linux-stable.git/include/linux, branch v4.14.237

kvm: fix previous commit for 32-bit builds

2021-06-16T09:53:04+00:00

commit 4422829e8053068e0225e4d0ef42dc41ea7c9ef5 upstream.

array_index_nospec does not work for uint64_t on 32-bit builds.
However, the size of a memory slot must be less than 20 bits wide
on those system, since the memory slot must fit in the user
address space.  So just store it in an unsigned long.

Signed-off-by: Paolo Bonzini 
Signed-off-by: Greg Kroah-Hartman

kvm: avoid speculation-based attacks from out-of-range memslot accesses

2021-06-16T09:53:02+00:00

commit da27a83fd6cc7780fea190e1f5c19e87019da65c upstream.

KVM's mechanism for accessing guest memory translates a guest physical
address (gpa) to a host virtual address using the right-shifted gpa
(also known as gfn) and a struct kvm_memory_slot.  The translation is
performed in __gfn_to_hva_memslot using the following formula:

      hva = slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE

It is expected that gfn falls within the boundaries of the guest's
physical memory.  However, a guest can access invalid physical addresses
in such a way that the gfn is invalid.

__gfn_to_hva_memslot is called from kvm_vcpu_gfn_to_hva_prot, which first
retrieves a memslot through __gfn_to_memslot.  While __gfn_to_memslot
does check that the gfn falls within the boundaries of the guest's
physical memory or not, a CPU can speculate the result of the check and
continue execution speculatively using an illegal gfn. The speculation
can result in calculating an out-of-bounds hva.  If the resulting host
virtual address is used to load another guest physical address, this
is effectively a Spectre gadget consisting of two consecutive reads,
the second of which is data dependent on the first.

Right now it's not clear if there are any cases in which this is
exploitable.  One interesting case was reported by the original author
of this patch, and involves visiting guest page tables on x86.  Right
now these are not vulnerable because the hva read goes through get_user(),
which contains an LFENCE speculation barrier.  However, there are
patches in progress for x86 uaccess.h to mask kernel addresses instead of
using LFENCE; once these land, a guest could use speculation to read
from the VMM's ring 3 address space.  Other architectures such as ARM
already use the address masking method, and would be susceptible to
this same kind of data-dependent access gadgets.  Therefore, this patch
proactively protects from these attacks by masking out-of-bounds gfns
in __gfn_to_hva_memslot, which blocks speculation of invalid hvas.

Sean Christopherson noted that this patch does not cover
kvm_read_guest_offset_cached.  This however is limited to a few bytes
past the end of the cache, and therefore it is unlikely to be useful in
the context of building a chain of data dependent accesses.

Reported-by: Artemiy Margaritov 
Co-developed-by: Artemiy Margaritov 
Signed-off-by: Paolo Bonzini 
Signed-off-by: Greg Kroah-Hartman

bpf: Fix leakage of uninitialized bpf stack under speculation

2021-06-10T10:43:53+00:00

commit 801c6058d14a82179a7ee17a4b532cac6fad067f upstream.

The current implemented mechanisms to mitigate data disclosure under
speculation mainly address stack and map value oob access from the
speculative domain. However, Piotr discovered that uninitialized BPF
stack is not protected yet, and thus old data from the kernel stack,
potentially including addresses of kernel structures, could still be
extracted from that 512 bytes large window. The BPF stack is special
compared to map values since it's not zero initialized for every
program invocation, whereas map values /are/ zero initialized upon
their initial allocation and thus cannot leak any prior data in either
domain. In the non-speculative domain, the verifier ensures that every
stack slot read must have a prior stack slot write by the BPF program
to avoid such data leaking issue.

However, this is not enough: for example, when the pointer arithmetic
operation moves the stack pointer from the last valid stack offset to
the first valid offset, the sanitation logic allows for any intermediate
offsets during speculative execution, which could then be used to
extract any restricted stack content via side-channel.

