@@ -63,6 +63,17 @@ config KVM_INTEL
To compile this as a module, choose M here: the module
will be called kvm-intel.
+config VMCSINFO_INTEL
+ tristate "Export VMCSINFO for Intel processors"
+ depends on KVM_INTEL
+ ---help---
+ Provides support for exporting VMCSINFO on Intel processors equipped
+ with the VT extensions. The VMCSINFO contains a VMCS revision
+ identifier and offsets of VMCS fields.
+
+ To compile this as a module, choose M here: the module
+ will be called vmcsinfo-intel.
+
config KVM_AMD
tristate "KVM for AMD processors support"
depends on KVM
@@ -4,6 +4,7 @@ ccflags-y += -Ivirt/kvm -Iarch/x86/kvm
CFLAGS_x86.o := -I.
CFLAGS_svm.o := -I.
CFLAGS_vmx.o := -I.
+CFLAGS_vmcsinfo.o := -I.
kvm-y += $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \
coalesced_mmio.o irq_comm.o eventfd.o \
@@ -15,7 +16,9 @@ kvm-y += x86.o mmu.o emulate.o i8259.o irq.o lapic.o \
i8254.o timer.o cpuid.o pmu.o
kvm-intel-y += vmx.o
kvm-amd-y += svm.o
+vmcsinfo-intel-y += vmcsinfo.o
obj-$(CONFIG_KVM) += kvm.o
obj-$(CONFIG_KVM_INTEL) += kvm-intel.o
obj-$(CONFIG_KVM_AMD) += kvm-amd.o
+obj-$(CONFIG_VMCSINFO_INTEL) += vmcsinfo-intel.o
new file mode 100644
@@ -0,0 +1,714 @@
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * This module enables machines with Intel VT-x extensions to export
+ * offsets of VMCS fields for guest debugging.
+ *
+ * Copyright (C) 2012 Fujitsu, Inc.
+ *
+ * Authors:
+ * Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2. See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/device.h>
+#include <linux/swab.h>
+#include <linux/cpu.h>
+
+#include <asm/vmx.h>
+
+MODULE_AUTHOR("Fujitsu");
+MODULE_LICENSE("GPL");
+
+static const struct x86_cpu_id vmcsinfo_cpu_id[] = {
+ X86_FEATURE_MATCH(X86_FEATURE_VMX),
+ {}
+};
+MODULE_DEVICE_TABLE(x86cpu, vmcsinfo_cpu_id);
+
+/*
+ * vmcs field offsets.
+ */
+static struct vmcsinfo {
+ u32 vmcs_revision_id;
+ u16 field_offset[][2];
+} vmcsinfo = {
+ 0,
+ {
+ {VIRTUAL_PROCESSOR_ID, 0},
+ {GUEST_ES_SELECTOR, 0},
+ {GUEST_CS_SELECTOR, 0},
+ {GUEST_SS_SELECTOR, 0},
+ {GUEST_DS_SELECTOR, 0},
+ {GUEST_FS_SELECTOR, 0},
+ {GUEST_GS_SELECTOR, 0},
+ {GUEST_LDTR_SELECTOR, 0},
+ {GUEST_TR_SELECTOR, 0},
+ {HOST_ES_SELECTOR, 0},
+ {HOST_CS_SELECTOR, 0},
+ {HOST_SS_SELECTOR, 0},
+ {HOST_DS_SELECTOR, 0},
+ {HOST_FS_SELECTOR, 0},
+ {HOST_GS_SELECTOR, 0},
+ {HOST_TR_SELECTOR, 0},
+ {IO_BITMAP_A, 0},
+ {IO_BITMAP_A_HIGH, 0},
+ {IO_BITMAP_B, 0},
+ {IO_BITMAP_B_HIGH, 0},
+ {MSR_BITMAP, 0},
+ {MSR_BITMAP_HIGH, 0},
+ {VM_EXIT_MSR_STORE_ADDR, 0},
+ {VM_EXIT_MSR_STORE_ADDR_HIGH, 0},
+ {VM_EXIT_MSR_LOAD_ADDR, 0},
+ {VM_EXIT_MSR_LOAD_ADDR_HIGH, 0},
+ {VM_ENTRY_MSR_LOAD_ADDR, 0},
+ {VM_ENTRY_MSR_LOAD_ADDR_HIGH, 0},
+ {TSC_OFFSET, 0},
+ {TSC_OFFSET_HIGH, 0},
+ {VIRTUAL_APIC_PAGE_ADDR, 0},
+ {VIRTUAL_APIC_PAGE_ADDR_HIGH, 0},
+ {APIC_ACCESS_ADDR, 0},
+ {APIC_ACCESS_ADDR_HIGH, 0},
+ {EPT_POINTER, 0},
