enlighten_pv.c

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 (xen_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 = {
	.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,

	.read_cr0 = xen_read_cr0,
	.write_cr0 = xen_write_cr0,

	.read_cr4 = native_read_cr4,
	.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,
	.write_msr = xen_write_msr,

	.read_msr_safe = xen_read_msr_safe,
	.write_msr_safe = xen_write_msr_safe,

	.read_pmc = xen_read_pmc,

	.iret = xen_iret,
#ifdef CONFIG_X86_64
	.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 void xen_restart(char *msg)
{
	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 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_platform_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_platform_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.
 */
static void xen_setup_gdt(int cpu)
{
	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;
}

static void __init xen_dom0_set_legacy_features(void)
{
	x86_platform.legacy.rtc = 1;
}

/* 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();

	xen_setup_machphys_mapping();

	/* Install Xen paravirt ops */
	pv_info = xen_info;
	pv_init_ops = xen_init_ops;
	pv_cpu_ops = xen_cpu_ops;

	x86_platform.get_nmi_reason = xen_get_nmi_reason;

	x86_init.resources.memory_setup = xen_memory_setup;
	x86_init.oem.arch_setup = xen_arch_setup;
	x86_init.oem.banner = xen_banner;

	/*
	 * 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_capabilities();

#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];

	WARN_ON(xen_cpuhp_setup(xen_cpu_up_prepare_pv, xen_cpu_dead_pv));

	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);
	xen_reserve_special_pages();

	/* 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();

	/*
	 * 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.x86_capability[CPUID_1_EDX] = 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);
	boot_params.hdr.hardware_subarch = X86_SUBARCH_XEN;

	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,
		};

		x86_platform.set_legacy_features =
				xen_dom0_set_legacy_features;
		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_platform_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");

	/* Let's presume PV guests always boot on vCPU with id 0. */
	per_cpu(xen_vcpu_id, 0) = 0;

	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
}

static int xen_cpu_up_prepare_pv(unsigned int cpu)
{
	int rc;

	xen_setup_timer(cpu);

	rc = xen_smp_intr_init(cpu);
	if (rc) {
		WARN(1, "xen_smp_intr_init() for CPU %d failed: %d\n",
		     cpu, rc);
		return rc;
	}

	rc = xen_smp_intr_init_pv(cpu);
	if (rc) {
		WARN(1, "xen_smp_intr_init_pv() for CPU %d failed: %d\n",
		     cpu, rc);
		return rc;
	}

	return 0;
}

static int xen_cpu_dead_pv(unsigned int cpu)
{
	xen_smp_intr_free(cpu);
	xen_smp_intr_free_pv(cpu);

	xen_teardown_timer(cpu);

	return 0;
}

static uint32_t __init xen_platform_pv(void)
{
	if (xen_pv_domain())
		return xen_cpuid_base();

	return 0;
}

const struct hypervisor_x86 x86_hyper_xen_pv = {
	.name                   = "Xen PV",
	.detect                 = xen_platform_pv,
	.pin_vcpu               = xen_pin_vcpu,
};
EXPORT_SYMBOL(x86_hyper_xen_pv);