When running KVM on POWER, we allow the user to pass "-cpu POWERx" instead
of "-cpu host". This is achieved by patching the ppc_cpu_aliases[] array
so that "POWERx" points to the CPU class with the same PVR as the host CPU.
This causes CPUs to be instantiated from this CPU class instead of the
TYPE_HOST_POWERPC_CPU class which is used with "-cpu host". These CPUs thus
miss all the KVM specific tuning from kvmppc_host_cpu_class_init().
This currently causes QEMU with "-cpu POWER9" to fail when running KVM on a
POWER9 DD1 host:
qemu-system-ppc64: Register sync failed... If you're using kvm-hv.ko, only
"-cpu host" is possible
kvm_init_vcpu failed: Invalid argument
Let's have the "POWERx" alias to point to TYPE_HOST_POWERPC_CPU directly,
so that "-cpu POWERx" instantiates CPUs from the same class as "-cpu host".
Signed-off-by: Greg Kurz <groug@kaod.org>
Tested-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
In target/ppc/mmu-hash64.c there already exists the function
ppc_hash64_get_phys_page_debug() to get the physical (real) address for
a given effective address in hash mode.
Implement the function ppc_radix64_get_phys_page_debug() to allow a real
address to be obtained for a given effective address in radix mode.
This is used when a debugger is attached to qemu.
Previously we just had a comment saying this is unimplemented which then
fell through to the default case and caused an abort due to
unrecognised mmu model as the default had no case for the V3 mmu, which
was misleading at best.
We reuse ppc_radix64_walk_tree() which is used by the radix fault
handler since the process of walking the radix tree is identical.
Reported-by: Balbir Singh <bsingharora@gmail.com>
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The mmu-radix64.c file implements functions to enable the radix mmu
emulation in tcg mode. There is a function ppc_radix64_walk_tree() which
performs the radix tree walk and also implicitly checks the pte
protection.
Move the protection checking of the pte from the ppc_radix64_walk_tree()
function into the caller. This means the ppc_radix64_walk_tree() function
can be used without protection checking which is useful for debugging.
ppc_radix64_walk_tree() no longer needs to take the rwx and prot variables.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Properly set the book E exception syndrome register when a floating
point exception occurs.
Currently on a book E processor, the POWERPC_EXCP_FP exception handler
fails to set "env->spr[SPR_BOOKE_ESR] = ESR_FP;" as required by the
book E specification.
Signed-off-by: Aaron Larson <alarson@ddci.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
If ppc_cpu_realizefn() fails after cpu_exec_realizefn() has been
called, we will have to undo whatever cpu_exec_realizefn() did
by explicitly calling cpu_exec_unrealizeffn() which is currently
missing. Failure to do this proper cleanup will result in CPU
which was never fully realized to linger on the cpus list causing
SIGSEGV later (for eg when running "info cpus").
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The mmu fault handler should return 0 if it was able to successfully
handle the fault and a positive value otherwise.
Currently the tcg radix mmu fault handler will return 1 after
successfully handling a fault in virtual mode. This is incorrect
so fix it so that it returns 0 in this case.
The handler already correctly returns 0 when a fault was handled
in real mode and 1 if an interrupt was generated.
Fixes: d5fee0bbe6 ("target/ppc: Implement ISA V3.00 radix page fault handler")
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Since the introduction of MTTCG, using the msgsnd instruction
abort()s if being called without holding the BQL. So let's protect
that part of the code now with qemu_mutex_lock_iothread().
Buglink: https://bugs.launchpad.net/qemu/+bug/1694998
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Migrating between different CPU versions is a bit complicated for ppc.
A long time ago, we ensured identical CPU versions at either end by
checking the PVR had the same value. However, this breaks under KVM
HV, because we always have to use the host's PVR - it's not
virtualized. That would mean we couldn't migrate between hosts with
different PVRs, even if the CPUs are close enough to compatible in
practice (sometimes identical cores with different surrounding logic
have different PVRs, so this happens in practice quite often).
So, we removed the PVR check, but instead checked that several flags
indicating supported instructions matched. This turns out to be a bad
idea, because those instruction masks are not architected information, but
essentially a TCG implementation detail. So changes to qemu internal CPU
modelling can break migration - this happened between qemu-2.6 and
qemu-2.7. That was addressed by 146c11f1 "target-ppc: Allow eventual
removal of old migration mistakes".
Now, verification of CPU compatibility across a migration basically doesn't
happen. We simply ignore the PVR of the incoming migration, and hope the
cpu on the destination is close enough to work.
Now that we've cleaned up handling of processor compatibility modes
for pseries machine type, we can do better. For new machine types
(pseries-2.10+) We allow migration if:
* The source and destination PVRs are for the same type of CPU, as
determined by CPU class's pvr_match function
OR * When the source was in a compatibility mode, and the destination CPU
supports the same compatibility mode
For older machine types we retain the existing behaviour - current CAS
code will usually set a compat mode which would break backwards
migration if we made them use the new behaviour. [Fixed from an
earlier version by Greg Kurz].
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Tested-by: Andrea Bolognani <abologna@redhat.com>
Server class POWER CPUs have a "compat" property, which is used to set the
backwards compatibility mode for the processor. However, this only makes
sense for machine types which don't give the guest access to hypervisor
privilege - otherwise the compatibility level is under the guest's control.
To reflect this, this removes the CPU 'compat' property and instead
creates a 'max-cpu-compat' property on the pseries machine. Strictly
speaking this breaks compatibility, but AFAIK the 'compat' option was
never (directly) used with -device or device_add.
The option was used with -cpu. So, to maintain compatibility, this
patch adds a hack to the cpu option parsing to strip out any compat
options supplied with -cpu and set them on the machine property
instead of the now deprecated cpu property.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Tested-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Tested-by: Greg Kurz <groug@kaod.org>
Tested-by: Andrea Bolognani <abologna@redhat.com>
In some cases a failing VMSTATE_*_EQUAL does not mean we detected a bug,
but it's actually the best we can do. Especially in these cases a verbose
error message is required.
Let's introduce infrastructure for specifying a error hint to be used if
equal check fails. Let's do this by adding a parameter to the _EQUAL
macros called _err_hint. Also change all current users to pass NULL as
last parameter so nothing changes for them.
Signed-off-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Message-Id: <20170623144823.42936-1-pasic@linux.vnet.ibm.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
The string returned by object_property_get_str() is dynamically allocated.
