OSDN Git Service
David S. Miller [Fri, 20 Dec 2019 01:53:05 +0000 (17:53 -0800)]
Merge branch 'cls_u32-fix-refcount-leak'
Davide Caratti says:
====================
net/sched: cls_u32: fix refcount leak
a refcount leak in the error path of u32_change() has been recently
introduced. It can be observed with the following commands:
[root@f31 ~]# tc filter replace dev eth0 ingress protocol ip prio 97 \
> u32 match ip src 127.0.0.1/32 indev notexist20 flowid 1:1 action drop
RTNETLINK answers: Invalid argument
We have an error talking to the kernel
[root@f31 ~]# tc filter replace dev eth0 ingress protocol ip prio 98 \
> handle 42:42 u32 divisor 256
Error: cls_u32: Divisor can only be used on a hash table.
We have an error talking to the kernel
[root@f31 ~]# tc filter replace dev eth0 ingress protocol ip prio 99 \
> u32 ht 47:47
Error: cls_u32: Specified hash table not found.
We have an error talking to the kernel
they all legitimately return -EINVAL; however, they leave semi-configured
filters at eth0 tc ingress:
[root@f31 ~]# tc filter show dev eth0 ingress
filter protocol ip pref 97 u32 chain 0
filter protocol ip pref 97 u32 chain 0 fh 800: ht divisor 1
filter protocol ip pref 98 u32 chain 0
filter protocol ip pref 98 u32 chain 0 fh 801: ht divisor 1
filter protocol ip pref 99 u32 chain 0
filter protocol ip pref 99 u32 chain 0 fh 802: ht divisor 1
With older kernels, filters were unconditionally considered empty (and
thus de-refcounted) on the error path of ->change().
After commit
8b64678e0af8 ("net: sched: refactor tp insert/delete for
concurrent execution"), filters were considered empty when the walk()
function didn't set 'walker.stop' to 1.
Finally, with commit
6676d5e416ee ("net: sched: set dedicated tcf_walker
flag when tp is empty"), tc filters are considered empty unless the walker
function is called with a non-NULL handle. This last change doesn't fit
cls_u32 design, because at least the "root hnode" is (almost) always
non-NULL, as it's allocated in u32_init().
- patch 1/2 is a proposal to restore the original kernel behavior, where
no filter was installed in the error path of u32_change().
- patch 2/2 adds tdc selftests that can be ued to verify the correct
behavior of u32 in the error path of ->change().
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Davide Caratti [Tue, 17 Dec 2019 23:00:05 +0000 (00:00 +0100)]
tc-testing: initial tdc selftests for cls_u32
- move test "e9a3 - Add u32 with source match" to u32.json, and change the
match pattern to catch all hnodes
- add testcases for relevant error paths of cls_u32 module
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Davide Caratti [Tue, 17 Dec 2019 23:00:04 +0000 (00:00 +0100)]
net/sched: cls_u32: fix refcount leak in the error path of u32_change()
when users replace cls_u32 filters with new ones having wrong parameters,
so that u32_change() fails to validate them, the kernel doesn't roll-back
correctly, and leaves semi-configured rules.
Fix this in u32_walk(), avoiding a call to the walker function on filters
that don't have a match rule connected. The side effect is, these "empty"
filters are not even dumped when present; but that shouldn't be a problem
as long as we are restoring the original behaviour, where semi-configured
filters were not even added in the error path of u32_change().
Fixes:
6676d5e416ee ("net: sched: set dedicated tcf_walker flag when tp is empty")
Signed-off-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Aditya Pakki [Tue, 17 Dec 2019 20:43:00 +0000 (14:43 -0600)]
nfc: s3fwrn5: replace the assertion with a WARN_ON
In s3fwrn5_fw_recv_frame, if fw_info->rsp is not empty, the
current code causes a crash via BUG_ON. However, s3fwrn5_fw_send_msg
does not crash in such a scenario. The patch replaces the BUG_ON
by returning the error to the callers and frees up skb.
Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 20 Dec 2019 01:32:21 +0000 (17:32 -0800)]
Merge branch 'macb-fix-probing-of-PHY-not-described-in-the-dt'
Antoine Tenart says:
====================
net: macb: fix probing of PHY not described in the dt
The macb Ethernet driver supports various ways of referencing its
network PHY. When a device tree is used the PHY can be referenced with
a phy-handle or, if connected to its internal MDIO bus, described in
a child node. Some platforms omitted the PHY description while
connecting the PHY to the internal MDIO bus and in such cases the MDIO
bus has to be scanned "manually" by the macb driver.
Prior to the phylink conversion the driver registered the MDIO bus with
of_mdiobus_register and then in case the PHY couldn't be retrieved
using dt or using phy_find_first (because registering an MDIO bus with
of_mdiobus_register masks all PHYs) the macb driver was "manually"
scanning the MDIO bus (like mdiobus_register does). The phylink
conversion did break this particular case but reimplementing the manual
scan of the bus in the macb driver wouldn't be very clean. The solution
seems to be registering the MDIO bus based on if the PHYs are described
in the device tree or not.
There are multiple ways to do this, none is perfect. I chose to check if
any of the child nodes of the macb node was a network PHY and based on
this to register the MDIO bus with the of_ helper or not. The drawback
is boards referencing the PHY through phy-handle, would scan the entire
MDIO bus of the macb at boot time (as the MDIO bus would be registered
with mdiobus_register). For this solution to work properly
of_mdiobus_child_is_phy has to be exported, which means the patch doing
so has to be backported to -stable as well.
Another possible solution could have been to simply check if the macb
node has a child node by counting its sub-nodes. This isn't techically
perfect, as there could be other sub-nodes (in practice this should be
fine, fixed-link being taken care of in the driver). We could also
simply s/of_mdiobus_register/mdiobus_register/ but that could break
boards using the PHY description in child node as a selector (which
really would be not a proper way to do this...).
The real issue here being having PHYs not described in the dt but we
have dt backward compatibility, so we have to live with that.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Tue, 17 Dec 2019 17:07:42 +0000 (18:07 +0100)]
net: macb: fix probing of PHY not described in the dt
This patch fixes the case where the PHY isn't described in the device
tree. This is due to the way the MDIO bus is registered in the driver:
whether the PHY is described in the device tree or not, the bus is
registered through of_mdiobus_register. The function masks all the PHYs
and only allow probing the ones described in the device tree. Prior to
the Phylink conversion this was also done but later on in the driver
the MDIO bus was manually scanned to circumvent the fact that the PHY
wasn't described.
This patch fixes it in a proper way, by registering the MDIO bus based
on if the PHY attached to a given interface is described in the device
tree or not.
Fixes:
7897b071ac3b ("net: macb: convert to phylink")
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Tue, 17 Dec 2019 17:07:41 +0000 (18:07 +0100)]
of: mdio: export of_mdiobus_child_is_phy
This patch exports of_mdiobus_child_is_phy, allowing to check if a child
node is a network PHY.
Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 19 Dec 2019 22:20:47 +0000 (14:20 -0800)]
Merge git://git./pub/scm/linux/kernel/git/bpf/bpf
Daniel Borkmann says:
====================
pull-request: bpf 2019-12-19
The following pull-request contains BPF updates for your *net* tree.
We've added 10 non-merge commits during the last 8 day(s) which contain
a total of 21 files changed, 269 insertions(+), 108 deletions(-).
The main changes are:
1) Fix lack of synchronization between xsk wakeup and destroying resources
used by xsk wakeup, from Maxim Mikityanskiy.
2) Fix pruning with tail call patching, untrack programs in case of verifier
error and fix a cgroup local storage tracking bug, from Daniel Borkmann.
3) Fix clearing skb->tstamp in bpf_redirect() when going from ingress to
egress which otherwise cause issues e.g. on fq qdisc, from Lorenz Bauer.
4) Fix compile warning of unused proc_dointvec_minmax_bpf_restricted() when
only cBPF is present, from Alexander Lobakin.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Thu, 19 Dec 2019 21:19:51 +0000 (22:19 +0100)]
bpf: Add further test_verifier cases for record_func_key
Expand dummy prog generation such that we can easily check on return
codes and add few more test cases to make sure we keep on tracking
pruning behavior.
