When explicitly enabling unmigratable flags using "-cpu host"
(e.g. "-cpu host,+invtsc"), the requested feature won't be
enabled because cpu->migratable is true by default.
This is inconsistent with all other CPU models, which don't have
the "migratable" option, making "+invtsc" work without the need
for extra options.
This happens because x86_cpu_filter_features() uses
cpu->migratable as an argument for
x86_cpu_get_supported_feature_word(). This is not useful
because:
2) on "-cpu host" it only makes QEMU disable features that were
explicitly enabled in the command-line;
1) on all the other CPU models, cpu->migratable is already false.
The fix is to just use 'false' as an argument to
x86_cpu_get_supported_feature_word() in
x86_cpu_filter_features().
Note that:
* This won't change anything for people using using
"-cpu host" or "-cpu host,migratable=<on|off>" (with no extra
features) because the x86_cpu_get_supported_feature_word() call
on the cpu->host_features check uses cpu->migratable as
argument.
* This won't change anything for any CPU model except "host"
because they all have cpu->migratable == false (and only "host"
has the "migratable" property that allows it to be changed).
* This will only change things for people using "-cpu host,+<feature>",
where <feature> is a non-migratable feature. The only existing
named non-migratable feature is "invtsc".
In other words, this change will only affect people using
"-cpu host,+invtsc" (that will now get what they asked for: the
invtsc flag will be enabled). All other use cases are unaffected.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Fill the "unavailable-features" field on the x86 implementation
of query-cpu-definitions.
Cc: Jiri Denemark <jdenemar@redhat.com>
Cc: libvir-list@redhat.com
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
When probing for CPU model information, we need to reuse the code
that initializes CPUID fields, but not the remaining side-effects
of x86_cpu_realizefn(). Move that code to a separate function
that can be reused later.
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
TYPE_X86_CPU now call cpu_exec_init() on realize, so we don't
need to set cannot_destroy_with_object_finalize_yet anymore.
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Assume that KVM would have returned the same on subsequent runs.
Abstract the memoizaiton pattern into macros and call it memorize as
adding the r makes it less obscure.
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Cluster x2APIC cannot work without KVM's x2apic API when the maximal
APIC ID is greater than 8 and only KVM's LAPIC can support x2APIC, so we
forbid other APICs and also the old KVM case with less than 9, to
simplify the code.
There is no point in enabling EIM in forbidden APICs, so we keep it
enabled only for the KVM APIC; unconditionally, because making the
option depend on KVM version would be a maintanance burden.
Old QEMUs would enable eim whenever intremap was on, which would trick
guests into thinking that they can enable cluster x2APIC even if any
interrupt destination would get clamped to 8 bits.
Depending on your configuration, QEMU could notice that the destination
LAPIC is not present and report it with a very non-obvious:
KVM: injection failed, MSI lost (Operation not permitted)
Or the guest could say something about unexpected interrupts, because
clamping leads to aliasing so interrupts were being delivered to
incorrect VCPUs.
KVM_X2APIC_API is the feature that allows us to enable EIM for KVM.
QEMU 2.7 allowed EIM whenever interrupt remapping was enabled. In order
to keep backward compatibility, we again allow guests to misbehave in
non-obvious ways, and make it the default for old machine types.
A user can enable the buggy mode it with "x-buggy-eim=on".
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Every configuration has only up to one APIC class and we'll be extending
the class with a function that can be called without an instanced
object, so a direct access to the class is convenient.
This patch will break compilation if some code uses apic_get_class()
with CONFIG_USER_ONLY.
Suggested-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
x86_cpu_filter_features() will be reused by code that shouldn't
print any warning. Move the warning code to a new
x86_cpu_report_filtered_features() function, and call it from
x86_cpu_realizefn().
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of treating the FP and SSE bits as special cases, add
them to the x86_ext_save_areas array. This will simplify the code
that calculates the supported xsave components and the size of
the xsave area.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of keeping the aliases inside the feature name arrays and
require parsing the strings, just register alias properties
manually. This simplifies the code for property registration and
lookup.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of translating the feature name entries when adding
property names, store the actual property names in the feature
name array.
