If the non-RAM regions in the e820 memory map are larger than the size
of the initial balloon, a BUG was triggered as the frames are remaped
beyond the limit of the linear p2m.  The frames are remapped into the
initial balloon area (xen_extra_mem) but not enough of this is
available.

Ensure enough extra memory regions are added for these remapped
frames.

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Juergen Gross <jgross@suse.com>

This accounting is just used to print a diagnostic message that isn't
very useful.

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Juergen Gross <jgross@suse.com>

With recent changes in p2m we now have legitimate cases when
p2m memory needs to be freed during early boot (i.e. before
slab is initialized).

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

There is no reason for having it and, with commit 250a1ac6

 ("x86,
smpboot: Remove pointless preempt_disable() in
native_smp_prepare_cpus()"), it prevents HVM guests from booting.

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

With the virtual mapped linear p2m list the post-init mmu operations
must be used for setting up the p2m mappings, as in case of
CONFIG_FLATMEM the init routines may trigger BUGs.

paging_init() sets up all infrastructure needed to switch to the
post-init mmu ops done by xen_post_allocator_init(). With the virtual
mapped linear p2m list we need some mmu ops during setup of this list,
so we have to switch to the correct mmu ops as soon as possible.

The p2m list is usable from the beginning, just expansion requires to
have established the new linear mapping. So the call of
xen_remap_memory() had to be introduced, but this is not due to the
mmu ops requiring this.

Summing it up: calling xen_post_allocator_init() not directly after
paging_init() was conceptually wrong in the beginning, it just didn't
matter up to now as no functions used between the two calls needed
some critical mmu ops (e.g. alloc_pte). This has changed now, so I
corrected it.

Reported-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

This reverts commit 2c3fc8d2

.

This commit broke on x86 PV because entries in the generic SWIOTLB are
indexed using (pseudo-)physical address not DMA address and these are
not the same in a x86 PV guest.

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>

This reverts commit 2c3fc8d2

.

This commit broke on x86 PV because entries in the generic SWIOTLB are
indexed using (pseudo-)physical address not DMA address and these are
not the same in a x86 PV guest.

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>

Commit 5b8e7d80

 removed the __init
annotation from xen_set_identity_and_remap_chunk(). Add it again.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Introduce two helper functions to safely read and write unsigned long
values from or to memory when the access may fault because the mapping
is non-present or read-only.

These helpers can be used instead of open coded uses of __get_user()
and __put_user() avoiding the need to do casts to fix sparse warnings.

Use the helpers in page.h and p2m.c. This will fix the sparse
warnings when doing "make C=1".

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Instead of checking at each call of set_phys_to_machine() whether a
new p2m page has to be allocated due to writing an entry in a large
invalid or identity area, just map those areas read only and react
to a page fault on write by allocating the new page.

This change will make the common path with no allocation much
faster as it only requires a single write of the new mfn instead
of walking the address translation tables and checking for the
special cases.

Suggested-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

At start of the day the Xen hypervisor presents a contiguous mfn list
to a pv-domain. In order to support sparse memory this mfn list is
accessed via a three level p2m tree built early in the boot process.
Whenever the system needs the mfn associated with a pfn this tree is
used to find the mfn.

Instead of using a software walked tree for accessing a specific mfn
list entry this patch is creating a virtual address area for the
entire possible mfn list including memory holes. The holes are
covered by mapping a pre-defined  page consisting only of "invalid
mfn" entries. Access to a mfn entry is possible by just using the
virtual base address of the mfn list and the pfn as index into that
list. This speeds up the (hot) path of determining the mfn of a
pfn.

Kernel build on a Dell Latitude E6440 (2 cores, HT) in 64 bit Dom0
showed following improvements:

Elapsed time: 32:50 ->  32:35
System:       18:07 ->  17:47
User:        104:00 -> 103:30

Tested with following configurations:
- 64 bit dom0, 8GB RAM
- 64 bit dom0, 128 GB RAM, PCI-area above 4 GB
- 32 bit domU, 512 MB, 8 GB, 43 GB (more wouldn't work even without
                                    the patch)
- 32 bit domU, ballooning up and down
- 32 bit domU, save and restore
- 32 bit domU with PCI passthrough
- 64 bit domU, 8 GB, 2049 MB, 5000 MB
- 64 bit domU, ballooning up and down
- 64 bit domU, save and restore
- 64 bit domU with PCI passthrough

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Today get_phys_to_machine() is always called when the mfn for a pfn
is to be obtained. Add a wrapper __pfn_to_mfn() as inline function
to be able to avoid calling get_phys_to_machine() when possible as
soon as the switch to a linear mapped p2m list has been done.