+ {EPT_POINTER_HIGH, 0},
+ {GUEST_PHYSICAL_ADDRESS, 0},
+ {GUEST_PHYSICAL_ADDRESS_HIGH, 0},
+ {VMCS_LINK_POINTER, 0},
+ {VMCS_LINK_POINTER_HIGH, 0},
+ {GUEST_IA32_DEBUGCTL, 0},
+ {GUEST_IA32_DEBUGCTL_HIGH, 0},
+ {GUEST_IA32_PAT, 0},
+ {GUEST_IA32_PAT_HIGH, 0},
+ {GUEST_IA32_EFER, 0},
+ {GUEST_IA32_EFER_HIGH, 0},
+ {GUEST_IA32_PERF_GLOBAL_CTRL, 0},
+ {GUEST_IA32_PERF_GLOBAL_CTRL_HIGH, 0},
+ {GUEST_PDPTR0, 0},
+ {GUEST_PDPTR0_HIGH, 0},
+ {GUEST_PDPTR1, 0},
+ {GUEST_PDPTR1_HIGH, 0},
+ {GUEST_PDPTR2, 0},
+ {GUEST_PDPTR2_HIGH, 0},
+ {GUEST_PDPTR3, 0},
+ {GUEST_PDPTR3_HIGH, 0},
+ {HOST_IA32_PAT, 0},
+ {HOST_IA32_PAT_HIGH, 0},
+ {HOST_IA32_EFER, 0},
+ {HOST_IA32_EFER_HIGH, 0},
+ {HOST_IA32_PERF_GLOBAL_CTRL, 0},
+ {HOST_IA32_PERF_GLOBAL_CTRL_HIGH, 0},
+ {PIN_BASED_VM_EXEC_CONTROL, 0},
+ {CPU_BASED_VM_EXEC_CONTROL, 0},
+ {EXCEPTION_BITMAP, 0},
+ {PAGE_FAULT_ERROR_CODE_MASK, 0},
+ {PAGE_FAULT_ERROR_CODE_MATCH, 0},
+ {CR3_TARGET_COUNT, 0},
+ {VM_EXIT_CONTROLS, 0},
+ {VM_EXIT_MSR_STORE_COUNT, 0},
+ {VM_EXIT_MSR_LOAD_COUNT, 0},
+ {VM_ENTRY_CONTROLS, 0},
+ {VM_ENTRY_MSR_LOAD_COUNT, 0},
+ {VM_ENTRY_INTR_INFO_FIELD, 0},
+ {VM_ENTRY_EXCEPTION_ERROR_CODE, 0},
+ {VM_ENTRY_INSTRUCTION_LEN, 0},
+ {TPR_THRESHOLD, 0},
+ {SECONDARY_VM_EXEC_CONTROL, 0},
+ {PLE_GAP, 0},
+ {PLE_WINDOW, 0},
+ {VM_INSTRUCTION_ERROR, 0},
+ {VM_EXIT_REASON, 0},
+ {VM_EXIT_INTR_INFO, 0},
+ {VM_EXIT_INTR_ERROR_CODE, 0},
+ {IDT_VECTORING_INFO_FIELD, 0},
+ {IDT_VECTORING_ERROR_CODE, 0},
+ {VM_EXIT_INSTRUCTION_LEN, 0},
+ {VMX_INSTRUCTION_INFO, 0},
+ {GUEST_ES_LIMIT, 0},
+ {GUEST_CS_LIMIT, 0},
+ {GUEST_SS_LIMIT, 0},
+ {GUEST_DS_LIMIT, 0},
+ {GUEST_FS_LIMIT, 0},
+ {GUEST_GS_LIMIT, 0},
+ {GUEST_LDTR_LIMIT, 0},
+ {GUEST_TR_LIMIT, 0},
+ {GUEST_GDTR_LIMIT, 0},
+ {GUEST_IDTR_LIMIT, 0},
+ {GUEST_ES_AR_BYTES, 0},
+ {GUEST_CS_AR_BYTES, 0},
+ {GUEST_SS_AR_BYTES, 0},
+ {GUEST_DS_AR_BYTES, 0},
+ {GUEST_FS_AR_BYTES, 0},
+ {GUEST_GS_AR_BYTES, 0},
+ {GUEST_LDTR_AR_BYTES, 0},
+ {GUEST_TR_AR_BYTES, 0},
+ {GUEST_INTERRUPTIBILITY_INFO, 0},
+ {GUEST_ACTIVITY_STATE, 0},
+ {GUEST_SYSENTER_CS, 0},
+ {HOST_IA32_SYSENTER_CS, 0},
+ {CR0_GUEST_HOST_MASK, 0},
+ {CR4_GUEST_HOST_MASK, 0},
+ {CR0_READ_SHADOW, 0},
+ {CR4_READ_SHADOW, 0},
+ {CR3_TARGET_VALUE0, 0},
+ {CR3_TARGET_VALUE1, 0},
+ {CR3_TARGET_VALUE2, 0},
+ {CR3_TARGET_VALUE3, 0},
+ {EXIT_QUALIFICATION, 0},
+ {GUEST_LINEAR_ADDRESS, 0},
+ {GUEST_CR0, 0},
+ {GUEST_CR3, 0},
+ {GUEST_CR4, 0},
+ {GUEST_ES_BASE, 0},
+ {GUEST_CS_BASE, 0},
+ {GUEST_SS_BASE, 0},
+ {GUEST_DS_BASE, 0},
+ {GUEST_FS_BASE, 0},
+ {GUEST_GS_BASE, 0},
+ {GUEST_LDTR_BASE, 0},
+ {GUEST_TR_BASE, 0},
+ {GUEST_GDTR_BASE, 0},
+ {GUEST_IDTR_BASE, 0},
+ {GUEST_DR7, 0},
+ {GUEST_RSP, 0},
+ {GUEST_RIP, 0},
+ {GUEST_RFLAGS, 0},
+ {GUEST_PENDING_DBG_EXCEPTIONS, 0},
+ {GUEST_SYSENTER_ESP, 0},
+ {GUEST_SYSENTER_EIP, 0},
+ {HOST_CR0, 0},
+ {HOST_CR3, 0},
+ {HOST_CR4, 0},
+ {HOST_FS_BASE, 0},
+ {HOST_GS_BASE, 0},