For reference, here is the full list of functions that use
FeatureWordInfo::feat_names:
* x86_cpu_get_migratable_flags(): not affected, as it just
check for non-NULL values.
* report_unavailable_features(): informative only. It will
start printing feature names with hyphens.
* x86_cpu_list(): informative only. It will start printing
feature names with hyphens
* x86_cpu_register_feature_bit_props(): not affected, as it
was already calling feat2prop(). Now we can remove the
feat2prop() calls safely.
So, the only user-visible effect of this patch are the new names
being used in help and error messages for users.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of using custom feature name lookup code for
plus_features/minus_features, save the property names used in
"[+-]feature" and use object_property_set_bool() to set them.
We don't need a feat2prop() call because we now have alias
properties for the old names containing underscores.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Registering the actual names containing underscores as aliases
will allow management software to be aware that the old
compatibility names are suported, and will make feat2prop() calls
unnecessary when using feature names.
Also, this will help us avoid making the code support underscores
on feature names that never had them in the first place. e.g.
"+tsc_deadline" was never supported and doesn't need to be
translated to "+tsc-deadline".
In other word: this will require less magic translation of
strings, and simple 1:1 match between the config options and
actual QOM properties.
Note that the underscores are still present in the
FeatureWordInfo::feat_names arrays, because
add_flagname_to_bitmaps() needs them to be kept. The next patches
will remove add_flagname_to_bitmaps() and will allow us to
finally remove the aliases from feat_names.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
VME is already disabled automatically when using TCG. So, instead
of pretending it is there when reporting CPU model data on
query-cpu-* QMP commands (making every CPU model to be reported
as not runnable), we can disable it by default on all CPU models
when using TCG.
Do that by adding a tcg_default_props array that will work like
kvm_default_props.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of using the builtin_x86_defs array, use the QOM subclass
list to list CPU models on "-cpu ?" and "query-cpu-definitions".
Signed-off-by: Andreas Färber <afaerber@suse.de>
[ehabkost: copied code from a patch by Andreas:
"target-i386: QOM'ify CPU", from March 2012]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Unlike the other sections, they are pretty specific to a particular CPU.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The qemu will crash when info ioapic through hmp if irqchip
is split. Below message is splat:
KVM_GET_IRQCHIP failed: Unknown error -6
This patch fix it by dumping the ioapic state from the qemu
emulated ioapic if irqchip is split.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Message-Id: <1474602456-3232-1-git-send-email-wanpeng.li@hotmail.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Message-ID: <20160923090824.GF15411@pxdev.xzpeter.org>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Current CPU definition for AMD Opteron third generation includes
features like SSE4a and LAHF_LM support in emulated CPUID. These
features are present in K8 rev.E or K10 CPUs and later. However,
current G3 family and model describe 2nd generation K8 cores instead.
This is incorrect but was considered harmless until our tests found a
problem with linux kernels >= 3.10 (and maybe earlier) which specifically
check for Opteron K8 model when parsing CPUID leaf 0x80000001:
http://lxr.free-electrons.com/source/arch/x86/kernel/cpu/amd.c?v=3.16#L552
This code will disable LAHF_LM feature in /proc/cpuinfo if model number
is inconsistent.
This change sets Opteron_G3 family/model/stepping to 16/2/3 which is
a proper Opteron 3rd generation 2350 CPU.
Signed-off-by: Evgeny Yakovlev <eyakovlev@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: Richard Henderson <rth@twiddle.net>
CC: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
A regression was introduced by commit 96193c22a "target-i386:
Move xsave component mask to features array": all
CPUID[EAX=0xD,ECX=0]:EAX bits were being reported as unmigratable
because they don't have feature names defined. This broke
"-cpu host" because it enables only migratable features by
default.
This adds a new field to FeatureWordInfo: migratable_flags, which
will make those features be reported as migratable even if they
don't have a property name defined.
Reported-by: Wanpeng Li <wanpeng.li@hotmail.com>
Cc: Paolo Bonzini <bonzini@gnu.org>
Reviewed-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The global variables are not necessary because we can check KVM
feature flags in X86CPU directly.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
This will ensure all checks for features[FEAT_KVM] in the code
will be correct in case the KVM CPUID leaf is completely
disabled.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The global variable is not necessary because we can check
cpu->hyperv_time directly.
We just need to ensure cpu->hyperv_time will be cleared if the
feature is not really being exposed to the guest due to missing
KVM_CAP_HYPERV_TIME capability.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The global variable is not necessary because we can check
cpu->hyperv_vapic directly.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The global variable is not necessary because we can check the CPU
feature flags directly.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
This will reuse the existing check/enforce logic in
x86_cpu_filter_features() to check the xsave component bits
against GET_SUPPORTED_CPUID.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of doing complex calculations and calling
kvm_arch_get_supported_cpuid() inside cpu_x86_cpuid(), calculate
the set of required XSAVE components earlier, at realize time.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Move the xsave area size calculation from cpu_x86_cpuid() inside
its own function. While doing it, change it to use the XSAVE area
struct sizes for the initial size, instead of the magic 0x240
number.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of assigning individual bits in a loop, just copy the
values from ena_mask.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of checking both env->features and ena_mask at two
different places in the CPUID code, initialize ena_mask based on
the features that are enabled for the CPU, and then clear
unsupported bits based on kvm_arch_get_supported_cpuid().
The results should be exactly the same, but it will make it
easier to move the mask calculation elsewhare, and reuse
x86_cpu_filter_features() for the kvm_arch_get_supported_cpuid()
check.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The code that calculates the set of supported XSAVE components on
CPUID looks at ext_save_areas to find out which components should
be enabled. However, if there are zeroed entries in the
ext_save_areas array, the
((env->features[esa->feature] & esa->bits) == esa->bits)
check will always succeed and QEMU will unconditionally try to
enable the component.
Luckily this never caused any problems because the only missing
entry in ext_save_areas is the PT State component (bit 8), and
KVM currently doesn't support it (so it was cleared on ena_mask).
But the code was still incorrect and would break if KVM starts
returning CPUID[EAX=0xD,ECX=0].EAX[bit 8] as supported on
GET_SUPPORTED_CPUID.
Fix the problem by changing the code to not enable a XSAVE
component if ExtSaveArea::bits is zero.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
It makes it easier to guarantee the arrays are the right size,
and to find information when looking at the code.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Those are unneeded now that CPUState nr_{cores,threads} is always
initialized.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
SVM needs CPUID[0x8000000A] to be available. So if SVM is enabled
in a CPU model or explicitly in the command-line, adjust CPUID
xlevel to expose the CPUID[0x8000000A] leaf.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of requiring users and management software to be aware of
required CPUID level/xlevel/xlevel2 values for each feature,
automatically increase those values when features need them.
This was already done for CPUID[7].EBX, and is now made generic
for all CPUID feature flags. Unit test included, to make sure we
don't break ABI on older machine-types and don't mess with the
CPUID level values if they are explicitly set by the user.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Instead of using cpuid_level, use an empty struct as a marker
(like we already did with {start,end}_init_save). This will avoid
accidentaly resetting the wrong fields if we change the field
ordering on CPUX86State.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
No CPU model in builtin_x86_defs has xlevel2 set, so it is always
zero. Delete the field.
Note that this is not an user-visible change. It doesn't remove
the ability to set xlevel2 on the command-line, it just removes
an unused field in builtin_x86_defs.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
CPUState is a fairly common pointer to pass to these helpers. This means
if you need other arguments for the async_run_on_cpu case you end up
having to do a g_malloc to stuff additional data into the routine. For
the current users this isn't a massive deal but for MTTCG this gets
cumbersome when the only other parameter is often an address.
This adds the typedef run_on_cpu_func for helper functions which has an
explicit CPUState * passed as the first parameter. All the users of
run_on_cpu and async_run_on_cpu have had their helpers updated to use
CPUState where available.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
[Sergey Fedorov:
- eliminate more CPUState in user data;
- remove unnecessary user data passing;
- fix target-s390x/kvm.c and target-s390x/misc_helper.c]
Signed-off-by: Sergey Fedorov <sergey.fedorov@linaro.org>
Acked-by: David Gibson <david@gibson.dropbear.id.au> (ppc parts)
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> (s390 parts)
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <1470158864-17651-3-git-send-email-alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The parsing of KVM_SET_LAPIC's input depends on the current value of the
APIC base MSR---which indeed is stored in APICCommonState---but for historical
reasons APIC base is set through KVM_SET_SREGS together with cr8 (which is
really just the APIC TPR) and the actual "special CPU registers".