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This function has three implementations. Two are stubs that do nothing
and the third one only passes the obj_path argument to:
Object *object_resolve_path(const char *path, bool *ambiguous);
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Move vcpu's associated numa_node field out of generic CPUState
into inherited classes that actually care about cpu<->numa mapping,
i.e: ARMCPU, PowerPCCPU, X86CPU.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1496161442-96665-6-git-send-email-imammedo@redhat.com>
[ehabkost: s/CPU is belonging to/CPU belongs to/ on comments]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
For transitioning back to userspace after the interrupt.
Suggested-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
When running with KVM, we update the "family" CPU alias to point
to the right host CPU type, so that it for example possible to
use "-cpu POWER8" on a POWER8NVL host. However, the function for
printing the list of available CPU models is called earlier than
the KVM setup code, so the output of "-cpu help" is wrong in that
case. Since it would be somewhat ugly anyway to have different
help texts depending on whether "-enable-kvm" has been specified
or not, we should better always print the same text, so fix this
issue by printing "alias for preferred XXX CPU" instead.
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
POWER9 DD1 silicon has some bugs which mean it a) isn't really compliant
with the ISA v3.00 and b) require a number of special workarounds in the
kernel.
At the moment, qemu isn't aware of DD1. For TCG we don't really want it to
be (why bother emulating buggy silicon). But with KVM, the guest does need
to be aware of DD1 so it can apply the necessary workarounds.
Meanwhile, the feature negotiation between qemu and the guest strongly
favours architected compatibility modes to "raw" CPU modes. In combination
with the above, this means the guest sees architected POWER9 mode, and
doesn't apply the DD1 workarounds. Well, unless it has yet another
workaround to partially ignore what qemu tells it.
This patch addresses this by disabling support for compatibility modes when
using KVM on a POWER9 DD1 host.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
ISA V3.00 introduced a new radix mmu model. Implement the page fault
handler for this so we can run a tcg guest in radix mode and perform
address translation correctly.
In real mode (mmu turned off) addresses are masked to remove the top
4 bits and then are subject to partition scoped translation, since we only
support pseries at this stage it is only necessary to perform the masking
and then we're done.
In virtual mode (mmu turned on) address translation if performed as
follows:
1. Use the quadrant to determine the fully qualified address.
The fully qualified address is defined as the combination of the effective
address, the effective logical partition id (LPID) and the effective
process id (PID). Based on the quadrant (EA63:62) we set the pid and lpid
like so:
quadrant 0: lpid = LPIDR, pid = PIDR
quadrant 1: HV only (not allowed in pseries)
quadrant 2: HV only (not allowed in pseries)
quadrant 3: lpid = LPIDR, pid = 0
If we can't get the fully qualified address we raise a segment interrupt.
2. Find the guest radix tree
We ask the virtual hypervisor for the partition table which was registered
with H_REGISTER_PROC_TBL which points us to the process table in guest
memory. We then index this table by pid to get the process table entry
which points us to the appropriate radix tree to translate the address.
If the process table isn't big enough to contain an entry for the current
pid then we raise a storage interrupt.
3. Walk the radix tree
Next we walk the radix tree where each level is a table of page directory
entries indexed by some number of bits from the effective address, where
the number of bits is determined by the table size. We continue to walk
the tree (while entries are valid and the table is of minimum size) until
we reach a table of page table entries, indicated by having the leaf bit
set. The appropriate pte is then checked for sufficient access permissions,
the reference and change bits are updated and the real address is
calculated from the real page number bits of the pte and the low bits of
the effective address.
If we can't find an entry or can't access the entry bacause of permissions
then we raise a storage interrupt.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
[dwg: Add missing parentheses to macro]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The tlbie[l] instructions are used to invalidate TLB entries used to cache
address translations.
In ISAv3.00 (POWER9) more fields were added to the tblie[l] instructions
which were previously invalid. We don't care about any of these new fields
since we just invalidate the whole world anyway but we need to not
cause an illegal instruction exception when the instructions are called.
We also don't want to allow an older processor to have these fields set
since that would be invalid.
Add a new GEN_HANDLER for the ISAv3 instructions with the correct invalid
mask. These will only be generated to a POWER9 processor for now based on
the instruction flag. Also remove the PPC_MEM_TLBIE instruction flag from
the POWER9 processor definition to ensure the old tlbie isn't generated.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The Guest Translation Shootdown Enable (GTSE) bit in the Logical Partition
Control Register (LPCR) can be set to enable a guest to use the tlbie
instruction directly to invalidate translations.
When the GTSE bit is set then the tlbie instruction is supervisor
privileged, otherwise it is hypervisor privileged.
Add a guest translation shootdown enable (gtse) field to the diassembly
context and use this to check the correct privilege level at code
generation time.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
In case when atomic operation is not supported, exit_atomic is called
and we stop the world and execute the atomic operation. This results
in a following call chain:
tcg_gen_atomic_cmpxchg_tl()
-> gen_helper_exit_atomic()
-> HELPER(exit_atomic)
-> cpu_loop_exit_atomic() -> EXCP_ATOMIC
-> qemu_tcg_cpu_thread_fn() => case EXCP_ATOMIC
-> cpu_exec_step_atomic()
-> cpu_step_atomic()
-> cc->cpu_exec_enter() = ppc_cpu_exec_enter()
Sets env->reserve_addr = -1;
But by the time it return back, the reservation is erased and the code
fails, this continues forever and the lock is never taken.
Instead set this in powerpc_excp()
Now that ppc_cpu_exec_enter() doesn't have anything meaningful to do,
let us get rid of the function.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This enables the multi-threaded system emulation by default for PPC64
guests using the x86_64 TCG back-end.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Emulating LL/SC with cmpxchg is not correct, since it can suffer from
the ABA problem. However, portable parallel code is written assuming
only cmpxchg which means that in practice this is a viable alternative.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This makes a small step fixing one of many style problems that exist in
the older ppc code. This removes spaces between function (or macro) name
and the following '('.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch registers mfspr 259 for Book3S and e500 family cores
following this research:
mfspr 259 provides read-only mapped user access to SPRG3(SPR 275) according to:
- PowerISA 2.02, Book III (documents implementation starting with POWER4+ @ p20)
- IBM PowerPC 970MP RISC Microprocessor User's Manual v2.1, page 48
- Amit Singh: "Mac OS X Internals: A Systems Approach" on 970 and 970FX cores:
He demonstrates mfspr 259 reading TLS data from Mac OS X on G5 on page 588
- NXP documents it in the Core Reference Manuals of: e500, e500mc and e5500
- getcpu() of the 32 & 64-bit Book3S Linux vDSOs use it to read the core number
mfspr 259 does not appear to be implemented in these cores according to:
- 74xx series: MPC7410/MPC7400 and MPC7450 RISC Microprocessor Reference Manuals
- 4xx series: PPC440 Processor User's Manual, Revision 1.09 by AMCC
- 750 series: IBM PowerPC 750CL RISC Microprocessor User's Manual
- e200 series: e200z4 Power Architectureâ Core Reference Manual
Implementation: gen_spr_usprg3() is called from init_proc_book3s_common()
(covers the 970 and POWER cores) and init_proc_e500() (covers the e500 family)
to register spr_read_ureg() in the same way which it already provides
the mapped SPR access for SPR_USPRG4-7 in gen_spr_usprgh() for cores
which have the same read-only mapped SPRG register access for SPRG4-7.
Verified using Linux by pinning a thread to a core and checking sched_getcpu()
using qemu-system-ppc64 -M pseries -cpu POWER8 using MTTCG on a x86_64 host.
Signed-off-by: Bernhard Kaindl <bernhard.kaindl@thalesgroup.com>
Reviewed-by: Stefan Resch <stefan.resch@thalesgroup.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The PIDR (process id register) is used to store the id of the currently
running process, which is used to select the process table entry used to
perform address translation. This means that when we write to this register
all the translations in the TLB become outdated as they are for a
previously running process. Thus when this register is written to we need
to invalidate the TLB entries to ensure stale entries aren't used to
to perform translation for the new process, which would result in at best
segfaults or alternatively just random memory being accessed.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
[dwg: Fixed compile error for 32-bit targets]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
gdb refuses to parse QEMU memory dumps because struct PPCElfPrstatus
is the wrong size. Fix it.
Signed-off-by: Anton Blanchard <anton@samba.org>
Fixes: e62fbc54d4 ("target-ppc: dump-guest-memory support")
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Today, the ICPState array of the sPAPR machine is indexed with
'cpu_index' of the CPUState. This numbering of CPUs is internal to
QEMU and the guest only knows about what is exposed in the device
tree, that is the 'cpu_dt_id'. This is why sPAPR uses the helper
xics_get_cpu_index_by_dt_id() to do the mapping in a couple of places.
To provide a more generic XICS layer, we need to abstract the IRQ
'server' number and remove any assumption made on its nature. It
should not be used as a 'cpu_index' for lookups like xics_cpu_setup()
and xics_cpu_destroy() do.
To reach that goal, we choose to introduce a generic 'intc' backlink
under PowerPCCPU, and let the machine core init routine do the
ICPState lookup. The resulting object is passed on to xics_cpu_setup()
which does the store under PowerPCCPU. The IRQ 'server' number in XICS
is now generic. sPAPR uses 'cpu_dt_id' and PowerNV will use 'PIR'
number.
This also has the benefit of simplifying the sPAPR hcall routines
which do not need to do any ICPState lookups anymore.
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The ibm,processor-radix-AP-encodings device tree property of the cpu node
is used to specify the radix mode supported page sizes of the processor
to the guest os. Contained in the top 3 bits of the msb is the actual
page size (AP) encoding associated with the corresponding radix mode
supported page size. Add this property for a TCG guest, note the TCG code
is capable of translating any format so just add the 4 default page sizes.
The ibm,processor-radix-AP-encodings device tree property is defined as:
One to n cells in ascending order of radix mode supported page sizes
encoded as BE ints (32bit on ppc) in the form:
0bxxxyyyyyyyyyyyyyyyyyyyyyyyyyyyyy
- 0bxxx -> AP encoding
- 0byyyyyyyyyyyyyyyyyyyyyyyyyyyyy -> supported page size encoded as a shift
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This enables in-kernel handling of H_PUT_TCE_INDIRECT and
H_STUFF_TCE hypercalls. The host kernel support is there since v4.6,
in particular d3695aa4f452
("KVM: PPC: Add support for multiple-TCE hcalls").
H_PUT_TCE is already accelerated and does not need any special enablement.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The H_REGISTER_PROCESS_TABLE H_CALL is used by a guest to indicate to the
hypervisor where in memory its process table is and how translation should
be performed using this process table.
Provide the implementation of this H_CALL for a guest.
We first check for invalid flags, then parse the flags to determine the
operation, and then check the other parameters for valid values based on
the operation (register new table/deregister table/maintain registration).
The process table is then stored in the appropriate location and registered
with the hypervisor (if running under KVM), and the LPCR_[UPRT/GTSE] bits
are updated as required.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
[dwg: Correct missing prototype and uninitialized variable]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Query and cache the value of two new KVM capabilities that indicate
KVM's support for new radix and hash modes of the MMU.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Use the new ioctl, KVM_PPC_GET_RMMU_INFO, to fetch radix MMU
information from KVM and present the page encodings in the device tree
under ibm,processor-radix-AP-encodings. This provides page size
information to the guest which is necessary for it to use radix mode.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
[dwg: Compile fix for 32-bit targets, style nit fix]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
KVM_CAP_SPAPR_TCE capability allows creating TCE tables in KVM which
allows having in-kernel acceleration for H_PUT_TCE_xxx hypercalls.
However it only supports 32bit DMA windows at zero bus offset.
There is a new KVM_CAP_SPAPR_TCE_64 capability which supports 64bit
window size, variable page size and bus offset.
This makes use of the new capability. The kernel headers are already
updated as the kernel support went in to v4.6.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
On Power8 hosts it is currently theoretically possible for QEMU/KVM-HV guests
to receive a ibm,pa-features property indicating that HTM support is available
when it is not. The situation would occur if the platform firmware of
a Power8 host cleared the HTM bit of the ibm,pa-features property.
QEMU would query KVM for the availability of HTM, which will return no
support, but workaround code in kvm_arch_init_vcpu() would then
re-enable it because KVM_HV is in use and the processor is P8.
This patch adjusts the workaround in kvm_arch_init_vcpu() so that it does not
enable HTM (in the above case) unless the host kernel indicates to the QEMU
process, via the auxiliary vector, that userspace can use HTM (via the HWCAP2
bit KVM_FEATURE2_HTM).
The reason to use the value from the auxiliary vector is that it is
set based only on what the host kernel found in the ibm,pa-features
HTM bit at boot time.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This removes the assert(kvm_enabled()) from kvmppc_host_cpu_initfn()
This assert can never be triggered as the function is only registered
when KVM is available (see also 4c315c2
"qdev: Protect device-list-properties against broken devices").
So we can remove the cannot_destroy_with_object_finalize_yet from
kvmppc_host_cpu_class_init() without fear and beyond reproach.
(as it has already be done for i386 with 771a13e "i386: Unset
cannot_destroy_with_object_finalize_yet on "host" model" and
e435601 "target-i386: Remove assert(kvm_enabled()) from
host_x86_cpu_initfn()")
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Message-Id: <20170414083717.13641-3-lvivier@redhat.com>
Acked-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
A bug was introduced in following commit:
dc0ad84 target/ppc: update overflow flags for add/sub
As for 32-bit ppc target extracting bit 63 for overflow is not correct.
Made it dependent on TARGET_LOG_BITS. This had broken booting MacOS
9.2.1 image
Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
The SPR UAMR has the number 13, and not 12. (Fortunately it seems like
Linux is not using this register yet - only the privileged version with
number 29 ... that's why nobody noticed this problem yet)
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Use the helper routine float[32,64]_maddsub_update_excp() in VSX_MADD
macro.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Adds FPU_MADDSUB_UPDATE macro, this will be used for other routines
having float32/16
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Current order of checking does not confirm with the spec
(ISA 3.0: MultiplyAddDP page-469). Change the order and make them
independent of each other.
For example: a = infinity, b = zero, c = SNaN, this should set both
VXIMZ and VXNAN
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This will probably be my last pull request before the hard freeze. It
has some new work, but that has all been posted in draft before the
soft freeze, so I think it's reasonable to include in qemu-2.9.
This batch has:
* A substantial amount of POWER9 work
* Implements the legacy (hash) MMU for POWER9
* Some more preliminaries for implementing the POWER9 radix
MMU
* POWER9 has_work
* Basic POWER9 compatibility mode handling
* Removal of some premature tests
* Some cleanups and fixes to the existing MMU code to make the
POWER9 work simpler
* A bugfix for TCG multiply adds on power
* Allow pseries guests to access PCIe extended config space
This also includes a code-motion not strictly in ppc code - moving
getrampagesize() from ppc code to exec.c. This will make some future
VFIO improvements easier, Paolo said it was ok to merge via my tree.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=PcNw
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-2.9-20170303' into staging
ppc patch queuye for 2017-03-03
This will probably be my last pull request before the hard freeze. It
has some new work, but that has all been posted in draft before the
soft freeze, so I think it's reasonable to include in qemu-2.9.
This batch has:
* A substantial amount of POWER9 work
* Implements the legacy (hash) MMU for POWER9
* Some more preliminaries for implementing the POWER9 radix
MMU
* POWER9 has_work
* Basic POWER9 compatibility mode handling
* Removal of some premature tests
* Some cleanups and fixes to the existing MMU code to make the
POWER9 work simpler
* A bugfix for TCG multiply adds on power
* Allow pseries guests to access PCIe extended config space
This also includes a code-motion not strictly in ppc code - moving
getrampagesize() from ppc code to exec.c. This will make some future
VFIO improvements easier, Paolo said it was ok to merge via my tree.
# gpg: Signature made Fri 03 Mar 2017 03:20:36 GMT
# gpg: using RSA key 0x6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>"
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>"
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>"
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>"
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-2.9-20170303:
target/ppc: rewrite f[n]m[add,sub] using float64_muladd
spapr: Small cleanup of PPC MMU enums
spapr_pci: Advertise access to PCIe extended config space
target/ppc: Rework hash mmu page fault code and add defines for clarity
target/ppc: Move no-execute and guarded page checking into new function
target/ppc: Add execute permission checking to access authority check
target/ppc: Add Instruction Authority Mask Register Check
hw/ppc/spapr: Add POWER9 to pseries cpu models
target/ppc/POWER9: Add cpu_has_work function for POWER9
target/ppc/POWER9: Add POWER9 pa-features definition
target/ppc/POWER9: Add POWER9 mmu fault handler
target/ppc: Don't gen an SDR1 on POWER9 and rework register creation
target/ppc: Add patb_entry to sPAPRMachineState
target/ppc/POWER9: Add POWERPC_MMU_V3 bit
powernv: Don't test POWER9 CPU yet
exec, kvm, target-ppc: Move getrampagesize() to common code
target/ppc: Add POWER9/ISAv3.00 to compat_table
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Call kvm_on_sigbus_vcpu asynchronously from the VCPU thread.
Information for the SIGBUS can be stored in thread-local variables
and processed later in kvm_cpu_exec.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Build it on kvm_arch_on_sigbus_vcpu instead. They do the same
for "action optional" SIGBUSes, and the main thread should never get
"action required" SIGBUSes because it blocks the signal.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the softfloat api for fused multiply-add.
Introduce routine to set the FPSCR flags VXNAN, VXIMZ nad VMISI.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The PPC MMU types are sometimes treated as if they were a bit field
and sometime as if they were an enum which causes maintenance
problems: flipping bits in the MMU type (which is done on both the 1TB
segment and 64K segment bits) currently produces new MMU type
values that are not handled in every "switch" on it, sometimes causing
an abort().
This patch provides some macros that can be used to filter out the
"bit field-like" bits so that the remainder of the value can be
switched on, like an enum. This allows removal of all of the
"degraded" types from the list and should ease maintenance.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The hash mmu page fault handling code is responsible for generating ISIs
and DSIs when access permissions cause an access to fail. Part of this
involves setting the srr1 or dsisr registers to indicate what causes the
access to fail. Add defines for the bit fields of these registers and
rework the code to use these new defines in order to improve readability
and code clarity.
While we're here, update what is logged when an access fails to include
information as to what caused to access to fail for debug purposes.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
[dwg: Moved constants to cpu.h since they're not MMUv3 specific]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
A pte entry has bit fields which can be used to make a page no-execute or
guarded, if either of these bits are set then an instruction access to this
page will fail. Currently these bits are checked with the pp_prot function
however the ISA specifies that the access authority controlled by the
key-pp value pair should only be checked on an instruction access after
the no-execute and guard bits have already been verified to permit the
access.
Move the no-execute and guard bit checking into a new separate function.
Note that we can remove the check for the no-execute bit in the slb entry
since this check was already performed above when we obtained the slb
entry.
