openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity
linux/arch/x86/xen/enlighten.c

415 lines
11 KiB
C
Raw Normal View History

xen/balloon: Mark unallocated host memory as UNUSABLE Commit f5775e0b6116 ("x86/xen: discard RAM regions above the maximum reservation") left host memory not assigned to dom0 as available for memory hotplug. Unfortunately this also meant that those regions could be used by others. Specifically, commit fa564ad96366 ("x86/PCI: Enable a 64bit BAR on AMD Family 15h (Models 00-1f, 30-3f, 60-7f)") may try to map those addresses as MMIO. To prevent this mark unallocated host memory as E820_TYPE_UNUSABLE (thus effectively reverting f5775e0b6116) and keep track of that region as a hostmem resource that can be used for the hotplug. Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reviewed-by: Juergen Gross <jgross@suse.com>
2017-12-13 04:08:21 +08:00
#ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
#include <linux/bootmem.h>
#endif
x86/xen: event channels delivery on HVM. Set the callback to receive evtchns from Xen, using the callback vector delivery mechanism. The traditional way for receiving event channel notifications from Xen is via the interrupts from the platform PCI device. The callback vector is a newer alternative that allow us to receive notifications on any vcpu and doesn't need any PCI support: we allocate a vector exclusively to receive events, in the vector handler we don't need to interact with the vlapic, therefore we avoid a VMEXIT. Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Sheng Yang <sheng@linux.intel.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2010-05-14 19:40:51 +08:00
#include <linux/cpu.h>
x86/xen: Support kexec/kdump in HVM guests by doing a soft reset Currently there is a number of issues preventing PVHVM Xen guests from doing successful kexec/kdump: - Bound event channels. - Registered vcpu_info. - PIRQ/emuirq mappings. - shared_info frame after XENMAPSPACE_shared_info operation. - Active grant mappings. Basically, newly booted kernel stumbles upon already set up Xen interfaces and there is no way to reestablish them. In Xen-4.7 a new feature called 'soft reset' is coming. A guest performing kexec/kdump operation is supposed to call SCHEDOP_shutdown hypercall with SHUTDOWN_soft_reset reason before jumping to new kernel. Hypervisor (with some help from toolstack) will do full domain cleanup (but keeping its memory and vCPU contexts intact) returning the guest to the state it had when it was first booted and thus allowing it to start over. Doing SHUTDOWN_soft_reset on Xen hypervisors which don't support it is probably OK as by default all unknown shutdown reasons cause domain destroy with a message in toolstack log: 'Unknown shutdown reason code 5. Destroying domain.' which gives a clue to what the problem is and eliminates false expectations. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: David Vrabel <david.vrabel@citrix.com>
2015-09-25 17:59:52 +08:00
#include <linux/kexec.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
#include <xen/features.h>
#include <xen/page.h>
xen/balloon: Mark unallocated host memory as UNUSABLE Commit f5775e0b6116 ("x86/xen: discard RAM regions above the maximum reservation") left host memory not assigned to dom0 as available for memory hotplug. Unfortunately this also meant that those regions could be used by others. Specifically, commit fa564ad96366 ("x86/PCI: Enable a 64bit BAR on AMD Family 15h (Models 00-1f, 30-3f, 60-7f)") may try to map those addresses as MMIO. To prevent this mark unallocated host memory as E820_TYPE_UNUSABLE (thus effectively reverting f5775e0b6116) and keep track of that region as a hostmem resource that can be used for the hotplug. Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reviewed-by: Juergen Gross <jgross@suse.com>
2017-12-13 04:08:21 +08:00
#include <xen/interface/memory.h>
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
#include <asm/xen/hypercall.h>
#include <asm/xen/hypervisor.h>
xen/arm: Enable cpu_hotplug.c Build cpu_hotplug for ARM and ARM64 guests. Rename arch_(un)register_cpu to xen_(un)register_cpu and provide an empty implementation on ARM and ARM64. On x86 just call arch_(un)register_cpu as we are already doing. Initialize cpu_hotplug on ARM. Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Reviewed-by: Julien Grall <julien.grall@citrix.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2015-10-23 00:20:46 +08:00
#include <asm/cpu.h>
x86/xen: Update e820 table handling to the new core x86 E820 code Note that I restructured the Xen E820 logic a bit: instead of trying to sort the boot parameters, only the kernel's E820 table is sorted. This is how the x86 code does it and it reduces coupling between the in-kernel E820 code and the (unchanged) boot parameters. Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Juergen Gross <jgross@suse.com> Cc: <stefano.stabellini@eu.citrix.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-03-01 15:49:47 +08:00
#include <asm/e820/api.h>
xen/enlighten: Expose MWAIT and MWAIT_LEAF if hypervisor OKs it. For the hypervisor to take advantage of the MWAIT support it needs to extract from the ACPI _CST the register address. But the hypervisor does not have the support to parse DSDT so it relies on the initial domain (dom0) to parse the ACPI Power Management information and push it up to the hypervisor. The pushing of the data is done by the processor_harveset_xen module which parses the information that the ACPI parser has graciously exposed in 'struct acpi_processor'. For the ACPI parser to also expose the Cx states for MWAIT, we need to expose the MWAIT capability (leaf 1). Furthermore we also need to expose the MWAIT_LEAF capability (leaf 5) for cstate.c to properly function. The hypervisor could expose these flags when it traps the XEN_EMULATE_PREFIX operations, but it can't do it since it needs to be backwards compatible. Instead we choose to use the native CPUID to figure out if the MWAIT capability exists and use the XEN_SET_PDC query hypercall to figure out if the hypervisor wants us to expose the MWAIT_LEAF capability or not. Note: The XEN_SET_PDC query was implemented in c/s 23783: "ACPI: add _PDC input override mechanism". With this in place, instead of C3 ACPI IOPORT 415 we get now C3:ACPI FFH INTEL MWAIT 0x20 Note: The cpu_idle which would be calling the mwait variants for idling never gets set b/c we set the default pm_idle to be the hypercall variant. Acked-by: Jan Beulich <JBeulich@suse.com> [v2: Fix missing header file include and #ifdef] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-02-14 11:26:32 +08:00
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
#include "xen-ops.h"
xen: implement apic ipi interface Map native ipi vector to xen vector. Implement apic ipi interface with xen_send_IPI_one. Tested-by: Steven Noonan <steven@uplinklabs.net> Signed-off-by: Ben Guthro <ben@guthro.net> Signed-off-by: Lin Ming <mlin@ss.pku.edu.cn> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-04-21 00:11:04 +08:00
#include "smp.h"
xen/PMU: Initialization code for Xen PMU Map shared data structure that will hold CPU registers, VPMU context, V/PCPU IDs of the CPU interrupted by PMU interrupt. Hypervisor fills this information in its handler and passes it to the guest for further processing. Set up PMU VIRQ. Now that perf infrastructure will assume that PMU is available on a PV guest we need to be careful and make sure that accesses via RDPMC instruction don't cause fatal traps by the hypervisor. Provide a nop RDPMC handler. For the same reason avoid issuing a warning on a write to APIC's LVTPC. Both of these will be made functional in later patches. Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reviewed-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: David Vrabel <david.vrabel@citrix.com>
2015-08-11 04:34:34 +08:00
#include "pmu.h"
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
EXPORT_SYMBOL_GPL(hypercall_page);
xen/vcpu: Document the xen_vcpu_info and xen_vcpu They are important structures and it is not clear at first look what they are for. The xen_vcpu is a pointer. By default it points to the shared_info structure (at the CPU offset location). However if the VCPUOP_register_vcpu_info hypercall is implemented we can make the xen_vcpu pointer point to a per-CPU location. Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> [v1: Added comments from Ian Campbell] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2013-05-05 20:51:47 +08:00
/*
* Pointer to the xen_vcpu_info structure or
* &HYPERVISOR_shared_info->vcpu_info[cpu]. See xen_hvm_init_shared_info
* and xen_vcpu_setup for details. By default it points to share_info->vcpu_info
* but if the hypervisor supports VCPUOP_register_vcpu_info then it can point
* to xen_vcpu_info. The pointer is used in __xen_evtchn_do_upcall to
* acknowledge pending events.
