Guest user mode can generate a guest MSA Disabled exception on an MSA
capable core by simply trying to execute an MSA instruction. Since this
exception is unknown to KVM it will be passed on to the guest kernel.
However guest Linux kernels prior to v3.15 do not set up an exception
handler for the MSA Disabled exception as they don't support any MSA
capable cores. This results in a guest OS panic.

Since an older processor ID may be being emulated, and MSA support is
not advertised to the guest, the correct behaviour is to generate a
Reserved Instruction exception in the guest kernel so it can send the
guest process an illegal instruction signal (SIGILL), as would happen
with a non-MSA-capable core.

Fix this as minimally as reasonably possible by preventing
kvm_mips_check_privilege() from relaying MSA Disabled exceptions from
guest user mode to the guest kernel, and handling the MSA Disabled
exception by emulating a Reserved Instruction exception in the guest,
via a new handle_msa_disabled() KVM callback.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Paul Burton <paul.burton@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Gleb Natapov <gleb@kernel.org>
Cc: linux-mips@linux-mips.org
Cc: kvm@vger.kernel.org
Cc: <stable@vger.kernel.org> # v3.15+

MIPS FP/MSA fixes from the MIPS tree. Includes a fix to ensure that the
FPU is properly disabled by lose_fpu() when MSA is in use, and Paul
Burton's "FP/MSA fixes" patchset which is required for FP/MSA support in
KVM:

> This series fixes a bunch of bugs, both build & runtime, with FP & MSA
> support. Most of them only affect systems with the new FP modes & MSA
> support enabled but patch 6 in particular is more general, fixing
> problems for mips64 systems.

The maximum word size is 64-bits since MSA state is saved using st.d
which stores two 64-bit words, therefore reimplement FPR_IDX using xor,
and only within each 64-bit word.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9169/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

This reverts commit 02987633.

The basic premise of the patch was incorrect since MSA context
(including FP state) is saved using st.d which stores two consecutive
64-bit words in memory rather than a single 128-bit word. This means
that even with big endian MSA, the FP state is still in the first 64-bit
word.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9168/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

The expected semantics of __enable_fpu are for the FPU to be enabled
in the given mode if possible, otherwise for the FPU to be left
disabled and SIGFPE returned. The FPU was incorrectly being left
enabled in cases where the desired value for FR was unavailable.
Without ensuring the FPU is disabled in this case, it would be
possible for userland to go on to execute further FP instructions
natively in the incorrect mode, rather than those instructions being
trapped & emulated as they need to be.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9167/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

If a ptracee has not used the FPU and the ptracer sets its FP context
using PTRACE_POKEUSR, PTRACE_SETFPREGS or PTRACE_SETREGSET then that
context will be discarded upon either the ptracee using the FPU or a
further write to the context via ptrace. Prevent this loss by recording
that the task has "used" math once its FP context has been written to.
The context initialisation code that was present for the PTRACE_POKEUSR
case is reused for the other 2 cases to provide consistent behaviour
for the different ptrace requests.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9166/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

When running the emulator to handle an instruction that raised an FP
unimplemented operation exception, the FCSR cause bits were being
cleared. This is done to ensure that the kernel does not take an FP
exception when later restoring FP context to registers. However, this
was not being done when the emulator is invoked in response to a
coprocessor unusable exception. This happens in 2 cases:

  - There is no FPU present in the system. In this case things were
    OK, since the FP context is never restored to hardware registers
    and thus no FP exception may be raised when restoring FCSR.

  - The FPU could not be configured to the mode required by the task.
    In this case it would be possible for the emulator to set cause
    bits which are later restored to hardware if the task migrates
    to a CPU whose associated FPU does support its mode requirements,
    or if the tasks FP mode requirements change.

