This simplifies the way that the book3s_pr makes the transition to
real mode when entering the guest. We now call kvmppc_entry_trampoline
(renamed from kvmppc_rmcall) in the base kernel using a normal function
call instead of doing an indirect call through a pointer in the vcpu.
If kvm is a module, the module loader takes care of generating a
trampoline as it does for other calls to functions outside the module.
kvmppc_entry_trampoline then disables interrupts and jumps to
kvmppc_handler_trampoline_enter in real mode using an rfi[d].
That then uses the link register as the address to return to
(potentially in module space) when the guest exits.
This also simplifies the way that we call the Linux interrupt handler
when we exit the guest due to an external, decrementer or performance
monitor interrupt. Instead of turning on the MMU, then deciding that
we need to call the Linux handler and turning the MMU back off again,
we now go straight to the handler at the point where we would turn the
MMU on. The handler will then return to the virtual-mode code
(potentially in the module).
Along the way, this moves the setting and clearing of the HID5 DCBZ32
bit into real-mode interrupts-off code, and also makes sure that
we clear the MSR[RI] bit before loading values into SRR0/1.
The net result is that we no longer need any code addresses to be
stored in vcpu->arch.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
When running a PAPR guest, we need to handle a few hypercalls in kernel space,
most prominently the page table invalidation (to sync the shadows).
So this patch adds handling for a few PAPR hypercalls to PR mode KVM. I tried
to share the code with HV mode, but it ended up being a lot easier this way
around, as the two differ too much in those details.
Signed-off-by: Alexander Graf <agraf@suse.de>
---
v1 -> v2:
- whitespace fix
Until now, we always set HIOR based on the PVR, but this is just wrong.
Instead, we should be setting HIOR explicitly, so user space can decide
what the initial HIOR value is - just like on real hardware.
We keep the old PVR based way around for backwards compatibility, but
once user space uses the SREGS based method, we drop the PVR logic.
Signed-off-by: Alexander Graf <agraf@suse.de>
We need the compute_tlbie_rb in _pr and _hv implementations for papr
soon, so let's move it over to a common header file that both
implementations can leverage.
Signed-off-by: Alexander Graf <agraf@suse.de>
Commit c8f729d408 (KVM: PPC: Deliver program interrupts right away instead
of queueing them) made away with all users of prog_flags, so we can just
remove it from the headers.
Signed-off-by: Alexander Graf <agraf@suse.de>
This adds infrastructure which will be needed to allow book3s_hv KVM to
run on older POWER processors, including PPC970, which don't support
the Virtual Real Mode Area (VRMA) facility, but only the Real Mode
Offset (RMO) facility. These processors require a physically
contiguous, aligned area of memory for each guest. When the guest does
an access in real mode (MMU off), the address is compared against a
limit value, and if it is lower, the address is ORed with an offset
value (from the Real Mode Offset Register (RMOR)) and the result becomes
the real address for the access. The size of the RMA has to be one of
a set of supported values, which usually includes 64MB, 128MB, 256MB
and some larger powers of 2.
Since we are unlikely to be able to allocate 64MB or more of physically
contiguous memory after the kernel has been running for a while, we
allocate a pool of RMAs at boot time using the bootmem allocator. The
size and number of the RMAs can be set using the kvm_rma_size=xx and
kvm_rma_count=xx kernel command line options.
KVM exports a new capability, KVM_CAP_PPC_RMA, to signal the availability
of the pool of preallocated RMAs. The capability value is 1 if the
processor can use an RMA but doesn't require one (because it supports
the VRMA facility), or 2 if the processor requires an RMA for each guest.
This adds a new ioctl, KVM_ALLOCATE_RMA, which allocates an RMA from the
pool and returns a file descriptor which can be used to map the RMA. It
also returns the size of the RMA in the argument structure.
Having an RMA means we will get multiple KMV_SET_USER_MEMORY_REGION
ioctl calls from userspace. To cope with this, we now preallocate the
kvm->arch.ram_pginfo array when the VM is created with a size sufficient
for up to 64GB of guest memory. Subsequently we will get rid of this
array and use memory associated with each memslot instead.