* Also more subtly it is used by the patched version of irq enable/disable
* e.g. xen_irq_enable_direct and xen_iret in PV mode.
*
* The desire to be able to do those mask/unmask operations as a single
* instruction by using the per-cpu offset held in %gs is the real reason
* vcpu info is in a per-cpu pointer and the original reason for this
* hypercall.
*
*/
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
xen/vcpu: Document the xen_vcpu_info and xen_vcpu They are important structures and it is not clear at first look what they are for. The xen_vcpu is a pointer. By default it points to the shared_info structure (at the CPU offset location). However if the VCPUOP_register_vcpu_info hypercall is implemented we can make the xen_vcpu pointer point to a per-CPU location. Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> [v1: Added comments from Ian Campbell] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2013-05-05 20:51:47 +08:00
/*
* Per CPU pages used if hypervisor supports VCPUOP_register_vcpu_info
* hypercall. This can be used both in PV and PVHVM mode. The structure
* overrides the default per_cpu(xen_vcpu, cpu) value.
*/
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
xen: deal with stale cr3 values when unpinning pagetables When a pagetable is no longer in use, it must be unpinned so that its pages can be freed. However, this is only possible if there are no stray uses of the pagetable. The code currently deals with all the usual cases, but there's a rare case where a vcpu is changing cr3, but is doing so lazily, and the change hasn't actually happened by the time the pagetable is unpinned, even though it appears to have been completed. This change adds a second per-cpu cr3 variable - xen_current_cr3 - which tracks the actual state of the vcpu cr3. It is only updated once the actual hypercall to set cr3 has been completed. Other processors wishing to unpin a pagetable can check other vcpu's xen_current_cr3 values to see if any cross-cpu IPIs are needed to clean things up. [ Stable folks: 2.6.23 bugfix ] Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Stable Kernel <stable@kernel.org>
2007-10-17 02:51:30 +08:00
xen: introduce xen_vcpu_id mapping It may happen that Xen's and Linux's ideas of vCPU id diverge. In particular, when we crash on a secondary vCPU we may want to do kdump and unlike plain kexec where we do migrate_to_reboot_cpu() we try booting on the vCPU which crashed. This doesn't work very well for PVHVM guests as we have a number of hypercalls where we pass vCPU id as a parameter. These hypercalls either fail or do something unexpected. To solve the issue introduce percpu xen_vcpu_id mapping. ARM and PV guests get direct mapping for now. Boot CPU for PVHVM guest gets its id from CPUID. With secondary CPUs it is a bit more trickier. Currently, we initialize IPI vectors before these CPUs boot so we can't use CPUID. Use ACPI ids from MADT instead. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: David Vrabel <david.vrabel@citrix.com>
2016-06-30 23:56:37 +08:00
/* Linux <-> Xen vCPU id mapping */
xen: change the type of xen_vcpu_id to uint32_t We pass xen_vcpu_id mapping information to hypercalls which require uint32_t type so it would be cleaner to have it as uint32_t. The initializer to -1 can be dropped as we always do the mapping before using it and we never check the 'not set' value anyway. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: David Vrabel <david.vrabel@citrix.com>
2016-07-29 17:06:48 +08:00
DEFINE_PER_CPU(uint32_t, xen_vcpu_id);
xen: introduce xen_vcpu_id mapping It may happen that Xen's and Linux's ideas of vCPU id diverge. In particular, when we crash on a secondary vCPU we may want to do kdump and unlike plain kexec where we do migrate_to_reboot_cpu() we try booting on the vCPU which crashed. This doesn't work very well for PVHVM guests as we have a number of hypercalls where we pass vCPU id as a parameter. These hypercalls either fail or do something unexpected. To solve the issue introduce percpu xen_vcpu_id mapping. ARM and PV guests get direct mapping for now. Boot CPU for PVHVM guest gets its id from CPUID. With secondary CPUs it is a bit more trickier. Currently, we initialize IPI vectors before these CPUs boot so we can't use CPUID. Use ACPI ids from MADT instead. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by: David Vrabel <david.vrabel@citrix.com>
2016-06-30 23:56:37 +08:00
EXPORT_PER_CPU_SYMBOL(xen_vcpu_id);
xen: clean up domain mode predicates There are four operating modes Xen code may find itself running in: - native - hvm domain - pv dom0 - pv domU Clean up predicates for testing for these states to make them more consistent. Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Cc: Xen-devel <xen-devel@lists.xensource.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-08-20 04:16:17 +08:00
enum xen_domain_type xen_domain_type = XEN_NATIVE;
EXPORT_SYMBOL_GPL(xen_domain_type);
xen: implement XENMEM_machphys_mapping This hypercall allows Xen to specify a non-default location for the machine to physical mapping. This capability is used when running a 32 bit domain 0 on a 64 bit hypervisor to shrink the hypervisor hole to exactly the size required. [ Impact: add Xen hypercall definitions ] Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
2010-09-30 19:37:26 +08:00
unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START;
EXPORT_SYMBOL(machine_to_phys_mapping);
xen/x86: replace order-based range checking of M2P table by linear one The order-based approach is not only less efficient (requiring a shift and a compare, typical generated code looking like this mov eax, [machine_to_phys_order] mov ecx, eax shr ebx, cl test ebx, ebx jnz ... whereas a direct check requires just a compare, like in cmp ebx, [machine_to_phys_nr] jae ... ), but also slightly dangerous in the 32-on-64 case - the element address calculation can wrap if the next power of two boundary is sufficiently far away from the actual upper limit of the table, and hence can result in user space addresses being accessed (with it being unknown what may actually be mapped there). Additionally, the elimination of the mistaken use of fls() here (should have been __fls()) fixes a latent issue on x86-64 that would trigger if the code was run on a system with memory extending beyond the 44-bit boundary. CC: stable@kernel.org Signed-off-by: Jan Beulich <jbeulich@novell.com> [v1: Based on Jeremy's feedback] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2011-08-16 22:07:41 +08:00
unsigned long machine_to_phys_nr;
EXPORT_SYMBOL(machine_to_phys_nr);
xen: implement XENMEM_machphys_mapping This hypercall allows Xen to specify a non-default location for the machine to physical mapping. This capability is used when running a 32 bit domain 0 on a 64 bit hypervisor to shrink the hypervisor hole to exactly the size required. [ Impact: add Xen hypercall definitions ] Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
2010-09-30 19:37:26 +08:00
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
struct start_info *xen_start_info;
EXPORT_SYMBOL_GPL(xen_start_info);
xen: make dummy_shared_info non-static Rename dummy_shared_info to xen_dummy_shared_info and make it non-static, in anticipation of users outside of enlighten.c Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-05-27 06:31:21 +08:00
struct shared_info xen_dummy_shared_info;
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
xen: Move xen_have_vector_callback definition to enlighten.c Commit 84d582d236dc ("xen: Revert commits da72ff5bfcb0 and 72a9b186292d") defined xen_have_vector_callback in enlighten_hvm.c. Since guest-type-neutral code refers to this variable this causes build failures when CONFIG_XEN_PVHVM is not defined. Moving xen_have_vector_callback definition to enlighten.c resolves this issue. Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reported-by: Randy Dunlap <rdunlap@infradead.org> Acked-by: Randy Dunlap <rdunlap@infradead.org> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-05-03 01:23:06 +08:00
__read_mostly int xen_have_vector_callback;
EXPORT_SYMBOL_GPL(xen_have_vector_callback);
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
/*
* Point at some empty memory to start with. We map the real shared_info
* page as soon as fixmap is up and running.
*/
xen: remove cast from HYPERVISOR_shared_info assignment Both have type struct shared_info so no cast is needed. Signed-off-by: Olaf Hering <olaf@aepfle.de> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-17 17:57:41 +08:00
struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info;
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
/*
* Flag to determine whether vcpu info placement is available on all
* VCPUs. We assume it is to start with, and then set it to zero on
* the first failure. This is because it can succeed on some VCPUs
* and not others, since it can involve hypervisor memory allocation,
* or because the guest failed to guarantee all the appropriate
* constraints on all VCPUs (ie buffer can't cross a page boundary).
*
* Note that any particular CPU may be using a placed vcpu structure,
* but we can only optimise if the all are.
*
* 0: not available, 1: available
*/
x86/xen: globalize have_vcpu_info_placement have_vcpu_info_placement applies to both PV and HVM and as we're going to split the code we need to make it global. Rename to xen_have_vcpu_info_placement. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:37 +08:00
int xen_have_vcpu_info_placement = 1;
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
static int xen_cpu_up_online(unsigned int cpu)
{
xen_init_lock_cpu(cpu);
return 0;
}
xen/x86: avoid updating TLS descriptors if they haven't changed When switching tasks in a Xen PV guest, avoid updating the TLS descriptors if they haven't changed. This improves the speed of context switches by almost 10% as much of the time the descriptors are the same or only one is different. The descriptors written into the GDT by Xen are modified from the values passed in the update_descriptor hypercall so we keep shadow copies of the three TLS descriptors to compare against. lmbench3 test Before After Improvement -------------------------------------------- lat_ctx -s 32 24 7.19 6.52 9% lat_pipe 12.56 11.66 7% Signed-off-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:08 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
int xen_cpuhp_setup(int (*cpu_up_prepare_cb)(unsigned int),
int (*cpu_dead_cb)(unsigned int))
{
int rc;
rc = cpuhp_setup_state_nocalls(CPUHP_XEN_PREPARE,
xen/vcpu: Use a unified name about cpu hotplug state for pv and pvhvm As xen_cpuhp_setup is called by PV and PVHVM, the name of "x86/xen/hvm_guest" is confusing. Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-10-09 10:58:46 +08:00
"x86/xen/guest:prepare",
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
cpu_up_prepare_cb, cpu_dead_cb);
if (rc >= 0) {
rc = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
xen/vcpu: Use a unified name about cpu hotplug state for pv and pvhvm As xen_cpuhp_setup is called by PV and PVHVM, the name of "x86/xen/hvm_guest" is confusing. Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2017-10-09 10:58:46 +08:00
"x86/xen/guest:online",
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
xen_cpu_up_online, NULL);
if (rc < 0)
cpuhp_remove_state_nocalls(CPUHP_XEN_PREPARE);
}
return rc >= 0 ? 0 : rc;
}
xen/x86: avoid updating TLS descriptors if they haven't changed When switching tasks in a Xen PV guest, avoid updating the TLS descriptors if they haven't changed. This improves the speed of context switches by almost 10% as much of the time the descriptors are the same or only one is different. The descriptors written into the GDT by Xen are modified from the values passed in the update_descriptor hypercall so we keep shadow copies of the three TLS descriptors to compare against. lmbench3 test Before After Improvement -------------------------------------------- lat_ctx -s 32 24 7.19 6.52 9% lat_pipe 12.56 11.66 7% Signed-off-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:08 +08:00
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
static int xen_vcpu_setup_restore(int cpu)
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
{
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
int rc = 0;
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
/* Any per_cpu(xen_vcpu) is stale, so reset it */
xen_vcpu_info_reset(cpu);
/*
* For PVH and PVHVM, setup online VCPUs only. The rest will
* be handled by hotplug.
*/
if (xen_pv_domain() ||
(xen_hvm_domain() && cpu_online(cpu))) {
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
rc = xen_vcpu_setup(cpu);
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
}
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
return rc;
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
}
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
/*
* On restore, set the vcpu placement up again.