APIC base must now be set before the actual LAPIC registers, so do that
in kvm_apic_put. It will be set again to the same value with KVM_SET_SREGS,
but that's not a big issue.
This only happens since Linux 4.8, which checks for x2apic mode in
KVM_SET_LAPIC. However it's really a QEMU bug; until the recent
commit 78d6a05 ("x86/lapic: Load LAPIC state at post_load", 2016-09-13)
QEMU was indeed setting APIC base (via KVM_SET_SREGS) before the other
LAPIC registers.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This avoids a double hand-full of magic numbers in the
xsave and xrstor helper functions.
Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
In user-mode emulation env->idt.base memory is
allocated in linux-user/main.c with
size 8*512 = 4096 (for 64-bit).
When fake interrupt EXCP_SYSCALL is thrown
do_interrupt_user checks destination privilege level
for this fake exception, and tries to read 4 bytes
at address base + (256 * 2^4)=4096, that causes
segfault.
Privlege level was checked only for int's, so lets
read dpl from memory only for this case.
Signed-off-by: Stanislav Shmarov <snarpix@gmail.com>
Message-Id: <1473773008-2588376-1-git-send-email-snarpix@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Make sure reset zeroes TSC_AUX, XCR0, PKRU. Move XSTATE_BV from the
"vmstate only" section to the "KVM only" section.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Load the LAPIC state during post_load (rather than when the CPU
starts).
This allows an interrupt to be delivered from the ioapic to
the lapic prior to cpu loading, in particular the RTC that starts
ticking as soon as we load it's state.
Fixes a case where Windows hangs after migration due to RTC interrupts
disappearing.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some software algorithms are based on the hardware's cache info, for example,
for x86 linux kernel, when cpu1 want to wakeup a task on cpu2, cpu1 will trigger
a resched IPI and told cpu2 to do the wakeup if they don't share low level
cache. Oppositely, cpu1 will access cpu2's runqueue directly if they share llc.
The relevant linux-kernel code as bellow:
static void ttwu_queue(struct task_struct *p, int cpu)
{
struct rq *rq = cpu_rq(cpu);
......
if (... && !cpus_share_cache(smp_processor_id(), cpu)) {
......
ttwu_queue_remote(p, cpu); /* will trigger RES IPI */
return;
}
......
ttwu_do_activate(rq, p, 0); /* access target's rq directly */
......
}
In real hardware, the cpus on the same socket share L3 cache, so one won't
trigger a resched IPIs when wakeup a task on others. But QEMU doesn't present a
virtual L3 cache info for VM, then the linux guest will trigger lots of RES IPIs
under some workloads even if the virtual cpus belongs to the same virtual socket.
For KVM, there will be lots of vmexit due to guest send IPIs.
The workload is a SAP HANA's testsuite, we run it one round(about 40 minuates)
and observe the (Suse11sp3)Guest's amounts of RES IPIs which triggering during
the period:
No-L3 With-L3(applied this patch)
cpu0: 363890 44582
cpu1: 373405 43109
cpu2: 340783 43797
cpu3: 333854 43409
cpu4: 327170 40038
cpu5: 325491 39922
cpu6: 319129 42391
cpu7: 306480 41035
cpu8: 161139 32188
cpu9: 164649 31024
cpu10: 149823 30398
cpu11: 149823 32455
cpu12: 164830 35143
cpu13: 172269 35805
cpu14: 179979 33898
cpu15: 194505 32754
avg: 268963.6 40129.8
The VM's topology is "1*socket 8*cores 2*threads".
After present virtual L3 cache info for VM, the amounts of RES IPIs in guest
reduce 85%.
For KVM, vcpus send IPIs will cause vmexit which is expensive, so it can cause
severe performance degradation. We had tested the overall system performance if
vcpus actually run on sparate physical socket. With L3 cache, the performance
improves 7.2%~33.1%(avg:15.7%).
Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The kvm_pv_unhalt feature doesn't work if kernel_irqchip is
disabled, so we need to report it as unsupported.
Tested-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>