# ./test_verifier
[...]
#1066/p XDP pkt read, pkt_data <= pkt_meta', bad access 1 OK
#1067/p XDP pkt read, pkt_data <= pkt_meta', bad access 2 OK
Summary: 1580 PASSED, 0 SKIPPED, 0 FAILED
Also verified that JIT dump of added test cases looks good.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/df7200b6021444fd369376d227de917357285b65.1576789878.git.daniel@iogearbox.net
Daniel Borkmann [Thu, 19 Dec 2019 21:19:50 +0000 (22:19 +0100)]
bpf: Fix record_func_key to perform backtracking on r3
While testing Cilium with /unreleased/ Linus' tree under BPF-based NodePort
implementation, I noticed a strange BPF SNAT engine behavior from time to
time. In some cases it would do the correct SNAT/DNAT service translation,
but at a random point in time it would just stop and perform an unexpected
translation after SYN, SYN/ACK and stack would send a RST back. While initially
assuming that there is some sort of a race condition in BPF code, adding
trace_printk()s for debugging purposes at some point seemed to have resolved
the issue auto-magically.
Digging deeper on this Heisenbug and reducing the trace_printk() calls to
an absolute minimum, it turns out that a single call would suffice to
trigger / not trigger the seen RST issue, even though the logic of the
program itself remains unchanged. Turns out the single call changed verifier
pruning behavior to get everything to work. Reconstructing a minimal test
case, the incorrect JIT dump looked as follows:
# bpftool p d j i 11346
0xffffffffc0cba96c:
[...]
21: movzbq 0x30(%rdi),%rax
26: cmp $0xd,%rax
2a: je 0x000000000000003a
2c: xor %edx,%edx
2e: movabs $0xffff89cc74e85800,%rsi
38: jmp 0x0000000000000049
3a: mov $0x2,%edx
3f: movabs $0xffff89cc74e85800,%rsi
49: mov -0x224(%rbp),%eax
4f: cmp $0x20,%eax
52: ja 0x0000000000000062
54: add $0x1,%eax
57: mov %eax,-0x224(%rbp)
5d: jmpq 0xffffffffffff6911
62: mov $0x1,%eax
[...]
Hence, unexpectedly, JIT emitted a direct jump even though retpoline based
one would have been needed since in line 2c and 3a we have different slot
keys in BPF reg r3. Verifier log of the test case reveals what happened:
0: (b7) r0 = 14
1: (73) *(u8 *)(r1 +48) = r0
2: (71) r0 = *(u8 *)(r1 +48)
3: (15) if r0 == 0xd goto pc+4
R0_w=inv(id=0,umax_value=255,var_off=(0x0; 0xff)) R1=ctx(id=0,off=0,imm=0) R10=fp0
4: (b7) r3 = 0
5: (18) r2 = 0xffff89cc74d54a00
7: (05) goto pc+3
11: (85) call bpf_tail_call#12
12: (b7) r0 = 1
13: (95) exit
from 3 to 8: R0_w=inv13 R1=ctx(id=0,off=0,imm=0) R10=fp0
8: (b7) r3 = 2
9: (18) r2 = 0xffff89cc74d54a00
11: safe
processed 13 insns (limit
1000000) [...]
Second branch is pruned by verifier since considered safe, but issue is that
record_func_key() couldn't have seen the index in line 3a and therefore
decided that emitting a direct jump at this location was okay.
Fix this by reusing our backtracking logic for precise scalar verification
in order to prevent pruning on the slot key. This means verifier will track
content of r3 all the way backwards and only prune if both scalars were
unknown in state equivalence check and therefore poisoned in the first place
in record_func_key(). The range is [x,x] in record_func_key() case since
the slot always would have to be constant immediate. Correct verification
after fix:
0: (b7) r0 = 14
1: (73) *(u8 *)(r1 +48) = r0
2: (71) r0 = *(u8 *)(r1 +48)
3: (15) if r0 == 0xd goto pc+4
R0_w=invP(id=0,umax_value=255,var_off=(0x0; 0xff)) R1=ctx(id=0,off=0,imm=0) R10=fp0
4: (b7) r3 = 0
5: (18) r2 = 0x0
7: (05) goto pc+3
11: (85) call bpf_tail_call#12
12: (b7) r0 = 1
13: (95) exit
from 3 to 8: R0_w=invP13 R1=ctx(id=0,off=0,imm=0) R10=fp0
8: (b7) r3 = 2
9: (18) r2 = 0x0
11: (85) call bpf_tail_call#12
12: (b7) r0 = 1
13: (95) exit
processed 15 insns (limit
1000000) [...]
