Gleb Natapov [Tue, 24 Jan 2012 13:06:05 +0000 (15:06 +0200)]
KVM: x86: reset edge sense circuit of i8259 on init
The spec says that during initialization "The edge sense circuit is
reset which means that following initialization an interrupt request
(IR) input must make a low-to-high transition to generate an interrupt",
but currently if edge triggered interrupt is in IRR it is delivered
after i8259 initialization.
Alexander Graf [Mon, 16 Jan 2012 18:12:11 +0000 (19:12 +0100)]
KVM: PPC: Add HPT preallocator
We're currently allocating 16MB of linear memory on demand when creating
a guest. That does work some times, but finding 16MB of linear memory
available in the system at runtime is definitely not a given.
So let's add another command line option similar to the RMA preallocator,
that we can use to keep a pool of page tables around. Now, when a guest
gets created it has a pretty low chance of receiving an OOM.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Mon, 16 Jan 2012 15:50:10 +0000 (16:50 +0100)]
KVM: PPC: Convert RMA allocation into generic code
We have code to allocate big chunks of linear memory on bootup for later use.
This code is currently used for RMA allocation, but can be useful beyond that
extent.
Make it generic so we can reuse it for other stuff later.
Signed-off-by: Alexander Graf <agraf@suse.de> Acked-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Avi Kivity <avi@redhat.com>
Since it doesn't make sense to run the init function on any non-supported
platform, we can just call our "is this platform supported?" function and
bail out of init() if it's not.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Thu, 12 Jan 2012 20:09:51 +0000 (20:09 +0000)]
KVM: Move gfn_to_memslot() to kvm_host.h
This moves __gfn_to_memslot() and search_memslots() from kvm_main.c to
kvm_host.h to reduce the code duplication caused by the need for
non-modular code in arch/powerpc/kvm/book3s_hv_rm_mmu.c to call
gfn_to_memslot() in real mode.
Rather than putting gfn_to_memslot() itself in a header, which would
lead to increased code size, this puts __gfn_to_memslot() in a header.
Then, the non-modular uses of gfn_to_memslot() are changed to call
__gfn_to_memslot() instead. This way there is only one place in the
source code that needs to be changed should the gfn_to_memslot()
implementation need to be modified.
On powerpc, the Book3S HV style of KVM has code that is called from
real mode which needs to call gfn_to_memslot() and thus needs this.
(Module code is allocated in the vmalloc region, which can't be
accessed in real mode.)
With this, we can remove builtin_gfn_to_memslot() from book3s_hv_rm_mmu.c.
Signed-off-by: Paul Mackerras <paulus@samba.org> Acked-by: Avi Kivity <avi@redhat.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Julian Stecklina [Mon, 16 Jan 2012 13:02:20 +0000 (14:02 +0100)]
KVM: Don't mistreat edge-triggered INIT IPI as INIT de-assert. (LAPIC)
If the guest programs an IPI with level=0 (de-assert) and trig_mode=0 (edge),
it is erroneously treated as INIT de-assert and ignored, but to quote the
spec: "For this delivery mode [INIT de-assert], the level flag must be set to
0 and trigger mode flag to 1."
find_index_from_host_irq returns 0 on error
but callers assume < 0 on error. This should
not matter much: an out of range irq should never happen since
irq handler was registered with this irq #,
and even if it does we get a spurious msix irq in guest
and typically nothing terrible happens.
Still, better to make it consistent.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Avi Kivity [Mon, 16 Jan 2012 13:08:45 +0000 (15:08 +0200)]
KVM: x86 emulator: Remove byte-sized MOVSX/MOVZX hack
Currently we treat MOVSX/MOVZX with a byte source as a byte instruction,
and change the destination operand size with a hack. Change it to be
a word instruction, so the destination receives its natural size, and
change the source to be SrcMem8.
Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Scott Wood [Wed, 11 Jan 2012 13:37:35 +0000 (13:37 +0000)]
KVM: PPC: refer to paravirt docs in header file
Instead of keeping separate copies of struct kvm_vcpu_arch_shared (one in
the code, one in the docs) that inevitably fail to be kept in sync
(already sr[] is missing from the doc version), just point to the header
file as the source of documentation on the contents of the magic page.
Signed-off-by: Scott Wood <scottwood@freescale.com> Acked-by: Avi Kivity <avi@redhat.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Sat, 7 Jan 2012 01:07:38 +0000 (02:07 +0100)]
KVM: PPC: Rename MMIO register identifiers
We need the KVM_REG namespace for generic register settings now, so
let's rename the existing users to something different, enabling
us to reuse the namespace for more visible interfaces.
