2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-23 12:43:55 +08:00
linux-next/arch/s390/kvm/kvm-s390.c

3340 lines
86 KiB
C
Raw Normal View History

/*
* hosting zSeries kernel virtual machines
*
* Copyright IBM Corp. 2008, 2009
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License (version 2 only)
* as published by the Free Software Foundation.
*
* Author(s): Carsten Otte <cotte@de.ibm.com>
* Christian Borntraeger <borntraeger@de.ibm.com>
* Heiko Carstens <heiko.carstens@de.ibm.com>
* Christian Ehrhardt <ehrhardt@de.ibm.com>
* Jason J. Herne <jjherne@us.ibm.com>
*/
#include <linux/compiler.h>
#include <linux/err.h>
#include <linux/fs.h>
#include <linux/hrtimer.h>
#include <linux/init.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <linux/mman.h>
#include <linux/module.h>
#include <linux/random.h>
#include <linux/slab.h>
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
#include <linux/timer.h>
#include <linux/vmalloc.h>
#include <linux/bitmap.h>
#include <asm/asm-offsets.h>
#include <asm/lowcore.h>
#include <asm/stp.h>
#include <asm/pgtable.h>
#include <asm/gmap.h>
#include <asm/nmi.h>
#include <asm/switch_to.h>
#include <asm/isc.h>
#include <asm/sclp.h>
#include <asm/cpacf.h>
- ARM: GICv3 ITS emulation and various fixes. Removal of the old VGIC implementation. - s390: support for trapping software breakpoints, nested virtualization (vSIE), the STHYI opcode, initial extensions for CPU model support. - MIPS: support for MIPS64 hosts (32-bit guests only) and lots of cleanups, preliminary to this and the upcoming support for hardware virtualization extensions. - x86: support for execute-only mappings in nested EPT; reduced vmexit latency for TSC deadline timer (by about 30%) on Intel hosts; support for more than 255 vCPUs. - PPC: bugfixes. The ugly bit is the conflicts. A couple of them are simple conflicts due to 4.7 fixes, but most of them are with other trees. There was definitely too much reliance on Acked-by here. Some conflicts are for KVM patches where _I_ gave my Acked-by, but the worst are for this pull request's patches that touch files outside arch/*/kvm. KVM submaintainers should probably learn to synchronize better with arch maintainers, with the latter providing topic branches whenever possible instead of Acked-by. This is what we do with arch/x86. And I should learn to refuse pull requests when linux-next sends scary signals, even if that means that submaintainers have to rebase their branches. Anyhow, here's the list: - arch/x86/kvm/vmx.c: handle_pcommit and EXIT_REASON_PCOMMIT was removed by the nvdimm tree. This tree adds handle_preemption_timer and EXIT_REASON_PREEMPTION_TIMER at the same place. In general all mentions of pcommit have to go. There is also a conflict between a stable fix and this patch, where the stable fix removed the vmx_create_pml_buffer function and its call. - virt/kvm/kvm_main.c: kvm_cpu_notifier was removed by the hotplug tree. This tree adds kvm_io_bus_get_dev at the same place. - virt/kvm/arm/vgic.c: a few final bugfixes went into 4.7 before the file was completely removed for 4.8. - include/linux/irqchip/arm-gic-v3.h: this one is entirely our fault; this is a change that should have gone in through the irqchip tree and pulled by kvm-arm. I think I would have rejected this kvm-arm pull request. The KVM version is the right one, except that it lacks GITS_BASER_PAGES_SHIFT. - arch/powerpc: what a mess. For the idle_book3s.S conflict, the KVM tree is the right one; everything else is trivial. In this case I am not quite sure what went wrong. The commit that is causing the mess (fd7bacbca47a, "KVM: PPC: Book3S HV: Fix TB corruption in guest exit path on HMI interrupt", 2016-05-15) touches both arch/powerpc/kernel/ and arch/powerpc/kvm/. It's large, but at 396 insertions/5 deletions I guessed that it wasn't really possible to split it and that the 5 deletions wouldn't conflict. That wasn't the case. - arch/s390: also messy. First is hypfs_diag.c where the KVM tree moved some code and the s390 tree patched it. You have to reapply the relevant part of commits 6c22c9863760, plus all of e030c1125eab, to arch/s390/kernel/diag.c. Or pick the linux-next conflict resolution from http://marc.info/?l=kvm&m=146717549531603&w=2. Second, there is a conflict in gmap.c between a stable fix and 4.8. The KVM version here is the correct one. I have pushed my resolution at refs/heads/merge-20160802 (commit 3d1f53419842) at git://git.kernel.org/pub/scm/virt/kvm/kvm.git. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAABAgAGBQJXoGm7AAoJEL/70l94x66DugQIAIj703ePAFepB/fCrKHkZZia SGrsBdvAtNsOhr7FQ5qvvjLxiv/cv7CymeuJivX8H+4kuUHUllDzey+RPHYHD9X7 U6n1PdCH9F15a3IXc8tDjlDdOMNIKJixYuq1UyNZMU6NFwl00+TZf9JF8A2US65b x/41W98ilL6nNBAsoDVmCLtPNWAqQ3lajaZELGfcqRQ9ZGKcAYOaLFXHv2YHf2XC qIDMf+slBGSQ66UoATnYV2gAopNlWbZ7n0vO6tE2KyvhHZ1m399aBX1+k8la/0JI 69r+Tz7ZHUSFtmlmyByi5IAB87myy2WQHyAPwj+4vwJkDGPcl0TrupzbG7+T05Y= =42ti -----END PGP SIGNATURE----- Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm Pull KVM updates from Paolo Bonzini: - ARM: GICv3 ITS emulation and various fixes. Removal of the old VGIC implementation. - s390: support for trapping software breakpoints, nested virtualization (vSIE), the STHYI opcode, initial extensions for CPU model support. - MIPS: support for MIPS64 hosts (32-bit guests only) and lots of cleanups, preliminary to this and the upcoming support for hardware virtualization extensions. - x86: support for execute-only mappings in nested EPT; reduced vmexit latency for TSC deadline timer (by about 30%) on Intel hosts; support for more than 255 vCPUs. - PPC: bugfixes. * tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (302 commits) KVM: PPC: Introduce KVM_CAP_PPC_HTM MIPS: Select HAVE_KVM for MIPS64_R{2,6} MIPS: KVM: Reset CP0_PageMask during host TLB flush MIPS: KVM: Fix ptr->int cast via KVM_GUEST_KSEGX() MIPS: KVM: Sign extend MFC0/RDHWR results MIPS: KVM: Fix 64-bit big endian dynamic translation MIPS: KVM: Fail if ebase doesn't fit in CP0_EBase MIPS: KVM: Use 64-bit CP0_EBase when appropriate MIPS: KVM: Set CP0_Status.KX on MIPS64 MIPS: KVM: Make entry code MIPS64 friendly MIPS: KVM: Use kmap instead of CKSEG0ADDR() MIPS: KVM: Use virt_to_phys() to get commpage PFN MIPS: Fix definition of KSEGX() for 64-bit KVM: VMX: Add VMCS to CPU's loaded VMCSs before VMPTRLD kvm: x86: nVMX: maintain internal copy of current VMCS KVM: PPC: Book3S HV: Save/restore TM state in H_CEDE KVM: PPC: Book3S HV: Pull out TM state save/restore into separate procedures KVM: arm64: vgic-its: Simplify MAPI error handling KVM: arm64: vgic-its: Make vgic_its_cmd_handle_mapi similar to other handlers KVM: arm64: vgic-its: Turn device_id validation into generic ID validation ...
2016-08-03 04:11:27 +08:00
#include <asm/timex.h>
#include "kvm-s390.h"
#include "gaccess.h"
#define KMSG_COMPONENT "kvm-s390"
#undef pr_fmt
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
#define CREATE_TRACE_POINTS
#include "trace.h"
#include "trace-s390.h"
#define MEM_OP_MAX_SIZE 65536 /* Maximum transfer size for KVM_S390_MEM_OP */
#define LOCAL_IRQS 32
#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \
(KVM_MAX_VCPUS + LOCAL_IRQS))
#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
struct kvm_stats_debugfs_item debugfs_entries[] = {
{ "userspace_handled", VCPU_STAT(exit_userspace) },
{ "exit_null", VCPU_STAT(exit_null) },
{ "exit_validity", VCPU_STAT(exit_validity) },
{ "exit_stop_request", VCPU_STAT(exit_stop_request) },
{ "exit_external_request", VCPU_STAT(exit_external_request) },
{ "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) },
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
{ "exit_instruction", VCPU_STAT(exit_instruction) },
{ "exit_pei", VCPU_STAT(exit_pei) },
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
kvm: add halt_poll_ns module parameter This patch introduces a new module parameter for the KVM module; when it is present, KVM attempts a bit of polling on every HLT before scheduling itself out via kvm_vcpu_block. This parameter helps a lot for latency-bound workloads---in particular I tested it with O_DSYNC writes with a battery-backed disk in the host. In this case, writes are fast (because the data doesn't have to go all the way to the platters) but they cannot be merged by either the host or the guest. KVM's performance here is usually around 30% of bare metal, or 50% if you use cache=directsync or cache=writethrough (these parameters avoid that the guest sends pointless flush requests, and at the same time they are not slow because of the battery-backed cache). The bad performance happens because on every halt the host CPU decides to halt itself too. When the interrupt comes, the vCPU thread is then migrated to a new physical CPU, and in general the latency is horrible because the vCPU thread has to be scheduled back in. With this patch performance reaches 60-65% of bare metal and, more important, 99% of what you get if you use idle=poll in the guest. This means that the tunable gets rid of this particular bottleneck, and more work can be done to improve performance in the kernel or QEMU. Of course there is some price to pay; every time an otherwise idle vCPUs is interrupted by an interrupt, it will poll unnecessarily and thus impose a little load on the host. The above results were obtained with a mostly random value of the parameter (500000), and the load was around 1.5-2.5% CPU usage on one of the host's core for each idle guest vCPU. The patch also adds a new stat, /sys/kernel/debug/kvm/halt_successful_poll, that can be used to tune the parameter. It counts how many HLT instructions received an interrupt during the polling period; each successful poll avoids that Linux schedules the VCPU thread out and back in, and may also avoid a likely trip to C1 and back for the physical CPU. While the VM is idle, a Linux 4 VCPU VM halts around 10 times per second. Of these halts, almost all are failed polls. During the benchmark, instead, basically all halts end within the polling period, except a more or less constant stream of 50 per second coming from vCPUs that are not running the benchmark. The wasted time is thus very low. Things may be slightly different for Windows VMs, which have a ~10 ms timer tick. The effect is also visible on Marcelo's recently-introduced latency test for the TSC deadline timer. Though of course a non-RT kernel has awful latency bounds, the latency of the timer is around 8000-10000 clock cycles compared to 20000-120000 without setting halt_poll_ns. For the TSC deadline timer, thus, the effect is both a smaller average latency and a smaller variance. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-05 01:20:58 +08:00
{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
KVM: halt_polling: provide a way to qualify wakeups during poll Some wakeups should not be considered a sucessful poll. For example on s390 I/O interrupts are usually floating, which means that _ALL_ CPUs would be considered runnable - letting all vCPUs poll all the time for transactional like workload, even if one vCPU would be enough. This can result in huge CPU usage for large guests. This patch lets architectures provide a way to qualify wakeups if they should be considered a good/bad wakeups in regard to polls. For s390 the implementation will fence of halt polling for anything but known good, single vCPU events. The s390 implementation for floating interrupts does a wakeup for one vCPU, but the interrupt will be delivered by whatever CPU checks first for a pending interrupt. We prefer the woken up CPU by marking the poll of this CPU as "good" poll. This code will also mark several other wakeup reasons like IPI or expired timers as "good". This will of course also mark some events as not sucessful. As KVM on z runs always as a 2nd level hypervisor, we prefer to not poll, unless we are really sure, though. This patch successfully limits the CPU usage for cases like uperf 1byte transactional ping pong workload or wakeup heavy workload like OLTP while still providing a proper speedup. This also introduced a new vcpu stat "halt_poll_no_tuning" that marks wakeups that are considered not good for polling. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Radim Krčmář <rkrcmar@redhat.com> (for an earlier version) Cc: David Matlack <dmatlack@google.com> Cc: Wanpeng Li <kernellwp@gmail.com> [Rename config symbol. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2016-05-13 18:16:35 +08:00
{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
{ "halt_wakeup", VCPU_STAT(halt_wakeup) },
{ "instruction_lctlg", VCPU_STAT(instruction_lctlg) },
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
{ "instruction_lctl", VCPU_STAT(instruction_lctl) },
{ "instruction_stctl", VCPU_STAT(instruction_stctl) },
{ "instruction_stctg", VCPU_STAT(instruction_stctg) },
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
{ "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) },
{ "deliver_external_call", VCPU_STAT(deliver_external_call) },
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
{ "deliver_service_signal", VCPU_STAT(deliver_service_signal) },
{ "deliver_virtio_interrupt", VCPU_STAT(deliver_virtio_interrupt) },
{ "deliver_stop_signal", VCPU_STAT(deliver_stop_signal) },
{ "deliver_prefix_signal", VCPU_STAT(deliver_prefix_signal) },
{ "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) },
{ "deliver_program_interruption", VCPU_STAT(deliver_program_int) },
{ "exit_wait_state", VCPU_STAT(exit_wait_state) },
{ "instruction_pfmf", VCPU_STAT(instruction_pfmf) },
{ "instruction_stidp", VCPU_STAT(instruction_stidp) },
{ "instruction_spx", VCPU_STAT(instruction_spx) },
{ "instruction_stpx", VCPU_STAT(instruction_stpx) },
{ "instruction_stap", VCPU_STAT(instruction_stap) },
{ "instruction_storage_key", VCPU_STAT(instruction_storage_key) },
{ "instruction_ipte_interlock", VCPU_STAT(instruction_ipte_interlock) },
{ "instruction_stsch", VCPU_STAT(instruction_stsch) },
{ "instruction_chsc", VCPU_STAT(instruction_chsc) },
{ "instruction_essa", VCPU_STAT(instruction_essa) },
{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
{ "instruction_sie", VCPU_STAT(instruction_sie) },
{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
{ "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) },
{ "instruction_sigp_cond_emergency", VCPU_STAT(instruction_sigp_cond_emergency) },
{ "instruction_sigp_start", VCPU_STAT(instruction_sigp_start) },
{ "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) },
{ "instruction_sigp_stop_store_status", VCPU_STAT(instruction_sigp_stop_store_status) },
{ "instruction_sigp_store_status", VCPU_STAT(instruction_sigp_store_status) },
{ "instruction_sigp_store_adtl_status", VCPU_STAT(instruction_sigp_store_adtl_status) },
{ "instruction_sigp_set_arch", VCPU_STAT(instruction_sigp_arch) },
{ "instruction_sigp_set_prefix", VCPU_STAT(instruction_sigp_prefix) },
{ "instruction_sigp_restart", VCPU_STAT(instruction_sigp_restart) },
{ "instruction_sigp_cpu_reset", VCPU_STAT(instruction_sigp_cpu_reset) },
{ "instruction_sigp_init_cpu_reset", VCPU_STAT(instruction_sigp_init_cpu_reset) },
{ "instruction_sigp_unknown", VCPU_STAT(instruction_sigp_unknown) },
{ "diagnose_10", VCPU_STAT(diagnose_10) },
{ "diagnose_44", VCPU_STAT(diagnose_44) },
{ "diagnose_9c", VCPU_STAT(diagnose_9c) },
{ "diagnose_258", VCPU_STAT(diagnose_258) },
{ "diagnose_308", VCPU_STAT(diagnose_308) },
{ "diagnose_500", VCPU_STAT(diagnose_500) },
{ NULL }
};
/* allow nested virtualization in KVM (if enabled by user space) */
static int nested;
module_param(nested, int, S_IRUGO);
MODULE_PARM_DESC(nested, "Nested virtualization support");
/* upper facilities limit for kvm */
unsigned long kvm_s390_fac_list_mask[16] = { FACILITIES_KVM };
unsigned long kvm_s390_fac_list_mask_size(void)
{
BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64);
return ARRAY_SIZE(kvm_s390_fac_list_mask);
}
/* available cpu features supported by kvm */
static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
/* available subfunctions indicated via query / "test bit" */
static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
static struct gmap_notifier gmap_notifier;
static struct gmap_notifier vsie_gmap_notifier;
debug_info_t *kvm_s390_dbf;
/* Section: not file related */
int kvm_arch_hardware_enable(void)
{
/* every s390 is virtualization enabled ;-) */
return 0;
}
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
unsigned long end);
/*
* This callback is executed during stop_machine(). All CPUs are therefore
* temporarily stopped. In order not to change guest behavior, we have to
* disable preemption whenever we touch the epoch of kvm and the VCPUs,
* so a CPU won't be stopped while calculating with the epoch.
