Current kvm disables preemption while the new virtualization registers are
in use.  This of course is not very good for latency sensitive workloads (one
use of virtualization is to offload user interface and other latency
insensitive stuff to a container, so that it is easier to analyze the
remaining workload).  This patch re-enables preemption for kvm; preemption
is now only disabled when switching the registers in and out, and during
the switch to guest mode and back.

Contains fixes from Shaohua Li <shaohua.li@intel.com>.
Signed-off-by: Avi Kivity <avi@qumranet.com>

Add the hypercall number to kvm_run and initialize it.  This changes the ABI,
but as this particular ABI was unusable before this no users are affected.
Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Put cpu feature detecting part in hardware_setup, and stored the vmcs
condition in global variable for further check.

[glommer: fix for some i386-only machines not supporting CR8 load/store
 exiting]
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

This patch converts the vcpus array in "struct kvm" to a pointer
array, and changes the "vcpu_create" and "vcpu_setup" hooks into one
"vcpu_create" call which does the allocation and initialization of the
vcpu (calling back into the kvm_vcpu_init core helper).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

struct kvm_vcpu has vmx-specific members; remove them to a private structure.
Signed-off-by: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

load_pdptrs can be handed an invalid cr3, and it should not oops.
This can happen because we injected #gp in set_cr3() after we set
vcpu->cr3 to the invalid value, or from kvm_vcpu_ioctl_set_sregs(), or
memory configuration changes after the guest did set_cr3().

We should also copy the pdpte array once, before checking and
assigning, otherwise an SMP guest can potentially alter the values
between the check and the set.

Finally one nitpick: ret = 1 should be done as late as possible: this
allows GCC to check for unset "ret" should the function change in
future.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

The writeback fixes (02c03a32) let
some dead code in the cmpxchg instruction emulation. Remove it.
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Signed-off-by: Avi Kivity <avi@qumranet.com>

This patch mainly imports some constants and rename two exist constants
of vmcs according to IA32 SDM.

It also adds two constants to indicate Lock bit and Enable bit in
MSR_IA32_FEATURE_CONTROL, and replace the hardcode _5_ with these two
bits.
Signed-off-by: Sheng Yang <sheng.yang@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

gfn_to_page might sleep with swap support. Move it out of the kmap calls.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

vmx_cpu_run doesn't handle error correctly and kvm_mmu_reload might
sleep with mutex changes, so I move it above.
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Don't fall through and turn on PAE in this case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Right now, the bug is harmless as we never emulate one-byte 0xb6 or 0xb7.
But things may change.

Noted by the mysterious Gabriel C.
Signed-off-by: Avi Kivity <avi@qumranet.com>

The writeback fixes (02c03a32) broke
cmov emulation.  Fix.
Signed-off-by: Avi Kivity <avi@qumranet.com>

Intel manual (and KVM definition) say the TPR is 4 bits wide.  Also fix
CR8_RESEVED_BITS typo.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Signed-off-by: Jeff Dike <jdike@linux.intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Creating one's own BITMAP macro seems suboptimal: if we use manual
arithmetic in the one place exposed to userspace, we can use standard
macros elsewhere.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

On this machine (Intel), writing to the CR4 bits 0x00000800 and
0x00001000 cause a GPF.  The Intel manual is a little unclear, but
AFIACT they're reserved, too.

Also fix spelling of CR4_RESEVED_BITS.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

The kernel now has asm/cpu-features.h: use those macros instead of inventing
our own.

Also spell out definition of CR3_RESEVED_BITS, fix spelling and
tighten it for the non-PAE case.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

The kernel now has asm/cpu-features.h: use those macros instead of
inventing our own.