* If it fails, then we're in a bad state, since
* we can't back out from using it...
*/
void xen_vcpu_restore(void)
{
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
int cpu, rc;
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
for_each_possible_cpu(cpu) {
bool other_cpu = (cpu != smp_processor_id());
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
bool is_up;
if (xen_vcpu_nr(cpu) == XEN_VCPU_ID_INVALID)
continue;
/* Only Xen 4.5 and higher support this. */
is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up,
xen_vcpu_nr(cpu), NULL) > 0;
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
if (other_cpu && is_up &&
HYPERVISOR_vcpu_op(VCPUOP_down, xen_vcpu_nr(cpu), NULL))
BUG();
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
if (xen_pv_domain() || xen_feature(XENFEAT_hvm_safe_pvclock))
xen_setup_runstate_info(cpu);
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
rc = xen_vcpu_setup_restore(cpu);
if (rc)
pr_emerg_once("vcpu restore failed for cpu=%d err=%d. "
"System will hang.\n", cpu, rc);
/*
* In case xen_vcpu_setup_restore() fails, do not bring up the
* VCPU. This helps us avoid the resulting OOPS when the VCPU
* accesses pvclock_vcpu_time via xen_vcpu (which is NULL.)
* Note that this does not improve the situation much -- now the
* VM hangs instead of OOPSing -- with the VCPUs that did not
* fail, spinning in stop_machine(), waiting for the failed
* VCPUs to come up.
*/
if (other_cpu && is_up && (rc == 0) &&
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
HYPERVISOR_vcpu_op(VCPUOP_up, xen_vcpu_nr(cpu), NULL))
BUG();
}
}
void xen_vcpu_info_reset(int cpu)
{
if (xen_vcpu_nr(cpu) < MAX_VIRT_CPUS) {
per_cpu(xen_vcpu, cpu) =
&HYPERVISOR_shared_info->vcpu_info[xen_vcpu_nr(cpu)];
} else {
/* Set to NULL so that if somebody accesses it we get an OOPS */
per_cpu(xen_vcpu, cpu) = NULL;
}
}
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
int xen_vcpu_setup(int cpu)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
{
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
struct vcpu_register_vcpu_info info;
int err;
struct vcpu_info *vcpup;
xen: make dummy_shared_info non-static Rename dummy_shared_info to xen_dummy_shared_info and make it non-static, in anticipation of users outside of enlighten.c Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2008-05-27 06:31:21 +08:00
BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
xen/vcpu/pvhvm: Fix vcpu hotplugging hanging. If a user did: echo 0 > /sys/devices/system/cpu/cpu1/online echo 1 > /sys/devices/system/cpu/cpu1/online we would (this a build with DEBUG enabled) get to: smpboot: ++++++++++++++++++++=_---CPU UP 1 .. snip.. smpboot: Stack at about ffff880074c0ff44 smpboot: CPU1: has booted. and hang. The RCU mechanism would kick in an try to IPI the CPU1 but the IPIs (and all other interrupts) would never arrive at the CPU1. At first glance at least. A bit digging in the hypervisor trace shows that (using xenanalyze): [vla] d4v1 vec 243 injecting 0.043163027 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043163639 --|x d4v1 vmentry cycles 1468 ] 0.043164913 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043164913 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting 0.043164913 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043165526 --|x d4v1 vmentry cycles 1472 ] 0.043166800 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043166800 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting there is a pending event (subsequent debugging shows it is the IPI from the VCPU0 when smpboot.c on VCPU1 has done "set_cpu_online(smp_processor_id(), true)") and the guest VCPU1 is interrupted with the callback IPI (0xf3 aka 243) which ends up calling __xen_evtchn_do_upcall. The __xen_evtchn_do_upcall seems to do *something* but not acknowledge the pending events. And the moment the guest does a 'cli' (that is the ffffffff81673254 in the log above) the hypervisor is invoked again to inject the IPI (0xf3) to tell the guest it has pending interrupts. This repeats itself forever. The culprit was the per_cpu(xen_vcpu, cpu) pointer. At the bootup we set each per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] but later on use the VCPUOP_register_vcpu_info to register per-CPU structures (xen_vcpu_setup). This is used to allow events for more than 32 VCPUs and for performance optimizations reasons. When the user performs the VCPU hotplug we end up calling the the xen_vcpu_setup once more. We make the hypercall which returns -EINVAL as it does not allow multiple registration calls (and already has re-assigned where the events are being set). We pick the fallback case and set per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] (which is a good fallback during bootup). However the hypervisor is still setting events in the register per-cpu structure (per_cpu(xen_vcpu_info, cpu)). As such when the events are set by the hypervisor (such as timer one), and when we iterate in __xen_evtchn_do_upcall we end up reading stale events from the shared_info->vcpu_info[vcpu] instead of the per_cpu(xen_vcpu_info, cpu) structures. Hence we never acknowledge the events that the hypervisor has set and the hypervisor keeps on reminding us to ack the events which we never do. The fix is simple. Don't on the second time when xen_vcpu_setup is called over-write the per_cpu(xen_vcpu, cpu) if it points to per_cpu(xen_vcpu_info). Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> CC: stable@vger.kernel.org Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2013-05-05 21:30:09 +08:00
/*
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
* This path is called on PVHVM at bootup (xen_hvm_smp_prepare_boot_cpu)
* and at restore (xen_vcpu_restore). Also called for hotplugged
* VCPUs (cpu_init -> xen_hvm_cpu_prepare_hvm).
* However, the hypercall can only be done once (see below) so if a VCPU
* is offlined and comes back online then let's not redo the hypercall.