Given for unprivileged stack pointer arithmetic the use of unknown
but bounded scalars is generally forbidden, we can simply turn the
register-based arithmetic operation into an immediate-based arithmetic
operation without the need for masking. This also gives the benefit
of reducing the needed instructions for the operation. Given after
the work in 7fedb63a8307 ("bpf: Tighten speculative pointer arithmetic
mask"), the aux->alu_limit already holds the final immediate value for
the offset register with the known scalar. Thus, a simple mov of the
immediate to AX register with using AX as the source for the original
instruction is sufficient and possible now in this case.

Reported-by: Piotr Krysiuk 
Signed-off-by: Daniel Borkmann 
Tested-by: Piotr Krysiuk 
Reviewed-by: Piotr Krysiuk 
Reviewed-by: John Fastabend 
Acked-by: Alexei Starovoitov 
[fllinden@amazon.com: fixed minor 4.14 conflict because of renamed function]
Signed-off-by: Frank van der Linden 
Signed-off-by: Greg Kroah-Hartman

net: usb: cdc_ncm: don't spew notifications

2021-06-10T10:43:49+00:00

[ Upstream commit de658a195ee23ca6aaffe197d1d2ea040beea0a2 ]

RTL8156 sends notifications about every 32ms.
Only display/log notifications when something changes.

This issue has been reported by others:
	https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1832472
	https://lkml.org/lkml/2020/8/27/1083

...
[785962.779840] usb 1-1: new high-speed USB device number 5 using xhci_hcd
[785962.929944] usb 1-1: New USB device found, idVendor=0bda, idProduct=8156, bcdDevice=30.00
[785962.929949] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=6
[785962.929952] usb 1-1: Product: USB 10/100/1G/2.5G LAN
[785962.929954] usb 1-1: Manufacturer: Realtek
[785962.929956] usb 1-1: SerialNumber: 000000001
[785962.991755] usbcore: registered new interface driver cdc_ether
[785963.017068] cdc_ncm 1-1:2.0: MAC-Address: 00:24:27:88:08:15
[785963.017072] cdc_ncm 1-1:2.0: setting rx_max = 16384
[785963.017169] cdc_ncm 1-1:2.0: setting tx_max = 16384
[785963.017682] cdc_ncm 1-1:2.0 usb0: register 'cdc_ncm' at usb-0000:00:14.0-1, CDC NCM, 00:24:27:88:08:15
[785963.019211] usbcore: registered new interface driver cdc_ncm
[785963.023856] usbcore: registered new interface driver cdc_wdm
[785963.025461] usbcore: registered new interface driver cdc_mbim
[785963.038824] cdc_ncm 1-1:2.0 enx002427880815: renamed from usb0
[785963.089586] cdc_ncm 1-1:2.0 enx002427880815: network connection: disconnected
[785963.121673] cdc_ncm 1-1:2.0 enx002427880815: network connection: disconnected
[785963.153682] cdc_ncm 1-1:2.0 enx002427880815: network connection: disconnected
...

This is about 2KB per second and will overwrite all contents of a 1MB
dmesg buffer in under 10 minutes rendering them useless for debugging
many kernel problems.

This is also an extra 180 MB/day in /var/logs (or 1GB per week) rendering
the majority of those logs useless too.

When the link is up (expected state), spew amount is >2x higher:
...
[786139.600992] cdc_ncm 2-1:2.0 enx002427880815: network connection: connected
[786139.632997] cdc_ncm 2-1:2.0 enx002427880815: 2500 mbit/s downlink 2500 mbit/s uplink
[786139.665097] cdc_ncm 2-1:2.0 enx002427880815: network connection: connected
[786139.697100] cdc_ncm 2-1:2.0 enx002427880815: 2500 mbit/s downlink 2500 mbit/s uplink
[786139.729094] cdc_ncm 2-1:2.0 enx002427880815: network connection: connected
[786139.761108] cdc_ncm 2-1:2.0 enx002427880815: 2500 mbit/s downlink 2500 mbit/s uplink
...

Chrome OS cannot support RTL8156 until this is fixed.

Signed-off-by: Grant Grundler 
Reviewed-by: Hayes Wang 
Link: https://lore.kernel.org/r/20210120011208.3768105-1-grundler@chromium.org
Signed-off-by: Jakub Kicinski 
Signed-off-by: Sasha Levin

hugetlbfs: hugetlb_fault_mutex_hash() cleanup

2021-06-03T06:36:25+00:00

commit 552546366a30d88bd1d6f5efe848b2ab50fd57e5 upstream.