While at it, also move these private constants to a private header.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:26:50 +0000 (12:26 +0000)]
KVM: PPC: Move kvm_vcpu_ioctl_[gs]et_one_reg down to platform-specific code
This moves the get/set_one_reg implementation down from powerpc.c into
booke.c, book3s_pr.c and book3s_hv.c. This avoids #ifdefs in C code,
but more importantly, it fixes a bug on Book3s HV where we were
accessing beyond the end of the kvm_vcpu struct (via the to_book3s()
macro) and corrupting memory, causing random crashes and file corruption.
On Book3s HV we only accept setting the HIOR to zero, since the guest
runs in supervisor mode and its vectors are never offset from zero.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de>
[agraf update to apply on top of changed ONE_REG patches] Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Wed, 14 Sep 2011 19:45:23 +0000 (21:45 +0200)]
KVM: PPC: Add support for explicit HIOR setting
Until now, we always set HIOR based on the PVR, but this is just wrong.
Instead, we should be setting HIOR explicitly, so user space can decide
what the initial HIOR value is - just like on real hardware.
We keep the old PVR based way around for backwards compatibility, but
once user space uses the SET_ONE_REG based method, we drop the PVR logic.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Wed, 14 Sep 2011 08:02:41 +0000 (10:02 +0200)]
KVM: PPC: Add generic single register ioctls
Right now we transfer a static struct every time we want to get or set
registers. Unfortunately, over time we realize that there are more of
these than we thought of before and the extensibility and flexibility of
transferring a full struct every time is limited.
So this is a new approach to the problem. With these new ioctls, we can
get and set a single register that is identified by an ID. This allows for
very precise and limited transmittal of data. When we later realize that
it's a better idea to shove over multiple registers at once, we can reuse
most of the infrastructure and simply implement a GET_MANY_REGS / SET_MANY_REGS
interface.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Liu Yu [Tue, 20 Dec 2011 14:42:56 +0000 (14:42 +0000)]
KVM: PPC: booke: Add booke206 TLB trace
The existing kvm_stlb_write/kvm_gtlb_write were a poor match for
the e500/book3e MMU -- mas1 was passed as "tid", mas2 was limited
to "unsigned int" which will be a problem on 64-bit, mas3/7 got
split up rather than treated as a single 64-bit word, etc.
Signed-off-by: Liu Yu <yu.liu@freescale.com>
[scottwood@freescale.com: made mas2 64-bit, and added mas8 init] Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Thu, 15 Dec 2011 02:03:22 +0000 (02:03 +0000)]
KVM: PPC: Book3s HV: Implement get_dirty_log using hardware changed bit
This changes the implementation of kvm_vm_ioctl_get_dirty_log() for
Book3s HV guests to use the hardware C (changed) bits in the guest
hashed page table. Since this makes the implementation quite different
from the Book3s PR case, this moves the existing implementation from
book3s.c to book3s_pr.c and creates a new implementation in book3s_hv.c.
That implementation calls kvmppc_hv_get_dirty_log() to do the actual
work by calling kvm_test_clear_dirty on each page. It iterates over
the HPTEs, clearing the C bit if set, and returns 1 if any C bit was
set (including the saved C bit in the rmap entry).
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Thu, 15 Dec 2011 02:02:47 +0000 (02:02 +0000)]
KVM: PPC: Book3S HV: Use the hardware referenced bit for kvm_age_hva
This uses the host view of the hardware R (referenced) bit to speed
up kvm_age_hva() and kvm_test_age_hva(). Instead of removing all
the relevant HPTEs in kvm_age_hva(), we now just reset their R bits
if set. Also, kvm_test_age_hva() now scans the relevant HPTEs to
see if any of them have R set.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Thu, 15 Dec 2011 02:02:02 +0000 (02:02 +0000)]
KVM: PPC: Book3s HV: Maintain separate guest and host views of R and C bits
This allows both the guest and the host to use the referenced (R) and
changed (C) bits in the guest hashed page table. The guest has a view
of R and C that is maintained in the guest_rpte field of the revmap
entry for the HPTE, and the host has a view that is maintained in the
rmap entry for the associated gfn.
Both view are updated from the guest HPT. If a bit (R or C) is zero
in either view, it will be initially set to zero in the HPTE (or HPTEs),
until set to 1 by hardware. When an HPTE is removed for any reason,
the R and C bits from the HPTE are ORed into both views. We have to
be careful to read the R and C bits from the HPTE after invalidating
it, but before unlocking it, in case of any late updates by the hardware.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Thu, 15 Dec 2011 02:01:10 +0000 (02:01 +0000)]
KVM: PPC: Book3S HV: Keep HPTE locked when invalidating
This reworks the implementations of the H_REMOVE and H_BULK_REMOVE
hcalls to make sure that we keep the HPTE locked and in the reverse-
mapping chain until we have finished invalidating it. Previously
we would remove it from the chain and unlock it before invalidating
it, leaving a tiny window when the guest could access the page even
though we believe we have removed it from the guest (e.g.,
kvm_unmap_hva() has been called for the page and has found no HPTEs
in the chain). In addition, we'll need this for future patches where
we will need to read the R and C bits in the HPTE after invalidating
it.