xen/vcpu/pvhvm: Fix vcpu hotplugging hanging. If a user did: echo 0 > /sys/devices/system/cpu/cpu1/online echo 1 > /sys/devices/system/cpu/cpu1/online we would (this a build with DEBUG enabled) get to: smpboot: ++++++++++++++++++++=_---CPU UP 1 .. snip.. smpboot: Stack at about ffff880074c0ff44 smpboot: CPU1: has booted. and hang. The RCU mechanism would kick in an try to IPI the CPU1 but the IPIs (and all other interrupts) would never arrive at the CPU1. At first glance at least. A bit digging in the hypervisor trace shows that (using xenanalyze): [vla] d4v1 vec 243 injecting 0.043163027 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043163639 --|x d4v1 vmentry cycles 1468 ] 0.043164913 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043164913 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting 0.043164913 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043165526 --|x d4v1 vmentry cycles 1472 ] 0.043166800 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043166800 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting there is a pending event (subsequent debugging shows it is the IPI from the VCPU0 when smpboot.c on VCPU1 has done "set_cpu_online(smp_processor_id(), true)") and the guest VCPU1 is interrupted with the callback IPI (0xf3 aka 243) which ends up calling __xen_evtchn_do_upcall. The __xen_evtchn_do_upcall seems to do *something* but not acknowledge the pending events. And the moment the guest does a 'cli' (that is the ffffffff81673254 in the log above) the hypervisor is invoked again to inject the IPI (0xf3) to tell the guest it has pending interrupts. This repeats itself forever. The culprit was the per_cpu(xen_vcpu, cpu) pointer. At the bootup we set each per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] but later on use the VCPUOP_register_vcpu_info to register per-CPU structures (xen_vcpu_setup). This is used to allow events for more than 32 VCPUs and for performance optimizations reasons. When the user performs the VCPU hotplug we end up calling the the xen_vcpu_setup once more. We make the hypercall which returns -EINVAL as it does not allow multiple registration calls (and already has re-assigned where the events are being set). We pick the fallback case and set per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] (which is a good fallback during bootup). However the hypervisor is still setting events in the register per-cpu structure (per_cpu(xen_vcpu_info, cpu)). As such when the events are set by the hypervisor (such as timer one), and when we iterate in __xen_evtchn_do_upcall we end up reading stale events from the shared_info->vcpu_info[vcpu] instead of the per_cpu(xen_vcpu_info, cpu) structures. Hence we never acknowledge the events that the hypervisor has set and the hypervisor keeps on reminding us to ack the events which we never do. The fix is simple. Don't on the second time when xen_vcpu_setup is called over-write the per_cpu(xen_vcpu, cpu) if it points to per_cpu(xen_vcpu_info). Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> CC: stable@vger.kernel.org Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2013-05-05 21:30:09 +08:00
*
* For PV it is called during restore (xen_vcpu_restore) and bootup
* (xen_setup_vcpu_info_placement). The hotplug mechanism does not
* use this function.
*/
if (xen_hvm_domain()) {
if (per_cpu(xen_vcpu, cpu) == &per_cpu(xen_vcpu_info, cpu))
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
return 0;
xen/vcpu/pvhvm: Fix vcpu hotplugging hanging. If a user did: echo 0 > /sys/devices/system/cpu/cpu1/online echo 1 > /sys/devices/system/cpu/cpu1/online we would (this a build with DEBUG enabled) get to: smpboot: ++++++++++++++++++++=_---CPU UP 1 .. snip.. smpboot: Stack at about ffff880074c0ff44 smpboot: CPU1: has booted. and hang. The RCU mechanism would kick in an try to IPI the CPU1 but the IPIs (and all other interrupts) would never arrive at the CPU1. At first glance at least. A bit digging in the hypervisor trace shows that (using xenanalyze): [vla] d4v1 vec 243 injecting 0.043163027 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043163639 --|x d4v1 vmentry cycles 1468 ] 0.043164913 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043164913 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting 0.043164913 --|x d4v1 intr_window vec 243 src 5(vector) intr f3 ] 0.043165526 --|x d4v1 vmentry cycles 1472 ] 0.043166800 --|x d4v1 vmexit exit_reason PENDING_INTERRUPT eip ffffffff81673254 0.043166800 --|x d4v1 inj_virq vec 243 real [vla] d4v1 vec 243 injecting there is a pending event (subsequent debugging shows it is the IPI from the VCPU0 when smpboot.c on VCPU1 has done "set_cpu_online(smp_processor_id(), true)") and the guest VCPU1 is interrupted with the callback IPI (0xf3 aka 243) which ends up calling __xen_evtchn_do_upcall. The __xen_evtchn_do_upcall seems to do *something* but not acknowledge the pending events. And the moment the guest does a 'cli' (that is the ffffffff81673254 in the log above) the hypervisor is invoked again to inject the IPI (0xf3) to tell the guest it has pending interrupts. This repeats itself forever. The culprit was the per_cpu(xen_vcpu, cpu) pointer. At the bootup we set each per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] but later on use the VCPUOP_register_vcpu_info to register per-CPU structures (xen_vcpu_setup). This is used to allow events for more than 32 VCPUs and for performance optimizations reasons. When the user performs the VCPU hotplug we end up calling the the xen_vcpu_setup once more. We make the hypercall which returns -EINVAL as it does not allow multiple registration calls (and already has re-assigned where the events are being set). We pick the fallback case and set per_cpu(xen_vcpu, cpu) to point to the shared_info->vcpu_info[vcpu] (which is a good fallback during bootup). However the hypervisor is still setting events in the register per-cpu structure (per_cpu(xen_vcpu_info, cpu)). As such when the events are set by the hypervisor (such as timer one), and when we iterate in __xen_evtchn_do_upcall we end up reading stale events from the shared_info->vcpu_info[vcpu] instead of the per_cpu(xen_vcpu_info, cpu) structures. Hence we never acknowledge the events that the hypervisor has set and the hypervisor keeps on reminding us to ack the events which we never do. The fix is simple. Don't on the second time when xen_vcpu_setup is called over-write the per_cpu(xen_vcpu, cpu) if it points to per_cpu(xen_vcpu_info). Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> CC: stable@vger.kernel.org Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2013-05-05 21:30:09 +08:00
}
xen/vcpu: Simplify xen_vcpu related code Largely mechanical changes to aid unification of xen_vcpu_restore() logic for PV, PVH and PVHVM. xen_vcpu_setup(): the only change in logic is that clamp_max_cpus() is now handled inside the "if (!xen_have_vcpu_info_placement)" block. xen_vcpu_restore(): code movement from enlighten_pv.c to enlighten.c. xen_vcpu_info_reset(): pulls together all the code where xen_vcpu is set to default. Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:58 +08:00
if (xen_have_vcpu_info_placement) {
vcpup = &per_cpu(xen_vcpu_info, cpu);
info.mfn = arbitrary_virt_to_mfn(vcpup);
info.offset = offset_in_page(vcpup);
/*
* Check to see if the hypervisor will put the vcpu_info
* structure where we want it, which allows direct access via
* a percpu-variable.
* N.B. This hypercall can _only_ be called once per CPU.
* Subsequent calls will error out with -EINVAL. This is due to
* the fact that hypervisor has no unregister variant and this
* hypercall does not allow to over-write info.mfn and
* info.offset.
*/
err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info,
xen_vcpu_nr(cpu), &info);
if (err) {
pr_warn_once("register_vcpu_info failed: cpu=%d err=%d\n",
cpu, err);
xen_have_vcpu_info_placement = 0;
} else {
/*
* This cpu is using the registered vcpu info, even if
* later ones fail to.
*/
per_cpu(xen_vcpu, cpu) = vcpup;
}
}
xen: Place vcpu_info structure into per-cpu memory An experimental patch for Xen allows guests to place their vcpu_info structs anywhere. We try to use this to place the vcpu_info into the PDA, which allows direct access. If this works, then switch to using direct access operations for irq_enable, disable, save_fl and restore_fl. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Chris Wright <chrisw@sous-sol.org> Cc: Keir Fraser <keir@xensource.com>
2007-07-18 09:37:07 +08:00
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
if (!xen_have_vcpu_info_placement)
xen/pvh*: Support > 32 VCPUs at domain restore When Xen restores a PVHVM or PVH guest, its shared_info only holds up to 32 CPUs. The hypercall VCPUOP_register_vcpu_info allows us to setup per-page areas for VCPUs. This means we can boot PVH* guests with more than 32 VCPUs. During restore the per-cpu structure is allocated freshly by the hypervisor (vcpu_info_mfn is set to INVALID_MFN) so that the newly restored guest can make a VCPUOP_register_vcpu_info hypercall. However, we end up triggering this condition in Xen: /* Run this command on yourself or on other offline VCPUS. */ if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) ) which means we are unable to setup the per-cpu VCPU structures for running VCPUS. The Linux PV code paths makes this work by iterating over cpu_possible in xen_vcpu_restore() with: 1) is target CPU up (VCPUOP_is_up hypercall?) 2) if yes, then VCPUOP_down to pause it 3) VCPUOP_register_vcpu_info 4) if it was down, then VCPUOP_up to bring it back up With Xen commit 192df6f9122d ("xen/x86: allow HVM guests to use hypercalls to bring up vCPUs") this is available for non-PV guests. As such first check if VCPUOP_is_up is actually possible before trying this dance. As most of this dance code is done already in xen_vcpu_restore() let's make it callable on PV, PVH and PVHVM. Based-on-patch-by: Konrad Wilk <konrad.wilk@oracle.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:05:59 +08:00
xen_vcpu_info_reset(cpu);
xen/vcpu: Handle xen_vcpu_setup() failure in hotplug The hypercall VCPUOP_register_vcpu_info can fail. This failure is handled by making per_cpu(xen_vcpu, cpu) point to its shared_info slot and those without one (cpu >= MAX_VIRT_CPUS) be NULL. For PVH/PVHVM, this is not enough, because we also need to pull these VCPUs out of circulation. Fix for PVH/PVHVM: on registration failure in the cpuhp prepare callback (xen_cpu_up_prepare_hvm()), return an error to the cpuhp state-machine so it can fail the CPU init. Fix for PV: the registration happens before smp_init(), so, in the failure case we clamp setup_max_cpus and limit the number of VCPUs that smp_init() will bring-up to MAX_VIRT_CPUS. This is functionally correct but it makes the code a bit simpler if we get rid of this explicit clamping: for VCPUs that don't have valid xen_vcpu, fail the CPU init in the cpuhp prepare callback (xen_cpu_up_prepare_pv()). Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-06-03 08:06:01 +08:00
return ((per_cpu(xen_vcpu, cpu) == NULL) ? -ENODEV : 0);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
}
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
void xen_reboot(int reason)
xen: restore vcpu_info mapping If we're using vcpu_info mapping, then make sure its restored on all processors before relasing them from stop_machine. The only complication is that if this fails, we can't continue because we've already made assumptions that the mapping is available (baked in calls to the _direct versions of the functions, for example). Fortunately this can only happen with a 32-bit hypervisor, which may possibly run out of mapping space. On a 64-bit hypervisor, this is a non-issue. Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-05-31 08:33:02 +08:00
{
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
struct sched_shutdown r = { .reason = reason };
xen: restore runstate_info even if !have_vcpu_info_placement Even if have_vcpu_info_placement is not set, we still need to set up the runstate area on each resumed vcpu. Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Cc: Stable Kernel <stable@kernel.org>
2009-11-21 08:46:29 +08:00
int cpu;
xen: restore vcpu_info mapping If we're using vcpu_info mapping, then make sure its restored on all processors before relasing them from stop_machine. The only complication is that if this fails, we can't continue because we've already made assumptions that the mapping is available (baked in calls to the _direct versions of the functions, for example). Fortunately this can only happen with a 32-bit hypervisor, which may possibly run out of mapping space. On a 64-bit hypervisor, this is a non-issue. Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-05-31 08:33:02 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
for_each_online_cpu(cpu)
xen_pmu_finish(cpu);
x86/xen: Probe target addresses in set_aliased_prot() before the hypercall The update_va_mapping hypercall can fail if the VA isn't present in the guest's page tables. Under certain loads, this can result in an OOPS when the target address is in unpopulated vmap space. While we're at it, add comments to help explain what's going on. This isn't a great long-term fix. This code should probably be changed to use something like set_memory_ro. Signed-off-by: Andy Lutomirski <luto@kernel.org> Cc: Andrew Cooper <andrew.cooper3@citrix.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: David Vrabel <dvrabel@cantab.net> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jan Beulich <jbeulich@suse.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sasha Levin <sasha.levin@oracle.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: security@kernel.org <security@kernel.org> Cc: <stable@vger.kernel.org> Cc: xen-devel <xen-devel@lists.xen.org> Link: http://lkml.kernel.org/r/0b0e55b995cda11e7829f140b833ef932fcabe3a.1438291540.git.luto@kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-07-31 05:31:31 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
if (HYPERVISOR_sched_op(SCHEDOP_shutdown, &r))
xen: fix allocation and use of large ldts When the ldt gets to more than 1 page in size, the kernel uses vmalloc to allocate it. This means that: - when making the ldt RO, we must update the pages in both the vmalloc mapping and the linear mapping to make sure there are no RW aliases. - we need to use arbitrary_virt_to_machine to compute the machine addr for each update Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-27 23:45:02 +08:00
BUG();
x86/paravirt/xen: properly fill out the ldt ops LTP testing showed that Xen does not properly implement sys_modify_ldt(). This patch does the final little bits needed to make the ldt work properly. Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-24 05:21:18 +08:00
}
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
void xen_emergency_restart(void)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
{
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
xen_reboot(SHUTDOWN_reboot);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
}
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
static int
xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
{
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
if (!kexec_crash_loaded())
