]> git.proxmox.com Git - mirror_ubuntu-kernels.git/log
mirror_ubuntu-kernels.git
3 years agobpf: Simplify alu_limit masking for pointer arithmetic
Piotr Krysiuk [Tue, 16 Mar 2021 07:26:25 +0000 (08:26 +0100)]
bpf: Simplify alu_limit masking for pointer arithmetic

Instead of having the mov32 with aux->alu_limit - 1 immediate, move this
operation to retrieve_ptr_limit() instead to simplify the logic and to
allow for subsequent sanity boundary checks inside retrieve_ptr_limit().
This avoids in future that at the time of the verifier masking rewrite
we'd run into an underflow which would not sign extend due to the nature
of mov32 instruction.

Signed-off-by: Piotr Krysiuk <piotras@gmail.com>
Co-developed-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
3 years agobpf: Fix off-by-one for area size in creating mask to left
Piotr Krysiuk [Tue, 16 Mar 2021 07:20:16 +0000 (08:20 +0100)]
bpf: Fix off-by-one for area size in creating mask to left

retrieve_ptr_limit() computes the ptr_limit for registers with stack and
map_value type. ptr_limit is the size of the memory area that is still
valid / in-bounds from the point of the current position and direction
of the operation (add / sub). This size will later be used for masking
the operation such that attempting out-of-bounds access in the speculative
domain is redirected to remain within the bounds of the current map value.

When masking to the right the size is correct, however, when masking to
the left, the size is off-by-one which would lead to an incorrect mask
and thus incorrect arithmetic operation in the non-speculative domain.
Piotr found that if the resulting alu_limit value is zero, then the
BPF_MOV32_IMM() from the fixup_bpf_calls() rewrite will end up loading
0xffffffff into AX instead of sign-extending to the full 64 bit range,
and as a result, this allows abuse for executing speculatively out-of-
bounds loads against 4GB window of address space and thus extracting the
contents of kernel memory via side-channel.

Fixes: 979d63d50c0c ("bpf: prevent out of bounds speculation on pointer arithmetic")
Signed-off-by: Piotr Krysiuk <piotras@gmail.com>
Co-developed-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
3 years agobpf: Prohibit alu ops for pointer types not defining ptr_limit
Piotr Krysiuk [Tue, 16 Mar 2021 08:47:02 +0000 (09:47 +0100)]
bpf: Prohibit alu ops for pointer types not defining ptr_limit

The purpose of this patch is to streamline error propagation and in particular
to propagate retrieve_ptr_limit() errors for pointer types that are not defining
a ptr_limit such that register-based alu ops against these types can be rejected.

The main rationale is that a gap has been identified by Piotr in the existing
protection against speculatively out-of-bounds loads, for example, in case of
ctx pointers, unprivileged programs can still perform pointer arithmetic. This
can be abused to execute speculatively out-of-bounds loads without restrictions
and thus extract contents of kernel memory.

Fix this by rejecting unprivileged programs that attempt any pointer arithmetic
on unprotected pointer types. The two affected ones are pointer to ctx as well
as pointer to map. Field access to a modified ctx' pointer is rejected at a
later point in time in the verifier, and 7c6967326267 ("bpf: Permit map_ptr
arithmetic with opcode add and offset 0") only relevant for root-only use cases.
Risk of unprivileged program breakage is considered very low.

Fixes: 7c6967326267 ("bpf: Permit map_ptr arithmetic with opcode add and offset 0")
Fixes: b2157399cc98 ("bpf: prevent out-of-bounds speculation")
Signed-off-by: Piotr Krysiuk <piotras@gmail.com>
Co-developed-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
3 years agoftrace: Fix modify_ftrace_direct.
Alexei Starovoitov [Tue, 16 Mar 2021 19:58:15 +0000 (12:58 -0700)]
ftrace: Fix modify_ftrace_direct.

The following sequence of commands:
  register_ftrace_direct(ip, addr1);
  modify_ftrace_direct(ip, addr1, addr2);
  unregister_ftrace_direct(ip, addr2);
will cause the kernel to warn:
[   30.179191] WARNING: CPU: 2 PID: 1961 at kernel/trace/ftrace.c:5223 unregister_ftrace_direct+0x130/0x150
[   30.180556] CPU: 2 PID: 1961 Comm: test_progs    W  O      5.12.0-rc2-00378-g86bc10a0a711-dirty #3246
[   30.182453] RIP: 0010:unregister_ftrace_direct+0x130/0x150

When modify_ftrace_direct() changes the addr from old to new it should update
the addr stored in ftrace_direct_funcs. Otherwise the final
unregister_ftrace_direct() won't find the address and will cause the splat.

Fixes: 0567d6809182 ("ftrace: Add modify_ftrace_direct()")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Link: https://lore.kernel.org/bpf/20210316195815.34714-1-alexei.starovoitov@gmail.com
3 years agoselftests/bpf: Set gopt opt_class to 0 if get tunnel opt failed
Hangbin Liu [Tue, 9 Mar 2021 03:22:14 +0000 (11:22 +0800)]
selftests/bpf: Set gopt opt_class to 0 if get tunnel opt failed

When fixing the bpf test_tunnel.sh geneve failure. I only fixed the IPv4
part but forgot the IPv6 issue. Similar with the IPv4 fixes 557c223b643a
("selftests/bpf: No need to drop the packet when there is no geneve opt"),
when there is no tunnel option and bpf_skb_get_tunnel_opt() returns error,
there is no need to drop the packets and break all geneve rx traffic.
Just set opt_class to 0 and keep returning TC_ACT_OK at the end.

Fixes: 557c223b643a ("selftests/bpf: No need to drop the packet when there is no geneve opt")
Fixes: 933a741e3b82 ("selftests/bpf: bpf tunnel test.")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: William Tu <u9012063@gmail.com>
Link: https://lore.kernel.org/bpf/20210309032214.2112438-1-liuhangbin@gmail.com
3 years agoflow_dissector: fix byteorder of dissected ICMP ID
Alexander Lobakin [Fri, 12 Mar 2021 20:08:57 +0000 (20:08 +0000)]
flow_dissector: fix byteorder of dissected ICMP ID

flow_dissector_key_icmp::id is of type u16 (CPU byteorder),
ICMP header has its ID field in network byteorder obviously.
Sparse says:

net/core/flow_dissector.c:178:43: warning: restricted __be16 degrades to integer

Convert ID value to CPU byteorder when storing it into
flow_dissector_key_icmp.

Fixes: 5dec597e5cd0 ("flow_dissector: extract more ICMP information")
Signed-off-by: Alexander Lobakin <alobakin@pm.me>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: qrtr: fix a kernel-infoleak in qrtr_recvmsg()
Eric Dumazet [Fri, 12 Mar 2021 16:59:48 +0000 (08:59 -0800)]
net: qrtr: fix a kernel-infoleak in qrtr_recvmsg()

struct sockaddr_qrtr has a 2-byte hole, and qrtr_recvmsg() currently
does not clear it before copying kernel data to user space.

It might be too late to name the hole since sockaddr_qrtr structure is uapi.

BUG: KMSAN: kernel-infoleak in kmsan_copy_to_user+0x9c/0xb0 mm/kmsan/kmsan_hooks.c:249
CPU: 0 PID: 29705 Comm: syz-executor.3 Not tainted 5.11.0-rc7-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:79 [inline]
 dump_stack+0x21c/0x280 lib/dump_stack.c:120
 kmsan_report+0xfb/0x1e0 mm/kmsan/kmsan_report.c:118
 kmsan_internal_check_memory+0x202/0x520 mm/kmsan/kmsan.c:402
 kmsan_copy_to_user+0x9c/0xb0 mm/kmsan/kmsan_hooks.c:249
 instrument_copy_to_user include/linux/instrumented.h:121 [inline]
 _copy_to_user+0x1ac/0x270 lib/usercopy.c:33
 copy_to_user include/linux/uaccess.h:209 [inline]
 move_addr_to_user+0x3a2/0x640 net/socket.c:237
 ____sys_recvmsg+0x696/0xd50 net/socket.c:2575
 ___sys_recvmsg net/socket.c:2610 [inline]
 do_recvmmsg+0xa97/0x22d0 net/socket.c:2710
 __sys_recvmmsg net/socket.c:2789 [inline]
 __do_sys_recvmmsg net/socket.c:2812 [inline]
 __se_sys_recvmmsg+0x24a/0x410 net/socket.c:2805
 __x64_sys_recvmmsg+0x62/0x80 net/socket.c:2805
 do_syscall_64+0x9f/0x140 arch/x86/entry/common.c:48
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x465f69
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f43659d6188 EFLAGS: 00000246 ORIG_RAX: 000000000000012b
RAX: ffffffffffffffda RBX: 000000000056bf60 RCX: 0000000000465f69
RDX: 0000000000000008 RSI: 0000000020003e40 RDI: 0000000000000003
RBP: 00000000004bfa8f R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000010060 R11: 0000000000000246 R12: 000000000056bf60
R13: 0000000000a9fb1f R14: 00007f43659d6300 R15: 0000000000022000

Local variable ----addr@____sys_recvmsg created at:
 ____sys_recvmsg+0x168/0xd50 net/socket.c:2550
 ____sys_recvmsg+0x168/0xd50 net/socket.c:2550

Bytes 2-3 of 12 are uninitialized
Memory access of size 12 starts at ffff88817c627b40
Data copied to user address 0000000020000140

Fixes: bdabad3e363d ("net: Add Qualcomm IPC router")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Courtney Cavin <courtney.cavin@sonymobile.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: arcnet: com20020 fix error handling
Tong Zhang [Sun, 14 Mar 2021 18:08:36 +0000 (14:08 -0400)]
net: arcnet: com20020 fix error handling

There are two issues when handling error case in com20020pci_probe()

1. priv might be not initialized yet when calling com20020pci_remove()
from com20020pci_probe(), since the priv is set at the very last but it
can jump to error handling in the middle and priv remains NULL.
2. memory leak - the net device is allocated in alloc_arcdev but not
properly released if error happens in the middle of the big for loop

[    1.529110] BUG: kernel NULL pointer dereference, address: 0000000000000008
[    1.531447] RIP: 0010:com20020pci_remove+0x15/0x60 [com20020_pci]
[    1.536805] Call Trace:
[    1.536939]  com20020pci_probe+0x3f2/0x48c [com20020_pci]
[    1.537226]  local_pci_probe+0x48/0x80
[    1.539918]  com20020pci_init+0x3f/0x1000 [com20020_pci]

Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agodevlink: fix typo in documentation
Eva Dengler [Sat, 13 Mar 2021 00:04:13 +0000 (01:04 +0100)]
devlink: fix typo in documentation

This commit fixes three spelling typos in devlink-dpipe.rst and
devlink-port.rst.

Signed-off-by: Eva Dengler <eva.dengler@fau.de>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: ipa: terminate message handler arrays
Alex Elder [Fri, 12 Mar 2021 15:12:48 +0000 (09:12 -0600)]
net: ipa: terminate message handler arrays

When a QMI handle is initialized, an array of message handler
structures is provided, defining how any received message should
be handled based on its type and message ID.  The QMI core code
traverses this array when a message arrives and calls the function
associated with the (type, msg_id) found in the array.

The array is supposed to be terminated with an empty (all zero)
entry though.  Without it, an unsupported message will cause
the QMI core code to go past the end of the array.

Fix this bug, by properly terminating the message handler arrays
provided when QMI handles are set up by the IPA driver.

Fixes: 530f9216a9537 ("soc: qcom: ipa: AP/modem communications")
Reported-by: Sujit Kautkar <sujitka@chromium.org>
Signed-off-by: Alex Elder <elder@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git
David S. Miller [Sat, 13 Mar 2021 02:02:28 +0000 (18:02 -0800)]
Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git
/tnguy/net-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2021-03-12

This series contains updates to ice, i40e, ixgbe and igb drivers.

Magnus adjusts the return value for xsk allocation for ice. This fixes
reporting of napi work done and matches the behavior of other Intel NIC
drivers for xsk allocations.

Maciej moves storing of the rx_offset value to after the build_skb flag
is set as this flag affects the offset value for ice, i40e, and ixgbe.

Li RongQing resolves an issue where an Rx buffer can be reused
prematurely with XDP redirect for igb.
====================

3 years agoibmvnic: update MAINTAINERS
Lijun Pan [Fri, 12 Mar 2021 18:45:30 +0000 (12:45 -0600)]
ibmvnic: update MAINTAINERS

Tom wrote most of the driver code and his experience is valuable to us.
Add him as a Reviewer so that patches will be Cc'ed and reviewed by him.