In the event that the no-execute or guard bits are set, an ISI should be
generated with the SRR1_NOEXEC_GUARD (0x10000000) bit set in srr1. Add a
define for this for clarity.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
[dwg: Move constants to cpu.h since they're not MMUv3 specific]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Basic storage protection defines various access authority permissions
based on a slb storage key and pte pp value pair. This access authority
defines read, write and execute permissions however currently we only
use this to control read and write permissions and ignore the execute
control.
Fix the code to allow execute permissions based on the key-pp value pair.
Execute is allowed under the same conditions which enable reads.
(i.e. read permission -> execute permission)
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The instruction authority mask register (IAMR) can be used to restrict
permissions for instruction fetch accesses on a per key basis for each
of 32 different key values. Access permissions are derived based on the
specific key value stored in the relevant page table entry.
The IAMR was introduced in, and is present in processors since, POWER8
(ISA v2.07). Thus introduce a function to check access permissions based
on the pte key value and the contents of the IAMR when handling a page
fault to ensure sufficient access permissions for an instruction fetch.
A hash pte contains a key value in bits 2:3|52:54 of the second double word
of the pte, this key value gives an index into the IAMR which contains 32
2-bit access masks. If the least significant bit of the 2-bit access mask
corresponding to the given key value is set (IAMR[key] & 0x1 == 0x1) then
the instruction fetch is not permitted and an ISI is generated accordingly.
While we're here, add defines for the srr1 bits to be set for the ISI for
clarity.
e.g.
pte:
dw0 [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]
dw1 [XX01XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX010XXXXXXXXX]
^^ ^^^
key = 01010 (0x0a)
IAMR: [XXXXXXXXXXXXXXXXXXXX01XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX]
^^
Access mask = 0b01
Test access mask: 0b01 & 0x1 == 0x1
Least significant bit of the access mask is set, thus the instruction fetch
is not permitted. We should generate an instruction storage interrupt (ISI)
with bit 42 of SRR1 set to indicate access precluded by virtual page class
key protection.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
[dwg: Move new constants to cpu.h, since they're not MMUv3 specific]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The cpu has work function is used to mask interrupts used to determine
if there is work for the cpu based on the LPCR. Add a function to do this
for POWER9 and add it to the POWER9 cpu definition. This is similar to that
for POWER8 except using the LPCR bits as defined for POWER9.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Add a new mmu fault handler for the POWER9 cpu and add it as the handler
for the POWER9 cpu definition.
This handler checks if the guest is radix or hash based on the value in the
partition table entry and calls the correct fault handler accordingly.
The hash fault handling code has also been updated to check if the
partition is using segment tables.
Currently only legacy hash (no segment tables) is supported.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
POWER9 doesn't have a storage description register 1 (SDR1) which is used
to store the base and size of the hash table. Thus we don't need to
generate this register on the POWER9 cpu model. While we're here, the
register generation code for 970, POWER5+, POWER<7/8/9> in general is a
mess where we call a generic function from a model specific function which
then attempts to call model specific functions, so rework this for
readability.
We update ppc_cpu_dump_state so that "info registers" will only display
the value of sdr1 if the register has been generated.
As mentioned above the register generation for the pcc->init_proc
function for 970, POWER5+, POWER7, POWER8 and POWER9 has been reworked
for improved clarity. Instead of calling init_proc_book3s_64 which then
attempts to generate the correct registers through a mess of if statements,
we remove this function and instead call the appropriate register
generation functions directly. This follows the register generation model
used for earlier cpu models (pre-970) whereby cpu specific registers are
generated directly in the init_proc function and makes it easier to
add/remove specific registers for new cpu models.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
ISA v3.00 adds the idea of a partition table which is used to store the
address translation details for all partitions on the system. The partition
table consists of double word entries indexed by partition id where the second
double word contains the location of the process table in guest memory. The
process table is registered by the guest via a h-call.
We need somewhere to store the address of the process table so we add an entry
to the sPAPRMachineState struct called patb_entry to represent the second
doubleword of a single partition table entry corresponding to the current
guest. We need to store this value so we know if the guest is using radix or
hash translation and the location of the corresponding process table in guest
memory. Since we only have a single guest per qemu instance, we only need one
entry.
Since the partition table is technically a hypervisor resource we require that
access to it is abstracted by the virtual hypervisor through the get_patbe()
call. Currently the value of the entry is never set (and thus
defaults to 0 indicating hash), but it will be required to both implement
POWER9 kvm support and tcg radix support.
We also add this field to be migrated as part of the sPAPRMachineState as we
will need it on the receiving side as the guest will never tell us this
information again and we need it to perform translation.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
For easier handling of future processors using the POWER9 or something
close to it, add a new bit in the MMU model. This was originally from a
revised version of 86cf1e9 "target/ppc/POWER9: Add ISAv3.00 MMU definition"
but the older version of the patch was already merged. This makes the
change on top of the original version.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
getrampagesize() returns the largest supported page size and mainly
used to know if huge pages are enabled.
However is implemented in target-ppc/kvm.c and not available
in TCG or other architectures.
This renames and moves gethugepagesize() to mmap-alloc.c where
fd-based analog of it is already implemented. This renames and moves
getrampagesize() to exec.c as it seems to be the common place for
helpers like this.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
compat_table contains the list of logical pvr compat modes which a cpu can
operate in. It is a list of struct CompatInfo which contains the given pvr
value for a compat mode, the pcr bits which should be set to operate in
that compat mode, the pcr level which must be present in pcr_supported for
a processor to support that compat mode and the max threads possible in
that compat mode.
Add an entry for the POWER9/ISAv3.00 logical pvr which represents a
processor running with support for logical pvr 0x0f000005. A processor
running in this mode should have PCR_COMPAT_3_00 set in the pcr (if
available in pcr_mask) and should have PCR_COMPAT_3_00 in pcr_supported
to indicate that it is capable of running in this compat mode.
Also add PCR_COMPAT_3_00 to the bits which must be set for all previous
compat modes. Since no processor models contain this bit yet in pcr_mask
it will never be set, but this ensures we don't forget to in the future.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This patch extends support for the `dump-guest-memory` command to the
32-bit PowerPC architecture. It relies on the assumption that a 64-bit
guest will not dump a 32-bit core file (and vice versa).
[dwg: I suspect this patch won't cover all cases, in particular a
32-bit machine type on a 64-bit qemu build. However, it does strictly
more than what we had before, so might as well apply as a starting
point]
Signed-off-by: Mike Nawrocki <michael.nawrocki@gtri.gatech.edu>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
mcrxrx: Move to CR from XER Extended
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Add helper_div_compute_ov() in the int_helper for updating the overflow
flags.