*/
static int kvm_clock_sync(struct notifier_block *notifier, unsigned long val,
void *v)
{
struct kvm *kvm;
struct kvm_vcpu *vcpu;
int i;
unsigned long long *delta = v;
list_for_each_entry(kvm, &vm_list, vm_list) {
kvm->arch.epoch -= *delta;
kvm_for_each_vcpu(i, vcpu, kvm) {
vcpu->arch.sie_block->epoch -= *delta;
if (vcpu->arch.cputm_enabled)
vcpu->arch.cputm_start += *delta;
if (vcpu->arch.vsie_block)
vcpu->arch.vsie_block->epoch -= *delta;
}
}
return NOTIFY_OK;
}
static struct notifier_block kvm_clock_notifier = {
.notifier_call = kvm_clock_sync,
};
int kvm_arch_hardware_setup(void)
{
gmap_notifier.notifier_call = kvm_gmap_notifier;
gmap_register_pte_notifier(&gmap_notifier);
vsie_gmap_notifier.notifier_call = kvm_s390_vsie_gmap_notifier;
gmap_register_pte_notifier(&vsie_gmap_notifier);
atomic_notifier_chain_register(&s390_epoch_delta_notifier,
&kvm_clock_notifier);
return 0;
}
void kvm_arch_hardware_unsetup(void)
{
gmap_unregister_pte_notifier(&gmap_notifier);
gmap_unregister_pte_notifier(&vsie_gmap_notifier);
atomic_notifier_chain_unregister(&s390_epoch_delta_notifier,
&kvm_clock_notifier);
}
static void allow_cpu_feat(unsigned long nr)
{
set_bit_inv(nr, kvm_s390_available_cpu_feat);
}
static inline int plo_test_bit(unsigned char nr)
{
register unsigned long r0 asm("0") = (unsigned long) nr | 0x100;
int cc = 3; /* subfunction not available */
asm volatile(
/* Parameter registers are ignored for "test bit" */
" plo 0,0,0,0(0)\n"
" ipm %0\n"
" srl %0,28\n"
: "=d" (cc)
: "d" (r0)
: "cc");
return cc == 0;
}
static void kvm_s390_cpu_feat_init(void)
{
int i;
for (i = 0; i < 256; ++i) {
if (plo_test_bit(i))
kvm_s390_available_subfunc.plo[i >> 3] |= 0x80 >> (i & 7);
}
if (test_facility(28)) /* TOD-clock steering */
- ARM: GICv3 ITS emulation and various fixes. Removal of the old VGIC implementation. - s390: support for trapping software breakpoints, nested virtualization (vSIE), the STHYI opcode, initial extensions for CPU model support. - MIPS: support for MIPS64 hosts (32-bit guests only) and lots of cleanups, preliminary to this and the upcoming support for hardware virtualization extensions. - x86: support for execute-only mappings in nested EPT; reduced vmexit latency for TSC deadline timer (by about 30%) on Intel hosts; support for more than 255 vCPUs. - PPC: bugfixes. The ugly bit is the conflicts. A couple of them are simple conflicts due to 4.7 fixes, but most of them are with other trees. There was definitely too much reliance on Acked-by here. Some conflicts are for KVM patches where _I_ gave my Acked-by, but the worst are for this pull request's patches that touch files outside arch/*/kvm. KVM submaintainers should probably learn to synchronize better with arch maintainers, with the latter providing topic branches whenever possible instead of Acked-by. This is what we do with arch/x86. And I should learn to refuse pull requests when linux-next sends scary signals, even if that means that submaintainers have to rebase their branches. Anyhow, here's the list: - arch/x86/kvm/vmx.c: handle_pcommit and EXIT_REASON_PCOMMIT was removed by the nvdimm tree. This tree adds handle_preemption_timer and EXIT_REASON_PREEMPTION_TIMER at the same place. In general all mentions of pcommit have to go. There is also a conflict between a stable fix and this patch, where the stable fix removed the vmx_create_pml_buffer function and its call. - virt/kvm/kvm_main.c: kvm_cpu_notifier was removed by the hotplug tree. This tree adds kvm_io_bus_get_dev at the same place. - virt/kvm/arm/vgic.c: a few final bugfixes went into 4.7 before the file was completely removed for 4.8. - include/linux/irqchip/arm-gic-v3.h: this one is entirely our fault; this is a change that should have gone in through the irqchip tree and pulled by kvm-arm. I think I would have rejected this kvm-arm pull request. The KVM version is the right one, except that it lacks GITS_BASER_PAGES_SHIFT. - arch/powerpc: what a mess. For the idle_book3s.S conflict, the KVM tree is the right one; everything else is trivial. In this case I am not quite sure what went wrong. The commit that is causing the mess (fd7bacbca47a, "KVM: PPC: Book3S HV: Fix TB corruption in guest exit path on HMI interrupt", 2016-05-15) touches both arch/powerpc/kernel/ and arch/powerpc/kvm/. It's large, but at 396 insertions/5 deletions I guessed that it wasn't really possible to split it and that the 5 deletions wouldn't conflict. That wasn't the case. - arch/s390: also messy. First is hypfs_diag.c where the KVM tree moved some code and the s390 tree patched it. You have to reapply the relevant part of commits 6c22c9863760, plus all of e030c1125eab, to arch/s390/kernel/diag.c. Or pick the linux-next conflict resolution from http://marc.info/?l=kvm&m=146717549531603&w=2. Second, there is a conflict in gmap.c between a stable fix and 4.8. The KVM version here is the correct one. I have pushed my resolution at refs/heads/merge-20160802 (commit 3d1f53419842) at git://git.kernel.org/pub/scm/virt/kvm/kvm.git. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAABAgAGBQJXoGm7AAoJEL/70l94x66DugQIAIj703ePAFepB/fCrKHkZZia SGrsBdvAtNsOhr7FQ5qvvjLxiv/cv7CymeuJivX8H+4kuUHUllDzey+RPHYHD9X7 U6n1PdCH9F15a3IXc8tDjlDdOMNIKJixYuq1UyNZMU6NFwl00+TZf9JF8A2US65b x/41W98ilL6nNBAsoDVmCLtPNWAqQ3lajaZELGfcqRQ9ZGKcAYOaLFXHv2YHf2XC qIDMf+slBGSQ66UoATnYV2gAopNlWbZ7n0vO6tE2KyvhHZ1m399aBX1+k8la/0JI 69r+Tz7ZHUSFtmlmyByi5IAB87myy2WQHyAPwj+4vwJkDGPcl0TrupzbG7+T05Y= =42ti -----END PGP SIGNATURE----- Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm Pull KVM updates from Paolo Bonzini: - ARM: GICv3 ITS emulation and various fixes. Removal of the old VGIC implementation. - s390: support for trapping software breakpoints, nested virtualization (vSIE), the STHYI opcode, initial extensions for CPU model support. - MIPS: support for MIPS64 hosts (32-bit guests only) and lots of cleanups, preliminary to this and the upcoming support for hardware virtualization extensions. - x86: support for execute-only mappings in nested EPT; reduced vmexit latency for TSC deadline timer (by about 30%) on Intel hosts; support for more than 255 vCPUs. - PPC: bugfixes. * tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (302 commits) KVM: PPC: Introduce KVM_CAP_PPC_HTM MIPS: Select HAVE_KVM for MIPS64_R{2,6} MIPS: KVM: Reset CP0_PageMask during host TLB flush MIPS: KVM: Fix ptr->int cast via KVM_GUEST_KSEGX() MIPS: KVM: Sign extend MFC0/RDHWR results MIPS: KVM: Fix 64-bit big endian dynamic translation MIPS: KVM: Fail if ebase doesn't fit in CP0_EBase MIPS: KVM: Use 64-bit CP0_EBase when appropriate MIPS: KVM: Set CP0_Status.KX on MIPS64 MIPS: KVM: Make entry code MIPS64 friendly MIPS: KVM: Use kmap instead of CKSEG0ADDR() MIPS: KVM: Use virt_to_phys() to get commpage PFN MIPS: Fix definition of KSEGX() for 64-bit KVM: VMX: Add VMCS to CPU's loaded VMCSs before VMPTRLD kvm: x86: nVMX: maintain internal copy of current VMCS KVM: PPC: Book3S HV: Save/restore TM state in H_CEDE KVM: PPC: Book3S HV: Pull out TM state save/restore into separate procedures KVM: arm64: vgic-its: Simplify MAPI error handling KVM: arm64: vgic-its: Make vgic_its_cmd_handle_mapi similar to other handlers KVM: arm64: vgic-its: Turn device_id validation into generic ID validation ...
2016-08-03 04:11:27 +08:00
ptff(kvm_s390_available_subfunc.ptff,
sizeof(kvm_s390_available_subfunc.ptff),
PTFF_QAF);
if (test_facility(17)) { /* MSA */
__cpacf_query(CPACF_KMAC, (cpacf_mask_t *)
kvm_s390_available_subfunc.kmac);
__cpacf_query(CPACF_KMC, (cpacf_mask_t *)
kvm_s390_available_subfunc.kmc);
__cpacf_query(CPACF_KM, (cpacf_mask_t *)
kvm_s390_available_subfunc.km);
__cpacf_query(CPACF_KIMD, (cpacf_mask_t *)
kvm_s390_available_subfunc.kimd);
__cpacf_query(CPACF_KLMD, (cpacf_mask_t *)
kvm_s390_available_subfunc.klmd);
}
if (test_facility(76)) /* MSA3 */
__cpacf_query(CPACF_PCKMO, (cpacf_mask_t *)
kvm_s390_available_subfunc.pckmo);
if (test_facility(77)) { /* MSA4 */
__cpacf_query(CPACF_KMCTR, (cpacf_mask_t *)
kvm_s390_available_subfunc.kmctr);
__cpacf_query(CPACF_KMF, (cpacf_mask_t *)
kvm_s390_available_subfunc.kmf);
__cpacf_query(CPACF_KMO, (cpacf_mask_t *)
kvm_s390_available_subfunc.kmo);
__cpacf_query(CPACF_PCC, (cpacf_mask_t *)
kvm_s390_available_subfunc.pcc);
}
if (test_facility(57)) /* MSA5 */
__cpacf_query(CPACF_PPNO, (cpacf_mask_t *)
kvm_s390_available_subfunc.ppno);
if (MACHINE_HAS_ESOP)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
/*
* We need SIE support, ESOP (PROT_READ protection for gmap_shadow),
* 64bit SCAO (SCA passthrough) and IDTE (for gmap_shadow unshadowing).
*/
if (!sclp.has_sief2 || !MACHINE_HAS_ESOP || !sclp.has_64bscao ||
!test_facility(3) || !nested)
return;
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIEF2);
if (sclp.has_64bscao)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_64BSCAO);
if (sclp.has_siif)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_SIIF);
if (sclp.has_gpere)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GPERE);
if (sclp.has_gsls)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_GSLS);
if (sclp.has_ib)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IB);
if (sclp.has_cei)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_CEI);
if (sclp.has_ibs)
allow_cpu_feat(KVM_S390_VM_CPU_FEAT_IBS);
/*
* KVM_S390_VM_CPU_FEAT_SKEY: Wrong shadow of PTE.I bits will make
* all skey handling functions read/set the skey from the PGSTE
* instead of the real storage key.
*
* KVM_S390_VM_CPU_FEAT_CMMA: Wrong shadow of PTE.I bits will make
* pages being detected as preserved although they are resident.
*
* KVM_S390_VM_CPU_FEAT_PFMFI: Wrong shadow of PTE.I bits will
* have the same effect as for KVM_S390_VM_CPU_FEAT_SKEY.
*
* For KVM_S390_VM_CPU_FEAT_SKEY, KVM_S390_VM_CPU_FEAT_CMMA and
* KVM_S390_VM_CPU_FEAT_PFMFI, all PTE.I and PGSTE bits have to be
* correctly shadowed. We can do that for the PGSTE but not for PTE.I.
*
* KVM_S390_VM_CPU_FEAT_SIGPIF: Wrong SCB addresses in the SCA. We
* cannot easily shadow the SCA because of the ipte lock.