Signed-off-by: Lijun Pan <ljp@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoselftests: mptcp: Restore packet capture option in join tests
Mat Martineau [Sat, 13 Mar 2021 00:43:52 +0000 (16:43 -0800)]
selftests: mptcp: Restore packet capture option in join tests

The join self tests previously used the '-c' command line option to
enable creation of pcap files for the tests that run, but the change to
allow running a subset of the join tests made overlapping use of that
option.

Restore the capture functionality with '-c' and move the syncookie test
option to '-k'.

Fixes: 1002b89f23ea ("selftests: mptcp: add command line arguments for mptcp_join.sh")
Acked-and-tested-by: Geliang Tang <geliangtang@gmail.com>
Co-developed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: dsa: mt7530: setup core clock even in TRGMII mode
Ilya Lipnitskiy [Fri, 12 Mar 2021 08:07:03 +0000 (00:07 -0800)]
net: dsa: mt7530: setup core clock even in TRGMII mode

A recent change to MIPS ralink reset logic made it so mt7530 actually
resets the switch on platforms such as mt7621 (where bit 2 is the reset
line for the switch). That exposed an issue where the switch would not
function properly in TRGMII mode after a reset.

Reconfigure core clock in TRGMII mode to fix the issue.

Tested on Ubiquiti ER-X (MT7621) with TRGMII mode enabled.

Fixes: 3f9ef7785a9c ("MIPS: ralink: manage low reset lines")
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agomptcp: fix bit MPTCP_PUSH_PENDING tests
Dan Carpenter [Fri, 12 Mar 2021 07:41:12 +0000 (10:41 +0300)]
mptcp: fix bit MPTCP_PUSH_PENDING tests

The MPTCP_PUSH_PENDING define is 6 and these tests should be testing if
BIT(6) is set.

Fixes: c2e6048fa1cf ("mptcp: fix race in release_cb")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: phy: broadcom: Fix RGMII delays for BCM50160 and BCM50610M
Florian Fainelli [Fri, 12 Mar 2021 00:52:50 +0000 (16:52 -0800)]
net: phy: broadcom: Fix RGMII delays for BCM50160 and BCM50610M

The PHY driver entry for BCM50160 and BCM50610M calls
bcm54xx_config_init() but does not call bcm54xx_config_clock_delay() in
order to configuration appropriate clock delays on the PHY, fix that.

Fixes: 733336262b28 ("net: phy: Allow BCM5481x PHYs to setup internal TX/RX clock delay")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoftgmac100: Restart MAC HW once
Dylan Hung [Fri, 12 Mar 2021 00:34:05 +0000 (11:04 +1030)]
ftgmac100: Restart MAC HW once

The interrupt handler may set the flag to reset the mac in the future,
but that flag is not cleared once the reset has occurred.

Fixes: 10cbd6407609 ("ftgmac100: Rework NAPI & interrupts handling")
Signed-off-by: Dylan Hung <dylan_hung@aspeedtech.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: axienet: Fix probe error cleanup
Robert Hancock [Thu, 11 Mar 2021 20:05:18 +0000 (14:05 -0600)]
net: axienet: Fix probe error cleanup

The driver did not always clean up all allocated resources when probe
failed. Fix the probe cleanup path to clean up everything that was
allocated.

Fixes: 57baf8cc70ea ("net: axienet: Handle deferred probe on clock properly")
Signed-off-by: Robert Hancock <robert.hancock@calian.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: correct sk_acceptq_is_full()
liuyacan [Thu, 11 Mar 2021 16:32:25 +0000 (00:32 +0800)]
net: correct sk_acceptq_is_full()

The "backlog" argument in listen() specifies
the maximom length of pending connections,
so the accept queue should be considered full
if there are exactly "backlog" elements.

Signed-off-by: liuyacan <yacanliu@163.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoRevert "net: bonding: fix error return code of bond_neigh_init()"
David S. Miller [Fri, 12 Mar 2021 20:15:03 +0000 (12:15 -0800)]
Revert "net: bonding: fix error return code of bond_neigh_init()"

This reverts commit 2055a99da8a253a357bdfd359b3338ef3375a26c.

This change rejects legitimate configurations.

A slave doesn't need to exist nor implement ndo_slave_setup.

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoigb: avoid premature Rx buffer reuse
Li RongQing [Thu, 21 Jan 2021 21:54:23 +0000 (13:54 -0800)]
igb: avoid premature Rx buffer reuse

Igb needs a similar fix as commit 75aab4e10ae6a ("i40e: avoid
premature Rx buffer reuse")

The page recycle code, incorrectly, relied on that a page fragment
could not be freed inside xdp_do_redirect(). This assumption leads to
that page fragments that are used by the stack/XDP redirect can be
reused and overwritten.

To avoid this, store the page count prior invoking xdp_do_redirect().

Longer explanation:

Intel NICs have a recycle mechanism. The main idea is that a page is
split into two parts. One part is owned by the driver, one part might
be owned by someone else, such as the stack.

t0: Page is allocated, and put on the Rx ring
              +---------------
used by NIC ->| upper buffer
(rx_buffer)   +---------------
              | lower buffer
              +---------------
  page count  == USHRT_MAX
  rx_buffer->pagecnt_bias == USHRT_MAX

t1: Buffer is received, and passed to the stack (e.g.)
              +---------------
              | upper buff (skb)
              +---------------
used by NIC ->| lower buffer
(rx_buffer)   +---------------
  page count  == USHRT_MAX
  rx_buffer->pagecnt_bias == USHRT_MAX - 1

t2: Buffer is received, and redirected
              +---------------
              | upper buff (skb)
              +---------------
used by NIC ->| lower buffer
(rx_buffer)   +---------------

Now, prior calling xdp_do_redirect():
  page count  == USHRT_MAX
  rx_buffer->pagecnt_bias == USHRT_MAX - 2

This means that buffer *cannot* be flipped/reused, because the skb is
still using it.

The problem arises when xdp_do_redirect() actually frees the
segment. Then we get:
  page count  == USHRT_MAX - 1
  rx_buffer->pagecnt_bias == USHRT_MAX - 2

From a recycle perspective, the buffer can be flipped and reused,
which means that the skb data area is passed to the Rx HW ring!

To work around this, the page count is stored prior calling
xdp_do_redirect().

Fixes: 9cbc948b5a20 ("igb: add XDP support")
Signed-off-by: Li RongQing <lirongqing@baidu.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Tested-by: Vishakha Jambekar <vishakha.jambekar@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoixgbe: move headroom initialization to ixgbe_configure_rx_ring
Maciej Fijalkowski [Wed, 3 Mar 2021 15:39:28 +0000 (16:39 +0100)]
ixgbe: move headroom initialization to ixgbe_configure_rx_ring

ixgbe_rx_offset(), that is supposed to initialize the Rx buffer headroom,
relies on __IXGBE_RX_BUILD_SKB_ENABLED flag.

Currently, the callsite of mentioned function is placed incorrectly
within ixgbe_setup_rx_resources() where Rx ring's build skb flag is not
set yet. This causes the XDP_REDIRECT to be partially broken due to
inability to create xdp_frame in the headroom space, as the headroom is
0.

Fix this by moving ixgbe_rx_offset() to ixgbe_configure_rx_ring() after
the flag setting, which happens to be set in ixgbe_set_rx_buffer_len.

Fixes: c0d4e9d223c5 ("ixgbe: store the result of ixgbe_rx_offset() onto ixgbe_ring")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: Vishakha Jambekar <vishakha.jambekar@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoice: move headroom initialization to ice_setup_rx_ctx
Maciej Fijalkowski [Wed, 3 Mar 2021 15:39:27 +0000 (16:39 +0100)]
ice: move headroom initialization to ice_setup_rx_ctx

ice_rx_offset(), that is supposed to initialize the Rx buffer headroom,
relies on ICE_RX_FLAGS_RING_BUILD_SKB flag as well as XDP prog presence.

Currently, the callsite of mentioned function is placed incorrectly
within ice_setup_rx_ring() where Rx ring's build skb flag is not
set yet. This causes the XDP_REDIRECT to be partially broken due to
inability to create xdp_frame in the headroom space, as the headroom is
0.

Fix this by moving ice_rx_offset() to ice_setup_rx_ctx() after the flag
setting.

Fixes: f1b1f409bf79 ("ice: store the result of ice_rx_offset() onto ice_ring")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoi40e: move headroom initialization to i40e_configure_rx_ring
Maciej Fijalkowski [Wed, 3 Mar 2021 15:39:26 +0000 (16:39 +0100)]
i40e: move headroom initialization to i40e_configure_rx_ring

i40e_rx_offset(), that is supposed to initialize the Rx buffer headroom,
relies on I40E_RXR_FLAGS_BUILD_SKB_ENABLED flag.

Currently, the callsite of mentioned function is placed incorrectly
within i40e_setup_rx_descriptors() where Rx ring's build skb flag is not
set yet. This causes the XDP_REDIRECT to be partially broken due to
inability to create xdp_frame in the headroom space, as the headroom is
0.

For the record, below is the call graph:

i40e_vsi_open
 i40e_vsi_setup_rx_resources
  i40e_setup_rx_descriptors
   i40e_rx_offset() <-- sets offset to 0 as build_skb flag is set below

 i40e_vsi_configure_rx
  i40e_configure_rx_ring
   set_ring_build_skb_enabled(ring) <-- set build_skb flag

Fix this by moving i40e_rx_offset() to i40e_configure_rx_ring() after
the flag setting.

Fixes: f7bb0d71d658 ("i40e: store the result of i40e_rx_offset() onto i40e_ring")
Reported-by: Jesper Dangaard Brouer <brouer@redhat.com>
Co-developed-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoice: fix napi work done reporting in xsk path
Magnus Karlsson [Fri, 5 Feb 2021 09:09:04 +0000 (10:09 +0100)]
ice: fix napi work done reporting in xsk path

Fix the wrong napi work done reporting in the xsk path of the ice
driver. The code in the main Rx processing loop was written to assume
that the buffer allocation code returns true if all allocations where
successful and false if not. In contrast with all other Intel NIC xsk
drivers, the ice_alloc_rx_bufs_zc() has the inverted logic messing up
the work done reporting in the napi loop.

This can be fixed either by inverting the return value from
ice_alloc_rx_bufs_zc() in the function that uses this in an incorrect
way, or by changing the return value of ice_alloc_rx_bufs_zc(). We
chose the latter as it makes all the xsk allocation functions for
Intel NICs behave in the same way. My guess is that it was this
unexpected discrepancy that gave rise to this bug in the first place.

Fixes: 5bb0c4b5eb61 ("ice, xsk: Move Rx allocation out of while-loop")
Reported-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Tested-by: Kiran Bhandare <kiranx.bhandare@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoMerge branch 'htb-fixes'
David S. Miller [Fri, 12 Mar 2021 02:30:32 +0000 (18:30 -0800)]
Merge branch 'htb-fixes'

Maxim Mikityanskiy says:

====================
Bugfixes for HTB

The HTB offload feature introduced a few bugs in HTB. One affects the
non-offload mode, preventing attaching qdiscs to HTB classes, and the
other affects the error flow, when the netdev doesn't support the
offload, but it was requested. This short series fixes them.
====================

Acked-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agosch_htb: Fix offload cleanup in htb_destroy on htb_init failure
Maxim Mikityanskiy [Thu, 11 Mar 2021 14:42:06 +0000 (16:42 +0200)]
sch_htb: Fix offload cleanup in htb_destroy on htb_init failure

htb_init may fail to do the offload if it's not supported or if a
runtime error happens when allocating direct qdiscs. In those cases
TC_HTB_CREATE command is not sent to the driver, however, htb_destroy
gets called anyway and attempts to send TC_HTB_DESTROY.

It shouldn't happen, because the driver didn't receive TC_HTB_CREATE,
and also because the driver may not support ndo_setup_tc at all, while
q->offload is true, and htb_destroy mistakenly thinks the offload is
supported. Trying to call ndo_setup_tc in the latter case will lead to a
NULL pointer dereference.

This commit fixes the issues with htb_destroy by deferring assignment of
q->offload until after the TC_HTB_CREATE command. The necessary cleanup
of the offload entities is already done in htb_init.

Reported-by: syzbot+b53a709f04722ca12a3c@syzkaller.appspotmail.com
Fixes: d03b195b5aa0 ("sch_htb: Hierarchical QoS hardware offload")
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agosch_htb: Fix select_queue for non-offload mode
Maxim Mikityanskiy [Thu, 11 Mar 2021 14:42:05 +0000 (16:42 +0200)]
sch_htb: Fix select_queue for non-offload mode

htb_select_queue assumes it's always the offload mode, and it ends up in
calling ndo_setup_tc without any checks. It may lead to a NULL pointer
dereference if ndo_setup_tc is not implemented, or to an error returned
from the driver, which will prevent attaching qdiscs to HTB classes in
the non-offload mode.

This commit fixes the bug by adding the missing check to
htb_select_queue. In the non-offload mode it will return sch->dev_queue,
mimicking tc_modify_qdisc's behavior for the case where select_queue is
not implemented.

Reported-by: syzbot+b53a709f04722ca12a3c@syzkaller.appspotmail.com
Fixes: d03b195b5aa0 ("sch_htb: Hierarchical QoS hardware offload")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: phy: broadcom: Add power down exit reset state delay
Florian Fainelli [Thu, 11 Mar 2021 04:53:42 +0000 (20:53 -0800)]
net: phy: broadcom: Add power down exit reset state delay

Per the datasheet, when we clear the power down bit, the PHY remains in
an internal reset state for 40us and then resume normal operation.
Account for that delay to avoid any issues in the future if
genphy_resume() changes.

Fixes: fe26821fa614 ("net: phy: broadcom: Wire suspend/resume for BCM54810")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agomISDN: fix crash in fritzpci
Tong Zhang [Thu, 11 Mar 2021 04:27:35 +0000 (23:27 -0500)]
mISDN: fix crash in fritzpci

setup_fritz() in avmfritz.c might fail with -EIO and in this case the
isac.type and isac.write_reg is not initialized and remains 0(NULL).
A subsequent call to isac_release() will dereference isac->write_reg and
crash.

[    1.737444] BUG: kernel NULL pointer dereference, address: 0000000000000000
[    1.737809] #PF: supervisor instruction fetch in kernel mode
[    1.738106] #PF: error_code(0x0010) - not-present page
[    1.738378] PGD 0 P4D 0
[    1.738515] Oops: 0010 [#1] SMP NOPTI
[    1.738711] CPU: 0 PID: 180 Comm: systemd-udevd Not tainted 5.12.0-rc2+ #78
[    1.739077] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-48-gd9c812dda519-p
rebuilt.qemu.org 04/01/2014
[    1.739664] RIP: 0010:0x0
[    1.739807] Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
[    1.740200] RSP: 0018:ffffc9000027ba10 EFLAGS: 00010202
[    1.740478] RAX: 0000000000000000 RBX: ffff888102f41840 RCX: 0000000000000027
[    1.740853] RDX: 00000000000000ff RSI: 0000000000000020 RDI: ffff888102f41800
[    1.741226] RBP: ffffc9000027ba20 R08: ffff88817bc18440 R09: ffffc9000027b808
[    1.741600] R10: 0000000000000001 R11: 0000000000000001 R12: ffff888102f41840
[    1.741976] R13: 00000000fffffffb R14: ffff888102f41800 R15: ffff8881008b0000
[    1.742351] FS:  00007fda3a38a8c0(0000) GS:ffff88817bc00000(0000) knlGS:0000000000000000
[    1.742774] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    1.743076] CR2: ffffffffffffffd6 CR3: 00000001021ec000 CR4: 00000000000006f0
[    1.743452] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[    1.743828] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[    1.744206] Call Trace:
[    1.744339]  isac_release+0xcc/0xe0 [mISDNipac]
[    1.744582]  fritzpci_probe.cold+0x282/0x739 [avmfritz]
[    1.744861]  local_pci_probe+0x48/0x80
[    1.745063]  pci_device_probe+0x10f/0x1c0
[    1.745278]  really_probe+0xfb/0x420
[    1.745471]  driver_probe_device+0xe9/0x160
[    1.745693]  device_driver_attach+0x5d/0x70
[    1.745917]  __driver_attach+0x8f/0x150
[    1.746123]  ? device_driver_attach+0x70/0x70
[    1.746354]  bus_for_each_dev+0x7e/0xc0
[    1.746560]  driver_attach+0x1e/0x20
[    1.746751]  bus_add_driver+0x152/0x1f0
[    1.746957]  driver_register+0x74/0xd0
[    1.747157]  ? 0xffffffffc00d8000
[    1.747334]  __pci_register_driver+0x54/0x60
[    1.747562]  AVM_init+0x36/0x1000 [avmfritz]
[    1.747791]  do_one_initcall+0x48/0x1d0
[    1.747997]  ? __cond_resched+0x19/0x30
[    1.748206]  ? kmem_cache_alloc_trace+0x390/0x440
[    1.748458]  ? do_init_module+0x28/0x250
[    1.748669]  do_init_module+0x62/0x250
[    1.748870]  load_module+0x23ee/0x26a0
[    1.749073]  __do_sys_finit_module+0xc2/0x120
[    1.749307]  ? __do_sys_finit_module+0xc2/0x120
[    1.749549]  __x64_sys_finit_module+0x1a/0x20
[    1.749782]  do_syscall_64+0x38/0x90

Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet/qlcnic: Fix a use after free in qlcnic_83xx_get_minidump_template
Lv Yunlong [Thu, 11 Mar 2021 04:01:40 +0000 (20:01 -0800)]
net/qlcnic: Fix a use after free in qlcnic_83xx_get_minidump_template

In qlcnic_83xx_get_minidump_template, fw_dump->tmpl_hdr was freed by
vfree(). But unfortunately, it is used when extended is true.

Fixes: 7061b2bdd620e ("qlogic: Deletion of unnecessary checks before two function calls")
Signed-off-by: Lv Yunlong <lyl2019@mail.ustc.edu.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
David S. Miller [Fri, 12 Mar 2021 00:15:42 +0000 (16:15 -0800)]
Merge branch '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue

Tony Nguyen says:

====================
Intel Wired LAN Driver Updates 2021-03-11

This series contains updates to igc and e1000e drivers.

Sasha adds locking to reset task to prevent race condition for igc.

Muhammad fixes reporting of supported pause frame as well as advertised
pause frame for Tx/Rx off for igc.

Andre fixes timestamp retrieval from the wrong timer for igc.

Vitaly adds locking to reset task to prevent race condition for e1000e.

Dinghao Liu adds a missed check to return on error in
e1000_set_d0_lplu_state_82571.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: sock: simplify tw proto registration
Tonghao Zhang [Thu, 11 Mar 2021 02:57:36 +0000 (10:57 +0800)]
net: sock: simplify tw proto registration

Introduce the new function tw_prot_init (inspired by
req_prot_init) to simplify "proto_register" function.

tw_prot_cleanup will take care of a partially initialized
timewait_sock_ops.

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoe1000e: Fix error handling in e1000_set_d0_lplu_state_82571
Dinghao Liu [Sun, 28 Feb 2021 09:44:23 +0000 (17:44 +0800)]
e1000e: Fix error handling in e1000_set_d0_lplu_state_82571

There is one e1e_wphy() call in e1000_set_d0_lplu_state_82571
that we have caught its return value but lack further handling.
Check and terminate the execution flow just like other e1e_wphy()
in this function.

Fixes: bc7f75fa9788 ("[E1000E]: New pci-express e1000 driver (currently for ICH9 devices only)")
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoe1000e: add rtnl_lock() to e1000_reset_task
Vitaly Lifshits [Wed, 21 Oct 2020 11:59:37 +0000 (14:59 +0300)]
e1000e: add rtnl_lock() to e1000_reset_task

A possible race condition was found in e1000_reset_task,
after discovering a similar issue in igb driver via
commit 024a8168b749 ("igb: reinit_locked() should be called
with rtnl_lock").

Added rtnl_lock() and rtnl_unlock() to avoid this.

Fixes: bc7f75fa9788 ("[E1000E]: New pci-express e1000 driver (currently for ICH9 devices only)")
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Vitaly Lifshits <vitaly.lifshits@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoigc: Fix igc_ptp_rx_pktstamp()
Andre Guedes [Wed, 10 Mar 2021 06:42:56 +0000 (22:42 -0800)]
igc: Fix igc_ptp_rx_pktstamp()

The comment describing the timestamps layout in the packet buffer is
wrong and the code is actually retrieving the timestamp in Timer 1
reference instead of Timer 0. This hasn't been a big issue so far
because hardware is configured to report both timestamps using Timer 0
(see IGC_SRRCTL register configuration in igc_ptp_enable_rx_timestamp()
helper). This patch fixes the comment and the code so we retrieve the
timestamp in Timer 0 reference as expected.

This patch also takes the opportunity to get rid of the hw.mac.type check
since it is not required.

Fixes: 81b055205e8ba ("igc: Add support for RX timestamping")
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoigc: Fix Supported Pause Frame Link Setting
Muhammad Husaini Zulkifli [Fri, 19 Feb 2021 16:36:48 +0000 (00:36 +0800)]
igc: Fix Supported Pause Frame Link Setting

The Supported Pause Frame always display "No" even though the Advertised
pause frame showing the correct setting based on the pause parameters via
ethtool. Set bit in link_ksettings to "Supported" for Pause Frame.

Before output:
Supported pause frame use: No

Expected output:
Supported pause frame use: Symmetric

Fixes: 8c5ad0dae93c ("igc: Add ethtool support")
Signed-off-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Reviewed-by: Malli C <mallikarjuna.chilakala@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoigc: Fix Pause Frame Advertising
Muhammad Husaini Zulkifli [Fri, 19 Feb 2021 16:36:47 +0000 (00:36 +0800)]
igc: Fix Pause Frame Advertising

Fix Pause Frame Advertising when getting the advertisement via ethtool.
Remove setting the "advertising" bit in link_ksettings during default
case when Tx and Rx are in off state with Auto Negotiate off.

Below is the original output of advertisement link during Tx and Rx off:
Advertised pause frame use: Symmetric Receive-only

Expected output:
Advertised pause frame use: No

Fixes: 8c5ad0dae93c ("igc: Add ethtool support")
Signed-off-by: Muhammad Husaini Zulkifli <muhammad.husaini.zulkifli@intel.com>
Reviewed-by: Malli C <mallikarjuna.chilakala@intel.com>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agoigc: reinit_locked() should be called with rtnl_lock
Sasha Neftin [Tue, 20 Oct 2020 13:34:00 +0000 (16:34 +0300)]
igc: reinit_locked() should be called with rtnl_lock

This commit applies to the igc_reset_task the same changes that
were applied to the igb driver in commit 024a8168b749 ("igb:
reinit_locked() should be called with rtnl_lock")
and fix possible race in reset subtask.

Fixes: 0507ef8a0372 ("igc: Add transmit and receive fastpath and interrupt handlers")
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Dvora Fuxbrumer <dvorax.fuxbrumer@linux.intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
3 years agonet: dsa: bcm_sf2: Qualify phydev->dev_flags based on port
Florian Fainelli [Wed, 10 Mar 2021 22:17:58 +0000 (14:17 -0800)]
net: dsa: bcm_sf2: Qualify phydev->dev_flags based on port

Similar to commit 92696286f3bb37ba50e4bd8d1beb24afb759a799 ("net:
bcmgenet: Set phydev->dev_flags only for internal PHYs") we need to
qualify the phydev->dev_flags based on whether the port is connected to
an internal or external PHY otherwise we risk having a flags collision
with a completely different interpretation depending on the driver.

Fixes: aa9aef77c761 ("net: dsa: bcm_sf2: communicate integrated PHY revision to PHY driver")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: dsa: b53: VLAN filtering is global to all users
Florian Fainelli [Wed, 10 Mar 2021 18:46:10 +0000 (10:46 -0800)]
net: dsa: b53: VLAN filtering is global to all users

The bcm_sf2 driver uses the b53 driver as a library but does not make
usre of the b53_setup() function, this made it fail to inherit the
vlan_filtering_is_global attribute. Fix this by moving the assignment to
b53_switch_alloc() which is used by bcm_sf2.

Fixes: 7228b23e68f7 ("net: dsa: b53: Let DSA handle mismatched VLAN filtering settings")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: sched: validate stab values
Eric Dumazet [Wed, 10 Mar 2021 16:26:41 +0000 (08:26 -0800)]
net: sched: validate stab values

iproute2 package is well behaved, but malicious user space can
provide illegal shift values and trigger UBSAN reports.

Add stab parameter to red_check_params() to validate user input.

syzbot reported:

UBSAN: shift-out-of-bounds in ./include/net/red.h:312:18
shift exponent 111 is too large for 64-bit type 'long unsigned int'
CPU: 1 PID: 14662 Comm: syz-executor.3 Not tainted 5.12.0-rc2-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:79 [inline]
 dump_stack+0x141/0x1d7 lib/dump_stack.c:120
 ubsan_epilogue+0xb/0x5a lib/ubsan.c:148
 __ubsan_handle_shift_out_of_bounds.cold+0xb1/0x181 lib/ubsan.c:327
 red_calc_qavg_from_idle_time include/net/red.h:312 [inline]
 red_calc_qavg include/net/red.h:353 [inline]
 choke_enqueue.cold+0x18/0x3dd net/sched/sch_choke.c:221
 __dev_xmit_skb net/core/dev.c:3837 [inline]
 __dev_queue_xmit+0x1943/0x2e00 net/core/dev.c:4150
 neigh_hh_output include/net/neighbour.h:499 [inline]
 neigh_output include/net/neighbour.h:508 [inline]
 ip6_finish_output2+0x911/0x1700 net/ipv6/ip6_output.c:117
 __ip6_finish_output net/ipv6/ip6_output.c:182 [inline]
 __ip6_finish_output+0x4c1/0xe10 net/ipv6/ip6_output.c:161
 ip6_finish_output+0x35/0x200 net/ipv6/ip6_output.c:192
 NF_HOOK_COND include/linux/netfilter.h:290 [inline]
 ip6_output+0x1e4/0x530 net/ipv6/ip6_output.c:215
 dst_output include/net/dst.h:448 [inline]
 NF_HOOK include/linux/netfilter.h:301 [inline]
 NF_HOOK include/linux/netfilter.h:295 [inline]
 ip6_xmit+0x127e/0x1eb0 net/ipv6/ip6_output.c:320
 inet6_csk_xmit+0x358/0x630 net/ipv6/inet6_connection_sock.c:135
 dccp_transmit_skb+0x973/0x12c0 net/dccp/output.c:138
 dccp_send_reset+0x21b/0x2b0 net/dccp/output.c:535
 dccp_finish_passive_close net/dccp/proto.c:123 [inline]
 dccp_finish_passive_close+0xed/0x140 net/dccp/proto.c:118
 dccp_terminate_connection net/dccp/proto.c:958 [inline]
 dccp_close+0xb3c/0xe60 net/dccp/proto.c:1028
 inet_release+0x12e/0x280 net/ipv4/af_inet.c:431
 inet6_release+0x4c/0x70 net/ipv6/af_inet6.c:478
 __sock_release+0xcd/0x280 net/socket.c:599
 sock_close+0x18/0x20 net/socket.c:1258
 __fput+0x288/0x920 fs/file_table.c:280
 task_work_run+0xdd/0x1a0 kernel/task_work.c:140
 tracehook_notify_resume include/linux/tracehook.h:189 [inline]

Fixes: 8afa10cbe281 ("net_sched: red: Avoid illegal values")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge tag 'mlx5-fixes-2021-03-10' of git://git.kernel.org/pub/scm/linux/kernel/git...
David S. Miller [Wed, 10 Mar 2021 23:23:55 +0000 (15:23 -0800)]
Merge tag 'mlx5-fixes-2021-03-10' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

mlx5-fixes-2021-03-10

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: dsa: bcm_sf2: use 2 Gbps IMP port link on BCM4908
Rafaล‚ Miล‚ecki [Wed, 10 Mar 2021 12:51:59 +0000 (13:51 +0100)]
net: dsa: bcm_sf2: use 2 Gbps IMP port link on BCM4908

BCM4908 uses 2 Gbps link between switch and the Ethernet interface.
Without this BCM4908 devices were able to achieve only 2 x ~895 Mb/s.
This allows handling e.g. NAT traffic with 940 Mb/s.

Signed-off-by: Rafaล‚ Miล‚ecki <rafal@milecki.pl>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: pxa168_eth: Fix a potential data race in pxa168_eth_remove
Pavel Andrianov [Wed, 10 Mar 2021 08:10:46 +0000 (11:10 +0300)]
net: pxa168_eth: Fix a potential data race in pxa168_eth_remove

pxa168_eth_remove() firstly calls unregister_netdev(),
then cancels a timeout work. unregister_netdev() shuts down a device
interface and removes it from the kernel tables. If the timeout occurs
in parallel, the timeout work (pxa168_eth_tx_timeout_task) performs stop
and open of the device. It may lead to an inconsistent state and memory
leaks.

Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Pavel Andrianov <andrianov@ispras.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agomacvlan: macvlan_count_rx() needs to be aware of preemption
Eric Dumazet [Wed, 10 Mar 2021 09:56:36 +0000 (01:56 -0800)]
macvlan: macvlan_count_rx() needs to be aware of preemption

macvlan_count_rx() can be called from process context, it is thus
necessary to disable preemption before calling u64_stats_update_begin()

syzbot was able to spot this on 32bit arch:

WARNING: CPU: 1 PID: 4632 at include/linux/seqlock.h:271 __seqprop_assert include/linux/seqlock.h:271 [inline]
WARNING: CPU: 1 PID: 4632 at include/linux/seqlock.h:271 __seqprop_assert.constprop.0+0xf0/0x11c include/linux/seqlock.h:269
Modules linked in:
Kernel panic - not syncing: panic_on_warn set ...
CPU: 1 PID: 4632 Comm: kworker/1:3 Not tainted 5.12.0-rc2-syzkaller #0
Hardware name: ARM-Versatile Express
Workqueue: events macvlan_process_broadcast
Backtrace:
[<82740468>] (dump_backtrace) from [<827406dc>] (show_stack+0x18/0x1c arch/arm/kernel/traps.c:252)
 r7:00000080 r6:60000093 r5:00000000 r4:8422a3c4
[<827406c4>] (show_stack) from [<82751b58>] (__dump_stack lib/dump_stack.c:79 [inline])
[<827406c4>] (show_stack) from [<82751b58>] (dump_stack+0xb8/0xe8 lib/dump_stack.c:120)
[<82751aa0>] (dump_stack) from [<82741270>] (panic+0x130/0x378 kernel/panic.c:231)
 r7:830209b4 r6:84069ea4 r5:00000000 r4:844350d0
[<82741140>] (panic) from [<80244924>] (__warn+0xb0/0x164 kernel/panic.c:605)
 r3:8404ec8c r2:00000000 r1:00000000 r0:830209b4
 r7:0000010f
[<80244874>] (__warn) from [<82741520>] (warn_slowpath_fmt+0x68/0xd4 kernel/panic.c:628)
 r7:81363f70 r6:0000010f r5:83018e50 r4:00000000
[<827414bc>] (warn_slowpath_fmt) from [<81363f70>] (__seqprop_assert include/linux/seqlock.h:271 [inline])
[<827414bc>] (warn_slowpath_fmt) from [<81363f70>] (__seqprop_assert.constprop.0+0xf0/0x11c include/linux/seqlock.h:269)
 r8:5a109000 r7:0000000f r6:a568dac0 r5:89802300 r4:00000001
[<81363e80>] (__seqprop_assert.constprop.0) from [<81364af0>] (u64_stats_update_begin include/linux/u64_stats_sync.h:128 [inline])
[<81363e80>] (__seqprop_assert.constprop.0) from [<81364af0>] (macvlan_count_rx include/linux/if_macvlan.h:47 [inline])
[<81363e80>] (__seqprop_assert.constprop.0) from [<81364af0>] (macvlan_broadcast+0x154/0x26c drivers/net/macvlan.c:291)
 r5:89802300 r4:8a927740
[<8136499c>] (macvlan_broadcast) from [<81365020>] (macvlan_process_broadcast+0x258/0x2d0 drivers/net/macvlan.c:317)
 r10:81364f78 r9:8a86d000 r8:8a9c7e7c r7:8413aa5c r6:00000000 r5:00000000
 r4:89802840
[<81364dc8>] (macvlan_process_broadcast) from [<802696a4>] (process_one_work+0x2d4/0x998 kernel/workqueue.c:2275)
 r10:00000008 r9:8404ec98 r8:84367a02 r7:ddfe6400 r6:ddfe2d40 r5:898dac80
 r4:8a86d43c
[<802693d0>] (process_one_work) from [<80269dcc>] (worker_thread+0x64/0x54c kernel/workqueue.c:2421)
 r10:00000008 r9:8a9c6000 r8:84006d00 r7:ddfe2d78 r6:898dac94 r5:ddfe2d40
 r4:898dac80
[<80269d68>] (worker_thread) from [<80271f40>] (kthread+0x184/0x1a4 kernel/kthread.c:292)
 r10:85247e64 r9:898dac80 r8:80269d68 r7:00000000 r6:8a9c6000 r5:89a2ee40
 r4:8a97bd00
[<80271dbc>] (kthread) from [<80200114>] (ret_from_fork+0x14/0x20 arch/arm/kernel/entry-common.S:158)
Exception stack(0x8a9c7fb0 to 0x8a9c7ff8)

Fixes: 412ca1550cbe ("macvlan: Move broadcasts into a work queue")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Reported-by: syzbot <syzkaller@googlegroups.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agodrop_monitor: Perform cleanup upon probe registration failure
Ido Schimmel [Wed, 10 Mar 2021 10:28:01 +0000 (12:28 +0200)]
drop_monitor: Perform cleanup upon probe registration failure

In the rare case that drop_monitor fails to register its probe on the
'napi_poll' tracepoint, it will not deactivate its hysteresis timer as
part of the error path. If the hysteresis timer was armed by the shortly
lived 'kfree_skb' probe and user space retries to initiate tracing, a
warning will be emitted for trying to initialize an active object [1].

Fix this by properly undoing all the operations that were done prior to
probe registration, in both software and hardware code paths.

Note that syzkaller managed to fail probe registration by injecting a
slab allocation failure [2].

[1]
ODEBUG: init active (active state 0) object type: timer_list hint: sched_send_work+0x0/0x60 include/linux/list.h:135
WARNING: CPU: 1 PID: 8649 at lib/debugobjects.c:505 debug_print_object+0x16e/0x250 lib/debugobjects.c:505
Modules linked in:
CPU: 1 PID: 8649 Comm: syz-executor.0 Not tainted 5.11.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:debug_print_object+0x16e/0x250 lib/debugobjects.c:505
[...]
Call Trace:
 __debug_object_init+0x524/0xd10 lib/debugobjects.c:588
 debug_timer_init kernel/time/timer.c:722 [inline]
 debug_init kernel/time/timer.c:770 [inline]
 init_timer_key+0x2d/0x340 kernel/time/timer.c:814
 net_dm_trace_on_set net/core/drop_monitor.c:1111 [inline]
 set_all_monitor_traces net/core/drop_monitor.c:1188 [inline]
 net_dm_monitor_start net/core/drop_monitor.c:1295 [inline]
 net_dm_cmd_trace+0x720/0x1220 net/core/drop_monitor.c:1339
 genl_family_rcv_msg_doit+0x228/0x320 net/netlink/genetlink.c:739
 genl_family_rcv_msg net/netlink/genetlink.c:783 [inline]
 genl_rcv_msg+0x328/0x580 net/netlink/genetlink.c:800
 netlink_rcv_skb+0x153/0x420 net/netlink/af_netlink.c:2502
 genl_rcv+0x24/0x40 net/netlink/genetlink.c:811
 netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline]
 netlink_unicast+0x533/0x7d0 net/netlink/af_netlink.c:1338
 netlink_sendmsg+0x856/0xd90 net/netlink/af_netlink.c:1927
 sock_sendmsg_nosec net/socket.c:652 [inline]
 sock_sendmsg+0xcf/0x120 net/socket.c:672
 ____sys_sendmsg+0x6e8/0x810 net/socket.c:2348
 ___sys_sendmsg+0xf3/0x170 net/socket.c:2402
 __sys_sendmsg+0xe5/0x1b0 net/socket.c:2435
 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
 entry_SYSCALL_64_after_hwframe+0x44/0xae

[2]
 FAULT_INJECTION: forcing a failure.
 name failslab, interval 1, probability 0, space 0, times 1
 CPU: 1 PID: 8645 Comm: syz-executor.0 Not tainted 5.11.0-syzkaller #0
 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
 Call Trace:
  dump_stack+0xfa/0x151
  should_fail.cold+0x5/0xa
  should_failslab+0x5/0x10
  __kmalloc+0x72/0x3f0
  tracepoint_add_func+0x378/0x990
  tracepoint_probe_register+0x9c/0xe0
  net_dm_cmd_trace+0x7fc/0x1220
  genl_family_rcv_msg_doit+0x228/0x320
  genl_rcv_msg+0x328/0x580
  netlink_rcv_skb+0x153/0x420
  genl_rcv+0x24/0x40
  netlink_unicast+0x533/0x7d0
  netlink_sendmsg+0x856/0xd90
  sock_sendmsg+0xcf/0x120
  ____sys_sendmsg+0x6e8/0x810
  ___sys_sendmsg+0xf3/0x170
  __sys_sendmsg+0xe5/0x1b0
  do_syscall_64+0x2d/0x70
  entry_SYSCALL_64_after_hwframe+0x44/0xae

Fixes: 70c69274f354 ("drop_monitor: Initialize timer and work item upon tracing enable")
Fixes: 8ee2267ad33e ("drop_monitor: Convert to using devlink tracepoint")
Reported-by: syzbot+779559d6503f3a56213d@syzkaller.appspotmail.com
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
David S. Miller [Wed, 10 Mar 2021 23:14:56 +0000 (15:14 -0800)]
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf

Daniel Borkmann says:

====================
pull-request: bpf 2021-03-10

The following pull-request contains BPF updates for your *net* tree.