Doing this required restructuring kvmppc_h_bulk_remove() substantially.
Since we want to batch up the tlbies, we now need to keep several
HPTEs locked simultaneously. In order to avoid possible deadlocks,
we don't spin on the HPTE bitlock for any except the first HPTE in
a batch. If we can't acquire the HPTE bitlock for the second or
subsequent HPTE, we terminate the batch at that point, do the tlbies
that we have accumulated so far, unlock those HPTEs, and then start
a new batch to do the remaining invalidations.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Matt Evans [Wed, 7 Dec 2011 16:55:57 +0000 (16:55 +0000)]
KVM: PPC: Add KVM_CAP_NR_VCPUS and KVM_CAP_MAX_VCPUS
PPC KVM lacks these two capabilities, and as such a userland system must assume
a max of 4 VCPUs (following api.txt). With these, a userland can determine
a more realistic limit.
Signed-off-by: Matt Evans <matt@ozlabs.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:38:51 +0000 (12:38 +0000)]
KVM: PPC: Allow for read-only pages backing a Book3S HV guest
With this, if a guest does an H_ENTER with a read/write HPTE on a page
which is currently read-only, we make the actual HPTE inserted be a
read-only version of the HPTE. We now intercept protection faults as
well as HPTE not found faults, and for a protection fault we work out
whether it should be reflected to the guest (e.g. because the guest HPTE
didn't allow write access to usermode) or handled by switching to
kernel context and calling kvmppc_book3s_hv_page_fault, which will then
request write access to the page and update the actual HPTE.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:37:21 +0000 (12:37 +0000)]
KVM: Add barriers to allow mmu_notifier_retry to be used locklessly
This adds an smp_wmb in kvm_mmu_notifier_invalidate_range_end() and an
smp_rmb in mmu_notifier_retry() so that mmu_notifier_retry() will give
the correct answer when called without kvm->mmu_lock being held.
PowerPC Book3S HV KVM wants to use a bitlock per guest page rather than
a single global spinlock in order to improve the scalability of updates
to the guest MMU hashed page table, and so needs this.
Signed-off-by: Paul Mackerras <paulus@samba.org> Acked-by: Avi Kivity <avi@redhat.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:38:05 +0000 (12:38 +0000)]
KVM: PPC: Implement MMU notifiers for Book3S HV guests
This adds the infrastructure to enable us to page out pages underneath
a Book3S HV guest, on processors that support virtualized partition
memory, that is, POWER7. Instead of pinning all the guest's pages,
we now look in the host userspace Linux page tables to find the
mapping for a given guest page. Then, if the userspace Linux PTE
gets invalidated, kvm_unmap_hva() gets called for that address, and
we replace all the guest HPTEs that refer to that page with absent
HPTEs, i.e. ones with the valid bit clear and the HPTE_V_ABSENT bit
set, which will cause an HDSI when the guest tries to access them.
Finally, the page fault handler is extended to reinstantiate the
guest HPTE when the guest tries to access a page which has been paged
out.
Since we can't intercept the guest DSI and ISI interrupts on PPC970,
we still have to pin all the guest pages on PPC970. We have a new flag,
kvm->arch.using_mmu_notifiers, that indicates whether we can page
guest pages out. If it is not set, the MMU notifier callbacks do
nothing and everything operates as before.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:36:37 +0000 (12:36 +0000)]
KVM: PPC: Implement MMIO emulation support for Book3S HV guests
This provides the low-level support for MMIO emulation in Book3S HV
guests. When the guest tries to map a page which is not covered by
any memslot, that page is taken to be an MMIO emulation page. Instead
of inserting a valid HPTE, we insert an HPTE that has the valid bit
clear but another hypervisor software-use bit set, which we call
HPTE_V_ABSENT, to indicate that this is an absent page. An
absent page is treated much like a valid page as far as guest hcalls
(H_ENTER, H_REMOVE, H_READ etc.) are concerned, except of course that
an absent HPTE doesn't need to be invalidated with tlbie since it
was never valid as far as the hardware is concerned.
When the guest accesses a page for which there is an absent HPTE, it
will take a hypervisor data storage interrupt (HDSI) since we now set
the VPM1 bit in the LPCR. Our HDSI handler for HPTE-not-present faults
looks up the hash table and if it finds an absent HPTE mapping the
requested virtual address, will switch to kernel mode and handle the
fault in kvmppc_book3s_hv_page_fault(), which at present just calls
kvmppc_hv_emulate_mmio() to set up the MMIO emulation.
