Commit Graph

190172 Commits

Author SHA1 Message Date
Joerg Roedel
455716fa94 KVM: SVM: Move msrpm offset calculation to seperate function
The algorithm to find the offset in the msrpm for a given
msr is needed at other places too. Move that logic to its
own function.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-05-17 12:15:08 +03:00
Joerg Roedel
d24778265a KVM: SVM: Return correct values in nested_svm_exit_handled_msr
The nested_svm_exit_handled_msr() returned an bool which is
a bug. I worked by accident because the exected integer
return values match with the true and false values. This
patch changes the return value to int and let the function
return the correct values.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-05-17 12:15:07 +03:00
Andrea Gelmini
0fc5c3a54d KVM: arch/x86/kvm/kvm_timer.h checkpatch cleanup
arch/x86/kvm/kvm_timer.h:13: ERROR: code indent should use tabs where possible

Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2010-05-17 12:14:42 +03:00
Gleb Natapov
ea79849d4c KVM: x86 emulator: Implement jmp far opcode ff/5
Implement jmp far opcode ff/5. It is used by multiboot loader.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:45 +03:00
Gleb Natapov
e35b7b9c9e KVM: x86 emulator: Add decoding of 16bit second in memory argument
Add decoding of Ep type of argument used by callf/jmpf.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:42 +03:00
Gleb Natapov
2d49ec72d3 KVM: move segment_base() into vmx.c
segment_base() is used only by vmx so move it there.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:39 +03:00
Gleb Natapov
254d4d48a5 KVM: fix segment_base() error checking
fix segment_base() to properly check for null segment selector and
avoid accessing NULL pointer if ldt selector in null.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:35 +03:00
Gleb Natapov
d6ab1ed446 KVM: Drop kvm_get_gdt() in favor of generic linux function
Linux now has native_store_gdt() to do the same. Use it instead of
kvm local version.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:32 +03:00
Takuya Yoshikawa
f5c9803173 KVM: update gfn_to_hva() to use gfn_to_hva_memslot()
Marcelo introduced gfn_to_hva_memslot() when he implemented
gfn_to_pfn_memslot(). Let's use this for gfn_to_hva() too.

Note: also remove parentheses next to return as checkpatch said to do.

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:29 +03:00
Joerg Roedel
197717d581 KVM: SVM: Clear exit_info for injected INTR exits
When injecting an vmexit.intr into the nested hypervisor
there might be leftover values in the exit_info fields.
Clear them to not confuse nested hypervisors.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:26 +03:00
Joerg Roedel
7f5d8b5600 KVM: SVM: Handle nested selective_cr0 intercept correctly
If we have the following situation with nested svm:

1. Host KVM intercepts cr0 writes
2. Guest hypervisor intercepts only selective cr0 writes

Then we get an cr0 write intercept which is handled on the
host. But that intercepts may actually be a selective cr0
intercept for the guest. This patch checks for this
condition and injects a selective cr0 intercept if needed.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:23 +03:00
Joerg Roedel
b44ea385d8 KVM: x86: Don't set arch.cr0 in kvm_set_cr0
The vcpu->arch.cr0 variable is already set in the
architecture specific set_cr0 callbacks. There is no need to
set it in the common code.
This allows the architecture code to keep the old arch.cr0
value if it wants. This is required for nested svm to decide
if a selective_cr0 exit needs to be injected.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:20 +03:00
Joerg Roedel
82494028df KVM: SVM: Ignore write of hwcr.ignne
Hyper-V as a guest wants to write this bit. This patch
ignores it.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:17 +03:00
Joerg Roedel
4a810181c8 KVM: SVM: Implement emulation of vm_cr msr
This patch implements the emulation of the vm_cr msr for
nested svm.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:14 +03:00
Joerg Roedel
2e554e8d67 KVM: SVM: Add kvm_nested_intercepts tracepoint
This patch adds a tracepoint to get information about the
most important intercept bitmasks from the nested vmcb.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:10 +03:00
Joerg Roedel
ecf1405df2 KVM: SVM: Restore tracing of nested vmcb address
A recent change broke tracing of the nested vmcb address. It
was reported as 0 all the time. This patch fixes it.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:07 +03:00
Joerg Roedel
887f500ca1 KVM: SVM: Check for nested intercepts on NMI injection
This patch implements the NMI intercept checking for nested
svm.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:04 +03:00
Joerg Roedel
0e5cbe368b KVM: SVM: Reset MMU on nested_svm_vmrun for NPT too
Without resetting the MMU the gva_to_pga function will not
work reliably when the vcpu is running in nested context.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:53:01 +03:00
Joerg Roedel
e02317153e KVM: SVM: Coding style cleanup
This patch removes whitespace errors, fixes comment formats
and most of checkpatch warnings. Now vim does not show
c-space-errors anymore.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:52:58 +03:00
Jan Kiszka
83bf0002c9 KVM: x86: Preserve injected TF across emulation
Call directly into the vendor services for getting/setting rflags in
emulate_instruction to ensure injected TF survives the emulation.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:52:55 +03:00
Jan Kiszka
c310bac5a2 KVM: x86: Drop RF manipulation for guest single-stepping
RF is not required for injecting TF as the latter will trigger only
after an instruction execution anyway. So do not touch RF when arming or
disarming guest single-step mode.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:52:51 +03:00
Jan Kiszka
66b7138f91 KVM: SVM: Emulate nRIP feature when reinjecting INT3
When in guest debugging mode, we have to reinject those #BP software
exceptions that are caused by guest-injected INT3. As older AMD
processors do not support the required nRIP VMCB field, try to emulate
it by moving RIP past the instruction on exception injection. Fix it up
again in case the injection failed and we were able to catch this. This
does not work for unintercepted faults, but it is better than doing
nothing.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:00:43 +03:00
Jan Kiszka
f92653eeb4 KVM: x86: Add kvm_is_linear_rip
Based on Gleb's suggestion: Add a helper kvm_is_linear_rip that matches
a given linear RIP against the current one. Use this for guest
single-stepping, more users will follow.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:00:40 +03:00
Jan Kiszka
116a4752c8 KVM: SVM: Move svm_queue_exception
Move svm_queue_exception past skip_emulated_instruction to allow calling
it later on.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 13:00:37 +03:00
Jan Kiszka
50a085bdd4 KVM: x86: Kick VCPU outside PIC lock again
This restores the deferred VCPU kicking before 956f97cf. We need this
over -rt as wake_up* requires non-atomic context in this configuration.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:39:28 +03:00
Alexander Graf
a595405df9 KVM: PPC: Destory timer on vcpu destruction
When we destory a vcpu, we should also make sure to kill all pending
timers that could still be up. When not doing this, hrtimers might
dereference null pointers trying to call our code.

This patch fixes spontanious kernel panics seen after closing VMs.

Signed-off-by: Alexander Graf <alex@csgraf.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:39:25 +03:00
Alexander Graf
7e821d3920 KVM: PPC: Memset vcpu to zeros
While converting the kzalloc we used to allocate our vcpu struct to
vmalloc, I forgot to memset the contents to zeros. That broke quite
a lot.

This patch memsets it to zero again.

Signed-off-by: Alexander Graf <alex@csgraf.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:39:21 +03:00
Jan Kiszka
a1efbe77c1 KVM: x86: Add support for saving&restoring debug registers
So far user space was not able to save and restore debug registers for
migration or after reset. Plug this hole.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:39:10 +03:00
Jan Kiszka
48005f64d0 KVM: x86: Save&restore interrupt shadow mask
The interrupt shadow created by STI or MOV-SS-like operations is part of
the VCPU state and must be preserved across migration. Transfer it in
the spare padding field of kvm_vcpu_events.interrupt.