This moves most of the code that translates the user addresses into
host pfns (page frame numbers) out of kvmppc_prepare_vrma up one level
to kvmppc_core_prepare_memory_region. Also, instead of having to look
up the VMA for each page in order to check the page size, we now check
that the pages we get are compound pages of 16MB. However, if we are
adding memory that is mapped to an RMA, we don't bother with calling
get_user_pages_fast and instead just offset from the base pfn for the
RMA.
Typically the RMA gets added after vcpus are created, which makes it
inconvenient to have the LPCR (logical partition control register) value
in the vcpu->arch struct, since the LPCR controls whether the processor
uses RMA or VRMA for the guest. This moves the LPCR value into the
kvm->arch struct and arranges for the MER (mediated external request)
bit, which is the only bit that varies between vcpus, to be set in
assembly code when going into the guest if there is a pending external
interrupt request.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This adds support for KVM running on 64-bit Book 3S processors,
specifically POWER7, in hypervisor mode. Using hypervisor mode means
that the guest can use the processor's supervisor mode. That means
that the guest can execute privileged instructions and access privileged
registers itself without trapping to the host. This gives excellent
performance, but does mean that KVM cannot emulate a processor
architecture other than the one that the hardware implements.
This code assumes that the guest is running paravirtualized using the
PAPR (Power Architecture Platform Requirements) interface, which is the
interface that IBM's PowerVM hypervisor uses. That means that existing
Linux distributions that run on IBM pSeries machines will also run
under KVM without modification. In order to communicate the PAPR
hypercalls to qemu, this adds a new KVM_EXIT_PAPR_HCALL exit code
to include/linux/kvm.h.
Currently the choice between book3s_hv support and book3s_pr support
(i.e. the existing code, which runs the guest in user mode) has to be
made at kernel configuration time, so a given kernel binary can only
do one or the other.
This new book3s_hv code doesn't support MMIO emulation at present.
Since we are running paravirtualized guests, this isn't a serious
restriction.
With the guest running in supervisor mode, most exceptions go straight
to the guest. We will never get data or instruction storage or segment
interrupts, alignment interrupts, decrementer interrupts, program
interrupts, single-step interrupts, etc., coming to the hypervisor from
the guest. Therefore this introduces a new KVMTEST_NONHV macro for the
exception entry path so that we don't have to do the KVM test on entry
to those exception handlers.
We do however get hypervisor decrementer, hypervisor data storage,
hypervisor instruction storage, and hypervisor emulation assist
interrupts, so we have to handle those.
In hypervisor mode, real-mode accesses can access all of RAM, not just
a limited amount. Therefore we put all the guest state in the vcpu.arch
and use the shadow_vcpu in the PACA only for temporary scratch space.
We allocate the vcpu with kzalloc rather than vzalloc, and we don't use
anything in the kvmppc_vcpu_book3s struct, so we don't allocate it.
We don't have a shared page with the guest, but we still need a
kvm_vcpu_arch_shared struct to store the values of various registers,
so we include one in the vcpu_arch struct.
The POWER7 processor has a restriction that all threads in a core have
to be in the same partition. MMU-on kernel code counts as a partition
(partition 0), so we have to do a partition switch on every entry to and
exit from the guest. At present we require the host and guest to run
in single-thread mode because of this hardware restriction.
This code allocates a hashed page table for the guest and initializes
it with HPTEs for the guest's Virtual Real Memory Area (VRMA). We
require that the guest memory is allocated using 16MB huge pages, in
order to simplify the low-level memory management. This also means that
we can get away without tracking paging activity in the host for now,
since huge pages can't be paged or swapped.
This also adds a few new exports needed by the book3s_hv code.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
In preparation for adding code to enable KVM to use hypervisor mode
on 64-bit Book 3S processors, this splits book3s.c into two files,
book3s.c and book3s_pr.c, where book3s_pr.c contains the code that is
specific to running the guest in problem state (user mode) and book3s.c
contains code which should apply to all Book 3S processors.
In doing this, we abstract some details, namely the interrupt offset,
updating the interrupt pending flag, and detecting if the guest is
in a critical section. These are all things that will be different
when we use hypervisor mode.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This moves the slb field, which represents the state of the emulated
SLB, from the kvmppc_vcpu_book3s struct to the kvm_vcpu_arch, and the
hpte_hash_[v]pte[_long] fields from kvm_vcpu_arch to kvmppc_vcpu_book3s.
