minor); + + cpuid(base + 2, &pages, &msr, &ecx, &edx); + + pfn = __pa(hypercall_page); + wrmsr_safe(msr, (u32)pfn, (u32)(pfn >> 32)); + + xen_setup_features(); + + pv_info = xen_info; + pv_info.kernel_rpl = 0; + + xen_domain_type = XEN_HVM_DOMAIN; + + return 0; +} + +static void __init init_shared_info(void) +{ + struct xen_add_to_physmap xatp; + struct shared_info *shared_info_page; + + shared_info_page = (struct shared_info *) + extend_brk(PAGE_SIZE, PAGE_SIZE); + xatp.domid = DOMID_SELF; + xatp.idx = 0; + xatp.space = XENMAPSPACE_shared_info; + xatp.gpfn = __pa(shared_info_page) >> PAGE_SHIFT; + if (HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp)) + BUG(); + + HYPERVISOR_shared_info = (struct shared_info *)shared_info_page; + + per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; +} + +static void __init xen_hvm_guest_init(void) +{ + int r; + int major, minor; + + r = init_hvm_pv_info(&major, &minor); + if (r < 0) + return; + + init_shared_info(); +} + +static bool __init xen_hvm_platform(void) +{ + if (xen_pv_domain()) + return false; + + if (!xen_cpuid_base()) + return false; + + return true; +} + +const __refconst struct hypervisor_x86 x86_hyper_xen_hvm = { + .name = "Xen HVM", + .detect = xen_hvm_platform, + .init_platform = xen_hvm_guest_init, +}; +EXPORT_SYMBOL(x86_hyper_xen_hvm); diff --git a/drivers/input/xen-kbdfront.c b/drivers/input/xen-kbdfront.c index e14081675bb2..ebb11907d402 100644 --- a/drivers/input/xen-kbdfront.c +++ b/drivers/input/xen-kbdfront.c @@ -339,7 +339,7 @@ static struct xenbus_driver xenkbd_driver = { static int __init xenkbd_init(void) { - if (!xen_domain()) + if (!xen_pv_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c index fa97d3e7c21a..7c7f42a12796 100644 --- a/drivers/video/xen-fbfront.c +++ b/drivers/video/xen-fbfront.c @@ -684,7 +684,7 @@ static struct xenbus_driver xenfb_driver = { static int __init xenfb_init(void) { - if (!xen_domain()) + if (!xen_pv_domain()) return -ENODEV; /* Nothing to do if running in dom0. */ diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 3479332113e9..d96fa75b45ec 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -56,6 +56,8 @@ #include #include +#include + #include "xenbus_comms.h" #include "xenbus_probe.h" @@ -805,11 +807,24 @@ static int __init xenbus_probe_init(void) if (xen_initial_domain()) { /* dom0 not yet supported */ } else { + if (xen_hvm_domain()) { + uint64_t v = 0; + err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v); + if (err) + goto out_error; + xen_store_evtchn = (int)v; + err = hvm_get_parameter(HVM_PARAM_STORE_PFN, &v); + if (err) + goto out_error; + xen_store_mfn = (unsigned long)v; + xen_store_interface = ioremap(xen_store_mfn << PAGE_SHIFT, PAGE_SIZE); + } else { + xen_store_evtchn = xen_start_info->store_evtchn; + xen_store_mfn = xen_start_info->store_mfn; + xen_store_interface = mfn_to_virt(xen_store_mfn); + } xenstored_ready = 1; - xen_store_evtchn = xen_start_info->store_evtchn; - xen_store_mfn = xen_start_info->store_mfn; } - xen_store_interface = mfn_to_virt(xen_store_mfn); /* Initialize the interface to xenstore. */ err = xs_init(); -- cgit v1.2.3 From 38e20b07efd541a959de367dc90a17f92ce2e8a6 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Fri, 14 May 2010 12:40:51 +0100 Subject: x86/xen: event channels delivery on HVM. Set the callback to receive evtchns from Xen, using the callback vector delivery mechanism. The traditional way for receiving event channel notifications from Xen is via the interrupts from the platform PCI device. The callback vector is a newer alternative that allow us to receive notifications on any vcpu and doesn't need any PCI support: we allocate a vector exclusively to receive events, in the vector handler we don't need to interact with the vlapic, therefore we avoid a VMEXIT. Signed-off-by: Stefano Stabellini Signed-off-by: Sheng Yang Signed-off-by: Jeremy Fitzhardinge --- arch/x86/include/asm/irq_vectors.h | 3 ++ arch/x86/kernel/entry_32.S | 3 ++ arch/x86/kernel/entry_64.S | 3 ++ arch/x86/xen/enlighten.c | 28 +++++++++++++++ arch/x86/xen/xen-ops.h | 2 ++ drivers/xen/events.c | 70 ++++++++++++++++++++++++++++++++++---- include/xen/events.h | 7 ++++ include/xen/hvm.h | 6 ++++ include/xen/interface/features.h | 3 ++ 9 files changed, 118 insertions(+), 7 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h index 8767d99c4f64..e2ca30092557 100644 --- a/arch/x86/include/asm/irq_vectors.h +++ b/arch/x86/include/asm/irq_vectors.h @@ -125,6 +125,9 @@ */ #define MCE_SELF_VECTOR 0xeb +/* Xen vector callback to receive events in a HVM domain */ +#define XEN_HVM_EVTCHN_CALLBACK 0xe9 + #define NR_VECTORS 256 #define FPU_IRQ 13 diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index cd49141cf153..6b196834a0dd 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -1166,6 +1166,9 @@ ENTRY(xen_failsafe_callback) .previous ENDPROC(xen_failsafe_callback) +BUILD_INTERRUPT3(xen_hvm_callback_vector, XEN_HVM_EVTCHN_CALLBACK, + xen_evtchn_do_upcall) + #endif /* CONFIG_XEN */ #ifdef CONFIG_FUNCTION_TRACER diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 0697ff139837..490ae2bb18a8 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1329,6 +1329,9 @@ ENTRY(xen_failsafe_callback) CFI_ENDPROC END(xen_failsafe_callback) +apicinterrupt XEN_HVM_EVTCHN_CALLBACK \ + xen_hvm_callback_vector xen_evtchn_do_upcall + #endif /* CONFIG_XEN */ /* diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 09b36e9d507a..b211a04c4b2c 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -11,6 +11,7 @@ * Jeremy Fitzhardinge , XenSource Inc, 2007 */ +#include #include #include #include @@ -38,6 +39,7 @@ #include #include #include +#include #include #include @@ -80,6 +82,8 @@ struct shared_info xen_dummy_shared_info; void *xen_initial_gdt; RESERVE_BRK(shared_info_page_brk, PAGE_SIZE); +__read_mostly int xen_have_vector_callback; +EXPORT_SYMBOL_GPL(xen_have_vector_callback); /* * Point at some empty memory to start with. We map the real shared_info @@ -1277,6 +1281,24 @@ static void __init init_shared_info(void) per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; } +static int __cpuinit xen_hvm_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + int cpu = (long)hcpu; + switch (action) { + case CPU_UP_PREPARE: + per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu]; + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block __cpuinitdata xen_hvm_cpu_notifier = { + .notifier_call = xen_hvm_cpu_notify, +}; + static void __init xen_hvm_guest_init(void) { int r; @@ -1287,6 +1309,12 @@ static void __init xen_hvm_guest_init(void) return; init_shared_info(); + + if (xen_feature(XENFEAT_hvm_callback_vector)) + xen_have_vector_callback = 1; + register_cpu_notifier(&xen_hvm_cpu_notifier); + have_vcpu_info_placement = 0; + x86_init.irqs.intr_init = xen_init_IRQ; } static bool __init xen_hvm_platform(void) diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index f9153a300bce..0d0e0e6a7479 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -38,6 +38,8 @@ void xen_enable_sysenter(void); void xen_enable_syscall(void); void xen_vcpu_restore(void); +void xen_callback_vector(void); + void __init xen_build_dynamic_phys_to_machine(void); void xen_init_irq_ops(void); diff --git a/drivers/xen/events.c b/drivers/xen/events.c index db8f506817f0..d659480125f0 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -29,6 +29,7 @@ #include #include +#include #include #include #include @@ -36,10 +37,14 @@ #include #include +#include +#include #include #include #include #include +#include +#include /* * This lock protects updates to the following mapping and reference-count @@ -617,17 +622,13 @@ static DEFINE_PER_CPU(unsigned, xed_nesting_count); * a bitset of words which contain pending event bits. The second * level is a bitset of pending events themselves. */ -void xen_evtchn_do_upcall(struct pt_regs *regs) +static void __xen_evtchn_do_upcall(void) { int cpu = get_cpu(); - struct pt_regs *old_regs = set_irq_regs(regs); struct shared_info *s = HYPERVISOR_shared_info; struct vcpu_info *vcpu_info = __get_cpu_var(xen_vcpu); unsigned count; - exit_idle(); - irq_enter(); - do { unsigned long pending_words; @@ -667,10 +668,26 @@ void xen_evtchn_do_upcall(struct pt_regs *regs) } while(count != 1); out: + + put_cpu(); +} + +void xen_evtchn_do_upcall(struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + + exit_idle(); + irq_enter(); + + __xen_evtchn_do_upcall(); + irq_exit(); set_irq_regs(old_regs); +} - put_cpu(); +void xen_hvm_evtchn_do_upcall(void) +{ + __xen_evtchn_do_upcall(); } /* Rebind a new event channel to an existing irq. */ @@ -933,6 +950,40 @@ static struct irq_chip xen_dynamic_chip __read_mostly = { .retrigger = retrigger_dynirq, }; +int xen_set_callback_via(uint64_t via) +{ + struct xen_hvm_param a; + a.domid = DOMID_SELF; + a.index = HVM_PARAM_CALLBACK_IRQ; + a.value = via; + return HYPERVISOR_hvm_op(HVMOP_set_param, &a); +} +EXPORT_SYMBOL_GPL(xen_set_callback_via); + +/* Vector callbacks are better than PCI interrupts to receive event + * channel notifications because we can receive vector callbacks on any + * vcpu and we don't need PCI support or APIC interactions. */ +void xen_callback_vector(void) +{ + int rc; + uint64_t callback_via; + if (xen_have_vector_callback) { + callback_via = HVM_CALLBACK_VECTOR(XEN_HVM_EVTCHN_CALLBACK); + rc = xen_set_callback_via(callback_via); + if (rc) { + printk(KERN_ERR "Request for Xen HVM callback vector" + " failed.\n"); + xen_have_vector_callback = 0; + return; + } + printk(KERN_INFO "Xen HVM callback vector for event delivery is " + "enabled\n"); + /* in the restore case the vector has already been allocated */ + if (!test_bit(XEN_HVM_EVTCHN_CALLBACK, used_vectors)) + alloc_intr_gate(XEN_HVM_EVTCHN_CALLBACK, xen_hvm_callback_vector); + } +} + void __init xen_init_IRQ(void) { int i; @@ -947,5 +998,10 @@ void __init xen_init_IRQ(void) for (i = 0; i < NR_EVENT_CHANNELS; i++) mask_evtchn(i); - irq_ctx_init(smp_processor_id()); + if (xen_hvm_domain()) { + xen_callback_vector(); + native_init_IRQ(); + } else { + irq_ctx_init(smp_processor_id()); + } } diff --git a/include/xen/events.h b/include/xen/events.h index e68d59a90ca8..a15d93262e30 100644 --- a/include/xen/events.h +++ b/include/xen/events.h @@ -56,4 +56,11 @@ void xen_poll_irq(int irq); /* Determine the IRQ which is bound to an event channel */ unsigned irq_from_evtchn(unsigned int evtchn); +/* Xen HVM evtchn vector callback */ +extern void xen_hvm_callback_vector(void); +extern int xen_have_vector_callback; +int xen_set_callback_via(uint64_t via); +void xen_evtchn_do_upcall(struct pt_regs *regs); +void xen_hvm_evtchn_do_upcall(void); + #endif /* _XEN_EVENTS_H */ diff --git a/include/xen/hvm.h b/include/xen/hvm.h index 5dfe8fb86e67..b193fa2f9fdd 100644 --- a/include/xen/hvm.h +++ b/include/xen/hvm.h @@ -3,6 +3,7 @@ #define XEN_HVM_H__ #include +#include static inline int hvm_get_parameter(int idx, uint64_t *value) { @@ -21,4 +22,9 @@ static inline int hvm_get_parameter(int idx, uint64_t *value) return r; } +#define HVM_CALLBACK_VIA_TYPE_VECTOR 0x2 +#define HVM_CALLBACK_VIA_TYPE_SHIFT 56 +#define HVM_CALLBACK_VECTOR(x) (((uint64_t)HVM_CALLBACK_VIA_TYPE_VECTOR)<<\ + HVM_CALLBACK_VIA_TYPE_SHIFT | (x)) + #endif /* XEN_HVM_H__ */ diff --git a/include/xen/interface/features.h b/include/xen/interface/features.h index f51b6413b054..8ab08b91bf6f 100644 --- a/include/xen/interface/features.h +++ b/include/xen/interface/features.h @@ -41,6 +41,9 @@ /* x86: Does this Xen host support the MMU_PT_UPDATE_PRESERVE_AD hypercall? */ #define XENFEAT_mmu_pt_update_preserve_ad 5 +/* x86: Does this Xen host support the HVM callback vector type? */ +#define XENFEAT_hvm_callback_vector 8 + #define XENFEAT_NR_SUBMAPS 1 #endif /* __XEN_PUBLIC_FEATURES_H__ */ -- cgit v1.2.3 From c7f52cdc2f3e1733d3864e439ac2e92edd99ef31 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 22 Jul 2010 22:58:01 -0700 Subject: support multiple .discard.* sections to avoid section type conflicts gcc 4.4.4 will complain if you use a .discard section for both text and data ("causes a section type conflict"). Add support for ".discard.*" sections, and use .discard.text for a dummy function in the x86 RESERVE_BRK() macro. Signed-off-by: Jeremy Fitzhardinge --- arch/x86/include/asm/setup.h | 2 +- include/asm-generic/vmlinux.lds.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h index 86b1506f4179..ef292c792d74 100644 --- a/arch/x86/include/asm/setup.h +++ b/arch/x86/include/asm/setup.h @@ -82,7 +82,7 @@ void *extend_brk(size_t size, size_t align); * executable.) */ #define RESERVE_BRK(name,sz) \ - static void __section(.discard) __used \ + static void __section(.discard.text) __used \ __brk_reservation_fn_##name##__(void) { \ asm volatile ( \ ".pushsection .brk_reservation,\"aw\",@nobits;" \ diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 48c5299cbf26..ae6b88eb1de1 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -643,6 +643,7 @@ EXIT_DATA \ EXIT_CALL \ *(.discard) \ + *(.discard.*) \ } /** -- cgit v1.2.3 From 9de41573675cbace09b02ef386f3e9c8739d495c Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:22 +0300 Subject: KVM: x86 emulator: introduce read cache Introduce read cache which is needed for instruction that require more then one exit to userspace. After returning from userspace the instruction will be re-executed with cached read value. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 1 + arch/x86/kvm/emulate.c | 56 ++++++++++++++++++++++++++++---------- 2 files changed, 43 insertions(+), 14 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 0b2729bf2070..288cbedcab1c 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -186,6 +186,7 @@ struct decode_cache { unsigned long modrm_val; struct fetch_cache fetch; struct read_cache io_read; + struct read_cache mem_read; }; struct x86_emulate_ctxt { diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 5ac0bb465ed6..776874b8e50e 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -1263,6 +1263,33 @@ done: return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; } +static int read_emulated(struct x86_emulate_ctxt *ctxt, + struct x86_emulate_ops *ops, + unsigned long addr, void *dest, unsigned size) +{ + int rc; + struct read_cache *mc = &ctxt->decode.mem_read; + + while (size) { + int n = min(size, 8u); + size -= n; + if (mc->pos < mc->end) + goto read_cached; + + rc = ops->read_emulated(addr, mc->data + mc->end, n, ctxt->vcpu); + if (rc != X86EMUL_CONTINUE) + return rc; + mc->end += n; + + read_cached: + memcpy(dest, mc->data + mc->pos, n); + mc->pos += n; + dest += n; + addr += n; + } + return X86EMUL_CONTINUE; +} + static int pio_in_emulated(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned int size, unsigned short port, @@ -1504,9 +1531,9 @@ static int emulate_pop(struct x86_emulate_ctxt *ctxt, struct decode_cache *c = &ctxt->decode; int rc; - rc = ops->read_emulated(register_address(c, ss_base(ctxt), - c->regs[VCPU_REGS_RSP]), - dest, len, ctxt->vcpu); + rc = read_emulated(ctxt, ops, register_address(c, ss_base(ctxt), + c->regs[VCPU_REGS_RSP]), + dest, len); if (rc != X86EMUL_CONTINUE) return rc; @@ -2475,6 +2502,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) int saved_dst_type = c->dst.type; ctxt->interruptibility = 0; + ctxt->decode.mem_read.pos = 0; /* Shadow copy of register state. Committed on successful emulation. * NOTE: we can copy them from vcpu as x86_decode_insn() doesn't @@ -2529,20 +2557,16 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) } if (c->src.type == OP_MEM) { - rc = ops->read_emulated((unsigned long)c->src.ptr, - &c->src.val, - c->src.bytes, - ctxt->vcpu); + rc = read_emulated(ctxt, ops, (unsigned long)c->src.ptr, + &c->src.val, c->src.bytes); if (rc != X86EMUL_CONTINUE) goto done; c->src.orig_val = c->src.val; } if (c->src2.type == OP_MEM) { - rc = ops->read_emulated((unsigned long)c->src2.ptr, - &c->src2.val, - c->src2.bytes, - ctxt->vcpu); + rc = read_emulated(ctxt, ops, (unsigned long)c->src2.ptr, + &c->src2.val, c->src2.bytes); if (rc != X86EMUL_CONTINUE) goto done; } @@ -2553,8 +2577,8 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) if ((c->dst.type == OP_MEM) && !(c->d & Mov)) { /* optimisation - avoid slow emulated read if Mov */ - rc = ops->read_emulated((unsigned long)c->dst.ptr, &c->dst.val, - c->dst.bytes, ctxt->vcpu); + rc = read_emulated(ctxt, ops, (unsigned long)c->dst.ptr, + &c->dst.val, c->dst.bytes); if (rc != X86EMUL_CONTINUE) goto done; } @@ -2981,7 +3005,11 @@ writeback: (rc->end != 0 && rc->end == rc->pos)) ctxt->restart = false; } - + /* + * reset read cache here in case string instruction is restared + * without decoding + */ + ctxt->decode.mem_read.end = 0; /* Commit shadow register state. */ memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs); kvm_rip_write(ctxt->vcpu, c->eip); -- cgit v1.2.3 From 414e6277fd148f6470261cef50a7fed0d88a2825 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:26 +0300 Subject: KVM: x86 emulator: handle "far address" source operand ljmp/lcall instruction operand contains address and segment. It can be 10 bytes long. Currently we decode it as two different operands. Fix it by introducing new kind of operand that can hold entire far address. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 6 +++- arch/x86/kvm/emulate.c | 56 ++++++++++++++++++++++---------------- 2 files changed, 37 insertions(+), 25 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 288cbedcab1c..69a64a6a36f4 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -143,7 +143,11 @@ struct x86_emulate_ops { struct operand { enum { OP_REG, OP_MEM, OP_IMM, OP_NONE } type; unsigned int bytes; - unsigned long val, orig_val, *ptr; + unsigned long orig_val, *ptr; + union { + unsigned long val; + char valptr[sizeof(unsigned long) + 2]; + }; }; struct fetch_cache { diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 03a72912d7b9..687ea0906b79 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -67,6 +67,8 @@ #define SrcImmUByte (8<<4) /* 8-bit unsigned immediate operand. */ #define SrcImmU (9<<4) /* Immediate operand, unsigned */ #define SrcSI (0xa<<4) /* Source is in the DS:RSI */ +#define SrcImmFAddr (0xb<<4) /* Source is immediate far address */ +#define SrcMemFAddr (0xc<<4) /* Source is far address in memory */ #define SrcMask (0xf<<4) /* Generic ModRM decode. */ #define ModRM (1<<8) @@ -88,10 +90,6 @@ #define Src2CL (1<<29) #define Src2ImmByte (2<<29) #define Src2One (3<<29) -#define Src2Imm16 (4<<29) -#define Src2Mem16 (5<<29) /* Used for Ep encoding. First argument has to be - in memory and second argument is located - immediately after the first one in memory. */ #define Src2Mask (7<<29) enum { @@ -175,7 +173,7 @@ static u32 opcode_table[256] = { /* 0x90 - 0x97 */ DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, DstReg, /* 0x98 - 0x9F */ - 0, 0, SrcImm | Src2Imm16 | No64, 0, + 0, 0, SrcImmFAddr | No64, 0, ImplicitOps | Stack, ImplicitOps | Stack, 0, 0, /* 0xA0 - 0xA7 */ ByteOp | DstReg | SrcMem | Mov | MemAbs, DstReg | SrcMem | Mov | MemAbs, @@ -215,7 +213,7 @@ static u32 opcode_table[256] = { ByteOp | SrcImmUByte | DstAcc, SrcImmUByte | DstAcc, /* 0xE8 - 0xEF */ SrcImm | Stack, SrcImm | ImplicitOps, - SrcImmU | Src2Imm16 | No64, SrcImmByte | ImplicitOps, + SrcImmFAddr | No64, SrcImmByte | ImplicitOps, SrcNone | ByteOp | DstAcc, SrcNone | DstAcc, SrcNone | ByteOp | DstAcc, SrcNone | DstAcc, /* 0xF0 - 0xF7 */ @@ -350,7 +348,7 @@ static u32 group_table[] = { [Group5*8] = DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM, SrcMem | ModRM | Stack, 0, - SrcMem | ModRM | Stack, SrcMem | ModRM | Src2Mem16 | ImplicitOps, + SrcMem | ModRM | Stack, SrcMemFAddr | ModRM | ImplicitOps, SrcMem | ModRM | Stack, 0, [Group7*8] = 0, 0, ModRM | SrcMem | Priv, ModRM | SrcMem | Priv, @@ -576,6 +574,13 @@ static u32 group2_table[] = { (_type)_x; \ }) +#define insn_fetch_arr(_arr, _size, _eip) \ +({ rc = do_insn_fetch(ctxt, ops, (_eip), _arr, (_size)); \ + if (rc != X86EMUL_CONTINUE) \ + goto done; \ + (_eip) += (_size); \ +}) + static inline unsigned long ad_mask(struct decode_cache *c) { return (1UL << (c->ad_bytes << 3)) - 1; @@ -1160,6 +1165,17 @@ done_prefixes: c->regs[VCPU_REGS_RSI]); c->src.val = 0; break; + case SrcImmFAddr: + c->src.type = OP_IMM; + c->src.ptr = (unsigned long *)c->eip; + c->src.bytes = c->op_bytes + 2; + insn_fetch_arr(c->src.valptr, c->src.bytes, c->eip); + break; + case SrcMemFAddr: + c->src.type = OP_MEM; + c->src.ptr = (unsigned long *)c->modrm_ea; + c->src.bytes = c->op_bytes + 2; + break; } /* @@ -1179,22 +1195,10 @@ done_prefixes: c->src2.bytes = 1; c->src2.val = insn_fetch(u8, 1, c->eip); break; - case Src2Imm16: - c->src2.type = OP_IMM; - c->src2.ptr = (unsigned long *)c->eip; - c->src2.bytes = 2; - c->src2.val = insn_fetch(u16, 2, c->eip); - break; case Src2One: c->src2.bytes = 1; c->src2.val = 1; break; - case Src2Mem16: - c->src2.type = OP_MEM; - c->src2.bytes = 2; - c->src2.ptr = (unsigned long *)(c->modrm_ea + c->src.bytes); - c->src2.val = 0; - break; } /* Decode and fetch the destination operand: register or memory. */ @@ -2558,7 +2562,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) if (c->src.type == OP_MEM) { rc = read_emulated(ctxt, ops, (unsigned long)c->src.ptr, - &c->src.val, c->src.bytes); + c->src.valptr, c->src.bytes); if (rc != X86EMUL_CONTINUE) goto done; c->src.orig_val = c->src.val; @@ -2884,14 +2888,18 @@ special_insn: } case 0xe9: /* jmp rel */ goto jmp; - case 0xea: /* jmp far */ + case 0xea: { /* jmp far */ + unsigned short sel; jump_far: - if (load_segment_descriptor(ctxt, ops, c->src2.val, - VCPU_SREG_CS)) + memcpy(&sel, c->src.valptr + c->op_bytes, 2); + + if (load_segment_descriptor(ctxt, ops, sel, VCPU_SREG_CS)) goto done; - c->eip = c->src.val; + c->eip = 0; + memcpy(&c->eip, c->src.valptr, c->op_bytes); break; + } case 0xeb: jmp: /* jmp rel short */ jmp_rel(c, c->src.val); -- cgit v1.2.3 From 35aa5375d407ecadcc3adb5cb31d27044bf7f29f Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:27 +0300 Subject: KVM: x86 emulator: add (set|get)_dr callbacks to x86_emulate_ops Add (set|get)_dr callbacks to x86_emulate_ops instead of calling them directly. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 2 ++ arch/x86/include/asm/kvm_host.h | 4 ---- arch/x86/kvm/emulate.c | 7 +++++-- arch/x86/kvm/x86.c | 12 ++++++------ 4 files changed, 13 insertions(+), 12 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 69a64a6a36f4..c37296d0e909 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -137,6 +137,8 @@ struct x86_emulate_ops { void (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); int (*cpl)(struct kvm_vcpu *vcpu); void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); + int (*get_dr)(int dr, unsigned long *dest, struct kvm_vcpu *vcpu); + int (*set_dr)(int dr, unsigned long value, struct kvm_vcpu *vcpu); }; /* Type, address-of, and value of an instruction's operand. */ diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 76f5483cffec..97774ae3c874 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -591,10 +591,6 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu); int kvm_emulate_halt(struct kvm_vcpu *vcpu); int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address); int emulate_clts(struct kvm_vcpu *vcpu); -int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, - unsigned long *dest); -int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, - unsigned long value); void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg); int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg); diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 687ea0906b79..8a4aa73ff1e4 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -3132,7 +3132,7 @@ twobyte_insn: kvm_queue_exception(ctxt->vcpu, UD_VECTOR); goto done; } - emulator_get_dr(ctxt, c->modrm_reg, &c->regs[c->modrm_rm]); + ops->get_dr(c->modrm_reg, &c->regs[c->modrm_rm], ctxt->vcpu); c->dst.type = OP_NONE; /* no writeback */ break; case 0x22: /* mov reg, cr */ @@ -3145,7 +3145,10 @@ twobyte_insn: kvm_queue_exception(ctxt->vcpu, UD_VECTOR); goto done; } - emulator_set_dr(ctxt, c->modrm_reg, c->regs[c->modrm_rm]); + + ops->set_dr(c->modrm_reg,c->regs[c->modrm_rm] & + ((ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U), + ctxt->vcpu); c->dst.type = OP_NONE; /* no writeback */ break; case 0x30: diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 801afc6461ed..059d63de169b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3620,16 +3620,14 @@ int emulate_clts(struct kvm_vcpu *vcpu) return X86EMUL_CONTINUE; } -int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long *dest) +int emulator_get_dr(int dr, unsigned long *dest, struct kvm_vcpu *vcpu) { - return kvm_get_dr(ctxt->vcpu, dr, dest); + return kvm_get_dr(vcpu, dr, dest); } -int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) +int emulator_set_dr(int dr, unsigned long value, struct kvm_vcpu *vcpu) { - unsigned long mask = (ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U; - - return kvm_set_dr(ctxt->vcpu, dr, value & mask); + return kvm_set_dr(vcpu, dr, value); } void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) @@ -3811,6 +3809,8 @@ static struct x86_emulate_ops emulate_ops = { .set_cr = emulator_set_cr, .cpl = emulator_get_cpl, .set_rflags = emulator_set_rflags, + .get_dr = emulator_get_dr, + .set_dr = emulator_set_dr, }; static void cache_all_regs(struct kvm_vcpu *vcpu) -- cgit v1.2.3 From 3fb1b5dbd397d16a855c97c3fb80fe6e9196ce7c Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:28 +0300 Subject: KVM: x86 emulator: add (set|get)_msr callbacks to x86_emulate_ops Add (set|get)_msr callbacks to x86_emulate_ops instead of calling them directly. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 2 ++ arch/x86/kvm/emulate.c | 36 ++++++++++++++++++------------------ arch/x86/kvm/x86.c | 2 ++ 3 files changed, 22 insertions(+), 18 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index c37296d0e909..f751657be732 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -139,6 +139,8 @@ struct x86_emulate_ops { void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); int (*get_dr)(int dr, unsigned long *dest, struct kvm_vcpu *vcpu); int (*set_dr)(int dr, unsigned long value, struct kvm_vcpu *vcpu); + int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); + int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata); }; /* Type, address-of, and value of an instruction's operand. */ diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 8a4aa73ff1e4..7c8ed560fd41 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -1875,7 +1875,7 @@ setup_syscalls_segments(struct x86_emulate_ctxt *ctxt, } static int -emulate_syscall(struct x86_emulate_ctxt *ctxt) +emulate_syscall(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { struct decode_cache *c = &ctxt->decode; struct kvm_segment cs, ss; @@ -1890,7 +1890,7 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt) setup_syscalls_segments(ctxt, &cs, &ss); - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); msr_data >>= 32; cs.selector = (u16)(msr_data & 0xfffc); ss.selector = (u16)(msr_data + 8); @@ -1907,17 +1907,17 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt) #ifdef CONFIG_X86_64 c->regs[VCPU_REGS_R11] = ctxt->eflags & ~EFLG_RF; - kvm_x86_ops->get_msr(ctxt->vcpu, - ctxt->mode == X86EMUL_MODE_PROT64 ? - MSR_LSTAR : MSR_CSTAR, &msr_data); + ops->get_msr(ctxt->vcpu, + ctxt->mode == X86EMUL_MODE_PROT64 ? + MSR_LSTAR : MSR_CSTAR, &msr_data); c->eip = msr_data; - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_SYSCALL_MASK, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_SYSCALL_MASK, &msr_data); ctxt->eflags &= ~(msr_data | EFLG_RF); #endif } else { /* legacy mode */ - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); c->eip = (u32)msr_data; ctxt->eflags &= ~(EFLG_VM | EFLG_IF | EFLG_RF); @@ -1927,7 +1927,7 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt) } static int -emulate_sysenter(struct x86_emulate_ctxt *ctxt) +emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { struct decode_cache *c = &ctxt->decode; struct kvm_segment cs, ss; @@ -1949,7 +1949,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt) setup_syscalls_segments(ctxt, &cs, &ss); - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); switch (ctxt->mode) { case X86EMUL_MODE_PROT32: if ((msr_data & 0xfffc) == 0x0) { @@ -1979,17 +1979,17 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt) kvm_x86_ops->set_segment(ctxt->vcpu, &cs, VCPU_SREG_CS); kvm_x86_ops->set_segment(ctxt->vcpu, &ss, VCPU_SREG_SS); - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_EIP, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_EIP, &msr_data); c->eip = msr_data; - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_ESP, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_ESP, &msr_data); c->regs[VCPU_REGS_RSP] = msr_data; return X86EMUL_CONTINUE; } static int -emulate_sysexit(struct x86_emulate_ctxt *ctxt) +emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { struct decode_cache *c = &ctxt->decode; struct kvm_segment cs, ss; @@ -2012,7 +2012,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt) cs.dpl = 3; ss.dpl = 3; - kvm_x86_ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); + ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); switch (usermode) { case X86EMUL_MODE_PROT32: cs.selector = (u16)(msr_data + 16); @@ -3099,7 +3099,7 @@ twobyte_insn: } break; case 0x05: /* syscall */ - rc = emulate_syscall(ctxt); + rc = emulate_syscall(ctxt, ops); if (rc != X86EMUL_CONTINUE) goto done; else @@ -3155,7 +3155,7 @@ twobyte_insn: /* wrmsr */ msr_data = (u32)c->regs[VCPU_REGS_RAX] | ((u64)c->regs[VCPU_REGS_RDX] << 32); - if (kvm_set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data)) { + if (ops->set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data)) { kvm_inject_gp(ctxt->vcpu, 0); goto done; } @@ -3164,7 +3164,7 @@ twobyte_insn: break; case 0x32: /* rdmsr */ - if (kvm_get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data)) { + if (ops->get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data)) { kvm_inject_gp(ctxt->vcpu, 0); goto done; } else { @@ -3175,14 +3175,14 @@ twobyte_insn: c->dst.type = OP_NONE; break; case 0x34: /* sysenter */ - rc = emulate_sysenter(ctxt); + rc = emulate_sysenter(ctxt, ops); if (rc != X86EMUL_CONTINUE) goto done; else goto writeback; break; case 0x35: /* sysexit */ - rc = emulate_sysexit(ctxt); + rc = emulate_sysexit(ctxt, ops); if (rc != X86EMUL_CONTINUE) goto done; else diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 059d63de169b..e3a5455049b0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3811,6 +3811,8 @@ static struct x86_emulate_ops emulate_ops = { .set_rflags = emulator_set_rflags, .get_dr = emulator_get_dr, .set_dr = emulator_set_dr, + .set_msr = kvm_set_msr, + .get_msr = kvm_get_msr, }; static void cache_all_regs(struct kvm_vcpu *vcpu) -- cgit v1.2.3 From 5951c4423724759906b10a26aa6a8817c4afa615 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:29 +0300 Subject: KVM: x86 emulator: add get_cached_segment_base() callback to x86_emulate_ops On VMX it is expensive to call get_cached_descriptor() just to get segment base since multiple vmcs_reads are done instead of only one. Introduce new call back get_cached_segment_base() for efficiency. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 1 + arch/x86/kvm/emulate.c | 13 +------------ arch/x86/kvm/x86.c | 7 +++++++ 3 files changed, 9 insertions(+), 12 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index f751657be732..df53ba2294b6 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -132,6 +132,7 @@ struct x86_emulate_ops { int seg, struct kvm_vcpu *vcpu); u16 (*get_segment_selector)(int seg, struct kvm_vcpu *vcpu); void (*set_segment_selector)(u16 sel, int seg, struct kvm_vcpu *vcpu); + unsigned long (*get_cached_segment_base)(int seg, struct kvm_vcpu *vcpu); void (*get_gdt)(struct desc_ptr *dt, struct kvm_vcpu *vcpu); ulong (*get_cr)(int cr, struct kvm_vcpu *vcpu); void (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 7c8ed560fd41..8228778ace38 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -2097,17 +2097,6 @@ static bool emulator_io_permited(struct x86_emulate_ctxt *ctxt, return true; } -static u32 get_cached_descriptor_base(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops, - int seg) -{ - struct desc_struct desc; - if (ops->get_cached_descriptor(&desc, seg, ctxt->vcpu)) - return get_desc_base(&desc); - else - return ~0; -} - static void save_state_to_tss16(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, struct tss_segment_16 *tss) @@ -2383,7 +2372,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, int ret; u16 old_tss_sel = ops->get_segment_selector(VCPU_SREG_TR, ctxt->vcpu); ulong old_tss_base = - get_cached_descriptor_base(ctxt, ops, VCPU_SREG_TR); + ops->get_cached_segment_base(VCPU_SREG_TR, ctxt->vcpu); u32 desc_limit; /* FIXME: old_tss_base == ~0 ? */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e3a5455049b0..9a469df6011c 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3714,6 +3714,12 @@ static void emulator_get_gdt(struct desc_ptr *dt, struct kvm_vcpu *vcpu) kvm_x86_ops->get_gdt(vcpu, dt); } +static unsigned long emulator_get_cached_segment_base(int seg, + struct kvm_vcpu *vcpu) +{ + return get_segment_base(vcpu, seg); +} + static bool emulator_get_cached_descriptor(struct desc_struct *desc, int seg, struct kvm_vcpu *vcpu) { @@ -3804,6 +3810,7 @@ static struct x86_emulate_ops emulate_ops = { .set_cached_descriptor = emulator_set_cached_descriptor, .get_segment_selector = emulator_get_segment_selector, .set_segment_selector = emulator_set_segment_selector, + .get_cached_segment_base = emulator_get_cached_segment_base, .get_gdt = emulator_get_gdt, .get_cr = emulator_get_cr, .set_cr = emulator_set_cr, -- cgit v1.2.3 From 0f12244fe70e8a94a491f6cd7ed70a352ab6c26c Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:31 +0300 Subject: KVM: x86 emulator: make set_cr() callback return error if it fails Make set_cr() callback return error if it fails instead of injecting #GP behind emulator's back. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 2 +- arch/x86/kvm/emulate.c | 10 ++- arch/x86/kvm/x86.c | 148 +++++++++++++++++++------------------ 3 files changed, 84 insertions(+), 76 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index df53ba2294b6..6c4f4918db5e 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -135,7 +135,7 @@ struct x86_emulate_ops { unsigned long (*get_cached_segment_base)(int seg, struct kvm_vcpu *vcpu); void (*get_gdt)(struct desc_ptr *dt, struct kvm_vcpu *vcpu); ulong (*get_cr)(int cr, struct kvm_vcpu *vcpu); - void (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); + int (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); int (*cpl)(struct kvm_vcpu *vcpu); void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); int (*get_dr)(int dr, unsigned long *dest, struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index f56ec486393e..061f7d37c9f7 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -2272,7 +2272,10 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, struct decode_cache *c = &ctxt->decode; int ret; - ops->set_cr(3, tss->cr3, ctxt->vcpu); + if (ops->set_cr(3, tss->cr3, ctxt->vcpu)) { + kvm_inject_gp(ctxt->vcpu, 0); + return X86EMUL_PROPAGATE_FAULT; + } c->eip = tss->eip; ctxt->eflags = tss->eflags | 2; c->regs[VCPU_REGS_RAX] = tss->eax; @@ -3135,7 +3138,10 @@ twobyte_insn: c->dst.type = OP_NONE; /* no writeback */ break; case 0x22: /* mov reg, cr */ - ops->set_cr(c->modrm_reg, c->modrm_val, ctxt->vcpu); + if (ops->set_cr(c->modrm_reg, c->modrm_val, ctxt->vcpu)) { + kvm_inject_gp(ctxt->vcpu, 0); + goto done; + } c->dst.type = OP_NONE; break; case 0x23: /* mov from reg to dr */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 9a469df6011c..64c6e7a31411 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -414,57 +414,49 @@ out: return changed; } -void kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +static int __kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { cr0 |= X86_CR0_ET; #ifdef CONFIG_X86_64 - if (cr0 & 0xffffffff00000000UL) { - kvm_inject_gp(vcpu, 0); - return; - } + if (cr0 & 0xffffffff00000000UL) + return 1; #endif cr0 &= ~CR0_RESERVED_BITS; - if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) { - kvm_inject_gp(vcpu, 0); - return; - } + if ((cr0 & X86_CR0_NW) && !(cr0 & X86_CR0_CD)) + return 1; - if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) { - kvm_inject_gp(vcpu, 0); - return; - } + if ((cr0 & X86_CR0_PG) && !(cr0 & X86_CR0_PE)) + return 1; if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) { #ifdef CONFIG_X86_64 if ((vcpu->arch.efer & EFER_LME)) { int cs_db, cs_l; - if (!is_pae(vcpu)) { - kvm_inject_gp(vcpu, 0); - return; - } + if (!is_pae(vcpu)) + return 1; kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); - if (cs_l) { - kvm_inject_gp(vcpu, 0); - return; - - } + if (cs_l) + return 1; } else #endif - if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.cr3)) { - kvm_inject_gp(vcpu, 0); - return; - } - + if (is_pae(vcpu) && !load_pdptrs(vcpu, vcpu->arch.cr3)) + return 1; } kvm_x86_ops->set_cr0(vcpu, cr0); kvm_mmu_reset_context(vcpu); - return; + return 0; +} + +void kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +{ + if (__kvm_set_cr0(vcpu, cr0)) + kvm_inject_gp(vcpu, 0); } EXPORT_SYMBOL_GPL(kvm_set_cr0); @@ -474,61 +466,56 @@ void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw) } EXPORT_SYMBOL_GPL(kvm_lmsw); -void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { unsigned long old_cr4 = kvm_read_cr4(vcpu); unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE; - if (cr4 & CR4_RESERVED_BITS) { - kvm_inject_gp(vcpu, 0); - return; - } + if (cr4 & CR4_RESERVED_BITS) + return 1; if (is_long_mode(vcpu)) { - if (!(cr4 & X86_CR4_PAE)) { - kvm_inject_gp(vcpu, 0); - return; - } + if (!(cr4 & X86_CR4_PAE)) + return 1; } else if (is_paging(vcpu) && (cr4 & X86_CR4_PAE) && ((cr4 ^ old_cr4) & pdptr_bits) - && !load_pdptrs(vcpu, vcpu->arch.cr3)) { - kvm_inject_gp(vcpu, 0); - return; - } + && !load_pdptrs(vcpu, vcpu->arch.cr3)) + return 1; + + if (cr4 & X86_CR4_VMXE) + return 1; - if (cr4 & X86_CR4_VMXE) { - kvm_inject_gp(vcpu, 0); - return; - } kvm_x86_ops->set_cr4(vcpu, cr4); vcpu->arch.cr4 = cr4; kvm_mmu_reset_context(vcpu); + + return 0; +} + +void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +{ + if (__kvm_set_cr4(vcpu, cr4)) + kvm_inject_gp(vcpu, 0); } EXPORT_SYMBOL_GPL(kvm_set_cr4); -void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +static int __kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) { kvm_mmu_sync_roots(vcpu); kvm_mmu_flush_tlb(vcpu); - return; + return 0; } if (is_long_mode(vcpu)) { - if (cr3 & CR3_L_MODE_RESERVED_BITS) { - kvm_inject_gp(vcpu, 0); - return; - } + if (cr3 & CR3_L_MODE_RESERVED_BITS) + return 1; } else { if (is_pae(vcpu)) { - if (cr3 & CR3_PAE_RESERVED_BITS) { - kvm_inject_gp(vcpu, 0); - return; - } - if (is_paging(vcpu) && !load_pdptrs(vcpu, cr3)) { - kvm_inject_gp(vcpu, 0); - return; - } + if (cr3 & CR3_PAE_RESERVED_BITS) + return 1; + if (is_paging(vcpu) && !load_pdptrs(vcpu, cr3)) + return 1; } /* * We don't check reserved bits in nonpae mode, because @@ -546,24 +533,34 @@ void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) * to debug) behavior on the guest side. */ if (unlikely(!gfn_to_memslot(vcpu->kvm, cr3 >> PAGE_SHIFT))) + return 1; + vcpu->arch.cr3 = cr3; + vcpu->arch.mmu.new_cr3(vcpu); + return 0; +} + +void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +{ + if (__kvm_set_cr3(vcpu, cr3)) kvm_inject_gp(vcpu, 0); - else { - vcpu->arch.cr3 = cr3; - vcpu->arch.mmu.new_cr3(vcpu); - } } EXPORT_SYMBOL_GPL(kvm_set_cr3); -void kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) +int __kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) { - if (cr8 & CR8_RESERVED_BITS) { - kvm_inject_gp(vcpu, 0); - return; - } + if (cr8 & CR8_RESERVED_BITS) + return 1; if (irqchip_in_kernel(vcpu->kvm)) kvm_lapic_set_tpr(vcpu, cr8); else vcpu->arch.cr8 = cr8; + return 0; +} + +void kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) +{ + if (__kvm_set_cr8(vcpu, cr8)) + kvm_inject_gp(vcpu, 0); } EXPORT_SYMBOL_GPL(kvm_set_cr8); @@ -3681,27 +3678,32 @@ static unsigned long emulator_get_cr(int cr, struct kvm_vcpu *vcpu) return value; } -static void emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) +static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) { + int res = 0; + switch (cr) { case 0: - kvm_set_cr0(vcpu, mk_cr_64(kvm_read_cr0(vcpu), val)); + res = __kvm_set_cr0(vcpu, mk_cr_64(kvm_read_cr0(vcpu), val)); break; case 2: vcpu->arch.cr2 = val; break; case 3: - kvm_set_cr3(vcpu, val); + res = __kvm_set_cr3(vcpu, val); break; case 4: - kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); + res = __kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); break; case 8: - kvm_set_cr8(vcpu, val & 0xfUL); + res = __kvm_set_cr8(vcpu, val & 0xfUL); break; default: vcpu_printf(vcpu, "%s: unexpected cr %u\n", __func__, cr); + res = -1; } + + return res; } static int emulator_get_cpl(struct kvm_vcpu *vcpu) -- cgit v1.2.3 From e680080e653b8c8725ca620bf22a5f8480f40cb5 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:33 +0300 Subject: KVM: x86 emulator: fix X86EMUL_RETRY_INSTR and X86EMUL_CMPXCHG_FAILED values Currently X86EMUL_PROPAGATE_FAULT, X86EMUL_RETRY_INSTR and X86EMUL_CMPXCHG_FAILED have the same value so caller cannot distinguish why function such as emulator_cmpxchg_emulated() (which can return both X86EMUL_PROPAGATE_FAULT and X86EMUL_CMPXCHG_FAILED) failed. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 6c4f4918db5e..0cf4311db0d6 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -51,8 +51,9 @@ struct x86_emulate_ctxt; #define X86EMUL_UNHANDLEABLE 1 /* Terminate emulation but return success to the caller. */ #define X86EMUL_PROPAGATE_FAULT 2 /* propagate a generated fault to guest */ -#define X86EMUL_RETRY_INSTR 2 /* retry the instruction for some reason */ -#define X86EMUL_CMPXCHG_FAILED 2 /* cmpxchg did not see expected value */ +#define X86EMUL_RETRY_INSTR 3 /* retry the instruction for some reason */ +#define X86EMUL_CMPXCHG_FAILED 4 /* cmpxchg did not see expected value */ + struct x86_emulate_ops { /* * read_std: Read bytes of standard (non-emulated/special) memory. -- cgit v1.2.3 From c3cd7ffaf57ae6ead5b394cebaeb76164059a57f Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:35 +0300 Subject: KVM: x86 emulator: x86_emulate_insn() return -1 only in case of emulation failure Currently emulator returns -1 when emulation failed or IO is needed. Caller tries to guess whether emulation failed by looking at other variables. Make it easier for caller to recognise error condition by always returning -1 in case of failure. For this new emulator internal return value X86EMUL_IO_NEEDED is introduced. It is used to distinguish between error condition (which returns X86EMUL_UNHANDLEABLE) and condition that requires IO exit to userspace to continue emulation. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 1 + arch/x86/kvm/x86.c | 36 ++++++++++++++++++------------------ 2 files changed, 19 insertions(+), 18 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 0cf4311db0d6..777240d4524e 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -53,6 +53,7 @@ struct x86_emulate_ctxt; #define X86EMUL_PROPAGATE_FAULT 2 /* propagate a generated fault to guest */ #define X86EMUL_RETRY_INSTR 3 /* retry the instruction for some reason */ #define X86EMUL_CMPXCHG_FAILED 4 /* cmpxchg did not see expected value */ +#define X86EMUL_IO_NEEDED 5 /* IO is needed to complete emulation */ struct x86_emulate_ops { /* diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b976c4c1fa8f..4cb65d82abca 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3275,7 +3275,7 @@ static int kvm_read_guest_virt_helper(gva_t addr, void *val, unsigned int bytes, } ret = kvm_read_guest(vcpu->kvm, gpa, data, toread); if (ret < 0) { - r = X86EMUL_UNHANDLEABLE; + r = X86EMUL_IO_NEEDED; goto out; } @@ -3331,7 +3331,7 @@ static int kvm_write_guest_virt_system(gva_t addr, void *val, } ret = kvm_write_guest(vcpu->kvm, gpa, data, towrite); if (ret < 0) { - r = X86EMUL_UNHANDLEABLE; + r = X86EMUL_IO_NEEDED; goto out; } @@ -3391,7 +3391,7 @@ mmio: vcpu->run->mmio.len = vcpu->mmio_size = bytes; vcpu->run->mmio.is_write = vcpu->mmio_is_write = 0; - return X86EMUL_UNHANDLEABLE; + return X86EMUL_IO_NEEDED; } int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, @@ -3863,8 +3863,6 @@ int emulate_instruction(struct kvm_vcpu *vcpu, */ cache_all_regs(vcpu); - vcpu->mmio_is_write = 0; - if (!(emulation_type & EMULTYPE_NO_DECODE)) { int cs_db, cs_l; kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); @@ -3938,24 +3936,26 @@ restart: return EMULATE_DO_MMIO; } - if (r) { - if (kvm_mmu_unprotect_page_virt(vcpu, cr2)) - goto done; - if (!vcpu->mmio_needed) { - ++vcpu->stat.insn_emulation_fail; - trace_kvm_emulate_insn_failed(vcpu); - kvm_report_emulation_failure(vcpu, "mmio"); - return EMULATE_FAIL; - } + if (vcpu->mmio_needed) { + if (vcpu->mmio_is_write) + vcpu->mmio_needed = 0; return EMULATE_DO_MMIO; } - if (vcpu->mmio_is_write) { - vcpu->mmio_needed = 0; - return EMULATE_DO_MMIO; + if (r) { /* emulation failed */ + /* + * if emulation was due to access to shadowed page table + * and it failed try to unshadow page and re-entetr the + * guest to let CPU execute the instruction. + */ + if (kvm_mmu_unprotect_page_virt(vcpu, cr2)) + return EMULATE_DONE; + + trace_kvm_emulate_insn_failed(vcpu); + kvm_report_emulation_failure(vcpu, "mmio"); + return EMULATE_FAIL; } -done: if (vcpu->arch.exception.pending) vcpu->arch.emulate_ctxt.restart = false; -- cgit v1.2.3 From f181b96d4c769b8915849eb9070c18116fd8d44e Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:36 +0300 Subject: KVM: remove export of emulator_write_emulated() It is not called directly outside of the file it's defined in anymore. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 5 ----- arch/x86/kvm/x86.c | 1 - 2 files changed, 6 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 97774ae3c874..2ca1867ed97a 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -628,11 +628,6 @@ void kvm_inject_nmi(struct kvm_vcpu *vcpu); void fx_init(struct kvm_vcpu *vcpu); -int emulator_write_emulated(unsigned long addr, - const void *val, - unsigned int bytes, - struct kvm_vcpu *vcpu); - void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu); void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, const u8 *new, int bytes, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 4cb65d82abca..15a4b754a451 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3465,7 +3465,6 @@ int emulator_write_emulated(unsigned long addr, } return emulator_write_emulated_onepage(addr, val, bytes, vcpu); } -EXPORT_SYMBOL_GPL(emulator_write_emulated); #define CMPXCHG_TYPE(t, ptr, old, new) \ (cmpxchg((t *)(ptr), *(t *)(old), *(t *)(new)) == *(t *)(old)) -- cgit v1.2.3 From 8fe681e984b6505d4d12125c0776399304803ec7 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:37 +0300 Subject: KVM: do not inject #PF in (read|write)_emulated() callbacks Return error to x86 emulator instead of injection exception behind its back. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 3 +++ arch/x86/kvm/emulate.c | 12 +++++++++++- arch/x86/kvm/x86.c | 28 ++++++++++++++-------------- 3 files changed, 28 insertions(+), 15 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 777240d4524e..b7e00cb21c6e 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -94,6 +94,7 @@ struct x86_emulate_ops { int (*read_emulated)(unsigned long addr, void *val, unsigned int bytes, + unsigned int *error, struct kvm_vcpu *vcpu); /* @@ -106,6 +107,7 @@ struct x86_emulate_ops { int (*write_emulated)(unsigned long addr, const void *val, unsigned int bytes, + unsigned int *error, struct kvm_vcpu *vcpu); /* @@ -120,6 +122,7 @@ struct x86_emulate_ops { const void *old, const void *new, unsigned int bytes, + unsigned int *error, struct kvm_vcpu *vcpu); int (*pio_in_emulated)(int size, unsigned short port, void *val, diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index d5979ecc2521..d7a18a0f80a2 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -1277,6 +1277,7 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, { int rc; struct read_cache *mc = &ctxt->decode.mem_read; + u32 err; while (size) { int n = min(size, 8u); @@ -1284,7 +1285,10 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, if (mc->pos < mc->end) goto read_cached; - rc = ops->read_emulated(addr, mc->data + mc->end, n, ctxt->vcpu); + rc = ops->read_emulated(addr, mc->data + mc->end, n, &err, + ctxt->vcpu); + if (rc == X86EMUL_PROPAGATE_FAULT) + kvm_inject_page_fault(ctxt->vcpu, addr, err); if (rc != X86EMUL_CONTINUE) return rc; mc->end += n; @@ -1789,6 +1793,7 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, { int rc; struct decode_cache *c = &ctxt->decode; + u32 err; switch (c->dst.type) { case OP_REG: @@ -1817,13 +1822,18 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, &c->dst.orig_val, &c->dst.val, c->dst.bytes, + &err, ctxt->vcpu); else rc = ops->write_emulated( (unsigned long)c->dst.ptr, &c->dst.val, c->dst.bytes, + &err, ctxt->vcpu); + if (rc == X86EMUL_PROPAGATE_FAULT) + kvm_inject_page_fault(ctxt->vcpu, + (unsigned long)c->dst.ptr, err); if (rc != X86EMUL_CONTINUE) return rc; break; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 15a4b754a451..51402d8a46fa 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3346,10 +3346,10 @@ out: static int emulator_read_emulated(unsigned long addr, void *val, unsigned int bytes, + unsigned int *error_code, struct kvm_vcpu *vcpu) { gpa_t gpa; - u32 error_code; if (vcpu->mmio_read_completed) { memcpy(val, vcpu->mmio_data, bytes); @@ -3359,12 +3359,10 @@ static int emulator_read_emulated(unsigned long addr, return X86EMUL_CONTINUE; } - gpa = kvm_mmu_gva_to_gpa_read(vcpu, addr, &error_code); + gpa = kvm_mmu_gva_to_gpa_read(vcpu, addr, error_code); - if (gpa == UNMAPPED_GVA) { - kvm_inject_page_fault(vcpu, addr, error_code); + if (gpa == UNMAPPED_GVA) return X86EMUL_PROPAGATE_FAULT; - } /* For APIC access vmexit */ if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) @@ -3409,17 +3407,15 @@ int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, static int emulator_write_emulated_onepage(unsigned long addr, const void *val, unsigned int bytes, + unsigned int *error_code, struct kvm_vcpu *vcpu) { gpa_t gpa; - u32 error_code; - gpa = kvm_mmu_gva_to_gpa_write(vcpu, addr, &error_code); + gpa = kvm_mmu_gva_to_gpa_write(vcpu, addr, error_code); - if (gpa == UNMAPPED_GVA) { - kvm_inject_page_fault(vcpu, addr, error_code); + if (gpa == UNMAPPED_GVA) return X86EMUL_PROPAGATE_FAULT; - } /* For APIC access vmexit */ if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) @@ -3449,6 +3445,7 @@ mmio: int emulator_write_emulated(unsigned long addr, const void *val, unsigned int bytes, + unsigned int *error_code, struct kvm_vcpu *vcpu) { /* Crossing a page boundary? */ @@ -3456,14 +3453,16 @@ int emulator_write_emulated(unsigned long addr, int rc, now; now = -addr & ~PAGE_MASK; - rc = emulator_write_emulated_onepage(addr, val, now, vcpu); + rc = emulator_write_emulated_onepage(addr, val, now, error_code, + vcpu); if (rc != X86EMUL_CONTINUE) return rc; addr += now; val += now; bytes -= now; } - return emulator_write_emulated_onepage(addr, val, bytes, vcpu); + return emulator_write_emulated_onepage(addr, val, bytes, error_code, + vcpu); } #define CMPXCHG_TYPE(t, ptr, old, new) \ @@ -3480,6 +3479,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr, const void *old, const void *new, unsigned int bytes, + unsigned int *error_code, struct kvm_vcpu *vcpu) { gpa_t gpa; @@ -3533,7 +3533,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr, emul_write: printk_once(KERN_WARNING "kvm: emulating exchange as write\n"); - return emulator_write_emulated(addr, new, bytes, vcpu); + return emulator_write_emulated(addr, new, bytes, error_code, vcpu); } static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) @@ -4293,7 +4293,7 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) kvm_x86_ops->patch_hypercall(vcpu, instruction); - return emulator_write_emulated(rip, instruction, 3, vcpu); + return emulator_write_emulated(rip, instruction, 3, NULL, vcpu); } void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base) -- cgit v1.2.3 From ef050dc0390176ec6888f373edb776587c88be3d Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:40 +0300 Subject: KVM: x86 emulator: set RFLAGS outside x86 emulator code Removes the need for set_flags() callback. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 1 - arch/x86/kvm/emulate.c | 1 - arch/x86/kvm/x86.c | 7 +------ 3 files changed, 1 insertion(+), 8 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index b7e00cb21c6e..a87d95f09572 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -142,7 +142,6 @@ struct x86_emulate_ops { ulong (*get_cr)(int cr, struct kvm_vcpu *vcpu); int (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); int (*cpl)(struct kvm_vcpu *vcpu); - void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); int (*get_dr)(int dr, unsigned long *dest, struct kvm_vcpu *vcpu); int (*set_dr)(int dr, unsigned long value, struct kvm_vcpu *vcpu); int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 437f31bcffea..291e220c69a5 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -3034,7 +3034,6 @@ writeback: /* Commit shadow register state. */ memcpy(ctxt->vcpu->arch.regs, c->regs, sizeof c->regs); ctxt->eip = c->eip; - ops->set_rflags(ctxt->vcpu, ctxt->eflags); done: return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8f45cc712dda..04ca343ee512 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3806,11 +3806,6 @@ static void emulator_set_segment_selector(u16 sel, int seg, kvm_set_segment(vcpu, &kvm_seg, seg); } -static void emulator_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) -{ - kvm_x86_ops->set_rflags(vcpu, rflags); -} - static struct x86_emulate_ops emulate_ops = { .read_std = kvm_read_guest_virt_system, .write_std = kvm_write_guest_virt_system, @@ -3829,7 +3824,6 @@ static struct x86_emulate_ops emulate_ops = { .get_cr = emulator_get_cr, .set_cr = emulator_set_cr, .cpl = emulator_get_cpl, - .set_rflags = emulator_set_rflags, .get_dr = emulator_get_dr, .set_dr = emulator_set_dr, .set_msr = kvm_set_msr, @@ -3941,6 +3935,7 @@ restart: shadow_mask = vcpu->arch.emulate_ctxt.interruptibility; kvm_x86_ops->set_interrupt_shadow(vcpu, shadow_mask); + kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); if (vcpu->arch.pio.count) { -- cgit v1.2.3 From 54b8486f469475d6c8e8aec917b91239a54eb8c8 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:44 +0300 Subject: KVM: x86 emulator: do not inject exception directly into vcpu Return exception as a result of instruction emulation and handle injection in KVM code. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 6 ++ arch/x86/kvm/emulate.c | 124 +++++++++++++++++++++++-------------- arch/x86/kvm/x86.c | 20 +++++- 3 files changed, 100 insertions(+), 50 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index a87d95f09572..51cfd730ac5d 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -216,6 +216,12 @@ struct x86_emulate_ctxt { int interruptibility; bool restart; /* restart string instruction after writeback */ + + int exception; /* exception that happens during emulation or -1 */ + u32 error_code; /* error code for exception */ + bool error_code_valid; + unsigned long cr2; /* faulted address in case of #PF */ + /* decode cache */ struct decode_cache decode; }; diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index c40b40561dff..b43ac98ef790 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -653,6 +653,37 @@ static unsigned long ss_base(struct x86_emulate_ctxt *ctxt, return seg_base(ctxt, ops, VCPU_SREG_SS); } +static void emulate_exception(struct x86_emulate_ctxt *ctxt, int vec, + u32 error, bool valid) +{ + ctxt->exception = vec; + ctxt->error_code = error; + ctxt->error_code_valid = valid; + ctxt->restart = false; +} + +static void emulate_gp(struct x86_emulate_ctxt *ctxt, int err) +{ + emulate_exception(ctxt, GP_VECTOR, err, true); +} + +static void emulate_pf(struct x86_emulate_ctxt *ctxt, unsigned long addr, + int err) +{ + ctxt->cr2 = addr; + emulate_exception(ctxt, PF_VECTOR, err, true); +} + +static void emulate_ud(struct x86_emulate_ctxt *ctxt) +{ + emulate_exception(ctxt, UD_VECTOR, 0, false); +} + +static void emulate_ts(struct x86_emulate_ctxt *ctxt, int err) +{ + emulate_exception(ctxt, TS_VECTOR, err, true); +} + static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned long eip, u8 *dest) @@ -1285,7 +1316,7 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, rc = ops->read_emulated(addr, mc->data + mc->end, n, &err, ctxt->vcpu); if (rc == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); if (rc != X86EMUL_CONTINUE) return rc; mc->end += n; @@ -1366,13 +1397,13 @@ static int read_segment_descriptor(struct x86_emulate_ctxt *ctxt, get_descriptor_table_ptr(ctxt, ops, selector, &dt); if (dt.size < index * 8 + 7) { - kvm_inject_gp(ctxt->vcpu, selector & 0xfffc); + emulate_gp(ctxt, selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } addr = dt.address + index * 8; ret = ops->read_std(addr, desc, sizeof *desc, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); return ret; } @@ -1391,14 +1422,14 @@ static int write_segment_descriptor(struct x86_emulate_ctxt *ctxt, get_descriptor_table_ptr(ctxt, ops, selector, &dt); if (dt.size < index * 8 + 7) { - kvm_inject_gp(ctxt->vcpu, selector & 0xfffc); + emulate_gp(ctxt, selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } addr = dt.address + index * 8; ret = ops->write_std(addr, desc, sizeof *desc, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); return ret; } @@ -1517,7 +1548,7 @@ load: ops->set_cached_descriptor(&seg_desc, seg, ctxt->vcpu); return X86EMUL_CONTINUE; exception: - kvm_queue_exception_e(ctxt->vcpu, err_vec, err_code); + emulate_exception(ctxt, err_vec, err_code, true); return X86EMUL_PROPAGATE_FAULT; } @@ -1578,7 +1609,7 @@ static int emulate_popf(struct x86_emulate_ctxt *ctxt, break; case X86EMUL_MODE_VM86: if (iopl < 3) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } change_mask |= EFLG_IF; @@ -1829,7 +1860,7 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, &err, ctxt->vcpu); if (rc == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, + emulate_pf(ctxt, (unsigned long)c->dst.ptr, err); if (rc != X86EMUL_CONTINUE) return rc; @@ -1883,7 +1914,7 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* syscall is not available in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL || ctxt->mode == X86EMUL_MODE_VM86) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); return X86EMUL_PROPAGATE_FAULT; } @@ -1937,7 +1968,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* inject #GP if in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } @@ -1945,7 +1976,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) * Therefore, we inject an #UD. */ if (ctxt->mode == X86EMUL_MODE_PROT64) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); return X86EMUL_PROPAGATE_FAULT; } @@ -1955,13 +1986,13 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) switch (ctxt->mode) { case X86EMUL_MODE_PROT32: if ((msr_data & 0xfffc) == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } break; case X86EMUL_MODE_PROT64: if (msr_data == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } break; @@ -2004,7 +2035,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* inject #GP if in real mode or Virtual 8086 mode */ if (ctxt->mode == X86EMUL_MODE_REAL || ctxt->mode == X86EMUL_MODE_VM86) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } @@ -2022,7 +2053,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) case X86EMUL_MODE_PROT32: cs_sel = (u16)(msr_data + 16); if ((msr_data & 0xfffc) == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } ss_sel = (u16)(msr_data + 24); @@ -2030,7 +2061,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) case X86EMUL_MODE_PROT64: cs_sel = (u16)(msr_data + 32); if (msr_data == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } ss_sel = cs_sel + 8; @@ -2192,7 +2223,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2202,7 +2233,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2210,7 +2241,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } @@ -2223,7 +2254,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } } @@ -2266,7 +2297,7 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, int ret; if (ops->set_cr(3, tss->cr3, ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } c->eip = tss->eip; @@ -2334,7 +2365,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2344,7 +2375,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2352,7 +2383,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } @@ -2365,7 +2396,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } } @@ -2399,7 +2430,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, if (reason != TASK_SWITCH_IRET) { if ((tss_selector & 3) > next_tss_desc.dpl || ops->cpl(ctxt->vcpu) > next_tss_desc.dpl) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } } @@ -2408,8 +2439,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, if (!next_tss_desc.p || ((desc_limit < 0x67 && (next_tss_desc.type & 8)) || desc_limit < 0x2b)) { - kvm_queue_exception_e(ctxt->vcpu, TS_VECTOR, - tss_selector & 0xfffc); + emulate_ts(ctxt, tss_selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } @@ -2505,19 +2535,19 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) ctxt->decode.mem_read.pos = 0; if (ctxt->mode == X86EMUL_MODE_PROT64 && (c->d & No64)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } /* LOCK prefix is allowed only with some instructions */ if (c->lock_prefix && (!