+ {HOST_TR_BASE, 0},
+ {HOST_GDTR_BASE, 0},
+ {HOST_IDTR_BASE, 0},
+ {HOST_IA32_SYSENTER_ESP, 0},
+ {HOST_IA32_SYSENTER_EIP, 0},
+ {HOST_RSP, 0},
+ {HOST_RIP, 0}
+ }
+};
+
+const char vmcs_group_name[] = "vmcs";
+
+static ssize_t vmcs_id_show(struct device *dev,
+ struct device_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "%d\n", vmcsinfo.vmcs_revision_id);
+}
+static DEVICE_ATTR(id, 0444, vmcs_id_show, NULL);
+
+#define BUILD_OFFSET_SHOW(i, field_code) \
+static ssize_t _##field_code##_show(struct device *dev, \
+ struct device_attribute *attr, \
+ char *buf) \
+{ \
+ return sprintf(buf, "%d\n", vmcsinfo.field_offset[i][1]); \
+} \
+static DEVICE_ATTR(field_code, 0444, _##field_code##_show, NULL);
+
+BUILD_OFFSET_SHOW( 0, 0000); /* VIRTUAL_PROCESSOR_ID */
+BUILD_OFFSET_SHOW( 1, 0800); /* GUEST_ES_SELECTOR */
+BUILD_OFFSET_SHOW( 2, 0802); /* GUEST_CS_SELECTOR */
+BUILD_OFFSET_SHOW( 3, 0804); /* GUEST_SS_SELECTOR */
+BUILD_OFFSET_SHOW( 4, 0806); /* GUEST_DS_SELECTOR */
+BUILD_OFFSET_SHOW( 5, 0808); /* GUEST_FS_SELECTOR */
+BUILD_OFFSET_SHOW( 6, 080a); /* GUEST_GS_SELECTOR */
+BUILD_OFFSET_SHOW( 7, 080c); /* GUEST_LDTR_SELECTOR */
+BUILD_OFFSET_SHOW( 8, 080e); /* GUEST_TR_SELECTOR */
+BUILD_OFFSET_SHOW( 9, 0c00); /* HOST_ES_SELECTOR */
+BUILD_OFFSET_SHOW( 10, 0c02); /* HOST_CS_SELECTOR */
+BUILD_OFFSET_SHOW( 11, 0c04); /* HOST_SS_SELECTOR */
+BUILD_OFFSET_SHOW( 12, 0c06); /* HOST_DS_SELECTOR */
+BUILD_OFFSET_SHOW( 13, 0c08); /* HOST_FS_SELECTOR */
+BUILD_OFFSET_SHOW( 14, 0c0a); /* HOST_GS_SELECTOR */
+BUILD_OFFSET_SHOW( 15, 0c0c); /* HOST_TR_SELECTOR */
+BUILD_OFFSET_SHOW( 16, 2000); /* IO_BITMAP_A */
+BUILD_OFFSET_SHOW( 17, 2001); /* IO_BITMAP_A_HIGH */
+BUILD_OFFSET_SHOW( 18, 2002); /* IO_BITMAP_B */
+BUILD_OFFSET_SHOW( 19, 2003); /* IO_BITMAP_B_HIGH */
+BUILD_OFFSET_SHOW( 20, 2004); /* MSR_BITMAP */
+BUILD_OFFSET_SHOW( 21, 2005); /* MSR_BITMAP_HIGH */
+BUILD_OFFSET_SHOW( 22, 2006); /* VM_EXIT_MSR_STORE_ADDR */
+BUILD_OFFSET_SHOW( 23, 2007); /* VM_EXIT_MSR_STORE_ADDR_HIGH */
+BUILD_OFFSET_SHOW( 24, 2008); /* VM_EXIT_MSR_LOAD_ADDR */
+BUILD_OFFSET_SHOW( 25, 2009); /* VM_EXIT_MSR_LOAD_ADDR_HIGH */
+BUILD_OFFSET_SHOW( 26, 200a); /* VM_ENTRY_MSR_LOAD_ADDR */
+BUILD_OFFSET_SHOW( 27, 200b); /* VM_ENTRY_MSR_LOAD_ADDR_HIGH */
+BUILD_OFFSET_SHOW( 28, 2010); /* TSC_OFFSET */
+BUILD_OFFSET_SHOW( 29, 2011); /* TSC_OFFSET_HIGH */
+BUILD_OFFSET_SHOW( 30, 2012); /* VIRTUAL_APIC_PAGE_ADDR */
+BUILD_OFFSET_SHOW( 31, 2013); /* VIRTUAL_APIC_PAGE_ADDR_HIGH */
+BUILD_OFFSET_SHOW( 32, 2014); /* APIC_ACCESS_ADDR */
+BUILD_OFFSET_SHOW( 33, 2015); /* APIC_ACCESS_ADDR_HIGH */
+BUILD_OFFSET_SHOW( 34, 201a); /* EPT_POINTER */
+BUILD_OFFSET_SHOW( 35, 201b); /* EPT_POINTER_HIGH */
+BUILD_OFFSET_SHOW( 36, 2400); /* GUEST_PHYSICAL_ADDRESS */
+BUILD_OFFSET_SHOW( 37, 2401); /* GUEST_PHYSICAL_ADDRESS_HIGH */
+BUILD_OFFSET_SHOW( 38, 2800); /* VMCS_LINK_POINTER */
