linux-stable.git/drivers/xen, branch v5.3.8

xen/pci: reserve MCFG areas earlier

2019-10-11T16:36:30+00:00

[ Upstream commit a4098bc6eed5e31e0391bcc068e61804c98138df ]

If MCFG area is not reserved in E820, Xen by default will defer its usage
until Dom0 registers it explicitly after ACPI parser recognizes it as
a reserved resource in DSDT. Having it reserved in E820 is not
mandatory according to "PCI Firmware Specification, rev 3.2" (par. 4.1.2)
and firmware is free to keep a hole in E820 in that place. Xen doesn't know
what exactly is inside this hole since it lacks full ACPI view of the
platform therefore it's potentially harmful to access MCFG region
without additional checks as some machines are known to provide
inconsistent information on the size of the region.

Now xen_mcfg_late() runs after acpi_init() which is too late as some basic
PCI enumeration starts exactly there as well. Trying to register a device
prior to MCFG reservation causes multiple problems with PCIe extended
capability initializations in Xen (e.g. SR-IOV VF BAR sizing). There are
no convenient hooks for us to subscribe to so register MCFG areas earlier
upon the first invocation of xen_add_device(). It should be safe to do once
since all the boot time buses must have their MCFG areas in MCFG table
already and we don't support PCI bus hot-plug.

Signed-off-by: Igor Druzhinin 
Reviewed-by: Boris Ostrovsky 
Signed-off-by: Boris Ostrovsky 
Signed-off-by: Sasha Levin

xen/xenbus: fix self-deadlock after killing user process

2019-10-11T16:36:22+00:00

commit a8fabb38525c51a094607768bac3ba46b3f4a9d5 upstream.

In case a user process using xenbus has open transactions and is killed
e.g. via ctrl-C the following cleanup of the allocated resources might
result in a deadlock due to trying to end a transaction in the xenbus
worker thread:

[ 2551.474706] INFO: task xenbus:37 blocked for more than 120 seconds.
[ 2551.492215]       Tainted: P           OE     5.0.0-29-generic #5
[ 2551.510263] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 2551.528585] xenbus          D    0    37      2 0x80000080
[ 2551.528590] Call Trace:
[ 2551.528603]  __schedule+0x2c0/0x870
[ 2551.528606]  ? _cond_resched+0x19/0x40
[ 2551.528632]  schedule+0x2c/0x70
[ 2551.528637]  xs_talkv+0x1ec/0x2b0
[ 2551.528642]  ? wait_woken+0x80/0x80
[ 2551.528645]  xs_single+0x53/0x80
[ 2551.528648]  xenbus_transaction_end+0x3b/0x70
[ 2551.528651]  xenbus_file_free+0x5a/0x160
[ 2551.528654]  xenbus_dev_queue_reply+0xc4/0x220
[ 2551.528657]  xenbus_thread+0x7de/0x880
[ 2551.528660]  ? wait_woken+0x80/0x80
[ 2551.528665]  kthread+0x121/0x140
[ 2551.528667]  ? xb_read+0x1d0/0x1d0
[ 2551.528670]  ? kthread_park+0x90/0x90
[ 2551.528673]  ret_from_fork+0x35/0x40

Fix this by doing the cleanup via a workqueue instead.

Reported-by: James Dingwall 
Fixes: fd8aa9095a95c ("xen: optimize xenbus driver for multiple concurrent xenstore accesses")
Cc:  # 4.11
Signed-off-by: Juergen Gross 
Reviewed-by: Boris Ostrovsky 
Signed-off-by: Boris Ostrovsky 
Signed-off-by: Greg Kroah-Hartman

xen/balloon: Set pages PageOffline() in balloon_add_region()

2019-10-11T16:36:22+00:00

commit c5ad81eb029570c5ca5859539b0679f07a776d25 upstream.

We are missing a __SetPageOffline(), which is why we can get
!PageOffline() pages onto the balloon list, where
alloc_xenballooned_pages() will complain:

page:ffffea0003e7ffc0 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0
flags: 0xffffe00001000(reserved)
raw: 000ffffe00001000 dead000000000100 dead000000000200 0000000000000000
raw: 0000000000000000 0000000000000000 00000001ffffffff 0000000000000000
page dumped because: VM_BUG_ON_PAGE(!PageOffline(page))
------------[ cut here ]------------
kernel BUG at include/linux/page-flags.h:744!
invalid opcode: 0000 [#1] SMP NOPTI

Reported-by: Marek Marczykowski-Górecki 
Tested-by: Marek Marczykowski-Górecki 
Fixes: 77c4adf6a6df ("xen/balloon: mark inflated pages PG_offline")
Cc: stable@vger.kernel.org # v5.1+
Cc: Boris Ostrovsky 
Cc: Juergen Gross 
Cc: Stefano Stabellini 
Signed-off-by: David Hildenbrand 
Reviewed-by: Boris Ostrovsky 
Signed-off-by: Boris Ostrovsky 
Signed-off-by: Greg Kroah-Hartman

Merge tag 'for-linus-5.3a-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip

2019-08-02T22:26:48+00:00

Pull xen fixes from Juergen Gross:

 - a small cleanup

 - a fix for a build error on ARM with some configs

 - a fix of a patch for the Xen gntdev driver

 - three patches for fixing a potential problem in the swiotlb-xen
   driver which Konrad was fine with me carrying them through the Xen
   tree

* tag 'for-linus-5.3a-rc3-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/swiotlb: remember having called xen_create_contiguous_region()
  xen/swiotlb: simplify range_straddles_page_boundary()
  xen/swiotlb: fix condition for calling xen_destroy_contiguous_region()
  xen: avoid link error on ARM
  xen/gntdev.c: Replace vm_map_pages() with vm_map_pages_zero()
  xen/pciback: remove set but not used variable 'old_state'

xen/swiotlb: remember having called xen_create_contiguous_region()

2019-08-01T04:39:33+00:00

Instead of always calling xen_destroy_contiguous_region() in case the
memory is DMA-able for the used device, do so only in case it has been
made DMA-able via xen_create_contiguous_region() before.