xen_reboot(SHUTDOWN_crash);
return NOTIFY_DONE;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
}
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
static struct notifier_block xen_panic_block = {
.notifier_call = xen_panic_event,
.priority = INT_MIN
};
xen: make -fstack-protector work under Xen -fstack-protector uses a special per-cpu "stack canary" value. gcc generates special code in each function to test the canary to make sure that the function's stack hasn't been overrun. On x86-64, this is simply an offset of %gs, which is the usual per-cpu base segment register, so setting it up simply requires loading %gs's base as normal. On i386, the stack protector segment is %gs (rather than the usual kernel percpu %fs segment register). This requires setting up the full kernel GDT and then loading %gs accordingly. We also need to make sure %gs is initialized when bringing up secondary cpus too. To keep things consistent, we do the full GDT/segment register setup on both architectures. Because we need to avoid -fstack-protected code before setting up the GDT and because there's no way to disable it on a per-function basis, several files need to have stack-protector inhibited. [ Impact: allow Xen booting with stack-protector enabled ] Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2009-08-28 03:46:35 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
int xen_panic_handler_init(void)
xen/x86: add desc_equal() to compare GDT descriptors Signed-off-by: David Vrabel <david.vrabel@citrix.com> [v1: Moving it to the Xen file] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:07 +08:00
{
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
return 0;
xen/x86: add desc_equal() to compare GDT descriptors Signed-off-by: David Vrabel <david.vrabel@citrix.com> [v1: Moving it to the Xen file] Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:07 +08:00
}
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
void xen_pin_vcpu(int cpu)
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
{
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
static bool disable_pinning;
struct sched_pin_override pin_override;
int ret;
xen/x86: avoid updating TLS descriptors if they haven't changed When switching tasks in a Xen PV guest, avoid updating the TLS descriptors if they haven't changed. This improves the speed of context switches by almost 10% as much of the time the descriptors are the same or only one is different. The descriptors written into the GDT by Xen are modified from the values passed in the update_descriptor hypercall so we keep shadow copies of the three TLS descriptors to compare against. lmbench3 test Before After Improvement -------------------------------------------- lat_ctx -s 32 24 7.19 6.52 9% lat_pipe 12.56 11.66 7% Signed-off-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:08 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
if (disable_pinning)
xen/x86: avoid updating TLS descriptors if they haven't changed When switching tasks in a Xen PV guest, avoid updating the TLS descriptors if they haven't changed. This improves the speed of context switches by almost 10% as much of the time the descriptors are the same or only one is different. The descriptors written into the GDT by Xen are modified from the values passed in the update_descriptor hypercall so we keep shadow copies of the three TLS descriptors to compare against. lmbench3 test Before After Improvement -------------------------------------------- lat_ctx -s 32 24 7.19 6.52 9% lat_pipe 12.56 11.66 7% Signed-off-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
2012-07-09 18:39:08 +08:00
return;
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
pin_override.pcpu = cpu;
ret = HYPERVISOR_sched_op(SCHEDOP_pin_override, &pin_override);
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
/* Ignore errors when removing override. */
if (cpu < 0)
return;
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
x86/xen: split off enlighten_pv.c Basically, enlighten.c is renamed to enlighten_pv.c and some code moved out to common enlighten.c. Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> Reviewed-by: Juergen Gross <jgross@suse.com> Signed-off-by: Juergen Gross <jgross@suse.com>
2017-03-15 01:35:41 +08:00
switch (ret) {
case -ENOSYS:
pr_warn("Unable to pin on physical cpu %d. In case of problems consider vcpu pinning.\n",
cpu);
disable_pinning = true;
break;
case -EPERM:
WARN(1, "Trying to pin vcpu without having privilege to do so\n");
disable_pinning = true;
break;
case -EINVAL:
case -EBUSY:
pr_warn("Physical cpu %d not available for pinning. Check Xen cpu configuration.\n",
cpu);
break;
case 0:
break;
default:
WARN(1, "rc %d while trying to pin vcpu\n", ret);
disable_pinning = true;
xen64: Clear %fs on xen_load_tls() We need to do this, otherwise we can get a GPF on hypercall return after TLS descriptor is cleared but %fs is still pointing to it. Signed-off-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Cc: Stephen Tweedie <sct@redhat.com> Cc: Eduardo Habkost <ehabkost@redhat.com> Cc: Mark McLoughlin <markmc@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-09 06:07:10 +08:00
}
xen: Core Xen implementation This patch is a rollup of all the core pieces of the Xen implementation, including: - booting and setup - pagetable setup - privileged instructions - segmentation - interrupt flags - upcalls - multicall batching BOOTING AND SETUP The vmlinux image is decorated with ELF notes which tell the Xen domain builder what the kernel's requirements are; the domain builder then constructs the address space accordingly and starts the kernel. Xen has its own entrypoint for the kernel (contained in an ELF note). The ELF notes are set up by xen-head.S, which is included into head.S. In principle it could be linked separately, but it seems to provoke lots of binutils bugs. Because the domain builder starts the kernel in a fairly sane state (32-bit protected mode, paging enabled, flat segments set up), there's not a lot of setup needed before starting the kernel proper. The main steps are: 1. Install the Xen paravirt_ops, which is simply a matter of a structure assignment. 2. Set init_mm to use the Xen-supplied pagetables (analogous to the head.S generated pagetables in a native boot). 