+BUILD_OFFSET_SHOW( 39, 2801); /* VMCS_LINK_POINTER_HIGH */
+BUILD_OFFSET_SHOW( 40, 2802); /* GUEST_IA32_DEBUGCTL */
+BUILD_OFFSET_SHOW( 41, 2803); /* GUEST_IA32_DEBUGCTL_HIGH */
+BUILD_OFFSET_SHOW( 42, 2804); /* GUEST_IA32_PAT */
+BUILD_OFFSET_SHOW( 43, 2805); /* GUEST_IA32_PAT_HIGH */
+BUILD_OFFSET_SHOW( 44, 2806); /* GUEST_IA32_EFER */
+BUILD_OFFSET_SHOW( 45, 2807); /* GUEST_IA32_EFER_HIGH */
+BUILD_OFFSET_SHOW( 46, 2808); /* GUEST_IA32_PERF_GLOBAL_CTRL */
+BUILD_OFFSET_SHOW( 47, 2809); /* GUEST_IA32_PERF_GLOBAL_CTRL_HIGH */
+BUILD_OFFSET_SHOW( 48, 280a); /* GUEST_PDPTR0 */
+BUILD_OFFSET_SHOW( 49, 280b); /* GUEST_PDPTR0_HIGH */
+BUILD_OFFSET_SHOW( 50, 280c); /* GUEST_PDPTR1 */
+BUILD_OFFSET_SHOW( 51, 280d); /* GUEST_PDPTR1_HIGH */
+BUILD_OFFSET_SHOW( 52, 280e); /* GUEST_PDPTR2 */
+BUILD_OFFSET_SHOW( 53, 280f); /* GUEST_PDPTR2_HIGH */
+BUILD_OFFSET_SHOW( 54, 2810); /* GUEST_PDPTR3 */
+BUILD_OFFSET_SHOW( 55, 2811); /* GUEST_PDPTR3_HIGH */
+BUILD_OFFSET_SHOW( 56, 2c00); /* HOST_IA32_PAT */
+BUILD_OFFSET_SHOW( 57, 2c01); /* HOST_IA32_PAT_HIGH */
+BUILD_OFFSET_SHOW( 58, 2c02); /* HOST_IA32_EFER */
+BUILD_OFFSET_SHOW( 59, 2c03); /* HOST_IA32_EFER_HIGH */
+BUILD_OFFSET_SHOW( 60, 2c04); /* HOST_IA32_PERF_GLOBAL_CTRL */
+BUILD_OFFSET_SHOW( 61, 2c05); /* HOST_IA32_PERF_GLOBAL_CTRL_HIGH */
+BUILD_OFFSET_SHOW( 62, 4000); /* PIN_BASED_VM_EXEC_CONTROL */
+BUILD_OFFSET_SHOW( 63, 4002); /* CPU_BASED_VM_EXEC_CONTROL */
+BUILD_OFFSET_SHOW( 64, 4004); /* EXCEPTION_BITMAP */
+BUILD_OFFSET_SHOW( 65, 4006); /* PAGE_FAULT_ERROR_CODE_MASK */
+BUILD_OFFSET_SHOW( 66, 4008); /* PAGE_FAULT_ERROR_CODE_MATCH */
+BUILD_OFFSET_SHOW( 67, 400a); /* CR3_TARGET_COUNT */
+BUILD_OFFSET_SHOW( 68, 400c); /* VM_EXIT_CONTROLS */
+BUILD_OFFSET_SHOW( 69, 400e); /* VM_EXIT_MSR_STORE_COUNT */
+BUILD_OFFSET_SHOW( 70, 4010); /* VM_EXIT_MSR_LOAD_COUNT */
+BUILD_OFFSET_SHOW( 71, 4012); /* VM_ENTRY_CONTROLS */
+BUILD_OFFSET_SHOW( 72, 4014); /* VM_ENTRY_MSR_LOAD_COUNT */
+BUILD_OFFSET_SHOW( 73, 4016); /* VM_ENTRY_INTR_INFO_FIELD */
+BUILD_OFFSET_SHOW( 74, 4018); /* VM_ENTRY_EXCEPTION_ERROR_CODE */
+BUILD_OFFSET_SHOW( 75, 401a); /* VM_ENTRY_INSTRUCTION_LEN */
+BUILD_OFFSET_SHOW( 76, 401c); /* TPR_THRESHOLD */
+BUILD_OFFSET_SHOW( 77, 401e); /* SECONDARY_VM_EXEC_CONTROL */
+BUILD_OFFSET_SHOW( 78, 4020); /* PLE_GAP */
+BUILD_OFFSET_SHOW( 79, 4022); /* PLE_WINDOW */
+BUILD_OFFSET_SHOW( 80, 4400); /* VM_INSTRUCTION_ERROR */
+BUILD_OFFSET_SHOW( 81, 4402); /* VM_EXIT_REASON */
+BUILD_OFFSET_SHOW( 82, 4404); /* VM_EXIT_INTR_INFO */
+BUILD_OFFSET_SHOW( 83, 4406); /* VM_EXIT_INTR_ERROR_CODE */
+BUILD_OFFSET_SHOW( 84, 4408); /* IDT_VECTORING_INFO_FIELD */
+BUILD_OFFSET_SHOW( 85, 440a); /* IDT_VECTORING_ERROR_CODE */
+BUILD_OFFSET_SHOW( 86, 440c); /* VM_EXIT_INSTRUCTION_LEN */
+BUILD_OFFSET_SHOW( 87, 440e); /* VMX_INSTRUCTION_INFO */
+BUILD_OFFSET_SHOW( 88, 4800); /* GUEST_ES_LIMIT */
+BUILD_OFFSET_SHOW( 89, 4802); /* GUEST_CS_LIMIT */
+BUILD_OFFSET_SHOW( 90, 4804); /* GUEST_SS_LIMIT */
+BUILD_OFFSET_SHOW( 91, 4806); /* GUEST_DS_LIMIT */
+BUILD_OFFSET_SHOW( 92, 4808); /* GUEST_FS_LIMIT */
+BUILD_OFFSET_SHOW( 93, 480a); /* GUEST_GS_LIMIT */
+BUILD_OFFSET_SHOW( 94, 480c); /* GUEST_LDTR_LIMIT */
+BUILD_OFFSET_SHOW( 95, 480e); /* GUEST_TR_LIMIT */
+BUILD_OFFSET_SHOW( 96, 4810); /* GUEST_GDTR_LIMIT */
+BUILD_OFFSET_SHOW( 97, 4812); /* GUEST_IDTR_LIMIT */
+BUILD_OFFSET_SHOW( 98, 4814); /* GUEST_ES_AR_BYTES */
+BUILD_OFFSET_SHOW( 99, 4816); /* GUEST_CS_AR_BYTES */
+BUILD_OFFSET_SHOW(100, 4818); /* GUEST_SS_AR_BYTES */
+BUILD_OFFSET_SHOW(101, 481a); /* GUEST_DS_AR_BYTES */
+BUILD_OFFSET_SHOW(102, 481c); /* GUEST_FS_AR_BYTES */
+BUILD_OFFSET_SHOW(103, 481e); /* GUEST_GS_AR_BYTES */
+BUILD_OFFSET_SHOW(104, 4820); /* GUEST_LDTR_AR_BYTES */
+BUILD_OFFSET_SHOW(105, 4822); /* GUEST_TR_AR_BYTES */
+BUILD_OFFSET_SHOW(106, 4824); /* GUEST_INTERRUPTIBILITY_INFO */
+BUILD_OFFSET_SHOW(107, 4826); /* GUEST_ACTIVITY_STATE */
+BUILD_OFFSET_SHOW(108, 482A); /* GUEST_SYSENTER_CS */
+BUILD_OFFSET_SHOW(109, 4c00); /* HOST_IA32_SYSENTER_CS */
+BUILD_OFFSET_SHOW(110, 6000); /* CR0_GUEST_HOST_MASK */
+BUILD_OFFSET_SHOW(111, 6002); /* CR4_GUEST_HOST_MASK */
+BUILD_OFFSET_SHOW(112, 6004); /* CR0_READ_SHADOW */
+BUILD_OFFSET_SHOW(113, 6006); /* CR4_READ_SHADOW */
+BUILD_OFFSET_SHOW(114, 6008); /* CR3_TARGET_VALUE0 */
+BUILD_OFFSET_SHOW(115, 600a); /* CR3_TARGET_VALUE1 */
+BUILD_OFFSET_SHOW(116, 600c); /* CR3_TARGET_VALUE2 */
+BUILD_OFFSET_SHOW(117, 600e); /* CR3_TARGET_VALUE3 */
+BUILD_OFFSET_SHOW(118, 6400); /* EXIT_QUALIFICATION */
+BUILD_OFFSET_SHOW(119, 640a); /* GUEST_LINEAR_ADDRESS */
+BUILD_OFFSET_SHOW(120, 6800); /* GUEST_CR0 */
+BUILD_OFFSET_SHOW(121, 6802); /* GUEST_CR3 */
+BUILD_OFFSET_SHOW(122, 6804); /* GUEST_CR4 */
+BUILD_OFFSET_SHOW(123, 6806); /* GUEST_ES_BASE */
+BUILD_OFFSET_SHOW(124, 6808); /* GUEST_CS_BASE */
+BUILD_OFFSET_SHOW(125, 680a); /* GUEST_SS_BASE */
+BUILD_OFFSET_SHOW(126, 680c); /* GUEST_DS_BASE */
+BUILD_OFFSET_SHOW(127, 680e); /* GUEST_FS_BASE */
+BUILD_OFFSET_SHOW(128, 6810); /* GUEST_GS_BASE */
+BUILD_OFFSET_SHOW(129, 6812); /* GUEST_LDTR_BASE */
+BUILD_OFFSET_SHOW(130, 6814); /* GUEST_TR_BASE */
+BUILD_OFFSET_SHOW(131, 6816); /* GUEST_GDTR_BASE */
+BUILD_OFFSET_SHOW(132, 6818); /* GUEST_IDTR_BASE */
+BUILD_OFFSET_SHOW(133, 681a); /* GUEST_DR7 */
+BUILD_OFFSET_SHOW(134, 681c); /* GUEST_RSP */
+BUILD_OFFSET_SHOW(135, 681e); /* GUEST_RIP */
+BUILD_OFFSET_SHOW(136, 6820); /* GUEST_RFLAGS */
+BUILD_OFFSET_SHOW(137, 6822); /* GUEST_PENDING_DBG_EXCEPTIONS */
+BUILD_OFFSET_SHOW(138, 6824); /* GUEST_SYSENTER_ESP */
+BUILD_OFFSET_SHOW(139, 6826); /* GUEST_SYSENTER_EIP */
+BUILD_OFFSET_SHOW(140, 6c00); /* HOST_CR0 */
+BUILD_OFFSET_SHOW(141, 6c02); /* HOST_CR3 */
+BUILD_OFFSET_SHOW(142, 6c04); /* HOST_CR4 */
+BUILD_OFFSET_SHOW(143, 6c06); /* HOST_FS_BASE */