(c->d & Lock) || c->dst.type != OP_MEM)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } /* Privileged instruction can be executed only in CPL=0 */ if ((c->d & Priv) && ops->cpl(ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } @@ -2679,7 +2709,7 @@ special_insn: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->regs[VCPU_REGS_RDX], c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } if (!pio_in_emulated(ctxt, ops, c->dst.bytes, @@ -2691,7 +2721,7 @@ special_insn: c->src.bytes = min(c->src.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->regs[VCPU_REGS_RDX], c->src.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } ops->pio_out_emulated(c->src.bytes, c->regs[VCPU_REGS_RDX], @@ -2754,7 +2784,7 @@ special_insn: goto mov; case 0x8c: /* mov r/m, sreg */ if (c->modrm_reg > VCPU_SREG_GS) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } c->dst.val = ops->get_segment_selector(c->modrm_reg, ctxt->vcpu); @@ -2769,7 +2799,7 @@ special_insn: if (c->modrm_reg == VCPU_SREG_CS || c->modrm_reg > VCPU_SREG_GS) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } @@ -2895,7 +2925,7 @@ special_insn: do_io_in: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->src.val, c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } if (!pio_in_emulated(ctxt, ops, c->dst.bytes, c->src.val, @@ -2908,7 +2938,7 @@ special_insn: do_io_out: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->src.val, c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } ops->pio_out_emulated(c->dst.bytes, c->src.val, &c->dst.val, 1, @@ -2933,7 +2963,7 @@ special_insn: break; case 0xfa: /* cli */ if (emulator_bad_iopl(ctxt, ops)) - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); else { ctxt->eflags &= ~X86_EFLAGS_IF; c->dst.type = OP_NONE; /* Disable writeback. */ @@ -2941,7 +2971,7 @@ special_insn: break; case 0xfb: /* sti */ if (emulator_bad_iopl(ctxt, ops)) - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); else { ctxt->interruptibility = KVM_X86_SHADOW_INT_STI; ctxt->eflags |= X86_EFLAGS_IF; @@ -3069,7 +3099,7 @@ twobyte_insn: c->dst.type = OP_NONE; break; case 5: /* not defined */ - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; case 7: /* invlpg*/ emulate_invlpg(ctxt->vcpu, c->modrm_ea); @@ -3102,7 +3132,7 @@ twobyte_insn: case 1: case 5 ... 7: case 9 ... 15: - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } c->regs[c->modrm_rm] = ops->get_cr(c->modrm_reg, ctxt->vcpu); @@ -3111,7 +3141,7 @@ twobyte_insn: case 0x21: /* mov from dr to reg */ if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && (c->modrm_reg == 4 || c->modrm_reg == 5)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } ops->get_dr(c->modrm_reg, &c->regs[c->modrm_rm], ctxt->vcpu); @@ -3119,7 +3149,7 @@ twobyte_insn: break; case 0x22: /* mov reg, cr */ if (ops->set_cr(c->modrm_reg, c->modrm_val, ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } c->dst.type = OP_NONE; @@ -3127,7 +3157,7 @@ twobyte_insn: case 0x23: /* mov from reg to dr */ if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && (c->modrm_reg == 4 || c->modrm_reg == 5)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } @@ -3135,7 +3165,7 @@ twobyte_insn: ((ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U), ctxt->vcpu) < 0) { /* #UD condition is already handled by the code above */ - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } @@ -3146,7 +3176,7 @@ twobyte_insn: msr_data = (u32)c->regs[VCPU_REGS_RAX] | ((u64)c->regs[VCPU_REGS_RDX] << 32); if (ops->set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } rc = X86EMUL_CONTINUE; @@ -3155,7 +3185,7 @@ twobyte_insn: case 0x32: /* rdmsr */ if (ops->get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } else { c->regs[VCPU_REGS_RAX] = (u32)msr_data; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 91bfe7771f50..63c87adcec48 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3852,6 +3852,17 @@ static void toggle_interruptibility(struct kvm_vcpu *vcpu, u32 mask) kvm_x86_ops->set_interrupt_shadow(vcpu, mask); } +static void inject_emulated_exception(struct kvm_vcpu *vcpu) +{ + struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; + if (ctxt->exception == PF_VECTOR) + kvm_inject_page_fault(vcpu, ctxt->cr2, ctxt->error_code); + else if (ctxt->error_code_valid) + kvm_queue_exception_e(vcpu, ctxt->exception, ctxt->error_code); + else + kvm_queue_exception(vcpu, ctxt->exception); +} + int emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, u16 error_code, @@ -3886,6 +3897,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu, memset(c, 0, sizeof(struct decode_cache)); memcpy(c->regs, vcpu->arch.regs, sizeof c->regs); vcpu->arch.emulate_ctxt.interruptibility = 0; + vcpu->arch.emulate_ctxt.exception = -1; r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops); trace_kvm_emulate_insn_start(vcpu); @@ -3958,6 +3970,11 @@ restart: memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); + if (vcpu->arch.emulate_ctxt.exception >= 0) { + inject_emulated_exception(vcpu); + return EMULATE_DONE; + } + if (vcpu->arch.pio.count) { if (!vcpu->arch.pio.in) vcpu->arch.pio.count = 0; @@ -3970,9 +3987,6 @@ restart: return EMULATE_DO_MMIO; } - if (vcpu->arch.exception.pending) - vcpu->arch.emulate_ctxt.restart = false; - if (vcpu->arch.emulate_ctxt.restart) goto restart; -- cgit v1.2.3 From eec4b140c924b4c650e9a89e01d223266490e325 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 5 May 2010 16:04:44 +0200 Subject: KVM: SVM: Allow EFER.LMSLE to be set with nested svm This patch enables setting of efer bit 13 which is allowed in all SVM capable processors. This is necessary for the SLES11 version of Xen 4.0 to boot with nested svm. Signed-off-by: Joerg Roedel Signed-off-by: Avi Kivity --- arch/x86/include/asm/msr-index.h | 2 ++ arch/x86/kvm/svm.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 8c7ae4318629..509a42187dc2 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -20,6 +20,7 @@ #define _EFER_LMA 10 /* Long mode active (read-only) */ #define _EFER_NX 11 /* No execute enable */ #define _EFER_SVME 12 /* Enable virtualization */ +#define _EFER_LMSLE 13 /* Long Mode Segment Limit Enable */ #define _EFER_FFXSR 14 /* Enable Fast FXSAVE/FXRSTOR */ #define EFER_SCE (1<<_EFER_SCE) @@ -27,6 +28,7 @@ #define EFER_LMA (1<<_EFER_LMA) #define EFER_NX (1<<_EFER_NX) #define EFER_SVME (1<<_EFER_SVME) +#define EFER_LMSLE (1<<_EFER_LMSLE) #define EFER_FFXSR (1<<_EFER_FFXSR) /* Intel MSRs. Some also available on other CPUs */ diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 685cffff01f3..41fe0381a1ae 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -640,7 +640,7 @@ static __init int svm_hardware_setup(void) if (nested) { printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); - kvm_enable_efer_bits(EFER_SVME); + kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE); } for_each_possible_cpu(cpu) { -- cgit v1.2.3 From 6d77dbfc88e37c9efd5c5dd18445cfe819ae17ea Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Mon, 10 May 2010 11:16:56 +0300 Subject: KVM: inject #UD if instruction emulation fails and exit to userspace Do not kill VM when instruction emulation fails. Inject #UD and report failure to userspace instead. Userspace may choose to reenter guest if vcpu is in userspace (cpl == 3) in which case guest OS will kill offending process and continue running. Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 1 - arch/x86/kvm/mmu.c | 5 +---- arch/x86/kvm/svm.c | 10 +++------- arch/x86/kvm/vmx.c | 28 ++++----------------------- arch/x86/kvm/x86.c | 43 +++++++++++++++++------------------------ 5 files changed, 26 insertions(+), 61 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 2ca1867ed97a..0c06148fa3b1 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -576,7 +576,6 @@ enum emulation_result { #define EMULTYPE_SKIP (1 << 2) int emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, u16 error_code, int emulation_type); -void kvm_report_emulation_failure(struct kvm_vcpu *cvpu, const char *context); void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index be981b1f1881..4a02dee1f2b5 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2814,11 +2814,8 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code) return 1; case EMULATE_DO_MMIO: ++vcpu->stat.mmio_exits; - return 0; + /* fall through */ case EMULATE_FAIL: - vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; - vcpu->run->internal.ndata = 0; return 0; default: BUG(); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 41fe0381a1ae..134260c36ce2 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1535,7 +1535,7 @@ static int io_interception(struct vcpu_svm *svm) string = (io_info & SVM_IOIO_STR_MASK) != 0; in = (io_info & SVM_IOIO_TYPE_MASK) != 0; if (string || in) - return !(emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DO_MMIO); + return emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DONE; port = io_info >> 16; size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT; @@ -2386,16 +2386,12 @@ static int iret_interception(struct vcpu_svm *svm) static int invlpg_interception(struct vcpu_svm *svm) { - if (emulate_instruction(&svm->vcpu, 0, 0, 0) != EMULATE_DONE) - pr_unimpl(&svm->vcpu, "%s: failed\n", __func__); - return 1; + return emulate_instruction(&svm->vcpu, 0, 0, 0) == EMULATE_DONE; } static int emulate_on_interception(struct vcpu_svm *svm) { - if (emulate_instruction(&svm->vcpu, 0, 0, 0) != EMULATE_DONE) - pr_unimpl(&svm->vcpu, "%s: failed\n", __func__); - return 1; + return emulate_instruction(&svm->vcpu, 0, 0, 0) == EMULATE_DONE; } static int cr8_write_interception(struct vcpu_svm *svm) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 598931734251..a82cfa1e2a40 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3070,7 +3070,7 @@ static int handle_io(struct kvm_vcpu *vcpu) ++vcpu->stat.io_exits; if (string || in) - return !(emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DO_MMIO); + return emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DONE; port = exit_qualification >> 16; size = (exit_qualification & 7) + 1; @@ -3327,22 +3327,7 @@ static int handle_wbinvd(struct kvm_vcpu *vcpu) static int handle_apic_access(struct kvm_vcpu *vcpu) { - unsigned long exit_qualification; - enum emulation_result er; - unsigned long offset; - - exit_qualification = vmcs_readl(EXIT_QUALIFICATION); - offset = exit_qualification & 0xffful; - - er = emulate_instruction(vcpu, 0, 0, 0); - - if (er != EMULATE_DONE) { - printk(KERN_ERR - "Fail to handle apic access vmexit! Offset is 0x%lx\n", - offset); - return -ENOEXEC; - } - return 1; + return emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DONE; } static int handle_task_switch(struct kvm_vcpu *vcpu) @@ -3554,13 +3539,8 @@ static int handle_invalid_guest_state(struct kvm_vcpu *vcpu) goto out; } - if (err != EMULATE_DONE) { - vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; - vcpu->run->internal.ndata = 0; - ret = 0; - goto out; - } + if (err != EMULATE_DONE) + return 0; if (signal_pending(current)) goto out; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index fc5611b4007f..ae9d6f3e5d0d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3639,24 +3639,6 @@ int emulator_set_dr(int dr, unsigned long value, struct kvm_vcpu *vcpu) return __kvm_set_dr(vcpu, dr, value); } -void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) -{ - u8 opcodes[4]; - unsigned long rip = kvm_rip_read(vcpu); - unsigned long rip_linear; - - if (!printk_ratelimit()) - return; - - rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS); - - kvm_read_guest_virt(rip_linear, (void *)opcodes, 4, vcpu, NULL); - - printk(KERN_ERR "emulation failed (%s) rip %lx %02x %02x %02x %02x\n", - context, rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]); -} -EXPORT_SYMBOL_GPL(kvm_report_emulation_failure); - static u64 mk_cr_64(u64 curr_cr, u32 new_val) { return (curr_cr & ~((1ULL << 32) - 1)) | new_val; @@ -3863,6 +3845,19 @@ static void inject_emulated_exception(struct kvm_vcpu *vcpu) kvm_queue_exception(vcpu, ctxt->exception); } +static int handle_emulation_failure(struct kvm_vcpu *vcpu) +{ + struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; + + ++vcpu->stat.insn_emulation_fail; + trace_kvm_emulate_insn_failed(vcpu); + vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + vcpu->run->internal.suberror = KVM_INTERNAL_ERROR_EMULATION; + vcpu->run->internal.ndata = 0; + kvm_queue_exception(vcpu, UD_VECTOR); + return EMULATE_FAIL; +} + int emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, u16 error_code, @@ -3931,11 +3926,11 @@ int emulate_instruction(struct kvm_vcpu *vcpu, ++vcpu->stat.insn_emulation; if (r) { - ++vcpu->stat.insn_emulation_fail; - trace_kvm_emulate_insn_failed(vcpu); if (kvm_mmu_unprotect_page_virt(vcpu, cr2)) return EMULATE_DONE; - return EMULATE_FAIL; + if (emulation_type & EMULTYPE_SKIP) + return EMULATE_FAIL; + return handle_emulation_failure(vcpu); } } @@ -3960,9 +3955,7 @@ restart: if (kvm_mmu_unprotect_page_virt(vcpu, cr2)) return EMULATE_DONE; - trace_kvm_emulate_insn_failed(vcpu); - kvm_report_emulation_failure(vcpu, "mmio"); - return EMULATE_FAIL; + return handle_emulation_failure(vcpu); } toggle_interruptibility(vcpu, vcpu->arch.emulate_ctxt.interruptibility); @@ -4798,7 +4791,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); r = emulate_instruction(vcpu, 0, 0, EMULTYPE_NO_DECODE); srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); - if (r == EMULATE_DO_MMIO) { + if (r != EMULATE_DONE) { r = 0; goto out; } -- cgit v1.2.3 From 5ee481da7b62a992b91f958bf26aaaa92354c170 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Mon, 17 May 2010 17:22:23 +0800 Subject: x86: Export FPU API for KVM use Also add some constants. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/include/asm/i387.h | 2 ++ arch/x86/include/asm/xsave.h | 3 +++ arch/x86/kernel/i387.c | 3 ++- arch/x86/kernel/process.c | 1 + 4 files changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index c991b3a7b904..815c5b2b9f57 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -482,6 +482,8 @@ static inline void fpu_copy(struct fpu *dst, struct fpu *src) memcpy(dst->state, src->state, xstate_size); } +extern void fpu_finit(struct fpu *fpu); + #endif /* __ASSEMBLY__ */ #define PSHUFB_XMM5_XMM0 .byte 0x66, 0x0f, 0x38, 0x00, 0xc5 diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h index 2c4390cae228..29ee4e4c64cf 100644 --- a/arch/x86/include/asm/xsave.h +++ b/arch/x86/include/asm/xsave.h @@ -13,6 +13,9 @@ #define FXSAVE_SIZE 512 +#define XSTATE_YMM_SIZE 256 +#define XSTATE_YMM_OFFSET (512 + 64) + /* * These are the features that the OS can handle currently. */ diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 86cef6b32253..c4444bce8469 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -107,7 +107,7 @@ void __cpuinit fpu_init(void) } #endif /* CONFIG_X86_64 */ -static void fpu_finit(struct fpu *fpu) +void fpu_finit(struct fpu *fpu) { #ifdef CONFIG_X86_32 if (!HAVE_HWFP) { @@ -132,6 +132,7 @@ static void fpu_finit(struct fpu *fpu) fp->fos = 0xffff0000u; } } +EXPORT_SYMBOL_GPL(fpu_finit); /* * The _current_ task is using the FPU for the first time diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index e7e35219b32f..ebcfcceccc72 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -28,6 +28,7 @@ unsigned long idle_nomwait; EXPORT_SYMBOL(idle_nomwait); struct kmem_cache *task_xstate_cachep; +EXPORT_SYMBOL_GPL(task_xstate_cachep); int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { -- cgit v1.2.3 From 7cf30855e02be7a207ffebb8b9350986f2ba83e9 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Mon, 17 May 2010 17:08:27 +0800 Subject: KVM: x86: Use unlazy_fpu() for host FPU We can avoid unnecessary fpu load when userspace process didn't use FPU frequently. Derived from Avi's idea. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 1 - arch/x86/kvm/x86.c | 18 ++---------------- 2 files changed, 2 insertions(+), 17 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 0c06148fa3b1..d93601c52902 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -301,7 +301,6 @@ struct kvm_vcpu_arch { unsigned long mmu_seq; } update_pte; - struct i387_fxsave_struct host_fx_image; struct i387_fxsave_struct guest_fx_image; gva_t mmio_fault_cr2; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 4c2096f30d90..54ce77582eda 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -52,6 +52,7 @@ #include #include #include +#include #define MAX_IO_MSRS 256 #define CR0_RESERVED_BITS \ @@ -5134,21 +5135,10 @@ void fx_init(struct kvm_vcpu *vcpu) { unsigned after_mxcsr_mask; - /* - * Touch the fpu the first time in non atomic context as if - * this is the first fpu instruction the exception handler - * will fire before the instruction returns and it'll have to - * allocate ram with GFP_KERNEL. - */ - if (!used_math()) - kvm_fx_save(&vcpu->arch.host_fx_image); - /* Initialize guest FPU by resetting ours and saving into guest's */ preempt_disable(); - kvm_fx_save(&vcpu->arch.host_fx_image); kvm_fx_finit(); kvm_fx_save(&vcpu->arch.guest_fx_image); - kvm_fx_restore(&vcpu->arch.host_fx_image); preempt_enable(); vcpu->arch.cr0 |= X86_CR0_ET; @@ -5165,7 +5155,7 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) return; vcpu->guest_fpu_loaded = 1; - kvm_fx_save(&vcpu->arch.host_fx_image); + unlazy_fpu(current); kvm_fx_restore(&vcpu->arch.guest_fx_image); trace_kvm_fpu(1); } @@ -5177,7 +5167,6 @@ void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) vcpu->guest_fpu_loaded = 0; kvm_fx_save(&vcpu->arch.guest_fx_image); - kvm_fx_restore(&vcpu->arch.host_fx_image); ++vcpu->stat.fpu_reload; set_bit(KVM_REQ_DEACTIVATE_FPU, &vcpu->requests); trace_kvm_fpu(0); @@ -5203,9 +5192,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) { int r; - /* We do fxsave: this must be aligned. */ - BUG_ON((unsigned long)&vcpu->arch.host_fx_image & 0xF); - vcpu->arch.mtrr_state.have_fixed = 1; vcpu_load(vcpu); r = kvm_arch_vcpu_reset(vcpu); -- cgit v1.2.3 From 98918833a3e21ffc5619535955e7a003cb788163 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Mon, 17 May 2010 17:08:28 +0800 Subject: KVM: x86: Use FPU API Convert KVM to use generic FPU API. Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 17 +------------- arch/x86/kvm/x86.c | 52 +++++++++++++---------------------------- 2 files changed, 17 insertions(+), 52 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index d93601c52902..d08bb4a202de 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -301,7 +301,7 @@ struct kvm_vcpu_arch { unsigned long mmu_seq; } update_pte; - struct i387_fxsave_struct guest_fx_image; + struct fpu guest_fpu; gva_t mmio_fault_cr2; struct kvm_pio_request pio; @@ -708,21 +708,6 @@ static inline unsigned long read_msr(unsigned long msr) } #endif -static inline void kvm_fx_save(struct i387_fxsave_struct *image) -{ - asm("fxsave (%0)":: "r" (image)); -} - -static inline void kvm_fx_restore(struct i387_fxsave_struct *image) -{ - asm("fxrstor (%0)":: "r" (image)); -} - -static inline void kvm_fx_finit(void) -{ - asm("finit"); -} - static inline u32 get_rdx_init_val(void) { return 0x600; /* P6 family */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 54ce77582eda..84b1788489de 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -53,6 +53,7 @@ #include #include #include +#include #define MAX_IO_MSRS 256 #define CR0_RESERVED_BITS \ @@ -5057,27 +5058,6 @@ out: return r; } -/* - * fxsave fpu state. Taken from x86_64/processor.h. To be killed when - * we have asm/x86/processor.h - */ -struct fxsave { - u16 cwd; - u16 swd; - u16 twd; - u16 fop; - u64 rip; - u64 rdp; - u32 mxcsr; - u32 mxcsr_mask; - u32 st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */ -#ifdef CONFIG_X86_64 - u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg = 256 bytes */ -#else - u32 xmm_space[32]; /* 8*16 bytes for each XMM-reg = 128 bytes */ -#endif -}; - /* * Translate a guest virtual address to a guest physical address. */ @@ -5101,7 +5081,8 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) { - struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image; + struct i387_fxsave_struct *fxsave = + &vcpu->arch.guest_fpu.state->fxsave; memcpy(fpu->fpr, fxsave->st_space, 128); fpu->fcw = fxsave->cwd; @@ -5117,7 +5098,8 @@ int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) { - struct fxsave *fxsave = (struct fxsave *)&vcpu->arch.guest_fx_image; + struct i387_fxsave_struct *fxsave = + &vcpu->arch.guest_fpu.state->fxsave; memcpy(fxsave->st_space, fpu->fpr, 128); fxsave->cwd = fpu->fcw; @@ -5133,22 +5115,18 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) void fx_init(struct kvm_vcpu *vcpu) { - unsigned after_mxcsr_mask; - - /* Initialize guest FPU by resetting ours and saving into guest's */ - preempt_disable(); - kvm_fx_finit(); - kvm_fx_save(&vcpu->arch.guest_fx_image); - preempt_enable(); + fpu_alloc(&vcpu->arch.guest_fpu); + fpu_finit(&vcpu->arch.guest_fpu); vcpu->arch.cr0 |= X86_CR0_ET; - after_mxcsr_mask = offsetof(struct i387_fxsave_struct, st_space); - vcpu->arch.guest_fx_image.mxcsr = 0x1f80; - memset((void *)&vcpu->arch.guest_fx_image + after_mxcsr_mask, - 0, sizeof(struct i387_fxsave_struct) - after_mxcsr_mask); } EXPORT_SYMBOL_GPL(fx_init); +static void fx_free(struct kvm_vcpu *vcpu) +{ + fpu_free(&vcpu->arch.guest_fpu); +} + void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) { if (vcpu->guest_fpu_loaded) @@ -5156,7 +5134,7 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) vcpu->guest_fpu_loaded = 1; unlazy_fpu(current); - kvm_fx_restore(&vcpu->arch.guest_fx_image); + fpu_restore_checking(&vcpu->arch.guest_fpu); trace_kvm_fpu(1); } @@ -5166,7 +5144,7 @@ void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) return; vcpu->guest_fpu_loaded = 0; - kvm_fx_save(&vcpu->arch.guest_fx_image); + fpu_save_init(&vcpu->arch.guest_fpu); ++vcpu->stat.fpu_reload; set_bit(KVM_REQ_DEACTIVATE_FPU, &vcpu->requests); trace_kvm_fpu(0); @@ -5179,6 +5157,7 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) vcpu->arch.time_page = NULL; } + fx_free(vcpu); kvm_x86_ops->vcpu_free(vcpu); } @@ -5213,6 +5192,7 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) kvm_mmu_unload(vcpu); vcpu_put(vcpu); + fx_free(vcpu); kvm_x86_ops->vcpu_free(vcpu); } -- cgit v1.2.3 From c8174f7b35b3018c4c7b3237ed1c792e454fd5c3 Mon Sep 17 00:00:00 2001 From: Mohammed Gamal Date: Mon, 24 May 2010 01:01:04 +0300 Subject: KVM: VMX: Add constant for invalid guest state exit reason For the sake of completeness, this patch adds a symbolic constant for VMX exit reason 0x21 (invalid guest state). Signed-off-by: Mohammed Gamal Signed-off-by: Avi Kivity --- arch/x86/include/asm/vmx.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 9e6779f7cf2d..104cf86a7562 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -257,6 +257,7 @@ enum vmcs_field { #define EXIT_REASON_IO_INSTRUCTION 30 #define EXIT_REASON_MSR_READ 31 #define EXIT_REASON_MSR_WRITE 32 +#define EXIT_REASON_INVALID_STATE 33 #define EXIT_REASON_MWAIT_INSTRUCTION 36 #define EXIT_REASON_MONITOR_INSTRUCTION 39 #define EXIT_REASON_PAUSE_INSTRUCTION 40 -- cgit v1.2.3 From 10ab25cd6bf7ee4e5a55d81f203f7dc1a855c27e Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Tue, 25 May 2010 16:01:50 +0200 Subject: KVM: x86: Propagate fpu_alloc errors Memory allocation may fail. Propagate such errors. Signed-off-by: Jan Kiszka Reviewed-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/svm.c | 7 ++++++- arch/x86/kvm/vmx.c | 4 +++- arch/x86/kvm/x86.c | 11 +++++++++-- 4 files changed, 19 insertions(+), 5 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index d08bb4a202de..0cd0f2923af5 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -624,7 +624,7 @@ int kvm_pic_set_irq(void *opaque, int irq, int level); void kvm_inject_nmi(struct kvm_vcpu *vcpu); -void fx_init(struct kvm_vcpu *vcpu); +int fx_init(struct kvm_vcpu *vcpu); void kvm_mmu_flush_tlb(struct kvm_vcpu *vcpu); void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 9c68a650f57e..2ae0c3923293 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -904,13 +904,18 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id) svm->asid_generation = 0; init_vmcb(svm); - fx_init(&svm->vcpu); + err = fx_init(&svm->vcpu); + if (err) + goto free_page4; + svm->vcpu.arch.apic_base = 0xfee00000 | MSR_IA32_APICBASE_ENABLE; if (kvm_vcpu_is_bsp(&svm->vcpu)) svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP; return &svm->vcpu; +free_page4: + __free_page(hsave_page); free_page3: __free_pages(nested_msrpm_pages, MSRPM_ALLOC_ORDER); free_page2: diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 9c3ffc5fde44..e71c731433ee 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2659,7 +2659,9 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) msr |= MSR_IA32_APICBASE_BSP; kvm_set_apic_base(&vmx->vcpu, msr); - fx_init(&vmx->vcpu); + ret = fx_init(&vmx->vcpu); + if (ret != 0) + goto out; seg_setup(VCPU_SREG_CS); /* diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 033b9c207f93..e6e0d7781af7 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5114,12 +5114,19 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) return 0; } -void fx_init(struct kvm_vcpu *vcpu) +int fx_init(struct kvm_vcpu *vcpu) { - fpu_alloc(&vcpu->arch.guest_fpu); + int err; + + err = fpu_alloc(&vcpu->arch.guest_fpu); + if (err) + return err; + fpu_finit(&vcpu->arch.guest_fpu); vcpu->arch.cr0 |= X86_CR0_ET; + + return 0; } EXPORT_SYMBOL_GPL(fx_init); -- cgit v1.2.3 From 518c8aee5ca74fc03273fc6b4893cf456d65d545 Mon Sep 17 00:00:00 2001 From: Gui Jianfeng Date: Fri, 4 Jun 2010 08:51:39 +0800 Subject: KVM: VMX: Make sure single type invvpid is supported before issuing invvpid instruction According to SDM, we need check whether single-context INVVPID type is supported before issuing invvpid instruction. Signed-off-by: Gui Jianfeng Reviewed-by: Sheng Yang Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/vmx.h | 2 ++ arch/x86/kvm/vmx.c | 8 +++++++- 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 104cf86a7562..b4e28400c9ff 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -376,6 +376,8 @@ enum vmcs_field { #define VMX_EPT_EXTENT_CONTEXT_BIT (1ull << 25) #define VMX_EPT_EXTENT_GLOBAL_BIT (1ull << 26) +#define VMX_VPID_EXTENT_SINGLE_CONTEXT_BIT (1ull << 9) /* (41 - 32) */ + #define VMX_EPT_DEFAULT_GAW 3 #define VMX_EPT_MAX_GAW 0x4 #define VMX_EPT_MT_EPTE_SHIFT 3 diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 2201e3816205..945265361885 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -360,6 +360,11 @@ static inline bool cpu_has_vmx_invept_global(void) return vmx_capability.ept & VMX_EPT_EXTENT_GLOBAL_BIT; } +static inline bool cpu_has_vmx_invvpid_single(void) +{ + return vmx_capability.vpid & VMX_VPID_EXTENT_SINGLE_CONTEXT_BIT; +} + static inline bool cpu_has_vmx_ept(void) { return vmcs_config.cpu_based_2nd_exec_ctrl & @@ -504,7 +509,8 @@ static inline void vpid_sync_vcpu_all(struct vcpu_vmx *vmx) if (vmx->vpid == 0) return; - __invvpid(VMX_VPID_EXTENT_SINGLE_CONTEXT, vmx->vpid, 0); + if (cpu_has_vmx_invvpid_single()) + __invvpid(VMX_VPID_EXTENT_SINGLE_CONTEXT, vmx->vpid, 0); } static inline void ept_sync_global(void) -- cgit v1.2.3 From b9d762fa79f541ab480cdb733b46fdb0b4471c2d Mon Sep 17 00:00:00 2001 From: Gui Jianfeng Date: Mon, 7 Jun 2010 10:32:29 +0800 Subject: KVM: VMX: Add all-context INVVPID type support Add all-context INVVPID type support. Signed-off-by: Gui Jianfeng Signed-off-by: Avi Kivity --- arch/x86/include/asm/vmx.h | 1 + arch/x86/kvm/vmx.c | 23 +++++++++++++++++++++-- 2 files changed, 22 insertions(+), 2 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index b4e28400c9ff..96a5886d384e 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -377,6 +377,7 @@ enum vmcs_field { #define VMX_EPT_EXTENT_GLOBAL_BIT (1ull << 26) #define VMX_VPID_EXTENT_SINGLE_CONTEXT_BIT (1ull << 9) /* (41 - 32) */ +#define VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT (1ull << 10) /* (42 - 32) */ #define VMX_EPT_DEFAULT_GAW 3 #define VMX_EPT_MAX_GAW 0x4 diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 945265361885..622d83b0caf3 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -365,6 +365,11 @@ static inline bool cpu_has_vmx_invvpid_single(void) return vmx_capability.vpid & VMX_VPID_EXTENT_SINGLE_CONTEXT_BIT; } +static inline bool cpu_has_vmx_invvpid_global(void) +{ + return vmx_capability.vpid & VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT; +} + static inline bool cpu_has_vmx_ept(void) { return vmcs_config.cpu_based_2nd_exec_ctrl & @@ -513,6 +518,20 @@ static inline void vpid_sync_vcpu_all(struct vcpu_vmx *vmx) __invvpid(VMX_VPID_EXTENT_SINGLE_CONTEXT, vmx->vpid, 0); } +static inline void vpid_sync_vcpu_global(void) +{ + if (cpu_has_vmx_invvpid_global()) + __invvpid(VMX_VPID_EXTENT_ALL_CONTEXT, 0, 0); +} + +static inline void vpid_sync_context(struct vcpu_vmx *vmx) +{ + if (cpu_has_vmx_invvpid_single()) + vpid_sync_vcpu_all(vmx); + else + vpid_sync_vcpu_global(); +} + static inline void ept_sync_global(void) { if (cpu_has_vmx_invept_global()) @@ -1800,7 +1819,7 @@ static void exit_lmode(struct kvm_vcpu *vcpu) static void vmx_flush_tlb(struct kvm_vcpu *vcpu) { - vpid_sync_vcpu_all(to_vmx(vcpu)); + vpid_sync_context(to_vmx(vcpu)); if (enable_ept) ept_sync_context(construct_eptp(vcpu->arch.mmu.root_hpa)); } @@ -2756,7 +2775,7 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) vmx_fpu_activate(&vmx->vcpu); update_exception_bitmap(&vmx->vcpu); - vpid_sync_vcpu_all(vmx); + vpid_sync_context(vmx); ret = 0; -- cgit v1.2.3 From 2acf923e38fb6a4ce0c57115decbb38d334902ac Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Thu, 10 Jun 2010 11:27:12 +0800 Subject: KVM: VMX: Enable XSAVE/XRSTOR for guest This patch enable guest to use XSAVE/XRSTOR instructions. We assume that host_xcr0 would use all possible bits that OS supported. And we loaded xcr0 in the same way we handled fpu - do it as late as we can. Signed-off-by: Dexuan Cui Signed-off-by: Sheng Yang Reviewed-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 2 + arch/x86/include/asm/vmx.h | 1 + arch/x86/kvm/kvm_cache_regs.h | 6 ++ arch/x86/kvm/vmx.c | 13 ++++ arch/x86/kvm/x86.c | 130 +++++++++++++++++++++++++++++++++++++--- include/linux/kvm_host.h | 2 +- 6 files changed, 146 insertions(+), 8 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 0cd0f2923af5..91631b8b2090 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -302,6 +302,7 @@ struct kvm_vcpu_arch { } update_pte; struct fpu guest_fpu; + u64 xcr0; gva_t mmio_fault_cr2; struct kvm_pio_request pio; @@ -605,6 +606,7 @@ int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val); unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu); void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw); void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l); +int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr); int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata); int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data); diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 96a5886d384e..9f0cbd987d50 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -267,6 +267,7 @@ enum vmcs_field { #define EXIT_REASON_EPT_VIOLATION 48 #define EXIT_REASON_EPT_MISCONFIG 49 #define EXIT_REASON_WBINVD 54 +#define EXIT_REASON_XSETBV 55 /* * Interruption-information format diff --git a/arch/x86/kvm/kvm_cache_regs.h b/arch/x86/kvm/kvm_cache_regs.h index d2a98f8f9af5..6491ac8e755b 100644 --- a/arch/x86/kvm/kvm_cache_regs.h +++ b/arch/x86/kvm/kvm_cache_regs.h @@ -71,4 +71,10 @@ static inline ulong kvm_read_cr4(struct kvm_vcpu *vcpu) return kvm_read_cr4_bits(vcpu, ~0UL); } +static inline u64 kvm_read_edx_eax(struct kvm_vcpu *vcpu) +{ + return (kvm_register_read(vcpu, VCPU_REGS_RAX) & -1u) + | ((u64)(kvm_register_read(vcpu, VCPU_REGS_RDX) & -1u) << 32); +} + #endif diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 26ba61d6af8c..864a1b6d155a 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -37,6 +37,8 @@ #include #include #include +#include +#include #include "trace.h" @@ -3390,6 +3392,16 @@ static int handle_wbinvd(struct kvm_vcpu *vcpu) return 1; } +static int handle_xsetbv(struct kvm_vcpu *vcpu) +{ + u64 new_bv = kvm_read_edx_eax(vcpu); + u32 index = kvm_register_read(vcpu, VCPU_REGS_RCX); + + if (kvm_set_xcr(vcpu, index, new_bv) == 0) + skip_emulated_instruction(vcpu); + return 1; +} + static int handle_apic_access(struct kvm_vcpu *vcpu) { return emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DONE; @@ -3668,6 +3680,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { [EXIT_REASON_TPR_BELOW_THRESHOLD] = handle_tpr_below_threshold, [EXIT_REASON_APIC_ACCESS] = handle_apic_access, [EXIT_REASON_WBINVD] = handle_wbinvd, + [EXIT_REASON_XSETBV] = handle_xsetbv, [EXIT_REASON_TASK_SWITCH] = handle_task_switch, [EXIT_REASON_MCE_DURING_VMENTRY] = handle_machine_check, [EXIT_REASON_EPT_VIOLATION] = handle_ept_violation, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b08c0052e332..b5e644701cc1 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -65,6 +65,7 @@ (~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\ | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE \ | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR \ + | X86_CR4_OSXSAVE \ | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE)) #define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR) @@ -150,6 +151,13 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { NULL } }; +u64 __read_mostly host_xcr0; + +static inline u32 bit(int bitno) +{ + return 1 << (bitno & 31); +} + static void kvm_on_user_return(struct user_return_notifier *urn) { unsigned slot; @@ -474,6 +482,61 @@ void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw) } EXPORT_SYMBOL_GPL(kvm_lmsw); +int __kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr) +{ + u64 xcr0; + + /* Only support XCR_XFEATURE_ENABLED_MASK(xcr0) now */ + if (index != XCR_XFEATURE_ENABLED_MASK) + return 1; + xcr0 = xcr; + if (kvm_x86_ops->get_cpl(vcpu) != 0) + return 1; + if (!(xcr0 & XSTATE_FP)) + return 1; + if ((xcr0 & XSTATE_YMM) && !(xcr0 & XSTATE_SSE)) + return 1; + if (xcr0 & ~host_xcr0) + return 1; + vcpu->arch.xcr0 = xcr0; + vcpu->guest_xcr0_loaded = 0; + return 0; +} + +int kvm_set_xcr(struct kvm_vcpu *vcpu, u32 index, u64 xcr) +{ + if (__kvm_set_xcr(vcpu, index, xcr)) { + kvm_inject_gp(vcpu, 0); + return 1; + } + return 0; +} +EXPORT_SYMBOL_GPL(kvm_set_xcr); + +static bool guest_cpuid_has_xsave(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 1, 0); + return best && (best->ecx & bit(X86_FEATURE_XSAVE)); +} + +static void update_cpuid(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 1, 0); + if (!best) + return; + + /* Update OSXSAVE bit */ + if (cpu_has_xsave && best->function == 0x1) { + best->ecx &= ~(bit(X86_FEATURE_OSXSAVE)); + if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE)) + best->ecx |= bit(X86_FEATURE_OSXSAVE); + } +} + int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { unsigned long old_cr4 = kvm_read_cr4(vcpu); @@ -482,6 +545,9 @@ int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) if (cr4 & CR4_RESERVED_BITS) return 1; + if (!guest_cpuid_has_xsave(vcpu) && (cr4 & X86_CR4_OSXSAVE)) + return 1; + if (is_long_mode(vcpu)) { if (!(cr4 & X86_CR4_PAE)) return 1; @@ -498,6 +564,9 @@ int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) if ((cr4 ^ old_cr4) & pdptr_bits) kvm_mmu_reset_context(vcpu); + if ((cr4 ^ old_cr4) & X86_CR4_OSXSAVE) + update_cpuid(vcpu); + return 0; } @@ -666,11 +735,6 @@ int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val) } EXPORT_SYMBOL_GPL(kvm_get_dr); -static inline u32 bit(int bitno) -{ - return 1 << (bitno & 31); -} - /* * List of msr numbers which we expose to userspace through KVM_GET_MSRS * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST. @@ -1814,6 +1878,7 @@ static int kvm_vcpu_ioctl_set_cpuid(struct kvm_vcpu *vcpu, r = 0; kvm_apic_set_version(vcpu); kvm_x86_ops->cpuid_update(vcpu); + update_cpuid(vcpu); out_free: vfree(cpuid_entries); @@ -1837,6 +1902,7 @@ static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu, vcpu->arch.cpuid_nent = cpuid->nent; kvm_apic_set_version(vcpu); kvm_x86_ops->cpuid_update(vcpu); + update_cpuid(vcpu); return 0; out: @@ -1917,7 +1983,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, 0 /* Reserved */ | F(CX16) | 0 /* xTPR Update, PDCM */ | 0 /* Reserved, DCA */ | F(XMM4_1) | F(XMM4_2) | F(X2APIC) | F(MOVBE) | F(POPCNT) | - 0 /* Reserved, XSAVE, OSXSAVE */; + 0 /* Reserved, AES */ | F(XSAVE) | 0 /* OSXSAVE */; /* cpuid 0x80000001.ecx */ const u32 kvm_supported_word6_x86_features = F(LAHF_LM) | F(CMP_LEGACY) | F(SVM) | 0 /* ExtApicSpace */ | @@ -1932,7 +1998,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, switch (function) { case 0: - entry->eax = min(entry->eax, (u32)0xb); + entry->eax = min(entry->eax, (u32)0xd); break; case 1: entry->edx &= kvm_supported_word0_x86_features; @@ -1990,6 +2056,20 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, } break; } + case 0xd: { + int i; + + entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + for (i = 1; *nent < maxnent; ++i) { + if (entry[i - 1].eax == 0 && i != 2) + break; + do_cpuid_1_ent(&entry[i], function, i); + entry[i].flags |= + KVM_CPUID_FLAG_SIGNIFCANT_INDEX; + ++*nent; + } + break; + } case KVM_CPUID_SIGNATURE: { char signature[12] = "KVMKVMKVM\0\0"; u32 *sigptr = (u32 *)signature; @@ -4125,6 +4205,9 @@ int kvm_arch_init(void *opaque) perf_register_guest_info_callbacks(&kvm_guest_cbs); + if (cpu_has_xsave) + host_xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); + return 0; out: @@ -4523,6 +4606,25 @@ static void inject_pending_event(struct kvm_vcpu *vcpu) } } +static void kvm_load_guest_xcr0(struct kvm_vcpu *vcpu) +{ + if (kvm_read_cr4_bits(vcpu, X86_CR4_OSXSAVE) && + !vcpu->guest_xcr0_loaded) { + /* kvm_set_xcr() also depends on this */ + xsetbv(XCR_XFEATURE_ENABLED_MASK, vcpu->arch.xcr0); + vcpu->guest_xcr0_loaded = 1; + } +} + +static void kvm_put_guest_xcr0(struct kvm_vcpu *vcpu) +{ + if (vcpu->guest_xcr0_loaded) { + if (vcpu->arch.xcr0 != host_xcr0) + xsetbv(XCR_XFEATURE_ENABLED_MASK, host_xcr0); + vcpu->guest_xcr0_loaded = 0; + } +} + static int vcpu_enter_guest(struct kvm_vcpu *vcpu) { int r; @@ -4568,6 +4670,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_x86_ops->prepare_guest_switch(vcpu); if (vcpu->fpu_active) kvm_load_guest_fpu(vcpu); + kvm_load_guest_xcr0(vcpu); atomic_set(&vcpu->guest_mode, 1); smp_wmb(); @@ -5124,6 +5227,11 @@ int fx_init(struct kvm_vcpu *vcpu) fpu_finit(&vcpu->arch.guest_fpu); + /* + * Ensure guest xcr0 is valid for loading + */ + vcpu->arch.xcr0 = XSTATE_FP; + vcpu->arch.cr0 |= X86_CR0_ET; return 0; @@ -5140,6 +5248,12 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) if (vcpu->guest_fpu_loaded) return; + /* + * Restore all possible states in the guest, + * and assume host would use all available bits. + * Guest xcr0 would be loaded later. + */ + kvm_put_guest_xcr0(vcpu); vcpu->guest_fpu_loaded = 1; unlazy_fpu(current); fpu_restore_checking(&vcpu->arch.guest_fpu); @@ -5148,6 +5262,8 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) { + kvm_put_guest_xcr0(vcpu); + if (!vcpu->guest_fpu_loaded) return; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 2c62319727ef..2d96555cd4ed 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -88,7 +88,7 @@ struct kvm_vcpu { int srcu_idx; int fpu_active; - int guest_fpu_loaded; + int guest_fpu_loaded, guest_xcr0_loaded; wait_queue_head_t wq; int sigset_active; sigset_t sigset; -- cgit v1.2.3 From 49a9b07edcf4aff159c1f3d3a27e58cf38bc27cd Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 10 Jun 2010 17:02:14 +0300 Subject: KVM: Fix mov cr0 #GP at wrong instruction On Intel, we call skip_emulated_instruction() even if we injected a #GP, resulting in the #GP pointing at the wrong address. Fix by injecting the exception and skipping the instruction at the same place, so we can do just one or the other. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/svm.c | 2 +- arch/x86/kvm/vmx.c | 13 +++++++++++-- arch/x86/kvm/x86.c | 12 +++--------- 4 files changed, 16 insertions(+), 13 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 91631b8b2090..b23708450210 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -597,7 +597,7 @@ int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg); int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason, bool has_error_code, u32 error_code); -void kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); +int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3); void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); void kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 2ae0c3923293..6d1616d47c54 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -807,7 +807,7 @@ static void init_vmcb(struct vcpu_svm *svm) * svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0. */ svm->vcpu.arch.cr0 = X86_CR0_NW | X86_CR0_CD | X86_CR0_ET; - kvm_set_cr0(&svm->vcpu, svm->vcpu.arch.cr0); + (void)kvm_set_cr0(&svm->vcpu, svm->vcpu.arch.cr0); save->cr4 = X86_CR4_PAE; /* rdx = ?? */ diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 864a1b6d155a..1baf4b2d98ee 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3157,11 +3157,20 @@ vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall) hypercall[2] = 0xc1; } +static void complete_insn_gp(struct kvm_vcpu *vcpu, int err) +{ + if (err) + kvm_inject_gp(vcpu, 0); + else + skip_emulated_instruction(vcpu); +} + static int handle_cr(struct kvm_vcpu *vcpu) { unsigned long exit_qualification, val; int cr; int reg; + int err; exit_qualification = vmcs_readl(EXIT_QUALIFICATION); cr = exit_qualification & 15; @@ -3172,8 +3181,8 @@ static int handle_cr(struct kvm_vcpu *vcpu) trace_kvm_cr_write(cr, val); switch (cr) { case 0: - kvm_set_cr0(vcpu, val); - skip_emulated_instruction(vcpu); + err = kvm_set_cr0(vcpu, val); + complete_insn_gp(vcpu, err); return 1; case 3: kvm_set_cr3(vcpu, val); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b5e644701cc1..05e9b5dde646 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -425,7 +425,7 @@ out: return changed; } -static int __kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) +int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { unsigned long old_cr0 = kvm_read_cr0(vcpu); unsigned long update_bits = X86_CR0_PG | X86_CR0_WP | @@ -468,17 +468,11 @@ static int __kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) kvm_mmu_reset_context(vcpu); return 0; } - -void kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) -{ - if (__kvm_set_cr0(vcpu, cr0)) - kvm_inject_gp(vcpu, 0); -} EXPORT_SYMBOL_GPL(kvm_set_cr0); void kvm_lmsw(struct kvm_vcpu *vcpu, unsigned long msw) { - kvm_set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~0x0eul) | (msw & 0x0f)); + (void)kvm_set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~0x0eul) | (msw & 0x0f)); } EXPORT_SYMBOL_GPL(kvm_lmsw); @@ -3732,7 +3726,7 @@ static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) switch (cr) { case 0: - res = __kvm_set_cr0(vcpu, mk_cr_64(kvm_read_cr0(vcpu), val)); + res = kvm_set_cr0(vcpu, mk_cr_64(kvm_read_cr0(vcpu), val)); break; case 2: vcpu->arch.cr2 = val; -- cgit v1.2.3 From a83b29c6ad6d6497e569edbc29e556a384cebddd Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 10 Jun 2010 17:02:15 +0300 Subject: KVM: Fix mov cr4 #GP at wrong instruction On Intel, we call skip_emulated_instruction() even if we injected a #GP, resulting in the #GP pointing at the wrong address. Fix by injecting the exception and skipping the instruction at the same place, so we can do just one or the other. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/vmx.c | 4 ++-- arch/x86/kvm/x86.c | 10 ++-------- 3 files changed, 5 insertions(+), 11 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index b23708450210..ea8c319cdffc 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -599,7 +599,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason, int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3); -void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); +int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); void kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8); int kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val); int kvm_get_dr(struct kvm_vcpu *vcpu, int dr, unsigned long *val); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 1baf4b2d98ee..f64d65dc38c6 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3189,8 +3189,8 @@ static int handle_cr(struct kvm_vcpu *vcpu) skip_emulated_instruction(vcpu); return 1; case 4: - kvm_set_cr4(vcpu, val); - skip_emulated_instruction(vcpu); + err = kvm_set_cr4(vcpu, val); + complete_insn_gp(vcpu, err); return 1; case 8: { u8 cr8_prev = kvm_get_cr8(vcpu); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 05e9b5dde646..ed3af15d4404 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -531,7 +531,7 @@ static void update_cpuid(struct kvm_vcpu *vcpu) } } -int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) +int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) { unsigned long old_cr4 = kvm_read_cr4(vcpu); unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE; @@ -563,12 +563,6 @@ int __kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) return 0; } - -void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) -{ - if (__kvm_set_cr4(vcpu, cr4)) - kvm_inject_gp(vcpu, 0); -} EXPORT_SYMBOL_GPL(kvm_set_cr4); static int __kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) @@ -3735,7 +3729,7 @@ static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) res = __kvm_set_cr3(vcpu, val); break; case 4: - res = __kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); + res = kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); break; case 8: res = __kvm_set_cr8(vcpu, val & 0xfUL); -- cgit v1.2.3 From 2390218b6aa2eb3784b0a82fa811c19097dc793a Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 10 Jun 2010 17:02:16 +0300 Subject: KVM: Fix mov cr3 #GP at wrong instruction On Intel, we call skip_emulated_instruction() even if we injected a #GP, resulting in the #GP pointing at the wrong address. Fix by injecting the exception and skipping the instruction at the same place, so we can do just one or the other. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/mmu.c | 2 +- arch/x86/kvm/svm.c | 4 ++-- arch/x86/kvm/vmx.c | 4 ++-- arch/x86/kvm/x86.c | 10 ++-------- 5 files changed, 8 insertions(+), 14 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index ea8c319cdffc..c2813d658f3e 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -598,7 +598,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason, bool has_error_code, u32 error_code); int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); -void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3); +int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3); int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4); void kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8); int kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 4706a936e36f..aa98fca03ed7 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3203,7 +3203,7 @@ static int kvm_pv_mmu_write(struct kvm_vcpu *vcpu, static int kvm_pv_mmu_flush_tlb(struct kvm_vcpu *vcpu) { - kvm_set_cr3(vcpu, vcpu->arch.cr3); + (void)kvm_set_cr3(vcpu, vcpu->arch.cr3); return 1; } diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 6d1616d47c54..f7a6fdcf8ef3 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1963,7 +1963,7 @@ static int nested_svm_vmexit(struct vcpu_svm *svm) svm->vmcb->save.cr3 = hsave->save.cr3; svm->vcpu.arch.cr3 = hsave->save.cr3; } else { - kvm_set_cr3(&svm->vcpu, hsave->save.cr3); + (void)kvm_set_cr3(&svm->vcpu, hsave->save.cr3); } kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, hsave->save.rax); kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, hsave->save.rsp); @@ -2086,7 +2086,7 @@ static bool nested_svm_vmrun(struct vcpu_svm *svm) svm->vmcb->save.cr3 = nested_vmcb->save.cr3; svm->vcpu.arch.cr3 = nested_vmcb->save.cr3; } else - kvm_set_cr3(&svm->vcpu, nested_vmcb->save.cr3); + (void)kvm_set_cr3(&svm->vcpu, nested_vmcb->save.cr3); /* Guest paging mode is active - reset mmu */ kvm_mmu_reset_context(&svm->vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index f64d65dc38c6..345a35470511 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3185,8 +3185,8 @@ static int handle_cr(struct kvm_vcpu *vcpu) complete_insn_gp(vcpu, err); return 1; case 3: - kvm_set_cr3(vcpu, val); - skip_emulated_instruction(vcpu); + err = kvm_set_cr3(vcpu, val); + complete_insn_gp(vcpu, err); return 1; case 4: err = kvm_set_cr4(vcpu, val); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ed3af15d4404..795999e1ac19 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -565,7 +565,7 @@ int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) } EXPORT_SYMBOL_GPL(kvm_set_cr4); -static int __kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) +int kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { if (cr3 == vcpu->arch.cr3 && !pdptrs_changed(vcpu)) { kvm_mmu_sync_roots(vcpu); @@ -604,12 +604,6 @@ static int __kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) vcpu->arch.mmu.new_cr3(vcpu); return 0; } - -void kvm_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) -{ - if (__kvm_set_cr3(vcpu, cr3)) - kvm_inject_gp(vcpu, 0); -} EXPORT_SYMBOL_GPL(kvm_set_cr3); int __kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8) @@ -3726,7 +3720,7 @@ static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) vcpu->arch.cr2 = val; break; case 3: - res = __kvm_set_cr3(vcpu, val); + res = kvm_set_cr3(vcpu, val); break; case 4: res = kvm_set_cr4(vcpu, mk_cr_64(kvm_read_cr4(vcpu), val)); -- cgit v1.2.3 From 2d5b5a665508c60577c1088e0405850a965b6795 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Sun, 13 Jun 2010 17:29:39 +0800 Subject: KVM: x86: XSAVE/XRSTOR live migration support This patch enable save/restore of xsave state. Signed-off-by: Sheng Yang Signed-off-by: Marcelo Tosatti --- Documentation/kvm/api.txt | 74 +++++++++++++++++++++++ arch/x86/include/asm/kvm.h | 22 +++++++ arch/x86/include/asm/xsave.h | 7 ++- arch/x86/kvm/x86.c | 139 +++++++++++++++++++++++++++++++++++++++++++ include/linux/kvm.h | 12 ++++ 5 files changed, 252 insertions(+), 2 deletions(-) (limited to 'arch/x86/include') diff --git a/Documentation/kvm/api.txt b/Documentation/kvm/api.txt index 159b4efe1b0e..ffba03f55bdf 100644 --- a/Documentation/kvm/api.txt +++ b/Documentation/kvm/api.txt @@ -922,6 +922,80 @@ Define which vcpu is the Bootstrap Processor (BSP). Values are the same as the vcpu id in KVM_CREATE_VCPU. If this ioctl is not called, the default is vcpu 0. +4.41 KVM_GET_XSAVE + +Capability: KVM_CAP_XSAVE +Architectures: x86 +Type: vcpu ioctl +Parameters: struct kvm_xsave (out) +Returns: 0 on success, -1 on error + +struct kvm_xsave { + __u32 region[1024]; +}; + +This ioctl would copy current vcpu's xsave struct to the userspace. + +4.42 KVM_SET_XSAVE + +Capability: KVM_CAP_XSAVE +Architectures: x86 +Type: vcpu ioctl +Parameters: struct kvm_xsave (in) +Returns: 0 on success, -1 on error + +struct kvm_xsave { + __u32 region[1024]; +}; + +This ioctl would copy userspace's xsave struct to the kernel. + +4.43 KVM_GET_XCRS + +Capability: KVM_CAP_XCRS +Architectures: x86 +Type: vcpu ioctl +Parameters: struct kvm_xcrs (out) +Returns: 0 on success, -1 on error + +struct kvm_xcr { + __u32 xcr; + __u32 reserved; + __u64 value; +}; + +struct kvm_xcrs { + __u32 nr_xcrs; + __u32 flags; + struct kvm_xcr xcrs[KVM_MAX_XCRS]; + __u64 padding[16]; +}; + +This ioctl would copy current vcpu's xcrs to the userspace. + +4.44 KVM_SET_XCRS + +Capability: KVM_CAP_XCRS +Architectures: x86 +Type: vcpu ioctl +Parameters: struct kvm_xcrs (in) +Returns: 0 on success, -1 on error + +struct kvm_xcr { + __u32 xcr; + __u32 reserved; + __u64 value; +}; + +struct kvm_xcrs { + __u32 nr_xcrs; + __u32 flags; + struct kvm_xcr xcrs[KVM_MAX_XCRS]; + __u64 padding[16]; +}; + +This ioctl would set vcpu's xcr to the value userspace specified. + 5. The kvm_run structure Application code obtains a pointer to the kvm_run structure by diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index ff90055c7f0b..4d8dcbdfc120 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -22,6 +22,8 @@ #define __KVM_HAVE_XEN_HVM #define __KVM_HAVE_VCPU_EVENTS #define __KVM_HAVE_DEBUGREGS +#define __KVM_HAVE_XSAVE +#define __KVM_HAVE_XCRS /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 @@ -299,4 +301,24 @@ struct kvm_debugregs { __u64 reserved[9]; }; +/* for KVM_CAP_XSAVE */ +struct kvm_xsave { + __u32 region[1024]; +}; + +#define KVM_MAX_XCRS 16 + +struct kvm_xcr { + __u32 xcr; + __u32 reserved; + __u64 value; +}; + +struct kvm_xcrs { + __u32 nr_xcrs; + __u32 flags; + struct kvm_xcr xcrs[KVM_MAX_XCRS]; + __u64 padding[16]; +}; + #endif /* _ASM_X86_KVM_H */ diff --git a/arch/x86/include/asm/xsave.h b/arch/x86/include/asm/xsave.h index 29ee4e4c64cf..32c36668fa7b 100644 --- a/arch/x86/include/asm/xsave.h +++ b/arch/x86/include/asm/xsave.h @@ -13,8 +13,11 @@ #define FXSAVE_SIZE 512 -#define XSTATE_YMM_SIZE 256 -#define XSTATE_YMM_OFFSET (512 + 64) +#define XSAVE_HDR_SIZE 64 +#define XSAVE_HDR_OFFSET FXSAVE_SIZE + +#define XSAVE_YMM_SIZE 256 +#define XSAVE_YMM_OFFSET (XSAVE_HDR_SIZE + XSAVE_HDR_OFFSET) /* * These are the features that the OS can handle currently. diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 795999e1ac19..0c8dc9614e7d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1680,6 +1680,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_PCI_SEGMENT: case KVM_CAP_DEBUGREGS: case KVM_CAP_X86_ROBUST_SINGLESTEP: + case KVM_CAP_XSAVE: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -1703,6 +1704,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_MCE: r = KVM_MAX_MCE_BANKS; break; + case KVM_CAP_XCRS: + r = cpu_has_xsave; + break; default: r = 0; break; @@ -2355,6 +2359,77 @@ static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu, return 0; } +static void kvm_vcpu_ioctl_x86_get_xsave(struct kvm_vcpu *vcpu, + struct kvm_xsave *guest_xsave) +{ + if (cpu_has_xsave) + memcpy(guest_xsave->region, + &vcpu->arch.guest_fpu.state->xsave, + sizeof(struct xsave_struct)); + else { + memcpy(guest_xsave->region, + &vcpu->arch.guest_fpu.state->fxsave, + sizeof(struct i387_fxsave_struct)); + *(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)] = + XSTATE_FPSSE; + } +} + +static int kvm_vcpu_ioctl_x86_set_xsave(struct kvm_vcpu *vcpu, + struct kvm_xsave *guest_xsave) +{ + u64 xstate_bv = + *(u64 *)&guest_xsave->region[XSAVE_HDR_OFFSET / sizeof(u32)]; + + if (cpu_has_xsave) + memcpy(&vcpu->arch.guest_fpu.state->xsave, + guest_xsave->region, sizeof(struct xsave_struct)); + else { + if (xstate_bv & ~XSTATE_FPSSE) + return -EINVAL; + memcpy(&vcpu->arch.guest_fpu.state->fxsave, + guest_xsave->region, sizeof(struct i387_fxsave_struct)); + } + return 0; +} + +static void kvm_vcpu_ioctl_x86_get_xcrs(struct kvm_vcpu *vcpu, + struct kvm_xcrs *guest_xcrs) +{ + if (!cpu_has_xsave) { + guest_xcrs->nr_xcrs = 0; + return; + } + + guest_xcrs->nr_xcrs = 1; + guest_xcrs->flags = 0; + guest_xcrs->xcrs[0].xcr = XCR_XFEATURE_ENABLED_MASK; + guest_xcrs->xcrs[0].value = vcpu->arch.xcr0; +} + +static int kvm_vcpu_ioctl_x86_set_xcrs(struct kvm_vcpu *vcpu, + struct kvm_xcrs *guest_xcrs) +{ + int i, r = 0; + + if (!cpu_has_xsave) + return -EINVAL; + + if (guest_xcrs->nr_xcrs > KVM_MAX_XCRS || guest_xcrs->flags) + return -EINVAL; + + for (i = 0; i < guest_xcrs->nr_xcrs; i++) + /* Only support XCR0 currently */ + if (guest_xcrs->xcrs[0].xcr == XCR_XFEATURE_ENABLED_MASK) { + r = __kvm_set_xcr(vcpu, XCR_XFEATURE_ENABLED_MASK, + guest_xcrs->xcrs[0].value); + break; + } + if (r) + r = -EINVAL; + return r; +} + long kvm_arch_vcpu_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { @@ -2556,6 +2631,70 @@ long kvm_arch_vcpu_ioctl(struct file *filp, r = kvm_vcpu_ioctl_x86_set_debugregs(vcpu, &dbgregs); break; } + case KVM_GET_XSAVE: { + struct kvm_xsave *xsave; + + xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL); + r = -ENOMEM; + if (!xsave) + break; + + kvm_vcpu_ioctl_x86_get_xsave(vcpu, xsave); + + r = -EFAULT; + if (copy_to_user(argp, xsave, sizeof(struct kvm_xsave))) + break; + r = 0; + break; + } + case KVM_SET_XSAVE: { + struct kvm_xsave *xsave; + + xsave = kzalloc(sizeof(struct kvm_xsave), GFP_KERNEL); + r = -ENOMEM; + if (!xsave) + break; + + r = -EFAULT; + if (copy_from_user(xsave, argp, sizeof(struct kvm_xsave))) + break; + + r = kvm_vcpu_ioctl_x86_set_xsave(vcpu, xsave); + break; + } + case KVM_GET_XCRS: { + struct kvm_xcrs *xcrs; + + xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL); + r = -ENOMEM; + if (!xcrs) + break; + + kvm_vcpu_ioctl_x86_get_xcrs(vcpu, xcrs); + + r = -EFAULT; + if (copy_to_user(argp, xcrs, + sizeof(struct kvm_xcrs))) + break; + r = 0; + break; + } + case KVM_SET_XCRS: { + struct kvm_xcrs *xcrs; + + xcrs = kzalloc(sizeof(struct kvm_xcrs), GFP_KERNEL); + r = -ENOMEM; + if (!xcrs) + break; + + r = -EFAULT; + if (copy_from_user(xcrs, argp, + sizeof(struct kvm_xcrs))) + break; + + r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, xcrs); + break; + } default: r = -EINVAL; } diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 23ea02253900..6fd40f540a8e 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -524,6 +524,12 @@ struct kvm_enable_cap { #define KVM_CAP_PPC_OSI 52 #define KVM_CAP_PPC_UNSET_IRQ 53 #define KVM_CAP_ENABLE_CAP 54 +#ifdef __KVM_HAVE_XSAVE +#define KVM_CAP_XSAVE 55 +#endif +#ifdef __KVM_HAVE_XCRS +#define KVM_CAP_XCRS 56 +#endif #ifdef KVM_CAP_IRQ_ROUTING @@ -714,6 +720,12 @@ struct kvm_clock_data { #define KVM_GET_DEBUGREGS _IOR(KVMIO, 0xa1, struct kvm_debugregs) #define KVM_SET_DEBUGREGS _IOW(KVMIO, 0xa2, struct kvm_debugregs) #define KVM_ENABLE_CAP _IOW(KVMIO, 0xa3, struct kvm_enable_cap) +/* Available with KVM_CAP_XSAVE */ +#define KVM_GET_XSAVE _IOR(KVMIO, 0xa4, struct kvm_xsave) +#define KVM_SET_XSAVE _IOW(KVMIO, 0xa5, struct kvm_xsave) +/* Available with KVM_CAP_XCRS */ +#define KVM_GET_XCRS _IOR(KVMIO, 0xa6, struct kvm_xcrs) +#define KVM_SET_XCRS _IOW(KVMIO, 0xa7, struct kvm_xcrs) #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0) -- cgit v1.2.3 From be71e061d15c0aad4f8c2606f76c57b8a19792fd Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Fri, 11 Jun 2010 21:31:38 +0800 Subject: KVM: MMU: don't mark pte notrap if it's just sync transient If the sync-sp just sync transient, don't mark its pte notrap Signed-off-by: Xiao Guangrong Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/mmu.c | 11 ++++------- arch/x86/kvm/paging_tmpl.h | 5 +++-- 3 files changed, 8 insertions(+), 10 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index c2813d658f3e..2ec2e27a403e 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -241,7 +241,7 @@ struct kvm_mmu { void (*prefetch_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page); int (*sync_page)(struct kvm_vcpu *vcpu, - struct kvm_mmu_page *sp); + struct kvm_mmu_page *sp, bool clear_unsync); void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva); hpa_t root_hpa; int root_level; diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index ff333572be75..d1e09f3c5614 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -1103,7 +1103,7 @@ static void nonpaging_prefetch_page(struct kvm_vcpu *vcpu, } static int nonpaging_sync_page(struct kvm_vcpu *vcpu, - struct kvm_mmu_page *sp) + struct kvm_mmu_page *sp, bool clear_unsync) { return 1; } @@ -1228,7 +1228,7 @@ static int __kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, if (clear_unsync) kvm_unlink_unsync_page(vcpu->kvm, sp); - if (vcpu->arch.mmu.sync_page(vcpu, sp)) { + if (vcpu->arch.mmu.sync_page(vcpu, sp, clear_unsync)) { kvm_mmu_prepare_zap_page(vcpu->kvm, sp, invalid_list); return 1; } @@ -1237,7 +1237,6 @@ static int __kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, return 0; } -static void mmu_convert_notrap(struct kvm_mmu_page *sp); static int kvm_sync_page_transient(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) { @@ -1245,9 +1244,7 @@ static int kvm_sync_page_transient(struct kvm_vcpu *vcpu, int ret; ret = __kvm_sync_page(vcpu, sp, &invalid_list, false); - if (!ret) - mmu_convert_notrap(sp); - else + if (ret) kvm_mmu_commit_zap_page(vcpu->kvm, &invalid_list); return ret; @@ -1273,7 +1270,7 @@ static void kvm_sync_pages(struct kvm_vcpu *vcpu, gfn_t gfn) WARN_ON(s->role.level != PT_PAGE_TABLE_LEVEL); if ((s->role.cr4_pae != !!is_pae(vcpu)) || - (vcpu->arch.mmu.sync_page(vcpu, s))) { + (vcpu->arch.mmu.sync_page(vcpu, s, true))) { kvm_mmu_prepare_zap_page(vcpu->kvm, s, &invalid_list); continue; } diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index efba353369e7..863920f649fb 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -578,7 +578,8 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, * can't change unless all sptes pointing to it are nuked first. * - Alias changes zap the entire shadow cache. */ -static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) +static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, + bool clear_unsync) { int i, offset, nr_present; bool reset_host_protection; @@ -615,7 +616,7 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) u64 nonpresent; rmap_remove(vcpu->kvm, &sp->spt[i]); - if (is_present_gpte(gpte)) + if (is_present_gpte(gpte) || !clear_unsync) nonpresent = shadow_trap_nonpresent_pte; else nonpresent = shadow_notrap_nonpresent_pte; -- cgit v1.2.3 From a1f4d39500ad8ed61825eff061debff42386ab5b Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 21 Jun 2010 11:44:20 +0300 Subject: KVM: Remove memory alias support As advertised in feature-removal-schedule.txt. Equivalent support is provided by overlapping memory regions. Signed-off-by: Avi Kivity --- Documentation/feature-removal-schedule.txt | 11 --- Documentation/kvm/api.txt | 12 +-- arch/ia64/kvm/kvm-ia64.c | 5 -- arch/powerpc/kvm/powerpc.c | 5 -- arch/s390/kvm/kvm-s390.c | 5 -- arch/x86/include/asm/kvm_host.h | 21 ----- arch/x86/kvm/mmu.c | 17 +--- arch/x86/kvm/paging_tmpl.h | 3 +- arch/x86/kvm/x86.c | 125 ----------------------------- arch/x86/kvm/x86.h | 7 -- include/linux/kvm.h | 1 + include/linux/kvm_host.h | 6 -- virt/kvm/kvm_main.c | 18 +---- 13 files changed, 11 insertions(+), 225 deletions(-) (limited to 'arch/x86/include') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 1571c0c83dba..ad1e90dd2780 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -538,17 +538,6 @@ Who: Jan Kiszka ---------------------------- -What: KVM memory aliases support -When: July 2010 -Why: Memory aliasing support is used for speeding up guest vga access - through the vga windows. - - Modern userspace no longer uses this feature, so it's just bitrotted - code and can be removed with no impact. -Who: Avi Kivity - ----------------------------- - What: xtime, wall_to_monotonic When: 2.6.36+ Files: kernel/time/timekeeping.c include/linux/time.h diff --git a/Documentation/kvm/api.txt b/Documentation/kvm/api.txt index ffba03f55bdf..7e415943a11e 100644 --- a/Documentation/kvm/api.txt +++ b/Documentation/kvm/api.txt @@ -226,17 +226,7 @@ Type: vm ioctl Parameters: struct kvm_memory_alias (in) Returns: 0 (success), -1 (error) -struct kvm_memory_alias { - __u32 slot; /* this has a different namespace than memory slots */ - __u32 flags; - __u64 guest_phys_addr; - __u64 memory_size; - __u64 target_phys_addr; -}; - -Defines a guest physical address space region as an alias to another -region. Useful for aliased address, for example the VGA low memory -window. Should not be used with userspace memory. +This ioctl is obsolete and has been removed. 