*/
}
int kvm_arch_init(void *opaque)
{
kvm_s390_dbf = debug_register("kvm-trace", 32, 1, 7 * sizeof(long));
if (!kvm_s390_dbf)
return -ENOMEM;
if (debug_register_view(kvm_s390_dbf, &debug_sprintf_view)) {
debug_unregister(kvm_s390_dbf);
return -ENOMEM;
}
kvm_s390_cpu_feat_init();
/* Register floating interrupt controller interface. */
return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
}
void kvm_arch_exit(void)
{
debug_unregister(kvm_s390_dbf);
}
/* Section: device related */
long kvm_arch_dev_ioctl(struct file *filp,
unsigned int ioctl, unsigned long arg)
{
if (ioctl == KVM_S390_ENABLE_SIE)
return s390_enable_sie();
return -EINVAL;
}
int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
{
int r;
switch (ext) {
case KVM_CAP_S390_PSW:
case KVM_CAP_S390_GMAP:
case KVM_CAP_SYNC_MMU:
#ifdef CONFIG_KVM_S390_UCONTROL
case KVM_CAP_S390_UCONTROL:
#endif
case KVM_CAP_ASYNC_PF:
case KVM_CAP_SYNC_REGS:
case KVM_CAP_ONE_REG:
case KVM_CAP_ENABLE_CAP:
case KVM_CAP_S390_CSS_SUPPORT:
case KVM_CAP_IOEVENTFD:
case KVM_CAP_DEVICE_CTRL:
case KVM_CAP_ENABLE_CAP_VM:
case KVM_CAP_S390_IRQCHIP:
case KVM_CAP_VM_ATTRIBUTES:
case KVM_CAP_MP_STATE:
case KVM_CAP_S390_INJECT_IRQ:
case KVM_CAP_S390_USER_SIGP:
case KVM_CAP_S390_USER_STSI:
case KVM_CAP_S390_SKEYS:
case KVM_CAP_S390_IRQ_STATE:
case KVM_CAP_S390_USER_INSTR0:
r = 1;
break;
case KVM_CAP_S390_MEM_OP:
r = MEM_OP_MAX_SIZE;
break;
case KVM_CAP_NR_VCPUS:
case KVM_CAP_MAX_VCPUS:
r = KVM_S390_BSCA_CPU_SLOTS;
if (!kvm_s390_use_sca_entries())
r = KVM_MAX_VCPUS;
else if (sclp.has_esca && sclp.has_64bscao)
r = KVM_S390_ESCA_CPU_SLOTS;
break;
case KVM_CAP_NR_MEMSLOTS:
r = KVM_USER_MEM_SLOTS;
break;
case KVM_CAP_S390_COW:
s390/mm: implement software dirty bits The s390 architecture is unique in respect to dirty page detection, it uses the change bit in the per-page storage key to track page modifications. All other architectures track dirty bits by means of page table entries. This property of s390 has caused numerous problems in the past, e.g. see git commit ef5d437f71afdf4a "mm: fix XFS oops due to dirty pages without buffers on s390". To avoid future issues in regard to per-page dirty bits convert s390 to a fault based software dirty bit detection mechanism. All user page table entries which are marked as clean will be hardware read-only, even if the pte is supposed to be writable. A write by the user process will trigger a protection fault which will cause the user pte to be marked as dirty and the hardware read-only bit is removed. With this change the dirty bit in the storage key is irrelevant for Linux as a host, but the storage key is still required for KVM guests. The effect is that page_test_and_clear_dirty and the related code can be removed. The referenced bit in the storage key is still used by the page_test_and_clear_young primitive to provide page age information. For page cache pages of mappings with mapping_cap_account_dirty there will not be any change in behavior as the dirty bit tracking already uses read-only ptes to control the amount of dirty pages. Only for swap cache pages and pages of mappings without mapping_cap_account_dirty there can be additional protection faults. To avoid an excessive number of additional faults the mk_pte primitive checks for PageDirty if the pgprot value allows for writes and pre-dirties the pte. That avoids all additional faults for tmpfs and shmem pages until these pages are added to the swap cache. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2012-11-07 20:17:37 +08:00
r = MACHINE_HAS_ESOP;
break;
case KVM_CAP_S390_VECTOR_REGISTERS:
r = MACHINE_HAS_VX;
break;
case KVM_CAP_S390_RI:
r = test_facility(64);
break;
default:
r = 0;
}
return r;
}
static void kvm_s390_sync_dirty_log(struct kvm *kvm,
struct kvm_memory_slot *memslot)
{
gfn_t cur_gfn, last_gfn;
unsigned long address;
struct gmap *gmap = kvm->arch.gmap;
/* Loop over all guest pages */
last_gfn = memslot->base_gfn + memslot->npages;
for (cur_gfn = memslot->base_gfn; cur_gfn <= last_gfn; cur_gfn++) {
address = gfn_to_hva_memslot(memslot, cur_gfn);
if (test_and_clear_guest_dirty(gmap->mm, address))
mark_page_dirty(kvm, cur_gfn);
if (fatal_signal_pending(current))
return;
KVM: s390: do not block CPU on dirty logging When doing dirty logging on huge guests (e.g.600GB) we sometimes get rcu stall timeouts with backtraces like [ 2753.194083] ([<0000000000112fb2>] show_trace+0x12a/0x130) [ 2753.194092] [<0000000000113024>] show_stack+0x6c/0xe8 [ 2753.194094] [<00000000001ee6a8>] rcu_pending+0x358/0xa48 [ 2753.194099] [<00000000001f20cc>] rcu_check_callbacks+0x84/0x168 [ 2753.194102] [<0000000000167654>] update_process_times+0x54/0x80 [ 2753.194107] [<00000000001bdb5c>] tick_sched_handle.isra.16+0x4c/0x60 [ 2753.194113] [<00000000001bdbd8>] tick_sched_timer+0x68/0x90 [ 2753.194115] [<0000000000182a88>] __run_hrtimer+0x88/0x1f8 [ 2753.194119] [<00000000001838ba>] hrtimer_interrupt+0x122/0x2b0 [ 2753.194121] [<000000000010d034>] do_extint+0x16c/0x170 [ 2753.194123] [<00000000005e206e>] ext_skip+0x38/0x3e [ 2753.194129] [<000000000012157c>] gmap_test_and_clear_dirty+0xcc/0x118 [ 2753.194134] ([<00000000001214ea>] gmap_test_and_clear_dirty+0x3a/0x118) [ 2753.194137] [<0000000000132da4>] kvm_vm_ioctl_get_dirty_log+0xd4/0x1b0 [ 2753.194143] [<000000000012ac12>] kvm_vm_ioctl+0x21a/0x548 [ 2753.194146] [<00000000002b57f6>] do_vfs_ioctl+0x30e/0x518 [ 2753.194149] [<00000000002b5a9c>] SyS_ioctl+0x9c/0xb0 [ 2753.194151] [<00000000005e1ae6>] sysc_tracego+0x14/0x1a [ 2753.194153] [<000003ffb75f3972>] 0x3ffb75f3972 We should do a cond_resched in here. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-02 22:15:56 +08:00
cond_resched();
}
}
/* Section: vm related */
static void sca_del_vcpu(struct kvm_vcpu *vcpu);
/*
* Get (and clear) the dirty memory log for a memory slot.
*/
int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm,
struct kvm_dirty_log *log)
{
int r;
unsigned long n;
struct kvm_memslots *slots;
struct kvm_memory_slot *memslot;
int is_dirty = 0;
mutex_lock(&kvm->slots_lock);
r = -EINVAL;
if (log->slot >= KVM_USER_MEM_SLOTS)
goto out;
slots = kvm_memslots(kvm);
memslot = id_to_memslot(slots, log->slot);
r = -ENOENT;
if (!memslot->dirty_bitmap)
goto out;
kvm_s390_sync_dirty_log(kvm, memslot);
r = kvm_get_dirty_log(kvm, log, &is_dirty);
if (r)
goto out;
/* Clear the dirty log */
if (is_dirty) {
n = kvm_dirty_bitmap_bytes(memslot);
memset(memslot->dirty_bitmap, 0, n);
}
r = 0;
out:
mutex_unlock(&kvm->slots_lock);
return r;
}
static void icpt_operexc_on_all_vcpus(struct kvm *kvm)
{
unsigned int i;
struct kvm_vcpu *vcpu;
kvm_for_each_vcpu(i, vcpu, kvm) {
kvm_s390_sync_request(KVM_REQ_ICPT_OPEREXC, vcpu);
}
}
static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap)
{
int r;
if (cap->flags)
return -EINVAL;
switch (cap->cap) {
case KVM_CAP_S390_IRQCHIP:
VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_IRQCHIP");
kvm->arch.use_irqchip = 1;
r = 0;
break;
case KVM_CAP_S390_USER_SIGP:
VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_SIGP");
kvm->arch.user_sigp = 1;
r = 0;
break;
case KVM_CAP_S390_VECTOR_REGISTERS:
mutex_lock(&kvm->lock);
if (kvm->created_vcpus) {
r = -EBUSY;
} else if (MACHINE_HAS_VX) {
set_kvm_facility(kvm->arch.model.fac_mask, 129);
set_kvm_facility(kvm->arch.model.fac_list, 129);
r = 0;
} else
r = -EINVAL;
mutex_unlock(&kvm->lock);
VM_EVENT(kvm, 3, "ENABLE: CAP_S390_VECTOR_REGISTERS %s",
r ? "(not available)" : "(success)");
break;
case KVM_CAP_S390_RI:
r = -EINVAL;
mutex_lock(&kvm->lock);
if (kvm->created_vcpus) {
r = -EBUSY;
} else if (test_facility(64)) {
set_kvm_facility(kvm->arch.model.fac_mask, 64);
set_kvm_facility(kvm->arch.model.fac_list, 64);
r = 0;
}
mutex_unlock(&kvm->lock);
VM_EVENT(kvm, 3, "ENABLE: CAP_S390_RI %s",
r ? "(not available)" : "(success)");
break;
case KVM_CAP_S390_USER_STSI:
VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_STSI");
kvm->arch.user_stsi = 1;
r = 0;
break;
case KVM_CAP_S390_USER_INSTR0:
VM_EVENT(kvm, 3, "%s", "ENABLE: CAP_S390_USER_INSTR0");
kvm->arch.user_instr0 = 1;
icpt_operexc_on_all_vcpus(kvm);
r = 0;
break;
default:
r = -EINVAL;
break;
}
return r;
}
static int kvm_s390_get_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
switch (attr->attr) {
case KVM_S390_VM_MEM_LIMIT_SIZE:
ret = 0;
VM_EVENT(kvm, 3, "QUERY: max guest memory: %lu bytes",
kvm->arch.mem_limit);
if (put_user(kvm->arch.mem_limit, (u64 __user *)attr->addr))
ret = -EFAULT;
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
unsigned int idx;
switch (attr->attr) {
case KVM_S390_VM_MEM_ENABLE_CMMA:
ret = -ENXIO;
if (!sclp.has_cmma)
break;
ret = -EBUSY;
VM_EVENT(kvm, 3, "%s", "ENABLE: CMMA support");
mutex_lock(&kvm->lock);
if (!kvm->created_vcpus) {
kvm->arch.use_cmma = 1;
ret = 0;
}
mutex_unlock(&kvm->lock);
break;
case KVM_S390_VM_MEM_CLR_CMMA:
ret = -ENXIO;
if (!sclp.has_cmma)
break;
ret = -EINVAL;
if (!kvm->arch.use_cmma)
break;
VM_EVENT(kvm, 3, "%s", "RESET: CMMA states");
mutex_lock(&kvm->lock);
idx = srcu_read_lock(&kvm->srcu);
s390_reset_cmma(kvm->arch.gmap->mm);
srcu_read_unlock(&kvm->srcu, idx);
mutex_unlock(&kvm->lock);
ret = 0;
break;
case KVM_S390_VM_MEM_LIMIT_SIZE: {
unsigned long new_limit;
if (kvm_is_ucontrol(kvm))
return -EINVAL;
if (get_user(new_limit, (u64 __user *)attr->addr))
return -EFAULT;
if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT &&
new_limit > kvm->arch.mem_limit)
return -E2BIG;
if (!new_limit)
return -EINVAL;
/* gmap_create takes last usable address */
if (new_limit != KVM_S390_NO_MEM_LIMIT)
new_limit -= 1;
ret = -EBUSY;
mutex_lock(&kvm->lock);
if (!kvm->created_vcpus) {
/* gmap_create will round the limit up */
struct gmap *new = gmap_create(current->mm, new_limit);
if (!new) {
ret = -ENOMEM;
} else {
gmap_remove(kvm->arch.gmap);
new->private = kvm;
kvm->arch.gmap = new;
ret = 0;
}
}
mutex_unlock(&kvm->lock);
VM_EVENT(kvm, 3, "SET: max guest address: %lu", new_limit);
VM_EVENT(kvm, 3, "New guest asce: 0x%pK",
(void *) kvm->arch.gmap->asce);
break;
}
default:
ret = -ENXIO;
break;
}
return ret;
}
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu);
static int kvm_s390_vm_set_crypto(struct kvm *kvm, struct kvm_device_attr *attr)
{
struct kvm_vcpu *vcpu;
int i;
if (!test_kvm_facility(kvm, 76))
return -EINVAL;
mutex_lock(&kvm->lock);
switch (attr->attr) {
case KVM_S390_VM_CRYPTO_ENABLE_AES_KW:
get_random_bytes(
kvm->arch.crypto.crycb->aes_wrapping_key_mask,
sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
kvm->arch.crypto.aes_kw = 1;
VM_EVENT(kvm, 3, "%s", "ENABLE: AES keywrapping support");
break;
case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW:
get_random_bytes(
kvm->arch.crypto.crycb->dea_wrapping_key_mask,
sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
kvm->arch.crypto.dea_kw = 1;
VM_EVENT(kvm, 3, "%s", "ENABLE: DEA keywrapping support");
break;
case KVM_S390_VM_CRYPTO_DISABLE_AES_KW:
kvm->arch.crypto.aes_kw = 0;
memset(kvm->arch.crypto.crycb->aes_wrapping_key_mask, 0,
sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
VM_EVENT(kvm, 3, "%s", "DISABLE: AES keywrapping support");
break;
case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW:
kvm->arch.crypto.dea_kw = 0;
memset(kvm->arch.crypto.crycb->dea_wrapping_key_mask, 0,
sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
VM_EVENT(kvm, 3, "%s", "DISABLE: DEA keywrapping support");
break;
default:
mutex_unlock(&kvm->lock);
return -ENXIO;
}
kvm_for_each_vcpu(i, vcpu, kvm) {
kvm_s390_vcpu_crypto_setup(vcpu);
exit_sie(vcpu);
}
mutex_unlock(&kvm->lock);
return 0;
}
static int kvm_s390_set_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
{
u8 gtod_high;
if (copy_from_user(&gtod_high, (void __user *)attr->addr,
sizeof(gtod_high)))
return -EFAULT;
if (gtod_high != 0)
return -EINVAL;
VM_EVENT(kvm, 3, "SET: TOD extension: 0x%x", gtod_high);
return 0;
}
static int kvm_s390_set_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
u64 gtod;
if (copy_from_user(&gtod, (void __user *)attr->addr, sizeof(gtod)))
return -EFAULT;
kvm_s390_set_tod_clock(kvm, gtod);
VM_EVENT(kvm, 3, "SET: TOD base: 0x%llx", gtod);
return 0;
}
static int kvm_s390_set_tod(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
if (attr->flags)
return -EINVAL;
switch (attr->attr) {
case KVM_S390_VM_TOD_HIGH:
ret = kvm_s390_set_tod_high(kvm, attr);
break;
case KVM_S390_VM_TOD_LOW:
ret = kvm_s390_set_tod_low(kvm, attr);
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static int kvm_s390_get_tod_high(struct kvm *kvm, struct kvm_device_attr *attr)
{
u8 gtod_high = 0;
if (copy_to_user((void __user *)attr->addr, &gtod_high,
sizeof(gtod_high)))
return -EFAULT;
VM_EVENT(kvm, 3, "QUERY: TOD extension: 0x%x", gtod_high);
return 0;
}
static int kvm_s390_get_tod_low(struct kvm *kvm, struct kvm_device_attr *attr)
{
u64 gtod;
gtod = kvm_s390_get_tod_clock_fast(kvm);
if (copy_to_user((void __user *)attr->addr, &gtod, sizeof(gtod)))
return -EFAULT;
VM_EVENT(kvm, 3, "QUERY: TOD base: 0x%llx", gtod);
return 0;
}
static int kvm_s390_get_tod(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
if (attr->flags)
return -EINVAL;
switch (attr->attr) {
case KVM_S390_VM_TOD_HIGH:
ret = kvm_s390_get_tod_high(kvm, attr);
break;
case KVM_S390_VM_TOD_LOW:
ret = kvm_s390_get_tod_low(kvm, attr);
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_processor *proc;
u16 lowest_ibc, unblocked_ibc;
int ret = 0;
mutex_lock(&kvm->lock);
if (kvm->created_vcpus) {
ret = -EBUSY;
goto out;
}
proc = kzalloc(sizeof(*proc), GFP_KERNEL);
if (!proc) {
ret = -ENOMEM;
goto out;
}
if (!copy_from_user(proc, (void __user *)attr->addr,
sizeof(*proc))) {
kvm->arch.model.cpuid = proc->cpuid;
lowest_ibc = sclp.ibc >> 16 & 0xfff;
unblocked_ibc = sclp.ibc & 0xfff;
if (lowest_ibc && proc->ibc) {
if (proc->ibc > unblocked_ibc)
kvm->arch.model.ibc = unblocked_ibc;
else if (proc->ibc < lowest_ibc)
kvm->arch.model.ibc = lowest_ibc;
else
kvm->arch.model.ibc = proc->ibc;
}
memcpy(kvm->arch.model.fac_list, proc->fac_list,
S390_ARCH_FAC_LIST_SIZE_BYTE);
} else
ret = -EFAULT;
kfree(proc);
out:
mutex_unlock(&kvm->lock);
return ret;
}
static int kvm_s390_set_processor_feat(struct kvm *kvm,
struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_feat data;
int ret = -EBUSY;
if (copy_from_user(&data, (void __user *)attr->addr, sizeof(data)))
return -EFAULT;
if (!bitmap_subset((unsigned long *) data.feat,
kvm_s390_available_cpu_feat,
KVM_S390_VM_CPU_FEAT_NR_BITS))
return -EINVAL;
mutex_lock(&kvm->lock);
if (!atomic_read(&kvm->online_vcpus)) {
bitmap_copy(kvm->arch.cpu_feat, (unsigned long *) data.feat,
KVM_S390_VM_CPU_FEAT_NR_BITS);
ret = 0;
}
mutex_unlock(&kvm->lock);
return ret;
}
static int kvm_s390_set_processor_subfunc(struct kvm *kvm,
struct kvm_device_attr *attr)
{
/*
* Once supported by kernel + hw, we have to store the subfunctions
* in kvm->arch and remember that user space configured them.