We've added 8 non-merge commits during the last 5 day(s) which contain
a total of 11 files changed, 136 insertions(+), 17 deletions(-).

The main changes are:

1) Reject bogus use of vmlinux BTF as map/prog creation BTF, from Alexei Starovoitov.

2) Fix allocation failure splat in x86 JIT for large progs. Also fix overwriting
   percpu cgroup storage from tracing programs when nested, from Yonghong Song.

3) Fix rx queue retrieval in XDP for multi-queue veth, from Maciej Fijalkowski.

4) Fix bpf_check_mtu() helper API before freeze to have mtu_len as custom skb/xdp
   L3 input length, from Jesper Dangaard Brouer.

5) Fix inode_storage's lookup_elem return value upon having bad fd, from Tal Lossos.

6) Fix bpftool and libbpf cross-build on MacOS, from Georgi Valkov.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoipv6: fix suspecious RCU usage warning
Wei Wang [Wed, 10 Mar 2021 02:20:35 +0000 (18:20 -0800)]
ipv6: fix suspecious RCU usage warning

Syzbot reported the suspecious RCU usage in nexthop_fib6_nh() when
called from ipv6_route_seq_show(). The reason is ipv6_route_seq_start()
calls rcu_read_lock_bh(), while nexthop_fib6_nh() calls
rcu_dereference_rtnl().
The fix proposed is to add a variant of nexthop_fib6_nh() to use
rcu_dereference_bh_rtnl() for ipv6_route_seq_show().

The reported trace is as follows:
./include/net/nexthop.h:416 suspicious rcu_dereference_check() usage!

other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
2 locks held by syz-executor.0/17895:
     at: seq_read+0x71/0x12a0 fs/seq_file.c:169
     at: seq_file_net include/linux/seq_file_net.h:19 [inline]
     at: ipv6_route_seq_start+0xaf/0x300 net/ipv6/ip6_fib.c:2616

stack backtrace:
CPU: 1 PID: 17895 Comm: syz-executor.0 Not tainted 4.15.0-syzkaller #0
Call Trace:
 [<ffffffff849edf9e>] __dump_stack lib/dump_stack.c:17 [inline]
 [<ffffffff849edf9e>] dump_stack+0xd8/0x147 lib/dump_stack.c:53
 [<ffffffff8480b7fa>] lockdep_rcu_suspicious+0x153/0x15d kernel/locking/lockdep.c:5745
 [<ffffffff8459ada6>] nexthop_fib6_nh include/net/nexthop.h:416 [inline]
 [<ffffffff8459ada6>] ipv6_route_native_seq_show net/ipv6/ip6_fib.c:2488 [inline]
 [<ffffffff8459ada6>] ipv6_route_seq_show+0x436/0x7a0 net/ipv6/ip6_fib.c:2673
 [<ffffffff81c556df>] seq_read+0xccf/0x12a0 fs/seq_file.c:276
 [<ffffffff81dbc62c>] proc_reg_read+0x10c/0x1d0 fs/proc/inode.c:231
 [<ffffffff81bc28ae>] do_loop_readv_writev fs/read_write.c:714 [inline]
 [<ffffffff81bc28ae>] do_loop_readv_writev fs/read_write.c:701 [inline]
 [<ffffffff81bc28ae>] do_iter_read+0x49e/0x660 fs/read_write.c:935
 [<ffffffff81bc81ab>] vfs_readv+0xfb/0x170 fs/read_write.c:997
 [<ffffffff81c88847>] kernel_readv fs/splice.c:361 [inline]
 [<ffffffff81c88847>] default_file_splice_read+0x487/0x9c0 fs/splice.c:416
 [<ffffffff81c86189>] do_splice_to+0x129/0x190 fs/splice.c:879
 [<ffffffff81c86f66>] splice_direct_to_actor+0x256/0x890 fs/splice.c:951
 [<ffffffff81c8777d>] do_splice_direct+0x1dd/0x2b0 fs/splice.c:1060
 [<ffffffff81bc4747>] do_sendfile+0x597/0xce0 fs/read_write.c:1459
 [<ffffffff81bca205>] SYSC_sendfile64 fs/read_write.c:1520 [inline]
 [<ffffffff81bca205>] SyS_sendfile64+0x155/0x170 fs/read_write.c:1506
 [<ffffffff81015fcf>] do_syscall_64+0x1ff/0x310 arch/x86/entry/common.c:305
 [<ffffffff84a00076>] entry_SYSCALL_64_after_hwframe+0x42/0xb7

Fixes: f88d8ea67fbdb ("ipv6: Plumb support for nexthop object in a fib6_info")
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Wei Wang <weiwan@google.com>
Cc: David Ahern <dsahern@kernel.org>
Cc: Ido Schimmel <idosch@idosch.org>
Cc: Petr Machata <petrm@nvidia.com>
Cc: Eric Dumazet <edumazet@google.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch 'ip6ip6-crash'
David S. Miller [Wed, 10 Mar 2021 20:24:19 +0000 (12:24 -0800)]
Merge branch 'ip6ip6-crash'

Daniel Borkmann says:

====================
Fix ip6ip6 crash for collect_md skbs

Fix a NULL pointer deref panic I ran into for regular ip6ip6 tunnel devices
when collect_md populated skbs were redirected to them for xmit. See patches
for further details, thanks!
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet, bpf: Fix ip6ip6 crash with collect_md populated skbs
Daniel Borkmann [Wed, 10 Mar 2021 00:38:10 +0000 (01:38 +0100)]
net, bpf: Fix ip6ip6 crash with collect_md populated skbs

I ran into a crash where setting up a ip6ip6 tunnel device which was /not/
set to collect_md mode was receiving collect_md populated skbs for xmit.

The BPF prog was populating the skb via bpf_skb_set_tunnel_key() which is
assigning special metadata dst entry and then redirecting the skb to the
device, taking ip6_tnl_start_xmit() -> ipxip6_tnl_xmit() -> ip6_tnl_xmit()
and in the latter it performs a neigh lookup based on skb_dst(skb) where
we trigger a NULL pointer dereference on dst->ops->neigh_lookup() since
the md_dst_ops do not populate neigh_lookup callback with a fake handler.

Transform the md_dst_ops into generic dst_blackhole_ops that can also be
reused elsewhere when needed, and use them for the metadata dst entries as
callback ops.

Also, remove the dst_md_discard{,_out}() ops and rely on dst_discard{,_out}()
from dst_init() which free the skb the same way modulo the splat. Given we
will be able to recover just fine from there, avoid any potential splats
iff this gets ever triggered in future (or worse, panic on warns when set).

Fixes: f38a9eb1f77b ("dst: Metadata destinations")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: Consolidate common blackhole dst ops
Daniel Borkmann [Wed, 10 Mar 2021 00:38:09 +0000 (01:38 +0100)]
net: Consolidate common blackhole dst ops

Move generic blackhole dst ops to the core and use them from both
ipv4_dst_blackhole_ops and ip6_dst_blackhole_ops where possible. No
functional change otherwise. We need these also in other locations
and having to define them over and over again is not great.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet/mlx5: DR, Fix potential shift wrapping of 32-bit value in STEv1 getter
Yevgeny Kliteynik [Sat, 6 Feb 2021 20:11:52 +0000 (22:11 +0200)]
net/mlx5: DR, Fix potential shift wrapping of 32-bit value in STEv1 getter

Fix 32-bit variable shift wrapping in dr_ste_v1_get_miss_addr.

Fixes: a6098129c781 ("net/mlx5: DR, Add STEv1 setters and getters")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: SF: Fix error flow of SFs allocation flow
Shay Drory [Thu, 11 Feb 2021 13:41:35 +0000 (15:41 +0200)]
net/mlx5: SF: Fix error flow of SFs allocation flow

When SF id is unavailable, code jumps to wrong label that accesses
sw id array outside of its range.
Hence, when SF id is not allocated, avoid accessing such array.

Fixes: 8f0105418668 ("net/mlx5: SF, Add port add delete functionality")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: SF: Fix memory leak of work item
Shay Drory [Tue, 2 Feb 2021 10:21:04 +0000 (12:21 +0200)]
net/mlx5: SF: Fix memory leak of work item

Cited patch in the fixes tag missed to free the allocated work.
Fix it by freeing the work after work execution.

Fixes: f3196bb0f14c ("net/mlx5: Introduce vhca state event notifier")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: SF, Correct vhca context size
Parav Pandit [Fri, 26 Feb 2021 08:15:16 +0000 (10:15 +0200)]
net/mlx5: SF, Correct vhca context size

Fix vhca context size as defined by device interface specification.

Fixes: f3196bb0f14c ("net/mlx5: Introduce vhca state event notifier")
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: E-switch, Fix rate calculation division
Parav Pandit [Wed, 17 Feb 2021 06:23:29 +0000 (08:23 +0200)]
net/mlx5e: E-switch, Fix rate calculation division

do_div() returns reminder, while cited patch wanted to use
quotient.
Fix it by using quotient.

Fixes: 0e22bfb7c046 ("net/mlx5e: E-switch, Fix rate calculation for overflow")
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agoRDMA/mlx5: Fix timestamp default mode
Maor Gottlieb [Wed, 3 Mar 2021 12:39:20 +0000 (14:39 +0200)]
RDMA/mlx5: Fix timestamp default mode

1. Don't set the ts_format bit to default when it reserved - device is
   running in the old mode (free running).
2. XRC doesn't have a CQ therefore the ts format in the QP
   context should be default / free running.
3. Set ts_format to WQ.

Fixes: 2fe8d4b87802 ("RDMA/mlx5: Fail QP creation if the device can not support the CQE TS")
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: Set QP timestamp mode to default
Maor Gottlieb [Wed, 3 Mar 2021 12:36:16 +0000 (14:36 +0200)]
net/mlx5: Set QP timestamp mode to default

QPs which don't care from timestamp mode, should set the ts_format
to default, otherwise the QP creation could be failed if the timestamp
mode is not supported.

Fixes: 2fe8d4b87802 ("RDMA/mlx5: Fail QP creation if the device can not support the CQE TS")
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Fix error flow in change profile
Roi Dayan [Wed, 10 Feb 2021 08:33:13 +0000 (10:33 +0200)]
net/mlx5e: Fix error flow in change profile

Move priv memset from init to cleanup to avoid double priv cleanup
that can happen on profile change if also roolback fails.
Add missing cleanup flow in mlx5e_netdev_attach_profile().

Fixes: c4d7eb57687f ("net/mxl5e: Add change profile method")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: Disable VF tunnel TX offload if ignore_flow_level isn't supported
Maor Dickman [Mon, 1 Mar 2021 11:07:08 +0000 (13:07 +0200)]
net/mlx5: Disable VF tunnel TX offload if ignore_flow_level isn't supported

VF tunnel TX traffic offload is adding flow which forward to flow
tables with lower level, which isn't support on all FW versions
and may cause firmware to fail with syndrome.

Fixed by enabling VF tunnel TX offload only if flow table capability
ignore_flow_level is enabled.

Fixes: 10742efc20a4 ("net/mlx5e: VF tunnel TX traffic offloading")
Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Check correct ip_version in decapsulation route resolution
Roi Dayan [Tue, 2 Mar 2021 09:31:08 +0000 (11:31 +0200)]
net/mlx5e: Check correct ip_version in decapsulation route resolution

flow_attr->ip_version has the matching that should be done inner/outer.
When working with chains, decapsulation is done on chain0 and next chain
match on outer header which is the original inner which could be ipv4.
So in tunnel route resolution we cannot use that to know which ip version
we are at so save tun_ip_version when parsing the tunnel match and use
that.

Fixes: a508728a4c8b ("net/mlx5e: VF tunnel RX traffic offloading")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Dmytro Linkin <dlinkin@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5: Fix turn-off PPS command
Aya Levin [Mon, 1 Mar 2021 15:54:21 +0000 (17:54 +0200)]
net/mlx5: Fix turn-off PPS command

Fix a bug of uninitialized pin index when trying to turn off PPS out.

Fixes: de19cd6cc977 ("net/mlx5: Move some PPS logic into helper functions")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Don't match on Geneve options in case option masks are all zero
Maor Dickman [Tue, 16 Feb 2021 11:39:18 +0000 (13:39 +0200)]
net/mlx5e: Don't match on Geneve options in case option masks are all zero

The cited change added offload support for Geneve options without verifying
the validity of the options masks, this caused offload of rules with match
on Geneve options with class,type and data masks which are zero to fail.

Fix by ignoring the match on Geneve options in case option masks are
all zero.

Fixes: 9272e3df3023 ("net/mlx5e: Geneve, Add support for encap/decap flows offload")
Signed-off-by: Maor Dickman <maord@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Revert parameters on errors when changing PTP state without reset
Maxim Mikityanskiy [Fri, 29 Jan 2021 12:04:34 +0000 (14:04 +0200)]
net/mlx5e: Revert parameters on errors when changing PTP state without reset

Port timestamping for PTP can be enabled/disabled while the channels are
closed. In that case mlx5e_safe_switch_channels is skipped, and the
preactivate hook is called directly. However, if that hook returns an
error, the channel parameters must be reverted back to their old values.
This commit adds missing handling on this case.

Fixes: 145e5637d941 ("net/mlx5e: Add TX PTP port object support")
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: When changing XDP program without reset, take refs for XSK RQs
Maxim Mikityanskiy [Thu, 11 Feb 2021 13:51:11 +0000 (15:51 +0200)]
net/mlx5e: When changing XDP program without reset, take refs for XSK RQs

Each RQ (including XSK RQs) takes a reference to the XDP program. When
an XDP program is attached or detached, the channels and queues are
recreated, however, there is a special flow for changing an active XDP
program to another one. In that flow, channels and queues stay alive,
but the refcounts of the old and new XDP programs are adjusted. This
flow didn't increment refcount by the number of active XSK RQs, and this
commit fixes it.

Fixes: db05815b36cb ("net/mlx5e: Add XSK zero-copy support")
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Set PTP channel pointer explicitly to NULL
Aya Levin [Tue, 12 Jan 2021 11:34:29 +0000 (13:34 +0200)]
net/mlx5e: Set PTP channel pointer explicitly to NULL

When closing the PTP channel, set its pointer explicitly to NULL. PTP
channel is opened on demand, the code verify the pointer validity before
access. Nullify it when closing the PTP channel to avoid unexpected
behavior.

Fixes: 145e5637d941 ("net/mlx5e: Add TX PTP port object support")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Accumulate port PTP TX stats with other channels stats
Aya Levin [Sun, 28 Feb 2021 16:40:18 +0000 (18:40 +0200)]
net/mlx5e: Accumulate port PTP TX stats with other channels stats

In addition to .get_ethtool_stats, add port PTP TX stats to
.ndo_get_stats64.

Fixes: 145e5637d941 ("net/mlx5e: Add TX PTP port object support")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: RX, Mind the MPWQE gaps when calculating offsets
Tariq Toukan [Tue, 12 Jan 2021 11:21:17 +0000 (13:21 +0200)]
net/mlx5e: RX, Mind the MPWQE gaps when calculating offsets

Since cited patch, MLX5E_REQUIRED_WQE_MTTS is not a power of two.
Hence, usage of MLX5E_LOG_ALIGNED_MPWQE_PPW should be replaced,
as it lost some accuracy. Use the designated macro to calculate
the number of required MTTs.

This makes sure the solution in cited patch works properly.

While here, un-inline mlx5e_get_mpwqe_offset(), and remove the
unused RQ parameter.

Fixes: c3c9402373fe ("net/mlx5e: Add resiliency in Striding RQ mode for packets larger than MTU")
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agonet/mlx5e: Enforce minimum value check for ICOSQ size
Tariq Toukan [Tue, 12 Jan 2021 11:29:14 +0000 (13:29 +0200)]
net/mlx5e: Enforce minimum value check for ICOSQ size

The ICOSQ size should not go below MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE.
Enforce this where it's missing.

Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
3 years agoMerge git://git.kernel.org:/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Wed, 10 Mar 2021 01:15:56 +0000 (17:15 -0800)]
Merge git://git.kernel.org:/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from David Miller:

 1) Fix transmissions in dynamic SMPS mode in ath9k, from Felix Fietkau.

 2) TX skb error handling fix in mt76 driver, also from Felix.

 3) Fix BPF_FETCH atomic in x86 JIT, from Brendan Jackman.

 4) Avoid double free of percpu pointers when freeing a cloned bpf prog.
    From Cong Wang.

 5) Use correct printf format for dma_addr_t in ath11k, from Geert
    Uytterhoeven.

 6) Fix resolve_btfids build with older toolchains, from Kun-Chuan
    Hsieh.

 7) Don't report truncated frames to mac80211 in mt76 driver, from
    Lorenzop Bianconi.

 8) Fix watcdog timeout on suspend/resume of stmmac, from Joakim Zhang.

 9) mscc ocelot needs NET_DEVLINK selct in Kconfig, from Arnd Bergmann.

10) Fix sign comparison bug in TCP_ZEROCOPY_RECEIVE getsockopt(), from
    Arjun Roy.

11) Ignore routes with deleted nexthop object in mlxsw, from Ido
    Schimmel.

12) Need to undo tcp early demux lookup sometimes in nf_nat, from
    Florian Westphal.

13) Fix gro aggregation for udp encaps with zero csum, from Daniel
    Borkmann.

14) Make sure to always use imp*_ndo_send when necessaey, from Jason A.
    Donenfeld.

15) Fix TRSCER masks in sh_eth driver from Sergey Shtylyov.

16) prevent overly huge skb allocationsd in qrtr, from Pavel Skripkin.

17) Prevent rx ring copnsumer index loss of sync in enetc, from Vladimir
    Oltean.

18) Make sure textsearch copntrol block is large enough, from Wilem de
    Bruijn.

19) Revert MAC changes to r8152 leading to instability, from Hates Wang.

20) Advance iov in 9p even for empty reads, from Jissheng Zhang.

21) Double hook unregister in nftables, from PabloNeira Ayuso.

22) Fix memleak in ixgbe, fropm Dinghao Liu.

23) Avoid dups in pkt scheduler class dumps, from Maximilian Heyne.

24) Various mptcp fixes from Florian Westphal, Paolo Abeni, and Geliang
    Tang.

25) Fix DOI refcount bugs in cipso, from Paul Moore.

26) One too many irqsave in ibmvnic, from Junlin Yang.

27) Fix infinite loop with MPLS gso segmenting via virtio_net, from
    Balazs Nemeth.

* git://git.kernel.org:/pub/scm/linux/kernel/git/netdev/net: (164 commits)
  s390/qeth: fix notification for pending buffers during teardown
  s390/qeth: schedule TX NAPI on QAOB completion
  s390/qeth: improve completion of pending TX buffers
  s390/qeth: fix memory leak after failed TX Buffer allocation
  net: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0
  net: check if protocol extracted by virtio_net_hdr_set_proto is correct
  net: dsa: xrs700x: check if partner is same as port in hsr join
  net: lapbether: Remove netif_start_queue / netif_stop_queue
  atm: idt77252: fix null-ptr-dereference
  atm: uPD98402: fix incorrect allocation
  atm: fix a typo in the struct description
  net: qrtr: fix error return code of qrtr_sendmsg()
  mptcp: fix length of ADD_ADDR with port sub-option
  net: bonding: fix error return code of bond_neigh_init()
  net: enetc: allow hardware timestamping on TX queues with tc-etf enabled
  net: enetc: set MAC RX FIFO to recommended value
  net: davicom: Use platform_get_irq_optional()
  net: davicom: Fix regulator not turned off on driver removal
  net: davicom: Fix regulator not turned off on failed probe
  net: dsa: fix switchdev objects on bridge master mistakenly being applied on ports
  ...

3 years agoMerge git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc
Linus Torvalds [Wed, 10 Mar 2021 01:08:41 +0000 (17:08 -0800)]
Merge git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc

Pull sparc fixes from David Miller:
 "Fix opcode filtering for exceptions, and clean up defconfig"

* git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc:
  sparc: sparc64_defconfig: remove duplicate CONFIGs
  sparc64: Fix opcode filtering in handling of no fault loads

3 years agosparc: sparc64_defconfig: remove duplicate CONFIGs
Corentin Labbe [Mon, 8 Mar 2021 09:51:26 +0000 (09:51 +0000)]
sparc: sparc64_defconfig: remove duplicate CONFIGs

After my patch there is CONFIG_ATA defined twice.
Remove the duplicate one.
Same problem for CONFIG_HAPPYMEAL, except I added as builtin for boot
test with NFS.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Fixes: a57cdeb369ef ("sparc: sparc64_defconfig: add necessary configs for qemu")
Signed-off-by: Corentin Labbe <clabbe@baylibre.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agosparc64: Fix opcode filtering in handling of no fault loads
Rob Gardner [Mon, 1 Mar 2021 05:48:16 +0000 (22:48 -0700)]
sparc64: Fix opcode filtering in handling of no fault loads

is_no_fault_exception() has two bugs which were discovered via random
opcode testing with stress-ng. Both are caused by improper filtering
of opcodes.

The first bug can be triggered by a floating point store with a no-fault
ASI, for instance "sta %f0, [%g0] #ASI_PNF", opcode C1A01040.

The code first tests op3[5] (0x1000000), which denotes a floating
point instruction, and then tests op3[2] (0x200000), which denotes a
store instruction. But these bits are not mutually exclusive, and the
above mentioned opcode has both bits set. The intent is to filter out
stores, so the test for stores must be done first in order to have
any effect.

The second bug can be triggered by a floating point load with one of
the invalid ASI values 0x8e or 0x8f, which pass this check in
is_no_fault_exception():
     if ((asi & 0xf2) == ASI_PNF)

An example instruction is "ldqa [%l7 + %o7] #ASI 0x8f, %f38",
opcode CF95D1EF. Asi values greater than 0x8b (ASI_SNFL) are fatal
in handle_ldf_stq(), and is_no_fault_exception() must not allow these
invalid asi values to make it that far.

In both of these cases, handle_ldf_stq() reacts by calling
sun4v_data_access_exception() or spitfire_data_access_exception(),
which call is_no_fault_exception() and results in an infinite
recursion.

Signed-off-by: Rob Gardner <rob.gardner@oracle.com>
Tested-by: Anatoly Pugachev <matorola@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch 's390-qeth-fixes'
David S. Miller [Wed, 10 Mar 2021 00:14:54 +0000 (16:14 -0800)]
Merge branch 's390-qeth-fixes'

Julian Wiedmann says:

====================
s390/qeth: fixes 2021-03-09

please apply the following patch series to netdev's net tree.

This brings one fix for a memleak in an error path of the setup code.
Also several fixes for dealing with pending TX buffers - two for old
bugs in their completion handling, and one recent regression in a
teardown path.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agos390/qeth: fix notification for pending buffers during teardown
Julian Wiedmann [Tue, 9 Mar 2021 16:52:21 +0000 (17:52 +0100)]
s390/qeth: fix notification for pending buffers during teardown

The cited commit reworked the state machine for pending TX buffers.
In qeth_iqd_tx_complete() it turned PENDING into a transient state, and
uses NEED_QAOB for buffers that get parked while waiting for their QAOB
completion.

But it missed to adjust the check in qeth_tx_complete_buf(). So if
qeth_tx_complete_pending_bufs() is called during teardown to drain
the parked TX buffers, we no longer raise a notification for af_iucv.

Instead of updating the checked state, just move this code into
qeth_tx_complete_pending_bufs() itself. This also gets rid of the
special-case in the common TX completion path.

Fixes: 8908f36d20d8 ("s390/qeth: fix af_iucv notification race")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agos390/qeth: schedule TX NAPI on QAOB completion
Julian Wiedmann [Tue, 9 Mar 2021 16:52:20 +0000 (17:52 +0100)]
s390/qeth: schedule TX NAPI on QAOB completion

When a QAOB notifies us that a pending TX buffer has been delivered, the
actual TX completion processing by qeth_tx_complete_pending_bufs()
is done within the context of a TX NAPI instance. We shouldn't rely on
this instance being scheduled by some other TX event, but just do it
ourselves.

qeth_qdio_handle_aob() is called from qeth_poll(), ie. our main NAPI
instance. To avoid touching the TX queue's NAPI instance
before/after it is (un-)registered, reorder the code in qeth_open()
and qeth_stop() accordingly.

Fixes: 0da9581ddb0f ("qeth: exploit asynchronous delivery of storage blocks")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agos390/qeth: improve completion of pending TX buffers
Julian Wiedmann [Tue, 9 Mar 2021 16:52:19 +0000 (17:52 +0100)]
s390/qeth: improve completion of pending TX buffers

The current design attaches a pending TX buffer to a custom
single-linked list, which is anchored at the buffer's slot on the
TX ring. The buffer is then checked for final completion whenever
this slot is processed during a subsequent TX NAPI poll cycle.

But if there's insufficient traffic on the ring, we might never make
enough progress to get back to this ring slot and discover the pending
buffer's final TX completion. In particular if this missing TX
completion blocks the application from sending further traffic.

So convert the custom single-linked list code to a per-queue list_head,
and scan this list on every TX NAPI cycle.

Fixes: 0da9581ddb0f ("qeth: exploit asynchronous delivery of storage blocks")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agos390/qeth: fix memory leak after failed TX Buffer allocation
Julian Wiedmann [Tue, 9 Mar 2021 16:52:18 +0000 (17:52 +0100)]
s390/qeth: fix memory leak after failed TX Buffer allocation

When qeth_alloc_qdio_queues() fails to allocate one of the buffers that
back an Output Queue, the 'out_freeoutqbufs' path will free all
previously allocated buffers for this queue. But it misses to free the
half-finished queue struct itself.

Move the buffer allocation into qeth_alloc_output_queue(), and deal with
such errors internally.

Fixes: 0da9581ddb0f ("qeth: exploit asynchronous delivery of storage blocks")
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com>
Reviewed-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch 'virtio_net-infinite-loop'
David S. Miller [Wed, 10 Mar 2021 00:12:20 +0000 (16:12 -0800)]
Merge branch 'virtio_net-infinite-loop'

Balazs Nemeth says:

====================
net: prevent infinite loop caused by incorrect proto from virtio_net_hdr_set_proto

These patches prevent an infinite loop for gso packets with a protocol
from virtio net hdr that doesn't match the protocol in the packet.
Note that packets coming from a device without
header_ops->parse_protocol being implemented will not be caught by
the check in virtio_net_hdr_to_skb, but the infinite loop will still
be prevented by the check in the gso layer.

Changes from v2 to v3:
  - Remove unused *eth.
  - Use MPLS_HLEN to also check if the MPLS header length is a multiple
    of four.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0
Balazs Nemeth [Tue, 9 Mar 2021 11:31:01 +0000 (12:31 +0100)]
net: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0

A packet with skb_inner_network_header(skb) == skb_network_header(skb)
and ETH_P_MPLS_UC will prevent mpls_gso_segment from pulling any headers
from the packet. Subsequently, the call to skb_mac_gso_segment will
again call mpls_gso_segment with the same packet leading to an infinite
loop. In addition, ensure that the header length is a multiple of four,
which should hold irrespective of the number of stacked labels.

Signed-off-by: Balazs Nemeth <bnemeth@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: check if protocol extracted by virtio_net_hdr_set_proto is correct
Balazs Nemeth [Tue, 9 Mar 2021 11:31:00 +0000 (12:31 +0100)]
net: check if protocol extracted by virtio_net_hdr_set_proto is correct

For gso packets, virtio_net_hdr_set_proto sets the protocol (if it isn't
set) based on the type in the virtio net hdr, but the skb could contain
anything since it could come from packet_snd through a raw socket. If
there is a mismatch between what virtio_net_hdr_set_proto sets and
the actual protocol, then the skb could be handled incorrectly later
on.

An example where this poses an issue is with the subsequent call to
skb_flow_dissect_flow_keys_basic which relies on skb->protocol being set
correctly. A specially crafted packet could fool
skb_flow_dissect_flow_keys_basic preventing EINVAL to be returned.

Avoid blindly trusting the information provided by the virtio net header
by checking that the protocol in the packet actually matches the
protocol set by virtio_net_hdr_set_proto. Note that since the protocol
is only checked if skb->dev implements header_ops->parse_protocol,
packets from devices without the implementation are not checked at this
stage.

Fixes: 9274124f023b ("net: stricter validation of untrusted gso packets")
Signed-off-by: Balazs Nemeth <bnemeth@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: dsa: xrs700x: check if partner is same as port in hsr join
George McCollister [Mon, 8 Mar 2021 23:38:22 +0000 (17:38 -0600)]
net: dsa: xrs700x: check if partner is same as port in hsr join

Don't assign dp to partner if it's the same port that xrs700x_hsr_join
was called with. The partner port is supposed to be the other port in
the HSR/PRP redundant pair not the same port. This fixes an issue
observed in testing where forwarding between redundant HSR ports on this
switch didn't work depending on the order the ports were added to the
hsr device.

Fixes: bd62e6f5e6a9 ("net: dsa: xrs700x: add HSR offloading support")
Signed-off-by: George McCollister <george.mccollister@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agobpf, x86: Use kvmalloc_array instead kmalloc_array in bpf_jit_comp
Yonghong Song [Tue, 9 Mar 2021 01:56:47 +0000 (17:56 -0800)]
bpf, x86: Use kvmalloc_array instead kmalloc_array in bpf_jit_comp

x86 bpf_jit_comp.c used kmalloc_array to store jited addresses
for each bpf insn. With a large bpf program, we have see the
following allocation failures in our production server:

    page allocation failure: order:5, mode:0x40cc0(GFP_KERNEL|__GFP_COMP),
                             nodemask=(null),cpuset=/,mems_allowed=0"
    Call Trace:
    dump_stack+0x50/0x70
    warn_alloc.cold.120+0x72/0xd2
    ? __alloc_pages_direct_compact+0x157/0x160
    __alloc_pages_slowpath+0xcdb/0xd00
    ? get_page_from_freelist+0xe44/0x1600
    ? vunmap_page_range+0x1ba/0x340
    __alloc_pages_nodemask+0x2c9/0x320
    kmalloc_order+0x18/0x80
    kmalloc_order_trace+0x1d/0xa0
    bpf_int_jit_compile+0x1e2/0x484
    ? kmalloc_order_trace+0x1d/0xa0
    bpf_prog_select_runtime+0xc3/0x150
    bpf_prog_load+0x480/0x720
    ? __mod_memcg_lruvec_state+0x21/0x100
    __do_sys_bpf+0xc31/0x2040
    ? close_pdeo+0x86/0xe0
    do_syscall_64+0x42/0x110
    entry_SYSCALL_64_after_hwframe+0x44/0xa9
    RIP: 0033:0x7f2f300f7fa9
    Code: Bad RIP value.

Dumped assembly:

    ffffffff810b6d70 <bpf_int_jit_compile>:
    ; {
    ffffffff810b6d70: e8 eb a5 b4 00        callq   0xffffffff81c01360 <__fentry__>
    ffffffff810b6d75: 41 57                 pushq   %r15
    ...
    ffffffff810b6f39: e9 72 fe ff ff        jmp     0xffffffff810b6db0 <bpf_int_jit_compile+0x40>
    ;       addrs = kmalloc_array(prog->len + 1, sizeof(*addrs), GFP_KERNEL);
    ffffffff810b6f3e: 8b 45 0c              movl    12(%rbp), %eax
    ;       return __kmalloc(bytes, flags);
    ffffffff810b6f41: be c0 0c 00 00        movl    $3264, %esi
    ;       addrs = kmalloc_array(prog->len + 1, sizeof(*addrs), GFP_KERNEL);
    ffffffff810b6f46: 8d 78 01              leal    1(%rax), %edi
    ;       if (unlikely(check_mul_overflow(n, size, &bytes)))
    ffffffff810b6f49: 48 c1 e7 02           shlq    $2, %rdi
    ;       return __kmalloc(bytes, flags);
    ffffffff810b6f4d: e8 8e 0c 1d 00        callq   0xffffffff81287be0 <__kmalloc>
    ;       if (!addrs) {
    ffffffff810b6f52: 48 85 c0              testq   %rax, %rax

Change kmalloc_array() to kvmalloc_array() to avoid potential
allocation error for big bpf programs.

Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20210309015647.3657852-1-yhs@fb.com
3 years agobpf: Don't do bpf_cgroup_storage_set() for kuprobe/tp programs
Yonghong Song [Tue, 9 Mar 2021 18:50:28 +0000 (10:50 -0800)]
bpf: Don't do bpf_cgroup_storage_set() for kuprobe/tp programs

For kuprobe and tracepoint bpf programs, kernel calls
trace_call_bpf() which calls BPF_PROG_RUN_ARRAY_CHECK()
to run the program array. Currently, BPF_PROG_RUN_ARRAY_CHECK()
also calls bpf_cgroup_storage_set() to set percpu
cgroup local storage with NULL value. This is
due to Commit 394e40a29788 ("bpf: extend bpf_prog_array to store
pointers to the cgroup storage") which modified
__BPF_PROG_RUN_ARRAY() to call bpf_cgroup_storage_set()
and this macro is also used by BPF_PROG_RUN_ARRAY_CHECK().

kuprobe and tracepoint programs are not allowed to call
bpf_get_local_storage() helper hence does not
access percpu cgroup local storage. Let us
change BPF_PROG_RUN_ARRAY_CHECK() not to
modify percpu cgroup local storage.

The issue is observed when I tried to debug [1] where
percpu data is overwritten due to
  preempt_disable -> migration_disable
change. This patch does not completely fix the above issue,
which will be addressed separately, e.g., multiple cgroup
prog runs may preempt each other. But it does fix
any potential issue caused by tracing program
overwriting percpu cgroup storage:
 - in a busy system, a tracing program is to run between
   bpf_cgroup_storage_set() and the cgroup prog run.
 - a kprobe program is triggered by a helper in cgroup prog
   before bpf_get_local_storage() is called.

 [1] https://lore.kernel.org/bpf/CAKH8qBuXCfUz=w8L+Fj74OaUpbosO29niYwTki7e3Ag044_aww@mail.gmail.com/T

Fixes: 394e40a29788 ("bpf: extend bpf_prog_array to store pointers to the cgroup storage")
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Roman Gushchin <guro@fb.com>
Link: https://lore.kernel.org/bpf/20210309185028.3763817-1-yhs@fb.com
3 years agoMerge tag 'gpio-fixes-for-v5.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 9 Mar 2021 20:02:18 +0000 (12:02 -0800)]
Merge tag 'gpio-fixes-for-v5.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux

Pull gpio fixes from Bartosz Golaszewski:
 "A bunch of fixes for the GPIO subsystem. We have two regressions in
  the core code spotted right after the merge window, a series of fixes
  for ACPI GPIO and a subsequent fix for a related regression in
  gpio-pca953x + a minor tweak in .gitignore and a rework of handling of
  the gpio-line-names to remedy a regression in stm32mp151.

  Summary:

   - fix two regressions in core GPIO subsystem code: one NULL-pointer
     dereference and one list corruption

   - read GPIO line names from fwnode instead of using the generic
     device properties to fix a regression on stm32mp151

   - fixes to ACPI GPIO and gpio-pca953x to handle a regression in IRQ
     handling on Intel Galileo

   - update .gitignore in GPIO selftests"

* tag 'gpio-fixes-for-v5.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpiolib: Read "gpio-line-names" from a firmware node
  gpio: pca953x: Set IRQ type when handle Intel Galileo Gen 2
  gpiolib: acpi: Allow to find GpioInt() resource by name and index
  gpiolib: acpi: Add ACPI_GPIO_QUIRK_ABSOLUTE_NUMBER quirk
  gpiolib: acpi: Add missing IRQF_ONESHOT
  gpio: fix gpio-device list corruption
  gpio: fix NULL-deref-on-deregistration regression
  selftests: gpio: update .gitignore

3 years agoMerge tag 'mips-fixes_5.12_1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips...
Linus Torvalds [Tue, 9 Mar 2021 19:56:41 +0000 (11:56 -0800)]
Merge tag 'mips-fixes_5.12_1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux

Pull MIPS fixes from Thomas Bogendoerfer:

 - fixes for boot breakage because of misaligned FDTs

 - fix for overwritten exception handlers

 - enable MIPS optimized crypto for all MIPS CPUs to improve wireguard
   performance

* tag 'mips-fixes_5.12_1' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:
  MIPS: kernel: Reserve exception base early to prevent corruption
  MIPS: vmlinux.lds.S: align raw appended dtb to 8 bytes
  crypto: mips/poly1305 - enable for all MIPS processors
  MIPS: boot/compressed: Copy DTB to aligned address

3 years agonet: lapbether: Remove netif_start_queue / netif_stop_queue
Xie He [Sun, 7 Mar 2021 11:33:07 +0000 (03:33 -0800)]
net: lapbether: Remove netif_start_queue / netif_stop_queue

For the devices in this driver, the default qdisc is "noqueue",
because their "tx_queue_len" is 0.

