| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This patch adds runtime instrumentation support for KVM guest. We need to
setup a save area for the runtime instrumentation-controls control block(RICCB)
and implement the necessary interfaces to live migrate the guest settings.
We setup the sie control block in a way, that the runtime
instrumentation instructions of a guest are handled by hardware.
We also add a capability KVM_CAP_S390_RI to make this feature opt-in as
it needs migration support.
Signed-off-by: Fan Zhang <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
smp_mb on vcpu destroy isn't paired with anything, violating pairing
rules, and seems to be useless.
Drop it.
Signed-off-by: Michael S. Tsirkin <[email protected]>
Message-Id: <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD
KVM: s390 features and fixes for 4.5 (kvm/next)
Some small cleanups
- use assignment instead of memcpy
- use %pK for kernel pointers
Changes regarding guest memory size
- Fix an off-by-one error in our guest memory interface (we might
use unnecessarily big page tables, e.g. 3 levels for a 2GB guest
instead of 2 levels)
- We now ask the machine about the max. supported guest address
and limit accordingly.
|
|
Verify that the guest maximum storage address is below the MHA (maximum
host address) value allowed on the host.
Acked-by: Michael Holzheu <[email protected]>
Reviewed-by: Cornelia Huck <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Signed-off-by: Guenther Hutzl <[email protected]>
Signed-off-by: Dominik Dingel <[email protected]>
[adopt to match recent limit,size changes]
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
While the userspace interface requests the maximum size the gmap code
expects to get a maximum address.
This error resulted in bigger page tables than necessary for some guest
sizes, e.g. a 2GB guest used 3 levels instead of 2.
At the same time we introduce KVM_S390_NO_MEM_LIMIT, which allows in a
bright future that a guest spans the complete 64 bit address space.
We also switch to TASK_MAX_SIZE for the initial memory size, this is a
cosmetic change as the previous size also resulted in a 4 level pagetable
creation.
Reported-by: David Hildenbrand <[email protected]>
Reviewed-by: Cornelia Huck <[email protected]>
Signed-off-by: Dominik Dingel <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
The s390dbf and trace events provide a debugfs interface.
If kptr_restrict is active, we should not expose kernel
pointers. We can fence the debugfs output by using %pK
instead of %p.
Cc: Kees Cook <[email protected]>
Reviewed-by: Cornelia Huck <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Replace two memcpy with proper assignment.
Suggested-by: Paolo Bonzini <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD
KVM: s390 features, kvm_get_vcpu_by_id and stat
Several features for s390
1. ESCA support (up to 248 vCPUs)
2. KVM detection: we can now detect if we support KVM (e.g. does KVM
under KVM work?)
kvm_stat:
1. cleanup the exit path
kvm_get_vcpu_by_id:
1. Use kvm_get_vcpu_by_id where appropriate
2. Apply a heuristic to optimize for ID VCPU == No. VCPU
|
|
rc already contains -ENOMEM, no need to assign it twice.
Signed-off-by: Christian Borntraeger <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
|
|
This evaluates always to 'true'.
Signed-off-by: Heiko Carstens <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
If we don't have support for virtualization (SIE), e.g. when running under
a hypervisor not supporting execution of the SIE instruction, we should
immediately abort loading the kvm module, as the SIE instruction cannot
be enabled dynamically.
Currently, the SIE instructions fails with an exception on a non-SIE
host, resulting in the guest making no progress, instead of failing hard.
Reviewed-by: Cornelia Huck <[email protected]>
Acked-by: Martin Schwidefsky <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch exposes the SIE capability (aka virtualization support) via
/proc/cpuinfo -> "features" as "sie".
As we don't want to expose this hwcap via elf, let's add a second,
"internal"/non-elf capability list. The content is simply concatenated
to the existing features when printing /proc/cpuinfo.
We also add the defines to elf.h to keep the hwcap stuff at a common
place.
Acked-by: Martin Schwidefsky <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch adds a way to check if the SIE with zArchitecture support is
available.
Acked-by: Martin Schwidefsky <[email protected]>
Acked-by: Cornelia Huck <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
sca_add_vpcu is not called for ucontrol guests. We must also not
apply the sca checking for sca_can_add_vcpu as ucontrol guests
do not have to follow the sca limits.
As common code already checks that id < KVM_MAX_VCPUS all other
data structures are safe as well.
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Now that we already have kvm and the VCPU id set for the VCPU, we can
convert sda_add_vcpu to look much more like sda_del_vcpu.
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Let's always set and clear the sda when enabling/disabling a VCPU.
Dealing with sda being set to something else makes no sense anymore
as we enable a VCPU in the SCA now after it has been registered at
the VM.
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
If something goes wrong in kvm_arch_vcpu_create, the VCPU has already
been added to the sca but will never be removed. Trying to create VCPUs
with duplicate ids (e.g. after a failed attempt) is problematic.