This is in accord with the principle that the kvm_vcpu_arch struct
represents the state of the emulated CPU, and the kvmppc_vcpu_book3s
struct holds the auxiliary data structures used in the emulation.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Up until now, Book3S KVM had variables stored in the kernel that a kernel module
or the kvm code in the kernel could read from to figure out where some real mode
helper functions are located.
This is all unnecessary. The high bits of the EA get ignore in real mode, so we
can just use the pointer as is. Also, it's a lot easier on relocations when we
use the normal way of resolving the address to a function, instead of jumping
through hoops.
This patch fixes compilation with CONFIG_RELOCATABLE=y.
Signed-off-by: Alexander Graf <agraf@suse.de>
Up until now we were doing segment mappings wrong on Book3s_32. For Book3s_64
we were using a trick where we know that a single mmu_context gives us 16 bits
of context ids.
The mm system on Book3s_32 instead uses a clever algorithm to distribute VSIDs
across the available range, so a context id really only gives us 16 available
VSIDs.
To keep at least a few guest processes in the SID shadow, let's map a number of
contexts that we can use as VSID pool. This makes the code be actually correct
and shouldn't hurt performance too much.
Signed-off-by: Alexander Graf <agraf@suse.de>
Now that the actual mtsr doesn't do anything anymore, we can move the sr
contents over to the shared page, so a guest can directly read and write
its sr contents from guest context.
Signed-off-by: Alexander Graf <agraf@suse.de>
Right now we're examining the contents of Book3s_32's segment registers when
the register is written and put the interpreted contents into a struct.
There are two reasons this is bad. For starters, the struct has worse real-time
performance, as it occupies more ram. But the more important part is that with
segment registers being interpreted from their raw values, we can put them in
the shared page, allowing guests to mess with them directly.
This patch makes the internal representation of SRs be u32s.
Signed-off-by: Alexander Graf <agraf@suse.de>
On Book3S KVM we directly expose some asm pointers to C code as
variables. These need to be relocated and thus break on relocatable
kernels.
To make sure we can at least build, let's mark them as long instead
of u32 where 64bit relocations don't work.
This fixes the following build error:
WARNING: 2 bad relocations^M
> c000000000008590 R_PPC64_ADDR32 .text+0x4000000000008460^M
> c000000000008594 R_PPC64_ADDR32 .text+0x4000000000008598^M
Please keep in mind that actually using KVM on a relocated kernel
might still break. This only fixes the compile problem.
Reported-by: Subrata Modak <subrata@linux.vnet.ibm.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
We need to override EA as well as PA lookups for the magic page. When the guest
tells us to project it, the magic page overrides any guest mappings.
In order to reflect that, we need to hook into all the MMU layers of KVM to
force map the magic page if necessary.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
The DSISR register contains information about a data page fault. It is fully
read/write from inside the guest context and we don't need to worry about
interacting based on writes of this register.
This patch converts all users of the current field to the shared page.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
We just introduced generic functions to handle shadow pages on PPC.
This patch makes the respective backends make use of them, getting
rid of a lot of duplicate code along the way.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Initially we had to search for pte entries to invalidate them. Since
the logic has improved since then, we can just get rid of the search
function.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
When in split mode, instruction relocation and data relocation are not equal.
So far we implemented this mode by reserving a special pseudo-VSID for the
two cases and flushing all PTEs when going into split mode, which is slow.
Unfortunately 32bit Linux and Mac OS X use split mode extensively. So to not
slow down things too much, I came up with a different idea: Mark the split
mode with a bit in the VSID and then treat it like any other segment.
This means we can just flush the shadow segment cache, but keep the PTEs
intact. I verified that this works with ppc32 Linux and Mac OS X 10.4
guests and does speed them up.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
There are some pieces in the code that I overlooked that still use
u64s instead of longs. This slows down 32 bit hosts unnecessarily, so
let's just move them to ulong.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
We already have some inline fuctions we use to access vcpu or svcpu structs,
depending on whether we're on booke or book3s. Since we just put a few more
registers into the svcpu, we also need to make sure the respective callbacks
are available and get used.
So this patch moves direct use of the now in the svcpu struct fields to
inline function calls. While at it, it also moves the definition of those
inline function calls to respective header files for booke and book3s,
greatly improving readability.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
We have quite some code that can be used by Book3S_32 and Book3S_64 alike,
so let's call it "Book3S" instead of "Book3S_64", so we can later on
use it from the 32 bit port too.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Bool defaults to at least byte width. We usually only want to waste a single
bit on this. So let's move all the bool values to bitfields, potentially
saving memory.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Some constants were bigger than ints. Let's mark them as such so we don't
accidently truncate them.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
MOL uses its own hypercall interface to call back into userspace when
the guest wants to do something.
So let's implement that as an exit reason, specify it with a CAP and
only really use it when userspace wants us to.
The only user of it so far is MOL.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Mac OS X has some applications - namely the Finder - that require alignment
interrupts to work properly. So we need to implement them.
But the spec for 970 and 750 also looks different. While 750 requires the
DSISR and DAR fields to reflect some instruction bits (DSISR) and the fault
address (DAR), the 970 declares this as an optional feature. So we need
to reconstruct DSISR and DAR manually.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
This patch makes the VSID of mapped pages always reflecting all special cases
we have, like split mode.
It also changes the tlbie mask to 0x0ffff000 according to the spec. The mask
we used before was incorrect.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
DSISR is only defined as 32 bits wide. So let's reflect that in the
structs too.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
On PowerPC we can go into MMU Split Mode. That means that either
data relocation is on but instruction relocation is off or vice
versa.
That mode didn't work properly, as we weren't always flushing
entries when going into a new split mode, potentially mapping
different code or data that we're supposed to.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
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>
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>
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>
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>
Commit 7d01b4c3ed2bb33ceaf2d270cb4831a67a76b51b introduced PACA backed vcpu
values. With this patch, when a userspace app was setting GPRs before it was
actually first loaded, the set values get discarded.
This is because vcpu_load loads them from the vcpu backing store that we use
whenever we're not owning the PACA.
That behavior is not really a major problem, because we don't need it for
qemu. Other users (like kvmctl) do have problems with it though, so let's
better do it right.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Linux contains quite some bits of code to load FPU, Altivec and VSX lazily for
a task. It calls those bits in real mode, coming from an interrupt handler.
For KVM we better reuse those, so let's wrap a bit of trampoline magic around
them and then we can call them from normal module code.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
An SLB entry contains two pieces of information related to size:
1) PTE size
2) SLB size
The L bit defines the PTE be "large" (usually means 16MB),
SLB_VSID_B_1T defines that the SLB should span 1 GB instead of the
default 256MB.
Apparently I messed things up and just put those two in one box,
shaked it heavily and came up with the current code which handles
large pages incorrectly, because it also treats large page SLB entries
as "1TB" segment entries.
This patch splits those two features apart, making Linux guests boot
even when they have > 256MB.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Book3S needs some flags in SRR1 to get to know details about an interrupt.
One such example is the trap instruction. It tells the guest kernel that
a program interrupt is due to a trap using a bit in SRR1.
This patch implements above behavior, making WARN_ON behave like WARN_ON.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently we're racy when doing the transition from IR=1 to IR=0, from
the module memory entry code to the real mode SLB switching code.
To work around that I took a look at the RTAS entry code which is faced
with a similar problem and did the same thing:
A small helper in linear mapped memory that does mtmsr with IR=0 and
then RFIs info the actual handler.
Thanks to that trick we can safely take page faults in the entry code
and only need to be really wary of what to do as of the SLB switching
part.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
We're being horribly racy right now. All the entry and exit code hijacks
random fields from the PACA that could easily be used by different code in
case we get interrupted, for example by a #MC or even page fault.
After discussing this with Ben, we figured it's best to reserve some more
space in the PACA and just shove off some vcpu state to there.
That way we can drastically improve the readability of the code, make it
less racy and less complex.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Avi Kivity <avi@redhat.com>
Currently userspace has no chance to find out which virtual address space we're
in and resolve addresses. While that is a big problem for migration, it's also
unpleasent when debugging, as gdb and the monitor don't work on virtual
addresses.
This patch exports enough of the MMU segment state to userspace to make
debugging work and thus also includes the groundwork for migration.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This adds the book3s specific header file that contains structs that
are only valid on book3s specific code.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Paul Mackerras