And correct corresponding JIT dump:
# bpftool p d j i 11
0xffffffffc0dc34c4:
[...]
21: movzbq 0x30(%rdi),%rax
26: cmp $0xd,%rax
2a: je 0x000000000000003a
2c: xor %edx,%edx
2e: movabs $0xffff9928b4c02200,%rsi
38: jmp 0x0000000000000049
3a: mov $0x2,%edx
3f: movabs $0xffff9928b4c02200,%rsi
49: cmp $0x4,%rdx
4d: jae 0x0000000000000093
4f: and $0x3,%edx
52: mov %edx,%edx
54: cmp %edx,0x24(%rsi)
57: jbe 0x0000000000000093
59: mov -0x224(%rbp),%eax
5f: cmp $0x20,%eax
62: ja 0x0000000000000093
64: add $0x1,%eax
67: mov %eax,-0x224(%rbp)
6d: mov 0x110(%rsi,%rdx,8),%rax
75: test %rax,%rax
78: je 0x0000000000000093
7a: mov 0x30(%rax),%rax
7e: add $0x19,%rax
82: callq 0x000000000000008e
87: pause
89: lfence
8c: jmp 0x0000000000000087
8e: mov %rax,(%rsp)
92: retq
93: mov $0x1,%eax
[...]
Also explicitly adding explicit env->allow_ptr_leaks to fixup_bpf_calls() since
backtracking is enabled under former (direct jumps as well, but use different
test). In case of only tracking different map pointers as in
c93552c443eb ("bpf:
properly enforce index mask to prevent out-of-bounds speculation"), pruning
cannot make such short-cuts, neither if there are paths with scalar and non-scalar
types as r3. mark_chain_precision() is only needed after we know that
register_is_const(). If it was not the case, we already poison the key on first
path and non-const key in later paths are not matching the scalar range in regsafe()
either. Cilium NodePort testing passes fine as well now. Note, released kernels
not affected.
Fixes:
d2e4c1e6c294 ("bpf: Constant map key tracking for prog array pokes")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/ac43ffdeb7386c5bd688761ed266f3722bb39823.1576789878.git.daniel@iogearbox.net
Alexander Lobakin [Wed, 18 Dec 2019 09:18:21 +0000 (12:18 +0300)]
net, sysctl: Fix compiler warning when only cBPF is present
proc_dointvec_minmax_bpf_restricted() has been firstly introduced
in commit
2e4a30983b0f ("bpf: restrict access to core bpf sysctls")
under CONFIG_HAVE_EBPF_JIT. Then, this ifdef has been removed in
ede95a63b5e8 ("bpf: add bpf_jit_limit knob to restrict unpriv
allocations"), because a new sysctl, bpf_jit_limit, made use of it.
Finally, this parameter has become long instead of integer with
fdadd04931c2 ("bpf: fix bpf_jit_limit knob for PAGE_SIZE >= 64K")
and thus, a new proc_dolongvec_minmax_bpf_restricted() has been
added.
With this last change, we got back to that
proc_dointvec_minmax_bpf_restricted() is used only under
CONFIG_HAVE_EBPF_JIT, but the corresponding ifdef has not been
brought back.
So, in configurations like CONFIG_BPF_JIT=y && CONFIG_HAVE_EBPF_JIT=n
since v4.20 we have:
CC net/core/sysctl_net_core.o
net/core/sysctl_net_core.c:292:1: warning: ‘proc_dointvec_minmax_bpf_restricted’ defined but not used [-Wunused-function]
292 | proc_dointvec_minmax_bpf_restricted(struct ctl_table *table, int write,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Suppress this by guarding it with CONFIG_HAVE_EBPF_JIT again.
