diff options
Diffstat (limited to 'kernel/arch/x86/xen')
32 files changed, 10120 insertions, 0 deletions
diff --git a/kernel/arch/x86/xen/Kconfig b/kernel/arch/x86/xen/Kconfig new file mode 100644 index 000000000..e88fda867 --- /dev/null +++ b/kernel/arch/x86/xen/Kconfig @@ -0,0 +1,52 @@ +# +# This Kconfig describes xen options +# + +config XEN + bool "Xen guest support" + depends on PARAVIRT + select PARAVIRT_CLOCK + select XEN_HAVE_PVMMU + depends on X86_64 || (X86_32 && X86_PAE) + depends on X86_TSC + help + This is the Linux Xen port. Enabling this will allow the + kernel to boot in a paravirtualized environment under the + Xen hypervisor. + +config XEN_DOM0 + def_bool y + depends on XEN && PCI_XEN && SWIOTLB_XEN + depends on X86_LOCAL_APIC && X86_IO_APIC && ACPI && PCI + +config XEN_PVHVM + def_bool y + depends on XEN && PCI && X86_LOCAL_APIC + +config XEN_MAX_DOMAIN_MEMORY + int + default 500 if X86_64 + default 64 if X86_32 + depends on XEN + help + This only affects the sizing of some bss arrays, the unused + portions of which are freed. + +config XEN_SAVE_RESTORE + bool + depends on XEN + select HIBERNATE_CALLBACKS + default y + +config XEN_DEBUG_FS + bool "Enable Xen debug and tuning parameters in debugfs" + depends on XEN && DEBUG_FS + default n + help + Enable statistics output and various tuning options in debugfs. + Enabling this option may incur a significant performance overhead. + +config XEN_PVH + bool "Support for running as a PVH guest" + depends on X86_64 && XEN && XEN_PVHVM + def_bool n diff --git a/kernel/arch/x86/xen/Makefile b/kernel/arch/x86/xen/Makefile new file mode 100644 index 000000000..7322755f3 --- /dev/null +++ b/kernel/arch/x86/xen/Makefile @@ -0,0 +1,25 @@ +ifdef CONFIG_FUNCTION_TRACER +# Do not profile debug and lowlevel utilities +CFLAGS_REMOVE_spinlock.o = -pg +CFLAGS_REMOVE_time.o = -pg +CFLAGS_REMOVE_irq.o = -pg +endif + +# Make sure early boot has no stackprotector +nostackp := $(call cc-option, -fno-stack-protector) +CFLAGS_enlighten.o := $(nostackp) +CFLAGS_mmu.o := $(nostackp) + +obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \ + time.o xen-asm.o xen-asm_$(BITS).o \ + grant-table.o suspend.o platform-pci-unplug.o \ + p2m.o + +obj-$(CONFIG_EVENT_TRACING) += trace.o + +obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o +obj-$(CONFIG_XEN_DEBUG_FS) += debugfs.o +obj-$(CONFIG_XEN_DOM0) += apic.o vga.o +obj-$(CONFIG_SWIOTLB_XEN) += pci-swiotlb-xen.o +obj-$(CONFIG_XEN_EFI) += efi.o diff --git a/kernel/arch/x86/xen/apic.c b/kernel/arch/x86/xen/apic.c new file mode 100644 index 000000000..70e060ad8 --- /dev/null +++ b/kernel/arch/x86/xen/apic.c @@ -0,0 +1,214 @@ +#include <linux/init.h> + +#include <asm/x86_init.h> +#include <asm/apic.h> +#include <asm/xen/hypercall.h> + +#include <xen/xen.h> +#include <xen/interface/physdev.h> +#include "xen-ops.h" +#include "smp.h" + +static unsigned int xen_io_apic_read(unsigned apic, unsigned reg) +{ + struct physdev_apic apic_op; + int ret; + + apic_op.apic_physbase = mpc_ioapic_addr(apic); + apic_op.reg = reg; + ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op); + if (!ret) + return apic_op.value; + + /* fallback to return an emulated IO_APIC values */ + if (reg == 0x1) + return 0x00170020; + else if (reg == 0x0) + return apic << 24; + + return 0xfd; +} + +static unsigned long xen_set_apic_id(unsigned int x) +{ + WARN_ON(1); + return x; +} + +static unsigned int xen_get_apic_id(unsigned long x) +{ + return ((x)>>24) & 0xFFu; +} + +static u32 xen_apic_read(u32 reg) +{ + struct xen_platform_op op = { + .cmd = XENPF_get_cpuinfo, + .interface_version = XENPF_INTERFACE_VERSION, + .u.pcpu_info.xen_cpuid = 0, + }; + int ret = 0; + + /* Shouldn't need this as APIC is turned off for PV, and we only + * get called on the bootup processor. But just in case. */ + if (!xen_initial_domain() || smp_processor_id()) + return 0; + + if (reg == APIC_LVR) + return 0x10; +#ifdef CONFIG_X86_32 + if (reg == APIC_LDR) + return SET_APIC_LOGICAL_ID(1UL << smp_processor_id()); +#endif + if (reg != APIC_ID) + return 0; + + ret = HYPERVISOR_dom0_op(&op); + if (ret) + return 0; + + return op.u.pcpu_info.apic_id << 24; +} + +static void xen_apic_write(u32 reg, u32 val) +{ + /* Warn to see if there's any stray references */ + WARN(1,"register: %x, value: %x\n", reg, val); +} + +static u64 xen_apic_icr_read(void) +{ + return 0; +} + +static void xen_apic_icr_write(u32 low, u32 id) +{ + /* Warn to see if there's any stray references */ + WARN_ON(1); +} + +static u32 xen_safe_apic_wait_icr_idle(void) +{ + return 0; +} + +static int xen_apic_probe_pv(void) +{ + if (xen_pv_domain()) + return 1; + + return 0; +} + +static int xen_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return xen_pv_domain(); +} + +static int xen_id_always_valid(int apicid) +{ + return 1; +} + +static int xen_id_always_registered(void) +{ + return 1; +} + +static int xen_phys_pkg_id(int initial_apic_id, int index_msb) +{ + return initial_apic_id >> index_msb; +} + +#ifdef CONFIG_X86_32 +static int xen_x86_32_early_logical_apicid(int cpu) +{ + /* Match with APIC_LDR read. Otherwise setup_local_APIC complains. */ + return 1 << cpu; +} +#endif + +static void xen_noop(void) +{ +} + +static void xen_silent_inquire(int apicid) +{ +} + +static struct apic xen_pv_apic = { + .name = "Xen PV", + .probe = xen_apic_probe_pv, + .acpi_madt_oem_check = xen_madt_oem_check, + .apic_id_valid = xen_id_always_valid, + .apic_id_registered = xen_id_always_registered, + + /* .irq_delivery_mode - used in native_compose_msi_msg only */ + /* .irq_dest_mode - used in native_compose_msi_msg only */ + + .target_cpus = default_target_cpus, + .disable_esr = 0, + /* .dest_logical - default_send_IPI_ use it but we use our own. */ + .check_apicid_used = default_check_apicid_used, /* Used on 32-bit */ + + .vector_allocation_domain = flat_vector_allocation_domain, + .init_apic_ldr = xen_noop, /* setup_local_APIC calls it */ + + .ioapic_phys_id_map = default_ioapic_phys_id_map, /* Used on 32-bit */ + .setup_apic_routing = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = physid_set_mask_of_physid, /* Used on 32-bit */ + .check_phys_apicid_present = default_check_phys_apicid_present, /* smp_sanity_check needs it */ + .phys_pkg_id = xen_phys_pkg_id, /* detect_ht */ + + .get_apic_id = xen_get_apic_id, + .set_apic_id = xen_set_apic_id, /* Can be NULL on 32-bit. */ + .apic_id_mask = 0xFF << 24, /* Used by verify_local_APIC. Match with what xen_get_apic_id does. */ + + .cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and, + +#ifdef CONFIG_SMP + .send_IPI_mask = xen_send_IPI_mask, + .send_IPI_mask_allbutself = xen_send_IPI_mask_allbutself, + .send_IPI_allbutself = xen_send_IPI_allbutself, + .send_IPI_all = xen_send_IPI_all, + .send_IPI_self = xen_send_IPI_self, +#endif + /* .wait_for_init_deassert- used by AP bootup - smp_callin which we don't use */ + .inquire_remote_apic = xen_silent_inquire, + + .read = xen_apic_read, + .write = xen_apic_write, + .eoi_write = xen_apic_write, + + .icr_read = xen_apic_icr_read, + .icr_write = xen_apic_icr_write, + .wait_icr_idle = xen_noop, + .safe_wait_icr_idle = xen_safe_apic_wait_icr_idle, + +#ifdef CONFIG_X86_32 + /* generic_processor_info and setup_local_APIC. */ + .x86_32_early_logical_apicid = xen_x86_32_early_logical_apicid, +#endif +}; + +static void __init xen_apic_check(void) +{ + if (apic == &xen_pv_apic) + return; + + pr_info("Switched APIC routing from %s to %s.\n", apic->name, + xen_pv_apic.name); + apic = &xen_pv_apic; +} +void __init xen_init_apic(void) +{ + x86_io_apic_ops.read = xen_io_apic_read; + /* On PV guests the APIC CPUID bit is disabled so none of the + * routines end up executing. */ + if (!xen_initial_domain()) + apic = &xen_pv_apic; + + x86_platform.apic_post_init = xen_apic_check; +} +apic_driver(xen_pv_apic); diff --git a/kernel/arch/x86/xen/debugfs.c b/kernel/arch/x86/xen/debugfs.c new file mode 100644 index 000000000..c8377fb26 --- /dev/null +++ b/kernel/arch/x86/xen/debugfs.c @@ -0,0 +1,21 @@ +#include <linux/init.h> +#include <linux/debugfs.h> +#include <linux/slab.h> +#include <linux/module.h> + +#include "debugfs.h" + +static struct dentry *d_xen_debug; + +struct dentry * __init xen_init_debugfs(void) +{ + if (!d_xen_debug) { + d_xen_debug = debugfs_create_dir("xen", NULL); + + if (!d_xen_debug) + pr_warning("Could not create 'xen' debugfs directory\n"); + } + + return d_xen_debug; +} + diff --git a/kernel/arch/x86/xen/debugfs.h b/kernel/arch/x86/xen/debugfs.h new file mode 100644 index 000000000..12ebf3325 --- /dev/null +++ b/kernel/arch/x86/xen/debugfs.h @@ -0,0 +1,6 @@ +#ifndef _XEN_DEBUGFS_H +#define _XEN_DEBUGFS_H + +struct dentry * __init xen_init_debugfs(void); + +#endif /* _XEN_DEBUGFS_H */ diff --git a/kernel/arch/x86/xen/efi.c b/kernel/arch/x86/xen/efi.c new file mode 100644 index 000000000..be14cc3e4 --- /dev/null +++ b/kernel/arch/x86/xen/efi.c @@ -0,0 +1,45 @@ +/* + * Copyright (c) 2014 Oracle Co., Daniel Kiper + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/bitops.h> +#include <linux/efi.h> +#include <linux/init.h> +#include <linux/string.h> + +#include <xen/xen-ops.h> + +#include <asm/page.h> +#include <asm/setup.h> + +void __init xen_efi_init(void) +{ + efi_system_table_t *efi_systab_xen; + + efi_systab_xen = xen_efi_probe(); + + if (efi_systab_xen == NULL) + return; + + strncpy((char *)&boot_params.efi_info.efi_loader_signature, "Xen", + sizeof(boot_params.efi_info.efi_loader_signature)); + boot_params.efi_info.efi_systab = (__u32)__pa(efi_systab_xen); + boot_params.efi_info.efi_systab_hi = (__u32)(__pa(efi_systab_xen) >> 32); + + set_bit(EFI_BOOT, &efi.flags); + set_bit(EFI_PARAVIRT, &efi.flags); + set_bit(EFI_64BIT, &efi.flags); +} diff --git a/kernel/arch/x86/xen/enlighten.c b/kernel/arch/x86/xen/enlighten.c new file mode 100644 index 000000000..46957ead3 --- /dev/null +++ b/kernel/arch/x86/xen/enlighten.c @@ -0,0 +1,1828 @@ +/* + * Core of Xen paravirt_ops implementation. + * + * This file contains the xen_paravirt_ops structure itself, and the + * implementations for: + * - privileged instructions + * - interrupt flags + * - segment operations + * - booting and setup + * + * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007 + */ + +#include <linux/cpu.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/smp.h> +#include <linux/preempt.h> +#include <linux/hardirq.h> +#include <linux/percpu.h> +#include <linux/delay.h> +#include <linux/start_kernel.h> +#include <linux/sched.h> +#include <linux/kprobes.h> +#include <linux/bootmem.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/page-flags.h> +#include <linux/highmem.h> +#include <linux/console.h> +#include <linux/pci.h> +#include <linux/gfp.h> +#include <linux/memblock.h> +#include <linux/edd.h> + +#include <xen/xen.h> +#include <xen/events.h> +#include <xen/interface/xen.h> +#include <xen/interface/version.h> +#include <xen/interface/physdev.h> +#include <xen/interface/vcpu.h> +#include <xen/interface/memory.h> +#include <xen/interface/nmi.h> +#include <xen/interface/xen-mca.h> +#include <xen/features.h> +#include <xen/page.h> +#include <xen/hvm.h> +#include <xen/hvc-console.h> +#include <xen/acpi.h> + +#include <asm/paravirt.h> +#include <asm/apic.h> +#include <asm/page.h> +#include <asm/xen/pci.h> +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> +#include <asm/fixmap.h> +#include <asm/processor.h> +#include <asm/proto.h> +#include <asm/msr-index.h> +#include <asm/traps.h> +#include <asm/setup.h> +#include <asm/desc.h> +#include <asm/pgalloc.h> +#include <asm/pgtable.h> +#include <asm/tlbflush.h> +#include <asm/reboot.h> +#include <asm/stackprotector.h> +#include <asm/hypervisor.h> +#include <asm/mach_traps.h> +#include <asm/mwait.h> +#include <asm/pci_x86.h> +#include <asm/pat.h> + +#ifdef CONFIG_ACPI +#include <linux/acpi.h> +#include <asm/acpi.h> +#include <acpi/pdc_intel.h> +#include <acpi/processor.h> +#include <xen/interface/platform.h> +#endif + +#include "xen-ops.h" +#include "mmu.h" +#include "smp.h" +#include "multicalls.h" + +EXPORT_SYMBOL_GPL(hypercall_page); + +/* + * Pointer to the xen_vcpu_info structure or + * &HYPERVISOR_shared_info->vcpu_info[cpu]. See xen_hvm_init_shared_info + * and xen_vcpu_setup for details. By default it points to share_info->vcpu_info + * but if the hypervisor supports VCPUOP_register_vcpu_info then it can point + * to xen_vcpu_info. The pointer is used in __xen_evtchn_do_upcall to + * acknowledge pending events. + * Also more subtly it is used by the patched version of irq enable/disable + * e.g. xen_irq_enable_direct and xen_iret in PV mode. + * + * The desire to be able to do those mask/unmask operations as a single + * instruction by using the per-cpu offset held in %gs is the real reason + * vcpu info is in a per-cpu pointer and the original reason for this + * hypercall. + * + */ +DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu); + +/* + * Per CPU pages used if hypervisor supports VCPUOP_register_vcpu_info + * hypercall. This can be used both in PV and PVHVM mode. The structure + * overrides the default per_cpu(xen_vcpu, cpu) value. + */ +DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info); + +enum xen_domain_type xen_domain_type = XEN_NATIVE; +EXPORT_SYMBOL_GPL(xen_domain_type); + +unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START; +EXPORT_SYMBOL(machine_to_phys_mapping); +unsigned long machine_to_phys_nr; +EXPORT_SYMBOL(machine_to_phys_nr); + +struct start_info *xen_start_info; +EXPORT_SYMBOL_GPL(xen_start_info); + +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 + * page as soon as fixmap is up and running. + */ +struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info; + +/* + * Flag to determine whether vcpu info placement is available on all + * VCPUs. We assume it is to start with, and then set it to zero on + * the first failure. This is because it can succeed on some VCPUs + * and not others, since it can involve hypervisor memory allocation, + * or because the guest failed to guarantee all the appropriate + * constraints on all VCPUs (ie buffer can't cross a page boundary). + * + * Note that any particular CPU may be using a placed vcpu structure, + * but we can only optimise if the all are. + * + * 0: not available, 1: available + */ +static int have_vcpu_info_placement = 1; + +struct tls_descs { + struct desc_struct desc[3]; +}; + +/* + * Updating the 3 TLS descriptors in the GDT on every task switch is + * surprisingly expensive so we avoid updating them if they haven't + * changed. Since Xen writes different descriptors than the one + * passed in the update_descriptor hypercall we keep shadow copies to + * compare against. + */ +static DEFINE_PER_CPU(struct tls_descs, shadow_tls_desc); + +static void clamp_max_cpus(void) +{ +#ifdef CONFIG_SMP + if (setup_max_cpus > MAX_VIRT_CPUS) + setup_max_cpus = MAX_VIRT_CPUS; +#endif +} + +static void xen_vcpu_setup(int cpu) +{ + struct vcpu_register_vcpu_info info; + int err; + struct vcpu_info *vcpup; + + BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info); + + /* + * This path is called twice on PVHVM - first during bootup via + * smp_init -> xen_hvm_cpu_notify, and then if the VCPU is being + * hotplugged: cpu_up -> xen_hvm_cpu_notify. + * As we can only do the VCPUOP_register_vcpu_info once lets + * not over-write its result. + * + * For PV it is called during restore (xen_vcpu_restore) and bootup + * (xen_setup_vcpu_info_placement). The hotplug mechanism does not + * use this function. + */ + if (xen_hvm_domain()) { + if (per_cpu(xen_vcpu, cpu) == &per_cpu(xen_vcpu_info, cpu)) + return; + } + if (cpu < MAX_VIRT_CPUS) + per_cpu(xen_vcpu,cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu]; + + if (!have_vcpu_info_placement) { + if (cpu >= MAX_VIRT_CPUS) + clamp_max_cpus(); + return; + } + + vcpup = &per_cpu(xen_vcpu_info, cpu); + info.mfn = arbitrary_virt_to_mfn(vcpup); + info.offset = offset_in_page(vcpup); + + /* Check to see if the hypervisor will put the vcpu_info + structure where we want it, which allows direct access via + a percpu-variable. + N.B. This hypercall can _only_ be called once per CPU. Subsequent + calls will error out with -EINVAL. This is due to the fact that + hypervisor has no unregister variant and this hypercall does not + allow to over-write info.mfn and info.offset. + */ + err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info, cpu, &info); + + if (err) { + printk(KERN_DEBUG "register_vcpu_info failed: err=%d\n", err); + have_vcpu_info_placement = 0; + clamp_max_cpus(); + } else { + /* This cpu is using the registered vcpu info, even if + later ones fail to. */ + per_cpu(xen_vcpu, cpu) = vcpup; + } +} + +/* + * On restore, set the vcpu placement up again. + * If it fails, then we're in a bad state, since + * we can't back out from using it... + */ +void xen_vcpu_restore(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + bool other_cpu = (cpu != smp_processor_id()); + bool is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL); + + if (other_cpu && is_up && + HYPERVISOR_vcpu_op(VCPUOP_down, cpu, NULL)) + BUG(); + + xen_setup_runstate_info(cpu); + + if (have_vcpu_info_placement) + xen_vcpu_setup(cpu); + + if (other_cpu && is_up && + HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL)) + BUG(); + } +} + +static void __init xen_banner(void) +{ + unsigned version = HYPERVISOR_xen_version(XENVER_version, NULL); + struct xen_extraversion extra; + HYPERVISOR_xen_version(XENVER_extraversion, &extra); + + pr_info("Booting paravirtualized kernel %son %s\n", + xen_feature(XENFEAT_auto_translated_physmap) ? + "with PVH extensions " : "", pv_info.name); + printk(KERN_INFO "Xen version: %d.%d%s%s\n", + version >> 16, version & 0xffff, extra.extraversion, + xen_feature(XENFEAT_mmu_pt_update_preserve_ad) ? " (preserve-AD)" : ""); +} +/* Check if running on Xen version (major, minor) or later */ +bool +xen_running_on_version_or_later(unsigned int major, unsigned int minor) +{ + unsigned int version; + + if (!xen_domain()) + return false; + + version = HYPERVISOR_xen_version(XENVER_version, NULL); + if ((((version >> 16) == major) && ((version & 0xffff) >= minor)) || + ((version >> 16) > major)) + return true; + return false; +} + +#define CPUID_THERM_POWER_LEAF 6 +#define APERFMPERF_PRESENT 0 + +static __read_mostly unsigned int cpuid_leaf1_edx_mask = ~0; +static __read_mostly unsigned int cpuid_leaf1_ecx_mask = ~0; + +static __read_mostly unsigned int cpuid_leaf1_ecx_set_mask; +static __read_mostly unsigned int cpuid_leaf5_ecx_val; +static __read_mostly unsigned int cpuid_leaf5_edx_val; + +static void xen_cpuid(unsigned int *ax, unsigned int *bx, + unsigned int *cx, unsigned int *dx) +{ + unsigned maskebx = ~0; + unsigned maskecx = ~0; + unsigned maskedx = ~0; + unsigned setecx = 0; + /* + * Mask out inconvenient features, to try and disable as many + * unsupported kernel subsystems as possible. + */ + switch (*ax) { + case 1: + maskecx = cpuid_leaf1_ecx_mask; + setecx = cpuid_leaf1_ecx_set_mask; + maskedx = cpuid_leaf1_edx_mask; + break; + + case CPUID_MWAIT_LEAF: + /* Synthesize the values.. */ + *ax = 0; + *bx = 0; + *cx = cpuid_leaf5_ecx_val; + *dx = cpuid_leaf5_edx_val; + return; + + case CPUID_THERM_POWER_LEAF: + /* Disabling APERFMPERF for kernel usage */ + maskecx = ~(1 << APERFMPERF_PRESENT); + break; + + case 0xb: + /* Suppress extended topology stuff */ + maskebx = 0; + break; + } + + asm(XEN_EMULATE_PREFIX "cpuid" + : "=a" (*ax), + "=b" (*bx), + "=c" (*cx), + "=d" (*dx) + : "0" (*ax), "2" (*cx)); + + *bx &= maskebx; + *cx &= maskecx; + *cx |= setecx; + *dx &= maskedx; + +} + +static bool __init xen_check_mwait(void) +{ +#ifdef CONFIG_ACPI + struct xen_platform_op op = { + .cmd = XENPF_set_processor_pminfo, + .u.set_pminfo.id = -1, + .u.set_pminfo.type = XEN_PM_PDC, + }; + uint32_t buf[3]; + unsigned int ax, bx, cx, dx; + unsigned int mwait_mask; + + /* We need to determine whether it is OK to expose the MWAIT + * capability to the kernel to harvest deeper than C3 states from ACPI + * _CST using the processor_harvest_xen.c module. For this to work, we + * need to gather the MWAIT_LEAF values (which the cstate.c code + * checks against). The hypervisor won't expose the MWAIT flag because + * it would break backwards compatibility; so we will find out directly + * from the hardware and hypercall. + */ + if (!xen_initial_domain()) + return false; + + /* + * When running under platform earlier than Xen4.2, do not expose + * mwait, to avoid the risk of loading native acpi pad driver + */ + if (!xen_running_on_version_or_later(4, 2)) + return false; + + ax = 1; + cx = 0; + + native_cpuid(&ax, &bx, &cx, &dx); + + mwait_mask = (1 << (X86_FEATURE_EST % 32)) | + (1 << (X86_FEATURE_MWAIT % 32)); + + if ((cx & mwait_mask) != mwait_mask) + return false; + + /* We need to emulate the MWAIT_LEAF and for that we need both + * ecx and edx. The hypercall provides only partial information. + */ + + ax = CPUID_MWAIT_LEAF; + bx = 0; + cx = 0; + dx = 0; + + native_cpuid(&ax, &bx, &cx, &dx); + + /* Ask the Hypervisor whether to clear ACPI_PDC_C_C2C3_FFH. If so, + * don't expose MWAIT_LEAF and let ACPI pick the IOPORT version of C3. + */ + buf[0] = ACPI_PDC_REVISION_ID; + buf[1] = 1; + buf[2] = (ACPI_PDC_C_CAPABILITY_SMP | ACPI_PDC_EST_CAPABILITY_SWSMP); + + set_xen_guest_handle(op.u.set_pminfo.pdc, buf); + + if ((HYPERVISOR_dom0_op(&op) == 0) && + (buf[2] & (ACPI_PDC_C_C1_FFH | ACPI_PDC_C_C2C3_FFH))) { + cpuid_leaf5_ecx_val = cx; + cpuid_leaf5_edx_val = dx; + } + return true; +#else + return false; +#endif +} +static void __init xen_init_cpuid_mask(void) +{ + unsigned int ax, bx, cx, dx; + unsigned int xsave_mask; + + cpuid_leaf1_edx_mask = + ~((1 << X86_FEATURE_MTRR) | /* disable MTRR */ + (1 << X86_FEATURE_ACC)); /* thermal monitoring */ + + if (!xen_initial_domain()) + cpuid_leaf1_edx_mask &= + ~((1 << X86_FEATURE_ACPI)); /* disable ACPI */ + + cpuid_leaf1_ecx_mask &= ~(1 << (X86_FEATURE_X2APIC % 32)); + + ax = 1; + cx = 0; + cpuid(1, &ax, &bx, &cx, &dx); + + xsave_mask = + (1 << (X86_FEATURE_XSAVE % 32)) | + (1 << (X86_FEATURE_OSXSAVE % 32)); + + /* Xen will set CR4.OSXSAVE if supported and not disabled by force */ + if ((cx & xsave_mask) != xsave_mask) + cpuid_leaf1_ecx_mask &= ~xsave_mask; /* disable XSAVE & OSXSAVE */ + if (xen_check_mwait()) + cpuid_leaf1_ecx_set_mask = (1 << (X86_FEATURE_MWAIT % 32)); +} + +static void xen_set_debugreg(int reg, unsigned long val) +{ + HYPERVISOR_set_debugreg(reg, val); +} + +static unsigned long xen_get_debugreg(int reg) +{ + return HYPERVISOR_get_debugreg(reg); +} + +static void xen_end_context_switch(struct task_struct *next) +{ + xen_mc_flush(); + paravirt_end_context_switch(next); +} + +static unsigned long xen_store_tr(void) +{ + return 0; +} + +/* + * Set the page permissions for a particular virtual address. If the + * address is a vmalloc mapping (or other non-linear mapping), then + * find the linear mapping of the page and also set its protections to + * match. + */ +static void set_aliased_prot(void *v, pgprot_t prot) +{ + int level; + pte_t *ptep; + pte_t pte; + unsigned long pfn; + struct page *page; + + ptep = lookup_address((unsigned long)v, &level); + BUG_ON(ptep == NULL); + + pfn = pte_pfn(*ptep); + page = pfn_to_page(pfn); + + pte = pfn_pte(pfn, prot); + + if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0)) + BUG(); + + if (!PageHighMem(page)) { + void *av = __va(PFN_PHYS(pfn)); + + if (av != v) + if (HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0)) + BUG(); + } else + kmap_flush_unused(); +} + +static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries) +{ + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; + + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL_RO); +} + +static void xen_free_ldt(struct desc_struct *ldt, unsigned entries) +{ + const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE; + int i; + + for(i = 0; i < entries; i += entries_per_page) + set_aliased_prot(ldt + i, PAGE_KERNEL); +} + +static void xen_set_ldt(const void *addr, unsigned entries) +{ + struct mmuext_op *op; + struct multicall_space mcs = xen_mc_entry(sizeof(*op)); + + trace_xen_cpu_set_ldt(addr, entries); + + op = mcs.args; + op->cmd = MMUEXT_SET_LDT; + op->arg1.linear_addr = (unsigned long)addr; + op->arg2.nr_ents = entries; + + MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); + + xen_mc_issue(PARAVIRT_LAZY_CPU); +} + +static void xen_load_gdt(const struct desc_ptr *dtr) +{ + unsigned long va = dtr->address; + unsigned int size = dtr->size + 1; + unsigned pages = (size + PAGE_SIZE - 1) / PAGE_SIZE; + unsigned long frames[pages]; + int f; + + /* + * A GDT can be up to 64k in size, which corresponds to 8192 + * 8-byte entries, or 16 4k pages.. + */ + + BUG_ON(size > 65536); + BUG_ON(va & ~PAGE_MASK); + + for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) { + int level; + pte_t *ptep; + unsigned long pfn, mfn; + void *virt; + + /* + * The GDT is per-cpu and is in the percpu data area. + * That can be virtually mapped, so we need to do a + * page-walk to get the underlying MFN for the + * hypercall. The page can also be in the kernel's + * linear range, so we need to RO that mapping too. + */ + ptep = lookup_address(va, &level); + BUG_ON(ptep == NULL); + + pfn = pte_pfn(*ptep); + mfn = pfn_to_mfn(pfn); + virt = __va(PFN_PHYS(pfn)); + + frames[f] = mfn; + + make_lowmem_page_readonly((void *)va); + make_lowmem_page_readonly(virt); + } + + if (HYPERVISOR_set_gdt(frames, size / sizeof(struct desc_struct))) + BUG(); +} + +/* + * load_gdt for early boot, when the gdt is only mapped once + */ +static void __init xen_load_gdt_boot(const struct desc_ptr *dtr) +{ + unsigned long va = dtr->address; + unsigned int size = dtr->size + 1; + unsigned pages = (size + PAGE_SIZE - 1) / PAGE_SIZE; + unsigned long frames[pages]; + int f; + + /* + * A GDT can be up to 64k in size, which corresponds to 8192 + * 8-byte entries, or 16 4k pages.. + */ + + BUG_ON(size > 65536); + BUG_ON(va & ~PAGE_MASK); + + for (f = 0; va < dtr->address + size; va += PAGE_SIZE, f++) { + pte_t pte; + unsigned long pfn, mfn; + + pfn = virt_to_pfn(va); + mfn = pfn_to_mfn(pfn); + + pte = pfn_pte(pfn, PAGE_KERNEL_RO); + + if (HYPERVISOR_update_va_mapping((unsigned long)va, pte, 0)) + BUG(); + + frames[f] = mfn; + } + + if (HYPERVISOR_set_gdt(frames, size / sizeof(struct desc_struct))) + BUG(); +} + +static inline bool desc_equal(const struct desc_struct *d1, + const struct desc_struct *d2) +{ + return d1->a == d2->a && d1->b == d2->b; +} + +static void load_TLS_descriptor(struct thread_struct *t, + unsigned int cpu, unsigned int i) +{ + struct desc_struct *shadow = &per_cpu(shadow_tls_desc, cpu).desc[i]; + struct desc_struct *gdt; + xmaddr_t maddr; + struct multicall_space mc; + + if (desc_equal(shadow, &t->tls_array[i])) + return; + + *shadow = t->tls_array[i]; + + gdt = get_cpu_gdt_table(cpu); + maddr = arbitrary_virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]); + mc = __xen_mc_entry(0); + + MULTI_update_descriptor(mc.mc, maddr.maddr, t->tls_array[i]); +} + +static void xen_load_tls(struct thread_struct *t, unsigned int cpu) +{ + /* + * XXX sleazy hack: If we're being called in a lazy-cpu zone + * and lazy gs handling is enabled, it means we're in a + * context switch, and %gs has just been saved. This means we + * can zero it out to prevent faults on exit from the + * hypervisor if the next process has no %gs. Either way, it + * has been saved, and the new value will get loaded properly. + * This will go away as soon as Xen has been modified to not + * save/restore %gs for normal hypercalls. + * + * On x86_64, this hack is not used for %gs, because gs points + * to KERNEL_GS_BASE (and uses it for PDA references), so we + * must not zero %gs on x86_64 + * + * For x86_64, we need to zero %fs, otherwise we may get an + * exception between the new %fs descriptor being loaded and + * %fs being effectively cleared at __switch_to(). + */ + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) { +#ifdef CONFIG_X86_32 + lazy_load_gs(0); +#else + loadsegment(fs, 0); +#endif + } + + xen_mc_batch(); + + load_TLS_descriptor(t, cpu, 0); + load_TLS_descriptor(t, cpu, 1); + load_TLS_descriptor(t, cpu, 2); + + xen_mc_issue(PARAVIRT_LAZY_CPU); +} + +#ifdef CONFIG_X86_64 +static void xen_load_gs_index(unsigned int idx) +{ + if (HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, idx)) + BUG(); +} +#endif + +static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum, + const void *ptr) +{ + xmaddr_t mach_lp = arbitrary_virt_to_machine(&dt[entrynum]); + u64 entry = *(u64 *)ptr; + + trace_xen_cpu_write_ldt_entry(dt, entrynum, entry); + + preempt_disable(); + + xen_mc_flush(); + if (HYPERVISOR_update_descriptor(mach_lp.maddr, entry)) + BUG(); + + preempt_enable(); +} + +static int cvt_gate_to_trap(int vector, const gate_desc *val, + struct trap_info *info) +{ + unsigned long addr; + + if (val->type != GATE_TRAP && val->type != GATE_INTERRUPT) + return 0; + + info->vector = vector; + + addr = gate_offset(*val); +#ifdef CONFIG_X86_64 + /* + * Look for known traps using IST, and substitute them + * appropriately. The debugger ones are the only ones we care + * about. Xen will handle faults like double_fault, + * so we should never see them. Warn if + * there's an unexpected IST-using fault handler. + */ + if (addr == (unsigned long)debug) + addr = (unsigned long)xen_debug; + else if (addr == (unsigned long)int3) + addr = (unsigned long)xen_int3; + else if (addr == (unsigned long)stack_segment) + addr = (unsigned long)xen_stack_segment; + else if (addr == (unsigned long)double_fault) { + /* Don't need to handle these */ + return 0; +#ifdef CONFIG_X86_MCE + } else if (addr == (unsigned long)machine_check) { + /* + * when xen hypervisor inject vMCE to guest, + * use native mce handler to handle it + */ + ; +#endif + } else if (addr == (unsigned long)nmi) + /* + * Use the native version as well. + */ + ; + else { + /* Some other trap using IST? */ + if (WARN_ON(val->ist != 0)) + return 0; + } +#endif /* CONFIG_X86_64 */ + info->address = addr; + + info->cs = gate_segment(*val); + info->flags = val->dpl; + /* interrupt gates clear IF */ + if (val->type == GATE_INTERRUPT) + info->flags |= 1 << 2; + + return 1; +} + +/* Locations of each CPU's IDT */ +static DEFINE_PER_CPU(struct desc_ptr, idt_desc); + +/* Set an IDT entry. If the entry is part of the current IDT, then + also update Xen. */ +static void xen_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g) +{ + unsigned long p = (unsigned long)&dt[entrynum]; + unsigned long start, end; + + trace_xen_cpu_write_idt_entry(dt, entrynum, g); + + preempt_disable(); + + start = __this_cpu_read(idt_desc.address); + end = start + __this_cpu_read(idt_desc.size) + 1; + + xen_mc_flush(); + + native_write_idt_entry(dt, entrynum, g); + + if (p >= start && (p + 8) <= end) { + struct trap_info info[2]; + + info[1].address = 0; + + if (cvt_gate_to_trap(entrynum, g, &info[0])) + if (HYPERVISOR_set_trap_table(info)) + BUG(); + } + + preempt_enable(); +} + +static void xen_convert_trap_info(const struct desc_ptr *desc, + struct trap_info *traps) +{ + unsigned in, out, count; + + count = (desc->size+1) / sizeof(gate_desc); + BUG_ON(count > 256); + + for (in = out = 0; in < count; in++) { + gate_desc *entry = (gate_desc*)(desc->address) + in; + + if (cvt_gate_to_trap(in, entry, &traps[out])) + out++; + } + traps[out].address = 0; +} + +void xen_copy_trap_info(struct trap_info *traps) +{ + const struct desc_ptr *desc = this_cpu_ptr(&idt_desc); + + xen_convert_trap_info(desc, traps); +} + +/* Load a new IDT into Xen. In principle this can be per-CPU, so we + hold a spinlock to protect the static traps[] array (static because + it avoids allocation, and saves stack space). */ +static void xen_load_idt(const struct desc_ptr *desc) +{ + static DEFINE_SPINLOCK(lock); + static struct trap_info traps[257]; + + trace_xen_cpu_load_idt(desc); + + spin_lock(&lock); + + memcpy(this_cpu_ptr(&idt_desc), desc, sizeof(idt_desc)); + + xen_convert_trap_info(desc, traps); + + xen_mc_flush(); + if (HYPERVISOR_set_trap_table(traps)) + BUG(); + + spin_unlock(&lock); +} + +/* Write a GDT descriptor entry. Ignore LDT descriptors, since + they're handled differently. */ +static void xen_write_gdt_entry(struct desc_struct *dt, int entry, + const void *desc, int type) +{ + trace_xen_cpu_write_gdt_entry(dt, entry, desc, type); + + preempt_disable(); + + switch (type) { + case DESC_LDT: + case DESC_TSS: + /* ignore */ + break; + + default: { + xmaddr_t maddr = arbitrary_virt_to_machine(&dt[entry]); + + xen_mc_flush(); + if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc)) + BUG(); + } + + } + + preempt_enable(); +} + +/* + * Version of write_gdt_entry for use at early boot-time needed to + * update an entry as simply as possible. + */ +static void __init xen_write_gdt_entry_boot(struct desc_struct *dt, int entry, + const void *desc, int type) +{ + trace_xen_cpu_write_gdt_entry(dt, entry, desc, type); + + switch (type) { + case DESC_LDT: + case DESC_TSS: + /* ignore */ + break; + + default: { + xmaddr_t maddr = virt_to_machine(&dt[entry]); + + if (HYPERVISOR_update_descriptor(maddr.maddr, *(u64 *)desc)) + dt[entry] = *(struct desc_struct *)desc; + } + + } +} + +static void xen_load_sp0(struct tss_struct *tss, + struct thread_struct *thread) +{ + struct multicall_space mcs; + + mcs = xen_mc_entry(0); + MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0); + xen_mc_issue(PARAVIRT_LAZY_CPU); + tss->x86_tss.sp0 = thread->sp0; +} + +static void xen_set_iopl_mask(unsigned mask) +{ + struct physdev_set_iopl set_iopl; + + /* Force the change at ring 0. */ + set_iopl.iopl = (mask == 0) ? 