*/
return -ENXIO;
}
static int kvm_s390_set_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret = -ENXIO;
switch (attr->attr) {
case KVM_S390_VM_CPU_PROCESSOR:
ret = kvm_s390_set_processor(kvm, attr);
break;
case KVM_S390_VM_CPU_PROCESSOR_FEAT:
ret = kvm_s390_set_processor_feat(kvm, attr);
break;
case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
ret = kvm_s390_set_processor_subfunc(kvm, attr);
break;
}
return ret;
}
static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_processor *proc;
int ret = 0;
proc = kzalloc(sizeof(*proc), GFP_KERNEL);
if (!proc) {
ret = -ENOMEM;
goto out;
}
proc->cpuid = kvm->arch.model.cpuid;
proc->ibc = kvm->arch.model.ibc;
memcpy(&proc->fac_list, kvm->arch.model.fac_list,
S390_ARCH_FAC_LIST_SIZE_BYTE);
if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc)))
ret = -EFAULT;
kfree(proc);
out:
return ret;
}
static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_machine *mach;
int ret = 0;
mach = kzalloc(sizeof(*mach), GFP_KERNEL);
if (!mach) {
ret = -ENOMEM;
goto out;
}
get_cpu_id((struct cpuid *) &mach->cpuid);
mach->ibc = sclp.ibc;
memcpy(&mach->fac_mask, kvm->arch.model.fac_mask,
S390_ARCH_FAC_LIST_SIZE_BYTE);
memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list,
S390_ARCH_FAC_LIST_SIZE_BYTE);
if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach)))
ret = -EFAULT;
kfree(mach);
out:
return ret;
}
static int kvm_s390_get_processor_feat(struct kvm *kvm,
struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_feat data;
bitmap_copy((unsigned long *) data.feat, kvm->arch.cpu_feat,
KVM_S390_VM_CPU_FEAT_NR_BITS);
if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
return -EFAULT;
return 0;
}
static int kvm_s390_get_machine_feat(struct kvm *kvm,
struct kvm_device_attr *attr)
{
struct kvm_s390_vm_cpu_feat data;
bitmap_copy((unsigned long *) data.feat,
kvm_s390_available_cpu_feat,
KVM_S390_VM_CPU_FEAT_NR_BITS);
if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
return -EFAULT;
return 0;
}
static int kvm_s390_get_processor_subfunc(struct kvm *kvm,
struct kvm_device_attr *attr)
{
/*
* Once we can actually configure subfunctions (kernel + hw support),
* we have to check if they were already set by user space, if so copy
* them from kvm->arch.
*/
return -ENXIO;
}
static int kvm_s390_get_machine_subfunc(struct kvm *kvm,
struct kvm_device_attr *attr)
{
if (copy_to_user((void __user *)attr->addr, &kvm_s390_available_subfunc,
sizeof(struct kvm_s390_vm_cpu_subfunc)))
return -EFAULT;
return 0;
}
static int kvm_s390_get_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret = -ENXIO;
switch (attr->attr) {
case KVM_S390_VM_CPU_PROCESSOR:
ret = kvm_s390_get_processor(kvm, attr);
break;
case KVM_S390_VM_CPU_MACHINE:
ret = kvm_s390_get_machine(kvm, attr);
break;
case KVM_S390_VM_CPU_PROCESSOR_FEAT:
ret = kvm_s390_get_processor_feat(kvm, attr);
break;
case KVM_S390_VM_CPU_MACHINE_FEAT:
ret = kvm_s390_get_machine_feat(kvm, attr);
break;
case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
ret = kvm_s390_get_processor_subfunc(kvm, attr);
break;
case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
ret = kvm_s390_get_machine_subfunc(kvm, attr);
break;
}
return ret;
}
static int kvm_s390_vm_set_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
switch (attr->group) {
case KVM_S390_VM_MEM_CTRL:
ret = kvm_s390_set_mem_control(kvm, attr);
break;
case KVM_S390_VM_TOD:
ret = kvm_s390_set_tod(kvm, attr);
break;
case KVM_S390_VM_CPU_MODEL:
ret = kvm_s390_set_cpu_model(kvm, attr);
break;
case KVM_S390_VM_CRYPTO:
ret = kvm_s390_vm_set_crypto(kvm, attr);
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static int kvm_s390_vm_get_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
switch (attr->group) {
case KVM_S390_VM_MEM_CTRL:
ret = kvm_s390_get_mem_control(kvm, attr);
break;
case KVM_S390_VM_TOD:
ret = kvm_s390_get_tod(kvm, attr);
break;
case KVM_S390_VM_CPU_MODEL:
ret = kvm_s390_get_cpu_model(kvm, attr);
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr)
{
int ret;
switch (attr->group) {
case KVM_S390_VM_MEM_CTRL:
switch (attr->attr) {
case KVM_S390_VM_MEM_ENABLE_CMMA:
case KVM_S390_VM_MEM_CLR_CMMA:
ret = sclp.has_cmma ? 0 : -ENXIO;
break;
case KVM_S390_VM_MEM_LIMIT_SIZE:
ret = 0;
break;
default:
ret = -ENXIO;
break;
}
break;
case KVM_S390_VM_TOD:
switch (attr->attr) {
case KVM_S390_VM_TOD_LOW:
case KVM_S390_VM_TOD_HIGH:
ret = 0;
break;
default:
ret = -ENXIO;
break;
}
break;
case KVM_S390_VM_CPU_MODEL:
switch (attr->attr) {
case KVM_S390_VM_CPU_PROCESSOR:
case KVM_S390_VM_CPU_MACHINE:
case KVM_S390_VM_CPU_PROCESSOR_FEAT:
case KVM_S390_VM_CPU_MACHINE_FEAT:
case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
ret = 0;
break;
/* configuring subfunctions is not supported yet */
case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
default:
ret = -ENXIO;
break;
}
break;
case KVM_S390_VM_CRYPTO:
switch (attr->attr) {
case KVM_S390_VM_CRYPTO_ENABLE_AES_KW:
case KVM_S390_VM_CRYPTO_ENABLE_DEA_KW:
case KVM_S390_VM_CRYPTO_DISABLE_AES_KW:
case KVM_S390_VM_CRYPTO_DISABLE_DEA_KW:
ret = 0;
break;
default:
ret = -ENXIO;
break;
}
break;
default:
ret = -ENXIO;
break;
}
return ret;
}
static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
uint8_t *keys;
uint64_t hva;
int i, r = 0;
if (args->flags != 0)
return -EINVAL;
/* Is this guest using storage keys? */
if (!mm_use_skey(current->mm))
return KVM_S390_GET_SKEYS_NONE;
/* Enforce sane limit on memory allocation */
if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
return -EINVAL;
keys = kmalloc_array(args->count, sizeof(uint8_t),
GFP_KERNEL | __GFP_NOWARN);
if (!keys)
keys = vmalloc(sizeof(uint8_t) * args->count);
if (!keys)
return -ENOMEM;
down_read(&current->mm->mmap_sem);
for (i = 0; i < args->count; i++) {
hva = gfn_to_hva(kvm, args->start_gfn + i);
if (kvm_is_error_hva(hva)) {
r = -EFAULT;
break;
}
r = get_guest_storage_key(current->mm, hva, &keys[i]);
if (r)
break;
}
up_read(&current->mm->mmap_sem);
if (!r) {
r = copy_to_user((uint8_t __user *)args->skeydata_addr, keys,
sizeof(uint8_t) * args->count);
if (r)
r = -EFAULT;
}
kvfree(keys);
return r;
}
static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
{
uint8_t *keys;
uint64_t hva;
int i, r = 0;
if (args->flags != 0)
return -EINVAL;
/* Enforce sane limit on memory allocation */
if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
return -EINVAL;
keys = kmalloc_array(args->count, sizeof(uint8_t),
GFP_KERNEL | __GFP_NOWARN);
if (!keys)
keys = vmalloc(sizeof(uint8_t) * args->count);
if (!keys)
return -ENOMEM;
r = copy_from_user(keys, (uint8_t __user *)args->skeydata_addr,
sizeof(uint8_t) * args->count);
if (r) {
r = -EFAULT;
goto out;
}
/* Enable storage key handling for the guest */
r = s390_enable_skey();
if (r)
goto out;
down_read(&current->mm->mmap_sem);
for (i = 0; i < args->count; i++) {
hva = gfn_to_hva(kvm, args->start_gfn + i);
if (kvm_is_error_hva(hva)) {
r = -EFAULT;
break;
}
/* Lowest order bit is reserved */
if (keys[i] & 0x01) {
r = -EINVAL;
break;
}
r = set_guest_storage_key(current->mm, hva, keys[i], 0);
if (r)
break;
}
up_read(&current->mm->mmap_sem);
out:
kvfree(keys);
return r;
}
long kvm_arch_vm_ioctl(struct file *filp,
unsigned int ioctl, unsigned long arg)
{
struct kvm *kvm = filp->private_data;
void __user *argp = (void __user *)arg;
struct kvm_device_attr attr;
int r;
switch (ioctl) {
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
case KVM_S390_INTERRUPT: {
struct kvm_s390_interrupt s390int;
r = -EFAULT;
if (copy_from_user(&s390int, argp, sizeof(s390int)))
break;
r = kvm_s390_inject_vm(kvm, &s390int);
break;
}
case KVM_ENABLE_CAP: {
struct kvm_enable_cap cap;
r = -EFAULT;
if (copy_from_user(&cap, argp, sizeof(cap)))
break;
r = kvm_vm_ioctl_enable_cap(kvm, &cap);
break;
}
case KVM_CREATE_IRQCHIP: {
struct kvm_irq_routing_entry routing;
r = -EINVAL;
if (kvm->arch.use_irqchip) {
/* Set up dummy routing. */
memset(&routing, 0, sizeof(routing));
r = kvm_set_irq_routing(kvm, &routing, 0, 0);
}
break;
}
case KVM_SET_DEVICE_ATTR: {
r = -EFAULT;
if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
break;
r = kvm_s390_vm_set_attr(kvm, &attr);
break;
}
case KVM_GET_DEVICE_ATTR: {
r = -EFAULT;
if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
break;
r = kvm_s390_vm_get_attr(kvm, &attr);
break;
}
case KVM_HAS_DEVICE_ATTR: {
r = -EFAULT;
if (copy_from_user(&attr, (void __user *)arg, sizeof(attr)))
break;
r = kvm_s390_vm_has_attr(kvm, &attr);
break;
}
case KVM_S390_GET_SKEYS: {
struct kvm_s390_skeys args;
r = -EFAULT;
if (copy_from_user(&args, argp,
sizeof(struct kvm_s390_skeys)))
break;
r = kvm_s390_get_skeys(kvm, &args);
break;
}
case KVM_S390_SET_SKEYS: {
struct kvm_s390_skeys args;
r = -EFAULT;
if (copy_from_user(&args, argp,
sizeof(struct kvm_s390_skeys)))
break;
r = kvm_s390_set_skeys(kvm, &args);
break;
}
default:
r = -ENOTTY;
}
return r;
}
static int kvm_s390_query_ap_config(u8 *config)
{
u32 fcn_code = 0x04000000UL;
u32 cc = 0;
memset(config, 0, 128);
asm volatile(
"lgr 0,%1\n"
"lgr 2,%2\n"
".long 0xb2af0000\n" /* PQAP(QCI) */
"0: ipm %0\n"
"srl %0,28\n"
"1:\n"
EX_TABLE(0b, 1b)
: "+r" (cc)
: "r" (fcn_code), "r" (config)
: "cc", "0", "2", "memory"
);
return cc;
}
static int kvm_s390_apxa_installed(void)
{
u8 config[128];
int cc;
if (test_facility(12)) {
cc = kvm_s390_query_ap_config(config);
if (cc)
pr_err("PQAP(QCI) failed with cc=%d", cc);
else
return config[0] & 0x40;
}
return 0;
}
static void kvm_s390_set_crycb_format(struct kvm *kvm)
{
kvm->arch.crypto.crycbd = (__u32)(unsigned long) kvm->arch.crypto.crycb;
if (kvm_s390_apxa_installed())
kvm->arch.crypto.crycbd |= CRYCB_FORMAT2;
else
kvm->arch.crypto.crycbd |= CRYCB_FORMAT1;
}
static u64 kvm_s390_get_initial_cpuid(void)
{
struct cpuid cpuid;
get_cpu_id(&cpuid);
cpuid.version = 0xff;
return *((u64 *) &cpuid);
}
static void kvm_s390_crypto_init(struct kvm *kvm)
{
if (!test_kvm_facility(kvm, 76))
return;
kvm->arch.crypto.crycb = &kvm->arch.sie_page2->crycb;
kvm_s390_set_crycb_format(kvm);
/* Enable AES/DEA protected key functions by default */
kvm->arch.crypto.aes_kw = 1;
kvm->arch.crypto.dea_kw = 1;
get_random_bytes(kvm->arch.crypto.crycb->aes_wrapping_key_mask,
sizeof(kvm->arch.crypto.crycb->aes_wrapping_key_mask));
get_random_bytes(kvm->arch.crypto.crycb->dea_wrapping_key_mask,
sizeof(kvm->arch.crypto.crycb->dea_wrapping_key_mask));
}
static void sca_dispose(struct kvm *kvm)
{
if (kvm->arch.use_esca)
free_pages_exact(kvm->arch.sca, sizeof(struct esca_block));
else
free_page((unsigned long)(kvm->arch.sca));
kvm->arch.sca = NULL;
}
int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
{
gfp_t alloc_flags = GFP_KERNEL;
int i, rc;
char debug_name[16];
static unsigned long sca_offset;
rc = -EINVAL;
#ifdef CONFIG_KVM_S390_UCONTROL
if (type & ~KVM_VM_S390_UCONTROL)
goto out_err;
if ((type & KVM_VM_S390_UCONTROL) && (!capable(CAP_SYS_ADMIN)))
goto out_err;
#else
if (type)
goto out_err;