Also spell out definition of CR0_RESEVED_BITS (no code change) and fix typo.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Don't pre-declare hardware_disable: shuffle the reboot hook down.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Speling error in comment.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

I have shied away from touching x86_emulate.c (it could definitely use
some love, but it is forked from the Xen code, and it would be more
productive to cross-merge fixes).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

KVM interface is no longer experimental.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Add string pio write support to support some version of Windows.
Signed-off-by: Yaozu (Eddie) Dong <eddie.dong@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

Note that as the size of struct kvm_run is not part of the ABI, we can add
things at the end.
Signed-off-by: Avi Kivity <avi@qumranet.com>

This patch adds a `vcpu_id' field in `struct vcpu', so we can
differentiate BSP and APs without pointer comparison or arithmetic.
Signed-off-by: Qing He <qing.he@intel.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

*nopage() in kvm_main.c should only store the type of mmap() fault if
the pointers are not NULL. This patch fixes the problem.
Signed-off-by: Nguyen Anh Quynh <aquynh@gmail.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>

KVM reuses the IOAPIC register definitions, and needs them even if the
host is not compiled with IOAPIC support.  Move the #ifdef below so that only
the IOAPIC variables and functions are protected, and the register definitions
are available to all.
Signed-off-by: Avi Kivity <avi@qumranet.com>

Fix DMI const-ification fallout that appeared when merging subsystem
trees.
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

According to latest memory ordering specification documents from Intel
and AMD, both manufacturers are committed to in-order loads from
cacheable memory for the x86 architecture.  Hence, smp_rmb() may be a
simple barrier.

Also according to those documents, and according to existing practice in
Linux (eg.  spin_unlock doesn't enforce ordering), stores to cacheable
memory are visible in program order too.  Special string stores are safe
-- their constituent stores may be out of order, but they must complete
in order WRT surrounding stores.  Nontemporal stores to WB memory can go
out of order, and so they should be fenced explicitly to make them
appear in-order WRT other stores.  Hence, smp_wmb() may be a simple
barrier.

    http://developer.intel.com/products/processor/manuals/318147.pdf
    http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/24593.pdf

In userspace microbenchmarks on a core2 system, fence instructions range
anywhere from around 15 cycles to 50, which may not be totally
insignificant in performance critical paths (code size will go down
too).

However the primary motivation for this is to have the canonical barrier
implementation for x86 architecture.

smp_rmb on buggy pentium pros remains a locked op, which is apparently
required.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

wmb() on x86 must always include a barrier, because stores can go out of
order in many cases when dealing with devices (eg. WC memory).
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

movnt* instructions are not strongly ordered with respect to other stores,
so if we are to assume stores are strongly ordered in the rest of the 64
bit code, we must fence these off (see similar examples in 32 bit code).

[ The AMD memory ordering document seems to say that nontemporal stores can
  also pass earlier regular stores, so maybe we need sfences _before_
  movnt* everywhere too? ]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

IOW, it needs to depend on both CONFIG_BLOCK and CONFIG_COMPAT.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev: (119 commits)
  [libata] struct pci_dev related cleanups
  libata: use ata_exec_internal() for PMP register access
  libata: implement ATA_PFLAG_RESETTING
  libata: add @timeout to ata_exec_internal[_sg]()
  ahci: fix notification handling
  ahci: clean up PORT_IRQ_BAD_PMP enabling
  ahci: kill leftover from enabling NCQ over PMP
  libata: wrap schedule_timeout_uninterruptible() in loop
  libata: skip suppress reporting if ATA_EHI_QUIET
  libata: clear ehi description after initial host report
  pata_jmicron: match vendor and class code only
  libata: add ST9160821AS / 3.ALD to NCQ blacklist
  pata_acpi: ACPI driver support
  libata-core: Expose gtm methods for driver use
  libata: add HDT722516DLA380 to NCQ blacklist
  libata: blacklist NCQ on Seagate Barracuda ST380817AS
  [libata] Turn on ACPI by default
  libata_scsi: Fix ATAPI transfer lengths
  libata: correct handling of SRST reset sequences
  libata: Integrate ACPI-based PATA/SATA hotplug - version 5
  ...