A new clang diagnostic (-Wsizeof-array-div) warns about the calculation
to determine the number of u32's in an array of unsigned longs.
Suppress warning by adding parentheses.

While looking at the above issue, noticed that the 'address' parameter
to hugetlb_fault_mutex_hash is no longer used.  So, remove it from the
definition and all callers.

No functional change.

Link: http://lkml.kernel.org/r/20190919011847.18400-1-mike.kravetz@oracle.com
Signed-off-by: Mike Kravetz 
Reported-by: Nathan Chancellor 
Reviewed-by: Nathan Chancellor 
Reviewed-by: Davidlohr Bueso 
Reviewed-by: Andrew Morton 
Cc: Nick Desaulniers 
Cc: Ilie Halip 
Cc: David Bolvansky 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

netfilter: x_tables: Use correct memory barriers.

2021-06-03T06:36:11+00:00

commit 175e476b8cdf2a4de7432583b49c871345e4f8a1 upstream.

When a new table value was assigned, it was followed by a write memory
barrier. This ensured that all writes before this point would complete
before any writes after this point. However, to determine whether the
rules are unused, the sequence counter is read. To ensure that all
writes have been done before these reads, a full memory barrier is
needed, not just a write memory barrier. The same argument applies when
incrementing the counter, before the rules are read.

Changing to using smp_mb() instead of smp_wmb() fixes the kernel panic
reported in cc00bcaa5899 (which is still present), while still
maintaining the same speed of replacing tables.

The smb_mb() barriers potentially slow the packet path, however testing
has shown no measurable change in performance on a 4-core MIPS64
platform.

Fixes: 7f5c6d4f665b ("netfilter: get rid of atomic ops in fast path")
Signed-off-by: Mark Tomlinson 
Signed-off-by: Pablo Neira Ayuso 
[Ported to stable, affected barrier is added by d3d40f237480abf3268956daf18cdc56edd32834 in mainline]
Signed-off-by: Pavel Machek (CIP) 
Signed-off-by: Nobuhiro Iwamatsu (CIP) 
Signed-off-by: Greg Kroah-Hartman

vt: Fix character height handling with VT_RESIZEX

2021-05-26T09:47:01+00:00

commit 860dafa902595fb5f1d23bbcce1215188c3341e6 upstream.

Restore the original intent of the VT_RESIZEX ioctl's `v_clin' parameter
which is the number of pixel rows per character (cell) rather than the
height of the font used.

For framebuffer devices the two values are always the same, because the
former is inferred from the latter one.  For VGA used as a true text
mode device these two parameters are independent from each other: the
number of pixel rows per character is set in the CRT controller, while
font height is in fact hardwired to 32 pixel rows and fonts of heights
below that value are handled by padding their data with blanks when
loaded to hardware for use by the character generator.  One can change
the setting in the CRT controller and it will update the screen contents
accordingly regardless of the font loaded.

The `v_clin' parameter is used by the `vgacon' driver to set the height
of the character cell and then the cursor position within.  Make the
parameter explicit then, by defining a new `vc_cell_height' struct
member of `vc_data', set it instead of `vc_font.height' from `v_clin' in
the VT_RESIZEX ioctl, and then use it throughout the `vgacon' driver
except where actual font data is accessed which as noted above is
independent from the CRTC setting.

This way the framebuffer console driver is free to ignore the `v_clin'
parameter as irrelevant, as it always should have, avoiding any issues
attempts to give the parameter a meaning there could have caused, such
as one that has led to commit 988d0763361b ("vt_ioctl: make VT_RESIZEX
behave like VT_RESIZE"):

 "syzbot is reporting UAF/OOB read at bit_putcs()/soft_cursor() [1][2],
  for vt_resizex() from ioctl(VT_RESIZEX) allows setting font height
  larger than actual font height calculated by con_font_set() from
  ioctl(PIO_FONT). Since fbcon_set_font() from con_font_set() allocates
  minimal amount of memory based on actual font height calculated by
  con_font_set(), use of vt_resizex() can cause UAF/OOB read for font
  data."