This is based on an earlier patch by Benjamin Herrenschmidt, but since
heavily reworked.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:33:07 +0000 (12:33 +0000)]
KVM: PPC: Maintain a doubly-linked list of guest HPTEs for each gfn
This expands the reverse mapping array to contain two links for each
HPTE which are used to link together HPTEs that correspond to the
same guest logical page. Each circular list of HPTEs is pointed to
by the rmap array entry for the guest logical page, pointed to by
the relevant memslot. Links are 32-bit HPT entry indexes rather than
full 64-bit pointers, to save space. We use 3 of the remaining 32
bits in the rmap array entries as a lock bit, a referenced bit and
a present bit (the present bit is needed since HPTE index 0 is valid).
The bit lock for the rmap chain nests inside the HPTE lock bit.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:32:27 +0000 (12:32 +0000)]
KVM: PPC: Allow I/O mappings in memory slots
This provides for the case where userspace maps an I/O device into the
address range of a memory slot using a VM_PFNMAP mapping. In that
case, we work out the pfn from vma->vm_pgoff, and record the cache
enable bits from vma->vm_page_prot in two low-order bits in the
slot_phys array entries. Then, in kvmppc_h_enter() we check that the
cache bits in the HPTE that the guest wants to insert match the cache
bits in the slot_phys array entry. However, we do allow the guest to
create what it thinks is a non-cacheable or write-through mapping to
memory that is actually cacheable, so that we can use normal system
memory as part of an emulated device later on. In that case the actual
HPTE we insert is a cacheable HPTE.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:31:41 +0000 (12:31 +0000)]
KVM: PPC: Allow use of small pages to back Book3S HV guests
This relaxes the requirement that the guest memory be provided as
16MB huge pages, allowing it to be provided as normal memory, i.e.
in pages of PAGE_SIZE bytes (4k or 64k). To allow this, we index
the kvm->arch.slot_phys[] arrays with a small page index, even if
huge pages are being used, and use the low-order 5 bits of each
entry to store the order of the enclosing page with respect to
normal pages, i.e. log_2(enclosing_page_size / PAGE_SIZE).
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:31:00 +0000 (12:31 +0000)]
KVM: PPC: Only get pages when actually needed, not in prepare_memory_region()
This removes the code from kvmppc_core_prepare_memory_region() that
looked up the VMA for the region being added and called hva_to_page
to get the pfns for the memory. We have no guarantee that there will
be anything mapped there at the time of the KVM_SET_USER_MEMORY_REGION
ioctl call; userspace can do that ioctl and then map memory into the
region later.
Instead we defer looking up the pfn for each memory page until it is
needed, which generally means when the guest does an H_ENTER hcall on
the page. Since we can't call get_user_pages in real mode, if we don't
already have the pfn for the page, kvmppc_h_enter() will return
H_TOO_HARD and we then call kvmppc_virtmode_h_enter() once we get back
to kernel context. That calls kvmppc_get_guest_page() to get the pfn
for the page, and then calls back to kvmppc_h_enter() to redo the HPTE
insertion.
When the first vcpu starts executing, we need to have the RMO or VRMA
region mapped so that the guest's real mode accesses will work. Thus
we now have a check in kvmppc_vcpu_run() to see if the RMO/VRMA is set
up and if not, call kvmppc_hv_setup_rma(). It checks if the memslot
starting at guest physical 0 now has RMO memory mapped there; if so it
sets it up for the guest, otherwise on POWER7 it sets up the VRMA.
The function that does that, kvmppc_map_vrma, is now a bit simpler,
as it calls kvmppc_virtmode_h_enter instead of creating the HPTE itself.
Since we are now potentially updating entries in the slot_phys[]
arrays from multiple vcpu threads, we now have a spinlock protecting
those updates to ensure that we don't lose track of any references
to pages.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:30:16 +0000 (12:30 +0000)]
KVM: PPC: Make the H_ENTER hcall more reliable
At present, our implementation of H_ENTER only makes one try at locking
each slot that it looks at, and doesn't even retry the ldarx/stdcx.
atomic update sequence that it uses to attempt to lock the slot. Thus
it can return the H_PTEG_FULL error unnecessarily, particularly when
the H_EXACT flag is set, meaning that the caller wants a specific PTEG
slot.
This improves the situation by making a second pass when no free HPTE
slot is found, where we spin until we succeed in locking each slot in
turn and then check whether it is full while we hold the lock. If the
second pass fails, then we return H_PTEG_FULL.
This also moves lock_hpte to a header file (since later commits in this
series will need to use it from other source files) and renames it to
try_lock_hpte, which is a somewhat less misleading name.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:28:55 +0000 (12:28 +0000)]
KVM: PPC: Add an interface for pinning guest pages in Book3s HV guests
This adds two new functions, kvmppc_pin_guest_page() and
kvmppc_unpin_guest_page(), and uses them to pin the guest pages where
the guest has registered areas of memory for the hypervisor to update,
(i.e. the per-cpu virtual processor areas, SLB shadow buffers and
dispatch trace logs) and then unpin them when they are no longer
required.