#endif
rc = s390_enable_sie();
if (rc)
goto out_err;
rc = -ENOMEM;
ratelimit_state_init(&kvm->arch.sthyi_limit, 5 * HZ, 500);
kvm->arch.use_esca = 0; /* start with basic SCA */
if (!sclp.has_64bscao)
alloc_flags |= GFP_DMA;
rwlock_init(&kvm->arch.sca_lock);
kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
if (!kvm->arch.sca)
goto out_err;
spin_lock(&kvm_lock);
sca_offset += 16;
if (sca_offset + sizeof(struct bsca_block) > PAGE_SIZE)
sca_offset = 0;
kvm->arch.sca = (struct bsca_block *)
((char *) kvm->arch.sca + sca_offset);
spin_unlock(&kvm_lock);
sprintf(debug_name, "kvm-%u", current->pid);
kvm->arch.dbf = debug_register(debug_name, 32, 1, 7 * sizeof(long));
if (!kvm->arch.dbf)
goto out_err;
kvm->arch.sie_page2 =
(struct sie_page2 *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
if (!kvm->arch.sie_page2)
goto out_err;
/* Populate the facility mask initially. */
memcpy(kvm->arch.model.fac_mask, S390_lowcore.stfle_fac_list,
S390_ARCH_FAC_LIST_SIZE_BYTE);
for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) {
if (i < kvm_s390_fac_list_mask_size())
kvm->arch.model.fac_mask[i] &= kvm_s390_fac_list_mask[i];
else
kvm->arch.model.fac_mask[i] = 0UL;
}
/* Populate the facility list initially. */
kvm->arch.model.fac_list = kvm->arch.sie_page2->fac_list;
memcpy(kvm->arch.model.fac_list, kvm->arch.model.fac_mask,
S390_ARCH_FAC_LIST_SIZE_BYTE);
set_kvm_facility(kvm->arch.model.fac_mask, 74);
set_kvm_facility(kvm->arch.model.fac_list, 74);
kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
kvm->arch.model.ibc = sclp.ibc & 0x0fff;
kvm_s390_crypto_init(kvm);
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
spin_lock_init(&kvm->arch.float_int.lock);
for (i = 0; i < FIRQ_LIST_COUNT; i++)
INIT_LIST_HEAD(&kvm->arch.float_int.lists[i]);
init_waitqueue_head(&kvm->arch.ipte_wq);
mutex_init(&kvm->arch.ipte_mutex);
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
debug_register_view(kvm->arch.dbf, &debug_sprintf_view);
VM_EVENT(kvm, 3, "vm created with type %lu", type);
if (type & KVM_VM_S390_UCONTROL) {
kvm->arch.gmap = NULL;
kvm->arch.mem_limit = KVM_S390_NO_MEM_LIMIT;
} else {
if (sclp.hamax == U64_MAX)
kvm->arch.mem_limit = TASK_MAX_SIZE;
else
kvm->arch.mem_limit = min_t(unsigned long, TASK_MAX_SIZE,
sclp.hamax + 1);
kvm->arch.gmap = gmap_create(current->mm, kvm->arch.mem_limit - 1);
if (!kvm->arch.gmap)
goto out_err;
kvm->arch.gmap->private = kvm;
kvm->arch.gmap->pfault_enabled = 0;
}
kvm->arch.css_support = 0;
kvm->arch.use_irqchip = 0;
kvm->arch.epoch = 0;
spin_lock_init(&kvm->arch.start_stop_lock);
kvm_s390_vsie_init(kvm);
KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
return 0;
out_err:
free_page((unsigned long)kvm->arch.sie_page2);
debug_unregister(kvm->arch.dbf);
sca_dispose(kvm);
KVM_EVENT(3, "creation of vm failed: %d", rc);
return rc;
}
bool kvm_arch_has_vcpu_debugfs(void)
{
return false;
}
int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
{
return 0;
}
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
{
VCPU_EVENT(vcpu, 3, "%s", "free cpu");
trace_kvm_s390_destroy_vcpu(vcpu->vcpu_id);
kvm_s390_clear_local_irqs(vcpu);
kvm_clear_async_pf_completion_queue(vcpu);
if (!kvm_is_ucontrol(vcpu->kvm))
sca_del_vcpu(vcpu);
if (kvm_is_ucontrol(vcpu->kvm))
gmap_remove(vcpu->arch.gmap);
if (vcpu->kvm->arch.use_cmma)
kvm_s390_vcpu_unsetup_cmma(vcpu);
free_page((unsigned long)(vcpu->arch.sie_block));
kvm_vcpu_uninit(vcpu);
kmem_cache_free(kvm_vcpu_cache, vcpu);
}
static void kvm_free_vcpus(struct kvm *kvm)
{
unsigned int i;
struct kvm_vcpu *vcpu;
kvm_for_each_vcpu(i, vcpu, kvm)
kvm_arch_vcpu_destroy(vcpu);
mutex_lock(&kvm->lock);
for (i = 0; i < atomic_read(&kvm->online_vcpus); i++)
kvm->vcpus[i] = NULL;
atomic_set(&kvm->online_vcpus, 0);
mutex_unlock(&kvm->lock);
}
void kvm_arch_destroy_vm(struct kvm *kvm)
{
kvm_free_vcpus(kvm);
sca_dispose(kvm);
debug_unregister(kvm->arch.dbf);
free_page((unsigned long)kvm->arch.sie_page2);
if (!kvm_is_ucontrol(kvm))
gmap_remove(kvm->arch.gmap);
kvm_s390_destroy_adapters(kvm);
kvm_s390_clear_float_irqs(kvm);
kvm_s390_vsie_destroy(kvm);
KVM_EVENT(3, "vm 0x%pK destroyed", kvm);
}
/* Section: vcpu related */
static int __kvm_ucontrol_vcpu_init(struct kvm_vcpu *vcpu)
{
vcpu->arch.gmap = gmap_create(current->mm, -1UL);
if (!vcpu->arch.gmap)
return -ENOMEM;
vcpu->arch.gmap->private = vcpu->kvm;
return 0;
}
static void sca_del_vcpu(struct kvm_vcpu *vcpu)
{
if (!kvm_s390_use_sca_entries())
return;
read_lock(&vcpu->kvm->arch.sca_lock);
if (vcpu->kvm->arch.use_esca) {
struct esca_block *sca = vcpu->kvm->arch.sca;
clear_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
sca->cpu[vcpu->vcpu_id].sda = 0;
} else {
struct bsca_block *sca = vcpu->kvm->arch.sca;
clear_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
sca->cpu[vcpu->vcpu_id].sda = 0;
}
read_unlock(&vcpu->kvm->arch.sca_lock);
}
static void sca_add_vcpu(struct kvm_vcpu *vcpu)
{
if (!kvm_s390_use_sca_entries()) {
struct bsca_block *sca = vcpu->kvm->arch.sca;
/* we still need the basic sca for the ipte control */
vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
}
read_lock(&vcpu->kvm->arch.sca_lock);
if (vcpu->kvm->arch.use_esca) {
struct esca_block *sca = vcpu->kvm->arch.sca;
sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
vcpu->arch.sie_block->scaol = (__u32)(__u64)sca & ~0x3fU;
vcpu->arch.sie_block->ecb2 |= 0x04U;
set_bit_inv(vcpu->vcpu_id, (unsigned long *) sca->mcn);
} else {
struct bsca_block *sca = vcpu->kvm->arch.sca;
sca->cpu[vcpu->vcpu_id].sda = (__u64) vcpu->arch.sie_block;
vcpu->arch.sie_block->scaoh = (__u32)(((__u64)sca) >> 32);
vcpu->arch.sie_block->scaol = (__u32)(__u64)sca;
set_bit_inv(vcpu->vcpu_id, (unsigned long *) &sca->mcn);
}
read_unlock(&vcpu->kvm->arch.sca_lock);
}
/* Basic SCA to Extended SCA data copy routines */
static inline void sca_copy_entry(struct esca_entry *d, struct bsca_entry *s)
{
d->sda = s->sda;
d->sigp_ctrl.c = s->sigp_ctrl.c;
d->sigp_ctrl.scn = s->sigp_ctrl.scn;
}
static void sca_copy_b_to_e(struct esca_block *d, struct bsca_block *s)
{
int i;
d->ipte_control = s->ipte_control;
d->mcn[0] = s->mcn;
for (i = 0; i < KVM_S390_BSCA_CPU_SLOTS; i++)
sca_copy_entry(&d->cpu[i], &s->cpu[i]);
}
static int sca_switch_to_extended(struct kvm *kvm)
{
struct bsca_block *old_sca = kvm->arch.sca;
struct esca_block *new_sca;
struct kvm_vcpu *vcpu;
unsigned int vcpu_idx;
u32 scaol, scaoh;
new_sca = alloc_pages_exact(sizeof(*new_sca), GFP_KERNEL|__GFP_ZERO);
if (!new_sca)
return -ENOMEM;
scaoh = (u32)((u64)(new_sca) >> 32);
scaol = (u32)(u64)(new_sca) & ~0x3fU;
kvm_s390_vcpu_block_all(kvm);
write_lock(&kvm->arch.sca_lock);
sca_copy_b_to_e(new_sca, old_sca);
kvm_for_each_vcpu(vcpu_idx, vcpu, kvm) {
vcpu->arch.sie_block->scaoh = scaoh;
vcpu->arch.sie_block->scaol = scaol;
vcpu->arch.sie_block->ecb2 |= 0x04U;
}
kvm->arch.sca = new_sca;
kvm->arch.use_esca = 1;
write_unlock(&kvm->arch.sca_lock);
kvm_s390_vcpu_unblock_all(kvm);
free_page((unsigned long)old_sca);
VM_EVENT(kvm, 2, "Switched to ESCA (0x%pK -> 0x%pK)",
old_sca, kvm->arch.sca);
return 0;
}
static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
{
int rc;
if (!kvm_s390_use_sca_entries()) {
if (id < KVM_MAX_VCPUS)
return true;
return false;
}
if (id < KVM_S390_BSCA_CPU_SLOTS)
return true;
if (!sclp.has_esca || !sclp.has_64bscao)
return false;
mutex_lock(&kvm->lock);
rc = kvm->arch.use_esca ? 0 : sca_switch_to_extended(kvm);
mutex_unlock(&kvm->lock);
return rc == 0 && id < KVM_S390_ESCA_CPU_SLOTS;
}
int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
{
vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
kvm_clear_async_pf_completion_queue(vcpu);
vcpu->run->kvm_valid_regs = KVM_SYNC_PREFIX |
KVM_SYNC_GPRS |
KVM_SYNC_ACRS |
KVM_SYNC_CRS |
KVM_SYNC_ARCH0 |
KVM_SYNC_PFAULT;
kvm_s390_set_prefix(vcpu, 0);
if (test_kvm_facility(vcpu->kvm, 64))
vcpu->run->kvm_valid_regs |= KVM_SYNC_RICCB;
/* fprs can be synchronized via vrs, even if the guest has no vx. With
* MACHINE_HAS_VX, (load|store)_fpu_regs() will work with vrs format.
*/
if (MACHINE_HAS_VX)
vcpu->run->kvm_valid_regs |= KVM_SYNC_VRS;
else
vcpu->run->kvm_valid_regs |= KVM_SYNC_FPRS;
if (kvm_is_ucontrol(vcpu->kvm))
return __kvm_ucontrol_vcpu_init(vcpu);
return 0;
}
/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __start_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
WARN_ON_ONCE(vcpu->arch.cputm_start != 0);
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
vcpu->arch.cputm_start = get_tod_clock_fast();
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
}
/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __stop_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
WARN_ON_ONCE(vcpu->arch.cputm_start == 0);
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
vcpu->arch.sie_block->cputm -= get_tod_clock_fast() - vcpu->arch.cputm_start;
vcpu->arch.cputm_start = 0;
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
}
/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __enable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
WARN_ON_ONCE(vcpu->arch.cputm_enabled);
vcpu->arch.cputm_enabled = true;
__start_cpu_timer_accounting(vcpu);
}
/* needs disabled preemption to protect from TOD sync and vcpu_load/put */
static void __disable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
WARN_ON_ONCE(!vcpu->arch.cputm_enabled);
__stop_cpu_timer_accounting(vcpu);
vcpu->arch.cputm_enabled = false;
}
static void enable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
preempt_disable(); /* protect from TOD sync and vcpu_load/put */
__enable_cpu_timer_accounting(vcpu);
preempt_enable();
}
static void disable_cpu_timer_accounting(struct kvm_vcpu *vcpu)
{
preempt_disable(); /* protect from TOD sync and vcpu_load/put */
__disable_cpu_timer_accounting(vcpu);
preempt_enable();
}
/* set the cpu timer - may only be called from the VCPU thread itself */
void kvm_s390_set_cpu_timer(struct kvm_vcpu *vcpu, __u64 cputm)
{
preempt_disable(); /* protect from TOD sync and vcpu_load/put */
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_begin(&vcpu->arch.cputm_seqcount);
if (vcpu->arch.cputm_enabled)
vcpu->arch.cputm_start = get_tod_clock_fast();
vcpu->arch.sie_block->cputm = cputm;
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
raw_write_seqcount_end(&vcpu->arch.cputm_seqcount);
preempt_enable();
}
/* update and get the cpu timer - can also be called from other VCPU threads */
__u64 kvm_s390_get_cpu_timer(struct kvm_vcpu *vcpu)
{
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
unsigned int seq;
__u64 value;
if (unlikely(!vcpu->arch.cputm_enabled))
return vcpu->arch.sie_block->cputm;
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
preempt_disable(); /* protect from TOD sync and vcpu_load/put */
do {
seq = raw_read_seqcount(&vcpu->arch.cputm_seqcount);
/*
* If the writer would ever execute a read in the critical
* section, e.g. in irq context, we have a deadlock.
*/
WARN_ON_ONCE((seq & 1) && smp_processor_id() == vcpu->cpu);
value = vcpu->arch.sie_block->cputm;
/* if cputm_start is 0, accounting is being started/stopped */
if (likely(vcpu->arch.cputm_start))
value -= get_tod_clock_fast() - vcpu->arch.cputm_start;
} while (read_seqcount_retry(&vcpu->arch.cputm_seqcount, seq & ~1));
preempt_enable();
return value;
}
void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
{
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
/* Save host register state */
save_fpu_regs();
vcpu->arch.host_fpregs.fpc = current->thread.fpu.fpc;
vcpu->arch.host_fpregs.regs = current->thread.fpu.regs;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (MACHINE_HAS_VX)
current->thread.fpu.regs = vcpu->run->s.regs.vrs;
else
current->thread.fpu.regs = vcpu->run->s.regs.fprs;
current->thread.fpu.fpc = vcpu->run->s.regs.fpc;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (test_fp_ctl(current->thread.fpu.fpc))
/* User space provided an invalid FPC, let's clear it */
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
current->thread.fpu.fpc = 0;
save_access_regs(vcpu->arch.host_acrs);
restore_access_regs(vcpu->run->s.regs.acrs);
gmap_enable(vcpu->arch.enabled_gmap);
atomic_or(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
__start_cpu_timer_accounting(vcpu);
vcpu->cpu = cpu;
}
void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
{
vcpu->cpu = -1;
if (vcpu->arch.cputm_enabled && !is_vcpu_idle(vcpu))
__stop_cpu_timer_accounting(vcpu);
atomic_andnot(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags);
vcpu->arch.enabled_gmap = gmap_get_enabled();
gmap_disable(vcpu->arch.enabled_gmap);
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
/* Save guest register state */
save_fpu_regs();
vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
/* Restore host register state */
current->thread.fpu.fpc = vcpu->arch.host_fpregs.fpc;
current->thread.fpu.regs = vcpu->arch.host_fpregs.regs;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
save_access_regs(vcpu->run->s.regs.acrs);
restore_access_regs(vcpu->arch.host_acrs);
}
static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu)
{
/* this equals initial cpu reset in pop, but we don't switch to ESA */
vcpu->arch.sie_block->gpsw.mask = 0UL;
vcpu->arch.sie_block->gpsw.addr = 0UL;
kvm_s390_set_prefix(vcpu, 0);
kvm_s390_set_cpu_timer(vcpu, 0);
vcpu->arch.sie_block->ckc = 0UL;
vcpu->arch.sie_block->todpr = 0;
memset(vcpu->arch.sie_block->gcr, 0, 16 * sizeof(__u64));
vcpu->arch.sie_block->gcr[0] = 0xE0UL;
vcpu->arch.sie_block->gcr[14] = 0xC2000000UL;
/* make sure the new fpc will be lazily loaded */
save_fpu_regs();
current->thread.fpu.fpc = 0;
vcpu->arch.sie_block->gbea = 1;
vcpu->arch.sie_block->pp = 0;
vcpu->arch.pfault_token = KVM_S390_PFAULT_TOKEN_INVALID;
kvm_clear_async_pf_completion_queue(vcpu);
if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm))
kvm_s390_vcpu_stop(vcpu);
kvm_s390_clear_local_irqs(vcpu);
}
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
{
mutex_lock(&vcpu->kvm->lock);
preempt_disable();
vcpu->arch.sie_block->epoch = vcpu->kvm->arch.epoch;
preempt_enable();
mutex_unlock(&vcpu->kvm->lock);
if (!kvm_is_ucontrol(vcpu->kvm)) {
vcpu->arch.gmap = vcpu->kvm->arch.gmap;
sca_add_vcpu(vcpu);
}
if (test_kvm_facility(vcpu->kvm, 74) || vcpu->kvm->arch.user_instr0)
vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
/* make vcpu_load load the right gmap on the first trigger */
vcpu->arch.enabled_gmap = vcpu->arch.gmap;
}
static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu)
{
if (!test_kvm_facility(vcpu->kvm, 76))
return;
vcpu->arch.sie_block->ecb3 &= ~(ECB3_AES | ECB3_DEA);
if (vcpu->kvm->arch.crypto.aes_kw)
vcpu->arch.sie_block->ecb3 |= ECB3_AES;
if (vcpu->kvm->arch.crypto.dea_kw)
vcpu->arch.sie_block->ecb3 |= ECB3_DEA;
vcpu->arch.sie_block->crycbd = vcpu->kvm->arch.crypto.crycbd;
}
void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu)
{
free_page(vcpu->arch.sie_block->cbrlo);
vcpu->arch.sie_block->cbrlo = 0;
}
int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu)
{
vcpu->arch.sie_block->cbrlo = get_zeroed_page(GFP_KERNEL);
if (!vcpu->arch.sie_block->cbrlo)
return -ENOMEM;
vcpu->arch.sie_block->ecb2 |= 0x80;
vcpu->arch.sie_block->ecb2 &= ~0x08;
return 0;
}
static void kvm_s390_vcpu_setup_model(struct kvm_vcpu *vcpu)
{
struct kvm_s390_cpu_model *model = &vcpu->kvm->arch.model;
vcpu->arch.sie_block->ibc = model->ibc;
if (test_kvm_facility(vcpu->kvm, 7))
vcpu->arch.sie_block->fac = (u32)(u64) model->fac_list;
}
int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
{
int rc = 0;
atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH |
CPUSTAT_SM |
CPUSTAT_STOPPED);
if (test_kvm_facility(vcpu->kvm, 78))
atomic_or(CPUSTAT_GED2, &vcpu->arch.sie_block->cpuflags);
else if (test_kvm_facility(vcpu->kvm, 8))
atomic_or(CPUSTAT_GED, &vcpu->arch.sie_block->cpuflags);
kvm_s390_vcpu_setup_model(vcpu);
/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
if (MACHINE_HAS_ESOP)
vcpu->arch.sie_block->ecb |= 0x02;
if (test_kvm_facility(vcpu->kvm, 9))
vcpu->arch.sie_block->ecb |= 0x04;
if (test_kvm_facility(vcpu->kvm, 73))
vcpu->arch.sie_block->ecb |= 0x10;
if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi)
vcpu->arch.sie_block->ecb2 |= 0x08;
vcpu->arch.sie_block->eca = 0x1002000U;
if (sclp.has_cei)
vcpu->arch.sie_block->eca |= 0x80000000U;
if (sclp.has_ib)
vcpu->arch.sie_block->eca |= 0x40000000U;
if (sclp.has_siif)
vcpu->arch.sie_block->eca |= 1;
if (sclp.has_sigpif)
vcpu->arch.sie_block->eca |= 0x10000000U;
if (test_kvm_facility(vcpu->kvm, 129)) {
vcpu->arch.sie_block->eca |= 0x00020000;
vcpu->arch.sie_block->ecd |= 0x20000000;
}
vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
if (vcpu->kvm->arch.use_cmma) {
rc = kvm_s390_vcpu_setup_cmma(vcpu);
if (rc)
return rc;
}
hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
kvm_s390_vcpu_crypto_setup(vcpu);
return rc;
}
struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
unsigned int id)
{
struct kvm_vcpu *vcpu;
struct sie_page *sie_page;
int rc = -EINVAL;
if (!kvm_is_ucontrol(kvm) && !sca_can_add_vcpu(kvm, id))
goto out;
rc = -ENOMEM;
vcpu = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL);
if (!vcpu)
goto out;
sie_page = (struct sie_page *) get_zeroed_page(GFP_KERNEL);
if (!sie_page)
goto out_free_cpu;
vcpu->arch.sie_block = &sie_page->sie_block;
vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;
/* the real guest size will always be smaller than msl */
vcpu->arch.sie_block->mso = 0;
vcpu->arch.sie_block->msl = sclp.hamax;
vcpu->arch.sie_block->icpua = id;
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
spin_lock_init(&vcpu->arch.local_int.lock);
vcpu->arch.local_int.float_int = &kvm->arch.float_int;
vcpu->arch.local_int.wq = &vcpu->wq;
vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags;
KVM: s390: protect VCPU cpu timer with a seqcount For now, only the owning VCPU thread (that has loaded the VCPU) can get a consistent cpu timer value when calculating the delta. However, other threads might also be interested in a more recent, consistent value. Of special interest will be the timer callback of a VCPU that executes without having the VCPU loaded and could run in parallel with the VCPU thread. The cpu timer has a nice property: it is only updated by the owning VCPU thread. And speaking about accounting, a consistent value can only be calculated by looking at cputm_start and the cpu timer itself in one shot, otherwise the result might be wrong. As we only have one writing thread at a time (owning VCPU thread), we can use a seqcount instead of a seqlock and retry if the VCPU refreshed its cpu timer. This avoids any heavy locking and only introduces a counter update/check plus a handful of smp_wmb(). The owning VCPU thread should never have to retry on reads, and also for other threads this might be a very rare scenario. Please note that we have to use the raw_* variants for locking the seqcount as lockdep will produce false warnings otherwise. The rq->lock held during vcpu_load/put is also acquired from hardirq context. Lockdep cannot know that we avoid potential deadlocks by disabling preemption and thereby disable concurrent write locking attempts (via vcpu_put/load). Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
2016-02-18 04:53:33 +08:00
seqcount_init(&vcpu->arch.cputm_seqcount);
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
rc = kvm_vcpu_init(vcpu, kvm, id);
if (rc)
goto out_free_sie_block;
VM_EVENT(kvm, 3, "create cpu %d at 0x%pK, sie block at 0x%pK", id, vcpu,
vcpu->arch.sie_block);
trace_kvm_s390_create_vcpu(id, vcpu, vcpu->arch.sie_block);
return vcpu;
out_free_sie_block:
free_page((unsigned long)(vcpu->arch.sie_block));
out_free_cpu:
kmem_cache_free(kvm_vcpu_cache, vcpu);
out:
return ERR_PTR(rc);
}
int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
{
return kvm_s390_vcpu_has_irq(vcpu, 0);
}
void kvm_s390_vcpu_block(struct kvm_vcpu *vcpu)
{
atomic_or(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
exit_sie(vcpu);
}
void kvm_s390_vcpu_unblock(struct kvm_vcpu *vcpu)
{
atomic_andnot(PROG_BLOCK_SIE, &vcpu->arch.sie_block->prog20);
}
static void kvm_s390_vcpu_request(struct kvm_vcpu *vcpu)
{
atomic_or(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
exit_sie(vcpu);
}
static void kvm_s390_vcpu_request_handled(struct kvm_vcpu *vcpu)
{
atomic_andnot(PROG_REQUEST, &vcpu->arch.sie_block->prog20);
}
/*
* Kick a guest cpu out of SIE and wait until SIE is not running.
* If the CPU is not running (e.g. waiting as idle) the function will
* return immediately. */
void exit_sie(struct kvm_vcpu *vcpu)
{
atomic_or(CPUSTAT_STOP_INT, &vcpu->arch.sie_block->cpuflags);
while (vcpu->arch.sie_block->prog0c & PROG_IN_SIE)
cpu_relax();
}
/* Kick a guest cpu out of SIE to process a request synchronously */
void kvm_s390_sync_request(int req, struct kvm_vcpu *vcpu)
{
kvm_make_request(req, vcpu);
kvm_s390_vcpu_request(vcpu);
}
static void kvm_gmap_notifier(struct gmap *gmap, unsigned long start,
unsigned long end)
{
struct kvm *kvm = gmap->private;
struct kvm_vcpu *vcpu;
unsigned long prefix;
int i;
if (gmap_is_shadow(gmap))
return;
if (start >= 1UL << 31)
/* We are only interested in prefix pages */
return;
kvm_for_each_vcpu(i, vcpu, kvm) {
/* match against both prefix pages */
prefix = kvm_s390_get_prefix(vcpu);
if (prefix <= end && start <= prefix + 2*PAGE_SIZE - 1) {
VCPU_EVENT(vcpu, 2, "gmap notifier for %lx-%lx",
start, end);
kvm_s390_sync_request(KVM_REQ_MMU_RELOAD, vcpu);
}
}
}
int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu)
{
/* kvm common code refers to this, but never calls it */
BUG();
return 0;
}
static int kvm_arch_vcpu_ioctl_get_one_reg(struct kvm_vcpu *vcpu,
struct kvm_one_reg *reg)
{
int r = -EINVAL;
switch (reg->id) {
case KVM_REG_S390_TODPR:
r = put_user(vcpu->arch.sie_block->todpr,
(u32 __user *)reg->addr);
break;
case KVM_REG_S390_EPOCHDIFF:
r = put_user(vcpu->arch.sie_block->epoch,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_CPU_TIMER:
r = put_user(kvm_s390_get_cpu_timer(vcpu),
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_CLOCK_COMP:
r = put_user(vcpu->arch.sie_block->ckc,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PFTOKEN:
r = put_user(vcpu->arch.pfault_token,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PFCOMPARE:
r = put_user(vcpu->arch.pfault_compare,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PFSELECT:
r = put_user(vcpu->arch.pfault_select,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PP:
r = put_user(vcpu->arch.sie_block->pp,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_GBEA:
r = put_user(vcpu->arch.sie_block->gbea,
(u64 __user *)reg->addr);
break;
default:
break;
}
return r;
}
static int kvm_arch_vcpu_ioctl_set_one_reg(struct kvm_vcpu *vcpu,
struct kvm_one_reg *reg)
{
int r = -EINVAL;
__u64 val;
switch (reg->id) {
case KVM_REG_S390_TODPR:
r = get_user(vcpu->arch.sie_block->todpr,
(u32 __user *)reg->addr);
break;
case KVM_REG_S390_EPOCHDIFF:
r = get_user(vcpu->arch.sie_block->epoch,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_CPU_TIMER:
r = get_user(val, (u64 __user *)reg->addr);
if (!r)
kvm_s390_set_cpu_timer(vcpu, val);
break;
case KVM_REG_S390_CLOCK_COMP:
r = get_user(vcpu->arch.sie_block->ckc,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PFTOKEN:
r = get_user(vcpu->arch.pfault_token,
(u64 __user *)reg->addr);
if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
kvm_clear_async_pf_completion_queue(vcpu);
break;
case KVM_REG_S390_PFCOMPARE:
r = get_user(vcpu->arch.pfault_compare,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PFSELECT:
r = get_user(vcpu->arch.pfault_select,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_PP:
r = get_user(vcpu->arch.sie_block->pp,
(u64 __user *)reg->addr);
break;
case KVM_REG_S390_GBEA:
r = get_user(vcpu->arch.sie_block->gbea,
(u64 __user *)reg->addr);
break;
default:
break;
}
return r;
}
static int kvm_arch_vcpu_ioctl_initial_reset(struct kvm_vcpu *vcpu)
{
kvm_s390_vcpu_initial_reset(vcpu);
return 0;
}
int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
memcpy(&vcpu->run->s.regs.gprs, &regs->gprs, sizeof(regs->gprs));
return 0;
}
int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
{
memcpy(&regs->gprs, &vcpu->run->s.regs.gprs, sizeof(regs->gprs));
return 0;
}
int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
struct kvm_sregs *sregs)
{
memcpy(&vcpu->run->s.regs.acrs, &sregs->acrs, sizeof(sregs->acrs));
memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs));
restore_access_regs(vcpu->run->s.regs.acrs);
return 0;
}
int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu,
struct kvm_sregs *sregs)
{
memcpy(&sregs->acrs, &vcpu->run->s.regs.acrs, sizeof(sregs->acrs));
memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs));
return 0;
}
int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
/* make sure the new values will be lazily loaded */
save_fpu_regs();
if (test_fp_ctl(fpu->fpc))
return -EINVAL;
current->thread.fpu.fpc = fpu->fpc;
if (MACHINE_HAS_VX)
convert_fp_to_vx((__vector128 *) vcpu->run->s.regs.vrs,
(freg_t *) fpu->fprs);
else
memcpy(vcpu->run->s.regs.fprs, &fpu->fprs, sizeof(fpu->fprs));
return 0;
}
int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
{
/* make sure we have the latest values */
save_fpu_regs();
if (MACHINE_HAS_VX)
convert_vx_to_fp((freg_t *) fpu->fprs,
(__vector128 *) vcpu->run->s.regs.vrs);
else
memcpy(fpu->fprs, vcpu->run->s.regs.fprs, sizeof(fpu->fprs));
fpu->fpc = current->thread.fpu.fpc;
return 0;
}
static int kvm_arch_vcpu_ioctl_set_initial_psw(struct kvm_vcpu *vcpu, psw_t psw)
{
int rc = 0;
if (!is_vcpu_stopped(vcpu))
rc = -EBUSY;
else {
vcpu->run->psw_mask = psw.mask;
vcpu->run->psw_addr = psw.addr;
}
return rc;
}
int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
struct kvm_translation *tr)
{
return -EINVAL; /* not implemented yet */
}
#define VALID_GUESTDBG_FLAGS (KVM_GUESTDBG_SINGLESTEP | \
KVM_GUESTDBG_USE_HW_BP | \
KVM_GUESTDBG_ENABLE)
int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
struct kvm_guest_debug *dbg)
{
int rc = 0;
vcpu->guest_debug = 0;
kvm_s390_clear_bp_data(vcpu);
if (dbg->control & ~VALID_GUESTDBG_FLAGS)
return -EINVAL;
if (!sclp.has_gpere)
return -EINVAL;
if (dbg->control & KVM_GUESTDBG_ENABLE) {
vcpu->guest_debug = dbg->control;
/* enforce guest PER */
atomic_or(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
if (dbg->control & KVM_GUESTDBG_USE_HW_BP)
rc = kvm_s390_import_bp_data(vcpu, dbg);
} else {
atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
vcpu->arch.guestdbg.last_bp = 0;
}
if (rc) {
vcpu->guest_debug = 0;
kvm_s390_clear_bp_data(vcpu);
atomic_andnot(CPUSTAT_P, &vcpu->arch.sie_block->cpuflags);
}
return rc;
}
int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
struct kvm_mp_state *mp_state)
{
/* CHECK_STOP and LOAD are not supported yet */
return is_vcpu_stopped(vcpu) ? KVM_MP_STATE_STOPPED :
KVM_MP_STATE_OPERATING;
}
int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
struct kvm_mp_state *mp_state)
{
int rc = 0;
/* user space knows about this interface - let it control the state */
vcpu->kvm->arch.user_cpu_state_ctrl = 1;
switch (mp_state->mp_state) {
case KVM_MP_STATE_STOPPED:
kvm_s390_vcpu_stop(vcpu);
break;
case KVM_MP_STATE_OPERATING:
kvm_s390_vcpu_start(vcpu);
break;
case KVM_MP_STATE_LOAD:
case KVM_MP_STATE_CHECK_STOP:
/* fall through - CHECK_STOP and LOAD are not supported yet */
default:
rc = -ENXIO;
}
return rc;
}
static bool ibs_enabled(struct kvm_vcpu *vcpu)
{
return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_IBS;
}
static int kvm_s390_handle_requests(struct kvm_vcpu *vcpu)
{
retry:
kvm_s390_vcpu_request_handled(vcpu);
if (!vcpu->requests)
return 0;
/*
* We use MMU_RELOAD just to re-arm the ipte notifier for the
* guest prefix page. gmap_mprotect_notify will wait on the ptl lock.