Fixes:
fdadd04931c2 ("bpf: fix bpf_jit_limit knob for PAGE_SIZE >= 64K")
Signed-off-by: Alexander Lobakin <alobakin@dlink.ru>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191218091821.7080-1-alobakin@dlink.ru
Daniel Borkmann [Thu, 19 Dec 2019 15:20:49 +0000 (16:20 +0100)]
Merge branch 'bpf-fix-xsk-wakeup'
Maxim Mikityanskiy says:
====================
This series addresses the issue described in the commit message of the
first patch: lack of synchronization between XSK wakeup and destroying
the resources used by XSK wakeup. The idea is similar to napi_synchronize.
The series contains fixes for the drivers that implement XSK.
v2 incorporates changes suggested by Björn:
1. Call synchronize_rcu in Intel drivers only if the XDP program is
being unloaded.
2. Don't forget rcu_read_lock when wakeup is called from xsk_poll.
3. Use xs->zc as the condition to call ndo_xsk_wakeup.
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Maxim Mikityanskiy [Tue, 17 Dec 2019 16:20:47 +0000 (16:20 +0000)]
net/ixgbe: Fix concurrency issues between config flow and XSK
Use synchronize_rcu to wait until the XSK wakeup function finishes
before destroying the resources it uses:
1. ixgbe_down already calls synchronize_rcu after setting __IXGBE_DOWN.
2. After switching the XDP program, call synchronize_rcu to let
ixgbe_xsk_wakeup exit before the XDP program is freed.
3. Changing the number of channels brings the interface down.
4. Disabling UMEM sets __IXGBE_TX_DISABLED before closing hardware
resources and resetting xsk_umem. Check that bit in ixgbe_xsk_wakeup to
avoid using the XDP ring when it's already destroyed. synchronize_rcu is
called from ixgbe_txrx_ring_disable.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191217162023.16011-5-maximmi@mellanox.com
Maxim Mikityanskiy [Tue, 17 Dec 2019 16:20:45 +0000 (16:20 +0000)]
net/i40e: Fix concurrency issues between config flow and XSK
Use synchronize_rcu to wait until the XSK wakeup function finishes
before destroying the resources it uses:
1. i40e_down already calls synchronize_rcu. On i40e_down either
__I40E_VSI_DOWN or __I40E_CONFIG_BUSY is set. Check the latter in
i40e_xsk_wakeup (the former is already checked there).
2. After switching the XDP program, call synchronize_rcu to let
i40e_xsk_wakeup exit before the XDP program is freed.
3. Changing the number of channels brings the interface down (see
i40e_prep_for_reset and i40e_pf_quiesce_all_vsi).
4. Disabling UMEM sets __I40E_CONFIG_BUSY, too.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191217162023.16011-4-maximmi@mellanox.com
Maxim Mikityanskiy [Tue, 17 Dec 2019 16:20:44 +0000 (16:20 +0000)]
net/mlx5e: Fix concurrency issues between config flow and XSK
After disabling resources necessary for XSK (the XDP program, channels,
XSK queues), use synchronize_rcu to wait until the XSK wakeup function
finishes, before freeing the resources.
Suspend XSK wakeups during switching channels. If the XDP program is
being removed, synchronize_rcu before closing the old channels to allow
XSK wakeup to complete.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191217162023.16011-3-maximmi@mellanox.com
Maxim Mikityanskiy [Tue, 17 Dec 2019 16:20:42 +0000 (16:20 +0000)]
xsk: Add rcu_read_lock around the XSK wakeup
The XSK wakeup callback in drivers makes some sanity checks before
triggering NAPI. However, some configuration changes may occur during
this function that affect the result of those checks. For example, the
interface can go down, and all the resources will be destroyed after the
checks in the wakeup function, but before it attempts to use these
resources. Wrap this callback in rcu_read_lock to allow driver to
synchronize_rcu before actually destroying the resources.
xsk_wakeup is a new function that encapsulates calling ndo_xsk_wakeup
wrapped into the RCU lock. After this commit, xsk_poll starts using
xsk_wakeup and checks xs->zc instead of ndo_xsk_wakeup != NULL to decide
ndo_xsk_wakeup should be called. It also fixes a bug introduced with the
need_wakeup feature: a non-zero-copy socket may be used with a driver
supporting zero-copy, and in this case ndo_xsk_wakeup should not be
called, so the xs->zc check is the correct one.