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Introduces lookup_pmd_address() to get the address of the pmd entry
related to a virtual address in the current address space. This
function is needed for support of a virtual mapped sparse p2m list
in xen pv domains, as we need the address of the pmd entry, not the
one of the pte in that case.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

When the physical memory configuration is initialized the p2m entries
for not pouplated memory pages are set to "invalid". As those pages
are beyond the hypervisor built p2m list the p2m tree has to be
extended.

This patch delays processing the extra memory related p2m entries
during the boot process until some more basic memory management
functions are callable. This removes the need to create new p2m
entries until virtual memory management is available.

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

The m2p overrides are used to be able to find the local pfn for a
foreign mfn mapped into the domain. They are used by driver backends
having to access frontend data.

As this functionality isn't used in early boot it makes no sense to
initialize the m2p override functions very early. It can be done
later without doing any harm, removing the need for allocating memory
via extend_brk().

While at it make some m2p override functions static as they are only
used internally.

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Early in the boot process the memory layout of a pv-domain is changed
to match the E820 map (either the host one for Dom0 or the Xen one)
regarding placement of RAM and PCI holes. This requires removing memory
pages initially located at positions not suitable for RAM and adding
them later at higher addresses where no restrictions apply.

To be able to operate on the hypervisor supported p2m list until a
virtual mapped linear p2m list can be constructed, remapping must
be delayed until virtual memory management is initialized, as the
initial p2m list can't be extended unlimited at physical memory
initialization time due to it's fixed structure.

A further advantage is the reduction in complexity and code volume as
we don't have to be careful regarding memory restrictions during p2m
updates.

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

In arch/x86/xen/p2m.c three different allocation functions for
obtaining a memory page are used: extend_brk(), alloc_bootmem_align()
or __get_free_page().  Which of those functions is used depends on the
progress of the boot process of the system.

Introduce a common allocation routine selecting the to be called
allocation routine dynamically based on the boot progress. This allows
moving initialization steps without having to care about changing
allocation calls.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Some functions in arch/x86/xen/p2m.c are used locally only. Make them
static. Rearrange the functions in p2m.c to avoid forward declarations.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

The source arch/x86/xen/p2m.c has some coding style issues. Fix them.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

When hardware supports APIC/x2APIC virtualization we don't need to use
pirqs for MSI handling and instead use APIC since most APIC accesses
(MMIO or MSR) will now be processed without VMEXITs.

As an example, netperf on the original code produces this profile
(collected wih 'xentrace -e 0x0008ffff -T 5'):

    342 cpu_change
    260 CPUID
  34638 HLT
  64067 INJ_VIRQ
  28374 INTR
  82733 INTR_WINDOW
     10 NPF
  24337 TRAP
 370610 vlapic_accept_pic_intr
 307528 VMENTRY
 307527 VMEXIT
 140998 VMMCALL
    127 wrap_buffer

After applying this patch the same test shows

    230 cpu_change
    260 CPUID
  36542 HLT
    174 INJ_VIRQ
  27250 INTR
    222 INTR_WINDOW
     20 NPF
  24999 TRAP
 381812 vlapic_accept_pic_intr
 166480 VMENTRY
 166479 VMEXIT
  77208 VMMCALL
     81 wrap_buffer

ApacheBench results (ab -n 10000 -c 200) improve by about 10%

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

If the hardware supports APIC virtualization we may decide not to use
pirqs and instead use APIC/x2APIC directly, meaning that we don't want
to set x86_msi.setup_msi_irqs and x86_msi.teardown_msi_irq to
Xen-specific routines.  However, x2APIC is not set up by the time
pci_xen_hvm_init() is called so we need to postpone setting these ops
until later, when we know which APIC mode is used.

(Note that currently x2APIC is never initialized on HVM guests. This
may change in the future)

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

When a PF driver unloads, it may find it necessary to leave the VFs
around simply because of pciback having marked them as assigned to a
guest. Utilize a suitable notification to let go of the VFs, thus
allowing the PF to go back into the state it was before its driver
loaded (which in particular allows the driver to be loaded again with
it being able to create the VFs anew, but which also allows to then
pass through the PF instead of the VFs).

Don't do this however for any VFs currently in active use by a guest.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
[v2: Removed the switch statement, moved it about]
[v3: Redid it a bit differently]
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

The commit "xen/pciback: Don't deadlock when unbinding." was using
the version of pci_reset_function which would lock the device lock.
That is no good as we can dead-lock. As such we swapped to using
the lock-less version and requiring that the callers
of 'pcistub_put_pci_dev' take the device lock. And as such
this bug got exposed.