* This ensures that the ipte instruction for this request has
* already finished. We might race against a second unmapper that
* wants to set the blocking bit. Lets just retry the request loop.
*/
if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) {
int rc;
rc = gmap_mprotect_notify(vcpu->arch.gmap,
kvm_s390_get_prefix(vcpu),
PAGE_SIZE * 2, PROT_WRITE);
if (rc) {
kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu);
return rc;
}
goto retry;
}
if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) {
vcpu->arch.sie_block->ihcpu = 0xffff;
goto retry;
}
if (kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu)) {
if (!ibs_enabled(vcpu)) {
trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 1);
atomic_or(CPUSTAT_IBS,
&vcpu->arch.sie_block->cpuflags);
}
goto retry;
}
if (kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu)) {
if (ibs_enabled(vcpu)) {
trace_kvm_s390_enable_disable_ibs(vcpu->vcpu_id, 0);
atomic_andnot(CPUSTAT_IBS,
&vcpu->arch.sie_block->cpuflags);
}
goto retry;
}
if (kvm_check_request(KVM_REQ_ICPT_OPEREXC, vcpu)) {
vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
goto retry;
}
/* nothing to do, just clear the request */
clear_bit(KVM_REQ_UNHALT, &vcpu->requests);
return 0;
}
void kvm_s390_set_tod_clock(struct kvm *kvm, u64 tod)
{
struct kvm_vcpu *vcpu;
int i;
mutex_lock(&kvm->lock);
preempt_disable();
kvm->arch.epoch = tod - get_tod_clock();
kvm_s390_vcpu_block_all(kvm);
kvm_for_each_vcpu(i, vcpu, kvm)
vcpu->arch.sie_block->epoch = kvm->arch.epoch;
kvm_s390_vcpu_unblock_all(kvm);
preempt_enable();
mutex_unlock(&kvm->lock);
}
/**
* kvm_arch_fault_in_page - fault-in guest page if necessary
* @vcpu: The corresponding virtual cpu
* @gpa: Guest physical address
* @writable: Whether the page should be writable or not
*
* Make sure that a guest page has been faulted-in on the host.
*
* Return: Zero on success, negative error code otherwise.
*/
long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable)
{
return gmap_fault(vcpu->arch.gmap, gpa,
writable ? FAULT_FLAG_WRITE : 0);
}
static void __kvm_inject_pfault_token(struct kvm_vcpu *vcpu, bool start_token,
unsigned long token)
{
struct kvm_s390_interrupt inti;
struct kvm_s390_irq irq;
if (start_token) {
irq.u.ext.ext_params2 = token;
irq.type = KVM_S390_INT_PFAULT_INIT;
WARN_ON_ONCE(kvm_s390_inject_vcpu(vcpu, &irq));
} else {
inti.type = KVM_S390_INT_PFAULT_DONE;
inti.parm64 = token;
WARN_ON_ONCE(kvm_s390_inject_vm(vcpu->kvm, &inti));
}
}
void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu,
struct kvm_async_pf *work)
{
trace_kvm_s390_pfault_init(vcpu, work->arch.pfault_token);
__kvm_inject_pfault_token(vcpu, true, work->arch.pfault_token);
}
void kvm_arch_async_page_present(struct kvm_vcpu *vcpu,
struct kvm_async_pf *work)
{
trace_kvm_s390_pfault_done(vcpu, work->arch.pfault_token);
__kvm_inject_pfault_token(vcpu, false, work->arch.pfault_token);
}
void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu,
struct kvm_async_pf *work)
{
/* s390 will always inject the page directly */
}
bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu)
{
/*
* s390 will always inject the page directly,
* but we still want check_async_completion to cleanup
*/
return true;
}
static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu)
{
hva_t hva;
struct kvm_arch_async_pf arch;
int rc;
if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
return 0;
if ((vcpu->arch.sie_block->gpsw.mask & vcpu->arch.pfault_select) !=
vcpu->arch.pfault_compare)
return 0;
if (psw_extint_disabled(vcpu))
return 0;
if (kvm_s390_vcpu_has_irq(vcpu, 0))
return 0;
if (!(vcpu->arch.sie_block->gcr[0] & 0x200ul))
return 0;
if (!vcpu->arch.gmap->pfault_enabled)
return 0;
hva = gfn_to_hva(vcpu->kvm, gpa_to_gfn(current->thread.gmap_addr));
hva += current->thread.gmap_addr & ~PAGE_MASK;
if (read_guest_real(vcpu, vcpu->arch.pfault_token, &arch.pfault_token, 8))
return 0;
rc = kvm_setup_async_pf(vcpu, current->thread.gmap_addr, hva, &arch);
return rc;
}
static int vcpu_pre_run(struct kvm_vcpu *vcpu)
{
int rc, cpuflags;
/*
* On s390 notifications for arriving pages will be delivered directly
* to the guest but the house keeping for completed pfaults is
* handled outside the worker.
*/
kvm_check_async_pf_completion(vcpu);
vcpu->arch.sie_block->gg14 = vcpu->run->s.regs.gprs[14];
vcpu->arch.sie_block->gg15 = vcpu->run->s.regs.gprs[15];
if (need_resched())
schedule();
if (test_cpu_flag(CIF_MCCK_PENDING))
s390_handle_mcck();
if (!kvm_is_ucontrol(vcpu->kvm)) {
rc = kvm_s390_deliver_pending_interrupts(vcpu);
if (rc)
return rc;
}
rc = kvm_s390_handle_requests(vcpu);
if (rc)
return rc;
if (guestdbg_enabled(vcpu)) {
kvm_s390_backup_guest_per_regs(vcpu);
kvm_s390_patch_guest_per_regs(vcpu);
}
vcpu->arch.sie_block->icptcode = 0;
cpuflags = atomic_read(&vcpu->arch.sie_block->cpuflags);
VCPU_EVENT(vcpu, 6, "entering sie flags %x", cpuflags);
trace_kvm_s390_sie_enter(vcpu, cpuflags);
return 0;
}
static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu)
{
struct kvm_s390_pgm_info pgm_info = {
.code = PGM_ADDRESSING,
};
u8 opcode, ilen;
int rc;
VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction");
trace_kvm_s390_sie_fault(vcpu);
/*
* We want to inject an addressing exception, which is defined as a
* suppressing or terminating exception. However, since we came here
* by a DAT access exception, the PSW still points to the faulting
* instruction since DAT exceptions are nullifying. So we've got
* to look up the current opcode to get the length of the instruction
* to be able to forward the PSW.
*/
rc = read_guest_instr(vcpu, &opcode, 1);
ilen = insn_length(opcode);
if (rc < 0) {
return rc;
} else if (rc) {
/* Instruction-Fetching Exceptions - we can't detect the ilen.
* Forward by arbitrary ilc, injection will take care of
* nullification if necessary.
*/
pgm_info = vcpu->arch.pgm;
ilen = 4;
}
pgm_info.flags = ilen | KVM_S390_PGM_FLAGS_ILC_VALID;
kvm_s390_forward_psw(vcpu, ilen);
return kvm_s390_inject_prog_irq(vcpu, &pgm_info);
}
static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
{
VCPU_EVENT(vcpu, 6, "exit sie icptcode %d",
vcpu->arch.sie_block->icptcode);
trace_kvm_s390_sie_exit(vcpu, vcpu->arch.sie_block->icptcode);
if (guestdbg_enabled(vcpu))
kvm_s390_restore_guest_per_regs(vcpu);
vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14;
vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15;
if (vcpu->arch.sie_block->icptcode > 0) {
int rc = kvm_handle_sie_intercept(vcpu);
if (rc != -EOPNOTSUPP)
return rc;
vcpu->run->exit_reason = KVM_EXIT_S390_SIEIC;
vcpu->run->s390_sieic.icptcode = vcpu->arch.sie_block->icptcode;
vcpu->run->s390_sieic.ipa = vcpu->arch.sie_block->ipa;
vcpu->run->s390_sieic.ipb = vcpu->arch.sie_block->ipb;
return -EREMOTE;
} else if (exit_reason != -EFAULT) {
vcpu->stat.exit_null++;
return 0;
} else if (kvm_is_ucontrol(vcpu->kvm)) {
vcpu->run->exit_reason = KVM_EXIT_S390_UCONTROL;
vcpu->run->s390_ucontrol.trans_exc_code =
current->thread.gmap_addr;
vcpu->run->s390_ucontrol.pgm_code = 0x10;
return -EREMOTE;
} else if (current->thread.gmap_pfault) {
trace_kvm_s390_major_guest_pfault(vcpu);
current->thread.gmap_pfault = 0;
if (kvm_arch_setup_async_pf(vcpu))
return 0;
return kvm_arch_fault_in_page(vcpu, current->thread.gmap_addr, 1);
}
return vcpu_post_run_fault_in_sie(vcpu);
}
static int __vcpu_run(struct kvm_vcpu *vcpu)
{
int rc, exit_reason;
/*
* We try to hold kvm->srcu during most of vcpu_run (except when run-
* ning the guest), so that memslots (and other stuff) are protected
*/
vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
do {
rc = vcpu_pre_run(vcpu);
if (rc)
break;
srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
/*
* As PF_VCPU will be used in fault handler, between
* guest_enter and guest_exit should be no uaccess.
*/
local_irq_disable();
guest_enter_irqoff();
__disable_cpu_timer_accounting(vcpu);
local_irq_enable();
exit_reason = sie64a(vcpu->arch.sie_block,
vcpu->run->s.regs.gprs);
local_irq_disable();
__enable_cpu_timer_accounting(vcpu);
guest_exit_irqoff();
local_irq_enable();
vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
rc = vcpu_post_run(vcpu, exit_reason);
} while (!signal_pending(current) && !guestdbg_exit_pending(vcpu) && !rc);
srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx);
return rc;
}
static void sync_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
vcpu->arch.sie_block->gpsw.mask = kvm_run->psw_mask;
vcpu->arch.sie_block->gpsw.addr = kvm_run->psw_addr;
if (kvm_run->kvm_dirty_regs & KVM_SYNC_PREFIX)
kvm_s390_set_prefix(vcpu, kvm_run->s.regs.prefix);
if (kvm_run->kvm_dirty_regs & KVM_SYNC_CRS) {
memcpy(&vcpu->arch.sie_block->gcr, &kvm_run->s.regs.crs, 128);
/* some control register changes require a tlb flush */
kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
}
if (kvm_run->kvm_dirty_regs & KVM_SYNC_ARCH0) {
kvm_s390_set_cpu_timer(vcpu, kvm_run->s.regs.cputm);
vcpu->arch.sie_block->ckc = kvm_run->s.regs.ckc;
vcpu->arch.sie_block->todpr = kvm_run->s.regs.todpr;
vcpu->arch.sie_block->pp = kvm_run->s.regs.pp;
vcpu->arch.sie_block->gbea = kvm_run->s.regs.gbea;
}
if (kvm_run->kvm_dirty_regs & KVM_SYNC_PFAULT) {
vcpu->arch.pfault_token = kvm_run->s.regs.pft;
vcpu->arch.pfault_select = kvm_run->s.regs.pfs;
vcpu->arch.pfault_compare = kvm_run->s.regs.pfc;
if (vcpu->arch.pfault_token == KVM_S390_PFAULT_TOKEN_INVALID)
kvm_clear_async_pf_completion_queue(vcpu);
}
/*
* If userspace sets the riccb (e.g. after migration) to a valid state,
* we should enable RI here instead of doing the lazy enablement.
*/
if ((kvm_run->kvm_dirty_regs & KVM_SYNC_RICCB) &&
test_kvm_facility(vcpu->kvm, 64)) {
struct runtime_instr_cb *riccb =
(struct runtime_instr_cb *) &kvm_run->s.regs.riccb;
if (riccb->valid)
vcpu->arch.sie_block->ecb3 |= 0x01;
}
kvm_run->kvm_dirty_regs = 0;
}
static void store_regs(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
kvm_run->psw_mask = vcpu->arch.sie_block->gpsw.mask;
kvm_run->psw_addr = vcpu->arch.sie_block->gpsw.addr;
kvm_run->s.regs.prefix = kvm_s390_get_prefix(vcpu);
memcpy(&kvm_run->s.regs.crs, &vcpu->arch.sie_block->gcr, 128);
kvm_run->s.regs.cputm = kvm_s390_get_cpu_timer(vcpu);
kvm_run->s.regs.ckc = vcpu->arch.sie_block->ckc;
kvm_run->s.regs.todpr = vcpu->arch.sie_block->todpr;
kvm_run->s.regs.pp = vcpu->arch.sie_block->pp;
kvm_run->s.regs.gbea = vcpu->arch.sie_block->gbea;
kvm_run->s.regs.pft = vcpu->arch.pfault_token;
kvm_run->s.regs.pfs = vcpu->arch.pfault_select;
kvm_run->s.regs.pfc = vcpu->arch.pfault_compare;
}
int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
{
int rc;
sigset_t sigsaved;
if (guestdbg_exit_pending(vcpu)) {
kvm_s390_prepare_debug_exit(vcpu);
return 0;
}
if (vcpu->sigset_active)
sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
if (!kvm_s390_user_cpu_state_ctrl(vcpu->kvm)) {
kvm_s390_vcpu_start(vcpu);
} else if (is_vcpu_stopped(vcpu)) {
pr_err_ratelimited("can't run stopped vcpu %d\n",
vcpu->vcpu_id);
return -EINVAL;
}
sync_regs(vcpu, kvm_run);
enable_cpu_timer_accounting(vcpu);
might_fault();
rc = __vcpu_run(vcpu);
if (signal_pending(current) && !rc) {
kvm_run->exit_reason = KVM_EXIT_INTR;
rc = -EINTR;
}
if (guestdbg_exit_pending(vcpu) && !rc) {
kvm_s390_prepare_debug_exit(vcpu);
rc = 0;
}
if (rc == -EREMOTE) {
/* userspace support is needed, kvm_run has been prepared */
rc = 0;
}
disable_cpu_timer_accounting(vcpu);
store_regs(vcpu, kvm_run);
if (vcpu->sigset_active)
sigprocmask(SIG_SETMASK, &sigsaved, NULL);
vcpu->stat.exit_userspace++;
return rc;
}
/*
* store status at address
* we use have two special cases:
* KVM_S390_STORE_STATUS_NOADDR: -> 0x1200 on 64 bit
* KVM_S390_STORE_STATUS_PREFIXED: -> prefix
*/
int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
{
unsigned char archmode = 1;
freg_t fprs[NUM_FPRS];
unsigned int px;
u64 clkcomp, cputm;
int rc;
px = kvm_s390_get_prefix(vcpu);
if (gpa == KVM_S390_STORE_STATUS_NOADDR) {
if (write_guest_abs(vcpu, 163, &archmode, 1))
return -EFAULT;
gpa = 0;
} else if (gpa == KVM_S390_STORE_STATUS_PREFIXED) {
if (write_guest_real(vcpu, 163, &archmode, 1))
return -EFAULT;
gpa = px;
} else
gpa -= __LC_FPREGS_SAVE_AREA;
/* manually convert vector registers if necessary */
if (MACHINE_HAS_VX) {
convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
fprs, 128);
} else {
rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
vcpu->run->s.regs.fprs, 128);
}
rc |= write_guest_abs(vcpu, gpa + __LC_GPREGS_SAVE_AREA,
vcpu->run->s.regs.gprs, 128);
rc |= write_guest_abs(vcpu, gpa + __LC_PSW_SAVE_AREA,
&vcpu->arch.sie_block->gpsw, 16);
rc |= write_guest_abs(vcpu, gpa + __LC_PREFIX_SAVE_AREA,
&px, 4);
rc |= write_guest_abs(vcpu, gpa + __LC_FP_CREG_SAVE_AREA,
&vcpu->run->s.regs.fpc, 4);
rc |= write_guest_abs(vcpu, gpa + __LC_TOD_PROGREG_SAVE_AREA,
&vcpu->arch.sie_block->todpr, 4);
cputm = kvm_s390_get_cpu_timer(vcpu);
rc |= write_guest_abs(vcpu, gpa + __LC_CPU_TIMER_SAVE_AREA,
&cputm, 8);
clkcomp = vcpu->arch.sie_block->ckc >> 8;
rc |= write_guest_abs(vcpu, gpa + __LC_CLOCK_COMP_SAVE_AREA,
&clkcomp, 8);
rc |= write_guest_abs(vcpu, gpa + __LC_AREGS_SAVE_AREA,
&vcpu->run->s.regs.acrs, 64);
rc |= write_guest_abs(vcpu, gpa + __LC_CREGS_SAVE_AREA,
&vcpu->arch.sie_block->gcr, 128);
return rc ? -EFAULT : 0;
}
int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
{
/*
* The guest FPRS and ACRS are in the host FPRS/ACRS due to the lazy
* copying in vcpu load/put. Lets update our copies before we save
* it into the save area
*/
save_fpu_regs();
vcpu->run->s.regs.fpc = current->thread.fpu.fpc;
save_access_regs(vcpu->run->s.regs.acrs);
return kvm_s390_store_status_unloaded(vcpu, addr);
}
static void __disable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
kvm_check_request(KVM_REQ_ENABLE_IBS, vcpu);
kvm_s390_sync_request(KVM_REQ_DISABLE_IBS, vcpu);
}
static void __disable_ibs_on_all_vcpus(struct kvm *kvm)
{
unsigned int i;
struct kvm_vcpu *vcpu;
kvm_for_each_vcpu(i, vcpu, kvm) {
__disable_ibs_on_vcpu(vcpu);
}
}
static void __enable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
{
if (!sclp.has_ibs)
return;
kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
}
void kvm_s390_vcpu_start(struct kvm_vcpu *vcpu)
{
int i, online_vcpus, started_vcpus = 0;
if (!is_vcpu_stopped(vcpu))
return;
trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 1);
/* Only one cpu at a time may enter/leave the STOPPED state. */
spin_lock(&vcpu->kvm->arch.start_stop_lock);
online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);
for (i = 0; i < online_vcpus; i++) {
if (!is_vcpu_stopped(vcpu->kvm->vcpus[i]))
started_vcpus++;
}
if (started_vcpus == 0) {
/* we're the only active VCPU -> speed it up */
__enable_ibs_on_vcpu(vcpu);
} else if (started_vcpus == 1) {
/*
* As we are starting a second VCPU, we have to disable
* the IBS facility on all VCPUs to remove potentially
* oustanding ENABLE requests.