+BUILD_OFFSET_SHOW(144, 6c08); /* HOST_GS_BASE */
+BUILD_OFFSET_SHOW(145, 6c0a); /* HOST_TR_BASE */
+BUILD_OFFSET_SHOW(146, 6c0c); /* HOST_GDTR_BASE */
+BUILD_OFFSET_SHOW(147, 6c0e); /* HOST_IDTR_BASE */
+BUILD_OFFSET_SHOW(148, 6c10); /* HOST_IA32_SYSENTER_ESP */
+BUILD_OFFSET_SHOW(149, 6c12); /* HOST_IA32_SYSENTER_EIP */
+BUILD_OFFSET_SHOW(150, 6c14); /* HOST_RSP */
+BUILD_OFFSET_SHOW(151, 6c16); /* HOST_RIP */
+
+static struct attribute *vmcs_attrs[] = {
+ &dev_attr_id.attr,
+ &dev_attr_0000.attr,
+ &dev_attr_0800.attr,
+ &dev_attr_0802.attr,
+ &dev_attr_0804.attr,
+ &dev_attr_0806.attr,
+ &dev_attr_0808.attr,
+ &dev_attr_080a.attr,
+ &dev_attr_080c.attr,
+ &dev_attr_080e.attr,
+ &dev_attr_0c00.attr,
+ &dev_attr_0c02.attr,
+ &dev_attr_0c04.attr,
+ &dev_attr_0c06.attr,
+ &dev_attr_0c08.attr,
+ &dev_attr_0c0a.attr,
+ &dev_attr_0c0c.attr,
+ &dev_attr_2000.attr,
+ &dev_attr_2001.attr,
+ &dev_attr_2002.attr,
+ &dev_attr_2003.attr,
+ &dev_attr_2004.attr,
+ &dev_attr_2005.attr,
+ &dev_attr_2006.attr,
+ &dev_attr_2007.attr,
+ &dev_attr_2008.attr,
+ &dev_attr_2009.attr,
+ &dev_attr_200a.attr,
+ &dev_attr_200b.attr,
+ &dev_attr_2010.attr,
+ &dev_attr_2011.attr,
+ &dev_attr_2012.attr,
+ &dev_attr_2013.attr,
+ &dev_attr_2014.attr,
+ &dev_attr_2015.attr,
+ &dev_attr_201a.attr,
+ &dev_attr_201b.attr,
+ &dev_attr_2400.attr,
+ &dev_attr_2401.attr,
+ &dev_attr_2800.attr,
+ &dev_attr_2801.attr,
+ &dev_attr_2802.attr,
+ &dev_attr_2803.attr,
+ &dev_attr_2804.attr,
+ &dev_attr_2805.attr,
+ &dev_attr_2806.attr,
+ &dev_attr_2807.attr,
+ &dev_attr_2808.attr,
+ &dev_attr_2809.attr,
+ &dev_attr_280a.attr,
+ &dev_attr_280b.attr,
+ &dev_attr_280c.attr,
+ &dev_attr_280d.attr,
+ &dev_attr_280e.attr,
+ &dev_attr_280f.attr,
+ &dev_attr_2810.attr,
+ &dev_attr_2811.attr,
+ &dev_attr_2c00.attr,
+ &dev_attr_2c01.attr,
+ &dev_attr_2c02.attr,
+ &dev_attr_2c03.attr,
+ &dev_attr_2c04.attr,
+ &dev_attr_2c05.attr,
+ &dev_attr_4000.attr,
+ &dev_attr_4002.attr,
+ &dev_attr_4004.attr,
+ &dev_attr_4006.attr,
+ &dev_attr_4008.attr,
+ &dev_attr_400a.attr,
+ &dev_attr_400c.attr,
+ &dev_attr_400e.attr,
+ &dev_attr_4010.attr,
+ &dev_attr_4012.attr,
+ &dev_attr_4014.attr,
+ &dev_attr_4016.attr,
+ &dev_attr_4018.attr,
+ &dev_attr_401a.attr,
+ &dev_attr_401c.attr,
+ &dev_attr_401e.attr,
+ &dev_attr_4020.attr,
+ &dev_attr_4022.attr,
+ &dev_attr_4400.attr,
+ &dev_attr_4402.attr,
+ &dev_attr_4404.attr,
+ &dev_attr_4406.attr,
+ &dev_attr_4408.attr,
+ &dev_attr_440a.attr,
+ &dev_attr_440c.attr,
+ &dev_attr_440e.attr,
+ &dev_attr_4800.attr,
+ &dev_attr_4802.attr,
+ &dev_attr_4804.attr,
+ &dev_attr_4806.attr,
+ &dev_attr_4808.attr,
+ &dev_attr_480a.attr,
+ &dev_attr_480c.attr,
+ &dev_attr_480e.attr,
+ &dev_attr_4810.attr,
+ &dev_attr_4812.attr,
+ &dev_attr_4814.attr,
+ &dev_attr_4816.attr,
+ &dev_attr_4818.attr,
+ &dev_attr_481a.attr,
+ &dev_attr_481c.attr,
+ &dev_attr_481e.attr,
+ &dev_attr_4820.attr,
+ &dev_attr_4822.attr,
+ &dev_attr_4824.attr,
+ &dev_attr_4826.attr,
+ &dev_attr_482A.attr,
+ &dev_attr_4c00.attr,
+ &dev_attr_6000.attr,
+ &dev_attr_6002.attr,
+ &dev_attr_6004.attr,
+ &dev_attr_6006.attr,
+ &dev_attr_6008.attr,
+ &dev_attr_600a.attr,
+ &dev_attr_600c.attr,
+ &dev_attr_600e.attr,
+ &dev_attr_6400.attr,
+ &dev_attr_640a.attr,
+ &dev_attr_6800.attr,
+ &dev_attr_6802.attr,
+ &dev_attr_6804.attr,
+ &dev_attr_6806.attr,
+ &dev_attr_6808.attr,
+ &dev_attr_680a.attr,
+ &dev_attr_680c.attr,
+ &dev_attr_680e.attr,
+ &dev_attr_6810.attr,
+ &dev_attr_6812.attr,
+ &dev_attr_6814.attr,
+ &dev_attr_6816.attr,
+ &dev_attr_6818.attr,
+ &dev_attr_681a.attr,
+ &dev_attr_681c.attr,
+ &dev_attr_681e.attr,
+ &dev_attr_6820.attr,
+ &dev_attr_6822.attr,
+ &dev_attr_6824.attr,
+ &dev_attr_6826.attr,
+ &dev_attr_6c00.attr,
+ &dev_attr_6c02.attr,
+ &dev_attr_6c04.attr,
+ &dev_attr_6c06.attr,
+ &dev_attr_6c08.attr,
+ &dev_attr_6c0a.attr,
+ &dev_attr_6c0c.attr,
+ &dev_attr_6c0e.attr,
+ &dev_attr_6c10.attr,
+ &dev_attr_6c12.attr,
+ &dev_attr_6c14.attr,
+ &dev_attr_6c16.attr,
+ NULL,
+};
+
+static struct attribute_group vmcs_attr_group = {
+ .name = vmcs_group_name,
+ .attrs = vmcs_attrs,
+};
+
+#define VMCSINFO_MAX_FIELD (ARRAY_SIZE(vmcs_attrs) - 1)
+
+static inline void vmcsinfo_revision_id(u32 id)
+{
+ vmcsinfo.vmcs_revision_id = id;
+}
+
+static inline void vmcsinfo_field(int i, u16 offset)
+{
+ if (i < VMCSINFO_MAX_FIELD)
+ vmcsinfo.field_offset[i][1] = offset;
+}
+
+/*
+ * For caculating offsets of fields in VMCS data, we index every 16-bit
+ * field by this kind of format:
+ * | --------- 16 bits ---------- |
+ * +-------------+-+------------+-+
+ * | high 7 bits |1| low 7 bits |0|
+ * +-------------+-+------------+-+
+ * In high byte, the lowest bit must be 1; In low byte, the lowest bit
+ * must be 0. The two bits are set like this in case indexes in VMCS
+ * data are read as big endian mode.
+ * The remaining 14 bits of the index indicate the real offset of the
+ * field. Because the size of a VMCS region is at most 4 KBytes, so
+ * 14 bits are enough to index the whole VMCS region.
+ *
+ * ENCODING_OFFSET: encode the offset into the index of this kind.
+ * DECODING_OFFSET: decode the index of this kind into real offset.
+ */
+#define OFFSET_HIGH_SHIFT (7)
+#define OFFSET_LOW_MASK ((1 << OFFSET_HIGH_SHIFT) - 1) /* 0x7f */
+#define OFFSET_HIGH_MASK (OFFSET_LOW_MASK << OFFSET_HIGH_SHIFT) /* 0x3f80 */
+#define ENCODING_OFFSET(offset) \
+ ((((offset) & OFFSET_LOW_MASK) << 1) + \
+ ((((offset) & OFFSET_HIGH_MASK) << 2) | 0x100))
+/*
+ * index here should be always read in little endian mode.
+ */
+#define DECODING_OFFSET_LE(index) \
+ ((((index) >> 1) & OFFSET_LOW_MASK) + \
+ (((index) >> 2) & OFFSET_HIGH_MASK))
+/*
+ * n indicates the bits of index. We first check if index
+ * is read in big endian mode.