1 : (mask >> 12) & 3; + HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl); +} + +static void xen_io_delay(void) +{ +} + +static void xen_clts(void) +{ + struct multicall_space mcs; + + mcs = xen_mc_entry(0); + + MULTI_fpu_taskswitch(mcs.mc, 0); + + xen_mc_issue(PARAVIRT_LAZY_CPU); +} + +static DEFINE_PER_CPU(unsigned long, xen_cr0_value); + +static unsigned long xen_read_cr0(void) +{ + unsigned long cr0 = this_cpu_read(xen_cr0_value); + + if (unlikely(cr0 == 0)) { + cr0 = native_read_cr0(); + this_cpu_write(xen_cr0_value, cr0); + } + + return cr0; +} + +static void xen_write_cr0(unsigned long cr0) +{ + struct multicall_space mcs; + + this_cpu_write(xen_cr0_value, cr0); + + /* Only pay attention to cr0.TS; everything else is + ignored. */ + mcs = xen_mc_entry(0); + + MULTI_fpu_taskswitch(mcs.mc, (cr0 & X86_CR0_TS) != 0); + + xen_mc_issue(PARAVIRT_LAZY_CPU); +} + +static void xen_write_cr4(unsigned long cr4) +{ + cr4 &= ~X86_CR4_PGE; + cr4 &= ~X86_CR4_PSE; + + native_write_cr4(cr4); +} +#ifdef CONFIG_X86_64 +static inline unsigned long xen_read_cr8(void) +{ + return 0; +} +static inline void xen_write_cr8(unsigned long val) +{ + BUG_ON(val); +} +#endif + +static u64 xen_read_msr_safe(unsigned int msr, int *err) +{ + u64 val; + + val = native_read_msr_safe(msr, err); + switch (msr) { + case MSR_IA32_APICBASE: +#ifdef CONFIG_X86_X2APIC + if (!(cpuid_ecx(1) & (1 << (X86_FEATURE_X2APIC & 31)))) +#endif + val &= ~X2APIC_ENABLE; + break; + } + return val; +} + +static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) +{ + int ret; + + ret = 0; + + switch (msr) { +#ifdef CONFIG_X86_64 + unsigned which; + u64 base; + + case MSR_FS_BASE: which = SEGBASE_FS; goto set; + case MSR_KERNEL_GS_BASE: which = SEGBASE_GS_USER; goto set; + case MSR_GS_BASE: which = SEGBASE_GS_KERNEL; goto set; + + set: + base = ((u64)high << 32) | low; + if (HYPERVISOR_set_segment_base(which, base) != 0) + ret = -EIO; + break; +#endif + + case MSR_STAR: + case MSR_CSTAR: + case MSR_LSTAR: + case MSR_SYSCALL_MASK: + case MSR_IA32_SYSENTER_CS: + case MSR_IA32_SYSENTER_ESP: + case MSR_IA32_SYSENTER_EIP: + /* Fast syscall setup is all done in hypercalls, so + these are all ignored. Stub them out here to stop + Xen console noise. */ + + default: + ret = native_write_msr_safe(msr, low, high); + } + + return ret; +} + +void xen_setup_shared_info(void) +{ + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + set_fixmap(FIX_PARAVIRT_BOOTMAP, + xen_start_info->shared_info); + + HYPERVISOR_shared_info = + (struct shared_info *)fix_to_virt(FIX_PARAVIRT_BOOTMAP); + } else + HYPERVISOR_shared_info = + (struct shared_info *)__va(xen_start_info->shared_info); + +#ifndef CONFIG_SMP + /* In UP this is as good a place as any to set up shared info */ + xen_setup_vcpu_info_placement(); +#endif + + xen_setup_mfn_list_list(); +} + +/* This is called once we have the cpu_possible_mask */ +void xen_setup_vcpu_info_placement(void) +{ + int cpu; + + for_each_possible_cpu(cpu) + xen_vcpu_setup(cpu); + + /* xen_vcpu_setup managed to place the vcpu_info within the + * percpu area for all cpus, so make use of it. Note that for + * PVH we want to use native IRQ mechanism. */ + if (have_vcpu_info_placement && !xen_pvh_domain()) { + pv_irq_ops.save_fl = __PV_IS_CALLEE_SAVE(xen_save_fl_direct); + pv_irq_ops.restore_fl = __PV_IS_CALLEE_SAVE(xen_restore_fl_direct); + pv_irq_ops.irq_disable = __PV_IS_CALLEE_SAVE(xen_irq_disable_direct); + pv_irq_ops.irq_enable = __PV_IS_CALLEE_SAVE(xen_irq_enable_direct); + pv_mmu_ops.read_cr2 = xen_read_cr2_direct; + } +} + +static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf, + unsigned long addr, unsigned len) +{ + char *start, *end, *reloc; + unsigned ret; + + start = end = reloc = NULL; + +#define SITE(op, x) \ + case PARAVIRT_PATCH(op.x): \ + if (have_vcpu_info_placement) { \ + start = (char *)xen_##x##_direct; \ + end = xen_##x##_direct_end; \ + reloc = xen_##x##_direct_reloc; \ + } \ + goto patch_site + + switch (type) { + SITE(pv_irq_ops, irq_enable); + SITE(pv_irq_ops, irq_disable); + SITE(pv_irq_ops, save_fl); + SITE(pv_irq_ops, restore_fl); +#undef SITE + + patch_site: + if (start == NULL || (end-start) > len) + goto default_patch; + + ret = paravirt_patch_insns(insnbuf, len, start, end); + + /* Note: because reloc is assigned from something that + appears to be an array, gcc assumes it's non-null, + but doesn't know its relationship with start and + end. */ + if (reloc > start && reloc < end) { + int reloc_off = reloc - start; + long *relocp = (long *)(insnbuf + reloc_off); + long delta = start - (char *)addr; + + *relocp += delta; + } + break; + + default_patch: + default: + ret = paravirt_patch_default(type, clobbers, insnbuf, + addr, len); + break; + } + + return ret; +} + +static const struct pv_info xen_info __initconst = { + .paravirt_enabled = 1, + .shared_kernel_pmd = 0, + +#ifdef CONFIG_X86_64 + .extra_user_64bit_cs = FLAT_USER_CS64, +#endif + + .name = "Xen", +}; + +static const struct pv_init_ops xen_init_ops __initconst = { + .patch = xen_patch, +}; + +static const struct pv_cpu_ops xen_cpu_ops __initconst = { + .cpuid = xen_cpuid, + + .set_debugreg = xen_set_debugreg, + .get_debugreg = xen_get_debugreg, + + .clts = xen_clts, + + .read_cr0 = xen_read_cr0, + .write_cr0 = xen_write_cr0, + + .read_cr4 = native_read_cr4, + .read_cr4_safe = native_read_cr4_safe, + .write_cr4 = xen_write_cr4, + +#ifdef CONFIG_X86_64 + .read_cr8 = xen_read_cr8, + .write_cr8 = xen_write_cr8, +#endif + + .wbinvd = native_wbinvd, + + .read_msr = xen_read_msr_safe, + .write_msr = xen_write_msr_safe, + + .read_tsc = native_read_tsc, + .read_pmc = native_read_pmc, + + .read_tscp = native_read_tscp, + + .iret = xen_iret, + .irq_enable_sysexit = xen_sysexit, +#ifdef CONFIG_X86_64 + .usergs_sysret32 = xen_sysret32, + .usergs_sysret64 = xen_sysret64, +#endif + + .load_tr_desc = paravirt_nop, + .set_ldt = xen_set_ldt, + .load_gdt = xen_load_gdt, + .load_idt = xen_load_idt, + .load_tls = xen_load_tls, +#ifdef CONFIG_X86_64 + .load_gs_index = xen_load_gs_index, +#endif + + .alloc_ldt = xen_alloc_ldt, + .free_ldt = xen_free_ldt, + + .store_idt = native_store_idt, + .store_tr = xen_store_tr, + + .write_ldt_entry = xen_write_ldt_entry, + .write_gdt_entry = xen_write_gdt_entry, + .write_idt_entry = xen_write_idt_entry, + .load_sp0 = xen_load_sp0, + + .set_iopl_mask = xen_set_iopl_mask, + .io_delay = xen_io_delay, + + /* Xen takes care of %gs when switching to usermode for us */ + .swapgs = paravirt_nop, + + .start_context_switch = paravirt_start_context_switch, + .end_context_switch = xen_end_context_switch, +}; + +static const struct pv_apic_ops xen_apic_ops __initconst = { +#ifdef CONFIG_X86_LOCAL_APIC + .startup_ipi_hook = paravirt_nop, +#endif +}; + +static void xen_reboot(int reason) +{ + struct sched_shutdown r = { .reason = reason }; + + if (HYPERVISOR_sched_op(SCHEDOP_shutdown, &r)) + BUG(); +} + +static void xen_restart(char *msg) +{ + xen_reboot(SHUTDOWN_reboot); +} + +static void xen_emergency_restart(void) +{ + xen_reboot(SHUTDOWN_reboot); +} + +static void xen_machine_halt(void) +{ + xen_reboot(SHUTDOWN_poweroff); +} + +static void xen_machine_power_off(void) +{ + if (pm_power_off) + pm_power_off(); + xen_reboot(SHUTDOWN_poweroff); +} + +static void xen_crash_shutdown(struct pt_regs *regs) +{ + xen_reboot(SHUTDOWN_crash); +} + +static int +xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr) +{ + xen_reboot(SHUTDOWN_crash); + return NOTIFY_DONE; +} + +static struct notifier_block xen_panic_block = { + .notifier_call= xen_panic_event, + .priority = INT_MIN +}; + +int xen_panic_handler_init(void) +{ + atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block); + return 0; +} + +static const struct machine_ops xen_machine_ops __initconst = { + .restart = xen_restart, + .halt = xen_machine_halt, + .power_off = xen_machine_power_off, + .shutdown = xen_machine_halt, + .crash_shutdown = xen_crash_shutdown, + .emergency_restart = xen_emergency_restart, +}; + +static unsigned char xen_get_nmi_reason(void) +{ + unsigned char reason = 0; + + /* Construct a value which looks like it came from port 0x61. */ + if (test_bit(_XEN_NMIREASON_io_error, + &HYPERVISOR_shared_info->arch.nmi_reason)) + reason |= NMI_REASON_IOCHK; + if (test_bit(_XEN_NMIREASON_pci_serr, + &HYPERVISOR_shared_info->arch.nmi_reason)) + reason |= NMI_REASON_SERR; + + return reason; +} + +static void __init xen_boot_params_init_edd(void) +{ +#if IS_ENABLED(CONFIG_EDD) + struct xen_platform_op op; + struct edd_info *edd_info; + u32 *mbr_signature; + unsigned nr; + int ret; + + edd_info = boot_params.eddbuf; + mbr_signature = boot_params.edd_mbr_sig_buffer; + + op.cmd = XENPF_firmware_info; + + op.u.firmware_info.type = XEN_FW_DISK_INFO; + for (nr = 0; nr < EDDMAXNR; nr++) { + struct edd_info *info = edd_info + nr; + + op.u.firmware_info.index = nr; + info->params.length = sizeof(info->params); + set_xen_guest_handle(op.u.firmware_info.u.disk_info.edd_params, + &info->params); + ret = HYPERVISOR_dom0_op(&op); + if (ret) + break; + +#define C(x) info->x = op.u.firmware_info.u.disk_info.x + C(device); + C(version); + C(interface_support); + C(legacy_max_cylinder); + C(legacy_max_head); + C(legacy_sectors_per_track); +#undef C + } + boot_params.eddbuf_entries = nr; + + op.u.firmware_info.type = XEN_FW_DISK_MBR_SIGNATURE; + for (nr = 0; nr < EDD_MBR_SIG_MAX; nr++) { + op.u.firmware_info.index = nr; + ret = HYPERVISOR_dom0_op(&op); + if (ret) + break; + mbr_signature[nr] = op.u.firmware_info.u.disk_mbr_signature.mbr_signature; + } + boot_params.edd_mbr_sig_buf_entries = nr; +#endif +} + +/* + * Set up the GDT and segment registers for -fstack-protector. Until + * we do this, we have to be careful not to call any stack-protected + * function, which is most of the kernel. + * + * Note, that it is __ref because the only caller of this after init + * is PVH which is not going to use xen_load_gdt_boot or other + * __init functions. + */ +static void __ref xen_setup_gdt(int cpu) +{ + if (xen_feature(XENFEAT_auto_translated_physmap)) { +#ifdef CONFIG_X86_64 + unsigned long dummy; + + load_percpu_segment(cpu); /* We need to access per-cpu area */ + switch_to_new_gdt(cpu); /* GDT and GS set */ + + /* We are switching of the Xen provided GDT to our HVM mode + * GDT. The new GDT has __KERNEL_CS with CS.L = 1 + * and we are jumping to reload it. + */ + asm volatile ("pushq %0\n" + "leaq 1f(%%rip),%0\n" + "pushq %0\n" + "lretq\n" + "1:\n" + : "=&r" (dummy) : "0" (__KERNEL_CS)); + + /* + * While not needed, we also set the %es, %ds, and %fs + * to zero. We don't care about %ss as it is NULL. + * Strictly speaking this is not needed as Xen zeros those + * out (and also MSR_FS_BASE, MSR_GS_BASE, MSR_KERNEL_GS_BASE) + * + * Linux zeros them in cpu_init() and in secondary_startup_64 + * (for BSP). + */ + loadsegment(es, 0); + loadsegment(ds, 0); + loadsegment(fs, 0); +#else + /* PVH: TODO Implement. */ + BUG(); +#endif + return; /* PVH does not need any PV GDT ops. */ + } + pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry_boot; + pv_cpu_ops.load_gdt = xen_load_gdt_boot; + + setup_stack_canary_segment(0); + switch_to_new_gdt(0); + + pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry; + pv_cpu_ops.load_gdt = xen_load_gdt; +} + +#ifdef CONFIG_XEN_PVH +/* + * A PV guest starts with default flags that are not set for PVH, set them + * here asap. + */ +static void xen_pvh_set_cr_flags(int cpu) +{ + + /* Some of these are setup in 'secondary_startup_64'. The others: + * X86_CR0_TS, X86_CR0_PE, X86_CR0_ET are set by Xen for HVM guests + * (which PVH shared codepaths), while X86_CR0_PG is for PVH. */ + write_cr0(read_cr0() | X86_CR0_MP | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM); + + if (!cpu) + return; + /* + * For BSP, PSE PGE are set in probe_page_size_mask(), for APs + * set them here. For all, OSFXSR OSXMMEXCPT are set in fpu_init. + */ + if (cpu_has_pse) + cr4_set_bits_and_update_boot(X86_CR4_PSE); + + if (cpu_has_pge) + cr4_set_bits_and_update_boot(X86_CR4_PGE); +} + +/* + * Note, that it is ref - because the only caller of this after init + * is PVH which is not going to use xen_load_gdt_boot or other + * __init functions. + */ +void __ref xen_pvh_secondary_vcpu_init(int cpu) +{ + xen_setup_gdt(cpu); + xen_pvh_set_cr_flags(cpu); +} + +static void __init xen_pvh_early_guest_init(void) +{ + if (!xen_feature(XENFEAT_auto_translated_physmap)) + return; + + if (!xen_feature(XENFEAT_hvm_callback_vector)) + return; + + xen_have_vector_callback = 1; + + xen_pvh_early_cpu_init(0, false); + xen_pvh_set_cr_flags(0); + +#ifdef CONFIG_X86_32 + BUG(); /* PVH: Implement proper support. */ +#endif +} +#endif /* CONFIG_XEN_PVH */ + +/* First C function to be called on Xen boot */ +asmlinkage __visible void __init xen_start_kernel(void) +{ + struct physdev_set_iopl set_iopl; + unsigned long initrd_start = 0; + int rc; + + if (!xen_start_info) + return; + + xen_domain_type = XEN_PV_DOMAIN; + + xen_setup_features(); +#ifdef CONFIG_XEN_PVH + xen_pvh_early_guest_init(); +#endif + xen_setup_machphys_mapping(); + + /* Install Xen paravirt ops */ + pv_info = xen_info; + pv_init_ops = xen_init_ops; + pv_apic_ops = xen_apic_ops; + if (!xen_pvh_domain()) { + pv_cpu_ops = xen_cpu_ops; + + x86_platform.get_nmi_reason = xen_get_nmi_reason; + } + + if (xen_feature(XENFEAT_auto_translated_physmap)) + x86_init.resources.memory_setup = xen_auto_xlated_memory_setup; + else + x86_init.resources.memory_setup = xen_memory_setup; + x86_init.oem.arch_setup = xen_arch_setup; + x86_init.oem.banner = xen_banner; + + xen_init_time_ops(); + + /* + * Set up some pagetable state before starting to set any ptes. + */ + + xen_init_mmu_ops(); + + /* Prevent unwanted bits from being set in PTEs. */ + __supported_pte_mask &= ~_PAGE_GLOBAL; + + /* + * Prevent page tables from being allocated in highmem, even + * if CONFIG_HIGHPTE is enabled. + */ + __userpte_alloc_gfp &= ~__GFP_HIGHMEM; + + /* Work out if we support NX */ + x86_configure_nx(); + + /* Get mfn list */ + xen_build_dynamic_phys_to_machine(); + + /* + * Set up kernel GDT and segment registers, mainly so that + * -fstack-protector code can be executed. + */ + xen_setup_gdt(0); + + xen_init_irq_ops(); + xen_init_cpuid_mask(); + +#ifdef CONFIG_X86_LOCAL_APIC + /* + * set up the basic apic ops. + */ + xen_init_apic(); +#endif + + if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) { + pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start; + pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit; + } + + machine_ops = xen_machine_ops; + + /* + * The only reliable way to retain the initial address of the + * percpu gdt_page is to remember it here, so we can go and + * mark it RW later, when the initial percpu area is freed. + */ + xen_initial_gdt = &per_cpu(gdt_page, 0); + + xen_smp_init(); + +#ifdef CONFIG_ACPI_NUMA + /* + * The pages we from Xen are not related to machine pages, so + * any NUMA information the kernel tries to get from ACPI will + * be meaningless. Prevent it from trying. + */ + acpi_numa = -1; +#endif + /* Don't do the full vcpu_info placement stuff until we have a + possible map and a non-dummy shared_info. */ + per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; + + local_irq_disable(); + early_boot_irqs_disabled = true; + + xen_raw_console_write("mapping kernel into physical memory\n"); + xen_setup_kernel_pagetable((pgd_t *)xen_start_info->pt_base, xen_start_info->nr_pages); + + /* + * Modify the cache mode translation tables to match Xen's PAT + * configuration. + */ + + pat_init_cache_modes(); + + /* keep using Xen gdt for now; no urgent need to change it */ + +#ifdef CONFIG_X86_32 + pv_info.kernel_rpl = 1; + if (xen_feature(XENFEAT_supervisor_mode_kernel)) + pv_info.kernel_rpl = 0; +#else + pv_info.kernel_rpl = 0; +#endif + /* set the limit of our address space */ + xen_reserve_top(); + + /* PVH: runs at default kernel iopl of 0 */ + if (!xen_pvh_domain()) { + /* + * We used to do this in xen_arch_setup, but that is too late + * on AMD were early_cpu_init (run before ->arch_setup()) calls + * early_amd_init which pokes 0xcf8 port. + */ + set_iopl.iopl = 1; + rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl); + if (rc != 0) + xen_raw_printk("physdev_op failed %d\n", rc); + } + +#ifdef CONFIG_X86_32 + /* set up basic CPUID stuff */ + cpu_detect(&new_cpu_data); + set_cpu_cap(&new_cpu_data, X86_FEATURE_FPU); + new_cpu_data.wp_works_ok = 1; + new_cpu_data.x86_capability[0] = cpuid_edx(1); +#endif + + if (xen_start_info->mod_start) { + if (xen_start_info->flags & SIF_MOD_START_PFN) + initrd_start = PFN_PHYS(xen_start_info->mod_start); + else + initrd_start = __pa(xen_start_info->mod_start); + } + + /* Poke various useful things into boot_params */ + boot_params.hdr.type_of_loader = (9 << 4) | 0; + boot_params.hdr.ramdisk_image = initrd_start; + boot_params.hdr.ramdisk_size = xen_start_info->mod_len; + boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line); + + if (!xen_initial_domain()) { + add_preferred_console("xenboot", 0, NULL); + add_preferred_console("tty", 0, NULL); + add_preferred_console("hvc", 0, NULL); + if (pci_xen) + x86_init.pci.arch_init = pci_xen_init; + } else { + const struct dom0_vga_console_info *info = + (void *)((char *)xen_start_info + + xen_start_info->console.dom0.info_off); + struct xen_platform_op op = { + .cmd = XENPF_firmware_info, + .interface_version = XENPF_INTERFACE_VERSION, + .u.firmware_info.type = XEN_FW_KBD_SHIFT_FLAGS, + }; + + xen_init_vga(info, xen_start_info->console.dom0.info_size); + xen_start_info->console.domU.mfn = 0; + xen_start_info->console.domU.evtchn = 0; + + if (HYPERVISOR_dom0_op(&op) == 0) + boot_params.kbd_status = op.u.firmware_info.u.kbd_shift_flags; + + /* Make sure ACS will be enabled */ + pci_request_acs(); + + xen_acpi_sleep_register(); + + /* Avoid searching for BIOS MP tables */ + x86_init.mpparse.find_smp_config = x86_init_noop; + x86_init.mpparse.get_smp_config = x86_init_uint_noop; + + xen_boot_params_init_edd(); + } +#ifdef CONFIG_PCI + /* PCI BIOS service won't work from a PV guest. */ + pci_probe &= ~PCI_PROBE_BIOS; +#endif + xen_raw_console_write("about to get started...\n"); + + xen_setup_runstate_info(0); + + xen_efi_init(); + + /* Start the world */ +#ifdef CONFIG_X86_32 + i386_start_kernel(); +#else + cr4_init_shadow(); /* 32b kernel does this in i386_start_kernel() */ + x86_64_start_reservations((char *)__pa_symbol(&boot_params)); +#endif +} + +void __ref xen_hvm_init_shared_info(void) +{ + int cpu; + struct xen_add_to_physmap xatp; + static struct shared_info *shared_info_page = 0; + + if (!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; + + /* xen_vcpu is a pointer to the vcpu_info struct in the shared_info + * page, we use it in the event channel upcall and in some pvclock + * related functions. We don't need the vcpu_info placement + * optimizations because we don't use any pv_mmu or pv_irq op on + * HVM. + * When xen_hvm_init_shared_info is run at boot time only vcpu 0 is + * online but xen_hvm_init_shared_info is run at resume time too and + * in that case multiple vcpus might be online. */ + for_each_online_cpu(cpu) { + /* Leave it to be NULL. */ + if (cpu >= MAX_VIRT_CPUS) + continue; + per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu]; + } +} + +#ifdef CONFIG_XEN_PVHVM +static void __init init_hvm_pv_info(void) +{ + int major, 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.name = "Xen HVM"; + + xen_domain_type = XEN_HVM_DOMAIN; +} + +static int xen_hvm_cpu_notify(struct notifier_block *self, unsigned long action, + void *hcpu) +{ + int cpu = (long)hcpu; + switch (action) { + case CPU_UP_PREPARE: + xen_vcpu_setup(cpu); + if (xen_have_vector_callback) { + if (xen_feature(XENFEAT_hvm_safe_pvclock)) + xen_setup_timer(cpu); + } + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block xen_hvm_cpu_notifier = { + .notifier_call = xen_hvm_cpu_notify, +}; + +static void __init xen_hvm_guest_init(void) +{ + if (xen_pv_domain()) + return; + + init_hvm_pv_info(); + + xen_hvm_init_shared_info(); + + xen_panic_handler_init(); + + if (xen_feature(XENFEAT_hvm_callback_vector)) + xen_have_vector_callback = 1; + xen_hvm_smp_init(); + register_cpu_notifier(&xen_hvm_cpu_notifier); + xen_unplug_emulated_devices(); + x86_init.irqs.intr_init = xen_init_IRQ; + xen_hvm_init_time_ops(); + xen_hvm_init_mmu_ops(); +} +#endif + +static bool xen_nopv = false; +static __init int xen_parse_nopv(char *arg) +{ + xen_nopv = true; + return 0; +} +early_param("xen_nopv", xen_parse_nopv); + +static uint32_t __init xen_platform(void) +{ + if (xen_nopv) + return 0; + + return xen_cpuid_base(); +} + +bool xen_hvm_need_lapic(void) +{ + if (xen_nopv) + return false; + if (xen_pv_domain()) + return false; + if (!xen_hvm_domain()) + return false; + if (xen_feature(XENFEAT_hvm_pirqs) && xen_have_vector_callback) + return false; + return true; +} +EXPORT_SYMBOL_GPL(xen_hvm_need_lapic); + +static void xen_set_cpu_features(struct cpuinfo_x86 *c) +{ + if (xen_pv_domain()) + clear_cpu_bug(c, X86_BUG_SYSRET_SS_ATTRS); +} + +const struct hypervisor_x86 x86_hyper_xen = { + .name = "Xen", + .detect = xen_platform, +#ifdef CONFIG_XEN_PVHVM + .init_platform = xen_hvm_guest_init, +#endif + .x2apic_available = xen_x2apic_para_available, + .set_cpu_features = xen_set_cpu_features, +}; +EXPORT_SYMBOL(x86_hyper_xen); diff --git a/kernel/arch/x86/xen/grant-table.c b/kernel/arch/x86/xen/grant-table.c new file mode 100644 index 000000000..1580e7a5a --- /dev/null +++ b/kernel/arch/x86/xen/grant-table.c @@ -0,0 +1,175 @@ +/****************************************************************************** + * grant_table.c + * x86 specific part + * + * Granting foreign access to our memory reservation. + * + * Copyright (c) 2005-2006, Christopher Clark + * Copyright (c) 2004-2005, K A Fraser + * Copyright (c) 2008 Isaku Yamahata <yamahata at valinux co jp> + * VA Linux Systems Japan. Split out x86 specific part. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (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. + */ + +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> + +#include <xen/interface/xen.h> +#include <xen/page.h> +#include <xen/grant_table.h> +#include <xen/xen.h> + +#include <asm/pgtable.h> + +static struct gnttab_vm_area { + struct vm_struct *area; + pte_t **ptes; +} gnttab_shared_vm_area; + +int arch_gnttab_map_shared(unsigned long *frames, unsigned long nr_gframes, + unsigned long max_nr_gframes, + void **__shared) +{ + void *shared = *__shared; + unsigned long addr; + unsigned long i; + + if (shared == NULL) + *__shared = shared = gnttab_shared_vm_area.area->addr; + + addr = (unsigned long)shared; + + for (i = 0; i < nr_gframes; i++) { + set_pte_at(&init_mm, addr, gnttab_shared_vm_area.ptes[i], + mfn_pte(frames[i], PAGE_KERNEL)); + addr += PAGE_SIZE; + } + + return 0; +} + +void arch_gnttab_unmap(void *shared, unsigned long nr_gframes) +{ + unsigned long addr; + unsigned long i; + + addr = (unsigned long)shared; + + for (i = 0; i < nr_gframes; i++) { + set_pte_at(&init_mm, addr, gnttab_shared_vm_area.ptes[i], + __pte(0)); + addr += PAGE_SIZE; + } +} + +static int arch_gnttab_valloc(struct gnttab_vm_area *area, unsigned nr_frames) +{ + area->ptes = kmalloc(sizeof(pte_t *) * nr_frames, GFP_KERNEL); + if (area->ptes == NULL) + return -ENOMEM; + + area->area = alloc_vm_area(PAGE_SIZE * nr_frames, area->ptes); + if (area->area == NULL) { + kfree(area->ptes); + return -ENOMEM; + } + + return 0; +} + +int arch_gnttab_init(unsigned long nr_shared) +{ + if (!xen_pv_domain()) + return 0; + + return arch_gnttab_valloc(&gnttab_shared_vm_area, nr_shared); +} + +#ifdef CONFIG_XEN_PVH +#include <xen/balloon.h> +#include <xen/events.h> +#include <linux/slab.h> +static int __init xlated_setup_gnttab_pages(void) +{ + struct page **pages; + xen_pfn_t *pfns; + void *vaddr; + int rc; + unsigned int i; + unsigned long nr_grant_frames = gnttab_max_grant_frames(); + + BUG_ON(nr_grant_frames == 0); + pages = kcalloc(nr_grant_frames, sizeof(pages[0]), GFP_KERNEL); + if (!pages) + return -ENOMEM; + + pfns = kcalloc(nr_grant_frames, sizeof(pfns[0]), GFP_KERNEL); + if (!pfns) { + kfree(pages); + return -ENOMEM; + } + rc = alloc_xenballooned_pages(nr_grant_frames, pages, 0 /* lowmem */); + if (rc) { + pr_warn("%s Couldn't balloon alloc %ld pfns rc:%d\n", __func__, + nr_grant_frames, rc); + kfree(pages); + kfree(pfns); + return rc; + } + for (i = 0; i < nr_grant_frames; i++) + pfns[i] = page_to_pfn(pages[i]); + + vaddr = vmap(pages, nr_grant_frames, 0, PAGE_KERNEL); + if (!vaddr) { + pr_warn("%s Couldn't map %ld pfns rc:%d\n", __func__, + nr_grant_frames, rc); + free_xenballooned_pages(nr_grant_frames, pages); + kfree(pages); + kfree(pfns); + return -ENOMEM; + } + kfree(pages); + + xen_auto_xlat_grant_frames.pfn = pfns; + xen_auto_xlat_grant_frames.count = nr_grant_frames; + xen_auto_xlat_grant_frames.vaddr = vaddr; + + return 0; +} + +static int __init xen_pvh_gnttab_setup(void) +{ + if (!xen_pvh_domain()) + return -ENODEV; + + return xlated_setup_gnttab_pages(); +} +/* Call it _before_ __gnttab_init as we need to initialize the + * xen_auto_xlat_grant_frames first. */ +core_initcall(xen_pvh_gnttab_setup); +#endif diff --git a/kernel/arch/x86/xen/irq.c b/kernel/arch/x86/xen/irq.c new file mode 100644 index 000000000..a1207cb64 --- /dev/null +++ b/kernel/arch/x86/xen/irq.c @@ -0,0 +1,136 @@ +#include <linux/hardirq.h> + +#include <asm/x86_init.h> + +#include <xen/interface/xen.h> +#include <xen/interface/sched.h> +#include <xen/interface/vcpu.h> +#include <xen/features.h> +#include <xen/events.h> + +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> + +#include "xen-ops.h" + +/* + * Force a proper event-channel callback from Xen after clearing the + * callback mask. We do this in a very simple manner, by making a call + * down into Xen. The pending flag will be checked by Xen on return. + */ +void xen_force_evtchn_callback(void) +{ + (void)HYPERVISOR_xen_version(0, NULL); +} + +asmlinkage __visible unsigned long xen_save_fl(void) +{ + struct vcpu_info *vcpu; + unsigned long flags; + + vcpu = this_cpu_read(xen_vcpu); + + /* flag has opposite sense of mask */ + flags = !vcpu->evtchn_upcall_mask; + + /* convert to IF type flag + -0 -> 0x00000000 + -1 -> 0xffffffff + */ + return (-flags) & X86_EFLAGS_IF; +} +PV_CALLEE_SAVE_REGS_THUNK(xen_save_fl); + +__visible void xen_restore_fl(unsigned long flags) +{ + struct vcpu_info *vcpu; + + /* convert from IF type flag */ + flags = !(flags & X86_EFLAGS_IF); + + /* See xen_irq_enable() for why preemption must be disabled. */ + preempt_disable(); + vcpu = this_cpu_read(xen_vcpu); + vcpu->evtchn_upcall_mask = flags; + + if (flags == 0) { + barrier(); /* unmask then check (avoid races) */ + if (unlikely(vcpu->evtchn_upcall_pending)) + xen_force_evtchn_callback(); + preempt_enable(); + } else + preempt_enable_no_resched(); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_restore_fl); + +asmlinkage __visible void xen_irq_disable(void) +{ + /* There's a one instruction preempt window here. We need to + make sure we're don't switch CPUs between getting the vcpu + pointer and updating the mask. */ + preempt_disable(); + this_cpu_read(xen_vcpu)->evtchn_upcall_mask = 1; + preempt_enable_no_resched(); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_irq_disable); + +asmlinkage __visible void xen_irq_enable(void) +{ + struct vcpu_info *vcpu; + + /* + * We may be preempted as soon as vcpu->evtchn_upcall_mask is + * cleared, so disable preemption to ensure we check for + * events on the VCPU we are still running on. + */ + preempt_disable(); + + vcpu = this_cpu_read(xen_vcpu); + vcpu->evtchn_upcall_mask = 0; + + /* Doesn't matter if we get preempted here, because any + pending event will get dealt with anyway. */ + + barrier(); /* unmask then check (avoid races) */ + if (unlikely(vcpu->evtchn_upcall_pending)) + xen_force_evtchn_callback(); + + preempt_enable(); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_irq_enable); + +static void xen_safe_halt(void) +{ + /* Blocking includes an implicit local_irq_enable(). */ + if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0) + BUG(); +} + +static void xen_halt(void) +{ + if (irqs_disabled()) + HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL); + else + xen_safe_halt(); +} + +static const struct pv_irq_ops xen_irq_ops __initconst = { + .save_fl = PV_CALLEE_SAVE(xen_save_fl), + .restore_fl = PV_CALLEE_SAVE(xen_restore_fl), + .irq_disable = PV_CALLEE_SAVE(xen_irq_disable), + .irq_enable = PV_CALLEE_SAVE(xen_irq_enable), + + .safe_halt = xen_safe_halt, + .halt = xen_halt, +#ifdef CONFIG_X86_64 + .adjust_exception_frame = xen_adjust_exception_frame, +#endif +}; + +void __init xen_init_irq_ops(void) +{ + /* For PVH we use default pv_irq_ops settings. */ + if (!xen_feature(XENFEAT_hvm_callback_vector)) + pv_irq_ops = xen_irq_ops; + x86_init.irqs.intr_init = xen_init_IRQ; +} diff --git a/kernel/arch/x86/xen/mmu.c b/kernel/arch/x86/xen/mmu.c new file mode 100644 index 000000000..dd151b204 --- /dev/null +++ b/kernel/arch/x86/xen/mmu.c @@ -0,0 +1,2591 @@ +/* + * Xen mmu operations + * + * This file contains the various mmu fetch and update operations. + * The most important job they must perform is the mapping between the + * domain's pfn and the overall machine mfns. + * + * Xen allows guests to directly update the pagetable, in a controlled + * fashion. In other words, the guest modifies the same pagetable + * that the CPU actually uses, which eliminates the overhead of having + * a separate shadow pagetable. + * + * In order to allow this, it falls on the guest domain to map its + * notion of a "physical" pfn - which is just a domain-local linear + * address - into a real "machine address" which the CPU's MMU can + * use. + * + * A pgd_t/pmd_t/pte_t will typically contain an mfn, and so can be + * inserted directly into the pagetable. When creating a new + * pte/pmd/pgd, it converts the passed pfn into an mfn. Conversely, + * when reading the content back with __(pgd|pmd|pte)_val, it converts + * the mfn back into a pfn. + * + * The other constraint is that all pages which make up a pagetable + * must be mapped read-only in the guest. This prevents uncontrolled + * guest updates to the pagetable. Xen strictly enforces this, and + * will disallow any pagetable update which will end up mapping a + * pagetable page RW, and will disallow using any writable page as a + * pagetable. + * + * Naively, when loading %cr3 with the base of a new pagetable, Xen + * would need to validate the whole pagetable before going on. + * Naturally, this is quite slow. The solution is to "pin" a + * pagetable, which enforces all the constraints on the pagetable even + * when it is not actively in use. This menas that Xen can be assured + * that it is still valid when you do load it into %cr3, and doesn't + * need to revalidate it. + * + * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007 + */ +#include <linux/sched.h> +#include <linux/highmem.h> +#include <linux/debugfs.h> +#include <linux/bug.h> +#include <linux/vmalloc.h> +#include <linux/module.h> +#include <linux/gfp.h> +#include <linux/memblock.h> +#include <linux/seq_file.h> +#include <linux/crash_dump.h> + +#include <trace/events/xen.h> + +#include <asm/pgtable.h> +#include <asm/tlbflush.h> +#include <asm/fixmap.h> +#include <asm/mmu_context.h> +#include <asm/setup.h> +#include <asm/paravirt.h> +#include <asm/e820.h> +#include <asm/linkage.h> +#include <asm/page.h> +#include <asm/init.h> +#include <asm/pat.h> +#include <asm/smp.h> + +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> + +#include <xen/xen.h> +#include <xen/page.h> +#include <xen/interface/xen.h> +#include <xen/interface/hvm/hvm_op.h> +#include <xen/interface/version.h> +#include <xen/interface/memory.h> +#include <xen/hvc-console.h> + +#include "multicalls.h" +#include "mmu.h" +#include "debugfs.h" + +/* + * Protects atomic reservation decrease/increase against concurrent increases. + * Also protects non-atomic updates of current_pages and balloon lists. + */ +DEFINE_SPINLOCK(xen_reservation_lock); + +#ifdef CONFIG_X86_32 +/* + * Identity map, in addition to plain kernel map. This needs to be + * large enough to allocate page table pages to allocate the rest. + * Each page can map 2MB. + */ +#define LEVEL1_IDENT_ENTRIES (PTRS_PER_PTE * 4) +static RESERVE_BRK_ARRAY(pte_t, level1_ident_pgt, LEVEL1_IDENT_ENTRIES); +#endif +#ifdef CONFIG_X86_64 +/* l3 pud for userspace vsyscall mapping */ +static pud_t level3_user_vsyscall[PTRS_PER_PUD] __page_aligned_bss; +#endif /* CONFIG_X86_64 */ + +/* + * Note about cr3 (pagetable base) values: + * + * xen_cr3 contains the current logical cr3 value; it contains the + * last set cr3. This may not be the current effective cr3, because + * its update may be being lazily deferred. However, a vcpu looking + * at its own cr3 can use this value knowing that it everything will + * be self-consistent. + * + * xen_current_cr3 contains the actual vcpu cr3; it is set once the + * hypercall to set the vcpu cr3 is complete (so it may be a little + * out of date, but it will never be set early). If one vcpu is + * looking at another vcpu's cr3 value, it should use this variable. + */ +DEFINE_PER_CPU(unsigned