^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) #ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) #include <linux/memblock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) #include <linux/cpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) #include <linux/kexec.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <xen/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <xen/features.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <xen/page.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <asm/xen/hypercall.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <asm/xen/hypervisor.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <asm/cpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <asm/e820/api.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include "xen-ops.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include "smp.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include "pmu.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) EXPORT_SYMBOL_GPL(hypercall_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * Pointer to the xen_vcpu_info structure or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) * &HYPERVISOR_shared_info->vcpu_info[cpu]. See xen_hvm_init_shared_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * and xen_vcpu_setup for details. By default it points to share_info->vcpu_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) * but if the hypervisor supports VCPUOP_register_vcpu_info then it can point
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * to xen_vcpu_info. The pointer is used in __xen_evtchn_do_upcall to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) * acknowledge pending events.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) * Also more subtly it is used by the patched version of irq enable/disable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) * e.g. xen_irq_enable_direct and xen_iret in PV mode.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * The desire to be able to do those mask/unmask operations as a single
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) * instruction by using the per-cpu offset held in %gs is the real reason
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * vcpu info is in a per-cpu pointer and the original reason for this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) * hypercall.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) * Per CPU pages used if hypervisor supports VCPUOP_register_vcpu_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * hypercall. This can be used both in PV and PVHVM mode. The structure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) * overrides the default per_cpu(xen_vcpu, cpu) value.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) /* Linux <-> Xen vCPU id mapping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) DEFINE_PER_CPU(uint32_t, xen_vcpu_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) EXPORT_PER_CPU_SYMBOL(xen_vcpu_id);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) EXPORT_SYMBOL(machine_to_phys_mapping);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) unsigned long machine_to_phys_nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) EXPORT_SYMBOL(machine_to_phys_nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) struct start_info *xen_start_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) EXPORT_SYMBOL_GPL(xen_start_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) struct shared_info xen_dummy_shared_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) __read_mostly int xen_have_vector_callback;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) EXPORT_SYMBOL_GPL(xen_have_vector_callback);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) * NB: These need to live in .data or alike because they're used by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) * xen_prepare_pvh() which runs before clearing the bss.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) enum xen_domain_type __ro_after_init xen_domain_type = XEN_NATIVE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) EXPORT_SYMBOL_GPL(xen_domain_type);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) uint32_t xen_start_flags __section(".data") = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) EXPORT_SYMBOL(xen_start_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) * Point at some empty memory to start with. We map the real shared_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) * page as soon as fixmap is up and running.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) * Flag to determine whether vcpu info placement is available on all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) * VCPUs. We assume it is to start with, and then set it to zero on
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) * the first failure. This is because it can succeed on some VCPUs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) * and not others, since it can involve hypervisor memory allocation,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) * or because the guest failed to guarantee all the appropriate
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) * constraints on all VCPUs (ie buffer can't cross a page boundary).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) * Note that any particular CPU may be using a placed vcpu structure,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) * but we can only optimise if the all are.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) * 0: not available, 1: available
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) int xen_have_vcpu_info_placement = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) static int xen_cpu_up_online(unsigned int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) xen_init_lock_cpu(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) int xen_cpuhp_setup(int (*cpu_up_prepare_cb)(unsigned int),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) int (*cpu_dead_cb)(unsigned int))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) int rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) rc = cpuhp_setup_state_nocalls(CPUHP_XEN_PREPARE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) "x86/xen/guest:prepare",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) cpu_up_prepare_cb, cpu_dead_cb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) if (rc >= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) rc = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) "x86/xen/guest:online",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) xen_cpu_up_online, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) if (rc < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) cpuhp_remove_state_nocalls(CPUHP_XEN_PREPARE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) return rc >= 0 ? 0 : rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) static int xen_vcpu_setup_restore(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) int rc = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) /* Any per_cpu(xen_vcpu) is stale, so reset it */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) xen_vcpu_info_reset(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) * For PVH and PVHVM, setup online VCPUs only. The rest will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) * be handled by hotplug.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) if (xen_pv_domain() ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) (xen_hvm_domain() && cpu_online(cpu))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) rc = xen_vcpu_setup(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) return rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) * On restore, set the vcpu placement up again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) * If it fails, then we're in a bad state, since
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) * we can't back out from using it...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) void xen_vcpu_restore(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) int cpu, rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) for_each_possible_cpu(cpu) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) bool other_cpu = (cpu != smp_processor_id());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) bool is_up;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) if (xen_vcpu_nr(cpu) == XEN_VCPU_ID_INVALID)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) /* Only Xen 4.5 and higher support this. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) xen_vcpu_nr(cpu), NULL) > 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) if (other_cpu && is_up &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) HYPERVISOR_vcpu_op(VCPUOP_down, xen_vcpu_nr(cpu), NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) if (xen_pv_domain() || xen_feature(XENFEAT_hvm_safe_pvclock))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) xen_setup_runstate_info(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) rc = xen_vcpu_setup_restore(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) if (rc)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) pr_emerg_once("vcpu restore failed for cpu=%d err=%d. "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) "System will hang.\n", cpu, rc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) * In case xen_vcpu_setup_restore() fails, do not bring up the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) * VCPU. This helps us avoid the resulting OOPS when the VCPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) * accesses pvclock_vcpu_time via xen_vcpu (which is NULL.