Although it is not strictly necessary to pin the pages at this point,
since all guest pages are already pinned, later commits in this series
will mean that guest pages aren't all pinned.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:28:21 +0000 (12:28 +0000)]
KVM: PPC: Keep page physical addresses in per-slot arrays
This allocates an array for each memory slot that is added to store
the physical addresses of the pages in the slot. This array is
vmalloc'd and accessed in kvmppc_h_enter using real_vmalloc_addr().
This allows us to remove the ram_pginfo field from the kvm_arch
struct, and removes the 64GB guest RAM limit that we had.
We use the low-order bits of the array entries to store a flag
indicating that we have done get_page on the corresponding page,
and therefore need to call put_page when we are finished with the
page. Currently this is set for all pages except those in our
special RMO regions.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:27:39 +0000 (12:27 +0000)]
KVM: PPC: Keep a record of HV guest view of hashed page table entries
This adds an array that parallels the guest hashed page table (HPT),
that is, it has one entry per HPTE, used to store the guest's view
of the second doubleword of the corresponding HPTE. The first
doubleword in the HPTE is the same as the guest's idea of it, so we
don't need to store a copy, but the second doubleword in the HPTE has
the real page number rather than the guest's logical page number.
This allows us to remove the back_translate() and reverse_xlate()
functions.
This "reverse mapping" array is vmalloc'd, meaning that to access it
in real mode we have to walk the kernel's page tables explicitly.
That is done by the new real_vmalloc_addr() function. (In fact this
returns an address in the linear mapping, so the result is usable
both in real mode and in virtual mode.)
There are also some minor cleanups here: moving the definitions of
HPT_ORDER etc. to a header file and defining HPT_NPTE for HPT_NPTEG << 3.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Paul Mackerras [Mon, 12 Dec 2011 12:24:48 +0000 (12:24 +0000)]
KVM: PPC: Make wakeups work again for Book3S HV guests
When commit f43fdc15fa ("KVM: PPC: booke: Improve timer register
emulation") factored out some code in arch/powerpc/kvm/powerpc.c
into a new helper function, kvm_vcpu_kick(), an error crept in
which causes Book3s HV guest vcpus to stall. This fixes it.
On POWER7 machines, guest vcpus are grouped together into virtual
CPU cores that share a single waitqueue, so it's important to use
vcpu->arch.wqp rather than &vcpu->wq.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Tue, 29 Nov 2011 10:40:23 +0000 (10:40 +0000)]
KVM: PPC: e500: use hardware hint when loading TLB0 entries
The hardware maintains a per-set next victim hint. Using this
reduces conflicts, especially on e500v2 where a single guest
TLB entry is mapped to two shadow TLB entries (user and kernel).
We want those two entries to go to different TLB ways.
sesel is now only used for TLB1.
Reported-by: Liu Yu <yu.liu@freescale.com> Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Mon, 19 Dec 2011 12:36:55 +0000 (13:36 +0100)]
KVM: PPC: Book3S: PR: Fix signal check race
As Scott put it:
> If we get a signal after the check, we want to be sure that we don't
> receive the reschedule IPI until after we're in the guest, so that it
> will cause another signal check.
we need to have interrupts disabled from the point we do signal_check()
all the way until we actually enter the guest.
This patch fixes potential signal loss races.
Reported-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Fri, 9 Dec 2011 14:20:46 +0000 (15:20 +0100)]
KVM: PPC: align vcpu_kick with x86
Our vcpu kick implementation differs a bit from x86 which resulted in us not
disabling preemption during the kick. Get it a bit closer to what x86 does.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Fri, 9 Dec 2011 13:44:13 +0000 (14:44 +0100)]
KVM: PPC: Use get/set for to_svcpu to help preemption
When running the 64-bit Book3s PR code without CONFIG_PREEMPT_NONE, we were
doing a few things wrong, most notably access to PACA fields without making
sure that the pointers stay stable accross the access (preempt_disable()).
This patch moves to_svcpu towards a get/put model which allows us to disable
preemption while accessing the shadow vcpu fields in the PACA. That way we
can run preemptible and everyone's happy!
Reported-by: Jörg Sommer <joerg@alea.gnuu.de> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Alexander Graf [Fri, 9 Dec 2011 14:46:21 +0000 (15:46 +0100)]
KVM: PPC: Book3s: PR: Disable preemption in vcpu_run
When entering the guest, we want to make sure we're not getting preempted
away, so let's disable preemption on entry, but enable it again while handling
guest exits.
Reported-by: Jörg Sommer <joerg@alea.gnuu.de> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Thu, 17 Nov 2011 12:39:59 +0000 (12:39 +0000)]
KVM: PPC: booke: Improve timer register emulation
Decrementers are now properly driven by TCR/TSR, and the guest
has full read/write access to these registers.
The decrementer keeps ticking (and setting the TSR bit) regardless of
whether the interrupts are enabled with TCR.
The decrementer stops at zero, rather than going negative.
Decrementers (and FITs, once implemented) are delivered as
level-triggered interrupts -- dequeued when the TSR bit is cleared, not
on delivery.
Signed-off-by: Liu Yu <yu.liu@freescale.com>
[scottwood@freescale.com: significant changes] Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Wed, 9 Nov 2011 00:23:30 +0000 (18:23 -0600)]
KVM: PPC: Paravirtualize SPRG4-7, ESR, PIR, MASn
This allows additional registers to be accessed by the guest
in PR-mode KVM without trapping.
SPRG4-7 are readable from userspace. On booke, KVM will sync
these registers when it enters the guest, so that accesses from
guest userspace will work. The guest kernel, OTOH, must consistently
use either the real registers or the shared area between exits. This
also applies to the already-paravirted SPRG3.
On non-booke, it's not clear to what extent SPRG4-7 are supported
(they're not architected for book3s, but exist on at least some classic
chips). They are copied in the get/set regs ioctls, but I do not see any
non-booke emulation. I also do not see any syncing with real registers
(in PR-mode) including the user-readable SPRG3. This patch should not
make that situation any worse.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Wed, 9 Nov 2011 00:23:27 +0000 (18:23 -0600)]
KVM: PPC: booke: Fix int_pending calculation for MSR[EE] paravirt
int_pending was only being lowered if a bit in pending_exceptions
was cleared during exception delivery -- but for interrupts, we clear
it during IACK/TSR emulation. This caused paravirt for enabling
MSR[EE] to be ineffective.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Wed, 9 Nov 2011 00:23:23 +0000 (18:23 -0600)]
KVM: PPC: Move prepare_to_enter call site into subarch code
This function should be called with interrupts disabled, to avoid
a race where an exception is delivered after we check, but the
resched kick is received before we disable interrupts (and thus doesn't
actually trigger the exit code that would recheck exceptions).
booke already does this properly in the lightweight exit case, but
not on initial entry.
For now, move the call of prepare_to_enter into subarch-specific code so
that booke can do the right thing here. Ideally book3s would do the same
thing, but I'm having a hard time seeing where it does any interrupt
disabling of this sort (plus it has several additional call sites), so
I'm deferring the book3s fix to someone more familiar with that code.
book3s behavior should be unchanged by this patch.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Wed, 9 Nov 2011 00:23:20 +0000 (18:23 -0600)]
KVM: PPC: Rename deliver_interrupts to prepare_to_enter
This function also updates paravirt int_pending, so rename it
to be more obvious that this is a collection of checks run prior
to (re)entering a guest.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Bharat Bhushan [Mon, 31 Oct 2011 08:52:08 +0000 (14:22 +0530)]
KVM: PPC: booke: Do Not start decrementer when SPRN_DEC set 0
As per specification the decrementer interrupt not happen when DEC is written
with 0. Also when DEC is zero, no decrementer running. So we should not start
hrtimer for decrementer when DEC = 0.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Bharat Bhushan [Wed, 19 Oct 2011 04:16:06 +0000 (09:46 +0530)]
KVM: PPC: Fix DEC truncation for greater than 0xffff_ffff/1000
kvmppc_emulate_dec() uses dec_nsec of type unsigned long and does below calculation:
dec_nsec = vcpu->arch.dec;
dec_nsec *= 1000;
This will truncate if DEC value "vcpu->arch.dec" is greater than 0xffff_ffff/1000.
For example : For tb_ticks_per_usec = 4a, we can not set decrementer more than ~58ms.
Signed-off-by: Bharat Bhushan <bharat.bhushan@freescale.com> Acked-by: Liu Yu <yu.liu@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
Scott Wood [Thu, 18 Aug 2011 20:25:18 +0000 (15:25 -0500)]
KVM: PPC: e500: clear up confusion between host and guest entries
Split out the portions of tlbe_priv that should be associated with host
entries into tlbe_ref. Base victim selection on the number of hardware
entries, not guest entries.
For TLB1, where one guest entry can be mapped by multiple host entries,
we use the host tlbe_ref for tracking page references. For the guest
TLB0 entries, we still track it with gtlb_priv, to avoid having to
retranslate if the entry is evicted from the host TLB but not the
guest TLB.
Signed-off-by: Scott Wood <scottwood@freescale.com> Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: s390: provide the prefix register via kvm_run
Add the prefix register to the synced register field in kvm_run.
While we need the prefix register most of the time read-only, this
patch also adds handling for guest dirtying of the prefix register.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
On some cpus the overhead for virtualization instructions is in the same
range as a system call. Having to call multiple ioctls to get set registers
will make certain userspace handled exits more expensive than necessary.
Lets provide a section in kvm_run that works as a shared save area
for guest registers.
We also provide two 64bit flags fields (architecture specific), that will
specify
1. which parts of these fields are valid.