Also, when creating multiple VCPUs in parallel, we could theoretically
forget to set the correct SCA when the switch to ESCA happens just
before the VCPU is registered.
Let's add the VCPU to the SCA in kvm_arch_vcpu_postcreate, where we can
be sure that no duplicate VCPU with the same id is around and the VCPU
has already been registered at the VM. We also have to make sure to update
ECB at that point.
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Having no sca can never happen, even when something goes wrong when
switching to ESCA. Otherwise we would have a serious bug.
Let's remove this superfluous check.
Acked-by: Dominik Dingel <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
If CPUSTAT_ECALL_PEND isn't set, we can't have an external call pending,
so we can directly avoid taking the lock.
Acked-by: Christian Borntraeger <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch allows s390 to have more than 64 VCPUs for a guest (up to
248 for memory usage considerations), if supported by the underlaying
hardware (sclp.has_esca).
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch adds code that performs transparent switch to Extended
SCA on addition of 65th VCPU in a VM. Disposal of ESCA is added too.
The entier ESCA functionality, however, is still not enabled.
The enablement will be provided in a separate patch.
This patch also uses read/write lock protection of SCA and its subfields for
possible disposal at the BSCA-to-ESCA transition. While only Basic SCA needs such
a protection (for the swap), any SCA access is now guarded.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch updates the routines (sca_*) to provide transparent access
to and manipulation on the data for both Basic and Extended SCA in use.
The kvm.arch.sca is generalized to (void *) to handle BSCA/ESCA cases.
Also the kvm.arch.use_esca flag is provided.
The actual functionality is kept the same.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch adds new structures and updates some existing ones to
provide the base for Extended SCA functionality.
The old sca_* structures were renamed to bsca_* to keep things uniform.
The access to fields of SIGP controls were turned into bitfields instead
of hardcoded bitmasks.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch provides SCA-aware helpers to create/delete a VCPU.
This is to prepare for upcoming introduction of Extended SCA support.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch generalizes access to the SIGP controls, which is a part of SCA.
This is to prepare for upcoming introduction of Extended SCA support.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
This patch generalizes access to the IPTE controls, which is a part of SCA.
This is to prepare for upcoming introduction of Extended SCA support.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Introduce sclp.has_hvs and sclp.has_esca to provide a way for kvm to check
whether the extended-SCA and the home-virtual-SCA facilities are available.
Signed-off-by: Eugene (jno) Dvurechenski <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Let's rewrite this function to better reflect how we actually handle
exit_code. By dropping out early we can save a few cycles. This
especially speeds up sie exits caused by host irqs.
Also, let's move the special -EOPNOTSUPP for intercepts to
the place where it belongs and convert it to -EREMOTE.
Reviewed-by: Dominik Dingel <[email protected]>
Reviewed-by: Cornelia Huck <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
KVM creates debugfs files to export VM statistics to userland. To be
able to remove them on kvm exit it tracks the files' dentries.
Since their parent directory is also tracked and since each parent
direntry knows its children we can easily remove them by using
debugfs_remove_recursive(kvm_debugfs_dir). Therefore we don't
need the extra tracking in the kvm_stats_debugfs_item anymore.
Signed-off-by: Janosch Frank <[email protected]>
Reviewed-By: Sascha Silbe <[email protected]>
Acked-by: Christian Borntraeger <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Usually, VCPU ids match the array index. So let's try a fast
lookup first before falling back to the slow iteration.
Suggested-by: Christian Borntraeger <[email protected]>
Reviewed-by: Dominik Dingel <[email protected]>
Reviewed-by: Christian Borntraeger <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
|
|
Let's reuse the new common function for VPCU lookup by id.
Reviewed-by: Christian Borntraeger <[email protected]>
Reviewed-by: Dominik Dingel <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Christian Borntraeger <[email protected]>
[split out the new function into a separate patch]
|
|
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
As kvm_mmu_get_page() was changed so that every parent pointer would not
get into the sp->parent_ptes chain before the entry pointed to by it was
set properly, we can use the for_each_rmap_spte macro instead of
pte_list_walk().
Signed-off-by: Takuya Yoshikawa <[email protected]>
Cc: Xiao Guangrong <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
link_shadow_page()
Every time kvm_mmu_get_page() is called with a non-NULL parent_pte
argument, link_shadow_page() follows that to set the parent entry so
that the new mapping will point to the returned page table.
Moving parent_pte handling there allows to clean up the code because
parent_pte is passed to kvm_mmu_get_page() just for mark_unsync() and
mmu_page_add_parent_pte().
In addition, the patch avoids calling mark_unsync() for other parents in
the sp->parent_ptes chain than the newly added parent_pte, because they
have been there since before the current page fault handling started.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Cc: Xiao Guangrong <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
Make kvm_mmu_alloc_page() do just what its name tells to do, and remove
the extra allocation error check and zero-initialization of parent_ptes:
shadow page headers allocated by kmem_cache_zalloc() are always in the
per-VCPU pools.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
At some call sites of rmap_get_first() and rmap_get_next(), BUG_ON is
placed right after the call to detect unrelated sptes which must not be
found in the reverse-mapping list.