4.9 KVM_RUN diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 91760e80e268..bd510beb43af 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1946,11 +1946,6 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) return vcpu->arch.timer_fired; } -gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) -{ - return gfn; -} - int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE) || diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index b5ebdfbed20b..72a4ad86ee91 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -36,11 +36,6 @@ #define CREATE_TRACE_POINTS #include "trace.h" -gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) -{ - return gfn; -} - int kvm_arch_vcpu_runnable(struct kvm_vcpu *v) { return !(v->arch.msr & MSR_WE) || !!(v->arch.pending_exceptions); diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 08a3b35d30be..4fe68650535c 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -723,11 +723,6 @@ void kvm_arch_flush_shadow(struct kvm *kvm) { } -gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) -{ - return gfn; -} - static int __init kvm_s390_init(void) { int ret; diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 2ec2e27a403e..a57cdeacc4d2 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -69,8 +69,6 @@ #define IOPL_SHIFT 12 -#define KVM_ALIAS_SLOTS 4 - #define KVM_PERMILLE_MMU_PAGES 20 #define KVM_MIN_ALLOC_MMU_PAGES 64 #define KVM_MMU_HASH_SHIFT 10 @@ -362,24 +360,7 @@ struct kvm_vcpu_arch { u64 hv_vapic; }; -struct kvm_mem_alias { - gfn_t base_gfn; - unsigned long npages; - gfn_t target_gfn; -#define KVM_ALIAS_INVALID 1UL - unsigned long flags; -}; - -#define KVM_ARCH_HAS_UNALIAS_INSTANTIATION - -struct kvm_mem_aliases { - struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS]; - int naliases; -}; - struct kvm_arch { - struct kvm_mem_aliases *aliases; - unsigned int n_free_mmu_pages; unsigned int n_requested_mmu_pages; unsigned int n_alloc_mmu_pages; @@ -655,8 +636,6 @@ void kvm_disable_tdp(void); int complete_pio(struct kvm_vcpu *vcpu); bool kvm_check_iopl(struct kvm_vcpu *vcpu); -struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn); - static inline struct kvm_mmu_page *page_header(hpa_t shadow_page) { struct page *page = pfn_to_page(shadow_page >> PAGE_SHIFT); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 8c2f580956d9..c5501bc10106 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -434,9 +434,7 @@ static void account_shadowed(struct kvm *kvm, gfn_t gfn) int *write_count; int i; - gfn = unalias_gfn(kvm, gfn); - - slot = gfn_to_memslot_unaliased(kvm, gfn); + slot = gfn_to_memslot(kvm, gfn); for (i = PT_DIRECTORY_LEVEL; i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { write_count = slot_largepage_idx(gfn, slot, i); @@ -450,8 +448,7 @@ static void unaccount_shadowed(struct kvm *kvm, gfn_t gfn) int *write_count; int i; - gfn = unalias_gfn(kvm, gfn); - slot = gfn_to_memslot_unaliased(kvm, gfn); + slot = gfn_to_memslot(kvm, gfn); for (i = PT_DIRECTORY_LEVEL; i < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++i) { write_count = slot_largepage_idx(gfn, slot, i); @@ -467,8 +464,7 @@ static int has_wrprotected_page(struct kvm *kvm, struct kvm_memory_slot *slot; int *largepage_idx; - gfn = unalias_gfn(kvm, gfn); - slot = gfn_to_memslot_unaliased(kvm, gfn); + slot = gfn_to_memslot(kvm, gfn); if (slot) { largepage_idx = slot_largepage_idx(gfn, slot, level); return *largepage_idx; @@ -521,7 +517,6 @@ static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn) /* * Take gfn and return the reverse mapping to it. - * Note: gfn must be unaliased before this function get called */ static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int level) @@ -561,7 +556,6 @@ static int rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) if (!is_rmap_spte(*spte)) return count; - gfn = unalias_gfn(vcpu->kvm, gfn); sp = page_header(__pa(spte)); kvm_mmu_page_set_gfn(sp, spte - sp->spt, gfn); rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level); @@ -698,7 +692,6 @@ static int rmap_write_protect(struct kvm *kvm, u64 gfn) u64 *spte; int i, write_protected = 0; - gfn = unalias_gfn(kvm, gfn); rmapp = gfn_to_rmap(kvm, gfn, PT_PAGE_TABLE_LEVEL); spte = rmap_next(kvm, rmapp, NULL); @@ -885,7 +878,6 @@ static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) sp = page_header(__pa(spte)); - gfn = unalias_gfn(vcpu->kvm, gfn); rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level); kvm_unmap_rmapp(vcpu->kvm, rmapp, 0); @@ -3510,8 +3502,7 @@ static void audit_write_protection(struct kvm_vcpu *vcpu) if (sp->unsync) continue; - gfn = unalias_gfn(vcpu->kvm, sp->gfn); - slot = gfn_to_memslot_unaliased(vcpu->kvm, sp->gfn); + slot = gfn_to_memslot(vcpu->kvm, sp->gfn); rmapp = &slot->rmap[gfn - slot->base_gfn]; spte = rmap_next(vcpu->kvm, rmapp, NULL); diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 863920f649fb..a21a86ef9e20 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -576,7 +576,6 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, * Using the cached information from sp->gfns is safe because: * - The spte has a reference to the struct page, so the pfn for a given gfn * can't change unless all sptes pointing to it are nuked first. - * - Alias changes zap the entire shadow cache. */ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, bool clear_unsync) @@ -611,7 +610,7 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, return -EINVAL; gfn = gpte_to_gfn(gpte); - if (unalias_gfn(vcpu->kvm, gfn) != sp->gfns[i] || + if (gfn != sp->gfns[i] || !is_present_gpte(gpte) || !(gpte & PT_ACCESSED_MASK)) { u64 nonpresent; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8e60b6c9c0b0..62596d373a49 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2740,115 +2740,6 @@ static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm) return kvm->arch.n_alloc_mmu_pages; } -gfn_t unalias_gfn_instantiation(struct kvm *kvm, gfn_t gfn) -{ - int i; - struct kvm_mem_alias *alias; - struct kvm_mem_aliases *aliases; - - aliases = kvm_aliases(kvm); - - for (i = 0; i < aliases->naliases; ++i) { - alias = &aliases->aliases[i]; - if (alias->flags & KVM_ALIAS_INVALID) - continue; - if (gfn >= alias->base_gfn - && gfn < alias->base_gfn + alias->npages) - return alias->target_gfn + gfn - alias->base_gfn; - } - return gfn; -} - -gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) -{ - int i; - struct kvm_mem_alias *alias; - struct kvm_mem_aliases *aliases; - - aliases = kvm_aliases(kvm); - - for (i = 0; i < aliases->naliases; ++i) { - alias = &aliases->aliases[i]; - if (gfn >= alias->base_gfn - && gfn < alias->base_gfn + alias->npages) - return alias->target_gfn + gfn - alias->base_gfn; - } - return gfn; -} - -/* - * Set a new alias region. Aliases map a portion of physical memory into - * another portion. This is useful for memory windows, for example the PC - * VGA region. - */ -static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm, - struct kvm_memory_alias *alias) -{ - int r, n; - struct kvm_mem_alias *p; - struct kvm_mem_aliases *aliases, *old_aliases; - - r = -EINVAL; - /* General sanity checks */ - if (alias->memory_size & (PAGE_SIZE - 1)) - goto out; - if (alias->guest_phys_addr & (PAGE_SIZE - 1)) - goto out; - if (alias->slot >= KVM_ALIAS_SLOTS) - goto out; - if (alias->guest_phys_addr + alias->memory_size - < alias->guest_phys_addr) - goto out; - if (alias->target_phys_addr + alias->memory_size - < alias->target_phys_addr) - goto out; - - r = -ENOMEM; - aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL); - if (!aliases) - goto out; - - mutex_lock(&kvm->slots_lock); - - /* invalidate any gfn reference in case of deletion/shrinking */ - memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases)); - aliases->aliases[alias->slot].flags |= KVM_ALIAS_INVALID; - old_aliases = kvm->arch.aliases; - rcu_assign_pointer(kvm->arch.aliases, aliases); - synchronize_srcu_expedited(&kvm->srcu); - kvm_mmu_zap_all(kvm); - kfree(old_aliases); - - r = -ENOMEM; - aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL); - if (!aliases) - goto out_unlock; - - memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases)); - - p = &aliases->aliases[alias->slot]; - p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT; - p->npages = alias->memory_size >> PAGE_SHIFT; - p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT; - p->flags &= ~(KVM_ALIAS_INVALID); - - for (n = KVM_ALIAS_SLOTS; n > 0; --n) - if (aliases->aliases[n - 1].npages) - break; - aliases->naliases = n; - - old_aliases = kvm->arch.aliases; - rcu_assign_pointer(kvm->arch.aliases, aliases); - synchronize_srcu_expedited(&kvm->srcu); - kfree(old_aliases); - r = 0; - -out_unlock: - mutex_unlock(&kvm->slots_lock); -out: - return r; -} - static int kvm_vm_ioctl_get_irqchip(struct kvm *kvm, struct kvm_irqchip *chip) { int r; @@ -3056,7 +2947,6 @@ long kvm_arch_vm_ioctl(struct file *filp, union { struct kvm_pit_state ps; struct kvm_pit_state2 ps2; - struct kvm_memory_alias alias; struct kvm_pit_config pit_config; } u; @@ -3101,14 +2991,6 @@ long kvm_arch_vm_ioctl(struct file *filp, case KVM_GET_NR_MMU_PAGES: r = kvm_vm_ioctl_get_nr_mmu_pages(kvm); break; - case KVM_SET_MEMORY_ALIAS: - r = -EFAULT; - if (copy_from_user(&u.alias, argp, sizeof(struct kvm_memory_alias))) - goto out; - r = kvm_vm_ioctl_set_memory_alias(kvm, &u.alias); - if (r) - goto out; - break; case KVM_CREATE_IRQCHIP: { struct kvm_pic *vpic; @@ -5559,12 +5441,6 @@ struct kvm *kvm_arch_create_vm(void) if (!kvm) return ERR_PTR(-ENOMEM); - kvm->arch.aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL); - if (!kvm->arch.aliases) { - kfree(kvm); - return ERR_PTR(-ENOMEM); - } - INIT_LIST_HEAD(&kvm->arch.active_mmu_pages); INIT_LIST_HEAD(&kvm->arch.assigned_dev_head); @@ -5622,7 +5498,6 @@ void kvm_arch_destroy_vm(struct kvm *kvm) if (kvm->arch.ept_identity_pagetable) put_page(kvm->arch.ept_identity_pagetable); cleanup_srcu_struct(&kvm->srcu); - kfree(kvm->arch.aliases); kfree(kvm); } diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h index f4b54458285b..b7a404722d2b 100644 --- a/arch/x86/kvm/x86.h +++ b/arch/x86/kvm/x86.h @@ -65,13 +65,6 @@ static inline int is_paging(struct kvm_vcpu *vcpu) return kvm_read_cr0_bits(vcpu, X86_CR0_PG); } -static inline struct kvm_mem_aliases *kvm_aliases(struct kvm *kvm) -{ - return rcu_dereference_check(kvm->arch.aliases, - srcu_read_lock_held(&kvm->srcu) - || lockdep_is_held(&kvm->slots_lock)); -} - void kvm_before_handle_nmi(struct kvm_vcpu *vcpu); void kvm_after_handle_nmi(struct kvm_vcpu *vcpu); diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 6fd40f540a8e..636fc381c897 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -619,6 +619,7 @@ struct kvm_clock_data { */ #define KVM_CREATE_VCPU _IO(KVMIO, 0x41) #define KVM_GET_DIRTY_LOG _IOW(KVMIO, 0x42, struct kvm_dirty_log) +/* KVM_SET_MEMORY_ALIAS is obsolete: */ #define KVM_SET_MEMORY_ALIAS _IOW(KVMIO, 0x43, struct kvm_memory_alias) #define KVM_SET_NR_MMU_PAGES _IO(KVMIO, 0x44) #define KVM_GET_NR_MMU_PAGES _IO(KVMIO, 0x45) diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 2d96555cd4ed..240e460777bc 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -286,8 +286,6 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, int user_alloc); void kvm_disable_largepages(void); void kvm_arch_flush_shadow(struct kvm *kvm); -gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn); -gfn_t unalias_gfn_instantiation(struct kvm *kvm, gfn_t gfn); struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn); unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn); @@ -564,10 +562,6 @@ static inline int mmu_notifier_retry(struct kvm_vcpu *vcpu, unsigned long mmu_se } #endif -#ifndef KVM_ARCH_HAS_UNALIAS_INSTANTIATION -#define unalias_gfn_instantiation unalias_gfn -#endif - #ifdef CONFIG_HAVE_KVM_IRQCHIP #define KVM_MAX_IRQ_ROUTES 1024 diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 84a090644d9d..65417e3d8462 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -841,7 +841,7 @@ int kvm_is_error_hva(unsigned long addr) } EXPORT_SYMBOL_GPL(kvm_is_error_hva); -struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn) +struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) { int i; struct kvm_memslots *slots = kvm_memslots(kvm); @@ -855,20 +855,13 @@ struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn) } return NULL; } -EXPORT_SYMBOL_GPL(gfn_to_memslot_unaliased); - -struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) -{ - gfn = unalias_gfn(kvm, gfn); - return gfn_to_memslot_unaliased(kvm, gfn); -} +EXPORT_SYMBOL_GPL(gfn_to_memslot); int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) { int i; struct kvm_memslots *slots = kvm_memslots(kvm); - gfn = unalias_gfn_instantiation(kvm, gfn); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { struct kvm_memory_slot *memslot = &slots->memslots[i]; @@ -913,7 +906,6 @@ int memslot_id(struct kvm *kvm, gfn_t gfn) struct kvm_memslots *slots = kvm_memslots(kvm); struct kvm_memory_slot *memslot = NULL; - gfn = unalias_gfn(kvm, gfn); for (i = 0; i < slots->nmemslots; ++i) { memslot = &slots->memslots[i]; @@ -934,8 +926,7 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) { struct kvm_memory_slot *slot; - gfn = unalias_gfn_instantiation(kvm, gfn); - slot = gfn_to_memslot_unaliased(kvm, gfn); + slot = gfn_to_memslot(kvm, gfn); if (!slot || slot->flags & KVM_MEMSLOT_INVALID) return bad_hva(); return gfn_to_hva_memslot(slot, gfn); @@ -1202,8 +1193,7 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn) { struct kvm_memory_slot *memslot; - gfn = unalias_gfn(kvm, gfn); - memslot = gfn_to_memslot_unaliased(kvm, gfn); + memslot = gfn_to_memslot(kvm, gfn); if (memslot && memslot->dirty_bitmap) { unsigned long rel_gfn = gfn - memslot->base_gfn; -- cgit v1.2.3 From f5f48ee15c2ee3e44cf429e34b16c6fa9b900246 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Wed, 30 Jun 2010 12:25:15 +0800 Subject: KVM: VMX: Execute WBINVD to keep data consistency with assigned devices Some guest device driver may leverage the "Non-Snoop" I/O, and explicitly WBINVD or CLFLUSH to a RAM space. Since migration may occur before WBINVD or CLFLUSH, we need to maintain data consistency either by: 1: flushing cache (wbinvd) when the guest is scheduled out if there is no wbinvd exit, or 2: execute wbinvd on all dirty physical CPUs when guest wbinvd exits. Signed-off-by: Yaozu (Eddie) Dong Signed-off-by: Sheng Yang Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 6 ++++++ arch/x86/kvm/emulate.c | 5 ++++- arch/x86/kvm/svm.c | 7 +++++++ arch/x86/kvm/vmx.c | 10 +++++++++- arch/x86/kvm/x86.c | 41 +++++++++++++++++++++++++++++++++++++++++ 5 files changed, 67 insertions(+), 2 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index a57cdeacc4d2..2bda62485c4c 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -358,6 +359,8 @@ struct kvm_vcpu_arch { /* fields used by HYPER-V emulation */ u64 hv_vapic; + + cpumask_var_t wbinvd_dirty_mask; }; struct kvm_arch { @@ -514,6 +517,8 @@ struct kvm_x86_ops { void (*set_supported_cpuid)(u32 func, struct kvm_cpuid_entry2 *entry); + bool (*has_wbinvd_exit)(void); + const struct trace_print_flags *exit_reasons_str; }; @@ -571,6 +576,7 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu); int kvm_emulate_halt(struct kvm_vcpu *vcpu); int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address); int emulate_clts(struct kvm_vcpu *vcpu); +int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu); void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg); int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, int seg); diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index abb8cec420a2..e8bdddc4509e 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -3138,8 +3138,11 @@ twobyte_insn: emulate_clts(ctxt->vcpu); c->dst.type = OP_NONE; break; - case 0x08: /* invd */ case 0x09: /* wbinvd */ + kvm_emulate_wbinvd(ctxt->vcpu); + c->dst.type = OP_NONE; + break; + case 0x08: /* invd */ case 0x0d: /* GrpP (prefetch) */ case 0x18: /* Grp16 (prefetch/nop) */ c->dst.type = OP_NONE; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 587b99d37d44..56c9b6bd7655 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -3424,6 +3424,11 @@ static bool svm_rdtscp_supported(void) return false; } +static bool svm_has_wbinvd_exit(void) +{ + return true; +} + static void svm_fpu_deactivate(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -3508,6 +3513,8 @@ static struct kvm_x86_ops svm_x86_ops = { .rdtscp_supported = svm_rdtscp_supported, .set_supported_cpuid = svm_set_supported_cpuid, + + .has_wbinvd_exit = svm_has_wbinvd_exit, }; static int __init svm_init(void) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 661c6e199b4a..4dfb1dc09c88 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -412,6 +412,12 @@ static inline bool cpu_has_virtual_nmis(void) return vmcs_config.pin_based_exec_ctrl & PIN_BASED_VIRTUAL_NMIS; } +static inline bool cpu_has_vmx_wbinvd_exit(void) +{ + return vmcs_config.cpu_based_2nd_exec_ctrl & + SECONDARY_EXEC_WBINVD_EXITING; +} + static inline bool report_flexpriority(void) { return flexpriority_enabled; @@ -3397,7 +3403,7 @@ static int handle_invlpg(struct kvm_vcpu *vcpu) static int handle_wbinvd(struct kvm_vcpu *vcpu) { skip_emulated_instruction(vcpu); - /* TODO: Add support for VT-d/pass-through device */ + kvm_emulate_wbinvd(vcpu); return 1; } @@ -4347,6 +4353,8 @@ static struct kvm_x86_ops vmx_x86_ops = { .rdtscp_supported = vmx_rdtscp_supported, .set_supported_cpuid = vmx_set_supported_cpuid, + + .has_wbinvd_exit = cpu_has_vmx_wbinvd_exit, }; static int __init vmx_init(void) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 27322d341232..3d72fc067059 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1783,8 +1783,28 @@ out: return r; } +static void wbinvd_ipi(void *garbage) +{ + wbinvd(); +} + +static bool need_emulate_wbinvd(struct kvm_vcpu *vcpu) +{ + return vcpu->kvm->arch.iommu_domain && + !