Using the lock-less version is  OK, except that we tried to
use 'pci_restore_state' after the lock-less version of
__pci_reset_function_locked - which won't work as 'state_saved'
is set to false. Said 'state_saved' is a toggle boolean that
is to be used by the sequence of a) pci_save_state/pci_restore_state
or b) pci_load_and_free_saved_state/pci_restore_state. We don't
want to use a) as the guest might have messed up the PCI
configuration space and we want it to revert to the state
when the PCI device was binded to us. Therefore we pick
b) to restore the configuration space.

We restore from our 'golden' version of PCI configuration space, when an:
 - Device is unbinded from pciback
 - Device is detached from a guest.

Reported-by: Sander Eikelenboom <linux@eikelenboom.it>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

We have the pci_load_and_free_saved_state, and pci_store_saved_state
but are missing the functionality to just load the state
multiple times in the PCI device without having to free/save
the state.

This patch makes it possible to use this function.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

A little cleanup. No functional difference.

Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

We had been printing it only if the device was built with
debug enabled. But this information is useful in the field
to troubleshoot.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Cleanup the function a bit - also include the id of the
domain that is using the device.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Instead of open-coding it in drivers that want to double check
that their functions are indeed holding the device lock.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Suggested-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

As commit 0a9fd015


'xen/pciback: Document the entry points for 'pcistub_put_pci_dev''
explained there are four entry points in this function.
Two of them are when the user fiddles in the SysFS to
unbind a device which might be in use by a guest or not.

Both 'unbind' states will cause a deadlock as the the PCI lock has
already been taken, which then pci_device_reset tries to take.

We can simplify this by requiring that all callers of
pcistub_put_pci_dev MUST hold the device lock. And then
we can just call the lockless version of pci_device_reset.

To make it even simpler we will modify xen_pcibk_release_pci_dev
to quality whether it should take a lock or not - as it ends
up calling xen_pcibk_release_pci_dev and needs to hold the lock.

Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: David Vrabel <david.vrabel@citrix.com>

Need to pass the pointer within the swiotlb internal buffer to the
swiotlb library, that in the case of xen_unmap_single is dev_addr, not
paddr.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
CC: stable@vger.kernel.org

In xen_swiotlb_sync_single we always call xen_dma_sync_single_for_cpu,
even when we should call xen_dma_sync_single_for_device. Fix that.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
CC: stable@vger.kernel.org

On x86 truncation cannot occur because config XEN depends on X86_64 ||
(X86_32 && X86_PAE).

On ARM truncation can occur without CONFIG_ARM_LPAE, when the dma
operation involves foreign grants. However in that case the physical
address returned by xen_bus_to_phys is actually invalid (there is no mfn
to pfn tracking for foreign grants on ARM) and it is not used.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
CC: stable@vger.kernel.org

xen_dma_unmap_page and xen_dma_sync_single_for_cpu take a dma_addr_t
handle as argument, not a physical address.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
CC: stable@vger.kernel.org

Introduce support for new hypercall GNTTABOP_cache_flush.
Use it to perform cache flashing on pages used for dma when necessary.

If GNTTABOP_cache_flush is supported by the hypervisor, we don't need to
bounce dma map operations that involve foreign grants and non-coherent
devices.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>

Introduce an arch specific function to find out whether a particular dma
mapping operation needs to bounce on the swiotlb buffer.

On ARM and ARM64, if the page involved is a foreign page and the device
is not coherent, we need to bounce because at unmap time we cannot
execute any required cache maintenance operations (we don't know how to
find the pfn from the mfn).

No change of behaviour for x86.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: David Vrabel <david.vrabel@citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Merge xen/mm32.c into xen/mm.c.
As a consequence the code gets compiled on arm64 too.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>

In xen_dma_map_page, if the page is a local page, call the native
map_page dma_ops. If the page is foreign, call __xen_dma_map_page that
issues any required cache maintenane operations via hypercall.

The reason for doing this is that the native dma_ops map_page could
allocate buffers than need to be freed. If the page is foreign we don't
call the native unmap_page dma_ops function, resulting in a memory leak.

Suggested-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>

dev_addr is the machine address of the page.

The new parameter can be used by the ARM and ARM64 implementations of
xen_dma_map_page to find out if the page is a local page (pfn == mfn) or
a foreign page (pfn != mfn).

dev_addr could be retrieved again from the physical address, using
pfn_to_mfn, but it requires accessing an rbtree. Since we already have
the dev_addr in our hands at the call site there is no need to get the
mfn twice.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Use is_device_dma_coherent to check whether we need to issue cache
maintenance operations rather than checking on the existence of a
particular dma_ops function for the device.

This is correct because coherent devices don't need cache maintenance
operations - arm_coherent_dma_ops does not set the hooks that we
were previously checking for existance.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>