For Divide Word:
SO, OV, and OV32 bits reflects overflow of the 32-bit result
For Divide DoubleWord:
SO, OV, and OV32 bits reflects overflow of the 64-bit result
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
For Multiply Word:
SO, OV, and OV32 bits reflects overflow of the 32-bit result
For Multiply DoubleWord:
SO, OV, and OV32 bits reflects overflow of the 64-bit result
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
* SO and OV reflects overflow of the 64-bit result in 64-bit mode and
overflow of the low-order 32-bit result in 32-bit mode
* OV32 reflects overflow of the low-order 32-bit independent of the mode
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Adds routine to compute ca32 - gen_op_arith_compute_ca32
For 64-bit mode use the compute ca32 routine. While for 32-bit mode, CA
and CA32 will have same value.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
POWER ISA 3.0 adds CA32 and OV32 status in 64-bit mode. Add the flags
and corresponding defines.
Moreover, CA32 is updated when CA is updated and OV32 is updated when OV
is updated.
Arithmetic instructions:
* Addition and Substractions:
addic, addic., subfic, addc, subfc, adde, subfe, addme, subfme,
addze, and subfze always updates CA and CA32.
=> CA reflects the carry out of bit 0 in 64-bit mode and out of
bit 32 in 32-bit mode.
=> CA32 reflects the carry out of bit 32 independent of the
mode.
=> SO and OV reflects overflow of the 64-bit result in 64-bit
mode and overflow of the low-order 32-bit result in 32-bit
mode
=> OV32 reflects overflow of the low-order 32-bit independent of
the mode
* Multiply Low and Divide:
For mulld, divd, divde, divdu and divdeu: SO, OV, and OV32 bits
reflects overflow of the 64-bit result
For mullw, divw, divwe, divwu and divweu: SO, OV, and OV32 bits
reflects overflow of the 32-bit result
* Negate with OE=1 (nego)
For 64-bit mode if the register RA contains
0x8000_0000_0000_0000, OV and OV32 are set to 1.
For 32-bit mode if the register RA contains 0x8000_0000, OV and
OV32 are set to 1.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
SDR_64_HTABORG, which indicates the bits of the SDR1 register to use for
the base of a 64-bit machine's hashed page table (HPT) isn't correct. It
includes the top 46 bits of the register, but in fact the top 4 bits must
be zero (according to the ISA v2.07). No actual implementation has
supported close to 2^60 bytes of physical address space, so it's kind of
irrelevant, but we might as well correct this.
In addition, although we checked for bad size values in SDR1, we never
reported an error if entirely invalid bits were set there. Add this check
to ppc_store_sdr1().
Reported-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The function ppc_hash64_set_sdr1 basically checked the htabsize and set an
error if it was too big, otherwise it just stored the value in SPR_SDR1.
Given that the only function which calls ppc_hash64_set_sdr1() is
ppc_store_sdr1(), why not handle the checking in ppc_store_sdr1() to avoid
the extra function call. Note that ppc_store_sdr1() already stores the
value in SPR_SDR1 anyway, so we were doing it twice.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
[dwg: Remove unnecessary error temporary]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The pseries machine type implements the behaviour of a PAPR compliant
hypervisor, without actually executing such a hypervisor on the virtual
CPU. To do this we need some hooks in the CPU code to make hypervisor
facilities get redirected to the machine instead of emulated internally.
For hypercalls this is managed through the cpu->vhyp field, which points
to a QOM interface with a method implementing the hypercall.
For the hashed page table (HPT) - also a hypervisor resource - we use an
older hack. CPUPPCState has an 'external_htab' field which when non-NULL
indicates that the HPT is stored in qemu memory, rather than within the
guest's address space.
For consistency - and to make some future extensions easier - this merges
the external HPT mechanism into the vhyp mechanism. Methods are added
to vhyp for the basic operations the core hash MMU code needs: map_hptes()
and unmap_hptes() for reading the HPT, store_hpte() for updating it and
hpt_mask() to retrieve its size.
To match this, the pseries machine now sets these vhyp fields in its
existing vhyp class, rather than reaching into the cpu object to set the
external_htab field.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
CPUPPCState includes fields htab_base and htab_mask which store the base
address (GPA) and size (as a mask) of the guest's hashed page table (HPT).
These are set when the SDR1 register is updated.
Keeping these in sync with the SDR1 is actually a little bit fiddly, and
probably not useful for performance, since keeping them expands the size of
CPUPPCState. It also makes some upcoming changes harder to implement.
This patch removes these fields, in favour of calculating them directly
from the SDR1 contents when necessary.
This does make a change to the behaviour of attempting to write a bad value
(invalid HPT size) to the SDR1 with an mtspr instruction. Previously, the
bad value would be stored in SDR1 and could be retrieved with a later
mfspr, but the HPT size as used by the softmmu would be, clamped to the
allowed values. Now, writing a bad value is treated as a no-op. An error
message is printed in both new and old versions.
I'm not sure which behaviour, if either, matches real hardware. I don't
think it matters that much, since it's pretty clear that if an OS writes
a bad value to SDR1, it's not going to boot.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Accesses to the hashed page table (HPT) are complicated by the fact that
the HPT could be in one of three places:
1) Within guest memory - when we're emulating a full guest CPU at the
hardware level (e.g. powernv, mac99, g3beige)
2) Within qemu, but outside guest memory - when we're emulating user and
supervisor instructions within TCG, but instead of emulating
the CPU's hypervisor mode, we just emulate a hypervisor's behaviour
(pseries in TCG or KVM-PR)
3) Within the host kernel - a pseries machine using KVM-HV
acceleration. Mostly accesses to the HPT are handled by KVM,
but there are a few cases where qemu needs to access it via a
special fd for the purpose.
In order to batch accesses to the fd in case (3), we use a somewhat awkward
ppc_hash64_start_access() / ppc_hash64_stop_access() pair, which for case
(3) reads / releases several HPTEs from the kernel as a batch (usually a
whole PTEG). For cases (1) & (2) it just returns an address value. The
actual HPTE load helpers then need to interpret the returned token
differently in the 3 cases.
This patch keeps the same basic structure, but simplfiies the details.
First start_access() / stop_access() are renamed to map_hptes() and
unmap_hptes() to make their operation more obvious. Second, map_hptes()
now always returns a qemu pointer, which can always be used in the same way
by the load_hpte() helpers. In case (1) it comes from address_space_map()
in case (2) directly from qemu's HPT buffer and in case (3) from a
temporary buffer read from the KVM fd.
While we're at it, make things a bit more consistent in terms of types and
variable names: avoid variables named 'index' (it shadows index(3) which
can lead to confusing results), use 'hwaddr ptex' for HPTE indices and
uint64_t for each of the HPTE words, use ptex throughout the call stack
instead of pte_offset in some places (we still need that at the bottom
layer, but nowhere else).
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At present the SDR1 register - the base of the system's hashed page table
(HPT) - is represented as an SPR with supervisor read and write permission.
However, on CPUs which have a hypervisor mode, the SDR1 is a hypervisor
only resource. Change the permission checking on the SPR to reflect this.
Now that this is done, we don't need to check for an external HPT executing
mtsdr1: an external HPT only applies when we're emulating the behaviour of
a hypervisor, rather than modelling the CPU's hypervisor mode internally,
so if we're permitted to execute mtsdr1, we don't have an external HPT.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
cpu_ppc_set_papr() sets up various aspects of CPU state for use with PAPR
paravirtualized guests. However, it doesn't set the virtual hypervisor,
so callers must also call cpu_ppc_set_vhyp() so that PAPR hypercalls are
handled properly. This is a bit silly, so fold setting the virtual
hypervisor into cpu_ppc_set_papr().
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
When a 'pseries' guest is running with KVM-HV, the guest's hashed page
table (HPT) is stored within the host kernel, so it is not directly
accessible to qemu. Most of the time, qemu doesn't need to access it:
we're using the hardware MMU, and KVM itself implements the guest
hypercalls for manipulating the HPT.
However, qemu does need access to the in-KVM HPT to implement
get_phys_page_debug() for the benefit of the gdbstub, and maybe for
other debug operations.
To allow this, 7c43bca "target-ppc: Fix page table lookup with kvm
enabled" added kvmppc_hash64_read_pteg() to target/ppc/kvm.c to read
in a batch of HPTEs from the KVM table. Unfortunately, there are a
couple of problems with this:
First, the name of the function implies it always reads a whole PTEG
from the HPT, but in fact in some cases it's used to grab individual
HPTEs (which ends up pulling 8 HPTEs, not aligned to a PTEG from the
kernel).
Second, and more importantly, the code to read the HPTEs from KVM is
simply wrong, in general. The data from the fd that KVM provides is
designed mostly for compact migration rather than this sort of one-off
access, and so needs some decoding for this purpose. The current code
will work in some cases, but if there are invalid HPTEs then it will
not get sane results.
This patch rewrite the HPTE reading function to have a simpler
interface (just read n HPTEs into a caller provided buffer), and to
correctly decode the stream from the kernel.
For consistency we also clean up the similar function for altering
HPTEs within KVM (introduced in c138593 "target-ppc: Update
ppc_hash64_store_hpte to support updating in-kernel htab").
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Removes duplicate code and will be useful for consolidating flags
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This pull request has:
* Yet more POWER9 instruction implementations
* Some extensions to the softfloat code which are necesssary for
some of those instructions
* Some preliminary patches in preparation for POWER9 softmmu
implementation
* Igor Mammedov's cleanups to unify hotplug cpu handling across
architectures
* Assorted bugfixes
The softfloat and cpu hotplug changes aren't entirely ppc specific (in
fact the hotplug stuff contains some pc specific patches). However
they're included here because ppc is one of the main beneficiaries,
and the series depend on some ppc specific patches.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
iQIcBAABCAAGBQJYrS/bAAoJEGw4ysog2bOSJ3MP/AjoTGTP5MHPwWBZKAxpEtie
vEXudVbOelr3QV06vHMH4YHVAncuzt9Hmz/RDgs5Uynp4vLdmEo5IdiFP9PMjrFg
oMAndku9icU8PG+XNF5pNrKy10n6k8dVRBR/19UxnRWMuxywOZO208WkICF/6kDK
IpFT96MubqbReLcVhdl2N8d2rP7/lRQmz6aPxhRLFBuAe8iheAQLq/QeZLIZaWEJ
i4mPWVu/CDYP9nMAgv56MW0yY5p2o5MCh+f80+7jvKXZBoeo83KOTaZeZbGb/byr
rCfyLTR24tj6WUGRvzyB+FJ8rbWKcox4UCx17239gAjXtLxhlYaQDo28S5gwinpQ
b/CaEgb8x2kl97tZT/M1mamr7PdFxachCA20oizguwFJ9oeukAPUvkVBpEtVYK8K
a+VrRHxVJwSi/ZD3N6WRZMXR4D+Oc8DcXoEzMu4CFtIzQ/WJroZCa4JCcdv4N1nw
9u1m+C2QbQ9sGBtTSGCy0KZyT3sZHoFT6aD4zpkV7s3BJKk+AXSLRpL4z8FP2sDB
Wh/Qk5q06P1pPZzvuU9QJmrpIE9EFcOQW4IQhyViut+BXzBlp7cWxeGcPM5PuJ7V
6FcMSchZeVOiLi9Y51csluDrecTKIQ3yFEgLW7j50Lg/WqmdwlwkcW39MzlWgjgQ
OIoVgvGmGovPTGIIYyY9
=bsJJ
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-2.9-20170222' into staging
ppc patch queue for 2017-02-22
This pull request has:
* Yet more POWER9 instruction implementations
* Some extensions to the softfloat code which are necesssary for
some of those instructions
* Some preliminary patches in preparation for POWER9 softmmu
implementation
* Igor Mammedov's cleanups to unify hotplug cpu handling across
architectures
* Assorted bugfixes
The softfloat and cpu hotplug changes aren't entirely ppc specific (in
fact the hotplug stuff contains some pc specific patches). However
they're included here because ppc is one of the main beneficiaries,
and the series depend on some ppc specific patches.