Move this check in rmap_get_first/next() so that all call sites, not
just the users of the for_each_rmap_spte() macro, will be checked the
same way.
One thing to keep in mind is that kvm_mmu_unlink_parents() also uses
rmap_get_first() to handle parent sptes. The change will not break it
because parent sptes are present, at least until drop_parent_pte()
actually unlinks them, and not mmio-sptes.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
is_rmap_spte(), originally named is_rmap_pte(), was introduced when the
simple reverse mapping was implemented by commit cd4a4e5374110444
("[PATCH] KVM: MMU: Implement simple reverse mapping"). At that point,
its role was clear and only rmap_add() and rmap_remove() were using it
to select sptes that need to be reverse-mapped.
Independently of that, is_shadow_present_pte() was first introduced by
commit c7addb902054195b ("KVM: Allow not-present guest page faults to
bypass kvm") to do bypass_guest_pf optimization, which does not exist
any more.
These two seem to have changed their roles somewhat, and is_rmap_spte()
just calls is_shadow_present_pte() now.
Since using both of them without clear distinction just makes the code
confusing, remove is_rmap_spte().
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
mmu_set_spte()'s code is based on the assumption that the emulate
parameter has a valid pointer value if set_spte() returns true and
write_fault is not zero. In other cases, emulate may be NULL, so a
NULL-check is needed.
Stop passing emulate pointer and make mmu_set_spte() return the emulate
value instead to clean up this complex interface. Prefetch functions
can just throw away the return value.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
Both __mmu_unsync_walk() and mmu_pages_clear_parents() have three line
code which clears a bit in the unsync child bitmap; the former places it
inside a loop block and uses a few goto statements to jump to it.
A new helper function, clear_unsync_child_bit(), makes the code cleaner.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
New struct kvm_rmap_head makes the code type-safe to some extent.
Signed-off-by: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
In another patch kvm_is_visible_gfn is maken return bool due to this
function only returns zero or one as its return value, let's also make
kvmppc_visible_gpa return bool to keep consistent.
No functional change.
Signed-off-by: Yaowei Bai <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
This patch makes kvm_para_has_feature return bool due to this
particular function only using either one or zero as its return
value.
No functional change.
Signed-off-by: Yaowei Bai <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
This patch makes kvm_is_visible_gfn return bool due to this particular
function only using either one or zero as its return value.
No functional change.
Signed-off-by: Yaowei Bai <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
"kmem_cache_destroy"
The kmem_cache_destroy() function tests whether its argument is NULL
and then returns immediately. Thus the test around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
Commit 7a1638ce4220 ("nEPT: Redefine EPT-specific link_shadow_page()",
2013-08-05) says:
Since nEPT doesn't support A/D bit, we should not set those bit
when building the shadow page table.
but this is not necessary. Even though nEPT doesn't support A/D
bits, and hence the vmcs12 EPT pointer will never enable them,
we always use them for shadow page tables if available (see
construct_eptp in vmx.c). So we can set the A/D bits freely
in the shadow page table.
This patch hence basically reverts commit 7a1638ce4220.
Cc: Yang Zhang <[email protected]>
Cc: Takuya Yoshikawa <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
Poor #AC was so unimportant until a few days ago that we were
not even tracing its name correctly. But now it's all over
the place.
Cc: [email protected]
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
RDTSCP was never supported for AMD CPUs, which nobody noticed because
Linux does not use it. But exactly the fact that Linux does not
use it makes the implementation very simple; we can freely trash
MSR_TSC_AUX while running the guest.
Cc: Joerg Roedel <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
If we do not do this, it is not properly saved and restored across
migration. Windows notices due to its self-protection mechanisms,
and is very upset about it (blue screen of death).
Cc: Radim Krcmar <[email protected]>
Cc: [email protected]
Signed-off-by: Paolo Bonzini <[email protected]>
|
|
A new vcpu exit is introduced to notify the userspace of the
changes in Hyper-V SynIC configuration triggered by guest writing to the
corresponding MSRs.
Changes v4:
* exit into userspace only if guest writes into SynIC MSR's
Changes v3:
* added KVM_EXIT_HYPERV types and structs notes into docs
Signed-off-by: Andrey Smetanin <[email protected]>
Reviewed-by: Roman Kagan <[email protected]>
Signed-off-by: Denis V. Lunev <[email protected]>
CC: Gleb Natapov <[email protected]>
CC: Paolo Bonzini <[email protected]>
CC: Roman Kagan <[email protected]>
CC: Denis V. Lunev <[email protected]>
CC: [email protected]
Signed-off-by: Paolo Bonzini <[email protected]>
|