3. Reserve address space for Xen, since it takes a chunk at the top of the address space for its own use. 4. Call start_kernel() PAGETABLE SETUP Once we hit the main kernel boot sequence, it will end up calling back via paravirt_ops to set up various pieces of Xen specific state. One of the critical things which requires a bit of extra care is the construction of the initial init_mm pagetable. Because Xen places tight constraints on pagetables (an active pagetable must always be valid, and must always be mapped read-only to the guest domain), we need to be careful when constructing the new pagetable to keep these constraints in mind. It turns out that the easiest way to do this is use the initial Xen-provided pagetable as a template, and then just insert new mappings for memory where a mapping doesn't already exist. This means that during pagetable setup, it uses a special version of xen_set_pte which ignores any attempt to remap a read-only page as read-write (since Xen will map its own initial pagetable as RO), but lets other changes to the ptes happen, so that things like NX are set properly. PRIVILEGED INSTRUCTIONS AND SEGMENTATION When the kernel runs under Xen, it runs in ring 1 rather than ring 0. This means that it is more privileged than user-mode in ring 3, but it still can't run privileged instructions directly. Non-performance critical instructions are dealt with by taking a privilege exception and trapping into the hypervisor and emulating the instruction, but more performance-critical instructions have their own specific paravirt_ops. In many cases we can avoid having to do any hypercalls for these instructions, or the Xen implementation is quite different from the normal native version. The privileged instructions fall into the broad classes of: Segmentation: setting up the GDT and the GDT entries, LDT, TLS and so on. Xen doesn't allow the GDT to be directly modified; all GDT updates are done via hypercalls where the new entries can be validated. This is important because Xen uses segment limits to prevent the guest kernel from damaging the hypervisor itself. Traps and exceptions: Xen uses a special format for trap entrypoints, so when the kernel wants to set an IDT entry, it needs to be converted to the form Xen expects. Xen sets int 0x80 up specially so that the trap goes straight from userspace into the guest kernel without going via the hypervisor. sysenter isn't supported. Kernel stack: The esp0 entry is extracted from the tss and provided to Xen. TLB operations: the various TLB calls are mapped into corresponding Xen hypercalls. Control registers: all the control registers are privileged. The most important is cr3, which points to the base of the current pagetable, and we handle it specially. Another instruction we treat specially is CPUID, even though its not privileged. We want to control what CPU features are visible to the rest of the kernel, and so CPUID ends up going into a paravirt_op. Xen implements this mainly to disable the ACPI and APIC subsystems. INTERRUPT FLAGS Xen maintains its own separate flag for masking events, which is contained within the per-cpu vcpu_info structure. Because the guest kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely ignored (and must be, because even if a guest domain disables interrupts for itself, it can't disable them overall). (A note on terminology: "events" and interrupts are effectively synonymous. However, rather than using an "enable flag", Xen uses a "mask flag", which blocks event delivery when it is non-zero.) There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which are implemented to manage the Xen event mask state. The only thing worth noting is that when events are unmasked, we need to explicitly see if there's a pending event and call into the hypervisor to make sure it gets delivered. UPCALLS Xen needs a couple of upcall (or callback) functions to be implemented by each guest. One is the event upcalls, which is how events (interrupts, effectively) are delivered to the guests. The other is the failsafe callback, which is used to report errors in either reloading a segment register, or caused by iret. These are implemented in i386/kernel/entry.S so they can jump into the normal iret_exc path when necessary. MULTICALL BATCHING Xen provides a multicall mechanism, which allows multiple hypercalls to be issued at once in order to mitigate the cost of trapping into the hypervisor. This is particularly useful for context switches, since the 4-5 hypercalls they would normally need (reload cr3, update TLS, maybe update LDT) can be reduced to one. This patch implements a generic batching mechanism for hypercalls, which gets used in many places in the Xen code. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org> Cc: Ian Pratt <ian.pratt@xensource.com> Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk> Cc: Adrian Bunk <bunk@stusta.de>
2007-07-18 09:37:04 +08:00
}
xen/arm: Enable cpu_hotplug.c Build cpu_hotplug for ARM and ARM64 guests. Rename arch_(un)register_cpu to xen_(un)register_cpu and provide an empty implementation on ARM and ARM64. On x86 just call arch_(un)register_cpu as we are already doing. Initialize cpu_hotplug on ARM. Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Reviewed-by: Julien Grall <julien.grall@citrix.com> Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
2015-10-23 00:20:46 +08:00
#ifdef CONFIG_HOTPLUG_CPU
void xen_arch_register_cpu(int num)
{
arch_register_cpu(num);
}
EXPORT_SYMBOL(xen_arch_register_cpu);
void xen_arch_unregister_cpu(int num)
{
arch_unregister_cpu(num);
}
EXPORT_SYMBOL(xen_arch_unregister_cpu);
#endif
xen/balloon: Mark unallocated host memory as UNUSABLE Commit f5775e0b6116 ("x86/xen: discard RAM regions above the maximum reservation") left host memory not assigned to dom0 as available for memory hotplug. Unfortunately this also meant that those regions could be used by others. Specifically, commit fa564ad96366 ("x86/PCI: Enable a 64bit BAR on AMD Family 15h (Models 00-1f, 30-3f, 60-7f)") may try to map those addresses as MMIO. To prevent this mark unallocated host memory as E820_TYPE_UNUSABLE (thus effectively reverting f5775e0b6116) and keep track of that region as a hostmem resource that can be used for the hotplug. Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reviewed-by: Juergen Gross <jgross@suse.com>
2017-12-13 04:08:21 +08:00
#ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
void __init arch_xen_balloon_init(struct resource *hostmem_resource)
{
struct xen_memory_map memmap;
int rc;
unsigned int i, last_guest_ram;
phys_addr_t max_addr = PFN_PHYS(max_pfn);
struct e820_table *xen_e820_table;
const struct e820_entry *entry;
struct resource *res;
if (!xen_initial_domain())
return;
xen_e820_table = kmalloc(sizeof(*xen_e820_table), GFP_KERNEL);
if (!xen_e820_table)
return;
memmap.nr_entries = ARRAY_SIZE(xen_e820_table->entries);
set_xen_guest_handle(memmap.buffer, xen_e820_table->entries);
rc = HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap);
if (rc) {
pr_warn("%s: Can't read host e820 (%d)\n", __func__, rc);
goto out;
}
last_guest_ram = 0;
for (i = 0; i < memmap.nr_entries; i++) {
if (xen_e820_table->entries[i].addr >= max_addr)
break;
if (xen_e820_table->entries[i].type == E820_TYPE_RAM)
last_guest_ram = i;
}
entry = &xen_e820_table->entries[last_guest_ram];
if (max_addr >= entry->addr + entry->size)
goto out; /* No unallocated host RAM. */
hostmem_resource->start = max_addr;
hostmem_resource->end = entry->addr + entry->size;
/*
* Mark non-RAM regions between the end of dom0 RAM and end of host RAM
* as unavailable. The rest of that region can be used for hotplug-based
* ballooning.
*/
for (; i < memmap.nr_entries; i++) {
entry = &xen_e820_table->entries[i];
if (entry->type == E820_TYPE_RAM)
continue;
if (entry->addr >= hostmem_resource->end)
break;
res = kzalloc(sizeof(*res), GFP_KERNEL);
if (!res)
goto out;
res->name = "Unavailable host RAM";
res->start = entry->addr;
res->end = (entry->addr + entry->size < hostmem_resource->end) ?
entry->addr + entry->size : hostmem_resource->end;
rc = insert_resource(hostmem_resource, res);
if (rc) {
pr_warn("%s: Can't insert [%llx - %llx) (%d)\n",
__func__, res->start, res->end, rc);
kfree(res);
goto out;
}
}
out:
kfree(xen_e820_table);
}
#endif /* CONFIG_XEN_BALLOON_MEMORY_HOTPLUG */