# gpg: Signature made Wed 22 Feb 2017 06:29:47 GMT
# gpg: using RSA key 0x6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>"
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>"
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>"
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>"
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dgibson/tags/ppc-for-2.9-20170222: (43 commits)
hw/ppc/ppc405_uc.c: Avoid integer overflows
hw/ppc/spapr: Check for valid page size when hot plugging memory
target-ppc: fix Book-E TLB matching
hw/net/spapr_llan: 6 byte mac address device tree entry
machine: replace query_hotpluggable_cpus() callback with has_hotpluggable_cpus flag
machine: unify [pc_|spapr_]query_hotpluggable_cpus() callbacks
spapr: reuse machine->possible_cpus instead of cores[]
change CPUArchId.cpu type to Object*
pc: pass apic_id to pc_find_cpu_slot() directly so lookup could be done without CPU object
pc: calculate topology only once when possible_cpus is initialised
pc: move pcms->possible_cpus init out of pc_cpus_init()
machine: move possible_cpus to MachineState
hw/pci-host/prep: Do not use hw_error() in realize function
target/ppc/POWER9: Direct all instr and data storage interrupts to the hypv
target/ppc/POWER9: Adapt LPCR handling for POWER9
target/ppc/POWER9: Add ISAv3.00 MMU definition
target/ppc: Fix LPCR DPFD mask define
target-ppc: Add xscvqpudz and xscvqpuwz instructions
target-ppc: Implement round to odd variants of quad FP instructions
softfloat: Add float128_to_uint32_round_to_zero()
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
On POWER, the valid page sizes that the guest can use are bound
to the CPU and not to the memory region. QEMU already has some
fancy logic to find out the right maximum memory size to tell
it to the guest during boot (see getrampagesize() in the file
target/ppc/kvm.c for more information).
However, once we're booted and the guest is using huge pages
already, it is currently still possible to hot-plug memory regions
that does not support huge pages - which of course does not work
on POWER, since the guest thinks that it is possible to use huge
pages everywhere. The KVM_RUN ioctl will then abort with -EFAULT,
QEMU spills out a not very helpful error message together with
a register dump and the user is annoyed that the VM unexpectedly
died.
To avoid this situation, we should check the page size of hot-plugged
DIMMs to see whether it is possible to use it in the current VM.
If it does not fit, we can print out a better error message and
refuse to add it, so that the VM does not die unexpectely and the
user has a second chance to plug a DIMM with a matching memory
backend instead.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1419466
Signed-off-by: Thomas Huth <thuth@redhat.com>
[dwg: Fix a build error on 32-bit builds with KVM]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The Book-E TLB matching process should bail out early when a TLB
entry matches, but the access permissions are wrong. The CPU
will then raise a DSI error instead of a Data TLB error, as
described for TLB matching in Freescale and IBM documents.
Signed-off-by: Alex Zuepke <azu@sysgo.de>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The vpm0 bit was removed from the LPCR in POWER9, this bit controlled
whether ISI and DSI interrupts were directed to the hypervisor or the
partition. These interrupts now go to the hypervisor irrespective, thus
it is no longer necessary to check the vmp0 bit in the LPCR.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The logical partitioning control register controls a threads operation
based on the partition it is currently executing. Add new definitions and
update the mask used when writing to the LPCR based on the POWER9 spec.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
POWER9 processors implement the mmu as defined in version 3.00 of the ISA.
Add a definition for this mmu model and set the POWER9 cpu model to use
this mmu model.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The DPFD field in the LPCR is 3 bits wide. This has always been defined
as 0x3 << shift which indicates a 2 bit field, which is incorrect.
Correct this.
Signed-off-by: Suraj Jitindar Singh <sjitindarsingh@gmail.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
xscvqpudz: VSX Scalar truncate & Convert Quad-Precision format to
Unsigned Doubleword format
xscvqpuwz: VSX Scalar truncate & Convert Quad-Precision format to
Unsigned Word format
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
xsaddqpo: VSX Scalar Add Quad-Precision using round to Odd
xsmulqo: VSX Scalar Multiply Quad-Precision using round to Odd
xsdivqpo: VSX Scalar Divide Quad-Precision using round to Odd
xscvqpdpo: VSX Scalar round & Convert Quad-Precision format to
Double-Precision format using round to Odd
xssqrtqpo: VSX Scalar Square Root Quad-Precision using round to Odd
xssubqpo: VSX Scalar Subtract Quad-Precision using round to Odd
In addition, fix the invalid bitmask in the instruction encoding
of xssqrtqp[o].
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
CC: Jose Ricardo Ziviani <joserz@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Use the available wait instruction implementation.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
slbsync: SLB Synchoronize
The instruction provides an ordering function for the effects of all
slbieg instructions executed by the thread executing the slbsync
instruction.
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
slbieg: SLB Invalidate Entry Global
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
stwat: Store Word Atomic
stdat: Store Doubleword Atomic
The instruction includes as function code (5 bits) which gives a detail
on the operation to be performed. The patch implements five such
functions.
Signed-off-by: Balamuruhan S <bala24@linux.vnet.ibm.com>
Signed-off-by: Harish S <harisrir@linux.vnet.ibm.com>
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
[ implement stdat, use macro and combine both implementation ]
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
lwat: Load Word Atomic
ldat: Load Doubleword Atomic
The instruction includes as function code (5 bits) which gives a detail
on the operation to be performed. The patch implements five such
functions.
Signed-off-by: Balamuruhan S <bala24@linux.vnet.ibm.com>
Signed-off-by: Harish S <harisrir@linux.vnet.ibm.com>
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
[ combine both lwat/ldat implementation using macro ]
Signed-off-by: Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
When running certain HMP commands ("info registers", "info cpustats",
"info tlb", "nmi", "memsave" or dumping virtual memory) with the "none"
machine, QEMU crashes with a segmentation fault. This happens because the
"none" machine does not have any CPUs by default, but these HMP commands
did not check for a valid CPU pointer yet. Add such checks now, so we get
an error message about the missing CPU instead.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1484309555-1935-1-git-send-email-thuth@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
When running with KVM on POWER, we are registering a "family" CPU
type for the host CPU that we are running on. For example, on all
POWER8-compatible hosts, we register a "POWER8" CPU type, so that
you can always start QEMU with "-cpu POWER8" there, without the
need to know whether you are running on a POWER8, POWER8E or POWER8NVL
host machine.
However, we also have a "POWER8" CPU alias in the ppc_cpu_aliases list
(that is mainly useful for TCG). This leads to two cosmetical drawbacks:
If the user runs QEMU with "-cpu ?", we always claim that POWER8 is an
"alias for POWER8_v2.0" - which is simply not true when running with
KVM on POWER. And when using the 'query-cpu-definitions' QMP call,
there are currently two entries for "POWER8", one for the alias, and
one for the additional registered type.
To solve the two problems, we should rather update the "family" alias
instead of registering a new types. We then only have one "POWER8"
CPU definition around, an alias, which also points to the right
destination.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1396536
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
