linux-toradex.git/arch/arm/xen, branch v6.7

arm/xen: fix xen_vcpu_info allocation alignment

2023-11-23T08:32:41+00:00

xen_vcpu_info is a percpu area than needs to be mapped by Xen.
Currently, it could cross a page boundary resulting in Xen being unable
to map it:

[    0.567318] kernel BUG at arch/arm64/xen/../../arm/xen/enlighten.c:164!
[    0.574002] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP

Fix the issue by using __alloc_percpu and requesting alignment for the
memory allocation.

Signed-off-by: Stefano Stabellini 

Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2311221501340.2053963@ubuntu-linux-20-04-desktop
Fixes: 24d5373dda7c ("arm/xen: Use alloc_percpu rather than __alloc_percpu")
Reviewed-by: Juergen Gross 
Signed-off-by: Juergen Gross

arm64/arm: xen: enlighten: Fix KPTI checks

2023-10-16T11:57:43+00:00

When KPTI is in use, we cannot register a runstate region as XEN
requires that this is always a valid VA, which we cannot guarantee. Due
to this, xen_starting_cpu() must avoid registering each CPU's runstate
region, and xen_guest_init() must avoid setting up features that depend
upon it.

We tried to ensure that in commit:

  f88af7229f6f22ce (" xen/arm: do not setup the runstate info page if kpti is enabled")

... where we added checks for xen_kernel_unmapped_at_usr(), which wraps
arm64_kernel_unmapped_at_el0() on arm64 and is always false on 32-bit
arm.

Unfortunately, as xen_guest_init() is an early_initcall, this happens
before secondary CPUs are booted and arm64 has finalized the
ARM64_UNMAP_KERNEL_AT_EL0 cpucap which backs
arm64_kernel_unmapped_at_el0(), and so this can subsequently be set as
secondary CPUs are onlined. On a big.LITTLE system where the boot CPU
does not require KPTI but some secondary CPUs do, this will result in
xen_guest_init() intializing features that depend on the runstate
region, and xen_starting_cpu() registering the runstate region on some
CPUs before KPTI is subsequent enabled, resulting the the problems the
aforementioned commit tried to avoid.

Handle this more robsutly by deferring the initialization of the
runstate region until secondary CPUs have been initialized and the
ARM64_UNMAP_KERNEL_AT_EL0 cpucap has been finalized. The per-cpu work is
moved into a new hotplug starting function which is registered later
when we're certain that KPTI will not be used.

Fixes: f88af7229f6f ("xen/arm: do not setup the runstate info page if kpti is enabled")
Signed-off-by: Mark Rutland 
Cc: Bertrand Marquis 
Cc: Boris Ostrovsky 
Cc: Juergen Gross 
Cc: Stefano Stabellini 
Cc: Suzuki K Poulose 
Cc: Will Deacon 
Signed-off-by: Catalin Marinas

xen: simplify evtchn_do_upcall() call maze

2023-09-19T05:04:49+00:00

There are several functions involved for performing the functionality
of evtchn_do_upcall():

- __xen_evtchn_do_upcall() doing the real work
- xen_hvm_evtchn_do_upcall() just being a wrapper for
  __xen_evtchn_do_upcall(), exposed for external callers
- xen_evtchn_do_upcall() calling __xen_evtchn_do_upcall(), too, but
  without any user

Simplify this maze by:

- removing the unused xen_evtchn_do_upcall()
- removing xen_hvm_evtchn_do_upcall() as the only left caller of
  __xen_evtchn_do_upcall(), while renaming __xen_evtchn_do_upcall() to
  xen_evtchn_do_upcall()

Signed-off-by: Juergen Gross 
Reviewed-by: Boris Ostrovsky 
Reviewed-by: Thomas Gleixner 
Signed-off-by: Juergen Gross

swiotlb: bail out of swiotlb_init_late() if swiotlb is already allocated

2023-08-01T16:02:02+00:00

If swiotlb is allocated, immediately return 0, so callers do not have to
check io_tlb_default_mem.nslabs explicitly.

Signed-off-by: Petr Tesarik 
Signed-off-by: Christoph Hellwig

xen/virtio: Optimize the setup of "xen-grant-dma" devices

2022-12-05T11:59:49+00:00

This is needed to avoid having to parse the same device-tree
several times for a given device.

For this to work we need to install the xen_virtio_restricted_mem_acc
callback in Arm's xen_guest_init() which is same callback as x86's
PV and HVM modes already use and remove the manual assignment in
xen_setup_dma_ops(). Also we need to split the code to initialize
backend_domid into a separate function.

Prior to current patch we parsed the device-tree three times:
1. xen_setup_dma_ops()->...->xen_is_dt_grant_dma_device()
2. xen_setup_dma_ops()->...->xen_dt_grant_init_backend_domid()
3. xen_virtio_mem_acc()->...->xen_is_dt_grant_dma_device()

With current patch we parse the device-tree only once in
xen_virtio_restricted_mem_acc()->...->xen_dt_grant_init_backend_domid()

Other benefits are:
- Not diverge from x86 when setting up Xen grant DMA ops
- Drop several global functions

Signed-off-by: Oleksandr Tyshchenko 
Reviewed-by: Xenia Ragiadakou 
Reviewed-by: Stefano Stabellini 
Link: https://lore.kernel.org/r/20221025162004.8501-2-olekstysh@gmail.com
Signed-off-by: Juergen Gross

xen: don't require virtio with grants for non-PV guests

2022-08-01T05:43:03+00:00

Commit fa1f57421e0b ("xen/virtio: Enable restricted memory access using
Xen grant mappings") introduced a new requirement for using virtio
devices: the backend now needs to support the VIRTIO_F_ACCESS_PLATFORM
feature.

This is an undue requirement for non-PV guests, as those can be operated
with existing backends without any problem, as long as those backends
are running in dom0.

Per default allow virtio devices without grant support for non-PV
guests.

On Arm require VIRTIO_F_ACCESS_PLATFORM for devices having been listed
in the device tree to use grants.

Add a new config item to always force use of grants for virtio.

Fixes: fa1f57421e0b ("xen/virtio: Enable restricted memory access using Xen grant mappings")
Reported-by: Viresh Kumar 
Signed-off-by: Juergen Gross 
Reviewed-by: Oleksandr Tyshchenko 
Tested-by: Oleksandr Tyshchenko  # Arm64 guest using Xen
Reviewed-by: Stefano Stabellini 
Link: https://lore.kernel.org/r/20220622063838.8854-4-jgross@suse.com
Signed-off-by: Juergen Gross

xen/arm: Fix race in RB-tree based P2M accounting

2022-07-01T08:02:03+00:00

During the PV driver life cycle the mappings are added to
the RB-tree by set_foreign_p2m_mapping(), which is called from
gnttab_map_refs() and are removed by clear_foreign_p2m_mapping()
which is called from gnttab_unmap_refs(). As both functions end
up calling __set_phys_to_machine_multi() which updates the RB-tree,
this function can be called concurrently.

There is already a "p2m_lock" to protect against concurrent accesses,
but the problem is that the first read of "phys_to_mach.rb_node"
in __set_phys_to_machine_multi() is not covered by it, so this might
lead to the incorrect mappings update (removing in our case) in RB-tree.

In my environment the related issue happens rarely and only when
PV net backend is running, the xen_add_phys_to_mach_entry() claims
that it cannot add new pfn <-> mfn mapping to the tree since it is
already exists which results in a failure when mapping foreign pages.

But there might be other bad consequences related to the non-protected
root reads such use-after-free, etc.

While at it, also fix the similar usage in __pfn_to_mfn(), so
initialize "struct rb_node *n" with the "p2m_lock" held in both
functions to avoid possible bad consequences.

This is CVE-2022-33744 / XSA-406.

Signed-off-by: Oleksandr Tyshchenko 
Reviewed-by: Stefano Stabellini 
Signed-off-by: Juergen Gross

xen/virtio: Enable restricted memory access using Xen grant mappings

2022-06-06T14:07:30+00:00

In order to support virtio in Xen guests add a config option XEN_VIRTIO
enabling the user to specify whether in all Xen guests virtio should
be able to access memory via Xen grant mappings only on the host side.

Also set PLATFORM_VIRTIO_RESTRICTED_MEM_ACCESS feature from the guest
initialization code on Arm and x86 if CONFIG_XEN_VIRTIO is enabled.

Signed-off-by: Juergen Gross 
Signed-off-by: Oleksandr Tyshchenko 
Reviewed-by: Stefano Stabellini 
Reviewed-by: Boris Ostrovsky 
Link: https://lore.kernel.org/r/1654197833-25362-5-git-send-email-olekstysh@gmail.com
Signed-off-by: Juergen Gross

Merge tag 'dma-mapping-5.19-2022-05-25' of git://git.infradead.org/users/hch/dma-mapping

2022-05-26T02:18:36+00:00

Pull dma-mapping updates from Christoph Hellwig:

 - don't over-decrypt memory (Robin Murphy)

 - takes min align mask into account for the swiotlb max mapping size
   (Tianyu Lan)

 - use GFP_ATOMIC in dma-debug (Mikulas Patocka)

 - fix DMA_ATTR_NO_KERNEL_MAPPING on xen/arm (me)

 - don't fail on highmem CMA pages in dma_direct_alloc_pages (me)

 - cleanup swiotlb initialization and share more code with swiotlb-xen
   (me, Stefano Stabellini)

* tag 'dma-mapping-5.19-2022-05-25' of git://git.infradead.org/users/hch/dma-mapping: (23 commits)
  dma-direct: don't over-decrypt memory
  swiotlb: max mapping size takes min align mask into account
  swiotlb: use the right nslabs-derived sizes in swiotlb_init_late
  swiotlb: use the right nslabs value in swiotlb_init_remap
  swiotlb: don't panic when the swiotlb buffer can't be allocated
  dma-debug: change allocation mode from GFP_NOWAIT to GFP_ATIOMIC
  dma-direct: don't fail on highmem CMA pages in dma_direct_alloc_pages
  swiotlb-xen: fix DMA_ATTR_NO_KERNEL_MAPPING on arm
  x86: remove cruft from 
  swiotlb: remove swiotlb_init_with_tbl and swiotlb_init_late_with_tbl
  swiotlb: merge swiotlb-xen initialization into swiotlb
  swiotlb: provide swiotlb_init variants that remap the buffer
  swiotlb: pass a gfp_mask argument to swiotlb_init_late
  swiotlb: add a SWIOTLB_ANY flag to lift the low memory restriction
  swiotlb: make the swiotlb_init interface more useful
  x86: centralize setting SWIOTLB_FORCE when guest memory encryption is enabled
  x86: remove the IOMMU table infrastructure
  MIPS/octeon: use swiotlb_init instead of open coding it
  arm/xen: don't check for xen_initial_domain() in xen_create_contiguous_region
  swiotlb: rename swiotlb_late_init_with_default_size
  ...

swiotlb-xen: fix DMA_ATTR_NO_KERNEL_MAPPING on arm

2022-05-11T17:48:32+00:00

swiotlb-xen uses very different ways to allocate coherent memory on x86
vs arm.  On the former it allocates memory from the page allocator, while
on the later it reuses the dma-direct allocator the handles the
complexities of non-coherent DMA on arm platforms.

Unfortunately the complexities of trying to deal with the two cases in
the swiotlb-xen.c code lead to a bug in the handling of
DMA_ATTR_NO_KERNEL_MAPPING on arm.  With the DMA_ATTR_NO_KERNEL_MAPPING
flag the coherent memory allocator does not actually allocate coherent
memory, but just a DMA handle for some memory that is DMA addressable
by the device, but which does not have to have a kernel mapping.  Thus
dereferencing the return value will lead to kernel crashed and memory
corruption.

Fix this by using the dma-direct allocator directly for arm, which works
perfectly fine because on arm swiotlb-xen is only used when the domain is
1:1 mapped, and then simplifying the remaining code to only cater for the
x86 case with DMA coherent device.

Reported-by: Rahul Singh 
Signed-off-by: Christoph Hellwig 
Reviewed-by: Rahul Singh 
Reviewed-by: Stefano Stabellini 
Tested-by: Rahul Singh