2. which registers were modified by userspace
Each bit for these flag fields will define a group of registers (like
general purpose) or a single register.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
There are several places in the kvm module, which set the prefix register.
Since we need to flush the cpu, lets combine this operation into a helper
function. This helper will also explicitely mask out the unused bits.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Boris Ostrovsky [Mon, 9 Jan 2012 19:00:35 +0000 (14:00 -0500)]
KVM: SVM: Add support for AMD's OSVW feature in guests
In some cases guests should not provide workarounds for errata even when the
physical processor is affected. For example, because of erratum 400 on family
10h processors a Linux guest will read an MSR (resulting in VMEXIT) before
going to idle in order to avoid getting stuck in a non-C0 state. This is not
necessary: HLT and IO instructions are intercepted and therefore there is no
reason for erratum 400 workaround in the guest.
This patch allows us to present a guest with certain errata as fixed,
regardless of the state of actual hardware.
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@amd.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Carsten Otte [Wed, 4 Jan 2012 09:25:28 +0000 (10:25 +0100)]
KVM: s390: fix assumption for KVM_MAX_VCPUS
This patch fixes definition of the idle_mask and the local_int array
in kvm_s390_float_interrupt. Previous definition had 64 cpus max
hardcoded instead of using KVM_MAX_VCPUS.
Carsten Otte [Wed, 4 Jan 2012 09:25:27 +0000 (10:25 +0100)]
KVM: s390: ucontrol: disable sca
This patch makes sure user controlled virtual machines do not use a
system control area (sca). This is needed in order to create
virtual machines with more cpus than the size of the sca [64].
Carsten Otte [Wed, 4 Jan 2012 09:25:26 +0000 (10:25 +0100)]
KVM: s390: ucontrol: interface to inject faults on a vcpu page table
This patch allows the user to fault in pages on a virtual cpus
address space for user controlled virtual machines. Typically this
is superfluous because userspace can just create a mapping and
let the kernel's page fault logic take are of it. There is one
exception: SIE won't start if the lowcore is not present. Normally
the kernel takes care of this [handle_validity() in
arch/s390/kvm/intercept.c] but since the kernel does not handle
intercepts for user controlled virtual machines, userspace needs to
be able to handle this condition.
Carsten Otte [Wed, 4 Jan 2012 09:25:25 +0000 (10:25 +0100)]
KVM: s390: ucontrol: disable in-kernel irq stack
This patch disables the in-kernel interrupt stack for KVM virtual
machines that are controlled by user. Userspace has to take care
of handling interrupts on its own.
Carsten Otte [Wed, 4 Jan 2012 09:25:24 +0000 (10:25 +0100)]
KVM: s390: ucontrol: disable in-kernel handling of SIE intercepts
This patch disables in-kernel handling of SIE intercepts for user
controlled virtual machines. All intercepts are passed to userspace
via KVM_EXIT_SIE exit reason just like SIE intercepts that cannot be
handled in-kernel for regular KVM guests.
Carsten Otte [Wed, 4 Jan 2012 09:25:23 +0000 (10:25 +0100)]
KVM: s390: ucontrol: export SIE control block to user
This patch exports the s390 SIE hardware control block to userspace
via the mapping of the vcpu file descriptor. In order to do so,
a new arch callback named kvm_arch_vcpu_fault is introduced for all
architectures. It allows to map architecture specific pages.
Carsten Otte [Wed, 4 Jan 2012 09:25:22 +0000 (10:25 +0100)]
KVM: s390: ucontrol: export page faults to user
This patch introduces a new exit reason in the kvm_run structure
named KVM_EXIT_S390_UCONTROL. This exit indicates, that a virtual cpu
has regognized a fault on the host page table. The idea is that
userspace can handle this fault by mapping memory at the fault
location into the cpu's address space and then continue to run the
virtual cpu.
Carsten Otte [Wed, 4 Jan 2012 09:25:21 +0000 (10:25 +0100)]
KVM: s390: ucontrol: per vcpu address spaces
This patch introduces two ioctls for virtual cpus, that are only
valid for kernel virtual machines that are controlled by userspace.
Each virtual cpu has its individual address space in this mode of
operation, and each address space is backed by the gmap
implementation just like the address space for regular KVM guests.
KVM_S390_UCAS_MAP allows to map a part of the user's virtual address
space to the vcpu. Starting offset and length in both the user and
the vcpu address space need to be aligned to 1M.
KVM_S390_UCAS_UNMAP can be used to unmap a range of memory from a
virtual cpu in a similar way.
Carsten Otte [Wed, 4 Jan 2012 09:25:20 +0000 (10:25 +0100)]
KVM: s390: add parameter for KVM_CREATE_VM
This patch introduces a new config option for user controlled kernel
virtual machines. It introduces a parameter to KVM_CREATE_VM that
allows to set bits that alter the capabilities of the newly created
virtual machine.
The parameter is passed to kvm_arch_init_vm for all architectures.