Fixes:
77cd0d7b3f25 ("xsk: add support for need_wakeup flag in AF_XDP rings")
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191217162023.16011-2-maximmi@mellanox.com
David S. Miller [Wed, 18 Dec 2019 20:04:18 +0000 (12:04 -0800)]
Merge branch 'stmmac-fixes'
Jose Abreu says:
====================
net: stmmac: Fixes for -net
Fixes for stmmac.
1) Fixes the filtering selftests (again) for cases when the number of multicast
filters are not enough.
2) Fixes SPH feature for MTU > default.
3) Fixes the behavior of accepting invalid MTU values.
4) Fixes FCS stripping for multi-descriptor packets.
5) Fixes the change of RX buffer size in XGMAC.
6) Fixes RX buffer size alignment.
7) Fixes the 16KB buffer alignment.
8) Fixes the enabling of 16KB buffer size feature.
9) Always arm the TX coalesce timer so that missed interrupts do not cause
a TX queue timeout.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:43 +0000 (11:17 +0100)]
net: stmmac: Always arm TX Timer at end of transmission start
If TX Coalesce timer is enabled we should always arm it, otherwise we
may hit the case where an interrupt is missed and the TX Queue will
timeout.
Arming the timer does not necessarly mean it will run the tx_clean()
because this function is wrapped around NAPI launcher.
Fixes:
9125cdd1be11 ("stmmac: add the initial tx coalesce schema")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:42 +0000 (11:17 +0100)]
net: stmmac: Enable 16KB buffer size
XGMAC supports maximum MTU that can go to 16KB. Lets add this check in
the calculation of RX buffer size.
Fixes:
7ac6653a085b ("stmmac: Move the STMicroelectronics driver")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:41 +0000 (11:17 +0100)]
net: stmmac: 16KB buffer must be 16 byte aligned
The 16KB RX Buffer must also be 16 byte aligned. Fix it.
Fixes:
7ac6653a085b ("stmmac: Move the STMicroelectronics driver")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:40 +0000 (11:17 +0100)]
net: stmmac: RX buffer size must be 16 byte aligned
We need to align the RX buffer size to at least 16 byte so that IP
doesn't mis-behave. This is required by HW.
Changes from v2:
- Align UP and not DOWN (David)
Fixes:
7ac6653a085b ("stmmac: Move the STMicroelectronics driver")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:39 +0000 (11:17 +0100)]
net: stmmac: xgmac: Clear previous RX buffer size
When switching between buffer sizes we need to clear the previous value.
Fixes:
d6ddfacd95c7 ("net: stmmac: Add DMA related callbacks for XGMAC2")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:38 +0000 (11:17 +0100)]
net: stmmac: Only the last buffer has the FCS field
Only the last received buffer contains the FCS field. Check for end of
packet before trying to strip the FCS field.
Fixes:
88ebe2cf7f3f ("net: stmmac: Rework stmmac_rx()")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:37 +0000 (11:17 +0100)]
net: stmmac: Do not accept invalid MTU values
The maximum MTU value is determined by the maximum size of TX FIFO so
that a full packet can fit in the FIFO. Add a check for this in the MTU
change callback.
Also check if provided and rounded MTU does not passes the maximum limit
of 16K.
Changes from v2:
- Align MTU before checking if its valid
Fixes:
7ac6653a085b ("stmmac: Move the STMicroelectronics driver")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:36 +0000 (11:17 +0100)]
net: stmmac: Determine earlier the size of RX buffer
Split Header feature needs to know the size of RX buffer but current
code is determining it too late. Fix this by moving the RX buffer
computation to earlier stage.