*/
__disable_ibs_on_all_vcpus(vcpu->kvm);
}
atomic_andnot(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
/*
* Another VCPU might have used IBS while we were offline.
* Let's play safe and flush the VCPU at startup.
*/
kvm_make_request(KVM_REQ_TLB_FLUSH, vcpu);
spin_unlock(&vcpu->kvm->arch.start_stop_lock);
return;
}
void kvm_s390_vcpu_stop(struct kvm_vcpu *vcpu)
{
int i, online_vcpus, started_vcpus = 0;
struct kvm_vcpu *started_vcpu = NULL;
if (is_vcpu_stopped(vcpu))
return;
trace_kvm_s390_vcpu_start_stop(vcpu->vcpu_id, 0);
/* Only one cpu at a time may enter/leave the STOPPED state. */
spin_lock(&vcpu->kvm->arch.start_stop_lock);
online_vcpus = atomic_read(&vcpu->kvm->online_vcpus);
/* SIGP STOP and SIGP STOP AND STORE STATUS has been fully processed */
kvm_s390_clear_stop_irq(vcpu);
atomic_or(CPUSTAT_STOPPED, &vcpu->arch.sie_block->cpuflags);
__disable_ibs_on_vcpu(vcpu);
for (i = 0; i < online_vcpus; i++) {
if (!is_vcpu_stopped(vcpu->kvm->vcpus[i])) {
started_vcpus++;
started_vcpu = vcpu->kvm->vcpus[i];
}
}
if (started_vcpus == 1) {
/*
* As we only have one VCPU left, we want to enable the
* IBS facility for that VCPU to speed it up.
*/
__enable_ibs_on_vcpu(started_vcpu);
}
spin_unlock(&vcpu->kvm->arch.start_stop_lock);
return;
}
static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
struct kvm_enable_cap *cap)
{
int r;
if (cap->flags)
return -EINVAL;
switch (cap->cap) {
case KVM_CAP_S390_CSS_SUPPORT:
if (!vcpu->kvm->arch.css_support) {
vcpu->kvm->arch.css_support = 1;
VM_EVENT(vcpu->kvm, 3, "%s", "ENABLE: CSS support");
trace_kvm_s390_enable_css(vcpu->kvm);
}
r = 0;
break;
default:
r = -EINVAL;
break;
}
return r;
}
static long kvm_s390_guest_mem_op(struct kvm_vcpu *vcpu,
struct kvm_s390_mem_op *mop)
{
void __user *uaddr = (void __user *)mop->buf;
void *tmpbuf = NULL;
int r, srcu_idx;
const u64 supported_flags = KVM_S390_MEMOP_F_INJECT_EXCEPTION
| KVM_S390_MEMOP_F_CHECK_ONLY;
if (mop->flags & ~supported_flags)
return -EINVAL;
if (mop->size > MEM_OP_MAX_SIZE)
return -E2BIG;
if (!(mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY)) {
tmpbuf = vmalloc(mop->size);
if (!tmpbuf)
return -ENOMEM;
}
srcu_idx = srcu_read_lock(&vcpu->kvm->srcu);
switch (mop->op) {
case KVM_S390_MEMOP_LOGICAL_READ:
if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) {
r = check_gva_range(vcpu, mop->gaddr, mop->ar,
mop->size, GACC_FETCH);
break;
}
r = read_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size);
if (r == 0) {
if (copy_to_user(uaddr, tmpbuf, mop->size))
r = -EFAULT;
}
break;
case KVM_S390_MEMOP_LOGICAL_WRITE:
if (mop->flags & KVM_S390_MEMOP_F_CHECK_ONLY) {
r = check_gva_range(vcpu, mop->gaddr, mop->ar,
mop->size, GACC_STORE);
break;
}
if (copy_from_user(tmpbuf, uaddr, mop->size)) {
r = -EFAULT;
break;
}
r = write_guest(vcpu, mop->gaddr, mop->ar, tmpbuf, mop->size);
break;
default:
r = -EINVAL;
}
srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx);
if (r > 0 && (mop->flags & KVM_S390_MEMOP_F_INJECT_EXCEPTION) != 0)
kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm);
vfree(tmpbuf);
return r;
}
long kvm_arch_vcpu_ioctl(struct file *filp,
unsigned int ioctl, unsigned long arg)
{
struct kvm_vcpu *vcpu = filp->private_data;
void __user *argp = (void __user *)arg;
int idx;
long r;
switch (ioctl) {
case KVM_S390_IRQ: {
struct kvm_s390_irq s390irq;
r = -EFAULT;
if (copy_from_user(&s390irq, argp, sizeof(s390irq)))
break;
r = kvm_s390_inject_vcpu(vcpu, &s390irq);
break;
}
case KVM_S390_INTERRUPT: {
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
struct kvm_s390_interrupt s390int;
struct kvm_s390_irq s390irq;
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
r = -EFAULT;
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
if (copy_from_user(&s390int, argp, sizeof(s390int)))
break;
if (s390int_to_s390irq(&s390int, &s390irq))
return -EINVAL;
r = kvm_s390_inject_vcpu(vcpu, &s390irq);
break;
KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Carsten Otte <cotte@de.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Avi Kivity <avi@qumranet.com>
2008-03-26 01:47:26 +08:00
}
case KVM_S390_STORE_STATUS:
idx = srcu_read_lock(&vcpu->kvm->srcu);
r = kvm_s390_vcpu_store_status(vcpu, arg);
srcu_read_unlock(&vcpu->kvm->srcu, idx);
break;
case KVM_S390_SET_INITIAL_PSW: {
psw_t psw;
r = -EFAULT;
if (copy_from_user(&psw, argp, sizeof(psw)))
break;
r = kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw);
break;
}
case KVM_S390_INITIAL_RESET:
r = kvm_arch_vcpu_ioctl_initial_reset(vcpu);
break;
case KVM_SET_ONE_REG:
case KVM_GET_ONE_REG: {
struct kvm_one_reg reg;
r = -EFAULT;
if (copy_from_user(&reg, argp, sizeof(reg)))
break;
if (ioctl == KVM_SET_ONE_REG)
r = kvm_arch_vcpu_ioctl_set_one_reg(vcpu, &reg);
else
r = kvm_arch_vcpu_ioctl_get_one_reg(vcpu, &reg);
break;
}
#ifdef CONFIG_KVM_S390_UCONTROL
case KVM_S390_UCAS_MAP: {
struct kvm_s390_ucas_mapping ucasmap;
if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) {
r = -EFAULT;
break;
}
if (!kvm_is_ucontrol(vcpu->kvm)) {
r = -EINVAL;
break;
}
r = gmap_map_segment(vcpu->arch.gmap, ucasmap.user_addr,
ucasmap.vcpu_addr, ucasmap.length);
break;
}
case KVM_S390_UCAS_UNMAP: {
struct kvm_s390_ucas_mapping ucasmap;
if (copy_from_user(&ucasmap, argp, sizeof(ucasmap))) {
r = -EFAULT;
break;
}
if (!kvm_is_ucontrol(vcpu->kvm)) {
r = -EINVAL;
break;
}
r = gmap_unmap_segment(vcpu->arch.gmap, ucasmap.vcpu_addr,
ucasmap.length);
break;
}
#endif
case KVM_S390_VCPU_FAULT: {
r = gmap_fault(vcpu->arch.gmap, arg, 0);
break;
}
case KVM_ENABLE_CAP:
{
struct kvm_enable_cap cap;
r = -EFAULT;
if (copy_from_user(&cap, argp, sizeof(cap)))
break;
r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
break;
}
case KVM_S390_MEM_OP: {
struct kvm_s390_mem_op mem_op;
if (copy_from_user(&mem_op, argp, sizeof(mem_op)) == 0)
r = kvm_s390_guest_mem_op(vcpu, &mem_op);
else
r = -EFAULT;
break;
}
case KVM_S390_SET_IRQ_STATE: {
struct kvm_s390_irq_state irq_state;
r = -EFAULT;
if (copy_from_user(&irq_state, argp, sizeof(irq_state)))
break;
if (irq_state.len > VCPU_IRQS_MAX_BUF ||
irq_state.len == 0 ||
irq_state.len % sizeof(struct kvm_s390_irq) > 0) {
r = -EINVAL;
break;
}
r = kvm_s390_set_irq_state(vcpu,
(void __user *) irq_state.buf,
irq_state.len);
break;
}
case KVM_S390_GET_IRQ_STATE: {
struct kvm_s390_irq_state irq_state;
r = -EFAULT;
if (copy_from_user(&irq_state, argp, sizeof(irq_state)))
break;
if (irq_state.len == 0) {
r = -EINVAL;
break;
}
r = kvm_s390_get_irq_state(vcpu,
(__u8 __user *) irq_state.buf,
irq_state.len);
break;
}
default:
r = -ENOTTY;
}
return r;
}
int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
{
#ifdef CONFIG_KVM_S390_UCONTROL
if ((vmf->pgoff == KVM_S390_SIE_PAGE_OFFSET)
&& (kvm_is_ucontrol(vcpu->kvm))) {
vmf->page = virt_to_page(vcpu->arch.sie_block);
get_page(vmf->page);
return 0;
}
#endif
return VM_FAULT_SIGBUS;
}
int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
unsigned long npages)
{
return 0;
}
/* Section: memory related */
int kvm_arch_prepare_memory_region(struct kvm *kvm,
struct kvm_memory_slot *memslot,
const struct kvm_userspace_memory_region *mem,
enum kvm_mr_change change)
{
/* A few sanity checks. We can have memory slots which have to be
located/ended at a segment boundary (1MB). The memory in userland is
ok to be fragmented into various different vmas. It is okay to mmap()
and munmap() stuff in this slot after doing this call at any time */
if (mem->userspace_addr & 0xffffful)
return -EINVAL;
if (mem->memory_size & 0xffffful)
return -EINVAL;
if (mem->guest_phys_addr + mem->memory_size > kvm->arch.mem_limit)
return -EINVAL;
return 0;
}
void kvm_arch_commit_memory_region(struct kvm *kvm,
const struct kvm_userspace_memory_region *mem,
const struct kvm_memory_slot *old,
const struct kvm_memory_slot *new,
enum kvm_mr_change change)
{
int rc;
/* If the basics of the memslot do not change, we do not want
* to update the gmap. Every update causes several unnecessary
* segment translation exceptions. This is usually handled just
* fine by the normal fault handler + gmap, but it will also
* cause faults on the prefix page of running guest CPUs.
*/
if (old->userspace_addr == mem->userspace_addr &&
old->base_gfn * PAGE_SIZE == mem->guest_phys_addr &&
old->npages * PAGE_SIZE == mem->memory_size)
return;
rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
mem->guest_phys_addr, mem->memory_size);
if (rc)
pr_warn("failed to commit memory region\n");
return;
}
static inline unsigned long nonhyp_mask(int i)
{
unsigned int nonhyp_fai = (sclp.hmfai << i * 2) >> 30;
return 0x0000ffffffffffffUL >> (nonhyp_fai << 4);
}
KVM: halt_polling: provide a way to qualify wakeups during poll Some wakeups should not be considered a sucessful poll. For example on s390 I/O interrupts are usually floating, which means that _ALL_ CPUs would be considered runnable - letting all vCPUs poll all the time for transactional like workload, even if one vCPU would be enough. This can result in huge CPU usage for large guests. This patch lets architectures provide a way to qualify wakeups if they should be considered a good/bad wakeups in regard to polls. For s390 the implementation will fence of halt polling for anything but known good, single vCPU events. The s390 implementation for floating interrupts does a wakeup for one vCPU, but the interrupt will be delivered by whatever CPU checks first for a pending interrupt. We prefer the woken up CPU by marking the poll of this CPU as "good" poll. This code will also mark several other wakeup reasons like IPI or expired timers as "good". This will of course also mark some events as not sucessful. As KVM on z runs always as a 2nd level hypervisor, we prefer to not poll, unless we are really sure, though. This patch successfully limits the CPU usage for cases like uperf 1byte transactional ping pong workload or wakeup heavy workload like OLTP while still providing a proper speedup. This also introduced a new vcpu stat "halt_poll_no_tuning" that marks wakeups that are considered not good for polling. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Radim Krčmář <rkrcmar@redhat.com> (for an earlier version) Cc: David Matlack <dmatlack@google.com> Cc: Wanpeng Li <kernellwp@gmail.com> [Rename config symbol. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2016-05-13 18:16:35 +08:00
void kvm_arch_vcpu_block_finish(struct kvm_vcpu *vcpu)
{
vcpu->valid_wakeup = false;
}
static int __init kvm_s390_init(void)
{
int i;
if (!sclp.has_sief2) {
pr_info("SIE not available\n");
return -ENODEV;
}
for (i = 0; i < 16; i++)
kvm_s390_fac_list_mask[i] |=
S390_lowcore.stfle_fac_list[i] & nonhyp_mask(i);
return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE);
}
static void __exit kvm_s390_exit(void)
{
kvm_exit();
}
module_init(kvm_s390_init);
module_exit(kvm_s390_exit);
/*
* Enable autoloading of the kvm module.
* Note that we add the module alias here instead of virt/kvm/kvm_main.c
* since x86 takes a different approach.
*/
#include <linux/miscdevice.h>
MODULE_ALIAS_MISCDEV(KVM_MINOR);
MODULE_ALIAS("devname:kvm");