long, xen_cr3); /* cr3 stored as physaddr */ +DEFINE_PER_CPU(unsigned long, xen_current_cr3); /* actual vcpu cr3 */ + + +/* + * Just beyond the highest usermode address. STACK_TOP_MAX has a + * redzone above it, so round it up to a PGD boundary. + */ +#define USER_LIMIT ((STACK_TOP_MAX + PGDIR_SIZE - 1) & PGDIR_MASK) + +unsigned long arbitrary_virt_to_mfn(void *vaddr) +{ + xmaddr_t maddr = arbitrary_virt_to_machine(vaddr); + + return PFN_DOWN(maddr.maddr); +} + +xmaddr_t arbitrary_virt_to_machine(void *vaddr) +{ + unsigned long address = (unsigned long)vaddr; + unsigned int level; + pte_t *pte; + unsigned offset; + + /* + * if the PFN is in the linear mapped vaddr range, we can just use + * the (quick) virt_to_machine() p2m lookup + */ + if (virt_addr_valid(vaddr)) + return virt_to_machine(vaddr); + + /* otherwise we have to do a (slower) full page-table walk */ + + pte = lookup_address(address, &level); + BUG_ON(pte == NULL); + offset = address & ~PAGE_MASK; + return XMADDR(((phys_addr_t)pte_mfn(*pte) << PAGE_SHIFT) + offset); +} +EXPORT_SYMBOL_GPL(arbitrary_virt_to_machine); + +void make_lowmem_page_readonly(void *vaddr) +{ + pte_t *pte, ptev; + unsigned long address = (unsigned long)vaddr; + unsigned int level; + + pte = lookup_address(address, &level); + if (pte == NULL) + return; /* vaddr missing */ + + ptev = pte_wrprotect(*pte); + + if (HYPERVISOR_update_va_mapping(address, ptev, 0)) + BUG(); +} + +void make_lowmem_page_readwrite(void *vaddr) +{ + pte_t *pte, ptev; + unsigned long address = (unsigned long)vaddr; + unsigned int level; + + pte = lookup_address(address, &level); + if (pte == NULL) + return; /* vaddr missing */ + + ptev = pte_mkwrite(*pte); + + if (HYPERVISOR_update_va_mapping(address, ptev, 0)) + BUG(); +} + + +static bool xen_page_pinned(void *ptr) +{ + struct page *page = virt_to_page(ptr); + + return PagePinned(page); +} + +void xen_set_domain_pte(pte_t *ptep, pte_t pteval, unsigned domid) +{ + struct multicall_space mcs; + struct mmu_update *u; + + trace_xen_mmu_set_domain_pte(ptep, pteval, domid); + + mcs = xen_mc_entry(sizeof(*u)); + u = mcs.args; + + /* ptep might be kmapped when using 32-bit HIGHPTE */ + u->ptr = virt_to_machine(ptep).maddr; + u->val = pte_val_ma(pteval); + + MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, domid); + + xen_mc_issue(PARAVIRT_LAZY_MMU); +} +EXPORT_SYMBOL_GPL(xen_set_domain_pte); + +static void xen_extend_mmu_update(const struct mmu_update *update) +{ + struct multicall_space mcs; + struct mmu_update *u; + + mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u)); + + if (mcs.mc != NULL) { + mcs.mc->args[1]++; + } else { + mcs = __xen_mc_entry(sizeof(*u)); + MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF); + } + + u = mcs.args; + *u = *update; +} + +static void xen_extend_mmuext_op(const struct mmuext_op *op) +{ + struct multicall_space mcs; + struct mmuext_op *u; + + mcs = xen_mc_extend_args(__HYPERVISOR_mmuext_op, sizeof(*u)); + + if (mcs.mc != NULL) { + mcs.mc->args[1]++; + } else { + mcs = __xen_mc_entry(sizeof(*u)); + MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF); + } + + u = mcs.args; + *u = *op; +} + +static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val) +{ + struct mmu_update u; + + preempt_disable(); + + xen_mc_batch(); + + /* ptr may be ioremapped for 64-bit pagetable setup */ + u.ptr = arbitrary_virt_to_machine(ptr).maddr; + u.val = pmd_val_ma(val); + xen_extend_mmu_update(&u); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} + +static void xen_set_pmd(pmd_t *ptr, pmd_t val) +{ + trace_xen_mmu_set_pmd(ptr, val); + + /* If page is not pinned, we can just update the entry + directly */ + if (!xen_page_pinned(ptr)) { + *ptr = val; + return; + } + + xen_set_pmd_hyper(ptr, val); +} + +/* + * Associate a virtual page frame with a given physical page frame + * and protection flags for that frame. + */ +void set_pte_mfn(unsigned long vaddr, unsigned long mfn, pgprot_t flags) +{ + set_pte_vaddr(vaddr, mfn_pte(mfn, flags)); +} + +static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval) +{ + struct mmu_update u; + + if (paravirt_get_lazy_mode() != PARAVIRT_LAZY_MMU) + return false; + + xen_mc_batch(); + + u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE; + u.val = pte_val_ma(pteval); + xen_extend_mmu_update(&u); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + return true; +} + +static inline void __xen_set_pte(pte_t *ptep, pte_t pteval) +{ + if (!xen_batched_set_pte(ptep, pteval)) { + /* + * Could call native_set_pte() here and trap and + * emulate the PTE write but with 32-bit guests this + * needs two traps (one for each of the two 32-bit + * words in the PTE) so do one hypercall directly + * instead. + */ + struct mmu_update u; + + u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE; + u.val = pte_val_ma(pteval); + HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF); + } +} + +static void xen_set_pte(pte_t *ptep, pte_t pteval) +{ + trace_xen_mmu_set_pte(ptep, pteval); + __xen_set_pte(ptep, pteval); +} + +static void xen_set_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pteval) +{ + trace_xen_mmu_set_pte_at(mm, addr, ptep, pteval); + __xen_set_pte(ptep, pteval); +} + +pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + /* Just return the pte as-is. We preserve the bits on commit */ + trace_xen_mmu_ptep_modify_prot_start(mm, addr, ptep, *ptep); + return *ptep; +} + +void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte) +{ + struct mmu_update u; + + trace_xen_mmu_ptep_modify_prot_commit(mm, addr, ptep, pte); + xen_mc_batch(); + + u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD; + u.val = pte_val_ma(pte); + xen_extend_mmu_update(&u); + + xen_mc_issue(PARAVIRT_LAZY_MMU); +} + +/* Assume pteval_t is equivalent to all the other *val_t types. */ +static pteval_t pte_mfn_to_pfn(pteval_t val) +{ + if (val & _PAGE_PRESENT) { + unsigned long mfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT; + unsigned long pfn = mfn_to_pfn(mfn); + + pteval_t flags = val & PTE_FLAGS_MASK; + if (unlikely(pfn == ~0)) + val = flags & ~_PAGE_PRESENT; + else + val = ((pteval_t)pfn << PAGE_SHIFT) | flags; + } + + return val; +} + +static pteval_t pte_pfn_to_mfn(pteval_t val) +{ + if (val & _PAGE_PRESENT) { + unsigned long pfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT; + pteval_t flags = val & PTE_FLAGS_MASK; + unsigned long mfn; + + if (!xen_feature(XENFEAT_auto_translated_physmap)) + mfn = __pfn_to_mfn(pfn); + else + mfn = pfn; + /* + * If there's no mfn for the pfn, then just create an + * empty non-present pte. Unfortunately this loses + * information about the original pfn, so + * pte_mfn_to_pfn is asymmetric. + */ + if (unlikely(mfn == INVALID_P2M_ENTRY)) { + mfn = 0; + flags = 0; + } else + mfn &= ~(FOREIGN_FRAME_BIT | IDENTITY_FRAME_BIT); + val = ((pteval_t)mfn << PAGE_SHIFT) | flags; + } + + return val; +} + +__visible pteval_t xen_pte_val(pte_t pte) +{ + pteval_t pteval = pte.pte; + + return pte_mfn_to_pfn(pteval); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_pte_val); + +__visible pgdval_t xen_pgd_val(pgd_t pgd) +{ + return pte_mfn_to_pfn(pgd.pgd); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_pgd_val); + +__visible pte_t xen_make_pte(pteval_t pte) +{ + pte = pte_pfn_to_mfn(pte); + + return native_make_pte(pte); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte); + +__visible pgd_t xen_make_pgd(pgdval_t pgd) +{ + pgd = pte_pfn_to_mfn(pgd); + return native_make_pgd(pgd); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_make_pgd); + +__visible pmdval_t xen_pmd_val(pmd_t pmd) +{ + return pte_mfn_to_pfn(pmd.pmd); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_pmd_val); + +static void xen_set_pud_hyper(pud_t *ptr, pud_t val) +{ + struct mmu_update u; + + preempt_disable(); + + xen_mc_batch(); + + /* ptr may be ioremapped for 64-bit pagetable setup */ + u.ptr = arbitrary_virt_to_machine(ptr).maddr; + u.val = pud_val_ma(val); + xen_extend_mmu_update(&u); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} + +static void xen_set_pud(pud_t *ptr, pud_t val) +{ + trace_xen_mmu_set_pud(ptr, val); + + /* If page is not pinned, we can just update the entry + directly */ + if (!xen_page_pinned(ptr)) { + *ptr = val; + return; + } + + xen_set_pud_hyper(ptr, val); +} + +#ifdef CONFIG_X86_PAE +static void xen_set_pte_atomic(pte_t *ptep, pte_t pte) +{ + trace_xen_mmu_set_pte_atomic(ptep, pte); + set_64bit((u64 *)ptep, native_pte_val(pte)); +} + +static void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +{ + trace_xen_mmu_pte_clear(mm, addr, ptep); + if (!xen_batched_set_pte(ptep, native_make_pte(0))) + native_pte_clear(mm, addr, ptep); +} + +static void xen_pmd_clear(pmd_t *pmdp) +{ + trace_xen_mmu_pmd_clear(pmdp); + set_pmd(pmdp, __pmd(0)); +} +#endif /* CONFIG_X86_PAE */ + +__visible pmd_t xen_make_pmd(pmdval_t pmd) +{ + pmd = pte_pfn_to_mfn(pmd); + return native_make_pmd(pmd); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_make_pmd); + +#if CONFIG_PGTABLE_LEVELS == 4 +__visible pudval_t xen_pud_val(pud_t pud) +{ + return pte_mfn_to_pfn(pud.pud); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_pud_val); + +__visible pud_t xen_make_pud(pudval_t pud) +{ + pud = pte_pfn_to_mfn(pud); + + return native_make_pud(pud); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_make_pud); + +static pgd_t *xen_get_user_pgd(pgd_t *pgd) +{ + pgd_t *pgd_page = (pgd_t *)(((unsigned long)pgd) & PAGE_MASK); + unsigned offset = pgd - pgd_page; + pgd_t *user_ptr = NULL; + + if (offset < pgd_index(USER_LIMIT)) { + struct page *page = virt_to_page(pgd_page); + user_ptr = (pgd_t *)page->private; + if (user_ptr) + user_ptr += offset; + } + + return user_ptr; +} + +static void __xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) +{ + struct mmu_update u; + + u.ptr = virt_to_machine(ptr).maddr; + u.val = pgd_val_ma(val); + xen_extend_mmu_update(&u); +} + +/* + * Raw hypercall-based set_pgd, intended for in early boot before + * there's a page structure. This implies: + * 1. The only existing pagetable is the kernel's + * 2. It is always pinned + * 3. It has no user pagetable attached to it + */ +static void __init xen_set_pgd_hyper(pgd_t *ptr, pgd_t val) +{ + preempt_disable(); + + xen_mc_batch(); + + __xen_set_pgd_hyper(ptr, val); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} + +static void xen_set_pgd(pgd_t *ptr, pgd_t val) +{ + pgd_t *user_ptr = xen_get_user_pgd(ptr); + + trace_xen_mmu_set_pgd(ptr, user_ptr, val); + + /* If page is not pinned, we can just update the entry + directly */ + if (!xen_page_pinned(ptr)) { + *ptr = val; + if (user_ptr) { + WARN_ON(xen_page_pinned(user_ptr)); + *user_ptr = val; + } + return; + } + + /* If it's pinned, then we can at least batch the kernel and + user updates together. */ + xen_mc_batch(); + + __xen_set_pgd_hyper(ptr, val); + if (user_ptr) + __xen_set_pgd_hyper(user_ptr, val); + + xen_mc_issue(PARAVIRT_LAZY_MMU); +} +#endif /* CONFIG_PGTABLE_LEVELS == 4 */ + +/* + * (Yet another) pagetable walker. This one is intended for pinning a + * pagetable. This means that it walks a pagetable and calls the + * callback function on each page it finds making up the page table, + * at every level. It walks the entire pagetable, but it only bothers + * pinning pte pages which are below limit. In the normal case this + * will be STACK_TOP_MAX, but at boot we need to pin up to + * FIXADDR_TOP. + * + * For 32-bit the important bit is that we don't pin beyond there, + * because then we start getting into Xen's ptes. + * + * For 64-bit, we must skip the Xen hole in the middle of the address + * space, just after the big x86-64 virtual hole. + */ +static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd, + int (*func)(struct mm_struct *mm, struct page *, + enum pt_level), + unsigned long limit) +{ + int flush = 0; + unsigned hole_low, hole_high; + unsigned pgdidx_limit, pudidx_limit, pmdidx_limit; + unsigned pgdidx, pudidx, pmdidx; + + /* The limit is the last byte to be touched */ + limit--; + BUG_ON(limit >= FIXADDR_TOP); + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return 0; + + /* + * 64-bit has a great big hole in the middle of the address + * space, which contains the Xen mappings. On 32-bit these + * will end up making a zero-sized hole and so is a no-op. + */ + hole_low = pgd_index(USER_LIMIT); + hole_high = pgd_index(PAGE_OFFSET); + + pgdidx_limit = pgd_index(limit); +#if PTRS_PER_PUD > 1 + pudidx_limit = pud_index(limit); +#else + pudidx_limit = 0; +#endif +#if PTRS_PER_PMD > 1 + pmdidx_limit = pmd_index(limit); +#else + pmdidx_limit = 0; +#endif + + for (pgdidx = 0; pgdidx <= pgdidx_limit; pgdidx++) { + pud_t *pud; + + if (pgdidx >= hole_low && pgdidx < hole_high) + continue; + + if (!pgd_val(pgd[pgdidx])) + continue; + + pud = pud_offset(&pgd[pgdidx], 0); + + if (PTRS_PER_PUD > 1) /* not folded */ + flush |= (*func)(mm, virt_to_page(pud), PT_PUD); + + for (pudidx = 0; pudidx < PTRS_PER_PUD; pudidx++) { + pmd_t *pmd; + + if (pgdidx == pgdidx_limit && + pudidx > pudidx_limit) + goto out; + + if (pud_none(pud[pudidx])) + continue; + + pmd = pmd_offset(&pud[pudidx], 0); + + if (PTRS_PER_PMD > 1) /* not folded */ + flush |= (*func)(mm, virt_to_page(pmd), PT_PMD); + + for (pmdidx = 0; pmdidx < PTRS_PER_PMD; pmdidx++) { + struct page *pte; + + if (pgdidx == pgdidx_limit && + pudidx == pudidx_limit && + pmdidx > pmdidx_limit) + goto out; + + if (pmd_none(pmd[pmdidx])) + continue; + + pte = pmd_page(pmd[pmdidx]); + flush |= (*func)(mm, pte, PT_PTE); + } + } + } + +out: + /* Do the top level last, so that the callbacks can use it as + a cue to do final things like tlb flushes. */ + flush |= (*func)(mm, virt_to_page(pgd), PT_PGD); + + return flush; +} + +static int xen_pgd_walk(struct mm_struct *mm, + int (*func)(struct mm_struct *mm, struct page *, + enum pt_level), + unsigned long limit) +{ + return __xen_pgd_walk(mm, mm->pgd, func, limit); +} + +/* If we're using split pte locks, then take the page's lock and + return a pointer to it. Otherwise return NULL. */ +static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm) +{ + spinlock_t *ptl = NULL; + +#if USE_SPLIT_PTE_PTLOCKS + ptl = ptlock_ptr(page); + spin_lock_nest_lock(ptl, &mm->page_table_lock); +#endif + + return ptl; +} + +static void xen_pte_unlock(void *v) +{ + spinlock_t *ptl = v; + spin_unlock(ptl); +} + +static void xen_do_pin(unsigned level, unsigned long pfn) +{ + struct mmuext_op op; + + op.cmd = level; + op.arg1.mfn = pfn_to_mfn(pfn); + + xen_extend_mmuext_op(&op); +} + +static int xen_pin_page(struct mm_struct *mm, struct page *page, + enum pt_level level) +{ + unsigned pgfl = TestSetPagePinned(page); + int flush; + + if (pgfl) + flush = 0; /* already pinned */ + else if (PageHighMem(page)) + /* kmaps need flushing if we found an unpinned + highpage */ + flush = 1; + else { + void *pt = lowmem_page_address(page); + unsigned long pfn = page_to_pfn(page); + struct multicall_space mcs = __xen_mc_entry(0); + spinlock_t *ptl; + + flush = 0; + + /* + * We need to hold the pagetable lock between the time + * we make the pagetable RO and when we actually pin + * it. If we don't, then other users may come in and + * attempt to update the pagetable by writing it, + * which will fail because the memory is RO but not + * pinned, so Xen won't do the trap'n'emulate. + * + * If we're using split pte locks, we can't hold the + * entire pagetable's worth of locks during the + * traverse, because we may wrap the preempt count (8 + * bits). The solution is to mark RO and pin each PTE + * page while holding the lock. This means the number + * of locks we end up holding is never more than a + * batch size (~32 entries, at present). + * + * If we're not using split pte locks, we needn't pin + * the PTE pages independently, because we're + * protected by the overall pagetable lock. + */ + ptl = NULL; + if (level == PT_PTE) + ptl = xen_pte_lock(page, mm); + + MULTI_update_va_mapping(mcs.mc, (unsigned long)pt, + pfn_pte(pfn, PAGE_KERNEL_RO), + level == PT_PGD ? UVMF_TLB_FLUSH : 0); + + if (ptl) { + xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn); + + /* Queue a deferred unlock for when this batch + is completed. */ + xen_mc_callback(xen_pte_unlock, ptl); + } + } + + return flush; +} + +/* This is called just after a mm has been created, but it has not + been used yet. We need to make sure that its pagetable is all + read-only, and can be pinned. */ +static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd) +{ + trace_xen_mmu_pgd_pin(mm, pgd); + + xen_mc_batch(); + + if (__xen_pgd_walk(mm, pgd, xen_pin_page, USER_LIMIT)) { + /* re-enable interrupts for flushing */ + xen_mc_issue(0); + + kmap_flush_unused(); + + xen_mc_batch(); + } + +#ifdef CONFIG_X86_64 + { + pgd_t *user_pgd = xen_get_user_pgd(pgd); + + xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(pgd))); + + if (user_pgd) { + xen_pin_page(mm, virt_to_page(user_pgd), PT_PGD); + xen_do_pin(MMUEXT_PIN_L4_TABLE, + PFN_DOWN(__pa(user_pgd))); + } + } +#else /* CONFIG_X86_32 */ +#ifdef CONFIG_X86_PAE + /* Need to make sure unshared kernel PMD is pinnable */ + xen_pin_page(mm, pgd_page(pgd[pgd_index(TASK_SIZE)]), + PT_PMD); +#endif + xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd))); +#endif /* CONFIG_X86_64 */ + xen_mc_issue(0); +} + +static void xen_pgd_pin(struct mm_struct *mm) +{ + __xen_pgd_pin(mm, mm->pgd); +} + +/* + * On save, we need to pin all pagetables to make sure they get their + * mfns turned into pfns. Search the list for any unpinned pgds and pin + * them (unpinned pgds are not currently in use, probably because the + * process is under construction or destruction). + * + * Expected to be called in stop_machine() ("equivalent to taking + * every spinlock in the system"), so the locking doesn't really + * matter all that much. + */ +void xen_mm_pin_all(void) +{ + struct page *page; + + spin_lock(&pgd_lock); + + list_for_each_entry(page, &pgd_list, lru) { + if (!PagePinned(page)) { + __xen_pgd_pin(&init_mm, (pgd_t *)page_address(page)); + SetPageSavePinned(page); + } + } + + spin_unlock(&pgd_lock); +} + +/* + * The init_mm pagetable is really pinned as soon as its created, but + * that's before we have page structures to store the bits. So do all + * the book-keeping now. + */ +static int __init xen_mark_pinned(struct mm_struct *mm, struct page *page, + enum pt_level level) +{ + SetPagePinned(page); + return 0; +} + +static void __init xen_mark_init_mm_pinned(void) +{ + xen_pgd_walk(&init_mm, xen_mark_pinned, FIXADDR_TOP); +} + +static int xen_unpin_page(struct mm_struct *mm, struct page *page, + enum pt_level level) +{ + unsigned pgfl = TestClearPagePinned(page); + + if (pgfl && !PageHighMem(page)) { + void *pt = lowmem_page_address(page); + unsigned long pfn = page_to_pfn(page); + spinlock_t *ptl = NULL; + struct multicall_space mcs; + + /* + * Do the converse to pin_page. If we're using split + * pte locks, we must be holding the lock for while + * the pte page is unpinned but still RO to prevent + * concurrent updates from seeing it in this + * partially-pinned state. + */ + if (level == PT_PTE) { + ptl = xen_pte_lock(page, mm); + + if (ptl) + xen_do_pin(MMUEXT_UNPIN_TABLE, pfn); + } + + mcs = __xen_mc_entry(0); + + MULTI_update_va_mapping(mcs.mc, (unsigned long)pt, + pfn_pte(pfn, PAGE_KERNEL), + level == PT_PGD ? UVMF_TLB_FLUSH : 0); + + if (ptl) { + /* unlock when batch completed */ + xen_mc_callback(xen_pte_unlock, ptl); + } + } + + return 0; /* never need to flush on unpin */ +} + +/* Release a pagetables pages back as normal RW */ +static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd) +{ + trace_xen_mmu_pgd_unpin(mm, pgd); + + xen_mc_batch(); + + xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd))); + +#ifdef CONFIG_X86_64 + { + pgd_t *user_pgd = xen_get_user_pgd(pgd); + + if (user_pgd) { + xen_do_pin(MMUEXT_UNPIN_TABLE, + PFN_DOWN(__pa(user_pgd))); + xen_unpin_page(mm, virt_to_page(user_pgd), PT_PGD); + } + } +#endif + +#ifdef CONFIG_X86_PAE + /* Need to make sure unshared kernel PMD is unpinned */ + xen_unpin_page(mm, pgd_page(pgd[pgd_index(TASK_SIZE)]), + PT_PMD); +#endif + + __xen_pgd_walk(mm, pgd, xen_unpin_page, USER_LIMIT); + + xen_mc_issue(0); +} + +static void xen_pgd_unpin(struct mm_struct *mm) +{ + __xen_pgd_unpin(mm, mm->pgd); +} + +/* + * On resume, undo any pinning done at save, so that the rest of the + * kernel doesn't see any unexpected pinned pagetables. + */ +void xen_mm_unpin_all(void) +{ + struct page *page; + + spin_lock(&pgd_lock); + + list_for_each_entry(page, &pgd_list, lru) { + if (PageSavePinned(page)) { + BUG_ON(!PagePinned(page)); + __xen_pgd_unpin(&init_mm, (pgd_t *)page_address(page)); + ClearPageSavePinned(page); + } + } + + spin_unlock(&pgd_lock); +} + +static void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next) +{ + spin_lock(&next->page_table_lock); + xen_pgd_pin(next); + spin_unlock(&next->page_table_lock); +} + +static void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm) +{ + spin_lock(&mm->page_table_lock); + xen_pgd_pin(mm); + spin_unlock(&mm->page_table_lock); +} + + +#ifdef CONFIG_SMP +/* Another cpu may still have their %cr3 pointing at the pagetable, so + we need to repoint it somewhere else before we can unpin it. */ +static void drop_other_mm_ref(void *info) +{ + struct mm_struct *mm = info; + struct mm_struct *active_mm; + + active_mm = this_cpu_read(cpu_tlbstate.active_mm); + + if (active_mm == mm && this_cpu_read(cpu_tlbstate.state) != TLBSTATE_OK) + leave_mm(smp_processor_id()); + + /* If this cpu still has a stale cr3 reference, then make sure + it has been flushed. */ + if (this_cpu_read(xen_current_cr3) == __pa(mm->pgd)) + load_cr3(swapper_pg_dir); +} + +static void xen_drop_mm_ref(struct mm_struct *mm) +{ + cpumask_var_t mask; + unsigned cpu; + + if (current->active_mm == mm) { + if (current->mm == mm) + load_cr3(swapper_pg_dir); + else + leave_mm(smp_processor_id()); + } + + /* Get the "official" set of cpus referring to our pagetable. */ + if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) { + for_each_online_cpu(cpu) { + if (!cpumask_test_cpu(cpu, mm_cpumask(mm)) + && per_cpu(xen_current_cr3, cpu) != __pa(mm->pgd)) + continue; + smp_call_function_single(cpu, drop_other_mm_ref, mm, 1); + } + return; + } + cpumask_copy(mask, mm_cpumask(mm)); + + /* It's possible that a vcpu may have a stale reference to our + cr3, because its in lazy mode, and it hasn't yet flushed + its set of pending hypercalls yet. In this case, we can + look at its actual current cr3 value, and force it to flush + if needed. */ + for_each_online_cpu(cpu) { + if (per_cpu(xen_current_cr3, cpu) == __pa(mm->pgd)) + cpumask_set_cpu(cpu, mask); + } + + if (!cpumask_empty(mask)) + smp_call_function_many(mask, drop_other_mm_ref, mm, 1); + free_cpumask_var(mask); +} +#else +static void xen_drop_mm_ref(struct mm_struct *mm) +{ + if (current->active_mm == mm) + load_cr3(swapper_pg_dir); +} +#endif + +/* + * While a process runs, Xen pins its pagetables, which means that the + * hypervisor forces it to be read-only, and it controls all updates + * to it. This means that all pagetable updates have to go via the + * hypervisor, which is moderately expensive. + * + * Since we're pulling the pagetable down, we switch to use init_mm, + * unpin old process pagetable and mark it all read-write, which + * allows further operations on it to be simple memory accesses. + * + * The only subtle point is that another CPU may be still using the + * pagetable because of lazy tlb flushing. This means we need need to + * switch all CPUs off this pagetable before we can unpin it. + */ +static void xen_exit_mmap(struct mm_struct *mm) +{ + get_cpu(); /* make sure we don't move around */ + xen_drop_mm_ref(mm); + put_cpu(); + + spin_lock(&mm->page_table_lock); + + /* pgd may not be pinned in the error exit path of execve */ + if (xen_page_pinned(mm->pgd)) + xen_pgd_unpin(mm); + + spin_unlock(&mm->page_table_lock); +} + +static void xen_post_allocator_init(void); + +#ifdef CONFIG_X86_64 +static void __init xen_cleanhighmap(unsigned long vaddr, + unsigned long vaddr_end) +{ + unsigned long kernel_end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1; + pmd_t *pmd = level2_kernel_pgt + pmd_index(vaddr); + + /* NOTE: The loop is more greedy than the cleanup_highmap variant. + * We include the PMD passed in on _both_ boundaries. */ + for (; vaddr <= vaddr_end && (pmd < (level2_kernel_pgt + PAGE_SIZE)); + pmd++, vaddr += PMD_SIZE) { + if (pmd_none(*pmd)) + continue; + if (vaddr < (unsigned long) _text || vaddr > kernel_end) + set_pmd(pmd, __pmd(0)); + } + /* In case we did something silly, we should crash in this function + * instead of somewhere later and be confusing. */ + xen_mc_flush(); +} + +static void __init xen_pagetable_p2m_free(void) +{ + unsigned long size; + unsigned long addr; + + size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long)); + + /* No memory or already called. */ + if ((unsigned long)xen_p2m_addr == xen_start_info->mfn_list) + return; + + /* using __ka address and sticking INVALID_P2M_ENTRY! */ + memset((void *)xen_start_info->mfn_list, 0xff, size); + + /* We should be in __ka space. */ + BUG_ON(xen_start_info->mfn_list < __START_KERNEL_map); + addr = xen_start_info->mfn_list; + /* We roundup to the PMD, which means that if anybody at this stage is + * using the __ka address of xen_start_info or xen_start_info->shared_info + * they are in going to crash. Fortunatly we have already revectored + * in xen_setup_kernel_pagetable and in xen_setup_shared_info. */ + size = roundup(size, PMD_SIZE); + xen_cleanhighmap(addr, addr + size); + + size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long)); + memblock_free(__pa(xen_start_info->mfn_list), size); + + /* At this stage, cleanup_highmap has already cleaned __ka space + * from _brk_limit way up to the max_pfn_mapped (which is the end of + * the ramdisk). We continue on, erasing PMD entries that point to page + * tables - do note that they are accessible at this stage via __va. + * For good measure we also round up to the PMD - which means that if + * anybody is using __ka address to the initial boot-stack - and try + * to use it - they are going to crash. The xen_start_info has been + * taken care of already in xen_setup_kernel_pagetable. */ + addr = xen_start_info->pt_base; + size = roundup(xen_start_info->nr_pt_frames * PAGE_SIZE, PMD_SIZE); + + xen_cleanhighmap(addr, addr + size); + xen_start_info->pt_base = (unsigned long)__va(__pa(xen_start_info->pt_base)); +#ifdef DEBUG + /* This is superflous and is not neccessary, but you know what + * lets do it. The MODULES_VADDR -> MODULES_END should be clear of + * anything at this stage. */ + xen_cleanhighmap(MODULES_VADDR, roundup(MODULES_VADDR, PUD_SIZE) - 1); +#endif +} +#endif + +static void __init xen_pagetable_p2m_setup(void) +{ + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + xen_vmalloc_p2m_tree(); + +#ifdef CONFIG_X86_64 + xen_pagetable_p2m_free(); +#endif + /* And revector! Bye bye old array */ + xen_start_info->mfn_list = (unsigned long)xen_p2m_addr; +} + +static void __init xen_pagetable_init(void) +{ + paging_init(); + xen_post_allocator_init(); + + xen_pagetable_p2m_setup(); + + /* Allocate and initialize top and mid mfn levels for p2m structure */ + xen_build_mfn_list_list(); + + /* Remap memory freed due to conflicts with E820 map */ + if (!xen_feature(XENFEAT_auto_translated_physmap)) + xen_remap_memory(); + + xen_setup_shared_info(); +} +static void xen_write_cr2(unsigned long cr2) +{ + this_cpu_read(xen_vcpu)->arch.cr2 = cr2; +} + +static unsigned long xen_read_cr2(void) +{ + return this_cpu_read(xen_vcpu)->arch.cr2; +} + +unsigned long xen_read_cr2_direct(void) +{ + return this_cpu_read(xen_vcpu_info.arch.cr2); +} + +void xen_flush_tlb_all(void) +{ + struct mmuext_op *op; + struct multicall_space mcs; + + trace_xen_mmu_flush_tlb_all(0); + + preempt_disable(); + + mcs = xen_mc_entry(sizeof(*op)); + + op = mcs.args; + op->cmd = MMUEXT_TLB_FLUSH_ALL; + MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} +static void xen_flush_tlb(void) +{ + struct mmuext_op *op; + struct multicall_space mcs; + + trace_xen_mmu_flush_tlb(0); + + preempt_disable(); + + mcs = xen_mc_entry(sizeof(*op)); + + op = mcs.args; + op->cmd = MMUEXT_TLB_FLUSH_LOCAL; + MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} + +static void xen_flush_tlb_single(unsigned long addr) +{ + struct mmuext_op *op; + struct multicall_space mcs; + + trace_xen_mmu_flush_tlb_single(addr); + + preempt_disable(); + + mcs = xen_mc_entry(sizeof(*op)); + op = mcs.args; + op->cmd = MMUEXT_INVLPG_LOCAL; + op->arg1.linear_addr = addr & PAGE_MASK; + MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + + preempt_enable(); +} + +static void xen_flush_tlb_others(const struct cpumask *cpus, + struct mm_struct *mm, unsigned long start, + unsigned long end) +{ + struct { + struct mmuext_op op; +#ifdef CONFIG_SMP + DECLARE_BITMAP(mask, num_processors); +#else + DECLARE_BITMAP(mask, NR_CPUS); +#endif + } *args; + struct multicall_space mcs; + + trace_xen_mmu_flush_tlb_others(cpus, mm, start, end); + + if (cpumask_empty(cpus)) + return; /* nothing to do */ + + mcs = xen_mc_entry(sizeof(*args)); + args = mcs.args; + args->op.arg2.vcpumask = to_cpumask(args->mask); + + /* Remove us, and any offline CPUS. */ + cpumask_and(to_cpumask(args->mask), cpus, cpu_online_mask); + cpumask_clear_cpu(smp_processor_id(), to_cpumask(args->mask)); + + args->op.cmd = MMUEXT_TLB_FLUSH_MULTI; + if (end != TLB_FLUSH_ALL && (end - start) <= PAGE_SIZE) { + args->op.cmd = MMUEXT_INVLPG_MULTI; + args->op.arg1.linear_addr = start; + } + + MULTI_mmuext_op(mcs.mc, &args->op, 1, NULL, DOMID_SELF); + + xen_mc_issue(PARAVIRT_LAZY_MMU); +} + +static unsigned long xen_read_cr3(void) +{ + return this_cpu_read(xen_cr3); +} + +static void set_current_cr3(void *v) +{ + this_cpu_write(xen_current_cr3, (unsigned long)v); +} + +static void __xen_write_cr3(bool kernel, unsigned long cr3) +{ + struct mmuext_op op; + unsigned long mfn; + + trace_xen_mmu_write_cr3(kernel, cr3); + + if (cr3) + mfn = pfn_to_mfn(PFN_DOWN(cr3)); + else + mfn = 0; + + WARN_ON(mfn == 0 && kernel); + + op.cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR; + op.arg1.mfn = mfn; + + xen_extend_mmuext_op(&op); + + if (kernel) { + this_cpu_write(xen_cr3, cr3); + + /* Update xen_current_cr3 once the batch has actually + been submitted. */ + xen_mc_callback(set_current_cr3, (void *)cr3); + } +} +static void xen_write_cr3(unsigned long cr3) +{ + BUG_ON(preemptible()); + + xen_mc_batch(); /* disables interrupts */ + + /* Update while interrupts are disabled, so its atomic with + respect to ipis */ + this_cpu_write(xen_cr3, cr3); + + __xen_write_cr3(true, cr3); + +#ifdef CONFIG_X86_64 + { + pgd_t *user_pgd = xen_get_user_pgd(__va(cr3)); + if (user_pgd) + __xen_write_cr3(false, __pa(user_pgd)); + else + __xen_write_cr3(false, 0); + } +#endif + + xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */ +} + +#ifdef CONFIG_X86_64 +/* + * At the start of the day - when Xen launches a guest, it has already + * built pagetables for the guest. We diligently look over them + * in xen_setup_kernel_pagetable and graft as appropiate them in the + * init_level4_pgt and its friends. Then when we are happy we load + * the new init_level4_pgt - and continue on. + * + * The generic code starts (start_kernel) and 'init_mem_mapping' sets + * up the rest of the pagetables. When it has completed it loads the cr3. + * N.B. that baremetal would start at 'start_kernel' (and the early + * #PF handler would create bootstrap pagetables) - so we are running + * with the same assumptions as what to do when write_cr3 is executed + * at this point. + * + * Since there are no user-page tables at all, we have two variants + * of xen_write_cr3 - the early bootup (this one), and the late one + * (xen_write_cr3). The reason we have to do that is that in 64-bit + * the Linux kernel and user-space are both in ring 3 while the + * hypervisor is in ring 0. + */ +static void __init xen_write_cr3_init(unsigned long cr3) +{ + BUG_ON(preemptible()); + + xen_mc_batch(); /* disables interrupts */ + + /* Update while interrupts are disabled, so its atomic with + respect to ipis */ + this_cpu_write(xen_cr3, cr3); + + __xen_write_cr3(true, cr3); + + xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */ +} +#endif + +static int xen_pgd_alloc(struct mm_struct *mm) +{ + pgd_t *pgd = mm->pgd; + int ret = 0; + + BUG_ON(PagePinned(virt_to_page(pgd))); + +#ifdef CONFIG_X86_64 + { + struct page *page = virt_to_page(pgd); + pgd_t *user_pgd; + + BUG_ON(page->private != 0); + + ret = -ENOMEM; + + user_pgd = (pgd_t *)__get_free_page(GFP_KERNEL | __GFP_ZERO); + page->private = (unsigned long)user_pgd; + + if (user_pgd != NULL) { +#ifdef CONFIG_X86_VSYSCALL_EMULATION + user_pgd[pgd_index(VSYSCALL_ADDR)] = + __pgd(__pa(level3_user_vsyscall) | _PAGE_TABLE); +#endif + ret = 0; + } + + BUG_ON(PagePinned(virt_to_page(xen_get_user_pgd(pgd)))); + } +#endif + + return ret; +} + +static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd) +{ +#ifdef CONFIG_X86_64 + pgd_t *user_pgd = xen_get_user_pgd(pgd); + + if (user_pgd) + free_page((unsigned long)user_pgd); +#endif +} + +#ifdef CONFIG_X86_32 +static pte_t __init mask_rw_pte(pte_t *ptep, pte_t pte) +{ + /* If there's an existing pte, then don't allow _PAGE_RW to be set */ + if (pte_val_ma(*ptep) & _PAGE_PRESENT) + pte = __pte_ma(((pte_val_ma(*ptep) & _PAGE_RW) | ~_PAGE_RW) & + pte_val_ma(pte)); + + return pte; +} +#else /* CONFIG_X86_64 */ +static pte_t __init mask_rw_pte(pte_t *ptep, pte_t pte) +{ + return pte; +} +#endif /* CONFIG_X86_64 */ + +/* + * Init-time set_pte while constructing initial pagetables, which + * doesn't allow RO page table pages to be remapped RW. + * + * If there is no MFN for this PFN then this page is initially + * ballooned out so clear the PTE (as in decrease_reservation() in + * drivers/xen/balloon.c). + * + * Many of these PTE updates are done on unpinned and writable pages + * and doing a hypercall for these is unnecessary and expensive. At + * this point it is not possible to tell if a page is pinned or not, + * so always write the PTE directly and rely on Xen trapping and + * emulating any updates as necessary. + */ +static void __init xen_set_pte_init(pte_t *ptep, pte_t pte) +{ + if (pte_mfn(pte) != INVALID_P2M_ENTRY) + pte = mask_rw_pte(ptep, pte); + else + pte = __pte_ma(0); + + native_set_pte(ptep, pte); +} + +static void __init pin_pagetable_pfn(unsigned cmd, unsigned long pfn) +{ + struct mmuext_op op; + op.cmd = cmd; + op.arg1.mfn = pfn_to_mfn(pfn); + if (HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF)) + BUG(); +} + +/* Early in boot, while setting up the initial pagetable, assume + everything is pinned. */ +static void __init xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) +{ +#ifdef CONFIG_FLATMEM + BUG_ON(mem_map); /* should only be used early */ +#endif + make_lowmem_page_readonly(__va(PFN_PHYS(pfn))); + pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn); +} + +/* Used for pmd and pud */ +static void __init xen_alloc_pmd_init(struct mm_struct *mm, unsigned long pfn) +{ +#ifdef CONFIG_FLATMEM + BUG_ON(mem_map); /* should only be used early */ +#endif + make_lowmem_page_readonly(__va(PFN_PHYS(pfn))); +} + +/* Early release_pte assumes that all pts are pinned, since there's + only init_mm and anything attached to that is pinned. */ +static void __init xen_release_pte_init(unsigned long pfn) +{ + pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn); + make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); +} + +static void __init xen_release_pmd_init(unsigned long pfn) +{ + make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); +} + +static inline void __pin_pagetable_pfn(unsigned cmd, unsigned long pfn) +{ + struct multicall_space mcs; + struct mmuext_op *op; + + mcs = __xen_mc_entry(sizeof(*op)); + op = mcs.args; + op->cmd = cmd; + op->arg1.mfn = pfn_to_mfn(pfn); + + MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF); +} + +static inline void __set_pfn_prot(unsigned long pfn, pgprot_t prot) +{ + struct multicall_space mcs; + unsigned long addr = (unsigned long)__va(pfn << PAGE_SHIFT); + + mcs = __xen_mc_entry(0); + MULTI_update_va_mapping(mcs.mc, (unsigned long)addr, + pfn_pte(pfn, prot), 0); +} + +/* This needs to make sure the new pte page is pinned iff its being + attached to a pinned pagetable. */ +static inline void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, + unsigned level) +{ + bool pinned = PagePinned(virt_to_page(mm->pgd)); + + trace_xen_mmu_alloc_ptpage(mm, pfn, level, pinned); + + if (pinned) { + struct page *page = pfn_to_page(pfn); + + SetPagePinned(page); + + if (!PageHighMem(page)) { + xen_mc_batch(); + + __set_pfn_prot(pfn, PAGE_KERNEL_RO); + + if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS) + __pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + } else { + /* make sure there are no stray mappings of + this page */ + kmap_flush_unused(); + } + } +} + +static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn) +{ + xen_alloc_ptpage(mm, pfn, PT_PTE); +} + +static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn) +{ + xen_alloc_ptpage(mm, pfn, PT_PMD); +} + +/* This should never happen until we're OK to use struct page */ +static inline void xen_release_ptpage(unsigned long pfn, unsigned level) +{ + struct page *page = pfn_to_page(pfn); + bool pinned = PagePinned(page); + + trace_xen_mmu_release_ptpage(pfn, level, pinned); + + if (pinned) { + if (!PageHighMem(page)) { + xen_mc_batch(); + + if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS) + __pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn); + + __set_pfn_prot(pfn, PAGE_KERNEL); + + xen_mc_issue(PARAVIRT_LAZY_MMU); + } + ClearPagePinned(page); + } +} + +static void xen_release_pte(unsigned long pfn) +{ + xen_release_ptpage(pfn, PT_PTE); +} + +static void xen_release_pmd(unsigned long pfn) +{ + xen_release_ptpage(pfn, PT_PMD); +} + +#if CONFIG_PGTABLE_LEVELS == 4 +static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn) +{ + xen_alloc_ptpage(mm, pfn, PT_PUD); +} + +static void xen_release_pud(unsigned long pfn) +{ + xen_release_ptpage(pfn, PT_PUD); +} +#endif + +void __init xen_reserve_top(void) +{ +#ifdef CONFIG_X86_32 + unsigned long top = HYPERVISOR_VIRT_START; + struct xen_platform_parameters pp; + + if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0) + top = pp.virt_start; + + reserve_top_address(-top); +#endif /* CONFIG_X86_32 */ +} + +/* + * Like __va(), but returns address in the kernel mapping (which is + * all we have until the physical memory mapping has been set up. + */ +static void * __init __ka(phys_addr_t paddr) +{ +#ifdef CONFIG_X86_64 + return (void *)(paddr + __START_KERNEL_map); +#else + return __va(paddr); +#endif +} + +/* Convert a machine address to physical address */ +static unsigned long __init m2p(phys_addr_t maddr) +{ + phys_addr_t paddr; + + maddr &= PTE_PFN_MASK; + paddr = mfn_to_pfn(maddr >> PAGE_SHIFT) << PAGE_SHIFT; + + return paddr; +} + +/* Convert a machine address to kernel virtual */ +static void * __init m2v(phys_addr_t maddr) +{ + return __ka(m2p(maddr)); +} + +/* Set the page permissions on an identity-mapped pages */ +static void __init set_page_prot_flags(void *addr, pgprot_t prot, + unsigned long flags) +{ + unsigned long pfn = __pa(addr) >> PAGE_SHIFT; + pte_t pte = pfn_pte(pfn, prot); + + /* For PVH no need to set R/O or R/W to pin them or unpin them. */ + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, flags)) + BUG(); +} +static void __init set_page_prot(void *addr, pgprot_t prot) +{ + return set_page_prot_flags(addr, prot, UVMF_NONE); +} +#ifdef CONFIG_X86_32 +static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) +{ + unsigned pmdidx, pteidx; + unsigned ident_pte; + unsigned long pfn; + + level1_ident_pgt = extend_brk(sizeof(pte_t) * LEVEL1_IDENT_ENTRIES, + PAGE_SIZE); + + ident_pte = 0; + pfn = 0; + for (pmdidx = 0; pmdidx < PTRS_PER_PMD && pfn < max_pfn; pmdidx++) { + pte_t *pte_page; + + /* Reuse or allocate a page of ptes */ + if (pmd_present(pmd[pmdidx])) + pte_page = m2v(pmd[pmdidx].pmd); + else { + /* Check for free pte pages */ + if (ident_pte == LEVEL1_IDENT_ENTRIES) + break; + + pte_page = &level1_ident_pgt[ident_pte]; + ident_pte += PTRS_PER_PTE; + + pmd[pmdidx] = __pmd(__pa(pte_page) | _PAGE_TABLE); + } + + /* Install mappings */ + for (pteidx = 0; pteidx < PTRS_PER_PTE; pteidx++, pfn++) { + pte_t pte; + + if (pfn > max_pfn_mapped) + max_pfn_mapped = pfn; + + if (!pte_none(pte_page[pteidx])) + continue; + + pte = pfn_pte(pfn, PAGE_KERNEL_EXEC); + pte_page[pteidx] = pte; + } + } + + for (pteidx = 0; pteidx < ident_pte; pteidx += PTRS_PER_PTE) + set_page_prot(&level1_ident_pgt[pteidx], PAGE_KERNEL_RO); + + set_page_prot(pmd, PAGE_KERNEL_RO); +} +#endif +void __init xen_setup_machphys_mapping(void) +{ + struct xen_machphys_mapping mapping; + + if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) { + machine_to_phys_mapping = (unsigned long *)mapping.v_start; + machine_to_phys_nr = mapping.max_mfn + 1; + } else { + machine_to_phys_nr = MACH2PHYS_NR_ENTRIES; + } +#ifdef CONFIG_X86_32 + WARN_ON((machine_to_phys_mapping + (machine_to_phys_nr - 1)) + < machine_to_phys_mapping); +#endif +} + +#ifdef CONFIG_X86_64 +static void __init convert_pfn_mfn(void *v) +{ + pte_t *pte = v; + int i; + + /* All levels are converted the same way, so just treat them + as ptes. */ + for (i = 0; i < PTRS_PER_PTE; i++) + pte[i] = xen_make_pte(pte[i].pte); +} +static void __init check_pt_base(unsigned long *pt_base, unsigned long *pt_end, + unsigned long addr) +{ + if (*pt_base == PFN_DOWN(__pa(addr))) { + set_page_prot_flags((void *)addr, PAGE_KERNEL, UVMF_INVLPG); + clear_page((void *)addr); + (*pt_base)++; + } + if (*pt_end == PFN_DOWN(__pa(addr))) { + set_page_prot_flags((void *)addr, PAGE_KERNEL, UVMF_INVLPG); + clear_page((void *)addr); + (*pt_end)--; + } +} +/* + * Set up the initial kernel pagetable. + * + * We can construct this by grafting the Xen provided pagetable into + * head_64.S's preconstructed pagetables. We copy the Xen L2's into + * level2_ident_pgt, and level2_kernel_pgt. This means that only the + * kernel has a physical mapping to start with - but that's enough to + * get __va working. We need to fill in the rest of the physical + * mapping once some sort of allocator has been set up. NOTE: for + * PVH, the page tables are native. + */ +void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) +{ + pud_t *l3; + pmd_t *l2; + unsigned long addr[3]; + unsigned long pt_base, pt_end; + unsigned i; + + /* max_pfn_mapped is the last pfn mapped in the initial memory + * mappings. Considering that on Xen after the kernel mappings we + * have the mappings of some pages that don't exist in pfn space, we + * set max_pfn_mapped to the last real pfn mapped. */ + max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->mfn_list)); + + pt_base = PFN_DOWN(__pa(xen_start_info->pt_base)); + pt_end = pt_base + xen_start_info->nr_pt_frames; + + /* Zap identity mapping */ + init_level4_pgt[0] = __pgd(0); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* Pre-constructed entries are in pfn, so convert to mfn */ + /* L4[272] -> level3_ident_pgt + * L4[511] -> level3_kernel_pgt */ + convert_pfn_mfn(init_level4_pgt); + + /* L3_i[0] -> level2_ident_pgt */ + convert_pfn_mfn(level3_ident_pgt); + /* L3_k[510] -> level2_kernel_pgt + * L3_k[511] -> level2_fixmap_pgt */ + convert_pfn_mfn(level3_kernel_pgt); + + /* L3_k[511][506] -> level1_fixmap_pgt */ + convert_pfn_mfn(level2_fixmap_pgt); + } + /* We get [511][511] and have Xen's version of level2_kernel_pgt */ + l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd); + l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud); + + addr[0] = (unsigned long)pgd; + addr[1] = (unsigned long)l3; + addr[2] = (unsigned long)l2; + /* Graft it onto L4[272][0]. Note that we creating an aliasing problem: + * Both L4[272][0] and L4[511][510] have entries that point to the same + * L2 (PMD) tables. Meaning that if you modify it in __va space + * it will be also modified in the __ka space! (But if you just + * modify the PMD table to point to other PTE's or none, then you + * are OK - which is what cleanup_highmap does) */ + copy_page(level2_ident_pgt, l2); + /* Graft it onto L4[511][510] */ + copy_page(level2_kernel_pgt, l2); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* Make pagetable pieces RO */ + set_page_prot(init_level4_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO); + set_page_prot(level2_ident_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO); + set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO); + set_page_prot(level1_fixmap_pgt, PAGE_KERNEL_RO); + + /* Pin down new L4 */ + pin_pagetable_pfn(MMUEXT_PIN_L4_TABLE, + PFN_DOWN(__pa_symbol(init_level4_pgt))); + + /* Unpin Xen-provided one */ + pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd))); + + /* + * At this stage there can be no user pgd, and no page + * structure to attach it to, so make sure we just set kernel + * pgd. + */ + xen_mc_batch(); + __xen_write_cr3(true, __pa(init_level4_pgt)); + xen_mc_issue(PARAVIRT_LAZY_CPU); + } else + native_write_cr3(__pa(init_level4_pgt)); + + /* We can't that easily rip out L3 and L2, as the Xen pagetables are + * set out this way: [L4], [L1], [L2], [L3], [L1], [L1] ... for + * the initial domain. For guests using the toolstack, they are in: + * [L4], [L3], [L2], [L1], [L1], order .. So for dom0 we can only + * rip out the [L4] (pgd), but for guests we shave off three pages. + */ + for (i = 0; i < ARRAY_SIZE(addr); i++) + check_pt_base(&pt_base, &pt_end, addr[i]); + + /* Our (by three pages) smaller Xen pagetable that we are using */ + memblock_reserve(PFN_PHYS(pt_base), (pt_end - pt_base) * PAGE_SIZE); + /* Revector the xen_start_info */ + xen_start_info = (struct start_info *)__va(__pa(xen_start_info)); +} +#else /* !CONFIG_X86_64 */ +static RESERVE_BRK_ARRAY(pmd_t, initial_kernel_pmd, PTRS_PER_PMD); +static RESERVE_BRK_ARRAY(pmd_t, swapper_kernel_pmd, PTRS_PER_PMD); + +static void __init xen_write_cr3_init(unsigned long cr3) +{ + unsigned long pfn = PFN_DOWN(__pa(swapper_pg_dir)); + + BUG_ON(read_cr3() != __pa(initial_page_table)); + BUG_ON(cr3 != __pa(swapper_pg_dir)); + + /* + * We are switching to swapper_pg_dir for the first time (from + * initial_page_table) and therefore need to mark that page + * read-only and then pin it. + * + * Xen disallows sharing of kernel PMDs for PAE + * guests. Therefore we must copy the kernel PMD from + * initial_page_table into a new kernel PMD to be used in + * swapper_pg_dir. + */ + swapper_kernel_pmd = + extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE); + copy_page(swapper_kernel_pmd, initial_kernel_pmd); + swapper_pg_dir[KERNEL_PGD_BOUNDARY] = + __pgd(__pa(swapper_kernel_pmd) | _PAGE_PRESENT); + set_page_prot(swapper_kernel_pmd, PAGE_KERNEL_RO); + + set_page_prot(swapper_pg_dir, PAGE_KERNEL_RO); + xen_write_cr3(cr3); + pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, pfn); + + pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, + PFN_DOWN(__pa(initial_page_table))); + set_page_prot(initial_page_table, PAGE_KERNEL); + set_page_prot(initial_kernel_pmd, PAGE_KERNEL); + + pv_mmu_ops.write_cr3 = &xen_write_cr3; +} + +void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) +{ + pmd_t *kernel_pmd; + + initial_kernel_pmd = + extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE); + + max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->pt_base) + + xen_start_info->nr_pt_frames * PAGE_SIZE + + 512*1024); + + kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd); + copy_page(initial_kernel_pmd, kernel_pmd); + + xen_map_identity_early(initial_kernel_pmd, max_pfn); + + copy_page(initial_page_table, pgd); + initial_page_table[KERNEL_PGD_BOUNDARY] = + __pgd(__pa(initial_kernel_pmd) | _PAGE_PRESENT); + + set_page_prot(initial_kernel_pmd, PAGE_KERNEL_RO); + set_page_prot(initial_page_table, PAGE_KERNEL_RO); + set_page_prot(empty_zero_page, PAGE_KERNEL_RO); + + pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd))); + + pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, + PFN_DOWN(__pa(initial_page_table))); + xen_write_cr3(__pa(initial_page_table)); + + memblock_reserve(__pa(xen_start_info->pt_base), + xen_start_info->nr_pt_frames * PAGE_SIZE); +} +#endif /* CONFIG_X86_64 */ + +static unsigned char dummy_mapping[PAGE_SIZE] __page_aligned_bss; + +static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot) +{ + pte_t pte; + + phys >>= PAGE_SHIFT; + + switch (idx) { + case FIX_BTMAP_END ... FIX_BTMAP_BEGIN: + case FIX_RO_IDT: +#ifdef CONFIG_X86_32 + case FIX_WP_TEST: +# ifdef CONFIG_HIGHMEM + case FIX_KMAP_BEGIN ... FIX_KMAP_END: +# endif +#elif defined(CONFIG_X86_VSYSCALL_EMULATION) + case VSYSCALL_PAGE: +#endif + case FIX_TEXT_POKE0: + case FIX_TEXT_POKE1: + /* All local page mappings */ + pte = pfn_pte(phys, prot); + break; + +#ifdef CONFIG_X86_LOCAL_APIC + case FIX_APIC_BASE: /* maps dummy local APIC */ + pte = pfn_pte(PFN_DOWN(__pa(dummy_mapping)), PAGE_KERNEL); + break; +#endif + +#ifdef CONFIG_X86_IO_APIC + case FIX_IO_APIC_BASE_0 ... FIX_IO_APIC_BASE_END: + /* + * We just don't map the IO APIC - all access is via + * hypercalls. Keep the address in the pte for reference. + */ + pte = pfn_pte(PFN_DOWN(__pa(dummy_mapping)), PAGE_KERNEL); + break; +#endif + + case FIX_PARAVIRT_BOOTMAP: + /* This is an MFN, but it isn't an IO mapping from the + IO domain */ + pte = mfn_pte(phys, prot); + break; + + default: + /* By default, set_fixmap is used for hardware mappings */ + pte = mfn_pte(phys, prot); + break; + } + + __native_set_fixmap(idx, pte); + +#ifdef CONFIG_X86_VSYSCALL_EMULATION + /* Replicate changes to map the vsyscall page into the user + pagetable vsyscall mapping. */ + if (idx == VSYSCALL_PAGE) { + unsigned long vaddr = __fix_to_virt(idx); + set_pte_vaddr_pud(level3_user_vsyscall, vaddr, pte); + } +#endif +} + +static void __init xen_post_allocator_init(void) +{ + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + pv_mmu_ops.set_pte = xen_set_pte; + pv_mmu_ops.set_pmd = xen_set_pmd; + pv_mmu_ops.set_pud = xen_set_pud; +#if CONFIG_PGTABLE_LEVELS == 4 + pv_mmu_ops.set_pgd = xen_set_pgd; +#endif + + /* This will work as long as patching hasn't happened yet + (which it hasn't) */ + pv_mmu_ops.alloc_pte = xen_alloc_pte; + pv_mmu_ops.alloc_pmd = xen_alloc_pmd; + pv_mmu_ops.release_pte = xen_release_pte; + pv_mmu_ops.release_pmd = xen_release_pmd; +#if CONFIG_PGTABLE_LEVELS == 4 + pv_mmu_ops.alloc_pud = xen_alloc_pud; + pv_mmu_ops.release_pud = xen_release_pud; +#endif + +#ifdef CONFIG_X86_64 + pv_mmu_ops.write_cr3 = &xen_write_cr3; + SetPagePinned(virt_to_page(level3_user_vsyscall)); +#endif + xen_mark_init_mm_pinned(); +} + +static void xen_leave_lazy_mmu(void) +{ + preempt_disable(); + xen_mc_flush(); + paravirt_leave_lazy_mmu(); + preempt_enable(); +} + +static const struct pv_mmu_ops xen_mmu_ops __initconst = { + .read_cr2 = xen_read_cr2, + .write_cr2 = xen_write_cr2, + + .read_cr3 = xen_read_cr3, + .write_cr3 = xen_write_cr3_init, + + .flush_tlb_user = xen_flush_tlb, + .flush_tlb_kernel = xen_flush_tlb, + .flush_tlb_single = xen_flush_tlb_single, + .flush_tlb_others = xen_flush_tlb_others, + + .pte_update = paravirt_nop, + .pte_update_defer = paravirt_nop, + + .pgd_alloc = xen_pgd_alloc, + .pgd_free = xen_pgd_free, + + .alloc_pte = xen_alloc_pte_init, + .release_pte = xen_release_pte_init, + .alloc_pmd = xen_alloc_pmd_init, + .release_pmd = xen_release_pmd_init, + + .set_pte = xen_set_pte_init, + .set_pte_at = xen_set_pte_at, + .set_pmd = xen_set_pmd_hyper, + + .ptep_modify_prot_start = __ptep_modify_prot_start, + .ptep_modify_prot_commit = __ptep_modify_prot_commit, + + .pte_val = PV_CALLEE_SAVE(xen_pte_val), + .pgd_val = PV_CALLEE_SAVE(xen_pgd_val), + + .make_pte = PV_CALLEE_SAVE(xen_make_pte), + .make_pgd = PV_CALLEE_SAVE(xen_make_pgd), + +#ifdef CONFIG_X86_PAE + .set_pte_atomic = xen_set_pte_atomic, + .pte_clear = xen_pte_clear, + .pmd_clear = xen_pmd_clear, +#endif /* CONFIG_X86_PAE */ + .set_pud = xen_set_pud_hyper, + + .make_pmd = PV_CALLEE_SAVE(xen_make_pmd), + .pmd_val = PV_CALLEE_SAVE(xen_pmd_val), + +#if CONFIG_PGTABLE_LEVELS == 4 + .pud_val = PV_CALLEE_SAVE(xen_pud_val), + .make_pud = PV_CALLEE_SAVE(xen_make_pud), + .set_pgd = xen_set_pgd_hyper, + + .alloc_pud = xen_alloc_pmd_init, + .release_pud = xen_release_pmd_init, +#endif /* CONFIG_PGTABLE_LEVELS == 4 */ + + .activate_mm = xen_activate_mm, + .dup_mmap = xen_dup_mmap, + .exit_mmap = xen_exit_mmap, + + .lazy_mode = { + .enter = paravirt_enter_lazy_mmu, + .leave = xen_leave_lazy_mmu, + .flush = paravirt_flush_lazy_mmu, + }, + + .set_fixmap = xen_set_fixmap, +}; + +void __init xen_init_mmu_ops(void) +{ + x86_init.paging.pagetable_init = xen_pagetable_init; + + /* Optimization - we can use the HVM one but it has no idea which + * VCPUs are descheduled - which means that it will needlessly IPI + * them. Xen knows so let it do the job. + */ + if (xen_feature(XENFEAT_auto_translated_physmap)) { + pv_mmu_ops.flush_tlb_others = xen_flush_tlb_others; + return; + } + pv_mmu_ops = xen_mmu_ops; + + memset(dummy_mapping, 0xff, PAGE_SIZE); +} + +/* Protected by xen_reservation_lock. */ +#define MAX_CONTIG_ORDER 9 /* 2MB */ +static unsigned long discontig_frames[1<<MAX_CONTIG_ORDER]; + +#define VOID_PTE (mfn_pte(0, __pgprot(0))) +static void xen_zap_pfn_range(unsigned long vaddr, unsigned int order, + unsigned long *in_frames, + unsigned long *out_frames) +{ + int i; + struct multicall_space mcs; + + xen_mc_batch(); + for (i = 0; i < (1UL<<order); i++, vaddr += PAGE_SIZE) { + mcs = __xen_mc_entry(0); + + if (in_frames) + in_frames[i] = virt_to_mfn(vaddr); + + MULTI_update_va_mapping(mcs.mc, vaddr, VOID_PTE, 0); + __set_phys_to_machine(virt_to_pfn(vaddr), INVALID_P2M_ENTRY); + + if (out_frames) + out_frames[i] = virt_to_pfn(vaddr); + } + xen_mc_issue(0); +} + +/* + * Update the pfn-to-mfn mappings for a virtual address range, either to + * point to an array of mfns, or contiguously from a single starting + * mfn. + */ +static void xen_remap_exchanged_ptes(unsigned long vaddr, int order, + unsigned long *mfns, + unsigned long first_mfn) +{ + unsigned i, limit; + unsigned long mfn; + + xen_mc_batch(); + + limit = 1u << order; + for (i = 0; i < limit; i++, vaddr += PAGE_SIZE) { + struct multicall_space mcs; + unsigned flags; + + mcs = __xen_mc_entry(0); + if (mfns) + mfn = mfns[i]; + else + mfn = first_mfn + i; + + if (i < (limit - 1)) + flags = 0; + else { + if (order == 0) + flags = UVMF_INVLPG | UVMF_ALL; + else + flags = UVMF_TLB_FLUSH | UVMF_ALL; + } + + MULTI_update_va_mapping(mcs.mc, vaddr, + mfn_pte(mfn, PAGE_KERNEL), flags); + + set_phys_to_machine(virt_to_pfn(vaddr), mfn); + } + + xen_mc_issue(0); +} + +/* + * Perform the hypercall to exchange a region of our pfns to point to + * memory with the required contiguous alignment. Takes the pfns as + * input, and populates mfns as output. + * + * Returns a success code indicating whether the hypervisor was able to + * satisfy the request or not. + */ +static int xen_exchange_memory(unsigned long extents_in, unsigned int order_in, + unsigned long *pfns_in, + unsigned long extents_out, + unsigned int order_out, + unsigned long *mfns_out, + unsigned int address_bits) +{ + long rc; + int success; + + struct xen_memory_exchange exchange = { + .in = { + .nr_extents = extents_in, + .extent_order = order_in, + .extent_start = pfns_in, + .domid = DOMID_SELF + }, + .out = { + .nr_extents = extents_out, + .extent_order = order_out, + .extent_start = mfns_out, + .address_bits = address_bits, + .domid = DOMID_SELF + } + }; + + BUG_ON(extents_in << order_in != extents_out << order_out); + + rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange); + success = (exchange.nr_exchanged == extents_in); + + BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0))); + BUG_ON(success && (rc != 0)); + + return success; +} + +int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order, + unsigned int address_bits, + dma_addr_t *dma_handle) +{ + unsigned long *in_frames = discontig_frames, out_frame; + unsigned long flags; + int success; + unsigned long vstart = (unsigned long)phys_to_virt(pstart); + + /* + * Currently an auto-translated guest will not perform I/O, nor will + * it require PAE page directories below 4GB. Therefore any calls to + * this function are redundant and can be ignored. + */ + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return 0; + + if (unlikely(order > MAX_CONTIG_ORDER)) + return -ENOMEM; + + memset((void *) vstart, 0, PAGE_SIZE << order); + + spin_lock_irqsave(&xen_reservation_lock, flags); + + /* 1. Zap current PTEs, remembering MFNs. */ + xen_zap_pfn_range(vstart, order, in_frames, NULL); + + /* 2. Get a new contiguous memory extent. */ + out_frame = virt_to_pfn(vstart); + success = xen_exchange_memory(1UL << order, 0, in_frames, + 1, order, &out_frame, + address_bits); + + /* 3. Map the new extent in place of old pages. */ + if (success) + xen_remap_exchanged_ptes(vstart, order, NULL, out_frame); + else + xen_remap_exchanged_ptes(vstart, order, in_frames, 0); + + spin_unlock_irqrestore(&xen_reservation_lock, flags); + + *dma_handle = virt_to_machine(vstart).maddr; + return success ? 0 : -ENOMEM; +} +EXPORT_SYMBOL_GPL(xen_create_contiguous_region); + +void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order) +{ + unsigned long *out_frames = discontig_frames, in_frame; + unsigned long flags; + int success; + unsigned long vstart; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + if (unlikely(order > MAX_CONTIG_ORDER)) + return; + + vstart = (unsigned long)phys_to_virt(pstart); + memset((void *) vstart, 0, PAGE_SIZE << order); + + spin_lock_irqsave(&xen_reservation_lock, flags); + + /* 1. Find start MFN of contiguous extent. */ + in_frame = virt_to_mfn(vstart); + + /* 2. Zap current PTEs. */ + xen_zap_pfn_range(vstart, order, NULL, out_frames); + + /* 3. Do the exchange for non-contiguous MFNs. */ + success = xen_exchange_memory(1, order, &in_frame, 1UL << order, + 0, out_frames, 0); + + /* 4. Map new pages in place of old pages. */ + if (success) + xen_remap_exchanged_ptes(vstart, order, out_frames, 0); + else + xen_remap_exchanged_ptes(vstart, order, NULL, in_frame); + + spin_unlock_irqrestore(&xen_reservation_lock, flags); +} +EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region); + +#ifdef CONFIG_XEN_PVHVM +#ifdef CONFIG_PROC_VMCORE +/* + * This function is used in two contexts: + * - the kdump kernel has to check whether a pfn of the crashed kernel + * was a ballooned page. vmcore is using this function to decide + * whether to access a pfn of the crashed kernel. + * - the kexec kernel has to check whether a pfn was ballooned by the + * previous kernel. If the pfn is ballooned, handle it properly. + * Returns 0 if the pfn is not backed by a RAM page, the caller may + * handle the pfn special in this case. + */ +static int xen_oldmem_pfn_is_ram(unsigned long pfn) +{ + struct xen_hvm_get_mem_type a = { + .domid = DOMID_SELF, + .pfn = pfn, + }; + int ram; + + if (HYPERVISOR_hvm_op(HVMOP_get_mem_type, &a)) + return -ENXIO; + + switch (a.mem_type) { + case HVMMEM_mmio_dm: + ram = 0; + break; + case HVMMEM_ram_rw: + case HVMMEM_ram_ro: + default: + ram = 1; + break; + } + + return ram; +} +#endif + +static void xen_hvm_exit_mmap(struct mm_struct *mm) +{ + struct xen_hvm_pagetable_dying a; + int rc; + + a.domid = DOMID_SELF; + a.gpa = __pa(mm->pgd); + rc = HYPERVISOR_hvm_op(HVMOP_pagetable_dying, &a); + WARN_ON_ONCE(rc < 0); +} + +static int is_pagetable_dying_supported(void) +{ + struct xen_hvm_pagetable_dying a; + int rc = 0; + + a.domid = DOMID_SELF; + a.gpa = 0x00; + rc = HYPERVISOR_hvm_op(HVMOP_pagetable_dying, &a); + if (rc < 0) { + printk(KERN_DEBUG "HVMOP_pagetable_dying not supported\n"); + return 0; + } + return 1; +} + +void __init xen_hvm_init_mmu_ops(void) +{ + if (is_pagetable_dying_supported()) + pv_mmu_ops.exit_mmap = xen_hvm_exit_mmap; +#ifdef CONFIG_PROC_VMCORE + register_oldmem_pfn_is_ram(&xen_oldmem_pfn_is_ram); +#endif +} +#endif + +#define REMAP_BATCH_SIZE 16 + +struct remap_data { + xen_pfn_t *mfn; + bool contiguous; + pgprot_t prot; + struct mmu_update *mmu_update; +}; + +static int remap_area_mfn_pte_fn(pte_t *ptep, pgtable_t token, + unsigned long addr, void *data) +{ + struct remap_data *rmd = data; + pte_t pte = pte_mkspecial(mfn_pte(*rmd->mfn, rmd->prot)); + + /* If we have a contigious range, just update the mfn itself, + else update pointer to be "next mfn". */ + if (rmd->contiguous) + (*rmd->mfn)++; + else + rmd->mfn++; + + rmd->mmu_update->ptr = virt_to_machine(ptep).maddr; + rmd->mmu_update->val = pte_val_ma(pte); + rmd->mmu_update++; + + return 0; +} + +static int do_remap_mfn(struct vm_area_struct *vma, + unsigned long addr, + xen_pfn_t *mfn, int nr, + int *err_ptr, pgprot_t prot, + unsigned domid, + struct page **pages) +{ + int err = 0; + struct remap_data rmd; + struct mmu_update mmu_update[REMAP_BATCH_SIZE]; + unsigned long range; + int mapped = 0; + + BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_IO)) == (VM_PFNMAP | VM_IO))); + + if (xen_feature(XENFEAT_auto_translated_physmap)) { +#ifdef CONFIG_XEN_PVH + /* We need to update the local page tables and the xen HAP */ + return xen_xlate_remap_gfn_array(vma, addr, mfn, nr, err_ptr, + prot, domid, pages); +#else + return -EINVAL; +#endif + } + + rmd.mfn = mfn; + rmd.prot = prot; + /* We use the err_ptr to indicate if there we are doing a contigious + * mapping or a discontigious mapping. */ + rmd.contiguous = !err_ptr; + + while (nr) { + int index = 0; + int done = 0; + int batch = min(REMAP_BATCH_SIZE, nr); + int batch_left = batch; + range = (unsigned long)batch << PAGE_SHIFT; + + rmd.mmu_update = mmu_update; + err = apply_to_page_range(vma->vm_mm, addr, range, + remap_area_mfn_pte_fn, &rmd); + if (err) + goto out; + + /* We record the error for each page that gives an error, but + * continue mapping until the whole set is done */ + do { + int i; + + err = HYPERVISOR_mmu_update(&mmu_update[index], + batch_left, &done, domid); + + /* + * @err_ptr may be the same buffer as @mfn, so + * only clear it after each chunk of @mfn is + * used. + */ + if (err_ptr) { + for (i = index; i < index + done; i++) + err_ptr[i] = 0; + } + if (err < 0) { + if (!err_ptr) + goto out; + err_ptr[i] = err; + done++; /* Skip failed frame. */ + } else + mapped += done; + batch_left -= done; + index += done; + } while (batch_left); + + nr -= batch; + addr += range; + if (err_ptr) + err_ptr += batch; + } +out: + + xen_flush_tlb_all(); + + return err < 0 ? err : mapped; +} + +int xen_remap_domain_mfn_range(struct vm_area_struct *vma, + unsigned long addr, + xen_pfn_t mfn, int nr, + pgprot_t prot, unsigned domid, + struct page **pages) +{ + return do_remap_mfn(vma, addr, &mfn, nr, NULL, prot, domid, pages); +} +EXPORT_SYMBOL_GPL(xen_remap_domain_mfn_range); + +int xen_remap_domain_mfn_array(struct vm_area_struct *vma, + unsigned long addr, + xen_pfn_t *mfn, int nr, + int *err_ptr, pgprot_t prot, + unsigned domid, struct page **pages) +{ + /* We BUG_ON because it's a programmer error to pass a NULL err_ptr, + * and the consequences later is quite hard to detect what the actual + * cause of "wrong memory was mapped in". + */ + BUG_ON(err_ptr == NULL); + return do_remap_mfn(vma, addr, mfn, nr, err_ptr, prot, domid, pages); +} +EXPORT_SYMBOL_GPL(xen_remap_domain_mfn_array); + + +/* Returns: 0 success */ +int xen_unmap_domain_mfn_range(struct vm_area_struct *vma, + int numpgs, struct page **pages) +{ + if (!pages || !xen_feature(XENFEAT_auto_translated_physmap)) + return 0; + +#ifdef CONFIG_XEN_PVH + return xen_xlate_unmap_gfn_range(vma, numpgs, pages); +#else + return -EINVAL; +#endif +} +EXPORT_SYMBOL_GPL(xen_unmap_domain_mfn_range); diff --git a/kernel/arch/x86/xen/mmu.h b/kernel/arch/x86/xen/mmu.h new file mode 100644 index 000000000..73809bb95 --- /dev/null +++ b/kernel/arch/x86/xen/mmu.h @@ -0,0 +1,26 @@ +#ifndef _XEN_MMU_H + +#include <linux/linkage.h> +#include <asm/page.h> + +enum pt_level { + PT_PGD, + PT_PUD, + PT_PMD, + PT_PTE +}; + + +bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn); + +void set_pte_mfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags); + +pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr, pte_t *ptep); +void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte); + +unsigned long xen_read_cr2_direct(void); + +extern void xen_init_mmu_ops(void); +extern void xen_hvm_init_mmu_ops(void); +#endif /* _XEN_MMU_H */ diff --git a/kernel/arch/x86/xen/multicalls.c b/kernel/arch/x86/xen/multicalls.c new file mode 100644 index 000000000..ea54a08d8 --- /dev/null +++ b/kernel/arch/x86/xen/multicalls.c @@ -0,0 +1,208 @@ +/* + * Xen hypercall batching. + * + * Xen allows multiple hypercalls to be issued at once, using the + * multicall interface. This allows the cost of trapping into the + * hypervisor to be amortized over several calls. + * + * This file implements a simple interface for multicalls. There's a + * per-cpu buffer of outstanding multicalls. When you want to queue a + * multicall for issuing, you can allocate a multicall slot for the + * call and its arguments, along with storage for space which is + * pointed to by the arguments (for passing pointers to structures, + * etc). When the multicall is actually issued, all the space for the + * commands and allocated memory is freed for reuse. + * + * Multicalls are flushed whenever any of the buffers get full, or + * when explicitly requested. There's no way to get per-multicall + * return results back. It will BUG if any of the multicalls fail. + * + * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007 + */ +#include <linux/percpu.h> +#include <linux/hardirq.h> +#include <linux/debugfs.h> + +#include <asm/xen/hypercall.h> + +#include "multicalls.h" +#include "debugfs.h" + +#define MC_BATCH 32 + +#define MC_DEBUG 0 + +#define MC_ARGS (MC_BATCH * 16) + + +struct mc_buffer { + unsigned mcidx, argidx, cbidx; + struct multicall_entry entries[MC_BATCH]; +#if MC_DEBUG + struct multicall_entry debug[MC_BATCH]; + void *caller[MC_BATCH]; +#endif + unsigned char args[MC_ARGS]; + struct callback { + void (*fn)(void *); + void *data; + } callbacks[MC_BATCH]; +}; + +static DEFINE_PER_CPU(struct mc_buffer, mc_buffer); +DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags); + +void xen_mc_flush(void) +{ + struct mc_buffer *b = this_cpu_ptr(&mc_buffer); + struct multicall_entry *mc; + int ret = 0; + unsigned long flags; + int i; + + BUG_ON(preemptible()); + + /* Disable interrupts in case someone comes in and queues + something in the middle */ + local_irq_save(flags); + + trace_xen_mc_flush(b->mcidx, b->argidx, b->cbidx); + + switch (b->mcidx) { + case 0: + /* no-op */ + BUG_ON(b->argidx != 0); + break; + + case 1: + /* Singleton multicall - bypass multicall machinery + and just do the call directly. */ + mc = &b->entries[0]; + + mc->result = privcmd_call(mc->op, + mc->args[0], mc->args[1], mc->args[2], + mc->args[3], mc->args[4]); + ret = mc->result < 0; + break; + + default: +#if MC_DEBUG + memcpy(b->debug, b->entries, + b->mcidx * sizeof(struct multicall_entry)); +#endif + + if (HYPERVISOR_multicall(b->entries, b->mcidx) != 0) + BUG(); + for (i = 0; i < b->mcidx; i++) + if (b->entries[i].result < 0) + ret++; + +#if MC_DEBUG + if (ret) { + printk(KERN_ERR "%d multicall(s) failed: cpu %d\n", + ret, smp_processor_id()); + dump_stack(); + for (i = 0; i < b->mcidx; i++) { + printk(KERN_DEBUG " call %2d/%d: op=%lu arg=[%lx] result=%ld\t%pF\n", + i+1, b->mcidx, + b->debug[i].op, + b->debug[i].args[0], + b->entries[i].result, + b->caller[i]); + } + } +#endif + } + + b->mcidx = 0; + b->argidx = 0; + + for (i = 0; i < b->cbidx; i++) { + struct callback *cb = &b->callbacks[i]; + + (*cb->fn)(cb->data); + } + b->cbidx = 0; + + local_irq_restore(flags); + + WARN_ON(ret); +} + +struct multicall_space __xen_mc_entry(size_t args) +{ + struct mc_buffer *b = this_cpu_ptr(&mc_buffer); + struct multicall_space ret; + unsigned argidx = roundup(b->argidx, sizeof(u64)); + + trace_xen_mc_entry_alloc(args); + + BUG_ON(preemptible()); + BUG_ON(b->argidx >= MC_ARGS); + + if (unlikely(b->mcidx == MC_BATCH || + (argidx + args) >= MC_ARGS)) { + trace_xen_mc_flush_reason((b->mcidx == MC_BATCH) ? + XEN_MC_FL_BATCH : XEN_MC_FL_ARGS); + xen_mc_flush(); + argidx = roundup(b->argidx, sizeof(u64)); + } + + ret.mc = &b->entries[b->mcidx]; +#if MC_DEBUG + b->caller[b->mcidx] = __builtin_return_address(0); +#endif + b->mcidx++; + ret.args = &b->args[argidx]; + b->argidx = argidx + args; + + BUG_ON(b->argidx >= MC_ARGS); + return ret; +} + +struct multicall_space xen_mc_extend_args(unsigned long op, size_t size) +{ + struct mc_buffer *b = this_cpu_ptr(&mc_buffer); + struct multicall_space ret = { NULL, NULL }; + + BUG_ON(preemptible()); + BUG_ON(b->argidx >= MC_ARGS); + + if (unlikely(b->mcidx == 0 || + b->entries[b->mcidx - 1].op != op)) { + trace_xen_mc_extend_args(op, size, XEN_MC_XE_BAD_OP); + goto out; + } + + if (unlikely((b->argidx + size) >= MC_ARGS)) { + trace_xen_mc_extend_args(op, size, XEN_MC_XE_NO_SPACE); + goto out; + } + + ret.mc = &b->entries[b->mcidx - 1]; + ret.args = &b->args[b->argidx]; + b->argidx += size; + + BUG_ON(b->argidx >= MC_ARGS); + + trace_xen_mc_extend_args(op, size, XEN_MC_XE_OK); +out: + return ret; +} + +void xen_mc_callback(void (*fn)(void *), void *data) +{ + struct mc_buffer *b = this_cpu_ptr(&mc_buffer); + struct callback *cb; + + if (b->cbidx == MC_BATCH) { + trace_xen_mc_flush_reason(XEN_MC_FL_CALLBACK); + xen_mc_flush(); + } + + trace_xen_mc_callback(fn, data); + + cb = &b->callbacks[b->cbidx++]; + cb->fn = fn; + cb->data = data; +} diff --git a/kernel/arch/x86/xen/multicalls.h b/kernel/arch/x86/xen/multicalls.h new file mode 100644 index 000000000..9c2e74f90 --- /dev/null +++ b/kernel/arch/x86/xen/multicalls.h @@ -0,0 +1,68 @@ +#ifndef _XEN_MULTICALLS_H +#define _XEN_MULTICALLS_H + +#include <trace/events/xen.h> + +#include "xen-ops.h" + +/* Multicalls */ +struct multicall_space +{ + struct multicall_entry *mc; + void *args; +}; + +/* Allocate room for a multicall and its args */ +struct multicall_space __xen_mc_entry(size_t args); + +DECLARE_PER_CPU(unsigned long, xen_mc_irq_flags); + +/* Call to start a batch of multiple __xen_mc_entry()s. Must be + paired with xen_mc_issue() */ +static inline void xen_mc_batch(void) +{ + unsigned long flags; + + /* need to disable interrupts until this entry is complete */ + local_irq_save(flags); + trace_xen_mc_batch(paravirt_get_lazy_mode()); + __this_cpu_write(xen_mc_irq_flags, flags); +} + +static inline struct multicall_space xen_mc_entry(size_t args) +{ + xen_mc_batch(); + return __xen_mc_entry(args); +} + +/* Flush all pending multicalls */ +void xen_mc_flush(void); + +/* Issue a multicall if we're not in a lazy mode */ +static inline void xen_mc_issue(unsigned mode) +{ + trace_xen_mc_issue(mode); + + if ((paravirt_get_lazy_mode() & mode) == 0) + xen_mc_flush(); + + /* restore flags saved in xen_mc_batch */ + local_irq_restore(this_cpu_read(xen_mc_irq_flags)); +} + +/* Set up a callback to be called when the current batch is flushed */ +void xen_mc_callback(void (*fn)(void *), void *data); + +/* + * Try to extend the arguments of the previous multicall command. The + * previous command's op must match. If it does, then it attempts to + * extend the argument space allocated to the multicall entry by + * arg_size bytes. + * + * The returned multicall_space will return with mc pointing to the + * command on success, or NULL on failure, and args pointing to the + * newly allocated space. + */ +struct multicall_space xen_mc_extend_args(unsigned long op, size_t arg_size); + +#endif /* _XEN_MULTICALLS_H */ diff --git a/kernel/arch/x86/xen/p2m.c b/kernel/arch/x86/xen/p2m.c new file mode 100644 index 000000000..b47124d4c --- /dev/null +++ b/kernel/arch/x86/xen/p2m.c @@ -0,0 +1,790 @@ +/* + * Xen leaves the responsibility for maintaining p2m mappings to the + * guests themselves, but it must also access and update the p2m array + * during suspend/resume when all the pages are reallocated. + * + * The logical flat p2m table is mapped to a linear kernel memory area. + * For accesses by Xen a three-level tree linked via mfns only is set up to + * allow the address space to be sparse. + * + * Xen + * | + * p2m_top_mfn + * / \ + * p2m_mid_mfn p2m_mid_mfn + * / / + * p2m p2m p2m ... + * + * The p2m_mid_mfn pages are mapped by p2m_top_mfn_p. + * + * The p2m_top_mfn level is limited to 1 page, so the maximum representable + * pseudo-physical address space is: + * P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE pages + * + * P2M_PER_PAGE depends on the architecture, as a mfn is always + * unsigned long (8 bytes on 64-bit, 4 bytes on 32), leading to + * 512 and 1024 entries respectively. + * + * In short, these structures contain the Machine Frame Number (MFN) of the PFN. + * + * However not all entries are filled with MFNs. Specifically for all other + * leaf entries, or for the top root, or middle one, for which there is a void + * entry, we assume it is "missing". So (for example) + * pfn_to_mfn(0x90909090)=INVALID_P2M_ENTRY. + * We have a dedicated page p2m_missing with all entries being + * INVALID_P2M_ENTRY. This page may be referenced multiple times in the p2m + * list/tree in case there are multiple areas with P2M_PER_PAGE invalid pfns. + * + * We also have the possibility of setting 1-1 mappings on certain regions, so + * that: + * pfn_to_mfn(0xc0000)=0xc0000 + * + * The benefit of this is, that we can assume for non-RAM regions (think + * PCI BARs, or ACPI spaces), we can create mappings easily because we + * get the PFN value to match the MFN. + * + * For this to work efficiently we have one new page p2m_identity. All entries + * in p2m_identity are set to INVALID_P2M_ENTRY type (Xen toolstack only + * recognizes that and MFNs, no other fancy value). + * + * On lookup we spot that the entry points to p2m_identity and return the + * identity value instead of dereferencing and returning INVALID_P2M_ENTRY. + * If the entry points to an allocated page, we just proceed as before and + * return the PFN. If the PFN has IDENTITY_FRAME_BIT set we unmask that in + * appropriate functions (pfn_to_mfn). + * + * The reason for having the IDENTITY_FRAME_BIT instead of just returning the + * PFN is that we could find ourselves where pfn_to_mfn(pfn)==pfn for a + * non-identity pfn. To protect ourselves against we elect to set (and get) the + * IDENTITY_FRAME_BIT on all identity mapped PFNs. + */ + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/list.h> +#include <linux/hash.h> +#include <linux/sched.h> +#include <linux/seq_file.h> +#include <linux/bootmem.h> +#include <linux/slab.h> + +#include <asm/cache.h> +#include <asm/setup.h> +#include <asm/uaccess.h> + +#include <asm/xen/page.h> +#include <asm/xen/hypercall.h> +#include <asm/xen/hypervisor.h> +#include <xen/balloon.h> +#include <xen/grant_table.h> + +#include "p2m.h" +#include "multicalls.h" +#include "xen-ops.h" + +#define PMDS_PER_MID_PAGE (P2M_MID_PER_PAGE / PTRS_PER_PTE) + +unsigned long *xen_p2m_addr __read_mostly; +EXPORT_SYMBOL_GPL(xen_p2m_addr); +unsigned long xen_p2m_size __read_mostly; +EXPORT_SYMBOL_GPL(xen_p2m_size); +unsigned long xen_max_p2m_pfn __read_mostly; +EXPORT_SYMBOL_GPL(xen_max_p2m_pfn); + +#ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG_LIMIT +#define P2M_LIMIT CONFIG_XEN_BALLOON_MEMORY_HOTPLUG_LIMIT +#else +#define P2M_LIMIT 0 +#endif + +static DEFINE_SPINLOCK(p2m_update_lock); + +static unsigned long *p2m_mid_missing_mfn; +static unsigned long *p2m_top_mfn; +static unsigned long **p2m_top_mfn_p; +static unsigned long *p2m_missing; +static unsigned long *p2m_identity; +static pte_t *p2m_missing_pte; +static pte_t *p2m_identity_pte; + +static inline unsigned p2m_top_index(unsigned long pfn) +{ + BUG_ON(pfn >= MAX_P2M_PFN); + return pfn / (P2M_MID_PER_PAGE * P2M_PER_PAGE); +} + +static inline unsigned p2m_mid_index(unsigned long pfn) +{ + return (pfn / P2M_PER_PAGE) % P2M_MID_PER_PAGE; +} + +static inline unsigned p2m_index(unsigned long pfn) +{ + return pfn % P2M_PER_PAGE; +} + +static void p2m_top_mfn_init(unsigned long *top) +{ + unsigned i; + + for (i = 0; i < P2M_TOP_PER_PAGE; i++) + top[i] = virt_to_mfn(p2m_mid_missing_mfn); +} + +static void p2m_top_mfn_p_init(unsigned long **top) +{ + unsigned i; + + for (i = 0; i < P2M_TOP_PER_PAGE; i++) + top[i] = p2m_mid_missing_mfn; +} + +static void p2m_mid_mfn_init(unsigned long *mid, unsigned long *leaf) +{ + unsigned i; + + for (i = 0; i < P2M_MID_PER_PAGE; i++) + mid[i] = virt_to_mfn(leaf); +} + +static void p2m_init(unsigned long *p2m) +{ + unsigned i; + + for (i = 0; i < P2M_PER_PAGE; i++) + p2m[i] = INVALID_P2M_ENTRY; +} + +static void p2m_init_identity(unsigned long *p2m, unsigned long pfn) +{ + unsigned i; + + for (i = 0; i < P2M_PER_PAGE; i++) + p2m[i] = IDENTITY_FRAME(pfn + i); +} + +static void * __ref alloc_p2m_page(void) +{ + if (unlikely(!slab_is_available())) + return alloc_bootmem_align(PAGE_SIZE, PAGE_SIZE); + + return (void *)__get_free_page(GFP_KERNEL | __GFP_REPEAT); +} + +static void __ref free_p2m_page(void *p) +{ + if (unlikely(!slab_is_available())) { + free_bootmem((unsigned long)p, PAGE_SIZE); + return; + } + + free_page((unsigned long)p); +} + +/* + * Build the parallel p2m_top_mfn and p2m_mid_mfn structures + * + * This is called both at boot time, and after resuming from suspend: + * - At boot time we're called rather early, and must use alloc_bootmem*() + * to allocate memory. + * + * - After resume we're called from within stop_machine, but the mfn + * tree should already be completely allocated. + */ +void __ref xen_build_mfn_list_list(void) +{ + unsigned long pfn, mfn; + pte_t *ptep; + unsigned int level, topidx, mididx; + unsigned long *mid_mfn_p; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + /* Pre-initialize p2m_top_mfn to be completely missing */ + if (p2m_top_mfn == NULL) { + p2m_mid_missing_mfn = alloc_p2m_page(); + p2m_mid_mfn_init(p2m_mid_missing_mfn, p2m_missing); + + p2m_top_mfn_p = alloc_p2m_page(); + p2m_top_mfn_p_init(p2m_top_mfn_p); + + p2m_top_mfn = alloc_p2m_page(); + p2m_top_mfn_init(p2m_top_mfn); + } else { + /* Reinitialise, mfn's all change after migration */ + p2m_mid_mfn_init(p2m_mid_missing_mfn, p2m_missing); + } + + for (pfn = 0; pfn < xen_max_p2m_pfn && pfn < MAX_P2M_PFN; + pfn += P2M_PER_PAGE) { + topidx = p2m_top_index(pfn); + mididx = p2m_mid_index(pfn); + + mid_mfn_p = p2m_top_mfn_p[topidx]; + ptep = lookup_address((unsigned long)(xen_p2m_addr + pfn), + &level); + BUG_ON(!ptep || level != PG_LEVEL_4K); + mfn = pte_mfn(*ptep); + ptep = (pte_t *)((unsigned long)ptep & ~(PAGE_SIZE - 1)); + + /* Don't bother allocating any mfn mid levels if + * they're just missing, just update the stored mfn, + * since all could have changed over a migrate. + */ + if (ptep == p2m_missing_pte || ptep == p2m_identity_pte) { + BUG_ON(mididx); + BUG_ON(mid_mfn_p != p2m_mid_missing_mfn); + p2m_top_mfn[topidx] = virt_to_mfn(p2m_mid_missing_mfn); + pfn += (P2M_MID_PER_PAGE - 1) * P2M_PER_PAGE; + continue; + } + + if (mid_mfn_p == p2m_mid_missing_mfn) { + mid_mfn_p = alloc_p2m_page(); + p2m_mid_mfn_init(mid_mfn_p, p2m_missing); + + p2m_top_mfn_p[topidx] = mid_mfn_p; + } + + p2m_top_mfn[topidx] = virt_to_mfn(mid_mfn_p); + mid_mfn_p[mididx] = mfn; + } +} + +void xen_setup_mfn_list_list(void) +{ + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info); + + HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list = + virt_to_mfn(p2m_top_mfn); + HYPERVISOR_shared_info->arch.max_pfn = xen_max_p2m_pfn; +} + +/* Set up p2m_top to point to the domain-builder provided p2m pages */ +void __init xen_build_dynamic_phys_to_machine(void) +{ + unsigned long pfn; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return; + + xen_p2m_addr = (unsigned long *)xen_start_info->mfn_list; + xen_p2m_size = ALIGN(xen_start_info->nr_pages, P2M_PER_PAGE); + + for (pfn = xen_start_info->nr_pages; pfn < xen_p2m_size; pfn++) + xen_p2m_addr[pfn] = INVALID_P2M_ENTRY; + + xen_max_p2m_pfn = xen_p2m_size; +} + +#define P2M_TYPE_IDENTITY 0 +#define P2M_TYPE_MISSING 1 +#define P2M_TYPE_PFN 2 +#define P2M_TYPE_UNKNOWN 3 + +static int xen_p2m_elem_type(unsigned long pfn) +{ + unsigned long mfn; + + if (pfn >= xen_p2m_size) + return P2M_TYPE_IDENTITY; + + mfn = xen_p2m_addr[pfn]; + + if (mfn == INVALID_P2M_ENTRY) + return P2M_TYPE_MISSING; + + if (mfn & IDENTITY_FRAME_BIT) + return P2M_TYPE_IDENTITY; + + return P2M_TYPE_PFN; +} + +static void __init xen_rebuild_p2m_list(unsigned long *p2m) +{ + unsigned int i, chunk; + unsigned long pfn; + unsigned long *mfns; + pte_t *ptep; + pmd_t *pmdp; + int type; + + p2m_missing = alloc_p2m_page(); + p2m_init(p2m_missing); + p2m_identity = alloc_p2m_page(); + p2m_init(p2m_identity); + + p2m_missing_pte = alloc_p2m_page(); + paravirt_alloc_pte(&init_mm, __pa(p2m_missing_pte) >> PAGE_SHIFT); + p2m_identity_pte = alloc_p2m_page(); + paravirt_alloc_pte(&init_mm, __pa(p2m_identity_pte) >> PAGE_SHIFT); + for (i = 0; i < PTRS_PER_PTE; i++) { + set_pte(p2m_missing_pte + i, + pfn_pte(PFN_DOWN(__pa(p2m_missing)), PAGE_KERNEL_RO)); + set_pte(p2m_identity_pte + i, + pfn_pte(PFN_DOWN(__pa(p2m_identity)), PAGE_KERNEL_RO)); + } + + for (pfn = 0; pfn < xen_max_p2m_pfn; pfn += chunk) { + /* + * Try to map missing/identity PMDs or p2m-pages if possible. + * We have to respect the structure of the mfn_list_list + * which will be built just afterwards. + * Chunk size to test is one p2m page if we are in the middle + * of a mfn_list_list mid page and the complete mid page area + * if we are at index 0 of the mid page. Please note that a + * mid page might cover more than one PMD, e.g. on 32 bit PAE + * kernels. + */ + chunk = (pfn & (P2M_PER_PAGE * P2M_MID_PER_PAGE - 1)) ? + P2M_PER_PAGE : P2M_PER_PAGE * P2M_MID_PER_PAGE; + + type = xen_p2m_elem_type(pfn); + i = 0; + if (type != P2M_TYPE_PFN) + for (i = 1; i < chunk; i++) + if (xen_p2m_elem_type(pfn + i) != type) + break; + if (i < chunk) + /* Reset to minimal chunk size. */ + chunk = P2M_PER_PAGE; + + if (type == P2M_TYPE_PFN || i < chunk) { + /* Use initial p2m page contents. */ +#ifdef CONFIG_X86_64 + mfns = alloc_p2m_page(); + copy_page(mfns, xen_p2m_addr + pfn); +#else + mfns = xen_p2m_addr + pfn; +#endif + ptep = populate_extra_pte((unsigned long)(p2m + pfn)); + set_pte(ptep, + pfn_pte(PFN_DOWN(__pa(mfns)), PAGE_KERNEL)); + continue; + } + + if (chunk == P2M_PER_PAGE) { + /* Map complete missing or identity p2m-page. */ + mfns = (type == P2M_TYPE_MISSING) ? + p2m_missing : p2m_identity; + ptep = populate_extra_pte((unsigned long)(p2m + pfn)); + set_pte(ptep, + pfn_pte(PFN_DOWN(__pa(mfns)), PAGE_KERNEL_RO)); + continue; + } + + /* Complete missing or identity PMD(s) can be mapped. */ + ptep = (type == P2M_TYPE_MISSING) ? + p2m_missing_pte : p2m_identity_pte; + for (i = 0; i < PMDS_PER_MID_PAGE; i++) { + pmdp = populate_extra_pmd( + (unsigned long)(p2m + pfn) + i * PMD_SIZE); + set_pmd(pmdp, __pmd(__pa(ptep) | _KERNPG_TABLE)); + } + } +} + +void __init xen_vmalloc_p2m_tree(void) +{ + static struct vm_struct vm; + unsigned long p2m_limit; + + p2m_limit = (phys_addr_t)P2M_LIMIT * 1024 * 1024 * 1024 / PAGE_SIZE; + vm.flags = VM_ALLOC; + vm.size = ALIGN(sizeof(unsigned long) * max(xen_max_p2m_pfn, p2m_limit), + PMD_SIZE * PMDS_PER_MID_PAGE); + vm_area_register_early(&vm, PMD_SIZE * PMDS_PER_MID_PAGE); + pr_notice("p2m virtual area at %p, size is %lx\n", vm.addr, vm.size); + + xen_max_p2m_pfn = vm.size / sizeof(unsigned long); + + xen_rebuild_p2m_list(vm.addr); + + xen_p2m_addr = vm.addr; + xen_p2m_size = xen_max_p2m_pfn; + + xen_inv_extra_mem(); +} + +unsigned long get_phys_to_machine(unsigned long pfn) +{ + pte_t *ptep; + unsigned int level; + + if (unlikely(pfn >= xen_p2m_size)) { + if (pfn < xen_max_p2m_pfn) + return xen_chk_extra_mem(pfn); + + return IDENTITY_FRAME(pfn); + } + + ptep = lookup_address((unsigned long)(xen_p2m_addr + pfn), &level); + BUG_ON(!ptep || level != PG_LEVEL_4K); + + /* + * The INVALID_P2M_ENTRY is filled in both p2m_*identity + * and in p2m_*missing, so returning the INVALID_P2M_ENTRY + * would be wrong. + */ + if (pte_pfn(*ptep) == PFN_DOWN(__pa(p2m_identity))) + return IDENTITY_FRAME(pfn); + + return xen_p2m_addr[pfn]; +} +EXPORT_SYMBOL_GPL(get_phys_to_machine); + +/* + * Allocate new pmd(s). It is checked whether the old pmd is still in place. + * If not, nothing is changed. This is okay as the only reason for allocating + * a new pmd is to replace p2m_missing_pte or p2m_identity_pte by a individual + * pmd. In case of PAE/x86-32 there are multiple pmds to allocate! + */ +static pte_t *alloc_p2m_pmd(unsigned long addr, pte_t *pte_pg) +{ + pte_t *ptechk; + pte_t *pte_newpg[PMDS_PER_MID_PAGE]; + pmd_t *pmdp; + unsigned int level; + unsigned long flags; + unsigned long vaddr; + int i; + + /* Do all allocations first to bail out in error case. */ + for (i = 0; i < PMDS_PER_MID_PAGE; i++) { + pte_newpg[i] = alloc_p2m_page(); + if (!pte_newpg[i]) { + for (i--; i >= 0; i--) + free_p2m_page(pte_newpg[i]); + + return NULL; + } + } + + vaddr = addr & ~(PMD_SIZE * PMDS_PER_MID_PAGE - 1); + + for (i = 0; i < PMDS_PER_MID_PAGE; i++) { + copy_page(pte_newpg[i], pte_pg); + paravirt_alloc_pte(&init_mm, __pa(pte_newpg[i]) >> PAGE_SHIFT); + + pmdp = lookup_pmd_address(vaddr); + BUG_ON(!pmdp); + + spin_lock_irqsave(&p2m_update_lock, flags); + + ptechk = lookup_address(vaddr, &level); + if (ptechk == pte_pg) { + set_pmd(pmdp, + __pmd(__pa(pte_newpg[i]) | _KERNPG_TABLE)); + pte_newpg[i] = NULL; + } + + spin_unlock_irqrestore(&p2m_update_lock, flags); + + if (pte_newpg[i]) { + paravirt_release_pte(__pa(pte_newpg[i]) >> PAGE_SHIFT); + free_p2m_page(pte_newpg[i]); + } + + vaddr += PMD_SIZE; + } + + return lookup_address(addr, &level); +} + +/* + * Fully allocate the p2m structure for a given pfn. We need to check + * that both the top and mid levels are allocated, and make sure the + * parallel mfn tree is kept in sync. We may race with other cpus, so + * the new pages are installed with cmpxchg; if we lose the race then + * simply free the page we allocated and use the one that's there. + */ +static bool alloc_p2m(unsigned long pfn) +{ + unsigned topidx, mididx; + unsigned long *top_mfn_p, *mid_mfn; + pte_t *ptep, *pte_pg; + unsigned int level; + unsigned long flags; + unsigned long addr = (unsigned long)(xen_p2m_addr + pfn); + unsigned long p2m_pfn; + + topidx = p2m_top_index(pfn); + mididx = p2m_mid_index(pfn); + + ptep = lookup_address(addr, &level); + BUG_ON(!ptep || level != PG_LEVEL_4K); + pte_pg = (pte_t *)((unsigned long)ptep & ~(PAGE_SIZE - 1)); + + if (pte_pg == p2m_missing_pte || pte_pg == p2m_identity_pte) { + /* PMD level is missing, allocate a new one */ + ptep = alloc_p2m_pmd(addr, pte_pg); + if (!ptep) + return false; + } + + if (p2m_top_mfn) { + top_mfn_p = &p2m_top_mfn[topidx]; + mid_mfn = ACCESS_ONCE(p2m_top_mfn_p[topidx]); + + BUG_ON(virt_to_mfn(mid_mfn) != *top_mfn_p); + + if (mid_mfn == p2m_mid_missing_mfn) { + /* Separately check the mid mfn level */ + unsigned long missing_mfn; + unsigned long mid_mfn_mfn; + unsigned long old_mfn; + + mid_mfn = alloc_p2m_page(); + if (!mid_mfn) + return false; + + p2m_mid_mfn_init(mid_mfn, p2m_missing); + + missing_mfn = virt_to_mfn(p2m_mid_missing_mfn); + mid_mfn_mfn = virt_to_mfn(mid_mfn); + old_mfn = cmpxchg(top_mfn_p, missing_mfn, mid_mfn_mfn); + if (old_mfn != missing_mfn) { + free_p2m_page(mid_mfn); + mid_mfn = mfn_to_virt(old_mfn); + } else { + p2m_top_mfn_p[topidx] = mid_mfn; + } + } + } else { + mid_mfn = NULL; + } + + p2m_pfn = pte_pfn(READ_ONCE(*ptep)); + if (p2m_pfn == PFN_DOWN(__pa(p2m_identity)) || + p2m_pfn == PFN_DOWN(__pa(p2m_missing))) { + /* p2m leaf page is missing */ + unsigned long *p2m; + + p2m = alloc_p2m_page(); + if (!p2m) + return false; + + if (p2m_pfn == PFN_DOWN(__pa(p2m_missing))) + p2m_init(p2m); + else + p2m_init_identity(p2m, pfn & ~(P2M_PER_PAGE - 1)); + + spin_lock_irqsave(&p2m_update_lock, flags); + + if (pte_pfn(*ptep) == p2m_pfn) { + set_pte(ptep, + pfn_pte(PFN_DOWN(__pa(p2m)), PAGE_KERNEL)); + if (mid_mfn) + mid_mfn[mididx] = virt_to_mfn(p2m); + p2m = NULL; + } + + spin_unlock_irqrestore(&p2m_update_lock, flags); + + if (p2m) + free_p2m_page(p2m); + } + + return true; +} + +unsigned long __init set_phys_range_identity(unsigned long pfn_s, + unsigned long pfn_e) +{ + unsigned long pfn; + + if (unlikely(pfn_s >= xen_p2m_size)) + return 0; + + if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) + return pfn_e - pfn_s; + + if (pfn_s > pfn_e) + return 0; + + if (pfn_e > xen_p2m_size) + pfn_e = xen_p2m_size; + + for (pfn = pfn_s; pfn < pfn_e; pfn++) + xen_p2m_addr[pfn] = IDENTITY_FRAME(pfn); + + return pfn - pfn_s; +} + +bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn) +{ + pte_t *ptep; + unsigned int level; + + /* don't track P2M changes in autotranslate guests */ + if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) + return true; + + if (unlikely(pfn >= xen_p2m_size)) { + BUG_ON(mfn != INVALID_P2M_ENTRY); + return true; + } + + if (likely(!xen_safe_write_ulong(xen_p2m_addr + pfn, mfn))) + return true; + + ptep = lookup_address((unsigned long)(xen_p2m_addr + pfn), &level); + BUG_ON(!ptep || level != PG_LEVEL_4K); + + if (pte_pfn(*ptep) == PFN_DOWN(__pa(p2m_missing))) + return mfn == INVALID_P2M_ENTRY; + + if (pte_pfn(*ptep) == PFN_DOWN(__pa(p2m_identity))) + return mfn == IDENTITY_FRAME(pfn); + + return false; +} + +bool set_phys_to_machine(unsigned long pfn, unsigned long mfn) +{ + if (unlikely(!__set_phys_to_machine(pfn, mfn))) { + if (!alloc_p2m(pfn)) + return false; + + return __set_phys_to_machine(pfn, mfn); + } + + return true; +} + +int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops, + struct gnttab_map_grant_ref *kmap_ops, + struct page **pages, unsigned int count) +{ + int i, ret = 0; + pte_t *pte; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return 0; + + if (kmap_ops) { + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, + kmap_ops, count); + if (ret) + goto out; + } + + for (i = 0; i < count; i++) { + unsigned long mfn, pfn; + + /* Do not add to override if the map failed. */ + if (map_ops[i].status) + continue; + + if (map_ops[i].flags & GNTMAP_contains_pte) { + pte = (pte_t *)(mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) + + (map_ops[i].host_addr & ~PAGE_MASK)); + mfn = pte_mfn(*pte); + } else { + mfn = PFN_DOWN(map_ops[i].dev_bus_addr); + } + pfn = page_to_pfn(pages[i]); + + WARN(pfn_to_mfn(pfn) != INVALID_P2M_ENTRY, "page must be ballooned"); + + if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) { + ret = -ENOMEM; + goto out; + } + } + +out: + return ret; +} +EXPORT_SYMBOL_GPL(set_foreign_p2m_mapping); + +int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops, + struct gnttab_unmap_grant_ref *kunmap_ops, + struct page **pages, unsigned int count) +{ + int i, ret = 0; + + if (xen_feature(XENFEAT_auto_translated_physmap)) + return 0; + + for (i = 0; i < count; i++) { + unsigned long mfn = __pfn_to_mfn(page_to_pfn(pages[i])); + unsigned long pfn = page_to_pfn(pages[i]); + + if (mfn == INVALID_P2M_ENTRY || !(mfn & FOREIGN_FRAME_BIT)) { + ret = -EINVAL; + goto out; + } + + set_phys_to_machine(pfn, INVALID_P2M_ENTRY); + } + if (kunmap_ops) + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, + kunmap_ops, count); +out: + return ret; +} +EXPORT_SYMBOL_GPL(clear_foreign_p2m_mapping); + +#ifdef CONFIG_XEN_DEBUG_FS +#include <linux/debugfs.h> +#include "debugfs.h" +static int p2m_dump_show(struct seq_file *m, void *v) +{ + static const char * const type_name[] = { + [P2M_TYPE_IDENTITY] = "identity", + [P2M_TYPE_MISSING] = "missing", + [P2M_TYPE_PFN] = "pfn", + [P2M_TYPE_UNKNOWN] = "abnormal"}; + unsigned long pfn, first_pfn; + int type, prev_type; + + prev_type = xen_p2m_elem_type(0); + first_pfn = 0; + + for (pfn = 0; pfn < xen_p2m_size; pfn++) { + type = xen_p2m_elem_type(pfn); + if (type != prev_type) { + seq_printf(m, " [0x%lx->0x%lx] %s\n", first_pfn, pfn, + type_name[prev_type]); + prev_type = type; + first_pfn = pfn; + } + } + seq_printf(m, " [0x%lx->0x%lx] %s\n", first_pfn, pfn, + type_name[prev_type]); + return 0; +} + +static int p2m_dump_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, p2m_dump_show, NULL); +} + +static const struct file_operations p2m_dump_fops = { + .open = p2m_dump_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static struct dentry *d_mmu_debug; + +static int __init xen_p2m_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + d_mmu_debug = debugfs_create_dir("mmu", d_xen); + + debugfs_create_file("p2m", 0600, d_mmu_debug, NULL, &p2m_dump_fops); + return 0; +} +fs_initcall(xen_p2m_debugfs); +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/kernel/arch/x86/xen/p2m.h b/kernel/arch/x86/xen/p2m.h new file mode 100644 index 000000000..ad8aee24a --- /dev/null +++ b/kernel/arch/x86/xen/p2m.h @@ -0,0 +1,15 @@ +#ifndef _XEN_P2M_H +#define _XEN_P2M_H + +#define P2M_PER_PAGE (PAGE_SIZE / sizeof(unsigned long)) +#define P2M_MID_PER_PAGE (PAGE_SIZE / sizeof(unsigned long *)) +#define P2M_TOP_PER_PAGE (PAGE_SIZE / sizeof(unsigned long **)) + +#define MAX_P2M_PFN (P2M_TOP_PER_PAGE * P2M_MID_PER_PAGE * P2M_PER_PAGE) + +#define MAX_REMAP_RANGES 10 + +extern unsigned long __init set_phys_range_identity(unsigned long pfn_s, + unsigned long pfn_e); + +#endif /* _XEN_P2M_H */ diff --git a/kernel/arch/x86/xen/pci-swiotlb-xen.c b/kernel/arch/x86/xen/pci-swiotlb-xen.c new file mode 100644 index 000000000..0e98e5d24 --- /dev/null +++ b/kernel/arch/x86/xen/pci-swiotlb-xen.c @@ -0,0 +1,109 @@ +/* Glue code to lib/swiotlb-xen.c */ + +#include <linux/dma-mapping.h> +#include <linux/pci.h> +#include <xen/swiotlb-xen.h> + +#include <asm/xen/hypervisor.h> +#include <xen/xen.h> +#include <asm/iommu_table.h> + + +#include <asm/xen/swiotlb-xen.h> +#ifdef CONFIG_X86_64 +#include <asm/iommu.h> +#include <asm/dma.h> +#endif +#include <linux/export.h> + +int xen_swiotlb __read_mostly; + +static struct dma_map_ops xen_swiotlb_dma_ops = { + .mapping_error = xen_swiotlb_dma_mapping_error, + .alloc = xen_swiotlb_alloc_coherent, + .free = xen_swiotlb_free_coherent, + .sync_single_for_cpu = xen_swiotlb_sync_single_for_cpu, + .sync_single_for_device = xen_swiotlb_sync_single_for_device, + .sync_sg_for_cpu = xen_swiotlb_sync_sg_for_cpu, + .sync_sg_for_device = xen_swiotlb_sync_sg_for_device, + .map_sg = xen_swiotlb_map_sg_attrs, + .unmap_sg = xen_swiotlb_unmap_sg_attrs, + .map_page = xen_swiotlb_map_page, + .unmap_page = xen_swiotlb_unmap_page, + .dma_supported = xen_swiotlb_dma_supported, +}; + +/* + * pci_xen_swiotlb_detect - set xen_swiotlb to 1 if necessary + * + * This returns non-zero if we are forced to use xen_swiotlb (by the boot + * option). + */ +int __init pci_xen_swiotlb_detect(void) +{ + + if (!xen_pv_domain()) + return 0; + + /* If running as PV guest, either iommu=soft, or swiotlb=force will + * activate this IOMMU. If running as PV privileged, activate it + * irregardless. + */ + if ((xen_initial_domain() || swiotlb || swiotlb_force)) + xen_swiotlb = 1; + + /* If we are running under Xen, we MUST disable the native SWIOTLB. + * Don't worry about swiotlb_force flag activating the native, as + * the 'swiotlb' flag is the only one turning it on. */ + swiotlb = 0; + +#ifdef CONFIG_X86_64 + /* pci_swiotlb_detect_4gb turns on native SWIOTLB if no_iommu == 0 + * (so no iommu=X command line over-writes). + * Considering that PV guests do not want the *native SWIOTLB* but + * only Xen SWIOTLB it is not useful to us so set no_iommu=1 here. + */ + if (max_pfn > MAX_DMA32_PFN) + no_iommu = 1; +#endif + return xen_swiotlb; +} + +void __init pci_xen_swiotlb_init(void) +{ + if (xen_swiotlb) { + xen_swiotlb_init(1, true /* early */); + dma_ops = &xen_swiotlb_dma_ops; + +#ifdef CONFIG_PCI + /* Make sure ACS will be enabled */ + pci_request_acs(); +#endif + } +} + +int pci_xen_swiotlb_init_late(void) +{ + int rc; + + if (xen_swiotlb) + return 0; + + rc = xen_swiotlb_init(1, false /* late */); + if (rc) + return rc; + + dma_ops = &xen_swiotlb_dma_ops; +#ifdef CONFIG_PCI + /* Make sure ACS will be enabled */ + pci_request_acs(); +#endif + + return 0; +} +EXPORT_SYMBOL_GPL(pci_xen_swiotlb_init_late); + +IOMMU_INIT_FINISH(pci_xen_swiotlb_detect, + NULL, + pci_xen_swiotlb_init, + NULL); diff --git a/kernel/arch/x86/xen/platform-pci-unplug.c b/kernel/arch/x86/xen/platform-pci-unplug.c new file mode 100644 index 000000000..a8261716d --- /dev/null +++ b/kernel/arch/x86/xen/platform-pci-unplug.c @@ -0,0 +1,217 @@ +/****************************************************************************** + * platform-pci-unplug.c + * + * Xen platform PCI device driver + * Copyright (c) 2010, Citrix + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. + * + */ + +#include <linux/init.h> +#include <linux/io.h> +#include <linux/module.h> + +#include <xen/platform_pci.h> +#include "xen-ops.h" + +#define XEN_PLATFORM_ERR_MAGIC -1 +#define XEN_PLATFORM_ERR_PROTOCOL -2 +#define XEN_PLATFORM_ERR_BLACKLIST -3 + +#ifdef CONFIG_XEN_PVHVM +/* store the value of xen_emul_unplug after the unplug is done */ +static int xen_platform_pci_unplug; +static int xen_emul_unplug; + +static int check_platform_magic(void) +{ + short magic; + char protocol; + + magic = inw(XEN_IOPORT_MAGIC); + if (magic != XEN_IOPORT_MAGIC_VAL) { + printk(KERN_ERR "Xen Platform PCI: unrecognised magic value\n"); + return XEN_PLATFORM_ERR_MAGIC; + } + + protocol = inb(XEN_IOPORT_PROTOVER); + + printk(KERN_DEBUG "Xen Platform PCI: I/O protocol version %d\n", + protocol); + + switch (protocol) { + case 1: + outw(XEN_IOPORT_LINUX_PRODNUM, XEN_IOPORT_PRODNUM); + outl(XEN_IOPORT_LINUX_DRVVER, XEN_IOPORT_DRVVER); + if (inw(XEN_IOPORT_MAGIC) != XEN_IOPORT_MAGIC_VAL) { + printk(KERN_ERR "Xen Platform: blacklisted by host\n"); + return XEN_PLATFORM_ERR_BLACKLIST; + } + break; + default: + printk(KERN_WARNING "Xen Platform PCI: unknown I/O protocol version"); + return XEN_PLATFORM_ERR_PROTOCOL; + } + + return 0; +} + +bool xen_has_pv_devices() +{ + if (!xen_domain()) + return false; + + /* PV domains always have them. */ + if (xen_pv_domain()) + return true; + + /* And user has xen_platform_pci=0 set in guest config as + * driver did not modify the value. */ + if (xen_platform_pci_unplug == 0) + return false; + + if (xen_platform_pci_unplug & XEN_UNPLUG_NEVER) + return false; + + if (xen_platform_pci_unplug & XEN_UNPLUG_ALL) + return true; + + /* This is an odd one - we are going to run legacy + * and PV drivers at the same time. */ + if (xen_platform_pci_unplug & XEN_UNPLUG_UNNECESSARY) + return true; + + /* And the caller has to follow with xen_pv_{disk,nic}_devices + * to be certain which driver can load. */ + return false; +} +EXPORT_SYMBOL_GPL(xen_has_pv_devices); + +static bool __xen_has_pv_device(int state) +{ + /* HVM domains might or might not */ + if (xen_hvm_domain() && (xen_platform_pci_unplug & state)) + return true; + + return xen_has_pv_devices(); +} + +bool xen_has_pv_nic_devices(void) +{ + return __xen_has_pv_device(XEN_UNPLUG_ALL_NICS | XEN_UNPLUG_ALL); +} +EXPORT_SYMBOL_GPL(xen_has_pv_nic_devices); + +bool xen_has_pv_disk_devices(void) +{ + return __xen_has_pv_device(XEN_UNPLUG_ALL_IDE_DISKS | + XEN_UNPLUG_AUX_IDE_DISKS | XEN_UNPLUG_ALL); +} +EXPORT_SYMBOL_GPL(xen_has_pv_disk_devices); + +/* + * This one is odd - it determines whether you want to run PV _and_ + * legacy (IDE) drivers together. This combination is only possible + * under HVM. + */ +bool xen_has_pv_and_legacy_disk_devices(void) +{ + if (!xen_domain()) + return false; + + /* N.B. This is only ever used in HVM mode */ + if (xen_pv_domain()) + return false; + + if (xen_platform_pci_unplug & XEN_UNPLUG_UNNECESSARY) + return true; + + return false; +} +EXPORT_SYMBOL_GPL(xen_has_pv_and_legacy_disk_devices); + +void xen_unplug_emulated_devices(void) +{ + int r; + + /* user explicitly requested no unplug */ + if (xen_emul_unplug & XEN_UNPLUG_NEVER) + return; + /* check the version of the xen platform PCI device */ + r = check_platform_magic(); + /* If the version matches enable the Xen platform PCI driver. + * Also enable the Xen platform PCI driver if the host does + * not support the unplug protocol (XEN_PLATFORM_ERR_MAGIC) + * but the user told us that unplugging is unnecessary. */ + if (r && !