Consistently clear the cause bits after invoking the emulator, by moving
the clearing to process_fpemu_return and ensuring this is always called
before the tasks FP context is restored. This will make it easier to
catch further paths invoking the emulator in future, as will be
introduced in further patches.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9165/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Much like for traditional scalar FP exceptions, the cause bits in the
MSACSR register need to be cleared following an MSA FP exception.
Without doing so the exception will simply be raised again whenever
the kernel restores MSACSR from a tasks saved context, leading to
undesirable spurious exceptions. Clear the cause bits from the
handle_msa_fpe function, mirroring the way handle_fpe clears the
cause bits in FCSR.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9164/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Uses of the cfcmsa & ctcmsa instructions were not being wrapped by a
macro in the case where the toolchain supports MSA, since the arguments
exactly match a typical use of the instructions. However using current
toolchains this leads to errors such as:

  arch/mips/kernel/genex.S:437: Error: opcode not supported on this processor: mips32r2 (mips32r2) `cfcmsa $5,1'

Thus uses of the instructions must be in the context of a ".set msa"
directive, however doing that from the users of the instructions would
be messy due to the possibility that the toolchain does not support
MSA. Fix this by renaming the macros (prepending an underscore) in order
to avoid recursion when attempting to emit the instructions, and provide
implementations for the TOOLCHAIN_SUPPORTS_MSA case which ".set msa" as
appropriate.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9163/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Recursive macros made the code more concise & worked great for the
case where the toolchain doesn't support MSA. However, with toolchains
which do support MSA they lead to build failures such as:

  arch/mips/kernel/r4k_switch.S: Assembler messages:
  arch/mips/kernel/r4k_switch.S:148: Error: invalid operands `insert.w $w(0+1)[2],$1'
  arch/mips/kernel/r4k_switch.S:148: Error: invalid operands `insert.w $w(0+1)[3],$1'
  arch/mips/kernel/r4k_switch.S:148: Error: invalid operands `insert.w $w((0+1)+1)[2],$1'
  arch/mips/kernel/r4k_switch.S:148: Error: invalid operands `insert.w $w((0+1)+1)[3],$1'
  ...

Drop the recursion from msa_init_all_upper invoking the msa_init_upper
macro explicitly for each vector register.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9162/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

Assuming at ($1) as the source or destination register of copy or
insert instructions:

  - Simplifies the macros providing those instructions for toolchains
    without MSA support.

  - Avoids an unnecessary move instruction when at is used as the source
    or destination register anyway.

  - Is sufficient for the uses to be introduced in the kernel by a
    subsequent patch.

Note that due to a patch ordering snafu on my part this also fixes the
currently broken build with MSA support enabled. The build has been
broken since commit c9017757 "MIPS: init upper 64b of vector
registers when MSA is first used", which this patch should have
preceeded.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9161/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

The {save,restore}_fp_context{,32} functions require that the assembler
allows the use of sdc instructions on any FP register, and this is
acomplished by setting the arch to mips64r2 or mips64r6
(using MIPS_ISA_ARCH_LEVEL_RAW).

However this has the effect of enabling the assembler to use mips64
instructions in the expansion of pseudo-instructions. This was done in
the (now-reverted) commit eec43a22 "MIPS: Save/restore MSA context
around signals" which led to my mistakenly believing that there was an
assembler bug, when in reality the assembler was just emitting mips64
instructions. Avoid the issue for future commits which will add code to
r4k_fpu.S by pushing the .set MIPS_ISA_ARCH_LEVEL_RAW directives into
the functions that require it, and remove the spurious assertion
declaring the assembler bug.
Signed-off-by: Paul Burton <paul.burton@imgtec.com>
[james.hogan@imgtec.com: Rebase on v4.0-rc1 and reword commit message to
 reflect use of MIPS_ISA_ARCH_LEVEL_RAW]
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9612/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

The lose_fpu() function only disables the FPU in CP0_Status.CU1 if the
FPU is in use and MSA isn't enabled.

This isn't necessarily a problem because KSTK_STATUS(current), the
version of CP0_Status stored on the kernel stack on entry from user
mode, does always get updated and gets restored when returning to user
mode, but I don't think it was intended, and it is inconsistent with the
case of only the FPU being in use. Sometimes leaving the FPU enabled may
also mask kernel bugs where FPU operations are executed when the FPU
might not be enabled.

So lets disable the FPU in the MSA case too.