As a side effect we now have to make vmx_set_interrupt_shadow robust
against both shadow types being set. Give MOV SS a higher priority and
skip STI in that case to avoid that VMX throws a fault on next entry.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:28 +03:00
Jan Kiszka
03b82a30ea KVM: x86: Do not return soft events in vcpu_events
To avoid that user space migrates a pending software exception or
interrupt, mask them out on KVM_GET_VCPU_EVENTS. Without this, user
space would try to reinject them, and we would have to reconstruct the
proper instruction length for VMX event injection. Now the pending event
will be reinjected via executing the triggering instruction again.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:14 +03:00
Joerg Roedel
8fe546547c KVM: SVM: Fix wrong interrupt injection in enable_irq_windows
The nested_svm_intr() function does not execute the vmexit
anymore. Therefore we may still be in the nested state after
that function ran. This patch changes the nested_svm_intr()
function to return wether the irq window could be enabled.

Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:11 +03:00
Gleb Natapov
112592da0d KVM: drop unneeded kvm_run check in emulate_instruction()
vcpu->run is initialized on vcpu creation and can never be NULL
here.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:08 +03:00
Alexander Graf
032c340731 KVM: PPC: Allocate vcpu struct using vmalloc
We used to use get_free_pages to allocate our vcpu struct. Unfortunately
that call failed on me several times after my machine had a big enough
uptime, as memory became too fragmented by then.

Fortunately, we don't need it to be page aligned any more! We can just
vmalloc it and everything's great.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:04 +03:00
Alexander Graf
964b6411af KVM: PPC: Simplify kvmppc_load_up_(FPU|VMX|VSX)
We don't need as complex code. I had some thinkos while writing it, figuring
I needed to support PPC32 paths on PPC64 which would have required DR=0, but
everything just runs fine with DR=1.

So let's make the functions simple C call wrappers that reserve some space on
the stack for the respective functions to clobber.

Fixes out-of-RMA-access (and thus guest FPU loading) on the PS3.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:38:01 +03:00
Alexander Graf
20a340abd3 KVM: PPC: Enable use of secondary htab bucket
We had code to make use of the secondary htab buckets, but kept that
disabled because it was unstable when I put it in.

I checked again if that's still the case and apparently it was only
exposing some instability that was there anyways before. I haven't
seen any badness related to usage of secondary htab entries so far.

This should speed up guest memory allocations by quite a bit, because
we now have more space to put PTEs in.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:37:58 +03:00
Alexander Graf
c10207fe86 KVM: PPC: Add capability for paired singles
We need to tell userspace that we can emulate paired single instructions.
So let's add a capability export.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:37:47 +03:00
Alexander Graf
831317b605 KVM: PPC: Implement Paired Single emulation
The one big thing about the Gekko is paired singles.

Paired singles are an extension to the instruction set, that adds 32 single
precision floating point registers (qprs), some SPRs to modify the behavior
of paired singled operations and instructions to deal with qprs to the
instruction set.

Unfortunately, it also changes semantics of existing operations that affect
single values in FPRs. In most cases they get mirrored to the coresponding
QPR.

Thanks to that we need to emulate all FPU operations and all the new paired
single operations too.

In order to achieve that, we use the just introduced FPU call helpers to
call the real FPU whenever the guest wants to modify an FPR. Additionally
we also fix up the QPR values along the way.

That way we can execute paired single FPU operations without implementing a
soft fpu.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:27 +03:00
Alexander Graf
e5c29e926c KVM: PPC: Enable program interrupt to do MMIO
When we get a program interrupt we usually don't expect it to perform an
MMIO operation. But why not? When we emulate paired singles, we can end
up loading or storing to an MMIO address - and the handling of those
happens in the program interrupt handler.

So let's teach the program interrupt handler how to deal with EMULATE_MMIO.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:24 +03:00
Alexander Graf
0564ee8a86 KVM: PPC: Add helpers to modify ppc fields
The PowerPC specification always lists bits from MSB to LSB. That is
really confusing when you're trying to write C code, because it fits
in pretty badly with the normal (1 << xx) schemes.