(vcpu->kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY); +} + void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { + /* Address WBINVD may be executed by guest */ + if (need_emulate_wbinvd(vcpu)) { + if (kvm_x86_ops->has_wbinvd_exit()) + cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask); + else if (vcpu->cpu != -1 && vcpu->cpu != cpu) + smp_call_function_single(vcpu->cpu, + wbinvd_ipi, NULL, 1); + } + kvm_x86_ops->vcpu_load(vcpu, cpu); if (unlikely(per_cpu(cpu_tsc_khz, cpu) == 0)) { unsigned long khz = cpufreq_quick_get(cpu); @@ -3660,6 +3680,21 @@ int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address) return X86EMUL_CONTINUE; } +int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) +{ + if (!need_emulate_wbinvd(vcpu)) + return X86EMUL_CONTINUE; + + if (kvm_x86_ops->has_wbinvd_exit()) { + smp_call_function_many(vcpu->arch.wbinvd_dirty_mask, + wbinvd_ipi, NULL, 1); + cpumask_clear(vcpu->arch.wbinvd_dirty_mask); + } + wbinvd(); + return X86EMUL_CONTINUE; +} +EXPORT_SYMBOL_GPL(kvm_emulate_wbinvd); + int emulate_clts(struct kvm_vcpu *vcpu) { kvm_x86_ops->set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~X86_CR0_TS)); @@ -5263,6 +5298,7 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) vcpu->arch.time_page = NULL; } + free_cpumask_var(vcpu->arch.wbinvd_dirty_mask); fx_free(vcpu); kvm_x86_ops->vcpu_free(vcpu); } @@ -5392,7 +5428,12 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) } vcpu->arch.mcg_cap = KVM_MAX_MCE_BANKS; + if (!zalloc_cpumask_var(&vcpu->arch.wbinvd_dirty_mask, GFP_KERNEL)) + goto fail_free_mce_banks; + return 0; +fail_free_mce_banks: + kfree(vcpu->arch.mce_banks); fail_free_lapic: kvm_free_lapic(vcpu); fail_mmu_destroy: -- cgit v1.2.3 From 828554136bbacae6e39fc31b9cd7e7c660ad7530 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 1 Jul 2010 16:00:11 +0200 Subject: KVM: Remove unnecessary divide operations This patch converts unnecessary divide and modulo operations in the KVM large page related code into logical operations. This allows to convert gfn_t to u64 while not breaking 32 bit builds. Signed-off-by: Joerg Roedel Signed-off-by: Marcelo Tosatti --- arch/ia64/include/asm/kvm_host.h | 1 + arch/powerpc/include/asm/kvm_host.h | 1 + arch/s390/include/asm/kvm_host.h | 3 ++- arch/x86/include/asm/kvm_host.h | 3 ++- arch/x86/kvm/mmu.c | 8 ++++---- virt/kvm/kvm_main.c | 10 +++++----- 6 files changed, 15 insertions(+), 11 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h index a362e67e0ca6..2f229e5de498 100644 --- a/arch/ia64/include/asm/kvm_host.h +++ b/arch/ia64/include/asm/kvm_host.h @@ -235,6 +235,7 @@ struct kvm_vm_data { #define KVM_REQ_PTC_G 32 #define KVM_REQ_RESUME 33 +#define KVM_HPAGE_GFN_SHIFT(x) 0 #define KVM_NR_PAGE_SIZES 1 #define KVM_PAGES_PER_HPAGE(x) 1 diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index e004eafcd3f0..b0b23c007d6e 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -35,6 +35,7 @@ #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 /* We don't currently support large pages. */ +#define KVM_HPAGE_GFN_SHIFT(x) 0 #define KVM_NR_PAGE_SIZES 1 #define KVM_PAGES_PER_HPAGE(x) (1UL<<31) diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index b95710a1f5dd..cef7dbf69dfc 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -41,7 +41,8 @@ struct sca_block { } __attribute__((packed)); #define KVM_NR_PAGE_SIZES 2 -#define KVM_HPAGE_SHIFT(x) (PAGE_SHIFT + ((x) - 1) * 8) +#define KVM_HPAGE_GFN_SHIFT(x) (((x) - 1) * 8) +#define KVM_HPAGE_SHIFT(x) (PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x)) #define KVM_HPAGE_SIZE(x) (1UL << KVM_HPAGE_SHIFT(x)) #define KVM_HPAGE_MASK(x) (~(KVM_HPAGE_SIZE(x) - 1)) #define KVM_PAGES_PER_HPAGE(x) (KVM_HPAGE_SIZE(x) / PAGE_SIZE) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 2bda62485c4c..50c79b9f5c38 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -44,7 +44,8 @@ /* KVM Hugepage definitions for x86 */ #define KVM_NR_PAGE_SIZES 3 -#define KVM_HPAGE_SHIFT(x) (PAGE_SHIFT + (((x) - 1) * 9)) +#define KVM_HPAGE_GFN_SHIFT(x) (((x) - 1) * 9) +#define KVM_HPAGE_SHIFT(x) (PAGE_SHIFT + KVM_HPAGE_GFN_SHIFT(x)) #define KVM_HPAGE_SIZE(x) (1UL << KVM_HPAGE_SHIFT(x)) #define KVM_HPAGE_MASK(x) (~(KVM_HPAGE_SIZE(x) - 1)) #define KVM_PAGES_PER_HPAGE(x) (KVM_HPAGE_SIZE(x) / PAGE_SIZE) diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index ca07ed083b59..a20fd613acfe 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -423,8 +423,8 @@ static int *slot_largepage_idx(gfn_t gfn, { unsigned long idx; - idx = (gfn / KVM_PAGES_PER_HPAGE(level)) - - (slot->base_gfn / KVM_PAGES_PER_HPAGE(level)); + idx = (gfn >> KVM_HPAGE_GFN_SHIFT(level)) - + (slot->base_gfn >> KVM_HPAGE_GFN_SHIFT(level)); return &slot->lpage_info[level - 2][idx].write_count; } @@ -528,8 +528,8 @@ static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int level) if (likely(level == PT_PAGE_TABLE_LEVEL)) return &slot->rmap[gfn - slot->base_gfn]; - idx = (gfn / KVM_PAGES_PER_HPAGE(level)) - - (slot->base_gfn / KVM_PAGES_PER_HPAGE(level)); + idx = (gfn >> KVM_HPAGE_GFN_SHIFT(level)) - + (slot->base_gfn >> KVM_HPAGE_GFN_SHIFT(level)); return &slot->lpage_info[level - 2][idx].rmap_pde; } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index ec2e3c6ac7ed..a60b6b053b6f 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -626,9 +626,9 @@ int __kvm_set_memory_region(struct kvm *kvm, if (new.lpage_info[i]) continue; - lpages = 1 + (base_gfn + npages - 1) / - KVM_PAGES_PER_HPAGE(level); - lpages -= base_gfn / KVM_PAGES_PER_HPAGE(level); + lpages = 1 + ((base_gfn + npages - 1) + >> KVM_HPAGE_GFN_SHIFT(level)); + lpages -= base_gfn >> KVM_HPAGE_GFN_SHIFT(level); new.lpage_info[i] = vmalloc(lpages * sizeof(*new.lpage_info[i])); @@ -638,9 +638,9 @@ int __kvm_set_memory_region(struct kvm *kvm, memset(new.lpage_info[i], 0, lpages * sizeof(*new.lpage_info[i])); - if (base_gfn % KVM_PAGES_PER_HPAGE(level)) + if (base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1)) new.lpage_info[i][0].write_count = 1; - if ((base_gfn+npages) % KVM_PAGES_PER_HPAGE(level)) + if ((base_gfn+npages) & (KVM_PAGES_PER_HPAGE(level) - 1)) new.lpage_info[i][lpages - 1].write_count = 1; ugfn = new.userspace_addr >> PAGE_SHIFT; /* -- cgit v1.2.3 From dd180b3e90253cb4ca95d603a8c17413f8daec69 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Sat, 3 Jul 2010 16:02:42 +0800 Subject: KVM: VMX: fix tlb flush with invalid root Commit 341d9b535b6c simplify reload logic while entry guest mode, it can avoid unnecessary sync-root if KVM_REQ_MMU_RELOAD and KVM_REQ_MMU_SYNC both set. But, it cause a issue that when we handle 'KVM_REQ_TLB_FLUSH', the root is invalid, it is triggered during my test: Kernel BUG at ffffffffa00212b8 [verbose debug info unavailable] ...... Fixed by directly return if the root is not ready. Signed-off-by: Xiao Guangrong Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 ++ arch/x86/kvm/mmu.c | 2 -- arch/x86/kvm/vmx.c | 5 ++++- 3 files changed, 6 insertions(+), 3 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 50c79b9f5c38..502e53f999cf 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -40,6 +40,8 @@ 0xFFFFFF0000000000ULL) #define INVALID_PAGE (~(hpa_t)0) +#define VALID_PAGE(x) ((x) != INVALID_PAGE) + #define UNMAPPED_GVA (~(gpa_t)0) /* KVM Hugepage definitions for x86 */ diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index a20fd613acfe..70cdf6876b5f 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -92,8 +92,6 @@ module_param(oos_shadow, bool, 0644); #define PT_FIRST_AVAIL_BITS_SHIFT 9 #define PT64_SECOND_AVAIL_BITS_SHIFT 52 -#define VALID_PAGE(x) ((x) != INVALID_PAGE) - #define PT64_LEVEL_BITS 9 #define PT64_LEVEL_SHIFT(level) \ diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 4dfb1dc09c88..2fdcc9819f36 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1828,8 +1828,11 @@ static void exit_lmode(struct kvm_vcpu *vcpu) static void vmx_flush_tlb(struct kvm_vcpu *vcpu) { vpid_sync_context(to_vmx(vcpu)); - if (enable_ept) + if (enable_ept) { + if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) + return; ept_sync_context(construct_eptp(vcpu->arch.mmu.root_hpa)); + } } static void vmx_decache_cr0_guest_bits(struct kvm_vcpu *vcpu) -- cgit v1.2.3 From 35f2915c3bd0cd6950bdd9d461de565e8feae852 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Tue, 1 Jun 2010 13:07:34 +0100 Subject: intel_scu_ipc: add definitions for vRTC related command Signed-off-by: Feng Tang Signed-off-by: Alan Cox Signed-off-by: Matthew Garrett --- arch/x86/include/asm/intel_scu_ipc.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/intel_scu_ipc.h b/arch/x86/include/asm/intel_scu_ipc.h index 4470c9ad4a3e..03200452069b 100644 --- a/arch/x86/include/asm/intel_scu_ipc.h +++ b/arch/x86/include/asm/intel_scu_ipc.h @@ -1,6 +1,12 @@ #ifndef _ASM_X86_INTEL_SCU_IPC_H_ #define _ASM_X86_INTEL_SCU_IPC_H_ +#define IPCMSG_VRTC 0xFA /* Set vRTC device */ + +/* Command id associated with message IPCMSG_VRTC */ +#define IPC_CMD_VRTC_SETTIME 1 /* Set time */ +#define IPC_CMD_VRTC_SETALARM 2 /* Set alarm */ + /* Read single register */ int intel_scu_ipc_ioread8(u16 addr, u8 *data); -- cgit v1.2.3 From 804f8681a99da2aa49bd7f0dab3750848d1ab1bc Mon Sep 17 00:00:00 2001 From: Sreedhara DS Date: Mon, 26 Jul 2010 10:03:10 +0100 Subject: Remove indirect read write api support. The firmware of production devices does not support this interface so this is dead code. Signed-off-by: Sreedhara DS Signed-off-by: Alan Cox Signed-off-by: Matthew Garrett --- arch/x86/include/asm/intel_scu_ipc.h | 14 ------ drivers/platform/x86/intel_scu_ipc.c | 82 ------------------------------------ 2 files changed, 96 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/intel_scu_ipc.h b/arch/x86/include/asm/intel_scu_ipc.h index 03200452069b..29f66793cc55 100644 --- a/arch/x86/include/asm/intel_scu_ipc.h +++ b/arch/x86/include/asm/intel_scu_ipc.h @@ -34,20 +34,6 @@ int intel_scu_ipc_writev(u16 *addr, u8 *data, int len); /* Update single register based on the mask */ int intel_scu_ipc_update_register(u16 addr, u8 data, u8 mask); -/* - * Indirect register read - * Can be used when SCCB(System Controller Configuration Block) register - * HRIM(Honor Restricted IPC Messages) is set (bit 23) - */ -int intel_scu_ipc_register_read(u32 addr, u32 *data); - -/* - * Indirect register write - * Can be used when SCCB(System Controller Configuration Block) register - * HRIM(Honor Restricted IPC Messages) is set (bit 23) - */ -int intel_scu_ipc_register_write(u32 addr, u32 data); - /* Issue commands to the SCU with or without data */ int intel_scu_ipc_simple_command(int cmd, int sub); int intel_scu_ipc_command(int cmd, int sub, u32 *in, int inlen, diff --git a/drivers/platform/x86/intel_scu_ipc.c b/drivers/platform/x86/intel_scu_ipc.c index a0dc41e27733..fd78386cd048 100644 --- a/drivers/platform/x86/intel_scu_ipc.c +++ b/drivers/platform/x86/intel_scu_ipc.c @@ -115,24 +115,6 @@ static inline void ipc_data_writel(u32 data, u32 offset) /* Write ipc data */ writel(data, ipcdev.ipc_base + 0x80 + offset); } -/* - * IPC destination Pointer (Write Only): - * Use content as pointer for destination write - */ -static inline void ipc_write_dptr(u32 data) /* Write dptr data */ -{ - writel(data, ipcdev.ipc_base + 0x0C); -} - -/* - * IPC Source Pointer (Write Only): - * Use content as pointer for read location -*/ -static inline void ipc_write_sptr(u32 data) /* Write dptr data */ -{ - writel(data, ipcdev.ipc_base + 0x08); -} - /* * Status Register (Read Only): * Driver will read this register to get the ready/busy status of the IPC @@ -413,70 +395,6 @@ int intel_scu_ipc_update_register(u16 addr, u8 bits, u8 mask) } EXPORT_SYMBOL(intel_scu_ipc_update_register); -/** - * intel_scu_ipc_register_read - 32bit indirect read - * @addr: register address - * @value: 32bit value return - * - * Performs IA 32 bit indirect read, returns 0 on success, or an - * error code. - * - * Can be used when SCCB(System Controller Configuration Block) register - * HRIM(Honor Restricted IPC Messages) is set (bit 23) - * - * This function may sleep. Locking for SCU accesses is handled for - * the caller. - */ -int intel_scu_ipc_register_read(u32 addr, u32 *value) -{ - u32 err = 0; - - mutex_lock(&ipclock); - if (ipcdev.pdev == NULL) { - mutex_unlock(&ipclock); - return -ENODEV; - } - ipc_write_sptr(addr); - ipc_command(4 << 16 | IPC_CMD_INDIRECT_RD); - err = busy_loop(); - *value = ipc_data_readl(0); - mutex_unlock(&ipclock); - return err; -} -EXPORT_SYMBOL(intel_scu_ipc_register_read); - -/** - * intel_scu_ipc_register_write - 32bit indirect write - * @addr: register address - * @value: 32bit value to write - * - * Performs IA 32 bit indirect write, returns 0 on success, or an - * error code. - * - * Can be used when SCCB(System Controller Configuration Block) register - * HRIM(Honor Restricted IPC Messages) is set (bit 23) - * - * This function may sleep. Locking for SCU accesses is handled for - * the caller. - */ -int intel_scu_ipc_register_write(u32 addr, u32 value) -{ - u32 err = 0; - - mutex_lock(&ipclock); - if (ipcdev.pdev == NULL) { - mutex_unlock(&ipclock); - return -ENODEV; - } - ipc_write_dptr(addr); - ipc_data_writel(value, 0); - ipc_command(4 << 16 | IPC_CMD_INDIRECT_WR); - err = busy_loop(); - mutex_unlock(&ipclock); - return err; -} -EXPORT_SYMBOL(intel_scu_ipc_register_write); - /** * intel_scu_ipc_simple_command - send a simple command * @cmd: command -- cgit v1.2.3 From 12bfa3de63504d879ae427ec1f2884fc46556157 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Thu, 5 Aug 2010 09:22:20 -0500 Subject: kgdb,x86: Individual register get/set for x86 Implement the ability to individually get and set registers for kdb and kgdb for x86. Signed-off-by: Jason Wessel Acked-by: H. Peter Anvin CC: Ingo Molnar CC: x86@kernel.org --- arch/x86/include/asm/kgdb.h | 20 +++--- arch/x86/kernel/kgdb.c | 168 ++++++++++++++++++++++---------------------- 2 files changed, 94 insertions(+), 94 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/kgdb.h b/arch/x86/include/asm/kgdb.h index 006da3687cdc..396f5b5fc4d7 100644 --- a/arch/x86/include/asm/kgdb.h +++ b/arch/x86/include/asm/kgdb.h @@ -39,9 +39,11 @@ enum regnames { GDB_FS, /* 14 */ GDB_GS, /* 15 */ }; +#define GDB_ORIG_AX 41 +#define DBG_MAX_REG_NUM 16 #define NUMREGBYTES ((GDB_GS+1)*4) #else /* ! CONFIG_X86_32 */ -enum regnames64 { +enum regnames { GDB_AX, /* 0 */ GDB_BX, /* 1 */ GDB_CX, /* 2 */ @@ -59,15 +61,15 @@ enum regnames64 { GDB_R14, /* 14 */ GDB_R15, /* 15 */ GDB_PC, /* 16 */ + GDB_PS, /* 17 */ + GDB_CS, /* 18 */ + GDB_SS, /* 19 */ }; - -enum regnames32 { - GDB_PS = 34, - GDB_CS, - GDB_SS, -}; -#define NUMREGBYTES ((GDB_SS+1)*4) -#endif /* CONFIG_X86_32 */ +#define GDB_ORIG_AX 57 +#define DBG_MAX_REG_NUM 20 +/* 17 64 bit regs and 3 32 bit regs */ +#define NUMREGBYTES ((17 * 8) + (3 * 4)) +#endif /* ! CONFIG_X86_32 */ static inline void arch_kgdb_breakpoint(void) { diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index 01ab17ae2ae7..bae89825e14e 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -49,55 +49,94 @@ #include #include -/** - * pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs - * @gdb_regs: A pointer to hold the registers in the order GDB wants. - * @regs: The &struct pt_regs of the current process. - * - * Convert the pt_regs in @regs into the format for registers that - * GDB expects, stored in @gdb_regs. - */ -void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) +struct dbg_reg_def_t dbg_reg_def[DBG_MAX_REG_NUM] = { -#ifndef CONFIG_X86_32 - u32 *gdb_regs32 = (u32 *)gdb_regs; +#ifdef CONFIG_X86_32 + { "ax", 4, offsetof(struct pt_regs, ax) }, + { "cx", 4, offsetof(struct pt_regs, cx) }, + { "dx", 4, offsetof(struct pt_regs, dx) }, + { "bx", 4, offsetof(struct pt_regs, bx) }, + { "sp", 4, offsetof(struct pt_regs, sp) }, + { "bp", 4, offsetof(struct pt_regs, bp) }, + { "si", 4, offsetof(struct pt_regs, si) }, + { "di", 4, offsetof(struct pt_regs, di) }, + { "ip", 4, offsetof(struct pt_regs, ip) }, + { "flags", 4, offsetof(struct pt_regs, flags) }, + { "cs", 4, offsetof(struct pt_regs, cs) }, + { "ss", 4, offsetof(struct pt_regs, ss) }, + { "ds", 4, offsetof(struct pt_regs, ds) }, + { "es", 4, offsetof(struct pt_regs, es) }, + { "fs", 4, -1 }, + { "gs", 4, -1 }, +#else + { "ax", 8, offsetof(struct pt_regs, ax) }, + { "bx", 8, offsetof(struct pt_regs, bx) }, + { "cx", 8, offsetof(struct pt_regs, cx) }, + { "dx", 8, offsetof(struct pt_regs, dx) }, + { "si", 8, offsetof(struct pt_regs, dx) }, + { "di", 8, offsetof(struct pt_regs, di) }, + { "bp", 8, offsetof(struct pt_regs, bp) }, + { "sp", 8, offsetof(struct pt_regs, sp) }, + { "r8", 8, offsetof(struct pt_regs, r8) }, + { "r9", 8, offsetof(struct pt_regs, r9) }, + { "r10", 8, offsetof(struct pt_regs, r10) }, + { "r11", 8, offsetof(struct pt_regs, r11) }, + { "r12", 8, offsetof(struct pt_regs, r12) }, + { "r13", 8, offsetof(struct pt_regs, r13) }, + { "r14", 8, offsetof(struct pt_regs, r14) }, + { "r15", 8, offsetof(struct pt_regs, r15) }, + { "ip", 8, offsetof(struct pt_regs, ip) }, + { "flags", 4, offsetof(struct pt_regs, flags) }, + { "cs", 4, offsetof(struct pt_regs, cs) }, + { "ss", 4, offsetof(struct pt_regs, ss) }, #endif - gdb_regs[GDB_AX] = regs->ax; - gdb_regs[GDB_BX] = regs->bx; - gdb_regs[GDB_CX] = regs->cx; - gdb_regs[GDB_DX] = regs->dx; - gdb_regs[GDB_SI] = regs->si; - gdb_regs[GDB_DI] = regs->di; - gdb_regs[GDB_BP] = regs->bp; - gdb_regs[GDB_PC] = regs->ip; +}; + +int dbg_set_reg(int regno, void *mem, struct pt_regs *regs) +{ + if ( #ifdef CONFIG_X86_32 - gdb_regs[GDB_PS] = regs->flags; - gdb_regs[GDB_DS] = regs->ds; - gdb_regs[GDB_ES] = regs->es; - gdb_regs[GDB_CS] = regs->cs; - gdb_regs[GDB_FS] = 0xFFFF; - gdb_regs[GDB_GS] = 0xFFFF; - if (user_mode_vm(regs)) { - gdb_regs[GDB_SS] = regs->ss; - gdb_regs[GDB_SP] = regs->sp; - } else { - gdb_regs[GDB_SS] = __KERNEL_DS; - gdb_regs[GDB_SP] = kernel_stack_pointer(regs); + regno == GDB_SS || regno == GDB_FS || regno == GDB_GS || +#endif + regno == GDB_SP || regno == GDB_ORIG_AX) + return 0; + + if (dbg_reg_def[regno].offset != -1) + memcpy((void *)regs + dbg_reg_def[regno].offset, mem, + dbg_reg_def[regno].size); + return 0; +} + +char *dbg_get_reg(int regno, void *mem, struct pt_regs *regs) +{ + if (regno == GDB_ORIG_AX) { + memcpy(mem, ®s->orig_ax, sizeof(regs->orig_ax)); + return "orig_ax"; } -#else - gdb_regs[GDB_R8] = regs->r8; - gdb_regs[GDB_R9] = regs->r9; - gdb_regs[GDB_R10] = regs->r10; - gdb_regs[GDB_R11] = regs->r11; - gdb_regs[GDB_R12] = regs->r12; - gdb_regs[GDB_R13] = regs->r13; - gdb_regs[GDB_R14] = regs->r14; - gdb_regs[GDB_R15] = regs->r15; - gdb_regs32[GDB_PS] = regs->flags; - gdb_regs32[GDB_CS] = regs->cs; - gdb_regs32[GDB_SS] = regs->ss; - gdb_regs[GDB_SP] = kernel_stack_pointer(regs); + if (regno >= DBG_MAX_REG_NUM || regno < 0) + return NULL; + + if (dbg_reg_def[regno].offset != -1) + memcpy(mem, (void *)regs + dbg_reg_def[regno].offset, + dbg_reg_def[regno].size); + + switch (regno) { +#ifdef CONFIG_X86_32 + case GDB_SS: + if (!user_mode_vm(regs)) + *(unsigned long *)mem = __KERNEL_DS; + break; + case GDB_SP: + if (!user_mode_vm(regs)) + *(unsigned long *)mem = kernel_stack_pointer(regs); + break; + case GDB_GS: + case GDB_FS: + *(unsigned long *)mem = 0xFFFF; + break; #endif + } + return dbg_reg_def[regno].name; } /** @@ -150,47 +189,6 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) gdb_regs[GDB_SP] = p->thread.sp; } -/** - * gdb_regs_to_pt_regs - Convert GDB regs to ptrace regs. - * @gdb_regs: A pointer to hold the registers we've received from GDB. - * @regs: A pointer to a &struct pt_regs to hold these values in. - * - * Convert the GDB regs in @gdb_regs into the pt_regs, and store them - * in @regs. - */ -void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs) -{ -#ifndef CONFIG_X86_32 - u32 *gdb_regs32 = (u32 *)gdb_regs; -#endif - regs->ax = gdb_regs[GDB_AX]; - regs->bx = gdb_regs[GDB_BX]; - regs->cx = gdb_regs[GDB_CX]; - regs->dx = gdb_regs[GDB_DX]; - regs->si = gdb_regs[GDB_SI]; - regs->di = gdb_regs[GDB_DI]; - regs->bp = gdb_regs[GDB_BP]; - regs->ip = gdb_regs[GDB_PC]; -#ifdef CONFIG_X86_32 - regs->flags = gdb_regs[GDB_PS]; - regs->ds = gdb_regs[GDB_DS]; - regs->es = gdb_regs[GDB_ES]; - regs->cs = gdb_regs[GDB_CS]; -#else - regs->r8 = gdb_regs[GDB_R8]; - regs->r9 = gdb_regs[GDB_R9]; - regs->r10 = gdb_regs[GDB_R10]; - regs->r11 = gdb_regs[GDB_R11]; - regs->r12 = gdb_regs[GDB_R12]; - regs->r13 = gdb_regs[GDB_R13]; - regs->r14 = gdb_regs[GDB_R14]; - regs->r15 = gdb_regs[GDB_R15]; - regs->flags = gdb_regs32[GDB_PS]; - regs->cs = gdb_regs32[GDB_CS]; - regs->ss = gdb_regs32[GDB_SS]; -#endif -} - static struct hw_breakpoint { unsigned enabled; unsigned long addr; -- cgit v1.2.3