Fixes: 33c771ba ("MIPS: save/disable MSA in lose_fpu")
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul Burton <paul.burton@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9323/Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

If the guest CPU is supposed to support rdtscp and the host has rdtscp
enabled in the secondary execution controls, we can also expose this
feature to L1. Just extend nested_vmx_exit_handled to properly route
EXIT_REASON_RDTSCP.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

An overhead from function call is not appropriate for its size and
frequency of execution.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Merge tag 'kvm-s390-next-20150318' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into queue

KVM: s390: Features and fixes for 4.1 (kvm/next)

1. Fixes
2. Implement access register mode in KVM
3. Provide a userspace post handler for the STSI instruction
4. Provide an interface for compliant memory accesses
5. Provide an interface for getting/setting the guest storage key
6. Fixup for the vector facility patches: do not announce the
   vector facility in the guest for old QEMUs.

1-5 were initially shown as RFC in

http://www.spinics.net/lists/kvm/msg114720.html

some small review changes
- added some ACKs
- have the AR mode patches first
- get rid of unnecessary AR_INVAL define
- typos and language

6. two new patches
The two new patches fixup the vector support patches that were
introduced in the last pull request for QEMU versions that dont
know about vector support and guests that do. (We announce the
facility bit, but dont enable the facility so vector aware guests
will crash on vector instructions).

The following point:

    2. per-CPU pvclock time info is updated if the
       underlying CPU changes.

Is not true anymore since "KVM: x86: update pvclock area conditionally,
on cpu migration".

Add task migration notification back.

Problem noticed by Andy Lutomirski.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
CC: stable@kernel.org # 3.11+

When all bits in mask are not set,
kvm_arch_mmu_enable_log_dirty_pt_masked() has nothing to do.  But since
it needs to be called from the generic code, it cannot be inlined, and
a few function calls, two when PML is enabled, are wasted.

Since it is common to see many pages remain clean, e.g. framebuffers can
stay calm for a long time, it is worth eliminating this overhead.
Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

I hit this path on a AMD box and thought
someone was playing a April Fool's joke on me.
Signed-off-by: Bandan Das <bsd@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

This patch fix the following sparse warnings:

for arch/x86/kvm/x86.c:
warning: symbol 'emulator_read_write' was not declared. Should it be static?
warning: symbol 'emulator_write_emulated' was not declared. Should it be static?
warning: symbol 'emulator_get_dr' was not declared. Should it be static?
warning: symbol 'emulator_set_dr' was not declared. Should it be static?

for arch/x86/kvm/pmu.c:
warning: symbol 'fixed_pmc_events' was not declared. Should it be static?
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

This patch fix the following sparse warning:

for file arch/x86/kvm/x86.c:
warning: Using plain integer as NULL pointer
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

The patch represents capability KVM_CAP_S390_VECTOR_REGISTERS by means
of the SIMD facility bit. This allows to a) disable the use of SIMD when
used in conjunction with a not-SIMD-aware QEMU, b) to enable SIMD when
used with a SIMD-aware version of QEMU and c) finally by means of a QEMU
version using the future cpu model ioctls.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Eric Farman <farman@linux.vnet.ibm.com>
Tested-by: Eric Farman <farman@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

Setting the SIMD bit in the KVM mask is an issue because it makes the
facility visible but not usable to the guest, thus it needs to be
removed again.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Eric Farman <farman@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

Provide the KVM_S390_GET_SKEYS and KVM_S390_SET_SKEYS ioctl which can be used
to get/set guest storage keys. This functionality is needed for live migration
of s390 guests that use storage keys.
Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

The Store System Information (STSI) instruction currently collects all
information it relays to the caller in the kernel. Some information,
however, is only available in user space. An example of this is the
guest name: The kernel always sets "KVMGuest", but user space knows the
actual guest name.

This patch introduces a new exit, KVM_EXIT_S390_STSI, guarded by a
capability that can be enabled by user space if it wants to be able to
insert such data. User space will be provided with the target buffer
and the requested STSI function code.
Reviewed-by: Eric Farman <farman@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Ekaterina Tumanova <tumanova@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

On s390, we've got to make sure to hold the IPTE lock while accessing
logical memory. So let's add an ioctl for reading and writing logical
memory to provide this feature for userspace, too.
The maximum transfer size of this call is limited to 64kB to prevent
that the guest can trigger huge copy_from/to_user transfers. QEMU
currently only requests up to one or two pages so far, so 16*4kB seems
to be a reasonable limit here.
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