The only valid modifier bit for now is KVM_VM_S390_UCONTROL.
This requires CAP_SYS_ADMIN privileges and creates a user controlled
virtual machine on s390 architectures.
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-fixes:
GFS2: Read resource groups on mount
GFS2: Ensure rindex is uptodate for fallocate
GFS2: Read in rindex if necessary during unlink
GFS2: Fix race between lru_list and glock ref count
Linus Torvalds [Tue, 28 Feb 2012 17:15:31 +0000 (09:15 -0800)]
Merge tag 'iommu-fixes-v3.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
IOMMU fixes for Linux 3.3-rc5
All the fixes are for the OMAP IOMMU driver. The first patch is the
biggest one. It fixes the calls of the function omap_find_iovm_area() in
the omap-iommu-debug module which expects a 'struct device' parameter
since commit fabdbca instead of an omap_iommu handle. The
omap-iommu-debug code still passed the handle to the function which
caused a crash.
The second patch fixes a NULL pointer dereference in the OMAP code and
the third patch makes sure that the omap-iommu is initialized before the
omap-isp driver, which relies on the iommu. The last patch is only a
workaround until defered probing is implemented.
* tag 'iommu-fixes-v3.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
ARM: OMAP: make iommu subsys_initcall to fix builtin omap3isp
iommu/omap: fix NULL pointer dereference
iommu/omap: fix erroneous omap-iommu-debug API calls
This makes mount take slightly longer, but at the same time, the first
write to the filesystem will be faster too. It also means that if there
is a problem in the resource index, then we can refuse to mount rather
than having to try and report that when the first write occurs.
In addition, to avoid recursive locking, we hvae to take account of
instances when the rindex glock may already be held when we are
trying to update the rbtree of resource groups.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Bob Peterson [Fri, 17 Feb 2012 14:15:52 +0000 (09:15 -0500)]
GFS2: Ensure rindex is uptodate for fallocate
This patch fixes a problem whereby gfs2_grow was failing and causing GFS2
to assert. The problem was that when GFS2's fallocate operation tried to
acquire an "allocation" it made sure the rindex was up to date, and if not,
it called gfs2_rindex_update. However, if the file being fallocated was
the rindex itself, it was already locked at that point. By calling
gfs2_rindex_update at an earlier point in time, we bring rindex up to date
and thereby avoid trying to lock it when the "allocation" is acquired.
Signed-off-by: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Bob Peterson [Thu, 16 Feb 2012 16:31:04 +0000 (11:31 -0500)]
GFS2: Read in rindex if necessary during unlink
This patch fixes a problem whereby you were unable to delete
files until other file system operations were done (such as
statfs, touch, writes, etc.) that caused the rindex to be
read in.
Signed-off-by: Bob Peterson <rpeterso@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Linus Torvalds [Mon, 27 Feb 2012 23:43:05 +0000 (15:43 -0800)]
Merge tag 'ktest-fix-make-min-failed-build-for-real' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest
While demoing ktest at ELC in 2012, it was embarrassing that the
make_min_config test failed to work because the snowball board I was
testing it against had a config that would not build. But the
make_min_config only tested the testing part and ignored build failures.
The end result was a config file that would not boot.
This time, for real.
* tag 'ktest-fix-make-min-failed-build-for-real' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest:
ktest: Fix make_min_config test when build fails
Steven Rostedt [Mon, 27 Feb 2012 18:58:49 +0000 (13:58 -0500)]
ktest: Fix make_min_config test when build fails
The make_min_config does not take into account when the build fails,
resulting in a invalid MIN_CONFIG .config file. When the build fails,
it is ignored and the boot test is executed, using the previous built
kernel. The configs that should be tested are not tested and they may
be added or removed depending on the result of the last kernel that
succeeded to be built.
If the build fails, mark the current config as a failure and the
configs that were disabled may still be needed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Here are some trivial NTFS changes (a spelling fix and two use before
NULL check cases found by Coverity as well as an update in MAINTAINERS
for the path to the ntfs git repo) together with a simple LDM fix for
parsing fragmented VBLKs.
* git://git.kernel.org/pub/scm/linux/kernel/git/aia21/ntfs:
NTFS: Update git repo path in MAINTAINERS file.
LDM: Fix reassembly of extended VBLKs.
NTFS: Correct two spelling errors "dealocate" to "deallocate" in mft.c.
NTFS: Do not dereference pointer before checking for NULL.
NTFS: Remove unused variable.
Linus Torvalds [Mon, 27 Feb 2012 15:55:51 +0000 (07:55 -0800)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mce/AMD: Fix UP build error
x86: Specify a size for the cmp in the NMI handler
x86/nmi: Test saved %cs in NMI to determine nested NMI case
x86/amd: Fix L1i and L2 cache sharing information for AMD family 15h processors
x86/microcode: Remove noisy AMD microcode warning
projects / linux-beck.git / log