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) * Note that this does not improve the situation much -- now the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) * VM hangs instead of OOPSing -- with the VCPUs that did not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) * fail, spinning in stop_machine(), waiting for the failed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) * VCPUs to come up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) if (other_cpu && is_up && (rc == 0) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) HYPERVISOR_vcpu_op(VCPUOP_up, xen_vcpu_nr(cpu), NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) void xen_vcpu_info_reset(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) if (xen_vcpu_nr(cpu) < MAX_VIRT_CPUS) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) per_cpu(xen_vcpu, cpu) =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) &HYPERVISOR_shared_info->vcpu_info[xen_vcpu_nr(cpu)];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) /* Set to NULL so that if somebody accesses it we get an OOPS */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) per_cpu(xen_vcpu, cpu) = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) int xen_vcpu_setup(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) struct vcpu_register_vcpu_info info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) struct vcpu_info *vcpup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) * This path is called on PVHVM at bootup (xen_hvm_smp_prepare_boot_cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) * and at restore (xen_vcpu_restore). Also called for hotplugged
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) * VCPUs (cpu_init -> xen_hvm_cpu_prepare_hvm).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) * However, the hypercall can only be done once (see below) so if a VCPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) * is offlined and comes back online then let's not redo the hypercall.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) * For PV it is called during restore (xen_vcpu_restore) and bootup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) * (xen_setup_vcpu_info_placement). The hotplug mechanism does not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) * use this function.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) if (xen_hvm_domain()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) if (per_cpu(xen_vcpu, cpu) == &per_cpu(xen_vcpu_info, cpu))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) if (xen_have_vcpu_info_placement) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) vcpup = &per_cpu(xen_vcpu_info, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) info.mfn = arbitrary_virt_to_mfn(vcpup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) info.offset = offset_in_page(vcpup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) * Check to see if the hypervisor will put the vcpu_info
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) * structure where we want it, which allows direct access via
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) * a percpu-variable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) * N.B. This hypercall can _only_ be called once per CPU.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) * Subsequent calls will error out with -EINVAL. This is due to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * the fact that hypervisor has no unregister variant and this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) * hypercall does not allow to over-write info.mfn and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) * info.offset.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) xen_vcpu_nr(cpu), &info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) if (err) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) pr_warn_once("register_vcpu_info failed: cpu=%d err=%d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) cpu, err);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) xen_have_vcpu_info_placement = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) * This cpu is using the registered vcpu info, even if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) * later ones fail to.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) per_cpu(xen_vcpu, cpu) = vcpup;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) if (!xen_have_vcpu_info_placement)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) xen_vcpu_info_reset(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) return ((per_cpu(xen_vcpu, cpu) == NULL) ? -ENODEV : 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) void xen_reboot(int reason)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) struct sched_shutdown r = { .reason = reason };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) int cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) for_each_online_cpu(cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) xen_pmu_finish(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) if (HYPERVISOR_sched_op(SCHEDOP_shutdown, &r))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) static int reboot_reason = SHUTDOWN_reboot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) static bool xen_legacy_crash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) void xen_emergency_restart(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) xen_reboot(reboot_reason);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) static int
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if (!kexec_crash_loaded()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) if (xen_legacy_crash)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) xen_reboot(SHUTDOWN_crash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) reboot_reason = SHUTDOWN_crash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) * If panic_timeout==0 then we are supposed to wait forever.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) * However, to preserve original dom0 behavior we have to drop
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) * into hypervisor. (domU behavior is controlled by its
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) * config file)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) if (panic_timeout == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) panic_timeout = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) return NOTIFY_DONE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) static int __init parse_xen_legacy_crash(char *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) xen_legacy_crash = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) early_param("xen_legacy_crash", parse_xen_legacy_crash);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) static struct notifier_block xen_panic_block = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) .notifier_call = xen_panic_event,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) .priority = INT_MIN
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) int xen_panic_handler_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) void xen_pin_vcpu(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) static bool disable_pinning;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) struct sched_pin_override pin_override;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) if (disable_pinning)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) pin_override.pcpu = cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) ret = HYPERVISOR_sched_op(SCHEDOP_pin_override, &pin_override);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) /* Ignore errors when removing override. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) if (cpu < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) switch (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) case -ENOSYS:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) pr_warn("Unable to pin on physical cpu %d. In case of problems consider vcpu pinning.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) disable_pinning = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) case -EPERM:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) WARN(1, "Trying to pin vcpu without having privilege to do so\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) disable_pinning = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) case -EINVAL:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) case -EBUSY:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) pr_warn("Physical cpu %d not available for pinning. Check Xen cpu configuration.\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) case 0:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) WARN(1, "rc %d while trying to pin vcpu\n", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) disable_pinning = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) #ifdef CONFIG_HOTPLUG_CPU
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) void xen_arch_register_cpu(int num)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) arch_register_cpu(num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) EXPORT_SYMBOL(xen_arch_register_cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) void xen_arch_unregister_cpu(int num)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) arch_unregister_cpu(num);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) EXPORT_SYMBOL(xen_arch_unregister_cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) #endif
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags