This adds the code to interpret 64k HPTEs in the guest hashed page
table (HPT), 64k SLB entries, and to tell the guest about 64k pages
in kvm_vm_ioctl_get_smmu_info(). Guest 64k pages are still shadowed
by 4k pages.
This also adds another hash table to the four we have already in
book3s_mmu_hpte.c to allow us to find all the PTEs that we have
instantiated that match a given 64k guest page.
The tlbie instruction changed starting with POWER6 to use a bit in
the RB operand to indicate large page invalidations, and to use other
RB bits to indicate the base and actual page sizes and the segment
size. 64k pages came in slightly earlier, with POWER5++.
We use one bit in vcpu->arch.hflags to indicate that the emulated
cpu supports 64k pages, and another to indicate that it has the new
tlbie definition.
The KVM_PPC_GET_SMMU_INFO ioctl presents a bit of a problem, because
the MMU capabilities depend on which CPU model we're emulating, but it
is a VM ioctl not a VCPU ioctl and therefore doesn't get passed a VCPU
fd. In addition, commonly-used userspace (QEMU) calls it before
setting the PVR for any VCPU. Therefore, as a best effort we look at
the first vcpu in the VM and return 64k pages or not depending on its
capabilities. We also make the PVR default to the host PVR on recent
CPUs that support 1TB segments (and therefore multiple page sizes as
well) so that KVM_PPC_GET_SMMU_INFO will include 64k page and 1TB
segment support on those CPUs.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Currently PR-style KVM keeps the volatile guest register values
(R0 - R13, CR, LR, CTR, XER, PC) in a shadow_vcpu struct rather than
the main kvm_vcpu struct. For 64-bit, the shadow_vcpu exists in two
places, a kmalloc'd struct and in the PACA, and it gets copied back
and forth in kvmppc_core_vcpu_load/put(), because the real-mode code
can't rely on being able to access the kmalloc'd struct.
This changes the code to copy the volatile values into the shadow_vcpu
as one of the last things done before entering the guest. Similarly
the values are copied back out of the shadow_vcpu to the kvm_vcpu
immediately after exiting the guest. We arrange for interrupts to be
still disabled at this point so that we can't get preempted on 64-bit
and end up copying values from the wrong PACA.
This means that the accessor functions in kvm_book3s.h for these
registers are greatly simplified, and are same between PR and HV KVM.
In places where accesses to shadow_vcpu fields are now replaced by
accesses to the kvm_vcpu, we can also remove the svcpu_get/put pairs.
Finally, on 64-bit, we don't need the kmalloc'd struct at all any more.
With this, the time to read the PVR one million times in a loop went
from 567.7ms to 575.5ms (averages of 6 values), an increase of about
1.4% for this worse-case test for guest entries and exits. The
standard deviation of the measurements is about 11ms, so the
difference is only marginally significant statistically.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Commit 9d1ffdd8f3 ("KVM: PPC: Book3S PR: Don't corrupt guest state
when kernel uses VMX") added a call to kvmppc_load_up_altivec() that
isn't guarded by CONFIG_ALTIVEC, causing a link failure when building
a kernel without CONFIG_ALTIVEC set. This adds an #ifdef to fix this.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
If we come out of a guest with an interrupt that we don't know about,
instead of crashing the host with a BUG(), we now return to userspace
with the exit reason set to KVM_EXIT_UNKNOWN and the trap vector in
the hw.hardware_exit_reason field of the kvm_run structure, as is done
on x86. Note that run->exit_reason is already set to KVM_EXIT_UNKNOWN
at the beginning of kvmppc_handle_exit().
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This enables us to use the Processor Compatibility Register (PCR) on
POWER7 to put the processor into architecture 2.05 compatibility mode
when running a guest. In this mode the new instructions and registers
that were introduced on POWER7 are disabled in user mode. This
includes all the VSX facilities plus several other instructions such
as ldbrx, stdbrx, popcntw, popcntd, etc.
To select this mode, we have a new register accessible through the
set/get_one_reg interface, called KVM_REG_PPC_ARCH_COMPAT. Setting
this to zero gives the full set of capabilities of the processor.
Setting it to one of the "logical" PVR values defined in PAPR puts
the vcpu into the compatibility mode for the corresponding
architecture level. The supported values are:
0x0f000002 Architecture 2.05 (POWER6)
0x0f000003 Architecture 2.06 (POWER7)
0x0f100003 Architecture 2.06+ (POWER7+)
Since the PCR is per-core, the architecture compatibility level and
the corresponding PCR value are stored in the struct kvmppc_vcore, and
are therefore shared between all vcpus in a virtual core.
Signed-off-by: Paul Mackerras <paulus@samba.org>
[agraf: squash in fix to add missing break statements and documentation]
Signed-off-by: Alexander Graf <agraf@suse.de>
POWER7 and later IBM server processors have a register called the
Program Priority Register (PPR), which controls the priority of
each hardware CPU SMT thread, and affects how fast it runs compared
to other SMT threads. This priority can be controlled by writing to
the PPR or by use of a set of instructions of the form or rN,rN,rN
which are otherwise no-ops but have been defined to set the priority
to particular levels.
This adds code to context switch the PPR when entering and exiting
guests and to make the PPR value accessible through the SET/GET_ONE_REG
interface. When entering the guest, we set the PPR as late as
possible, because if we are setting a low thread priority it will
make the code run slowly from that point on. Similarly, the
first-level interrupt handlers save the PPR value in the PACA very
early on, and set the thread priority to the medium level, so that
the interrupt handling code runs at a reasonable speed.
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This adds the ability to have a separate LPCR (Logical Partitioning
Control Register) value relating to a guest for each virtual core,
rather than only having a single value for the whole VM. This
corresponds to what real POWER hardware does, where there is a LPCR
per CPU thread but most of the fields are required to have the same
value on all active threads in a core.
The per-virtual-core LPCR can be read and written using the
GET/SET_ONE_REG interface. Userspace can can only modify the
following fields of the LPCR value:
DPFD Default prefetch depth
ILE Interrupt little-endian
TC Translation control (secondary HPT hash group search disable)
We still maintain a per-VM default LPCR value in kvm->arch.lpcr, which
contains bits relating to memory management, i.e. the Virtualized
Partition Memory (VPM) bits and the bits relating to guest real mode.
When this default value is updated, the update needs to be propagated
to the per-vcore values, so we add a kvmppc_update_lpcr() helper to do
that.
Signed-off-by: Paul Mackerras <paulus@samba.org>
[agraf: fix whitespace]
Signed-off-by: Alexander Graf <agraf@suse.de>
This makes the VRSAVE register value for a vcpu accessible through
the GET/SET_ONE_REG interface on Book E systems (in addition to the
existing GET/SET_SREGS interface), for consistency with Book 3S.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
The yield count in the VPA is supposed to be incremented every time
we enter the guest, and every time we exit the guest, so that its
value is even when the vcpu is running in the guest and odd when it
isn't. However, it's currently possible that we increment the yield
count on the way into the guest but then find that other CPU threads
are already exiting the guest, so we go back to nap mode via the
secondary_too_late label. In this situation we don't increment the
yield count again, breaking the relationship between the LSB of the
count and whether the vcpu is in the guest.
To fix this, we move the increment of the yield count to a point
after we have checked whether other CPU threads are exiting.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This moves the code in book3s_hv_rmhandlers.S that reads any pending
interrupt from the XICS interrupt controller, and works out whether
it is an IPI for the guest, an IPI for the host, or a device interrupt,
into a new function called kvmppc_read_intr. Later patches will
need this.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
We have two paths into and out of the low-level guest entry and exit
code: from a vcpu task via kvmppc_hv_entry_trampoline, and from the
system reset vector for an offline secondary thread on POWER7 via
kvm_start_guest. Currently both just branch to kvmppc_hv_entry to
enter the guest, and on guest exit, we test the vcpu physical thread
ID to detect which way we came in and thus whether we should return
to the vcpu task or go back to nap mode.
In order to make the code flow clearer, and to keep the code relating
to each flow together, this turns kvmppc_hv_entry into a subroutine
that follows the normal conventions for call and return. This means
that kvmppc_hv_entry_trampoline() and kvmppc_hv_entry() now establish
normal stack frames, and we use the normal stack slots for saving
return addresses rather than local_paca->kvm_hstate.vmhandler. Apart
from that this is mostly moving code around unchanged.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
The H_CONFER hypercall is used when a guest vcpu is spinning on a lock
held by another vcpu which has been preempted, and the spinning vcpu
wishes to give its timeslice to the lock holder. We implement this
in the straightforward way using kvm_vcpu_yield_to().
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
The VRSAVE register value for a vcpu is accessible through the
GET/SET_SREGS interface for Book E processors, but not for Book 3S
processors. In order to make this accessible for Book 3S processors,
this adds a new register identifier for GET/SET_ONE_REG, and adds
the code to implement it.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This allows guests to have a different timebase origin from the host.
This is needed for migration, where a guest can migrate from one host
to another and the two hosts might have a different timebase origin.
However, the timebase seen by the guest must not go backwards, and
should go forwards only by a small amount corresponding to the time
taken for the migration.
Therefore this provides a new per-vcpu value accessed via the one_reg
interface using the new KVM_REG_PPC_TB_OFFSET identifier. This value
defaults to 0 and is not modified by KVM. On entering the guest, this
value is added onto the timebase, and on exiting the guest, it is
subtracted from the timebase.
This is only supported for recent POWER hardware which has the TBU40
(timebase upper 40 bits) register. Writing to the TBU40 register only
alters the upper 40 bits of the timebase, leaving the lower 24 bits
unchanged. This provides a way to modify the timebase for guest
migration without disturbing the synchronization of the timebase
registers across CPU cores. The kernel rounds up the value given
to a multiple of 2^24.
Timebase values stored in KVM structures (struct kvm_vcpu, struct
kvmppc_vcore, etc.) are stored as host timebase values. The timebase
values in the dispatch trace log need to be guest timebase values,
however, since that is read directly by the guest. This moves the
setting of vcpu->arch.dec_expires on guest exit to a point after we
have restored the host timebase so that vcpu->arch.dec_expires is a
host timebase value.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Currently we are not saving and restoring the SIAR and SDAR registers in
the PMU (performance monitor unit) on guest entry and exit. The result
is that performance monitoring tools in the guest could get false
information about where a program was executing and what data it was
accessing at the time of a performance monitor interrupt. This fixes
it by saving and restoring these registers along with the other PMU
registers on guest entry/exit.
This also provides a way for userspace to access these values for a
vcpu via the one_reg interface.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
This reserves space in get/set_one_reg ioctl for the extra guest state
needed for POWER8. It doesn't implement these at all, it just reserves
them so that the ABI is defined now.
A few things to note here:
- This add *a lot* state for transactional memory. TM suspend mode,
this is unavoidable, you can't simply roll back all transactions and
store only the checkpointed state. I've added this all to
get/set_one_reg (including GPRs) rather than creating a new ioctl
which returns a struct kvm_regs like KVM_GET_REGS does. This means we
if we need to extract the TM state, we are going to need a bucket load
of IOCTLs. Hopefully most of the time this will not be needed as we
can look at the MSR to see if TM is active and only grab them when
needed. If this becomes a bottle neck in future we can add another
ioctl to grab all this state in one go.
- The TM state is offset by 0x80000000.
- For TM, I've done away with VMX and FP and created a single 64x128 bit
VSX register space.
- I've left a space of 1 (at 0x9c) since Paulus needs to add a value
which applies to POWER7 as well.
Signed-off-by: Michael Neuling <mikey@neuling.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Reserved fields of the sync instruction have been used for other
instructions (e.g. lwsync). On processors that do not support variants
of the sync instruction, emulate it by executing a sync to subsume the
effect of the intended instruction.
Signed-off-by: James Yang <James.Yang@freescale.com>
[scottwood@freescale.com: whitespace and subject line fix]
Signed-off-by: Scott Wood <scottwood@freescale.com>
On Book3E some SPE/FP/AltiVec interrupts share the same number. Use
common defines to indentify these numbers.
Signed-off-by: Mihai Caraman <mihai.caraman@freescale.com>
[scottwood@freescale.com: fixed space-before-tab]
Signed-off-by: Scott Wood <scottwood@freescale.com>
On Book3E some SPE/FP/AltiVec interrupts share the same number. Use
common defines to indentify these numbers.
Signed-off-by: Mihai Caraman <mihai.caraman@freescale.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
The Kconfig entry that allows to "Distribute interrupts on all CPUs by
default" has a (negative) dependency on MV64360. But that Kconfig symbol
was removed in v2.6.27, which means that this dependency has evaluated
to true ever since. It can be removed too.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Scott Wood <scottwood@freescale.com>
Now when the main kvm code relying on these defines has been moved to
the x86 specific part of the world, we can get rid of these.
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Gleb Natapov <gleb@redhat.com>
Topic branch for commits that the KVM tree might want to pull
in separately.
Hand merged a few files due to conflicts with the LE stuff
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This provides a facility which is intended for use by KVM, where the
contents of the FP/VSX and VMX (Altivec) registers can be saved away
to somewhere other than the thread_struct when kernel code wants to
use floating point or VMX instructions. This is done by providing a
pointer in the thread_struct to indicate where the state should be
saved to. The giveup_fpu() and giveup_altivec() functions test these
pointers and save state to the indicated location if they are non-NULL.
Note that the MSR_FP/VEC bits in task->thread.regs->msr are still used
to indicate whether the CPU register state is live, even when an
alternate save location is being used.
This also provides load_fp_state() and load_vr_state() functions, which
load up FP/VSX and VMX state from memory into the CPU registers, and
corresponding store_fp_state() and store_vr_state() functions, which
store FP/VSX and VMX state into memory from the CPU registers.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This creates new 'thread_fp_state' and 'thread_vr_state' structures
to store FP/VSX state (including FPSCR) and Altivec/VSX state
(including VSCR), and uses them in the thread_struct. In the
thread_fp_state, the FPRs and VSRs are represented as u64 rather
than double, since we rarely perform floating-point computations
on the values, and this will enable the structures to be used
in KVM code as well. Similarly FPSCR is now a u64 rather than
a structure of two 32-bit values.
This takes the offsets out of the macros such as SAVE_32FPRS,
REST_32FPRS, etc. This enables the same macros to be used for normal
and transactional state, enabling us to delete the transactional
versions of the macros. This also removes the unused do_load_up_fpu
and do_load_up_altivec, which were in fact buggy since they didn't
create large enough stack frames to account for the fact that
load_up_fpu and load_up_altivec are not designed to be called from C
and assume that their caller's stack frame is an interrupt frame.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The existing TCE machine calls (tce_build and tce_free) only support
virtual mode as they call __raw_writeq for TCE invalidation what
fails in real mode.
This introduces tce_build_rm and tce_free_rm real mode versions
which do mostly the same but use "Store Doubleword Caching Inhibited
Indexed" instruction for TCE invalidation.
This new feature is going to be utilized by real mode support of VFIO.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The current VFIO-on-POWER implementation supports only user mode
driven mapping, i.e. QEMU is sending requests to map/unmap pages.
However this approach is really slow, so we want to move that to KVM.
Since H_PUT_TCE can be extremely performance sensitive (especially with
network adapters where each packet needs to be mapped/unmapped) we chose
to implement that as a "fast" hypercall directly in "real
mode" (processor still in the guest context but MMU off).
To be able to do that, we need to provide some facilities to
access the struct page count within that real mode environment as things
like the sparsemem vmemmap mappings aren't accessible.
This adds an API function realmode_pfn_to_page() to get page struct when
MMU is off.
This adds to MM a new function put_page_unless_one() which drops a page
if counter is bigger than 1. It is going to be used when MMU is off
(for example, real mode on PPC64) and we want to make sure that page
release will not happen in real mode as it may crash the kernel in
a horrible way.
CONFIG_SPARSEMEM_VMEMMAP and CONFIG_FLATMEM are supported.
Cc: linux-mm@kvack.org
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We already had some output messages from EEH core. Occasionally,
we can see the output messages from EEH core before the stack
dump. That's not what we expected. The patch fixes that and shows
the stack dump prior to output messages from EEH core.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch adds function ioda_eeh_phb3_phb_diag() to dump PHB3
PHB diag-data. That's called while detecting informative errors
or frozen PE on the specific PHB.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Each PHB instance (struct pnv_phb) has its corresponding log blob,
which is used to hold the retrieved error log from firmware. The
current size of that (4096) isn't enough for PHB3 case and the patch
makes that double to 8192.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The patch prints the error number while failing to retrieve error
log from firmware. It's helpful for debugging.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
For now, we only support outbound error injection. Actually, the
hardware supports injecting inbound errors as well. The patch enables
to inject inbound errors.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The EEH isn't enabled for PHB3 and the patch intends to enable it.
Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This is the traditional name for device-tree path, used in sysfs,
do the same for the XSCOM debugfs files.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
OPAL v3 provides interfaces to access the chips XSCOM, expose
this via the existing scom infrastructure.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When creating the debugfs scom files, use "ibm,chip-id" as the scom%d
index rather than a simple made up number when possible.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
When devices are direct children of a scom controller node, they
should be able to use the normal "reg" property instead of "scom-reg".
In that case, they also use #address-cells rather than #scom-cells
to indicate the size of an entry.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
scom_read() now returns the read value via a pointer argument and
both functions return an int error code
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
isa_io_special is set when the platform provides a "special"
implementation of inX/outX via some FW interface for example.
Such a platform doesn't need an ISA bridge on PCI, and so /dev/port
should be made available even if one isn't present.
This makes the LPC bus IOs accessible via /dev/port on PowerNV Power8
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Since the CPU is generating an exception when accessing unaligned word, and
as this exception is not yet handled when running prom_init, data should be
copied from the architecture vector byte per byte.
Signed-off-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
The performance monitor interrupt is asynchronous, so we should check
if the current processor is in napping status in the handler of this
interrupt.
Signed-off-by: Kevin Hao <haokexin@gmail.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Add support for the arch_get_random_long() hook based on the H_RANDOM
hypervisor call. We trust the hypervisor to provide us with random data,
ie. we don't whiten it in anyway.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Add the plumbing to implement arch_get_random_long/int(). It didn't seem
worth adding an extra ppc_md hook for int, so we reuse the one for long.
Add an implementation for powernv based on the hwrng found in power7+
systems. We whiten the output of the hwrng, and the result passes all
the dieharder tests.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
This was missing on powerpc and I am getting compilation error
drivers/vfio/pci/vfio_pci_rdwr.c:193: undefined reference to `__cmpdi2'
drivers/vfio/pci/vfio_pci_rdwr.c:193: undefined reference to `__cmpdi2'
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
While cross-building for PPC64 I've got bunch of
WARNING: arch/powerpc/kernel/built-in.o(.text.unlikely+0x2d2): Section
mismatch in reference from the function .free_lppacas() to the variable
.init.data:lppaca_size The function .free_lppacas() references the variable
__initdata lppaca_size. This is often because .free_lppacas lacks a __initdata
annotation or the annotation of lppaca_size is wrong.
Fix it by using proper annotation for free_lppacas. Additionally, annotate
{allocate,new}_llpcas properly.
Signed-off-by: Vladimir Murzin <murzin.v@gmail.com>
Acked-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>