Access register mode is one of the modes that control dynamic address
translation. In this mode the address space is specified by values of
the access registers. The effective address-space-control element is
obtained from the result of the access register translation. See
the "Access-Register Introduction" section of the chapter 5 "Program
Execution" in "Principles of Operations" for more details.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Reviewed-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

During dynamic address translation the get_vcpu_asce()
function can be invoked several times. It's ok for usual modes, but will
be slow if CPUs are in AR mode. Let's call the get_vcpu_asce() once and
pass the result to the called functions.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Reviewed-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

In access register mode, the write_guest() read_guest() and other
functions will invoke the access register translation, which
requires an ar, designated by one of the instruction fields.
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Reviewed-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

The kvm_s390_check_low_addr_protection() function is used only with real
addresses. According to the POP (the "Low-Address Protection"
paragraph in chapter 3), if the effective address is real or absolute,
the low-address protection procedure should raise a PROTECTION exception
only when the low-address protection is enabled in the control register
0 and the address is low.
This patch removes ASCE checks from the function and renames it to
better reflect its behavior.

Cc: Thomas Huth <thuth@linux.vnet.ibm.com>
Signed-off-by: Alexander Yarygin <yarygin@linux.vnet.ibm.com>
Reviewed-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

As all cleanup functions can handle their respective NULL case
there is no need to have more than one error jump label.
Signed-off-by: Dominik Dingel <dingel@linux.vnet.ibm.com>
Acked-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Message-Id: <1425932832-6244-1-git-send-email-geert+renesas@glider.be>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>

While in L2, leave all #UD to L2 and do not try to emulate it. If L1 is
interested in doing this, it reports its interest via the exception
bitmap, and we never get into handle_exception of L0 anyway.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

For a very long time (since 2b3d2a20), the path handling a vmmcall
instruction of the guest on an Intel host only applied the patch but no
longer handled the hypercall. The reverse case, vmcall on AMD hosts, is
fine. As both em_vmcall and em_vmmcall actually have to do the same, we
can fix the issue by consolidating both into the same handler.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Another patch in my war on emulate_on_interception() use as a svm exit handler.

These were pulled out of a larger patch at the suggestion of Radim Krcmar, see
https://lkml.org/lkml/2015/2/25/559

Changes since v1:
	* fixed typo introduced after test, retested
Signed-off-by: David Kaplan <david.kaplan@amd.com>
[separated out just cr_interception part from larger removal of
INTERCEPT_CR0_WRITE, forward ported, tested]
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Merge tag 'kvm-s390-next-20150306' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into queue

KVM: s390: Features and Fixes for 4.1 (kvm/next)

1. Several Fixes and enhancements
---------------------------------
- These 3 patches have cc stable:
b75f4c9a KVM: s390: Zero out current VMDB of STSI before including level3 data.
261520dc KVM: s390: fix handling of write errors in the tpi handler
15462e37 KVM: s390: reinjection of irqs can fail in the tpi handler

2. SIMD support the kernel part (introduced with z13)
-----------------------------------------------------
- two KVM-generic changes in kvm.h:
1. New capability that can be enabled: KVM_CAP_S390_VECTOR_REGISTERS
2. increased padding size for sync regs in struct kvm_run to clarify that
   sync regs can be larger than 1k. This is fine as this is the last
   element in the structure.

The KVM list should be CCed on changes for arch/x86/kernel/kvm.c
and arch/x86/kernel/kvmclock.c
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

No need to re-decode WBINVD since we know what it is from the intercept.
Signed-off-by: David Kaplan <David.Kaplan@amd.com>
[extracted from larger unlrelated patch, forward ported, tested,style cleanup]
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Currently kvm_emulate() skips the instruction but kvm_emulate_* sometimes
don't.  The end reult is the caller ends up doing the skip themselves.
Let's make them consistant.
Signed-off-by: Joel Schopp <joel.schopp@amd.com>
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

WARNING: Prefer [subsystem eg: netdev]_info([subsystem]dev, ... then
dev_info(dev, ... then pr_info(...  to printk(KERN_INFO ...
+   printk(KERN_INFO "kvm: exiting hardware virtualization\n");

WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then
dev_err(dev, ... then pr_err(...  to printk(KERN_ERR ...
+	printk(KERN_ERR "kvm: misc device register failed\n");
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>