In function "__dev_queue_xmit" in "net/core/dev.c", devices with the
"noqueue" qdisc are specially handled. Packets are transmitted without
being queued after a "dev->flags & IFF_UP" check. However, it's possible
that even if this check succeeds, "ops->ndo_stop" may still have already
been called. This is because in "__dev_close_many", "ops->ndo_stop" is
called before clearing the "IFF_UP" flag.

If we call "netif_stop_queue" in "ops->ndo_stop", then it's possible in
"__dev_queue_xmit", it sees the "IFF_UP" flag is present, and then it
checks "netif_xmit_stopped" and finds that the queue is already stopped.
In this case, it will complain that:
"Virtual device ... asks to queue packet!"

To prevent "__dev_queue_xmit" from generating this complaint, we should
not call "netif_stop_queue" in "ops->ndo_stop".

We also don't need to call "netif_start_queue" in "ops->ndo_open",
because after a netdev is allocated and registered, the
"__QUEUE_STATE_DRV_XOFF" flag is initially not set, so there is no need
to call "netif_start_queue" to clear it.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Xie He <xie.he.0141@gmail.com>
Acked-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMIPS: kernel: Reserve exception base early to prevent corruption
Thomas Bogendoerfer [Mon, 8 Mar 2021 09:24:47 +0000 (10:24 +0100)]
MIPS: kernel: Reserve exception base early to prevent corruption

BMIPS is one of the few platforms that do change the exception base.
After commit 2dcb39645441 ("memblock: do not start bottom-up allocations
with kernel_end") we started seeing BMIPS boards fail to boot with the
built-in FDT being corrupted.

Before the cited commit, early allocations would be in the [kernel_end,
RAM_END] range, but after commit they would be within [RAM_START +
PAGE_SIZE, RAM_END].

The custom exception base handler that is installed by
bmips_ebase_setup() done for BMIPS5000 CPUs ends-up trampling on the
memory region allocated by unflatten_and_copy_device_tree() thus
corrupting the FDT used by the kernel.

To fix this, we need to perform an early reservation of the custom
exception space. Additional we reserve the first 4k (1k for R3k) for
either normal exception vector space (legacy CPUs) or special vectors
like cache exceptions.

Huge thanks to Serge for analysing and proposing a solution to this
issue.

Fixes: 2dcb39645441 ("memblock: do not start bottom-up allocations with kernel_end")
Reported-by: Kamal Dasu <kdasu.kdev@gmail.com>
Debugged-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
3 years agoMerge git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc
Linus Torvalds [Tue, 9 Mar 2021 06:01:58 +0000 (22:01 -0800)]
Merge git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc

Pull sparc updates from David Miller:
 "Just some more random bits from Al, including a conversion over to
  generic extables"

* git://git.kernel.org:/pub/scm/linux/kernel/git/davem/sparc:
  sparc32: take ->thread.flags out
  sparc32: get rid of fake_swapper_regs
  sparc64: get rid of fake_swapper_regs
  sparc32: switch to generic extables
  sparc32: switch copy_user.S away from range exception table entries
  sparc32: get rid of range exception table entries in checksum_32.S
  sparc32: switch __bzero() away from range exception table entries
  sparc32: kill lookup_fault()
  sparc32: don't bother with lookup_fault() in __bzero()

3 years agoatm: idt77252: fix null-ptr-dereference
Tong Zhang [Mon, 8 Mar 2021 03:25:30 +0000 (22:25 -0500)]
atm: idt77252: fix null-ptr-dereference

this one is similar to the phy_data allocation fix in uPD98402, the
driver allocate the idt77105_priv and store to dev_data but later
dereference using dev->dev_data, which will cause null-ptr-dereference.

fix this issue by changing dev_data to phy_data so that PRIV(dev) can
work correctly.

Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoatm: uPD98402: fix incorrect allocation
Tong Zhang [Mon, 8 Mar 2021 03:25:29 +0000 (22:25 -0500)]
atm: uPD98402: fix incorrect allocation

dev->dev_data is set in zatm.c, calling zatm_start() will overwrite this
dev->dev_data in uPD98402_start() and a subsequent PRIV(dev)->lock
(i.e dev->phy_data->lock) will result in a null-ptr-dereference.

I believe this is a typo and what it actually want to do is to allocate
phy_data instead of dev_data.

Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoatm: fix a typo in the struct description
Tong Zhang [Mon, 8 Mar 2021 03:25:28 +0000 (22:25 -0500)]
atm: fix a typo in the struct description

phy_data means private PHY data not date

Signed-off-by: Tong Zhang <ztong0001@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: qrtr: fix error return code of qrtr_sendmsg()
Jia-Ju Bai [Mon, 8 Mar 2021 09:13:55 +0000 (01:13 -0800)]
net: qrtr: fix error return code of qrtr_sendmsg()

When sock_alloc_send_skb() returns NULL to skb, no error return code of
qrtr_sendmsg() is assigned.
To fix this bug, rc is assigned with -ENOMEM in this case.

Fixes: 194ccc88297a ("net: qrtr: Support decoding incoming v2 packets")
Reported-by: TOTE Robot <oslab@tsinghua.edu.cn>
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agomptcp: fix length of ADD_ADDR with port sub-option
Davide Caratti [Mon, 8 Mar 2021 09:00:04 +0000 (10:00 +0100)]
mptcp: fix length of ADD_ADDR with port sub-option

in current Linux, MPTCP peers advertising endpoints with port numbers use
a sub-option length that wrongly accounts for the trailing TCP NOP. Also,
receivers will only process incoming ADD_ADDR with port having such wrong
sub-option length. Fix this, making ADD_ADDR compliant to RFC8684 ยง3.4.1.

this can be verified running tcpdump on the kselftests artifacts:

 unpatched kernel:
 [root@bottarga mptcp]# tcpdump -tnnr unpatched.pcap | grep add-addr
 reading from file unpatched.pcap, link-type LINUX_SLL (Linux cooked v1), snapshot length 65535
 IP 10.0.1.1.10000 > 10.0.1.2.53078: Flags [.], ack 101, win 509, options [nop,nop,TS val 214459678 ecr 521312851,mptcp add-addr v1 id 1 a00:201:2774:2d88:7436:85c3:17fd:101], length 0
 IP 10.0.1.2.53078 > 10.0.1.1.10000: Flags [.], ack 101, win 502, options [nop,nop,TS val 521312852 ecr 214459678,mptcp add-addr[bad opt]]

 patched kernel:
 [root@bottarga mptcp]# tcpdump -tnnr patched.pcap | grep add-addr
 reading from file patched.pcap, link-type LINUX_SLL (Linux cooked v1), snapshot length 65535
 IP 10.0.1.1.10000 > 10.0.1.2.38178: Flags [.], ack 101, win 509, options [nop,nop,TS val 3728873902 ecr 2732713192,mptcp add-addr v1 id 1 10.0.2.1:10100 hmac 0xbccdfcbe59292a1f,nop,nop], length 0
 IP 10.0.1.2.38178 > 10.0.1.1.10000: Flags [.], ack 101, win 502, options [nop,nop,TS val 2732713195 ecr 3728873902,mptcp add-addr v1-echo id 1 10.0.2.1:10100,nop,nop], length 0

Fixes: 22fb85ffaefb ("mptcp: add port support for ADD_ADDR suboption writing")
CC: stable@vger.kernel.org # 5.11+
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Acked-and-tested-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoselftests/bpf: Tests using bpf_check_mtu BPF-helper input mtu_len param
Jesper Dangaard Brouer [Mon, 8 Mar 2021 14:59:23 +0000 (15:59 +0100)]
selftests/bpf: Tests using bpf_check_mtu BPF-helper input mtu_len param

Add tests that use mtu_len as input parameter in BPF-helper
bpf_check_mtu().

The BPF-helper is avail from both XDP and TC context. Add two tests
per context, one that tests below MTU and one that exceeds the MTU.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/161521556358.3515614.5915221479709358964.stgit@firesoul
3 years agobpf: BPF-helper for MTU checking add length input
Jesper Dangaard Brouer [Mon, 8 Mar 2021 14:59:18 +0000 (15:59 +0100)]
bpf: BPF-helper for MTU checking add length input

The FIB lookup example[1] show how the IP-header field tot_len
(iph->tot_len) is used as input to perform the MTU check.

This patch extend the BPF-helper bpf_check_mtu() with the same ability
to provide the length as user parameter input, via mtu_len parameter.

This still needs to be done before the bpf_check_mtu() helper API
becomes frozen.

  [1] samples/bpf/xdp_fwd_kern.c

Fixes: 34b2021cc616 ("bpf: Add BPF-helper for MTU checking")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/161521555850.3515614.6533850861569774444.stgit@firesoul
3 years agonet: bonding: fix error return code of bond_neigh_init()
Jia-Ju Bai [Mon, 8 Mar 2021 03:11:02 +0000 (19:11 -0800)]
net: bonding: fix error return code of bond_neigh_init()

When slave is NULL or slave_ops->ndo_neigh_setup is NULL, no error
return code of bond_neigh_init() is assigned.
To fix this bug, ret is assigned with -EINVAL in these cases.

Fixes: 9e99bfefdbce ("bonding: fix bond_neigh_init()")
Reported-by: TOTE Robot <oslab@tsinghua.edu.cn>
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: enetc: allow hardware timestamping on TX queues with tc-etf enabled
Vladimir Oltean [Sun, 7 Mar 2021 13:23:39 +0000 (15:23 +0200)]
net: enetc: allow hardware timestamping on TX queues with tc-etf enabled

The txtime is passed to the driver in skb->skb_mstamp_ns, which is
actually in a union with skb->tstamp (the place where software
timestamps are kept).

Since commit b50a5c70ffa4 ("net: allow simultaneous SW and HW transmit
timestamping"), __sock_recv_timestamp has some logic for making sure
that the two calls to skb_tstamp_tx:

skb_tx_timestamp(skb) # Software timestamp in the driver
-> skb_tstamp_tx(skb, NULL)

and

skb_tstamp_tx(skb, &shhwtstamps) # Hardware timestamp in the driver

will both do the right thing and in a race-free manner, meaning that
skb_tx_timestamp will deliver a cmsg with the software timestamp only,
and skb_tstamp_tx with a non-NULL hwtstamps argument will deliver a cmsg
with the hardware timestamp only.

Why are races even possible? Well, because although the software timestamp
skb->tstamp is private per skb, the hardware timestamp skb_hwtstamps(skb)
lives in skb_shinfo(skb), an area which is shared between skbs and their
clones. And skb_tstamp_tx works by cloning the packets when timestamping
them, therefore attempting to perform hardware timestamping on an skb's
clone will also change the hardware timestamp of the original skb. And
the original skb might have been yet again cloned for software
timestamping, at an earlier stage.

So the logic in __sock_recv_timestamp can't be as simple as saying
"does this skb have a hardware timestamp? if yes I'll send the hardware
timestamp to the socket, otherwise I'll send the software timestamp",
precisely because the hardware timestamp is shared.
Instead, it's quite the other way around: __sock_recv_timestamp says
"does this skb have a software timestamp? if yes, I'll send the software
timestamp, otherwise the hardware one". This works because the software
timestamp is not shared with clones.

But that means we have a problem when we attempt hardware timestamping
with skbs that don't have the skb->tstamp == 0. __sock_recv_timestamp
will say "oh, yeah, this must be some sort of odd clone" and will not
deliver the hardware timestamp to the socket. And this is exactly what
is happening when we have txtime enabled on the socket: as mentioned,
that is put in a union with skb->tstamp, so it is quite easy to mistake
it.

Do what other drivers do (intel igb/igc) and write zero to skb->tstamp
before taking the hardware timestamp. It's of no use to us now (we're
already on the TX confirmation path).

Fixes: 0d08c9ec7d6e ("enetc: add support time specific departure base on the qos etf")
Cc: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: enetc: set MAC RX FIFO to recommended value
Alex Marginean [Sun, 7 Mar 2021 13:23:38 +0000 (15:23 +0200)]
net: enetc: set MAC RX FIFO to recommended value

On LS1028A, the MAC RX FIFO defaults to the value 2, which is too high
and may lead to RX lock-up under traffic at a rate higher than 6 Gbps.
Set it to 1 instead, as recommended by the hardware design team and by
later versions of the ENETC block guide.

Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Reviewed-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Reviewed-by: Jason Liu <jason.hui.liu@nxp.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>