(r == XEN_PLATFORM_ERR_MAGIC && + (xen_emul_unplug & XEN_UNPLUG_UNNECESSARY))) + return; + /* Set the default value of xen_emul_unplug depending on whether or + * not the Xen PV frontends and the Xen platform PCI driver have + * been compiled for this kernel (modules or built-in are both OK). */ + if (!xen_emul_unplug) { + if (xen_must_unplug_nics()) { + printk(KERN_INFO "Netfront and the Xen platform PCI driver have " + "been compiled for this kernel: unplug emulated NICs.\n"); + xen_emul_unplug |= XEN_UNPLUG_ALL_NICS; + } + if (xen_must_unplug_disks()) { + printk(KERN_INFO "Blkfront and the Xen platform PCI driver have " + "been compiled for this kernel: unplug emulated disks.\n" + "You might have to change the root device\n" + "from /dev/hd[a-d] to /dev/xvd[a-d]\n" + "in your root= kernel command line option\n"); + xen_emul_unplug |= XEN_UNPLUG_ALL_IDE_DISKS; + } + } + /* Now unplug the emulated devices */ + if (!(xen_emul_unplug & XEN_UNPLUG_UNNECESSARY)) + outw(xen_emul_unplug, XEN_IOPORT_UNPLUG); + xen_platform_pci_unplug = xen_emul_unplug; +} + +static int __init parse_xen_emul_unplug(char *arg) +{ + char *p, *q; + int l; + + for (p = arg; p; p = q) { + q = strchr(p, ','); + if (q) { + l = q - p; + q++; + } else { + l = strlen(p); + } + if (!strncmp(p, "all", l)) + xen_emul_unplug |= XEN_UNPLUG_ALL; + else if (!strncmp(p, "ide-disks", l)) + xen_emul_unplug |= XEN_UNPLUG_ALL_IDE_DISKS; + else if (!strncmp(p, "aux-ide-disks", l)) + xen_emul_unplug |= XEN_UNPLUG_AUX_IDE_DISKS; + else if (!strncmp(p, "nics", l)) + xen_emul_unplug |= XEN_UNPLUG_ALL_NICS; + else if (!strncmp(p, "unnecessary", l)) + xen_emul_unplug |= XEN_UNPLUG_UNNECESSARY; + else if (!strncmp(p, "never", l)) + xen_emul_unplug |= XEN_UNPLUG_NEVER; + else + printk(KERN_WARNING "unrecognised option '%s' " + "in parameter 'xen_emul_unplug'\n", p); + } + return 0; +} +early_param("xen_emul_unplug", parse_xen_emul_unplug); +#endif diff --git a/kernel/arch/x86/xen/setup.c b/kernel/arch/x86/xen/setup.c new file mode 100644 index 000000000..55f388ef4 --- /dev/null +++ b/kernel/arch/x86/xen/setup.c @@ -0,0 +1,864 @@ +/* + * Machine specific setup for xen + * + * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007 + */ + +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/pm.h> +#include <linux/memblock.h> +#include <linux/cpuidle.h> +#include <linux/cpufreq.h> + +#include <asm/elf.h> +#include <asm/vdso.h> +#include <asm/e820.h> +#include <asm/setup.h> +#include <asm/acpi.h> +#include <asm/numa.h> +#include <asm/xen/hypervisor.h> +#include <asm/xen/hypercall.h> + +#include <xen/xen.h> +#include <xen/page.h> +#include <xen/interface/callback.h> +#include <xen/interface/memory.h> +#include <xen/interface/physdev.h> +#include <xen/features.h> +#include "xen-ops.h" +#include "vdso.h" +#include "p2m.h" +#include "mmu.h" + +/* Amount of extra memory space we add to the e820 ranges */ +struct xen_memory_region xen_extra_mem[XEN_EXTRA_MEM_MAX_REGIONS] __initdata; + +/* Number of pages released from the initial allocation. */ +unsigned long xen_released_pages; + +/* + * Buffer used to remap identity mapped pages. We only need the virtual space. + * The physical page behind this address is remapped as needed to different + * buffer pages. + */ +#define REMAP_SIZE (P2M_PER_PAGE - 3) +static struct { + unsigned long next_area_mfn; + unsigned long target_pfn; + unsigned long size; + unsigned long mfns[REMAP_SIZE]; +} xen_remap_buf __initdata __aligned(PAGE_SIZE); +static unsigned long xen_remap_mfn __initdata = INVALID_P2M_ENTRY; + +/* + * The maximum amount of extra memory compared to the base size. The + * main scaling factor is the size of struct page. At extreme ratios + * of base:extra, all the base memory can be filled with page + * structures for the extra memory, leaving no space for anything + * else. + * + * 10x seems like a reasonable balance between scaling flexibility and + * leaving a practically usable system. + */ +#define EXTRA_MEM_RATIO (10) + +static void __init xen_add_extra_mem(phys_addr_t start, phys_addr_t size) +{ + int i; + + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) { + /* Add new region. */ + if (xen_extra_mem[i].size == 0) { + xen_extra_mem[i].start = start; + xen_extra_mem[i].size = size; + break; + } + /* Append to existing region. */ + if (xen_extra_mem[i].start + xen_extra_mem[i].size == start) { + xen_extra_mem[i].size += size; + break; + } + } + if (i == XEN_EXTRA_MEM_MAX_REGIONS) + printk(KERN_WARNING "Warning: not enough extra memory regions\n"); + + memblock_reserve(start, size); +} + +static void __init xen_del_extra_mem(phys_addr_t start, phys_addr_t size) +{ + int i; + phys_addr_t start_r, size_r; + + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) { + start_r = xen_extra_mem[i].start; + size_r = xen_extra_mem[i].size; + + /* Start of region. */ + if (start_r == start) { + BUG_ON(size > size_r); + xen_extra_mem[i].start += size; + xen_extra_mem[i].size -= size; + break; + } + /* End of region. */ + if (start_r + size_r == start + size) { + BUG_ON(size > size_r); + xen_extra_mem[i].size -= size; + break; + } + /* Mid of region. */ + if (start > start_r && start < start_r + size_r) { + BUG_ON(start + size > start_r + size_r); + xen_extra_mem[i].size = start - start_r; + /* Calling memblock_reserve() again is okay. */ + xen_add_extra_mem(start + size, start_r + size_r - + (start + size)); + break; + } + } + memblock_free(start, size); +} + +/* + * Called during boot before the p2m list can take entries beyond the + * hypervisor supplied p2m list. Entries in extra mem are to be regarded as + * invalid. + */ +unsigned long __ref xen_chk_extra_mem(unsigned long pfn) +{ + int i; + phys_addr_t addr = PFN_PHYS(pfn); + + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) { + if (addr >= xen_extra_mem[i].start && + addr < xen_extra_mem[i].start + xen_extra_mem[i].size) + return INVALID_P2M_ENTRY; + } + + return IDENTITY_FRAME(pfn); +} + +/* + * Mark all pfns of extra mem as invalid in p2m list. + */ +void __init xen_inv_extra_mem(void) +{ + unsigned long pfn, pfn_s, pfn_e; + int i; + + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) { + if (!xen_extra_mem[i].size) + continue; + pfn_s = PFN_DOWN(xen_extra_mem[i].start); + pfn_e = PFN_UP(xen_extra_mem[i].start + xen_extra_mem[i].size); + for (pfn = pfn_s; pfn < pfn_e; pfn++) + set_phys_to_machine(pfn, INVALID_P2M_ENTRY); + } +} + +/* + * Finds the next RAM pfn available in the E820 map after min_pfn. + * This function updates min_pfn with the pfn found and returns + * the size of that range or zero if not found. + */ +static unsigned long __init xen_find_pfn_range( + const struct e820entry *list, size_t map_size, + unsigned long *min_pfn) +{ + const struct e820entry *entry; + unsigned int i; + unsigned long done = 0; + + for (i = 0, entry = list; i < map_size; i++, entry++) { + unsigned long s_pfn; + unsigned long e_pfn; + + if (entry->type != E820_RAM) + continue; + + e_pfn = PFN_DOWN(entry->addr + entry->size); + + /* We only care about E820 after this */ + if (e_pfn < *min_pfn) + continue; + + s_pfn = PFN_UP(entry->addr); + + /* If min_pfn falls within the E820 entry, we want to start + * at the min_pfn PFN. + */ + if (s_pfn <= *min_pfn) { + done = e_pfn - *min_pfn; + } else { + done = e_pfn - s_pfn; + *min_pfn = s_pfn; + } + break; + } + + return done; +} + +static int __init xen_free_mfn(unsigned long mfn) +{ + struct xen_memory_reservation reservation = { + .address_bits = 0, + .extent_order = 0, + .domid = DOMID_SELF + }; + + set_xen_guest_handle(reservation.extent_start, &mfn); + reservation.nr_extents = 1; + + return HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); +} + +/* + * This releases a chunk of memory and then does the identity map. It's used + * as a fallback if the remapping fails. + */ +static void __init xen_set_identity_and_release_chunk(unsigned long start_pfn, + unsigned long end_pfn, unsigned long nr_pages, unsigned long *released) +{ + unsigned long pfn, end; + int ret; + + WARN_ON(start_pfn > end_pfn); + + /* Release pages first. */ + end = min(end_pfn, nr_pages); + for (pfn = start_pfn; pfn < end; pfn++) { + unsigned long mfn = pfn_to_mfn(pfn); + + /* Make sure pfn exists to start with */ + if (mfn == INVALID_P2M_ENTRY || mfn_to_pfn(mfn) != pfn) + continue; + + ret = xen_free_mfn(mfn); + WARN(ret != 1, "Failed to release pfn %lx err=%d\n", pfn, ret); + + if (ret == 1) { + (*released)++; + if (!__set_phys_to_machine(pfn, INVALID_P2M_ENTRY)) + break; + } else + break; + } + + set_phys_range_identity(start_pfn, end_pfn); +} + +/* + * Helper function to update the p2m and m2p tables and kernel mapping. + */ +static void __init xen_update_mem_tables(unsigned long pfn, unsigned long mfn) +{ + struct mmu_update update = { + .ptr = ((uint64_t)mfn << PAGE_SHIFT) | MMU_MACHPHYS_UPDATE, + .val = pfn + }; + + /* Update p2m */ + if (!set_phys_to_machine(pfn, mfn)) { + WARN(1, "Failed to set p2m mapping for pfn=%ld mfn=%ld\n", + pfn, mfn); + BUG(); + } + + /* Update m2p */ + if (HYPERVISOR_mmu_update(&update, 1, NULL, DOMID_SELF) < 0) { + WARN(1, "Failed to set m2p mapping for mfn=%ld pfn=%ld\n", + mfn, pfn); + BUG(); + } + + /* Update kernel mapping, but not for highmem. */ + if (pfn >= PFN_UP(__pa(high_memory - 1))) + return; + + if (HYPERVISOR_update_va_mapping((unsigned long)__va(pfn << PAGE_SHIFT), + mfn_pte(mfn, PAGE_KERNEL), 0)) { + WARN(1, "Failed to update kernel mapping for mfn=%ld pfn=%ld\n", + mfn, pfn); + BUG(); + } +} + +/* + * This function updates the p2m and m2p tables with an identity map from + * start_pfn to start_pfn+size and prepares remapping the underlying RAM of the + * original allocation at remap_pfn. The information needed for remapping is + * saved in the memory itself to avoid the need for allocating buffers. The + * complete remap information is contained in a list of MFNs each containing + * up to REMAP_SIZE MFNs and the start target PFN for doing the remap. + * This enables us to preserve the original mfn sequence while doing the + * remapping at a time when the memory management is capable of allocating + * virtual and physical memory in arbitrary amounts, see 'xen_remap_memory' and + * its callers. + */ +static void __init xen_do_set_identity_and_remap_chunk( + unsigned long start_pfn, unsigned long size, unsigned long remap_pfn) +{ + unsigned long buf = (unsigned long)&xen_remap_buf; + unsigned long mfn_save, mfn; + unsigned long ident_pfn_iter, remap_pfn_iter; + unsigned long ident_end_pfn = start_pfn + size; + unsigned long left = size; + unsigned int i, chunk; + + WARN_ON(size == 0); + + BUG_ON(xen_feature(XENFEAT_auto_translated_physmap)); + + mfn_save = virt_to_mfn(buf); + + for (ident_pfn_iter = start_pfn, remap_pfn_iter = remap_pfn; + ident_pfn_iter < ident_end_pfn; + ident_pfn_iter += REMAP_SIZE, remap_pfn_iter += REMAP_SIZE) { + chunk = (left < REMAP_SIZE) ? left : REMAP_SIZE; + + /* Map first pfn to xen_remap_buf */ + mfn = pfn_to_mfn(ident_pfn_iter); + set_pte_mfn(buf, mfn, PAGE_KERNEL); + + /* Save mapping information in page */ + xen_remap_buf.next_area_mfn = xen_remap_mfn; + xen_remap_buf.target_pfn = remap_pfn_iter; + xen_remap_buf.size = chunk; + for (i = 0; i < chunk; i++) + xen_remap_buf.mfns[i] = pfn_to_mfn(ident_pfn_iter + i); + + /* Put remap buf into list. */ + xen_remap_mfn = mfn; + + /* Set identity map */ + set_phys_range_identity(ident_pfn_iter, ident_pfn_iter + chunk); + + left -= chunk; + } + + /* Restore old xen_remap_buf mapping */ + set_pte_mfn(buf, mfn_save, PAGE_KERNEL); +} + +/* + * This function takes a contiguous pfn range that needs to be identity mapped + * and: + * + * 1) Finds a new range of pfns to use to remap based on E820 and remap_pfn. + * 2) Calls the do_ function to actually do the mapping/remapping work. + * + * The goal is to not allocate additional memory but to remap the existing + * pages. In the case of an error the underlying memory is simply released back + * to Xen and not remapped. + */ +static unsigned long __init xen_set_identity_and_remap_chunk( + const struct e820entry *list, size_t map_size, unsigned long start_pfn, + unsigned long end_pfn, unsigned long nr_pages, unsigned long remap_pfn, + unsigned long *released, unsigned long *remapped) +{ + unsigned long pfn; + unsigned long i = 0; + unsigned long n = end_pfn - start_pfn; + + while (i < n) { + unsigned long cur_pfn = start_pfn + i; + unsigned long left = n - i; + unsigned long size = left; + unsigned long remap_range_size; + + /* Do not remap pages beyond the current allocation */ + if (cur_pfn >= nr_pages) { + /* Identity map remaining pages */ + set_phys_range_identity(cur_pfn, cur_pfn + size); + break; + } + if (cur_pfn + size > nr_pages) + size = nr_pages - cur_pfn; + + remap_range_size = xen_find_pfn_range(list, map_size, + &remap_pfn); + if (!remap_range_size) { + pr_warning("Unable to find available pfn range, not remapping identity pages\n"); + xen_set_identity_and_release_chunk(cur_pfn, + cur_pfn + left, nr_pages, released); + break; + } + /* Adjust size to fit in current e820 RAM region */ + if (size > remap_range_size) + size = remap_range_size; + + xen_do_set_identity_and_remap_chunk(cur_pfn, size, remap_pfn); + + /* Update variables to reflect new mappings. */ + i += size; + remap_pfn += size; + *remapped += size; + } + + /* + * If the PFNs are currently mapped, the VA mapping also needs + * to be updated to be 1:1. + */ + for (pfn = start_pfn; pfn <= max_pfn_mapped && pfn < end_pfn; pfn++) + (void)HYPERVISOR_update_va_mapping( + (unsigned long)__va(pfn << PAGE_SHIFT), + mfn_pte(pfn, PAGE_KERNEL_IO), 0); + + return remap_pfn; +} + +static void __init xen_set_identity_and_remap( + const struct e820entry *list, size_t map_size, unsigned long nr_pages, + unsigned long *released, unsigned long *remapped) +{ + phys_addr_t start = 0; + unsigned long last_pfn = nr_pages; + const struct e820entry *entry; + unsigned long num_released = 0; + unsigned long num_remapped = 0; + int i; + + /* + * Combine non-RAM regions and gaps until a RAM region (or the + * end of the map) is reached, then set the 1:1 map and + * remap the memory in those non-RAM regions. + * + * The combined non-RAM regions are rounded to a whole number + * of pages so any partial pages are accessible via the 1:1 + * mapping. This is needed for some BIOSes that put (for + * example) the DMI tables in a reserved region that begins on + * a non-page boundary. + */ + for (i = 0, entry = list; i < map_size; i++, entry++) { + phys_addr_t end = entry->addr + entry->size; + if (entry->type == E820_RAM || i == map_size - 1) { + unsigned long start_pfn = PFN_DOWN(start); + unsigned long end_pfn = PFN_UP(end); + + if (entry->type == E820_RAM) + end_pfn = PFN_UP(entry->addr); + + if (start_pfn < end_pfn) + last_pfn = xen_set_identity_and_remap_chunk( + list, map_size, start_pfn, + end_pfn, nr_pages, last_pfn, + &num_released, &num_remapped); + start = end; + } + } + + *released = num_released; + *remapped = num_remapped; + + pr_info("Released %ld page(s)\n", num_released); +} + +/* + * Remap the memory prepared in xen_do_set_identity_and_remap_chunk(). + * The remap information (which mfn remap to which pfn) is contained in the + * to be remapped memory itself in a linked list anchored at xen_remap_mfn. + * This scheme allows to remap the different chunks in arbitrary order while + * the resulting mapping will be independant from the order. + */ +void __init xen_remap_memory(void) +{ + unsigned long buf = (unsigned long)&xen_remap_buf; + unsigned long mfn_save, mfn, pfn; + unsigned long remapped = 0; + unsigned int i; + unsigned long pfn_s = ~0UL; + unsigned long len = 0; + + mfn_save = virt_to_mfn(buf); + + while (xen_remap_mfn != INVALID_P2M_ENTRY) { + /* Map the remap information */ + set_pte_mfn(buf, xen_remap_mfn, PAGE_KERNEL); + + BUG_ON(xen_remap_mfn != xen_remap_buf.mfns[0]); + + pfn = xen_remap_buf.target_pfn; + for (i = 0; i < xen_remap_buf.size; i++) { + mfn = xen_remap_buf.mfns[i]; + xen_update_mem_tables(pfn, mfn); + remapped++; + pfn++; + } + if (pfn_s == ~0UL || pfn == pfn_s) { + pfn_s = xen_remap_buf.target_pfn; + len += xen_remap_buf.size; + } else if (pfn_s + len == xen_remap_buf.target_pfn) { + len += xen_remap_buf.size; + } else { + xen_del_extra_mem(PFN_PHYS(pfn_s), PFN_PHYS(len)); + pfn_s = xen_remap_buf.target_pfn; + len = xen_remap_buf.size; + } + + mfn = xen_remap_mfn; + xen_remap_mfn = xen_remap_buf.next_area_mfn; + } + + if (pfn_s != ~0UL && len) + xen_del_extra_mem(PFN_PHYS(pfn_s), PFN_PHYS(len)); + + set_pte_mfn(buf, mfn_save, PAGE_KERNEL); + + pr_info("Remapped %ld page(s)\n", remapped); +} + +static unsigned long __init xen_get_max_pages(void) +{ + unsigned long max_pages = MAX_DOMAIN_PAGES; + domid_t domid = DOMID_SELF; + int ret; + + /* + * For the initial domain we use the maximum reservation as + * the maximum page. + * + * For guest domains the current maximum reservation reflects + * the current maximum rather than the static maximum. In this + * case the e820 map provided to us will cover the static + * maximum region. + */ + if (xen_initial_domain()) { + ret = HYPERVISOR_memory_op(XENMEM_maximum_reservation, &domid); + if (ret > 0) + max_pages = ret; + } + + return min(max_pages, MAX_DOMAIN_PAGES); +} + +static void __init xen_align_and_add_e820_region(phys_addr_t start, + phys_addr_t size, int type) +{ + phys_addr_t end = start + size; + + /* Align RAM regions to page boundaries. */ + if (type == E820_RAM) { + start = PAGE_ALIGN(start); + end &= ~((phys_addr_t)PAGE_SIZE - 1); + } + + e820_add_region(start, end - start, type); +} + +static void __init xen_ignore_unusable(struct e820entry *list, size_t map_size) +{ + struct e820entry *entry; + unsigned int i; + + for (i = 0, entry = list; i < map_size; i++, entry++) { + if (entry->type == E820_UNUSABLE) + entry->type = E820_RAM; + } +} + +/** + * machine_specific_memory_setup - Hook for machine specific memory setup. + **/ +char * __init xen_memory_setup(void) +{ + static struct e820entry map[E820MAX] __initdata; + + unsigned long max_pfn = xen_start_info->nr_pages; + phys_addr_t mem_end; + int rc; + struct xen_memory_map memmap; + unsigned long max_pages; + unsigned long extra_pages = 0; + unsigned long remapped_pages; + int i; + int op; + + max_pfn = min(MAX_DOMAIN_PAGES, max_pfn); + mem_end = PFN_PHYS(max_pfn); + + memmap.nr_entries = E820MAX; + set_xen_guest_handle(memmap.buffer, map); + + op = xen_initial_domain() ? + XENMEM_machine_memory_map : + XENMEM_memory_map; + rc = HYPERVISOR_memory_op(op, &memmap); + if (rc == -ENOSYS) { + BUG_ON(xen_initial_domain()); + memmap.nr_entries = 1; + map[0].addr = 0ULL; + map[0].size = mem_end; + /* 8MB slack (to balance backend allocations). */ + map[0].size += 8ULL << 20; + map[0].type = E820_RAM; + rc = 0; + } + BUG_ON(rc); + BUG_ON(memmap.nr_entries == 0); + + /* + * Xen won't allow a 1:1 mapping to be created to UNUSABLE + * regions, so if we're using the machine memory map leave the + * region as RAM as it is in the pseudo-physical map. + * + * UNUSABLE regions in domUs are not handled and will need + * a patch in the future. + */ + if (xen_initial_domain()) + xen_ignore_unusable(map, memmap.nr_entries); + + /* Make sure the Xen-supplied memory map is well-ordered. */ + sanitize_e820_map(map, memmap.nr_entries, &memmap.nr_entries); + + max_pages = xen_get_max_pages(); + if (max_pages > max_pfn) + extra_pages += max_pages - max_pfn; + + /* + * Set identity map on non-RAM pages and prepare remapping the + * underlying RAM. + */ + xen_set_identity_and_remap(map, memmap.nr_entries, max_pfn, + &xen_released_pages, &remapped_pages); + + extra_pages += xen_released_pages; + extra_pages += remapped_pages; + + /* + * Clamp the amount of extra memory to a EXTRA_MEM_RATIO + * factor the base size. On non-highmem systems, the base + * size is the full initial memory allocation; on highmem it + * is limited to the max size of lowmem, so that it doesn't + * get completely filled. + * + * In principle there could be a problem in lowmem systems if + * the initial memory is also very large with respect to + * lowmem, but we won't try to deal with that here. + */ + extra_pages = min(EXTRA_MEM_RATIO * min(max_pfn, PFN_DOWN(MAXMEM)), + extra_pages); + i = 0; + while (i < memmap.nr_entries) { + phys_addr_t addr = map[i].addr; + phys_addr_t size = map[i].size; + u32 type = map[i].type; + + if (type == E820_RAM) { + if (addr < mem_end) { + size = min(size, mem_end - addr); + } else if (extra_pages) { + size = min(size, PFN_PHYS(extra_pages)); + extra_pages -= PFN_DOWN(size); + xen_add_extra_mem(addr, size); + xen_max_p2m_pfn = PFN_DOWN(addr + size); + } else + type = E820_UNUSABLE; + } + + xen_align_and_add_e820_region(addr, size, type); + + map[i].addr += size; + map[i].size -= size; + if (map[i].size == 0) + i++; + } + + /* + * Set the rest as identity mapped, in case PCI BARs are + * located here. + * + * PFNs above MAX_P2M_PFN are considered identity mapped as + * well. + */ + set_phys_range_identity(map[i-1].addr / PAGE_SIZE, ~0ul); + + /* + * In domU, the ISA region is normal, usable memory, but we + * reserve ISA memory anyway because too many things poke + * about in there. + */ + e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS, + E820_RESERVED); + + /* + * Reserve Xen bits: + * - mfn_list + * - xen_start_info + * See comment above "struct start_info" in <xen/interface/xen.h> + * We tried to make the the memblock_reserve more selective so + * that it would be clear what region is reserved. Sadly we ran + * in the problem wherein on a 64-bit hypervisor with a 32-bit + * initial domain, the pt_base has the cr3 value which is not + * neccessarily where the pagetable starts! As Jan put it: " + * Actually, the adjustment turns out to be correct: The page + * tables for a 32-on-64 dom0 get allocated in the order "first L1", + * "first L2", "first L3", so the offset to the page table base is + * indeed 2. When reading xen/include/public/xen.h's comment + * very strictly, this is not a violation (since there nothing is said + * that the first thing in the page table space is pointed to by + * pt_base; I admit that this seems to be implied though, namely + * do I think that it is implied that the page table space is the + * range [pt_base, pt_base + nt_pt_frames), whereas that + * range here indeed is [pt_base - 2, pt_base - 2 + nt_pt_frames), + * which - without a priori knowledge - the kernel would have + * difficulty to figure out)." - so lets just fall back to the + * easy way and reserve the whole region. + */ + memblock_reserve(__pa(xen_start_info->mfn_list), + xen_start_info->pt_base - xen_start_info->mfn_list); + + sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); + + return "Xen"; +} + +/* + * Machine specific memory setup for auto-translated guests. + */ +char * __init xen_auto_xlated_memory_setup(void) +{ + static struct e820entry map[E820MAX] __initdata; + + struct xen_memory_map memmap; + int i; + int rc; + + memmap.nr_entries = E820MAX; + set_xen_guest_handle(memmap.buffer, map); + + rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap); + if (rc < 0) + panic("No memory map (%d)\n", rc); + + sanitize_e820_map(map, ARRAY_SIZE(map), &memmap.nr_entries); + + for (i = 0; i < memmap.nr_entries; i++) + e820_add_region(map[i].addr, map[i].size, map[i].type); + + memblock_reserve(__pa(xen_start_info->mfn_list), + xen_start_info->pt_base - xen_start_info->mfn_list); + + return "Xen"; +} + +/* + * Set the bit indicating "nosegneg" library variants should be used. + * We only need to bother in pure 32-bit mode; compat 32-bit processes + * can have un-truncated segments, so wrapping around is allowed. + */ +static void __init fiddle_vdso(void) +{ +#ifdef CONFIG_X86_32 + /* + * This could be called before selected_vdso32 is initialized, so + * just fiddle with both possible images. vdso_image_32_syscall + * can't be selected, since it only exists on 64-bit systems. + */ + u32 *mask; + mask = vdso_image_32_int80.data + + vdso_image_32_int80.sym_VDSO32_NOTE_MASK; + *mask |= 1 << VDSO_NOTE_NONEGSEG_BIT; + mask = vdso_image_32_sysenter.data + + vdso_image_32_sysenter.sym_VDSO32_NOTE_MASK; + *mask |= 1 << VDSO_NOTE_NONEGSEG_BIT; +#endif +} + +static int register_callback(unsigned type, const void *func) +{ + struct callback_register callback = { + .type = type, + .address = XEN_CALLBACK(__KERNEL_CS, func), + .flags = CALLBACKF_mask_events, + }; + + return HYPERVISOR_callback_op(CALLBACKOP_register, &callback); +} + +void xen_enable_sysenter(void) +{ + int ret; + unsigned sysenter_feature; + +#ifdef CONFIG_X86_32 + sysenter_feature = X86_FEATURE_SEP; +#else + sysenter_feature = X86_FEATURE_SYSENTER32; +#endif + + if (!boot_cpu_has(sysenter_feature)) + return; + + ret = register_callback(CALLBACKTYPE_sysenter, xen_sysenter_target); + if(ret != 0) + setup_clear_cpu_cap(sysenter_feature); +} + +void xen_enable_syscall(void) +{ +#ifdef CONFIG_X86_64 + int ret; + + ret = register_callback(CALLBACKTYPE_syscall, xen_syscall_target); + if (ret != 0) { + printk(KERN_ERR "Failed to set syscall callback: %d\n", ret); + /* Pretty fatal; 64-bit userspace has no other + mechanism for syscalls. */ + } + + if (boot_cpu_has(X86_FEATURE_SYSCALL32)) { + ret = register_callback(CALLBACKTYPE_syscall32, + xen_syscall32_target); + if (ret != 0) + setup_clear_cpu_cap(X86_FEATURE_SYSCALL32); + } +#endif /* CONFIG_X86_64 */ +} + +void __init xen_pvmmu_arch_setup(void) +{ + HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments); + HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_writable_pagetables); + + HYPERVISOR_vm_assist(VMASST_CMD_enable, + VMASST_TYPE_pae_extended_cr3); + + if (register_callback(CALLBACKTYPE_event, xen_hypervisor_callback) || + register_callback(CALLBACKTYPE_failsafe, xen_failsafe_callback)) + BUG(); + + xen_enable_sysenter(); + xen_enable_syscall(); +} + +/* This function is not called for HVM domains */ +void __init xen_arch_setup(void) +{ + xen_panic_handler_init(); + if (!xen_feature(XENFEAT_auto_translated_physmap)) + xen_pvmmu_arch_setup(); + +#ifdef CONFIG_ACPI + if (!(xen_start_info->flags & SIF_INITDOMAIN)) { + printk(KERN_INFO "ACPI in unprivileged domain disabled\n"); + disable_acpi(); + } +#endif + + memcpy(boot_command_line, xen_start_info->cmd_line, + MAX_GUEST_CMDLINE > COMMAND_LINE_SIZE ? + COMMAND_LINE_SIZE : MAX_GUEST_CMDLINE); + + /* Set up idle, making sure it calls safe_halt() pvop */ + disable_cpuidle(); + disable_cpufreq(); + WARN_ON(xen_set_default_idle()); + fiddle_vdso(); +#ifdef CONFIG_NUMA + numa_off = 1; +#endif +} diff --git a/kernel/arch/x86/xen/smp.c b/kernel/arch/x86/xen/smp.c new file mode 100644 index 000000000..864843844 --- /dev/null +++ b/kernel/arch/x86/xen/smp.c @@ -0,0 +1,780 @@ +/* + * Xen SMP support + * + * This file implements the Xen versions of smp_ops. SMP under Xen is + * very straightforward. Bringing a CPU up is simply a matter of + * loading its initial context and setting it running. + * + * IPIs are handled through the Xen event mechanism. + * + * Because virtual CPUs can be scheduled onto any real CPU, there's no + * useful topology information for the kernel to make use of. As a + * result, all CPUs are treated as if they're single-core and + * single-threaded. + */ +#include <linux/sched.h> +#include <linux/err.h> +#include <linux/slab.h> +#include <linux/smp.h> +#include <linux/irq_work.h> +#include <linux/tick.h> + +#include <asm/paravirt.h> +#include <asm/desc.h> +#include <asm/pgtable.h> +#include <asm/cpu.h> + +#include <xen/interface/xen.h> +#include <xen/interface/vcpu.h> + +#include <asm/xen/interface.h> +#include <asm/xen/hypercall.h> + +#include <xen/xen.h> +#include <xen/page.h> +#include <xen/events.h> + +#include <xen/hvc-console.h> +#include "xen-ops.h" +#include "mmu.h" +#include "smp.h" + +cpumask_var_t xen_cpu_initialized_map; + +struct xen_common_irq { + int irq; + char *name; +}; +static DEFINE_PER_CPU(struct xen_common_irq, xen_resched_irq) = { .irq = -1 }; +static DEFINE_PER_CPU(struct xen_common_irq, xen_callfunc_irq) = { .irq = -1 }; +static DEFINE_PER_CPU(struct xen_common_irq, xen_callfuncsingle_irq) = { .irq = -1 }; +static DEFINE_PER_CPU(struct xen_common_irq, xen_irq_work) = { .irq = -1 }; +static DEFINE_PER_CPU(struct xen_common_irq, xen_debug_irq) = { .irq = -1 }; + +static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id); +static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id); +static irqreturn_t xen_irq_work_interrupt(int irq, void *dev_id); + +/* + * Reschedule call back. + */ +static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id) +{ + inc_irq_stat(irq_resched_count); + scheduler_ipi(); + + return IRQ_HANDLED; +} + +static void cpu_bringup(void) +{ + int cpu; + + cpu_init(); + touch_softlockup_watchdog(); + preempt_disable(); + + /* PVH runs in ring 0 and allows us to do native syscalls. Yay! */ + if (!xen_feature(XENFEAT_supervisor_mode_kernel)) { + xen_enable_sysenter(); + xen_enable_syscall(); + } + cpu = smp_processor_id(); + smp_store_cpu_info(cpu); + cpu_data(cpu).x86_max_cores = 1; + set_cpu_sibling_map(cpu); + + xen_setup_cpu_clockevents(); + + notify_cpu_starting(cpu); + + set_cpu_online(cpu, true); + + cpu_set_state_online(cpu); /* Implies full memory barrier. */ + + /* We can take interrupts now: we're officially "up". */ + local_irq_enable(); +} + +/* + * Note: cpu parameter is only relevant for PVH. The reason for passing it + * is we can't do smp_processor_id until the percpu segments are loaded, for + * which we need the cpu number! So we pass it in rdi as first parameter. + */ +asmlinkage __visible void cpu_bringup_and_idle(int cpu) +{ +#ifdef CONFIG_XEN_PVH + if (xen_feature(XENFEAT_auto_translated_physmap) && + xen_feature(XENFEAT_supervisor_mode_kernel)) + xen_pvh_secondary_vcpu_init(cpu); +#endif + cpu_bringup(); + cpu_startup_entry(CPUHP_ONLINE); +} + +static void xen_smp_intr_free(unsigned int cpu) +{ + if (per_cpu(xen_resched_irq, cpu).irq >= 0) { + unbind_from_irqhandler(per_cpu(xen_resched_irq, cpu).irq, NULL); + per_cpu(xen_resched_irq, cpu).irq = -1; + kfree(per_cpu(xen_resched_irq, cpu).name); + per_cpu(xen_resched_irq, cpu).name = NULL; + } + if (per_cpu(xen_callfunc_irq, cpu).irq >= 0) { + unbind_from_irqhandler(per_cpu(xen_callfunc_irq, cpu).irq, NULL); + per_cpu(xen_callfunc_irq, cpu).irq = -1; + kfree(per_cpu(xen_callfunc_irq, cpu).name); + per_cpu(xen_callfunc_irq, cpu).name = NULL; + } + if (per_cpu(xen_debug_irq, cpu).irq >= 0) { + unbind_from_irqhandler(per_cpu(xen_debug_irq, cpu).irq, NULL); + per_cpu(xen_debug_irq, cpu).irq = -1; + kfree(per_cpu(xen_debug_irq, cpu).name); + per_cpu(xen_debug_irq, cpu).name = NULL; + } + if (per_cpu(xen_callfuncsingle_irq, cpu).irq >= 0) { + unbind_from_irqhandler(per_cpu(xen_callfuncsingle_irq, cpu).irq, + NULL); + per_cpu(xen_callfuncsingle_irq, cpu).irq = -1; + kfree(per_cpu(xen_callfuncsingle_irq, cpu).name); + per_cpu(xen_callfuncsingle_irq, cpu).name = NULL; + } + if (xen_hvm_domain()) + return; + + if (per_cpu(xen_irq_work, cpu).irq >= 0) { + unbind_from_irqhandler(per_cpu(xen_irq_work, cpu).irq, NULL); + per_cpu(xen_irq_work, cpu).irq = -1; + kfree(per_cpu(xen_irq_work, cpu).name); + per_cpu(xen_irq_work, cpu).name = NULL; + } +}; +static int xen_smp_intr_init(unsigned int cpu) +{ + int rc; + char *resched_name, *callfunc_name, *debug_name; + + resched_name = kasprintf(GFP_KERNEL, "resched%d", cpu); + rc = bind_ipi_to_irqhandler(XEN_RESCHEDULE_VECTOR, + cpu, + xen_reschedule_interrupt, + IRQF_PERCPU|IRQF_NOBALANCING, + resched_name, + NULL); + if (rc < 0) + goto fail; + per_cpu(xen_resched_irq, cpu).irq = rc; + per_cpu(xen_resched_irq, cpu).name = resched_name; + + callfunc_name = kasprintf(GFP_KERNEL, "callfunc%d", cpu); + rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_VECTOR, + cpu, + xen_call_function_interrupt, + IRQF_PERCPU|IRQF_NOBALANCING, + callfunc_name, + NULL); + if (rc < 0) + goto fail; + per_cpu(xen_callfunc_irq, cpu).irq = rc; + per_cpu(xen_callfunc_irq, cpu).name = callfunc_name; + + debug_name = kasprintf(GFP_KERNEL, "debug%d", cpu); + rc = bind_virq_to_irqhandler(VIRQ_DEBUG, cpu, xen_debug_interrupt, + IRQF_PERCPU | IRQF_NOBALANCING, + debug_name, NULL); + if (rc < 0) + goto fail; + per_cpu(xen_debug_irq, cpu).irq = rc; + per_cpu(xen_debug_irq, cpu).name = debug_name; + + callfunc_name = kasprintf(GFP_KERNEL, "callfuncsingle%d", cpu); + rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_SINGLE_VECTOR, + cpu, + xen_call_function_single_interrupt, + IRQF_PERCPU|IRQF_NOBALANCING, + callfunc_name, + NULL); + if (rc < 0) + goto fail; + per_cpu(xen_callfuncsingle_irq, cpu).irq = rc; + per_cpu(xen_callfuncsingle_irq, cpu).name = callfunc_name; + + /* + * The IRQ worker on PVHVM goes through the native path and uses the + * IPI mechanism. + */ + if (xen_hvm_domain()) + return 0; + + callfunc_name = kasprintf(GFP_KERNEL, "irqwork%d", cpu); + rc = bind_ipi_to_irqhandler(XEN_IRQ_WORK_VECTOR, + cpu, + xen_irq_work_interrupt, + IRQF_PERCPU|IRQF_NOBALANCING, + callfunc_name, + NULL); + if (rc < 0) + goto fail; + per_cpu(xen_irq_work, cpu).irq = rc; + per_cpu(xen_irq_work, cpu).name = callfunc_name; + + return 0; + + fail: + xen_smp_intr_free(cpu); + return rc; +} + +static void __init xen_fill_possible_map(void) +{ + int i, rc; + + if (xen_initial_domain()) + return; + + for (i = 0; i < nr_cpu_ids; i++) { + rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL); + if (rc >= 0) { + num_processors++; + set_cpu_possible(i, true); + } + } +} + +static void __init xen_filter_cpu_maps(void) +{ + int i, rc; + unsigned int subtract = 0; + + if (!xen_initial_domain()) + return; + + num_processors = 0; + disabled_cpus = 0; + for (i = 0; i < nr_cpu_ids; i++) { + rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL); + if (rc >= 0) { + num_processors++; + set_cpu_possible(i, true); + } else { + set_cpu_possible(i, false); + set_cpu_present(i, false); + subtract++; + } + } +#ifdef CONFIG_HOTPLUG_CPU + /* This is akin to using 'nr_cpus' on the Linux command line. + * Which is OK as when we use 'dom0_max_vcpus=X' we can only + * have up to X, while nr_cpu_ids is greater than X. This + * normally is not a problem, except when CPU hotplugging + * is involved and then there might be more than X CPUs + * in the guest - which will not work as there is no + * hypercall to expand the max number of VCPUs an already + * running guest has. So cap it up to X. */ + if (subtract) + nr_cpu_ids = nr_cpu_ids - subtract; +#endif + +} + +static void __init xen_smp_prepare_boot_cpu(void) +{ + BUG_ON(smp_processor_id() != 0); + native_smp_prepare_boot_cpu(); + + if (xen_pv_domain()) { + if (!xen_feature(XENFEAT_writable_page_tables)) + /* We've switched to the "real" per-cpu gdt, so make + * sure the old memory can be recycled. */ + make_lowmem_page_readwrite(xen_initial_gdt); + +#ifdef CONFIG_X86_32 + /* + * Xen starts us with XEN_FLAT_RING1_DS, but linux code + * expects __USER_DS + */ + loadsegment(ds, __USER_DS); + loadsegment(es, __USER_DS); +#endif + + xen_filter_cpu_maps(); + xen_setup_vcpu_info_placement(); + } + /* + * The alternative logic (which patches the unlock/lock) runs before + * the smp bootup up code is activated. Hence we need to set this up + * the core kernel is being patched. Otherwise we will have only + * modules patched but not core code. + */ + xen_init_spinlocks(); +} + +static void __init xen_smp_prepare_cpus(unsigned int max_cpus) +{ + unsigned cpu; + unsigned int i; + + if (skip_ioapic_setup) { + char *m = (max_cpus == 0) ? + "The nosmp parameter is incompatible with Xen; " \ + "use Xen dom0_max_vcpus=1 parameter" : + "The noapic parameter is incompatible with Xen"; + + xen_raw_printk(m); + panic(m); + } + xen_init_lock_cpu(0); + + smp_store_boot_cpu_info(); + cpu_data(0).x86_max_cores = 1; + + for_each_possible_cpu(i) { + zalloc_cpumask_var(&per_cpu(cpu_sibling_map, i), GFP_KERNEL); + zalloc_cpumask_var(&per_cpu(cpu_core_map, i), GFP_KERNEL); + zalloc_cpumask_var(&per_cpu(cpu_llc_shared_map, i), GFP_KERNEL); + } + set_cpu_sibling_map(0); + + if (xen_smp_intr_init(0)) + BUG(); + + if (!alloc_cpumask_var(&xen_cpu_initialized_map, GFP_KERNEL)) + panic("could not allocate xen_cpu_initialized_map\n"); + + cpumask_copy(xen_cpu_initialized_map, cpumask_of(0)); + + /* Restrict the possible_map according to max_cpus. */ + while ((num_possible_cpus() > 1) && (num_possible_cpus() > max_cpus)) { + for (cpu = nr_cpu_ids - 1; !cpu_possible(cpu); cpu--) + continue; + set_cpu_possible(cpu, false); + } + + for_each_possible_cpu(cpu) + set_cpu_present(cpu, true); +} + +static int +cpu_initialize_context(unsigned int cpu, struct task_struct *idle) +{ + struct vcpu_guest_context *ctxt; + struct desc_struct *gdt; + unsigned long gdt_mfn; + + /* used to tell cpu_init() that it can proceed with initialization */ + cpumask_set_cpu(cpu, cpu_callout_mask); + if (cpumask_test_and_set_cpu(cpu, xen_cpu_initialized_map)) + return 0; + + ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL); + if (ctxt == NULL) + return -ENOMEM; + + gdt = get_cpu_gdt_table(cpu); + +#ifdef CONFIG_X86_32 + /* Note: PVH is not yet supported on x86_32. */ + ctxt->user_regs.fs = __KERNEL_PERCPU; + ctxt->user_regs.gs = __KERNEL_STACK_CANARY; +#endif + memset(&ctxt->fpu_ctxt, 0, sizeof(ctxt->fpu_ctxt)); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle; + ctxt->flags = VGCF_IN_KERNEL; + ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */ + ctxt->user_regs.ds = __USER_DS; + ctxt->user_regs.es = __USER_DS; + ctxt->user_regs.ss = __KERNEL_DS; + + xen_copy_trap_info(ctxt->trap_ctxt); + + ctxt->ldt_ents = 0; + + BUG_ON((unsigned long)gdt & ~PAGE_MASK); + + gdt_mfn = arbitrary_virt_to_mfn(gdt); + make_lowmem_page_readonly(gdt); + make_lowmem_page_readonly(mfn_to_virt(gdt_mfn)); + + ctxt->gdt_frames[0] = gdt_mfn; + ctxt->gdt_ents = GDT_ENTRIES; + + ctxt->kernel_ss = __KERNEL_DS; + ctxt->kernel_sp = idle->thread.sp0; + +#ifdef CONFIG_X86_32 + ctxt->event_callback_cs = __KERNEL_CS; + ctxt->failsafe_callback_cs = __KERNEL_CS; +#else + ctxt->gs_base_kernel = per_cpu_offset(cpu); +#endif + ctxt->event_callback_eip = + (unsigned long)xen_hypervisor_callback; + ctxt->failsafe_callback_eip = + (unsigned long)xen_failsafe_callback; + ctxt->user_regs.cs = __KERNEL_CS; + per_cpu(xen_cr3, cpu) = __pa(swapper_pg_dir); + } +#ifdef CONFIG_XEN_PVH + else { + /* + * The vcpu comes on kernel page tables which have the NX pte + * bit set. This means before DS/SS is touched, NX in + * EFER must be set. Hence the following assembly glue code. + */ + ctxt->user_regs.eip = (unsigned long)xen_pvh_early_cpu_init; + ctxt->user_regs.rdi = cpu; + ctxt->user_regs.rsi = true; /* entry == true */ + } +#endif + ctxt->user_regs.esp = idle->thread.sp0 - sizeof(struct pt_regs); + ctxt->ctrlreg[3] = xen_pfn_to_cr3(virt_to_mfn(swapper_pg_dir)); + if (HYPERVISOR_vcpu_op(VCPUOP_initialise, cpu, ctxt)) + BUG(); + + kfree(ctxt); + return 0; +} + +static int xen_cpu_up(unsigned int cpu, struct task_struct *idle) +{ + int rc; + + common_cpu_up(cpu, idle); + + xen_setup_runstate_info(cpu); + xen_setup_timer(cpu); + xen_init_lock_cpu(cpu); + + /* + * PV VCPUs are always successfully taken down (see 'while' loop + * in xen_cpu_die()), so -EBUSY is an error. + */ + rc = cpu_check_up_prepare(cpu); + if (rc) + return rc; + + /* make sure interrupts start blocked */ + per_cpu(xen_vcpu, cpu)->evtchn_upcall_mask = 1; + + rc = cpu_initialize_context(cpu, idle); + if (rc) + return rc; + + rc = xen_smp_intr_init(cpu); + if (rc) + return rc; + + rc = HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL); + BUG_ON(rc); + + while (cpu_report_state(cpu) != CPU_ONLINE) + HYPERVISOR_sched_op(SCHEDOP_yield, NULL); + + return 0; +} + +static void xen_smp_cpus_done(unsigned int max_cpus) +{ +} + +#ifdef CONFIG_HOTPLUG_CPU +static int xen_cpu_disable(void) +{ + unsigned int cpu = smp_processor_id(); + if (cpu == 0) + return -EBUSY; + + cpu_disable_common(); + + load_cr3(swapper_pg_dir); + return 0; +} + +static void xen_cpu_die(unsigned int cpu) +{ + while (xen_pv_domain() && HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) { + __set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(HZ/10); + } + + if (common_cpu_die(cpu) == 0) { + xen_smp_intr_free(cpu); + xen_uninit_lock_cpu(cpu); + xen_teardown_timer(cpu); + } +} + +static void xen_play_dead(void) /* used only with HOTPLUG_CPU */ +{ + play_dead_common(); + HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL); + cpu_bringup(); + /* + * commit 4b0c0f294 (tick: Cleanup NOHZ per cpu data on cpu down) + * clears certain data that the cpu_idle loop (which called us + * and that we return from) expects. The only way to get that + * data back is to call: + */ + tick_nohz_idle_enter(); +} + +#else /* !CONFIG_HOTPLUG_CPU */ +static int xen_cpu_disable(void) +{ + return -ENOSYS; +} + +static void xen_cpu_die(unsigned int cpu) +{ + BUG(); +} + +static void xen_play_dead(void) +{ + BUG(); +} + +#endif +static void stop_self(void *v) +{ + int cpu = smp_processor_id(); + + /* make sure we're not pinning something down */ + load_cr3(swapper_pg_dir); + /* should set up a minimal gdt */ + + set_cpu_online(cpu, false); + + HYPERVISOR_vcpu_op(VCPUOP_down, cpu, NULL); + BUG(); +} + +static void xen_stop_other_cpus(int wait) +{ + smp_call_function(stop_self, NULL, wait); +} + +static void xen_smp_send_reschedule(int cpu) +{ + xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR); +} + +static void __xen_send_IPI_mask(const struct cpumask *mask, + int vector) +{ + unsigned cpu; + + for_each_cpu_and(cpu, mask, cpu_online_mask) + xen_send_IPI_one(cpu, vector); +} + +static void xen_smp_send_call_function_ipi(const struct cpumask *mask) +{ + int cpu; + + __xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR); + + /* Make sure other vcpus get a chance to run if they need to. */ + for_each_cpu(cpu, mask) { + if (xen_vcpu_stolen(cpu)) { + HYPERVISOR_sched_op(SCHEDOP_yield, NULL); + break; + } + } +} + +static void xen_smp_send_call_function_single_ipi(int cpu) +{ + __xen_send_IPI_mask(cpumask_of(cpu), + XEN_CALL_FUNCTION_SINGLE_VECTOR); +} + +static inline int xen_map_vector(int vector) +{ + int xen_vector; + + switch (vector) { + case RESCHEDULE_VECTOR: + xen_vector = XEN_RESCHEDULE_VECTOR; + break; + case CALL_FUNCTION_VECTOR: + xen_vector = XEN_CALL_FUNCTION_VECTOR; + break; + case CALL_FUNCTION_SINGLE_VECTOR: + xen_vector = XEN_CALL_FUNCTION_SINGLE_VECTOR; + break; + case IRQ_WORK_VECTOR: + xen_vector = XEN_IRQ_WORK_VECTOR; + break; +#ifdef CONFIG_X86_64 + case NMI_VECTOR: + case APIC_DM_NMI: /* Some use that instead of NMI_VECTOR */ + xen_vector = XEN_NMI_VECTOR; + break; +#endif + default: + xen_vector = -1; + printk(KERN_ERR "xen: vector 0x%x is not implemented\n", + vector); + } + + return xen_vector; +} + +void xen_send_IPI_mask(const struct cpumask *mask, + int vector) +{ + int xen_vector = xen_map_vector(vector); + + if (xen_vector >= 0) + __xen_send_IPI_mask(mask, xen_vector); +} + +void xen_send_IPI_all(int vector) +{ + int xen_vector = xen_map_vector(vector); + + if (xen_vector >= 0) + __xen_send_IPI_mask(cpu_online_mask, xen_vector); +} + +void xen_send_IPI_self(int vector) +{ + int xen_vector = xen_map_vector(vector); + + if (xen_vector >= 0) + xen_send_IPI_one(smp_processor_id(), xen_vector); +} + +void xen_send_IPI_mask_allbutself(const struct cpumask *mask, + int vector) +{ + unsigned cpu; + unsigned int this_cpu = smp_processor_id(); + int xen_vector = xen_map_vector(vector); + + if (!(num_online_cpus() > 1) || (xen_vector < 0)) + return; + + for_each_cpu_and(cpu, mask, cpu_online_mask) { + if (this_cpu == cpu) + continue; + + xen_send_IPI_one(cpu, xen_vector); + } +} + +void xen_send_IPI_allbutself(int vector) +{ + xen_send_IPI_mask_allbutself(cpu_online_mask, vector); +} + +static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id) +{ + irq_enter(); + generic_smp_call_function_interrupt(); + inc_irq_stat(irq_call_count); + irq_exit(); + + return IRQ_HANDLED; +} + +static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id) +{ + irq_enter(); + generic_smp_call_function_single_interrupt(); + inc_irq_stat(irq_call_count); + irq_exit(); + + return IRQ_HANDLED; +} + +static irqreturn_t xen_irq_work_interrupt(int irq, void *dev_id) +{ + irq_enter(); + irq_work_run(); + inc_irq_stat(apic_irq_work_irqs); + irq_exit(); + + return IRQ_HANDLED; +} + +static const struct smp_ops xen_smp_ops __initconst = { + .smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu, + .smp_prepare_cpus = xen_smp_prepare_cpus, + .smp_cpus_done = xen_smp_cpus_done, + + .cpu_up = xen_cpu_up, + .cpu_die = xen_cpu_die, + .cpu_disable = xen_cpu_disable, + .play_dead = xen_play_dead, + + .stop_other_cpus = xen_stop_other_cpus, + .smp_send_reschedule = xen_smp_send_reschedule, + + .send_call_func_ipi = xen_smp_send_call_function_ipi, + .send_call_func_single_ipi = xen_smp_send_call_function_single_ipi, +}; + +void __init xen_smp_init(void) +{ + smp_ops = xen_smp_ops; + xen_fill_possible_map(); +} + +static void __init xen_hvm_smp_prepare_cpus(unsigned int max_cpus) +{ + native_smp_prepare_cpus(max_cpus); + WARN_ON(xen_smp_intr_init(0)); + + xen_init_lock_cpu(0); +} + +static int xen_hvm_cpu_up(unsigned int cpu, struct task_struct *tidle) +{ + int rc; + + /* + * This can happen if CPU was offlined earlier and + * offlining timed out in common_cpu_die(). + */ + if (cpu_report_state(cpu) == CPU_DEAD_FROZEN) { + xen_smp_intr_free(cpu); + xen_uninit_lock_cpu(cpu); + } + + /* + * xen_smp_intr_init() needs to run before native_cpu_up() + * so that IPI vectors are set up on the booting CPU before + * it is marked online in native_cpu_up(). + */ + rc = xen_smp_intr_init(cpu); + WARN_ON(rc); + if (!rc) + rc = native_cpu_up(cpu, tidle); + + /* + * We must initialize the slowpath CPU kicker _after_ the native + * path has executed. If we initialized it before none of the + * unlocker IPI kicks would reach the booting CPU as the booting + * CPU had not set itself 'online' in cpu_online_mask. That mask + * is checked when IPIs are sent (on HVM at least). + */ + xen_init_lock_cpu(cpu); + return rc; +} + +void __init xen_hvm_smp_init(void) +{ + if (!xen_have_vector_callback) + return; + smp_ops.smp_prepare_cpus = xen_hvm_smp_prepare_cpus; + smp_ops.smp_send_reschedule = xen_smp_send_reschedule; + smp_ops.cpu_up = xen_hvm_cpu_up; + smp_ops.cpu_die = xen_cpu_die; + smp_ops.send_call_func_ipi = xen_smp_send_call_function_ipi; + smp_ops.send_call_func_single_ipi = xen_smp_send_call_function_single_ipi; + smp_ops.smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu; +} diff --git a/kernel/arch/x86/xen/smp.h b/kernel/arch/x86/xen/smp.h new file mode 100644 index 000000000..963d62a35 --- /dev/null +++ b/kernel/arch/x86/xen/smp.h @@ -0,0 +1,19 @@ +#ifndef _XEN_SMP_H + +extern void xen_send_IPI_mask(const struct cpumask *mask, + int vector); +extern void xen_send_IPI_mask_allbutself(const struct cpumask *mask, + int vector); +extern void xen_send_IPI_allbutself(int vector); +extern void xen_send_IPI_all(int vector); +extern void xen_send_IPI_self(int vector); + +#ifdef CONFIG_XEN_PVH +extern void xen_pvh_early_cpu_init(int cpu, bool entry); +#else +static inline void xen_pvh_early_cpu_init(int cpu, bool entry) +{ +} +#endif + +#endif diff --git a/kernel/arch/x86/xen/spinlock.c b/kernel/arch/x86/xen/spinlock.c new file mode 100644 index 000000000..956374c1e --- /dev/null +++ b/kernel/arch/x86/xen/spinlock.c @@ -0,0 +1,353 @@ +/* + * Split spinlock implementation out into its own file, so it can be + * compiled in a FTRACE-compatible way. + */ +#include <linux/kernel_stat.h> +#include <linux/spinlock.h> +#include <linux/debugfs.h> +#include <linux/log2.h> +#include <linux/gfp.h> +#include <linux/slab.h> + +#include <asm/paravirt.h> + +#include <xen/interface/xen.h> +#include <xen/events.h> + +#include "xen-ops.h" +#include "debugfs.h" + +enum xen_contention_stat { + TAKEN_SLOW, + TAKEN_SLOW_PICKUP, + TAKEN_SLOW_SPURIOUS, + RELEASED_SLOW, + RELEASED_SLOW_KICKED, + NR_CONTENTION_STATS +}; + + +#ifdef CONFIG_XEN_DEBUG_FS +#define HISTO_BUCKETS 30 +static struct xen_spinlock_stats +{ + u32 contention_stats[NR_CONTENTION_STATS]; + u32 histo_spin_blocked[HISTO_BUCKETS+1]; + u64 time_blocked; +} spinlock_stats; + +static u8 zero_stats; + +static inline void check_zero(void) +{ + u8 ret; + u8 old = READ_ONCE(zero_stats); + if (unlikely(old)) { + ret = cmpxchg(&zero_stats, old, 0); + /* This ensures only one fellow resets the stat */ + if (ret == old) + memset(&spinlock_stats, 0, sizeof(spinlock_stats)); + } +} + +static inline void add_stats(enum xen_contention_stat var, u32 val) +{ + check_zero(); + spinlock_stats.contention_stats[var] += val; +} + +static inline u64 spin_time_start(void) +{ + return xen_clocksource_read(); +} + +static void __spin_time_accum(u64 delta, u32 *array) +{ + unsigned index = ilog2(delta); + + check_zero(); + + if (index < HISTO_BUCKETS) + array[index]++; + else + array[HISTO_BUCKETS]++; +} + +static inline void spin_time_accum_blocked(u64 start) +{ + u32 delta = xen_clocksource_read() - start; + + __spin_time_accum(delta, spinlock_stats.histo_spin_blocked); + spinlock_stats.time_blocked += delta; +} +#else /* !CONFIG_XEN_DEBUG_FS */ +static inline void add_stats(enum xen_contention_stat var, u32 val) +{ +} + +static inline u64 spin_time_start(void) +{ + return 0; +} + +static inline void spin_time_accum_blocked(u64 start) +{ +} +#endif /* CONFIG_XEN_DEBUG_FS */ + +struct xen_lock_waiting { + struct arch_spinlock *lock; + __ticket_t want; +}; + +static DEFINE_PER_CPU(int, lock_kicker_irq) = -1; +static DEFINE_PER_CPU(char *, irq_name); +static DEFINE_PER_CPU(struct xen_lock_waiting, lock_waiting); +static cpumask_t waiting_cpus; + +static bool xen_pvspin = true; +__visible void xen_lock_spinning(struct arch_spinlock *lock, __ticket_t want) +{ + int irq = __this_cpu_read(lock_kicker_irq); + struct xen_lock_waiting *w = this_cpu_ptr(&lock_waiting); + int cpu = smp_processor_id(); + u64 start; + __ticket_t head; + unsigned long flags; + + /* If kicker interrupts not initialized yet, just spin */ + if (irq == -1) + return; + + start = spin_time_start(); + + /* + * Make sure an interrupt handler can't upset things in a + * partially setup state. + */ + local_irq_save(flags); + /* + * We don't really care if we're overwriting some other + * (lock,want) pair, as that would mean that we're currently + * in an interrupt context, and the outer context had + * interrupts enabled. That has already kicked the VCPU out + * of xen_poll_irq(), so it will just return spuriously and + * retry with newly setup (lock,want). + * + * The ordering protocol on this is that the "lock" pointer + * may only be set non-NULL if the "want" ticket is correct. + * If we're updating "want", we must first clear "lock". + */ + w->lock = NULL; + smp_wmb(); + w->want = want; + smp_wmb(); + w->lock = lock; + + /* This uses set_bit, which atomic and therefore a barrier */ + cpumask_set_cpu(cpu, &waiting_cpus); + add_stats(TAKEN_SLOW, 1); + + /* clear pending */ + xen_clear_irq_pending(irq); + + /* Only check lock once pending cleared */ + barrier(); + + /* + * Mark entry to slowpath before doing the pickup test to make + * sure we don't deadlock with an unlocker. + */ + __ticket_enter_slowpath(lock); + + /* make sure enter_slowpath, which is atomic does not cross the read */ + smp_mb__after_atomic(); + + /* + * check again make sure it didn't become free while + * we weren't looking + */ + head = READ_ONCE(lock->tickets.head); + if (__tickets_equal(head, want)) { + add_stats(TAKEN_SLOW_PICKUP, 1); + goto out; + } + + /* Allow interrupts while blocked */ + local_irq_restore(flags); + + /* + * If an interrupt happens here, it will leave the wakeup irq + * pending, which will cause xen_poll_irq() to return + * immediately. + */ + + /* Block until irq becomes pending (or perhaps a spurious wakeup) */ + xen_poll_irq(irq); + add_stats(TAKEN_SLOW_SPURIOUS, !xen_test_irq_pending(irq)); + + local_irq_save(flags); + + kstat_incr_irq_this_cpu(irq); +out: + cpumask_clear_cpu(cpu, &waiting_cpus); + w->lock = NULL; + + local_irq_restore(flags); + + spin_time_accum_blocked(start); +} +PV_CALLEE_SAVE_REGS_THUNK(xen_lock_spinning); + +static void xen_unlock_kick(struct arch_spinlock *lock, __ticket_t next) +{ + int cpu; + + add_stats(RELEASED_SLOW, 1); + + for_each_cpu(cpu, &waiting_cpus) { + const struct xen_lock_waiting *w = &per_cpu(lock_waiting, cpu); + + /* Make sure we read lock before want */ + if (READ_ONCE(w->lock) == lock && + READ_ONCE(w->want) == next) { + add_stats(RELEASED_SLOW_KICKED, 1); + xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR); + break; + } + } +} + +static irqreturn_t dummy_handler(int irq, void *dev_id) +{ + BUG(); + return IRQ_HANDLED; +} + +void xen_init_lock_cpu(int cpu) +{ + int irq; + char *name; + + if (!xen_pvspin) + return; + + WARN(per_cpu(lock_kicker_irq, cpu) >= 0, "spinlock on CPU%d exists on IRQ%d!\n", + cpu, per_cpu(lock_kicker_irq, cpu)); + + name = kasprintf(GFP_KERNEL, "spinlock%d", cpu); + irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR, + cpu, + dummy_handler, + IRQF_PERCPU|IRQF_NOBALANCING, + name, + NULL); + + if (irq >= 0) { + disable_irq(irq); /* make sure it's never delivered */ + per_cpu(lock_kicker_irq, cpu) = irq; + per_cpu(irq_name, cpu) = name; + } + + printk("cpu %d spinlock event irq %d\n", cpu, irq); +} + +void xen_uninit_lock_cpu(int cpu) +{ + if (!xen_pvspin) + return; + + unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL); + per_cpu(lock_kicker_irq, cpu) = -1; + kfree(per_cpu(irq_name, cpu)); + per_cpu(irq_name, cpu) = NULL; +} + + +/* + * Our init of PV spinlocks is split in two init functions due to us + * using paravirt patching and jump labels patching and having to do + * all of this before SMP code is invoked. + * + * The paravirt patching needs to be done _before_ the alternative asm code + * is started, otherwise we would not patch the core kernel code. + */ +void __init xen_init_spinlocks(void) +{ + + if (!xen_pvspin) { + printk(KERN_DEBUG "xen: PV spinlocks disabled\n"); + return; + } + printk(KERN_DEBUG "xen: PV spinlocks enabled\n"); + pv_lock_ops.lock_spinning = PV_CALLEE_SAVE(xen_lock_spinning); + pv_lock_ops.unlock_kick = xen_unlock_kick; +} + +/* + * While the jump_label init code needs to happend _after_ the jump labels are + * enabled and before SMP is started. Hence we use pre-SMP initcall level + * init. We cannot do it in xen_init_spinlocks as that is done before + * jump labels are activated. + */ +static __init int xen_init_spinlocks_jump(void) +{ + if (!xen_pvspin) + return 0; + + if (!xen_domain()) + return 0; + + static_key_slow_inc(¶virt_ticketlocks_enabled); + return 0; +} +early_initcall(xen_init_spinlocks_jump); + +static __init int xen_parse_nopvspin(char *arg) +{ + xen_pvspin = false; + return 0; +} +early_param("xen_nopvspin", xen_parse_nopvspin); + +#ifdef CONFIG_XEN_DEBUG_FS + +static struct dentry *d_spin_debug; + +static int __init xen_spinlock_debugfs(void) +{ + struct dentry *d_xen = xen_init_debugfs(); + + if (d_xen == NULL) + return -ENOMEM; + + if (!xen_pvspin) + return 0; + + d_spin_debug = debugfs_create_dir("spinlocks", d_xen); + + debugfs_create_u8("zero_stats", 0644, d_spin_debug, &zero_stats); + + debugfs_create_u32("taken_slow", 0444, d_spin_debug, + &spinlock_stats.contention_stats[TAKEN_SLOW]); + debugfs_create_u32("taken_slow_pickup", 0444, d_spin_debug, + &spinlock_stats.contention_stats[TAKEN_SLOW_PICKUP]); + debugfs_create_u32("taken_slow_spurious", 0444, d_spin_debug, + &spinlock_stats.contention_stats[TAKEN_SLOW_SPURIOUS]); + + debugfs_create_u32("released_slow", 0444, d_spin_debug, + &spinlock_stats.contention_stats[RELEASED_SLOW]); + debugfs_create_u32("released_slow_kicked", 0444, d_spin_debug, + &spinlock_stats.contention_stats[RELEASED_SLOW_KICKED]); + + debugfs_create_u64("time_blocked", 0444, d_spin_debug, + &spinlock_stats.time_blocked); + + debugfs_create_u32_array("histo_blocked", 0444, d_spin_debug, + spinlock_stats.histo_spin_blocked, HISTO_BUCKETS + 1); + + return 0; +} +fs_initcall(xen_spinlock_debugfs); + +#endif /* CONFIG_XEN_DEBUG_FS */ diff --git a/kernel/arch/x86/xen/suspend.c b/kernel/arch/x86/xen/suspend.c new file mode 100644 index 000000000..53b4c0811 --- /dev/null +++ b/kernel/arch/x86/xen/suspend.c @@ -0,0 +1,104 @@ +#include <linux/types.h> +#include <linux/tick.h> + +#include <xen/interface/xen.h> +#include <xen/grant_table.h> +#include <xen/events.h> + +#include <asm/xen/hypercall.h> +#include <asm/xen/page.h> +#include <asm/fixmap.h> + +#include "xen-ops.h" +#include "mmu.h" + +static void xen_pv_pre_suspend(void) +{ + xen_mm_pin_all(); + + xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn); + xen_start_info->console.domU.mfn = + mfn_to_pfn(xen_start_info->console.domU.mfn); + + BUG_ON(!irqs_disabled()); + + HYPERVISOR_shared_info = &xen_dummy_shared_info; + if (HYPERVISOR_update_va_mapping(fix_to_virt(FIX_PARAVIRT_BOOTMAP), + __pte_ma(0), 0)) + BUG(); +} + +static void xen_hvm_post_suspend(int suspend_cancelled) +{ +#ifdef CONFIG_XEN_PVHVM + int cpu; + xen_hvm_init_shared_info(); + xen_callback_vector(); + xen_unplug_emulated_devices(); + if (xen_feature(XENFEAT_hvm_safe_pvclock)) { + for_each_online_cpu(cpu) { + xen_setup_runstate_info(cpu); + } + } +#endif +} + +static void xen_pv_post_suspend(int suspend_cancelled) +{ + xen_build_mfn_list_list(); + + xen_setup_shared_info(); + + if (suspend_cancelled) { + xen_start_info->store_mfn = + pfn_to_mfn(xen_start_info->store_mfn); + xen_start_info->console.domU.mfn = + pfn_to_mfn(xen_start_info->console.domU.mfn); + } else { +#ifdef CONFIG_SMP + BUG_ON(xen_cpu_initialized_map == NULL); + cpumask_copy(xen_cpu_initialized_map, cpu_online_mask); +#endif + xen_vcpu_restore(); + } + + xen_mm_unpin_all(); +} + +void xen_arch_pre_suspend(void) +{ + if (xen_pv_domain()) + xen_pv_pre_suspend(); +} + +void xen_arch_post_suspend(int cancelled) +{ + if (xen_pv_domain()) + xen_pv_post_suspend(cancelled); + else + xen_hvm_post_suspend(cancelled); +} + +static void xen_vcpu_notify_restore(void *data) +{ + /* Boot processor notified via generic timekeeping_resume() */ + if (smp_processor_id() == 0) + return; + + tick_resume_local(); +} + +static void xen_vcpu_notify_suspend(void *data) +{ + tick_suspend_local(); +} + +void xen_arch_resume(void) +{ + on_each_cpu(xen_vcpu_notify_restore, NULL, 1); +} + +void xen_arch_suspend(void) +{ + on_each_cpu(xen_vcpu_notify_suspend, NULL, 1); +} diff --git a/kernel/arch/x86/xen/time.c b/kernel/arch/x86/xen/time.c new file mode 100644 index 000000000..55da33b1d --- /dev/null +++ b/kernel/arch/x86/xen/time.c @@ -0,0 +1,558 @@ +/* + * Xen time implementation. + * + * This is implemented in terms of a clocksource driver which uses + * the hypervisor clock as a nanosecond timebase, and a clockevent + * driver which uses the hypervisor's timer mechanism. + * + * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007 + */ +#include <linux/kernel.h> +#include <linux/interrupt.h> +#include <linux/clocksource.h> +#include <linux/clockchips.h> +#include <linux/kernel_stat.h> +#include <linux/math64.h> +#include <linux/gfp.h> +#include <linux/slab.h> +#include <linux/pvclock_gtod.h> + +#include <asm/pvclock.h> +#include <asm/xen/hypervisor.h> +#include <asm/xen/hypercall.h> + +#include <xen/events.h> +#include <xen/features.h> +#include <xen/interface/xen.h> +#include <xen/interface/vcpu.h> + +#include "xen-ops.h" + +/* Xen may fire a timer up to this many ns early */ +#define TIMER_SLOP 100000 +#define NS_PER_TICK (1000000000LL / HZ) + +/* runstate info updated by Xen */ +static DEFINE_PER_CPU(struct vcpu_runstate_info, xen_runstate); + +/* snapshots of runstate info */ +static DEFINE_PER_CPU(struct vcpu_runstate_info, xen_runstate_snapshot); + +/* unused ns of stolen time */ +static DEFINE_PER_CPU(u64, xen_residual_stolen); + +/* return an consistent snapshot of 64-bit time/counter value */ +static u64 get64(const u64 *p) +{ + u64 ret; + + if (BITS_PER_LONG < 64) { + u32 *p32 = (u32 *)p; + u32 h, l; + + /* + * Read high then low, and then make sure high is + * still the same; this will only loop if low wraps + * and carries into high. + * XXX some clean way to make this endian-proof? + */ + do { + h = p32[1]; + barrier(); + l = p32[0]; + barrier(); + } while (p32[1] != h); + + ret = (((u64)h) << 32) | l; + } else + ret = *p; + + return ret; +} + +/* + * Runstate accounting + */ +static void get_runstate_snapshot(struct vcpu_runstate_info *res) +{ + u64 state_time; + struct vcpu_runstate_info *state; + + BUG_ON(preemptible()); + + state = this_cpu_ptr(&xen_runstate); + + /* + * The runstate info is always updated by the hypervisor on + * the current CPU, so there's no need to use anything + * stronger than a compiler barrier when fetching it. + */ + do { + state_time = get64(&state->state_entry_time); + barrier(); + *res = *state; + barrier(); + } while (get64(&state->state_entry_time) != state_time); +} + +/* return true when a vcpu could run but has no real cpu to run on */ +bool xen_vcpu_stolen(int vcpu) +{ + return per_cpu(xen_runstate, vcpu).state == RUNSTATE_runnable; +} + +void xen_setup_runstate_info(int cpu) +{ + struct vcpu_register_runstate_memory_area area; + + area.addr.v = &per_cpu(xen_runstate, cpu); + + if (HYPERVISOR_vcpu_op(VCPUOP_register_runstate_memory_area, + cpu, &area)) + BUG(); +} + +static void do_stolen_accounting(void) +{ + struct vcpu_runstate_info state; + struct vcpu_runstate_info *snap; + s64 runnable, offline, stolen; + cputime_t ticks; + + get_runstate_snapshot(&state); + + WARN_ON(state.state != RUNSTATE_running); + + snap = this_cpu_ptr(&xen_runstate_snapshot); + + /* work out how much time the VCPU has not been runn*ing* */ + runnable = state.time[RUNSTATE_runnable] - snap->time[RUNSTATE_runnable]; + offline = state.time[RUNSTATE_offline] - snap->time[RUNSTATE_offline]; + + *snap = state; + + /* Add the appropriate number of ticks of stolen time, + including any left-overs from last time. */ + stolen = runnable + offline + __this_cpu_read(xen_residual_stolen); + + if (stolen < 0) + stolen = 0; + + ticks = iter_div_u64_rem(stolen, NS_PER_TICK, &stolen); + __this_cpu_write(xen_residual_stolen, stolen); + account_steal_ticks(ticks); +} + +/* Get the TSC speed from Xen */ +static unsigned long xen_tsc_khz(void) +{ + struct pvclock_vcpu_time_info *info = + &HYPERVISOR_shared_info->vcpu_info[0].time; + + return pvclock_tsc_khz(info); +} + +cycle_t xen_clocksource_read(void) +{ + struct pvclock_vcpu_time_info *src; + cycle_t ret; + + preempt_disable_notrace(); + src = &__this_cpu_read(xen_vcpu)->time; + ret = pvclock_clocksource_read(src); + preempt_enable_notrace(); + return ret; +} + +static cycle_t xen_clocksource_get_cycles(struct clocksource *cs) +{ + return xen_clocksource_read(); +} + +static void xen_read_wallclock(struct timespec *ts) +{ + struct shared_info *s = HYPERVISOR_shared_info; + struct pvclock_wall_clock *wall_clock = &(s->wc); + struct pvclock_vcpu_time_info *vcpu_time; + + vcpu_time = &get_cpu_var(xen_vcpu)->time; + pvclock_read_wallclock(wall_clock, vcpu_time, ts); + put_cpu_var(xen_vcpu); +} + +static void xen_get_wallclock(struct timespec *now) +{ + xen_read_wallclock(now); +} + +static int xen_set_wallclock(const struct timespec *now) +{ + return -1; +} + +static int xen_pvclock_gtod_notify(struct notifier_block *nb, + unsigned long was_set, void *priv) +{ + /* Protected by the calling core code serialization */ + static struct timespec next_sync; + + struct xen_platform_op op; + struct timespec now; + + now = __current_kernel_time(); + + /* + * We only take the expensive HV call when the clock was set + * or when the 11 minutes RTC synchronization time elapsed. + */ + if (!was_set && timespec_compare(&now, &next_sync) < 0) + return NOTIFY_OK; + + op.cmd = XENPF_settime; + op.u.settime.secs = now.tv_sec; + op.u.settime.nsecs = now.tv_nsec; + op.u.settime.system_time = xen_clocksource_read(); + + (void)HYPERVISOR_dom0_op(&op); + + /* + * Move the next drift compensation time 11 minutes + * ahead. That's emulating the sync_cmos_clock() update for + * the hardware RTC. + */ + next_sync = now; + next_sync.tv_sec += 11 * 60; + + return NOTIFY_OK; +} + +static struct notifier_block xen_pvclock_gtod_notifier = { + .notifier_call = xen_pvclock_gtod_notify, +}; + +static struct clocksource xen_clocksource __read_mostly = { + .name = "xen", + .rating = 400, + .read = xen_clocksource_get_cycles, + .mask = ~0, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; + +/* + Xen clockevent implementation + + Xen has two clockevent implementations: + + The old timer_op one works with all released versions of Xen prior + to version 3.0.4. This version of the hypervisor provides a + single-shot timer with nanosecond resolution. However, sharing the + same event channel is a 100Hz tick which is delivered while the + vcpu is running. We don't care about or use this tick, but it will + cause the core time code to think the timer fired too soon, and + will end up resetting it each time. It could be filtered, but + doing so has complications when the ktime clocksource is not yet + the xen clocksource (ie, at boot time). + + The new vcpu_op-based timer interface allows the tick timer period + to be changed or turned off. The tick timer is not useful as a + periodic timer because events are only delivered to running vcpus. + The one-shot timer can report when a timeout is in the past, so + set_next_event is capable of returning -ETIME when appropriate. + This interface is used when available. +*/ + + +/* + Get a hypervisor absolute time. In theory we could maintain an + offset between the kernel's time and the hypervisor's time, and + apply that to a kernel's absolute timeout. Unfortunately the + hypervisor and kernel times can drift even if the kernel is using + the Xen clocksource, because ntp can warp the kernel's clocksource. +*/ +static s64 get_abs_timeout(unsigned long delta) +{ + return xen_clocksource_read() + delta; +} + +static void xen_timerop_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + /* unsupported */ + WARN_ON(1); + break; + + case CLOCK_EVT_MODE_ONESHOT: + case CLOCK_EVT_MODE_RESUME: + break; + + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + HYPERVISOR_set_timer_op(0); /* cancel timeout */ + break; + } +} + +static int xen_timerop_set_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + WARN_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT); + + if (HYPERVISOR_set_timer_op(get_abs_timeout(delta)) < 0) + BUG(); + + /* We may have missed the deadline, but there's no real way of + knowing for sure. If the event was in the past, then we'll + get an immediate interrupt. */ + + return 0; +} + +static const struct clock_event_device xen_timerop_clockevent = { + .name = "xen", + .features = CLOCK_EVT_FEAT_ONESHOT, + + .max_delta_ns = 0xffffffff, + .min_delta_ns = TIMER_SLOP, + + .mult = 1, + .shift = 0, + .rating = 500, + + .set_mode = xen_timerop_set_mode, + .set_next_event = xen_timerop_set_next_event, +}; + + + +static void xen_vcpuop_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + int cpu = smp_processor_id(); + + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + WARN_ON(1); /* unsupported */ + break; + + case CLOCK_EVT_MODE_ONESHOT: + if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL)) + BUG(); + break; + + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + if (HYPERVISOR_vcpu_op(VCPUOP_stop_singleshot_timer, cpu, NULL) || + HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL)) + BUG(); + break; + case CLOCK_EVT_MODE_RESUME: + break; + } +} + +static int xen_vcpuop_set_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + int cpu = smp_processor_id(); + struct vcpu_set_singleshot_timer single; + int ret; + + WARN_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT); + + single.timeout_abs_ns = get_abs_timeout(delta); + single.flags = VCPU_SSHOTTMR_future; + + ret = HYPERVISOR_vcpu_op(VCPUOP_set_singleshot_timer, cpu, &single); + + BUG_ON(ret != 0 && ret != -ETIME); + + return ret; +} + +static const struct clock_event_device xen_vcpuop_clockevent = { + .name = "xen", + .features = CLOCK_EVT_FEAT_ONESHOT, + + .max_delta_ns = 0xffffffff, + .min_delta_ns = TIMER_SLOP, + + .mult = 1, + .shift = 0, + .rating = 500, + + .set_mode = xen_vcpuop_set_mode, + .set_next_event = xen_vcpuop_set_next_event, +}; + +static const struct clock_event_device *xen_clockevent = + &xen_timerop_clockevent; + +struct xen_clock_event_device { + struct clock_event_device evt; + char name[16]; +}; +static DEFINE_PER_CPU(struct xen_clock_event_device, xen_clock_events) = { .evt.irq = -1 }; + +static irqreturn_t xen_timer_interrupt(int irq, void *dev_id) +{ + struct clock_event_device *evt = this_cpu_ptr(&xen_clock_events.evt); + irqreturn_t ret; + + ret = IRQ_NONE; + if (evt->event_handler) { + evt->event_handler(evt); + ret = IRQ_HANDLED; + } + + do_stolen_accounting(); + + return ret; +} + +void xen_teardown_timer(int cpu) +{ + struct clock_event_device *evt; + BUG_ON(cpu == 0); + evt = &per_cpu(xen_clock_events, cpu).evt; + + if (evt->irq >= 0) { + unbind_from_irqhandler(evt->irq, NULL); + evt->irq = -1; + } +} + +void xen_setup_timer(int cpu) +{ + struct xen_clock_event_device *xevt = &per_cpu(xen_clock_events, cpu); + struct clock_event_device *evt = &xevt->evt; + int irq; + + WARN(evt->irq >= 0, "IRQ%d for CPU%d is already allocated\n", evt->irq, cpu); + if (evt->irq >= 0) + xen_teardown_timer(cpu); + + printk(KERN_INFO "installing Xen timer for CPU %d\n", cpu); + + snprintf(xevt->name, sizeof(xevt->name), "timer%d", cpu); + + irq = bind_virq_to_irqhandler(VIRQ_TIMER, cpu, xen_timer_interrupt, + IRQF_PERCPU|IRQF_NOBALANCING|IRQF_TIMER| + IRQF_FORCE_RESUME|IRQF_EARLY_RESUME, + xevt->name, NULL); + (void)xen_set_irq_priority(irq, XEN_IRQ_PRIORITY_MAX); + + memcpy(evt, xen_clockevent, sizeof(*evt)); + + evt->cpumask = cpumask_of(cpu); + evt->irq = irq; +} + + +void xen_setup_cpu_clockevents(void) +{ + clockevents_register_device(this_cpu_ptr(&xen_clock_events.evt)); +} + +void xen_timer_resume(void) +{ + int cpu; + + pvclock_resume(); + + if (xen_clockevent != &xen_vcpuop_clockevent) + return; + + for_each_online_cpu(cpu) { + if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL)) + BUG(); + } +} + +static const struct pv_time_ops xen_time_ops __initconst = { + .sched_clock = xen_clocksource_read, +}; + +static void __init xen_time_init(void) +{ + int cpu = smp_processor_id(); + struct timespec tp; + + /* As Dom0 is never moved, no penalty on using TSC there */ + if (xen_initial_domain()) + xen_clocksource.rating = 275; + + clocksource_register_hz(&xen_clocksource, NSEC_PER_SEC); + + if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) { + /* Successfully turned off 100Hz tick, so we have the + vcpuop-based timer interface */ + printk(KERN_DEBUG "Xen: using vcpuop timer interface\n"); + xen_clockevent = &xen_vcpuop_clockevent; + } + + /* Set initial system time with full resolution */ + xen_read_wallclock(&tp); + do_settimeofday(&tp); + + setup_force_cpu_cap(X86_FEATURE_TSC); + + xen_setup_runstate_info(cpu); + xen_setup_timer(cpu); + xen_setup_cpu_clockevents(); + + if (xen_initial_domain()) + pvclock_gtod_register_notifier(&xen_pvclock_gtod_notifier); +} + +void __init xen_init_time_ops(void) +{ + pv_time_ops = xen_time_ops; + + x86_init.timers.timer_init = xen_time_init; + x86_init.timers.setup_percpu_clockev = x86_init_noop; + x86_cpuinit.setup_percpu_clockev = x86_init_noop; + + x86_platform.calibrate_tsc = xen_tsc_khz; + x86_platform.get_wallclock = xen_get_wallclock; + /* Dom0 uses the native method to set the hardware RTC. */ + if (!xen_initial_domain()) + x86_platform.set_wallclock = xen_set_wallclock; +} + +#ifdef CONFIG_XEN_PVHVM +static void xen_hvm_setup_cpu_clockevents(void) +{ + int cpu = smp_processor_id(); + xen_setup_runstate_info(cpu); + /* + * xen_setup_timer(cpu) - snprintf is bad in atomic context. Hence + * doing it xen_hvm_cpu_notify (which gets called by smp_init during + * early bootup and also during CPU hotplug events). + */ + xen_setup_cpu_clockevents(); +} + +void __init xen_hvm_init_time_ops(void) +{ + /* vector callback is needed otherwise we cannot receive interrupts + * on cpu > 0 and at this point we don't know how many cpus are + * available */ + if (!xen_have_vector_callback) + return; + if (!xen_feature(XENFEAT_hvm_safe_pvclock)) { + printk(KERN_INFO "Xen doesn't support pvclock on HVM," + "disable pv timer\n"); + return; + } + + pv_time_ops = xen_time_ops; + x86_init.timers.setup_percpu_clockev = xen_time_init; + x86_cpuinit.setup_percpu_clockev = xen_hvm_setup_cpu_clockevents; + + x86_platform.calibrate_tsc = xen_tsc_khz; + x86_platform.get_wallclock = xen_get_wallclock; + x86_platform.set_wallclock = xen_set_wallclock; +} +#endif diff --git a/kernel/arch/x86/xen/trace.c b/kernel/arch/x86/xen/trace.c new file mode 100644 index 000000000..a702ec2f5 --- /dev/null +++ b/kernel/arch/x86/xen/trace.c @@ -0,0 +1,20 @@ +#include <linux/ftrace.h> +#include <xen/interface/xen.h> +#include <xen/interface/xen-mca.h> + +#define HYPERCALL(x) [__HYPERVISOR_##x] = "("#x")", +static const char *xen_hypercall_names[] = { +#include <asm/xen-hypercalls.h> +}; +#undef HYPERCALL + +static const char *xen_hypercall_name(unsigned op) +{ + if (op < ARRAY_SIZE(xen_hypercall_names) && xen_hypercall_names[op] != NULL) + return xen_hypercall_names[op]; + + return ""; +} + +#define CREATE_TRACE_POINTS +#include <trace/events/xen.h> diff --git a/kernel/arch/x86/xen/vdso.h b/kernel/arch/x86/xen/vdso.h new file mode 100644 index 000000000..861fedfe5 --- /dev/null +++ b/kernel/arch/x86/xen/vdso.h @@ -0,0 +1,4 @@ +/* Bit used for the pseudo-hwcap for non-negative segments. We use + bit 1 to avoid bugs in some versions of glibc when bit 0 is + used; the choice is otherwise arbitrary. */ +#define VDSO_NOTE_NONEGSEG_BIT 1 diff --git a/kernel/arch/x86/xen/vga.c b/kernel/arch/x86/xen/vga.c new file mode 100644 index 000000000..6722e3733 --- /dev/null +++ b/kernel/arch/x86/xen/vga.c @@ -0,0 +1,74 @@ +#include <linux/screen_info.h> +#include <linux/init.h> + +#include <asm/bootparam.h> +#include <asm/setup.h> + +#include <xen/interface/xen.h> + +#include "xen-ops.h" + +void __init xen_init_vga(const struct dom0_vga_console_info *info, size_t size) +{ + struct screen_info *screen_info = &boot_params.screen_info; + + /* This is drawn from a dump from vgacon:startup in + * standard Linux. */ + screen_info->orig_video_mode = 3; + screen_info->orig_video_isVGA = 1; + screen_info->orig_video_lines = 25; + screen_info->orig_video_cols = 80; + screen_info->orig_video_ega_bx = 3; + screen_info->orig_video_points = 16; + screen_info->orig_y = screen_info->orig_video_lines - 1; + + switch (info->video_type) { + case XEN_VGATYPE_TEXT_MODE_3: + if (size < offsetof(struct dom0_vga_console_info, u.text_mode_3) + + sizeof(info->u.text_mode_3)) + break; + screen_info->orig_video_lines = info->u.text_mode_3.rows; + screen_info->orig_video_cols = info->u.text_mode_3.columns; + screen_info->orig_x = info->u.text_mode_3.cursor_x; + screen_info->orig_y = info->u.text_mode_3.cursor_y; + screen_info->orig_video_points = + info->u.text_mode_3.font_height; + break; + + case XEN_VGATYPE_EFI_LFB: + case XEN_VGATYPE_VESA_LFB: + if (size < offsetof(struct dom0_vga_console_info, + u.vesa_lfb.gbl_caps)) + break; + screen_info->orig_video_isVGA = VIDEO_TYPE_VLFB; + screen_info->lfb_width = info->u.vesa_lfb.width; + screen_info->lfb_height = info->u.vesa_lfb.height; + screen_info->lfb_depth = info->u.vesa_lfb.bits_per_pixel; + screen_info->lfb_base = info->u.vesa_lfb.lfb_base; + screen_info->lfb_size = info->u.vesa_lfb.lfb_size; + screen_info->lfb_linelength = info->u.vesa_lfb.bytes_per_line; + screen_info->red_size = info->u.vesa_lfb.red_size; + screen_info->red_pos = info->u.vesa_lfb.red_pos; + screen_info->green_size = info->u.vesa_lfb.green_size; + screen_info->green_pos = info->u.vesa_lfb.green_pos; + screen_info->blue_size = info->u.vesa_lfb.blue_size; + screen_info->blue_pos = info->u.vesa_lfb.blue_pos; + screen_info->rsvd_size = info->u.vesa_lfb.rsvd_size; + screen_info->rsvd_pos = info->u.vesa_lfb.rsvd_pos; + + if (info->video_type == XEN_VGATYPE_EFI_LFB) { + screen_info->orig_video_isVGA = VIDEO_TYPE_EFI; + break; + } + + if (size >= offsetof(struct dom0_vga_console_info, + u.vesa_lfb.gbl_caps) + + sizeof(info->u.vesa_lfb.gbl_caps)) + screen_info->capabilities = info->u.vesa_lfb.gbl_caps; + if (size >= offsetof(struct dom0_vga_console_info, + u.vesa_lfb.mode_attrs) + + sizeof(info->u.vesa_lfb.mode_attrs)) + screen_info->vesa_attributes = info->u.vesa_lfb.mode_attrs; + break; + } +} diff --git a/kernel/arch/x86/xen/xen-asm.S b/kernel/arch/x86/xen/xen-asm.S new file mode 100644 index 000000000..3e45aa000 --- /dev/null +++ b/kernel/arch/x86/xen/xen-asm.S @@ -0,0 +1,142 @@ +/* + * Asm versions of Xen pv-ops, suitable for either direct use or + * inlining. The inline versions are the same as the direct-use + * versions, with the pre- and post-amble chopped off. + * + * This code is encoded for size rather than absolute efficiency, with + * a view to being able to inline as much as possible. + * + * We only bother with direct forms (ie, vcpu in percpu data) of the + * operations here; the indirect forms are better handled in C, since + * they're generally too large to inline anyway. + */ + +#include <asm/asm-offsets.h> +#include <asm/percpu.h> +#include <asm/processor-flags.h> + +#include "xen-asm.h" + +/* + * Enable events. This clears the event mask and tests the pending + * event status with one and operation. If there are pending events, + * then enter the hypervisor to get them handled. + */ +ENTRY(xen_irq_enable_direct) + /* Unmask events */ + movb $0, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask + + /* + * Preempt here doesn't matter because that will deal with any + * pending interrupts. The pending check may end up being run + * on the wrong CPU, but that doesn't hurt. + */ + + /* Test for pending */ + testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_pending + jz 1f + +2: call check_events +1: +ENDPATCH(xen_irq_enable_direct) + ret + ENDPROC(xen_irq_enable_direct) + RELOC(xen_irq_enable_direct, 2b+1) + + +/* + * Disabling events is simply a matter of making the event mask + * non-zero. + */ +ENTRY(xen_irq_disable_direct) + movb $1, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask +ENDPATCH(xen_irq_disable_direct) + ret + ENDPROC(xen_irq_disable_direct) + RELOC(xen_irq_disable_direct, 0) + +/* + * (xen_)save_fl is used to get the current interrupt enable status. + * Callers expect the status to be in X86_EFLAGS_IF, and other bits + * may be set in the return value. We take advantage of this by + * making sure that X86_EFLAGS_IF has the right value (and other bits + * in that byte are 0), but other bits in the return value are + * undefined. We need to toggle the state of the bit, because Xen and + * x86 use opposite senses (mask vs enable). + */ +ENTRY(xen_save_fl_direct) + testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask + setz %ah + addb %ah, %ah +ENDPATCH(xen_save_fl_direct) + ret + ENDPROC(xen_save_fl_direct) + RELOC(xen_save_fl_direct, 0) + + +/* + * In principle the caller should be passing us a value return from + * xen_save_fl_direct, but for robustness sake we test only the + * X86_EFLAGS_IF flag rather than the whole byte. After setting the + * interrupt mask state, it checks for unmasked pending events and + * enters the hypervisor to get them delivered if so. + */ +ENTRY(xen_restore_fl_direct) +#ifdef CONFIG_X86_64 + testw $X86_EFLAGS_IF, %di +#else + testb $X86_EFLAGS_IF>>8, %ah +#endif + setz PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask + /* + * Preempt here doesn't matter because that will deal with any + * pending interrupts. The pending check may end up being run + * on the wrong CPU, but that doesn't hurt. + */ + + /* check for unmasked and pending */ + cmpw $0x0001, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_pending + jnz 1f +2: call check_events +1: +ENDPATCH(xen_restore_fl_direct) + ret + ENDPROC(xen_restore_fl_direct) + RELOC(xen_restore_fl_direct, 2b+1) + + +/* + * Force an event check by making a hypercall, but preserve regs + * before making the call. + */ +check_events: +#ifdef CONFIG_X86_32 + push %eax + push %ecx + push %edx + call xen_force_evtchn_callback + pop %edx + pop %ecx + pop %eax +#else + push %rax + push %rcx + push %rdx + push %rsi + push %rdi + push %r8 + push %r9 + push %r10 + push %r11 + call xen_force_evtchn_callback + pop %r11 + pop %r10 + pop %r9 + pop %r8 + pop %rdi + pop %rsi + pop %rdx + pop %rcx + pop %rax +#endif + ret diff --git a/kernel/arch/x86/xen/xen-asm.h b/kernel/arch/x86/xen/xen-asm.h new file mode 100644 index 000000000..465276467 --- /dev/null +++ b/kernel/arch/x86/xen/xen-asm.h @@ -0,0 +1,12 @@ +#ifndef _XEN_XEN_ASM_H +#define _XEN_XEN_ASM_H + +#include <linux/linkage.h> + +#define RELOC(x, v) .globl x##_reloc; x##_reloc=v +#define ENDPATCH(x) .globl x##_end; x##_end=. + +/* Pseudo-flag used for virtual NMI, which we don't implement yet */ +#define XEN_EFLAGS_NMI 0x80000000 + +#endif diff --git a/kernel/arch/x86/xen/xen-asm_32.S b/kernel/arch/x86/xen/xen-asm_32.S new file mode 100644 index 000000000..fd92a64d7 --- /dev/null +++ b/kernel/arch/x86/xen/xen-asm_32.S @@ -0,0 +1,237 @@ +/* + * Asm versions of Xen pv-ops, suitable for either direct use or + * inlining. The inline versions are the same as the direct-use + * versions, with the pre- and post-amble chopped off. + * + * This code is encoded for size rather than absolute efficiency, with + * a view to being able to inline as much as possible. + * + * We only bother with direct forms (ie, vcpu in pda) of the + * operations here; the indirect forms are better handled in C, since + * they're generally too large to inline anyway. + */ + +#include <asm/thread_info.h> +#include <asm/processor-flags.h> +#include <asm/segment.h> +#include <asm/asm.h> + +#include <xen/interface/xen.h> + +#include "xen-asm.h" + +/* + * Force an event check by making a hypercall, but preserve regs + * before making the call. + */ +check_events: + push %eax + push %ecx + push %edx + call xen_force_evtchn_callback + pop %edx + pop %ecx + pop %eax + ret + +/* + * We can't use sysexit directly, because we're not running in ring0. + * But we can easily fake it up using iret. Assuming xen_sysexit is + * jumped to with a standard stack frame, we can just strip it back to + * a standard iret frame and use iret. + */ +ENTRY(xen_sysexit) + movl PT_EAX(%esp), %eax /* Shouldn't be necessary? */ + orl $X86_EFLAGS_IF, PT_EFLAGS(%esp) + lea PT_EIP(%esp), %esp + + jmp xen_iret +ENDPROC(xen_sysexit) + +/* + * This is run where a normal iret would be run, with the same stack setup: + * 8: eflags + * 4: cs + * esp-> 0: eip + * + * This attempts to make sure that any pending events are dealt with + * on return to usermode, but there is a small window in which an + * event can happen just before entering usermode. If the nested + * interrupt ends up setting one of the TIF_WORK_MASK pending work + * flags, they will not be tested again before returning to + * usermode. This means that a process can end up with pending work, + * which will be unprocessed until the process enters and leaves the + * kernel again, which could be an unbounded amount of time. This + * means that a pending signal or reschedule event could be + * indefinitely delayed. + * + * The fix is to notice a nested interrupt in the critical window, and + * if one occurs, then fold the nested interrupt into the current + * interrupt stack frame, and re-process it iteratively rather than + * recursively. This means that it will exit via the normal path, and + * all pending work will be dealt with appropriately. + * + * Because the nested interrupt handler needs to deal with the current + * stack state in whatever form its in, we keep things simple by only + * using a single register which is pushed/popped on the stack. + */ + +.macro POP_FS +1: + popw %fs +.pushsection .fixup, "ax" +2: movw $0, (%esp) + jmp 1b +.popsection + _ASM_EXTABLE(1b,2b) +.endm + +ENTRY(xen_iret) + /* test eflags for special cases */ + testl $(X86_EFLAGS_VM | XEN_EFLAGS_NMI), 8(%esp) + jnz hyper_iret + + push %eax + ESP_OFFSET=4 # bytes pushed onto stack + + /* Store vcpu_info pointer for easy access */ +#ifdef CONFIG_SMP + pushw %fs + movl $(__KERNEL_PERCPU), %eax + movl %eax, %fs + movl %fs:xen_vcpu, %eax + POP_FS +#else + movl %ss:xen_vcpu, %eax +#endif + + /* check IF state we're restoring */ + testb $X86_EFLAGS_IF>>8, 8+1+ESP_OFFSET(%esp) + + /* + * Maybe enable events. Once this happens we could get a + * recursive event, so the critical region starts immediately + * afterwards. However, if that happens we don't end up + * resuming the code, so we don't have to be worried about + * being preempted to another CPU. + */ + setz %ss:XEN_vcpu_info_mask(%eax) +xen_iret_start_crit: + + /* check for unmasked and pending */ + cmpw $0x0001, %ss:XEN_vcpu_info_pending(%eax) + + /* + * If there's something pending, mask events again so we can + * jump back into xen_hypervisor_callback. Otherwise do not + * touch XEN_vcpu_info_mask. + */ + jne 1f + movb $1, %ss:XEN_vcpu_info_mask(%eax) + +1: popl %eax + + /* + * From this point on the registers are restored and the stack + * updated, so we don't need to worry about it if we're + * preempted + */ +iret_restore_end: + + /* + * Jump to hypervisor_callback after fixing up the stack. + * Events are masked, so jumping out of the critical region is + * OK. + */ + je xen_hypervisor_callback + +1: iret +xen_iret_end_crit: + _ASM_EXTABLE(1b, iret_exc) + +hyper_iret: + /* put this out of line since its very rarely used */ + jmp hypercall_page + __HYPERVISOR_iret * 32 + + .globl xen_iret_start_crit, xen_iret_end_crit + +/* + * This is called by xen_hypervisor_callback in entry.S when it sees + * that the EIP at the time of interrupt was between + * xen_iret_start_crit and xen_iret_end_crit. We're passed the EIP in + * %eax so we can do a more refined determination of what to do. + * + * The stack format at this point is: + * ---------------- + * ss : (ss/esp may be present if we came from usermode) + * esp : + * eflags } outer exception info + * cs } + * eip } + * ---------------- <- edi (copy dest) + * eax : outer eax if it hasn't been restored + * ---------------- + * eflags } nested exception info + * cs } (no ss/esp because we're nested + * eip } from the same ring) + * orig_eax }<- esi (copy src) + * - - - - - - - - + * fs } + * es } + * ds } SAVE_ALL state + * eax } + * : : + * ebx }<- esp + * ---------------- + * + * In order to deliver the nested exception properly, we need to shift + * everything from the return addr up to the error code so it sits + * just under the outer exception info. This means that when we + * handle the exception, we do it in the context of the outer + * exception rather than starting a new one. + * + * The only caveat is that if the outer eax hasn't been restored yet + * (ie, it's still on stack), we need to insert its value into the + * SAVE_ALL state before going on, since it's usermode state which we + * eventually need to restore. + */ +ENTRY(xen_iret_crit_fixup) + /* + * Paranoia: Make sure we're really coming from kernel space. + * One could imagine a case where userspace jumps into the + * critical range address, but just before the CPU delivers a + * GP, it decides to deliver an interrupt instead. Unlikely? + * Definitely. Easy to avoid? Yes. The Intel documents + * explicitly say that the reported EIP for a bad jump is the + * jump instruction itself, not the destination, but some + * virtual environments get this wrong. + */ + movl PT_CS(%esp), %ecx + andl $SEGMENT_RPL_MASK, %ecx + cmpl $USER_RPL, %ecx + je 2f + + lea PT_ORIG_EAX(%esp), %esi + lea PT_EFLAGS(%esp), %edi + + /* + * If eip is before iret_restore_end then stack + * hasn't been restored yet. + */ + cmp $iret_restore_end, %eax + jae 1f + + movl 0+4(%edi), %eax /* copy EAX (just above top of frame) */ + movl %eax, PT_EAX(%esp) + + lea ESP_OFFSET(%edi), %edi /* move dest up over saved regs */ + + /* set up the copy */ +1: std + mov $PT_EIP / 4, %ecx /* saved regs up to orig_eax */ + rep movsl + cld + + lea 4(%edi), %esp /* point esp to new frame */ +2: jmp xen_do_upcall + diff --git a/kernel/arch/x86/xen/xen-asm_64.S b/kernel/arch/x86/xen/xen-asm_64.S new file mode 100644 index 000000000..985fc3ee0 --- /dev/null +++ b/kernel/arch/x86/xen/xen-asm_64.S @@ -0,0 +1,159 @@ +/* + * Asm versions of Xen pv-ops, suitable for either direct use or + * inlining. The inline versions are the same as the direct-use + * versions, with the pre- and post-amble chopped off. + * + * This code is encoded for size rather than absolute efficiency, with + * a view to being able to inline as much as possible. + * + * We only bother with direct forms (ie, vcpu in pda) of the + * operations here; the indirect forms are better handled in C, since + * they're generally too large to inline anyway. + */ + +#include <asm/errno.h> +#include <asm/percpu.h> +#include <asm/processor-flags.h> +#include <asm/segment.h> + +#include <xen/interface/xen.h> + +#include "xen-asm.h" + +ENTRY(xen_adjust_exception_frame) + mov 8+0(%rsp), %rcx + mov 8+8(%rsp), %r11 + ret $16 + +hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32 +/* + * Xen64 iret frame: + * + * ss + * rsp + * rflags + * cs + * rip <-- standard iret frame + * + * flags + * + * rcx } + * r11 }<-- pushed by hypercall page + * rsp->rax } + */ +ENTRY(xen_iret) + pushq $0 +1: jmp hypercall_iret +ENDPATCH(xen_iret) +RELOC(xen_iret, 1b+1) + +/* + * sysexit is not used for 64-bit processes, so it's only ever used to + * return to 32-bit compat userspace. + */ +ENTRY(xen_sysexit) + pushq $__USER32_DS + pushq %rcx + pushq $X86_EFLAGS_IF + pushq $__USER32_CS + pushq %rdx + + pushq $0 +1: jmp hypercall_iret +ENDPATCH(xen_sysexit) +RELOC(xen_sysexit, 1b+1) + +ENTRY(xen_sysret64) + /* + * We're already on the usermode stack at this point, but + * still with the kernel gs, so we can easily switch back + */ + movq %rsp, PER_CPU_VAR(rsp_scratch) + movq PER_CPU_VAR(kernel_stack), %rsp + + pushq $__USER_DS + pushq PER_CPU_VAR(rsp_scratch) + pushq %r11 + pushq $__USER_CS + pushq %rcx + + pushq $VGCF_in_syscall +1: jmp hypercall_iret +ENDPATCH(xen_sysret64) +RELOC(xen_sysret64, 1b+1) + +ENTRY(xen_sysret32) + /* + * We're already on the usermode stack at this point, but + * still with the kernel gs, so we can easily switch back + */ + movq %rsp, PER_CPU_VAR(rsp_scratch) + movq PER_CPU_VAR(kernel_stack), %rsp + + pushq $__USER32_DS + pushq PER_CPU_VAR(rsp_scratch) + pushq %r11 + pushq $__USER32_CS + pushq %rcx + + pushq $0 +1: jmp hypercall_iret +ENDPATCH(xen_sysret32) +RELOC(xen_sysret32, 1b+1) + +/* + * Xen handles syscall callbacks much like ordinary exceptions, which + * means we have: + * - kernel gs + * - kernel rsp + * - an iret-like stack frame on the stack (including rcx and r11): + * ss + * rsp + * rflags + * cs + * rip + * r11 + * rsp->rcx + * + * In all the entrypoints, we undo all that to make it look like a + * CPU-generated syscall/sysenter and jump to the normal entrypoint. + */ + +.macro undo_xen_syscall + mov 0*8(%rsp), %rcx + mov 1*8(%rsp), %r11 + mov 5*8(%rsp), %rsp +.endm + +/* Normal 64-bit system call target */ +ENTRY(xen_syscall_target) + undo_xen_syscall + jmp system_call_after_swapgs +ENDPROC(xen_syscall_target) + +#ifdef CONFIG_IA32_EMULATION + +/* 32-bit compat syscall target */ +ENTRY(xen_syscall32_target) + undo_xen_syscall + jmp ia32_cstar_target +ENDPROC(xen_syscall32_target) + +/* 32-bit compat sysenter target */ +ENTRY(xen_sysenter_target) + undo_xen_syscall + jmp ia32_sysenter_target +ENDPROC(xen_sysenter_target) + +#else /* !CONFIG_IA32_EMULATION */ + +ENTRY(xen_syscall32_target) +ENTRY(xen_sysenter_target) + lea 16(%rsp), %rsp /* strip %rcx, %r11 */ + mov $-ENOSYS, %rax + pushq $0 + jmp hypercall_iret +ENDPROC(xen_syscall32_target) +ENDPROC(xen_sysenter_target) + +#endif /* CONFIG_IA32_EMULATION */ diff --git a/kernel/arch/x86/xen/xen-head.S b/kernel/arch/x86/xen/xen-head.S new file mode 100644 index 000000000..8afdfccf6 --- /dev/null +++ b/kernel/arch/x86/xen/xen-head.S @@ -0,0 +1,123 @@ +/* Xen-specific pieces of head.S, intended to be included in the right + place in head.S */ + +#ifdef CONFIG_XEN + +#include <linux/elfnote.h> +#include <linux/init.h> + +#include <asm/boot.h> +#include <asm/asm.h> +#include <asm/page_types.h> + +#include <xen/interface/elfnote.h> +#include <xen/interface/features.h> +#include <xen/interface/xen.h> +#include <xen/interface/xen-mca.h> +#include <asm/xen/interface.h> + +#ifdef CONFIG_XEN_PVH +#define PVH_FEATURES_STR "|writable_descriptor_tables|auto_translated_physmap|supervisor_mode_kernel" +/* Note the lack of 'hvm_callback_vector'. Older hypervisor will + * balk at this being part of XEN_ELFNOTE_FEATURES, so we put it in + * XEN_ELFNOTE_SUPPORTED_FEATURES which older hypervisors will ignore. + */ +#define PVH_FEATURES ((1 << XENFEAT_writable_page_tables) | \ + (1 << XENFEAT_auto_translated_physmap) | \ + (1 << XENFEAT_supervisor_mode_kernel) | \ + (1 << XENFEAT_hvm_callback_vector)) +/* The XENFEAT_writable_page_tables is not stricly neccessary as we set that + * up regardless whether this CONFIG option is enabled or not, but it + * clarifies what the right flags need to be. + */ +#else +#define PVH_FEATURES_STR "" +#define PVH_FEATURES (0) +#endif + + __INIT +ENTRY(startup_xen) + cld +#ifdef CONFIG_X86_32 + mov %esi,xen_start_info + mov $init_thread_union+THREAD_SIZE,%esp +#else + mov %rsi,xen_start_info + mov $init_thread_union+THREAD_SIZE,%rsp +#endif + jmp xen_start_kernel + + __FINIT + +#ifdef CONFIG_XEN_PVH +/* + * xen_pvh_early_cpu_init() - early PVH VCPU initialization + * @cpu: this cpu number (%rdi) + * @entry: true if this is a secondary vcpu coming up on this entry + * point, false if this is the boot CPU being initialized for + * the first time (%rsi) + * + * Note: This is called as a function on the boot CPU, and is the entry point + * on the secondary CPU. + */ +ENTRY(xen_pvh_early_cpu_init) + mov %rsi, %r11 + + /* Gather features to see if NX implemented. */ + mov $0x80000001, %eax + cpuid + mov %edx, %esi + + mov $MSR_EFER, %ecx + rdmsr + bts $_EFER_SCE, %eax + + bt $20, %esi + jnc 1f /* No NX, skip setting it */ + bts $_EFER_NX, %eax +1: wrmsr +#ifdef CONFIG_SMP + cmp $0, %r11b + jne cpu_bringup_and_idle +#endif + ret + +#endif /* CONFIG_XEN_PVH */ + +.pushsection .text + .balign PAGE_SIZE +ENTRY(hypercall_page) + .skip PAGE_SIZE + +#define HYPERCALL(n) \ + .equ xen_hypercall_##n, hypercall_page + __HYPERVISOR_##n * 32; \ + .type xen_hypercall_##n, @function; .size xen_hypercall_##n, 32 +#include <asm/xen-hypercalls.h> +#undef HYPERCALL + +.popsection + + ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz "linux") + ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION, .asciz "2.6") + ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION, .asciz "xen-3.0") +#ifdef CONFIG_X86_32 + ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, _ASM_PTR __PAGE_OFFSET) +#else + ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE, _ASM_PTR __START_KERNEL_map) +#endif + ELFNOTE(Xen, XEN_ELFNOTE_ENTRY, _ASM_PTR startup_xen) + ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, _ASM_PTR hypercall_page) + ELFNOTE(Xen, XEN_ELFNOTE_FEATURES, .ascii "!writable_page_tables|pae_pgdir_above_4gb"; .asciz PVH_FEATURES_STR) + ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long (PVH_FEATURES) | + (1 << XENFEAT_writable_page_tables) | + (1 << XENFEAT_dom0)) + ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE, .asciz "yes") + ELFNOTE(Xen, XEN_ELFNOTE_LOADER, .asciz "generic") + ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID, + .quad _PAGE_PRESENT; .quad _PAGE_PRESENT) + ELFNOTE(Xen, XEN_ELFNOTE_SUSPEND_CANCEL, .long 1) + ELFNOTE(Xen, XEN_ELFNOTE_MOD_START_PFN, .long 1) + ELFNOTE(Xen, XEN_ELFNOTE_HV_START_LOW, _ASM_PTR __HYPERVISOR_VIRT_START) + ELFNOTE(Xen, XEN_ELFNOTE_PADDR_OFFSET, _ASM_PTR 0) + +#endif /*CONFIG_XEN */ diff --git a/kernel/arch/x86/xen/xen-ops.h b/kernel/arch/x86/xen/xen-ops.h new file mode 100644 index 000000000..9e195c683 --- /dev/null +++ b/kernel/arch/x86/xen/xen-ops.h @@ -0,0 +1,145 @@ +#ifndef XEN_OPS_H +#define XEN_OPS_H + +#include <linux/init.h> +#include <linux/clocksource.h> +#include <linux/irqreturn.h> +#include <xen/xen-ops.h> + +/* These are code, but not functions. Defined in entry.S */ +extern const char xen_hypervisor_callback[]; +extern const char xen_failsafe_callback[]; + +void xen_sysenter_target(void); +#ifdef CONFIG_X86_64 +void xen_syscall_target(void); +void xen_syscall32_target(void); +#endif + +extern void *xen_initial_gdt; + +struct trap_info; +void xen_copy_trap_info(struct trap_info *traps); + +DECLARE_PER_CPU(struct vcpu_info, xen_vcpu_info); +DECLARE_PER_CPU(unsigned long, xen_cr3); +DECLARE_PER_CPU(unsigned long, xen_current_cr3); + +extern struct start_info *xen_start_info; +extern struct shared_info xen_dummy_shared_info; +extern struct shared_info *HYPERVISOR_shared_info; + +void xen_setup_mfn_list_list(void); +void xen_setup_shared_info(void); +void xen_build_mfn_list_list(void); +void xen_setup_machphys_mapping(void); +void xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn); +void xen_reserve_top(void); + +void xen_mm_pin_all(void); +void xen_mm_unpin_all(void); + +unsigned long __ref xen_chk_extra_mem(unsigned long pfn); +void __init xen_inv_extra_mem(void); +void __init xen_remap_memory(void); +char * __init xen_memory_setup(void); +char * xen_auto_xlated_memory_setup(void); +void __init xen_arch_setup(void); +void xen_enable_sysenter(void); +void xen_enable_syscall(void); +void xen_vcpu_restore(void); + +void xen_callback_vector(void); +void xen_hvm_init_shared_info(void); +void xen_unplug_emulated_devices(void); + +void __init xen_build_dynamic_phys_to_machine(void); +void __init xen_vmalloc_p2m_tree(void); + +void xen_init_irq_ops(void); +void xen_setup_timer(int cpu); +void xen_setup_runstate_info(int cpu); +void xen_teardown_timer(int cpu); +cycle_t xen_clocksource_read(void); +void xen_setup_cpu_clockevents(void); +void __init xen_init_time_ops(void); +void __init xen_hvm_init_time_ops(void); + +irqreturn_t xen_debug_interrupt(int irq, void *dev_id); + +bool xen_vcpu_stolen(int vcpu); + +void xen_setup_vcpu_info_placement(void); + +#ifdef CONFIG_SMP +void xen_smp_init(void); +void __init xen_hvm_smp_init(void); + +extern cpumask_var_t xen_cpu_initialized_map; +#else +static inline void xen_smp_init(void) {} +static inline void xen_hvm_smp_init(void) {} +#endif + +#ifdef CONFIG_PARAVIRT_SPINLOCKS +void __init xen_init_spinlocks(void); +void xen_init_lock_cpu(int cpu); +void xen_uninit_lock_cpu(int cpu); +#else +static inline void xen_init_spinlocks(void) +{ +} +static inline void xen_init_lock_cpu(int cpu) +{ +} +static inline void xen_uninit_lock_cpu(int cpu) +{ +} +#endif + +struct dom0_vga_console_info; + +#ifdef CONFIG_XEN_DOM0 +void __init xen_init_vga(const struct dom0_vga_console_info *, size_t size); +void __init xen_init_apic(void); +#else +static inline void __init xen_init_vga(const struct dom0_vga_console_info *info, + size_t size) +{ +} +static inline void __init xen_init_apic(void) +{ +} +#endif + +#ifdef CONFIG_XEN_EFI +extern void xen_efi_init(void); +#else +static inline void __init xen_efi_init(void) +{ +} +#endif + +/* Declare an asm function, along with symbols needed to make it + inlineable */ +#define DECL_ASM(ret, name, ...) \ + __visible ret name(__VA_ARGS__); \ + extern char name##_end[] __visible; \ + extern char name##_reloc[] __visible + +DECL_ASM(void, xen_irq_enable_direct, void); +DECL_ASM(void, xen_irq_disable_direct, void); +DECL_ASM(unsigned long, xen_save_fl_direct, void); +DECL_ASM(void, xen_restore_fl_direct, unsigned long); + +/* These are not functions, and cannot be called normally */ +__visible void xen_iret(void); +__visible void xen_sysexit(void); +__visible void xen_sysret32(void); +__visible void xen_sysret64(void); +__visible void xen_adjust_exception_frame(void); + +extern int xen_panic_handler_init(void); + +void xen_pvh_secondary_vcpu_init(int cpu); +#endif /* XEN_OPS_H */ |