Since there is no x86-64 architecture anymore it cannot be maintained.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6: (37 commits)
  PCI: merge almost all of pci_32.h and pci_64.h together
  PCI: X86: Introduce and enable PCI domain support
  PCI: Add 'nodomains' boot option, and pci_domains_supported global
  PCI: modify PCI bridge control ISA flag for clarity
  PCI: use _CRS for PCI resource allocation
  PCI: avoid P2P prefetch window for expansion ROMs
  PCI: skip ISA ioresource alignment on some systems
  PCI: remove transparent bridge sizing
  pci: write file size to inode on proc bus file write
  pci: use size stored in proc_dir_entry for proc bus files
  pci: implement "pci=noaer"
  PCI: fix IDE legacy mode resources
  MSI: Use correct data offset for 32-bit MSI in read_msi_msg()
  PCI: Fix incorrect argument order to list_add_tail() in PCI dynamic ID code
  PCI: i386: Compaq EVO N800c needs PCI bus renumbering
  PCI: Remove no longer correct documentation regarding MSI vector assignment
  PCI: re-enable onboard sound on "MSI K8T Neo2-FIR"
  PCI: quirk_vt82c586_acpi: Omit reading PCI revision ID
  PCI: quirk amd_8131_mmrbc: Omit reading pci revision ID
  cpqphp: Use PCI_CLASS_REVISION instead of PCI_REVISION_ID for read
  ...

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6: (75 commits)
  PM: merge device power-management source files
  sysfs: add copyrights
  kobject: update the copyrights
  kset: add some kerneldoc to help describe what these strange things are
  Driver core: rename ktype_edd and ktype_efivar
  Driver core: rename ktype_driver
  Driver core: rename ktype_device
  Driver core: rename ktype_class
  driver core: remove subsystem_init()
  sysfs: move sysfs file poll implementation to sysfs_open_dirent
  sysfs: implement sysfs_open_dirent
  sysfs: move sysfs_dirent->s_children into sysfs_dirent->s_dir
  sysfs: make sysfs_root a regular directory dirent
  sysfs: open code sysfs_attach_dentry()
  sysfs: make s_elem an anonymous union
  sysfs: make bin attr open get active reference of parent too
  sysfs: kill unnecessary NULL pointer check in sysfs_release()
  sysfs: kill unnecessary sysfs_get() in open paths
  sysfs: reposition sysfs_dirent->s_mode.
  sysfs: kill sysfs_update_file()
  ...

* master.kernel.org:/pub/scm/linux/kernel/git/gregkh/usb-2.6: (142 commits)
  USB: fix race in autosuspend reschedule
  atmel_usba_udc: Keep track of the device status
  USB: Nikon D40X unusual_devs entry
  USB: serial core should respect driver requirements
  USB: documentation for USB power management
  USB: skip autosuspended devices during system resume
  USB: mutual exclusion for EHCI init and port resets
  USB: allow usbstorage to have LUNS greater than 2Tb
  USB: Adding support for SHARP WS011SH to ipaq.c
  USB: add atmel_usba_udc driver
  USB: ohci SSB bus glue
  USB: ehci build fixes on au1xxx, ppc-soc
  USB: add runtime frame_no quirk for big-endian OHCI
  USB: funsoft: Fix termios
  USB: visor: termios bits
  USB: unusual_devs entry for Nikon DSC D2Xs
  USB: re-remove <linux/usb_sl811.h>
  USB: move <linux/usb_gadget.h> to <linux/usb/gadget.h>
  USB: Export URB statistics for powertop
  USB: serial gadget: Disable endpoints on unload
  ...

* master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreq:
  [CPUFREQ] Don't take semaphore in cpufreq_quick_get()
  [CPUFREQ] Support different families in fid/did to frequency conversion
  [CPUFREQ] cpufreq_stats: misc cpuinit section annotations
  [CPUFREQ] implement !CONFIG_CPU_FREQ stub for  cpufreq_unregister_notifier()
  [CPUFREQ] mark hotplug notifier callback as __cpuinit
  [CPUFREQ] Only check for transition latency on problematic governors (kconfig fix)
  [CPUFREQ] allow ondemand and conservative cpufreq governors to be used as default
  [CPUFREQ] move policy's governor initialisation out of low-level drivers into cpufreq core
  [CPUFREQ] Longhaul - Add support for PM133 northbridge
  [CPUFREQ] x86: use num_online_nodes to get physical cpus numbers for