The problem first appeared around Linux 2.5.66 which predates our repo
history, but the origin could be identified with the old MIPS/Linux repo
also at: 
as commit 9736a3546de7 ("Merge with Linux 2.5.66."), where VT_RESIZEX
code in `vt_ioctl' was updated as follows:

 		if (clin)
-			video_font_height = clin;
+			vc->vc_font.height = clin;

making the parameter apply to framebuffer devices as well, perhaps due
to the use of "font" in the name of the original `video_font_height'
variable.  Use "cell" in the new struct member then to avoid ambiguity.

References:

[1] https://syzkaller.appspot.com/bug?id=32577e96d88447ded2d3b76d71254fb855245837
[2] https://syzkaller.appspot.com/bug?id=6b8355d27b2b94fb5cedf4655e3a59162d9e48e3

Signed-off-by: Maciej W. Rozycki 
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org # v2.6.12+
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

smp: Fix smp_call_function_single_async prototype

2021-05-22T08:57:35+00:00

commit 1139aeb1c521eb4a050920ce6c64c36c4f2a3ab7 upstream.

As of commit 966a967116e6 ("smp: Avoid using two cache lines for struct
call_single_data"), the smp code prefers 32-byte aligned call_single_data
objects for performance reasons, but the block layer includes an instance
of this structure in the main 'struct request' that is more senstive
to size than to performance here, see 4ccafe032005 ("block: unalign
call_single_data in struct request").

The result is a violation of the calling conventions that clang correctly
points out:

block/blk-mq.c:630:39: warning: passing 8-byte aligned argument to 32-byte aligned parameter 2 of 'smp_call_function_single_async' may result in an unaligned pointer access [-Walign-mismatch]
                smp_call_function_single_async(cpu, &rq->csd);

It does seem that the usage of the call_single_data without cache line
alignment should still be allowed by the smp code, so just change the
function prototype so it accepts both, but leave the default alignment
unchanged for the other users. This seems better to me than adding
a local hack to shut up an otherwise correct warning in the caller.

Signed-off-by: Arnd Bergmann 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Jens Axboe 
Link: https://lkml.kernel.org/r/20210505211300.3174456-1-arnd@kernel.org
[nc: Fix conflicts]
Signed-off-by: Nathan Chancellor 
Signed-off-by: Greg Kroah-Hartman

HID: plantronics: Workaround for double volume key presses

2021-05-22T08:57:31+00:00

[ Upstream commit f567d6ef8606fb427636e824c867229ecb5aefab ]

Plantronics Blackwire 3220 Series (047f:c056) sends HID reports twice
for each volume key press. This patch adds a quirk to hid-plantronics
for this product ID, which will ignore the second volume key press if
it happens within 5 ms from the last one that was handled.

The patch was tested on the mentioned model only, it shouldn't affect
other models, however, this quirk might be needed for them too.
Auto-repeat (when a key is held pressed) is not affected, because the
rate is about 3 times per second, which is far less frequent than once
in 5 ms.

Fixes: 81bb773faed7 ("HID: plantronics: Update to map volume up/down controls")
Signed-off-by: Maxim Mikityanskiy 
Signed-off-by: Jiri Kosina 
Signed-off-by: Sasha Levin

tty: fix return value for unsupported ioctls

2021-05-22T08:57:29+00:00

[ Upstream commit 1b8b20868a6d64cfe8174a21b25b74367bdf0560 ]

Drivers should return -ENOTTY ("Inappropriate I/O control operation")
when an ioctl isn't supported, while -EINVAL is used for invalid
arguments.

Fix up the TIOCMGET, TIOCMSET and TIOCGICOUNT helpers which returned
-EINVAL when a tty driver did not implement the corresponding
operations.

Note that the TIOCMGET and TIOCMSET helpers predate git and do not get a
corresponding Fixes tag below.

Fixes: d281da7ff6f7 ("tty: Make tiocgicount a handler")
Signed-off-by: Johan Hovold 
Link: https://lore.kernel.org/r/20210407095208.31838-3-johan@kernel.org
Signed-off-by: Greg Kroah-Hartman 
Signed-off-by: Sasha Levin