+ */
+#define DECODING_OFFSET(index, n) \
+ ((index & 1) ? (DECODING_OFFSET_LE(__swab##n(index))) : \
+ (DECODING_OFFSET_LE(index)))
+
+#define FIELD_OFFSET16(i, offset) \
+ vmcsinfo_field(i, DECODING_OFFSET(offset, 16))
+#define FIELD_OFFSET64(i, offset) \
+ vmcsinfo_field(i, DECODING_OFFSET(offset, 64))
+#define FIELD_OFFSET32(i, offset) \
+ vmcsinfo_field(i, DECODING_OFFSET(offset, 32))
+#define FIELD_OFFSETNW(i, offset) \
+do { \
+ if (sizeof(offset) == 8) \
+ vmcsinfo_field(i, DECODING_OFFSET(offset, 64)); \
+ else \
+ vmcsinfo_field(i, DECODING_OFFSET(offset, 32)); \
+} while (0)
+
+#define VMCS_FIELD_CHECK(i, offset, type) \
+do { \
+ if (vmcs_read32(VM_INSTRUCTION_ERROR) != \
+ VMXERR_UNSUPPORTED_VMCS_COMPONENT) \
+ FIELD_OFFSET##type(i, offset); \
+} while (0)
+
+static inline void vmcs_read_checking(int i)
+{
+ unsigned long field;
+ u16 offset16;
+ u64 offset64;
+ u32 offset32;
+ unsigned long offsetnw;
+
+ field = vmcsinfo.field_offset[i][0];
+ switch (vmcs_field_type(field)) {
+ case VMCS_FIELD_TYPE_U16:
+ offset16 = vmcs_read16(field);
+ VMCS_FIELD_CHECK(i, offset16, 16);
+ break;
+ case VMCS_FIELD_TYPE_U64:
+ offset64 = vmcs_read64(field);
+ VMCS_FIELD_CHECK(i, offset64, 64);
+ break;
+ case VMCS_FIELD_TYPE_U32:
+ offset32 = vmcs_read32(field);
+ VMCS_FIELD_CHECK(i, offset32, 32);
+ break;
+ case VMCS_FIELD_TYPE_NATURAL_WIDTH:
+ offsetnw = vmcs_readl(field);
+ VMCS_FIELD_CHECK(i, offsetnw, NW);
+ break;
+ }
+}
+
+/*
+ * Note, offsets of fields that defined in Intel specification
+ * (Intel® 64 and IA-32 Architectures Software Developer’s Manual,
+ * Volume 3C) but not defined in *vmcs_field* will not be filled in
+ * VMCSINFO. And, some fields may be unsupported in some machines,
+ * in these machines, corresponding offsets will be zero.
+ */
+static int __init alloc_vmcsinfo_init(void)
+{
+/*
+ * The first 8 bytes in vmcs region are for
+ * VMCS revision identifier
+ * VMX-abort indicator
+ */
+#define FIELD_START (8)
+
+ int r, offset, i;
+ struct vmcs *vmcs;
+ int cpu;
+
+ vmcs = alloc_vmcs();
+ if (!vmcs) {
+ return -ENOMEM;
+ }
+
+ r = hardware_enable_all();
+ if (r)
+ goto out;
+
+ /*
+ * Write encoded offsets into VMCS data for later vmcs_read.
+ */
+ for (offset = FIELD_START; offset < vmcs_config.size;
+ offset += sizeof(u16))
+ *(u16 *)((char *)vmcs + offset) = ENCODING_OFFSET(offset);
+
+ cpu = get_cpu();
+ vmcs_clear(vmcs);
+ per_cpu(current_vmcs, cpu) = vmcs;
+ vmcs_load(vmcs);
+
+ vmcsinfo_revision_id(vmcs->revision_id);
+
+ for (i = 0; i < VMCSINFO_MAX_FIELD; ++i) {
+ if (vmcsinfo.field_offset[i][0] != VM_INSTRUCTION_ERROR)
+ continue;
+ vmcs_read_checking(i);
+ if (vmcsinfo.field_offset[i][1] == 0) {
+ goto out_clear;
+ } else {
+ offset = vmcsinfo.field_offset[i][1];
+ break;
+ }
+ }
+
+ for (i = 0; i < VMCSINFO_MAX_FIELD; ++i) {
+ if (vmcsinfo.field_offset[i][0] == VM_INSTRUCTION_ERROR)
+ continue;
+ /*
+ * Before each reading, zeroed field VM_INSTRUCTION_ERROR
+ */
+ *(u32 *)((char *)vmcs + offset) = 0;
+ vmcs_read_checking(i);
+ }
+
+ r = sysfs_create_group(&cpu_subsys.dev_root->kobj, &vmcs_attr_group);
+
+out_clear:
+ vmcs_clear(vmcs);
+ put_cpu();
+ hardware_disable_all();
+out:
+ free_vmcs(vmcs);
+ return r;
+}
+
+static void __exit alloc_vmcsinfo_exit(void)
+{
+ sysfs_remove_group(&cpu_subsys.dev_root->kobj, &vmcs_attr_group);
+}
+
+module_init(alloc_vmcsinfo_init);
+module_exit(alloc_vmcsinfo_exit);
This patch implements a new module named vmcsinfo-intel. The module fills VMCSINFO with the VMCS revision identifier, and offsets of VMCS fields. Note, offsets of fields that defined in Intel specification (Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 3C) but not defined in *vmcs_field* will not be filled in VMCSINFO. And, some fields may be unsupported in some machines, in these machines, corresponding offsets will be zero. Besides, this patch also exports vmcs revision identifier via /sys/devices/system/cpu/vmcs/id and offsets of fields via /sys/devices/system/cpu/vmcs/. Individual offsets are contained in subfiles named by the filed's encoding, e.g.: /sys/devices/cpu/vmcs/0800 Signed-off-by: zhangyanfei <zhangyanfei@cn.fujitsu.com> --- arch/x86/kvm/Kconfig | 11 + arch/x86/kvm/Makefile | 3 + arch/x86/kvm/vmcsinfo.c | 714 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 728 insertions(+), 0 deletions(-) create mode 100644 arch/x86/kvm/vmcsinfo.c