So I came up with some nice wrappers that allow to get and set fields
in a u64 with bit numbers exactly as given in the spec. That makes the
code in KVM and the spec easier comparable.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:21 +03:00
Alexander Graf
dba2e123e7 KVM: PPC: Fix error in BAT assignment
BATs didn't work. Well, they did, but only up to BAT3. As soon as we
came to BAT4 the offset calculation was screwed up and we ended up
overwriting BAT0-3.

Fortunately, Linux hasn't been using BAT4+. It's still a good
idea to write correct code though.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:18 +03:00
Alexander Graf
963cf3dc63 KVM: PPC: Add helpers to call FPU instructions
To emulate paired single instructions, we need to be able to call FPU
operations from within the kernel. Since we don't want gcc to spill
arbitrary FPU code everywhere, we tell it to use a soft fpu.

Since we know we can really call the FPU in safe areas, let's also add
some calls that we can later use to actually execute real world FPU
operations on the host's FPU.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:15 +03:00
Alexander Graf
aba3bd7ffe KVM: PPC: Make ext giveup non-static
We need to call the ext giveup handlers from code outside of book3s.c.
So let's make it non-static.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:12 +03:00
Alexander Graf
5467a97d0f KVM: PPC: Make software load/store return eaddr
The Book3S KVM implementation contains some helper functions to load and store
data from and to virtual addresses.

Unfortunately, this helper used to keep the physical address it so nicely
found out for us to itself. So let's change that and make it return the
physical address it resolved.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:09 +03:00
Alexander Graf
71db408936 KVM: PPC: Implement mtsr instruction emulation
The Book3S_32 specifications allows for two instructions to modify segment
registers: mtsrin and mtsr.

Most normal operating systems use mtsrin, because it allows to define which
segment it wants to change using a register. But since I was trying to run
an embedded guest, it turned out to be using mtsr with hardcoded values.

So let's also emulate mtsr. It's a valid instruction after all.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:06 +03:00
Alexander Graf
e425a6de1a KVM: PPC: Fix typo in book3s_32 debug code
There's a typo in the debug ifdef of the book3s_32 mmu emulation. While trying
to debug something I stumbled across that and wanted to save anyone after me
(or myself later) from having to debug that again.

So let's fix the ifdef.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:35:03 +03:00
Alexander Graf
d1bab74c51 KVM: PPC: Preload FPU when possible
There are some situations when we're pretty sure the guest will use the
FPU soon. So we can save the churn of going into the guest, finding out
it does want to use the FPU and going out again.

This patch adds preloading of the FPU when it's reasonable.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:34:59 +03:00
Alexander Graf
c8c0b6f2f7 KVM: PPC: Combine extension interrupt handlers
When we for example get an Altivec interrupt, but our guest doesn't support
altivec, we need to inject a program interrupt, not an altivec interrupt.

The same goes for paired singles. When an altivec interrupt arrives, we're
pretty sure we need to emulate the instruction because it's a paired single
operation.

So let's make all the ext handlers aware that they need to jump to the
program interrupt handler when an extension interrupt arrives that
was not supposed to arrive for the guest CPU.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:34:56 +03:00
Alexander Graf
d6d549b207 KVM: PPC: Add Gekko SPRs
The Gekko has some SPR values that differ from other PPC core values and
also some additional ones.

Let's add support for them in our mfspr/mtspr emulator.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:34:53 +03:00
Alexander Graf
3c402a75ea KVM: PPC: Add hidden flag for paired singles
The Gekko implements an extension called paired singles. When the guest wants
to use that extension, we need to make sure we're not running the host FPU,
because all FPU instructions need to get emulated to accomodate for additional
operations that occur.

This patch adds an hflag to track if we're in paired single mode or not.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:34:50 +03:00
Alexander Graf
37f5bca64e KVM: PPC: Add AGAIN type for emulation return
Emulation of an instruction can have different outcomes. It can succeed,
fail, require MMIO, do funky BookE stuff - or it can just realize something's
odd and will be fixed the next time around.

Exactly that is what EMULATE_AGAIN means. Using that flag we can now tell
the caller that nothing happened, but we still want to go back to the
guest and see what happens next time we come around.

Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
2010-04-25 12:34:47 +03:00