This will avoid a lot of xen_destroy_contiguous_region() calls for
64-bit capable devices.

As the memory in question is owned by swiotlb-xen the PG_owner_priv_1
flag of the first allocated page can be used for remembering.

Signed-off-by: Juergen Gross 
Acked-by: Konrad Rzeszutek Wilk 
Signed-off-by: Juergen Gross

xen/swiotlb: simplify range_straddles_page_boundary()

2019-08-01T04:39:30+00:00

range_straddles_page_boundary() is open coding several macros from
include/xen/page.h. Use those instead. Additionally there is no need
to have check_pages_physically_contiguous() as a separate function as
it is used only once, so merge it into range_straddles_page_boundary().

Signed-off-by: Juergen Gross 
Reviewed-by: Boris Ostrovsky 
Acked-by: Konrad Rzeszutek Wilk 
Signed-off-by: Juergen Gross

xen/swiotlb: fix condition for calling xen_destroy_contiguous_region()

2019-08-01T04:39:27+00:00

The condition in xen_swiotlb_free_coherent() for deciding whether to
call xen_destroy_contiguous_region() is wrong: in case the region to
be freed is not contiguous calling xen_destroy_contiguous_region() is
the wrong thing to do: it would result in inconsistent mappings of
multiple PFNs to the same MFN. This will lead to various strange
crashes or data corruption.

Instead of calling xen_destroy_contiguous_region() in that case a
warning should be issued as that situation should never occur.

Cc: stable@vger.kernel.org
Signed-off-by: Juergen Gross 
Reviewed-by: Boris Ostrovsky 
Reviewed-by: Jan Beulich 
Acked-by: Konrad Rzeszutek Wilk 
Signed-off-by: Juergen Gross

xen: avoid link error on ARM

2019-07-31T06:14:12+00:00

Building the privcmd code as a loadable module on ARM, we get
a link error due to the private cache management functions:

ERROR: "__sync_icache_dcache" [drivers/xen/xen-privcmd.ko] undefined!

Move the code into a new that is always built in when Xen is enabled,
as suggested by Juergen Gross and Boris Ostrovsky.

Signed-off-by: Arnd Bergmann 
Reviewed-by: Stefano Stabellini 
Signed-off-by: Juergen Gross

xen/gntdev.c: Replace vm_map_pages() with vm_map_pages_zero()

2019-07-31T06:13:59+00:00

'commit df9bde015a72 ("xen/gntdev.c: convert to use vm_map_pages()")'
breaks gntdev driver. If vma->vm_pgoff > 0, vm_map_pages()
will:
 - use map->pages starting at vma->vm_pgoff instead of 0
 - verify map->count against vma_pages()+vma->vm_pgoff instead of just
   vma_pages().

In practice, this breaks using a single gntdev FD for mapping multiple
grants.

relevant strace output:
[pid   857] ioctl(7, IOCTL_GNTDEV_MAP_GRANT_REF, 0x7ffd3407b6d0) = 0
[pid   857] mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, 7, 0) =
0x777f1211b000
[pid   857] ioctl(7, IOCTL_GNTDEV_SET_UNMAP_NOTIFY, 0x7ffd3407b710) = 0
[pid   857] ioctl(7, IOCTL_GNTDEV_MAP_GRANT_REF, 0x7ffd3407b6d0) = 0
[pid   857] mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, 7,
0x1000) = -1 ENXIO (No such device or address)

details here:
https://github.com/QubesOS/qubes-issues/issues/5199

The reason is -> ( copying Marek's word from discussion)

vma->vm_pgoff is used as index passed to gntdev_find_map_index. It's
basically using this parameter for "which grant reference to map".
map struct returned by gntdev_find_map_index() describes just the pages
to be mapped. Specifically map->pages[0] should be mapped at
vma->vm_start, not vma->vm_start+vma->vm_pgoff*PAGE_SIZE.

When trying to map grant with index (aka vma->vm_pgoff) > 1,
__vm_map_pages() will refuse to map it because it will expect map->count
to be at least vma_pages(vma)+vma->vm_pgoff, while it is exactly
vma_pages(vma).

Converting vm_map_pages() to use vm_map_pages_zero() will fix the
problem.

Marek has tested and confirmed the same.

Cc: stable@vger.kernel.org # v5.2+
Fixes: df9bde015a72 ("xen/gntdev.c: convert to use vm_map_pages()")

Reported-by: Marek Marczykowski-Górecki 
Signed-off-by: Souptick Joarder 
Tested-by: Marek Marczykowski-Górecki 
Reviewed-by: Boris Ostrovsky 
Signed-off-by: Juergen Gross

xen/pciback: remove set but not used variable 'old_state'

2019-07-26T08:29:05+00:00

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/xen/xen-pciback/conf_space_capability.c: In function pm_ctrl_write:
drivers/xen/xen-pciback/conf_space_capability.c:119:25: warning:
 variable old_state set but not used [-Wunused-but-set-variable]

It is never used so can be removed.

Reported-by: Hulk Robot 
Signed-off-by: YueHaibing 
Reviewed-by: Boris Ostrovsky 
Signed-off-by: Juergen Gross