Changes from v2:
- Do not try to align already aligned buffer size
Fixes:
67afd6d1cfdf ("net: stmmac: Add Split Header support and enable it in XGMAC cores")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jose Abreu [Wed, 18 Dec 2019 10:17:35 +0000 (11:17 +0100)]
net: stmmac: selftests: Needs to check the number of Multicast regs
When running the MC and UC filter tests we setup a multicast address
that its expected to be blocked. If the number of available multicast
registers is zero, driver will always pass the multicast packets which
will fail the test.
Check if available multicast addresses is enough before running the
tests.
Fixes:
091810dbded9 ("net: stmmac: Introduce selftests support")
Signed-off-by: Jose Abreu <Jose.Abreu@synopsys.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jia-Ju Bai [Wed, 18 Dec 2019 09:21:55 +0000 (17:21 +0800)]
net: nfc: nci: fix a possible sleep-in-atomic-context bug in nci_uart_tty_receive()
The kernel may sleep while holding a spinlock.
The function call path (from bottom to top) in Linux 4.19 is:
net/nfc/nci/uart.c, 349:
nci_skb_alloc in nci_uart_default_recv_buf
net/nfc/nci/uart.c, 255:
(FUNC_PTR)nci_uart_default_recv_buf in nci_uart_tty_receive
net/nfc/nci/uart.c, 254:
spin_lock in nci_uart_tty_receive
nci_skb_alloc(GFP_KERNEL) can sleep at runtime.
(FUNC_PTR) means a function pointer is called.
To fix this bug, GFP_KERNEL is replaced with GFP_ATOMIC for
nci_skb_alloc().
This bug is found by a static analysis tool STCheck written by myself.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jouni Hogander [Tue, 17 Dec 2019 11:46:34 +0000 (13:46 +0200)]
net-sysfs: Call dev_hold always in rx_queue_add_kobject
Dev_hold has to be called always in rx_queue_add_kobject.
Otherwise usage count drops below 0 in case of failure in
kobject_init_and_add.
Fixes:
b8eb718348b8 ("net-sysfs: Fix reference count leak in rx|netdev_queue_add_kobject")
Reported-by: syzbot <syzbot+30209ea299c09d8785c9@syzkaller.appspotmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: David Miller <davem@davemloft.net>
Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Jouni Hogander <jouni.hogander@unikie.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
John Hurley [Tue, 17 Dec 2019 11:28:56 +0000 (11:28 +0000)]
nfp: flower: fix stats id allocation
As flower rules are added, they are given a stats ID based on the number
of rules that can be supported in firmware. Only after the initial
allocation of all available IDs does the driver begin to reuse those that
have been released.
The initial allocation of IDs was modified to account for multiple memory
units on the offloaded device. However, this introduced a bug whereby the
counter that controls the IDs could be decremented before the ID was
assigned (where it is further decremented). This means that the stats ID
could be assigned as -1/0xfffffff which is out of range.
Fix this by only decrementing the main counter after the current ID has
been assigned.
Fixes:
467322e2627f ("nfp: flower: support multiple memory units for filter offloads")
Signed-off-by: John Hurley <john.hurley@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Dooks (Codethink) [Tue, 17 Dec 2019 11:20:38 +0000 (11:20 +0000)]
net: dsa: make unexported dsa_link_touch() static
dsa_link_touch() is not exported, or defined outside of the
file it is in so make it static to avoid the following warning:
net/dsa/dsa2.c:127:17: warning: symbol 'dsa_link_touch' was not declared. Should it be static?
Signed-off-by: Ben Dooks (Codethink) <ben.dooks@codethink.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Oleksij Rempel [Tue, 17 Dec 2019 06:51:45 +0000 (07:51 +0100)]
net: ag71xx: fix compile warnings
drivers/net/ethernet/atheros/ag71xx.c: In function 'ag71xx_probe':
drivers/net/ethernet/atheros/ag71xx.c:1776:30: warning: passing argument 2 of
'of_get_phy_mode' makes pointer from integer without a cast [-Wint-conversion]
In file included from drivers/net/ethernet/atheros/ag71xx.c:33:
./include/linux/of_net.h:15:69: note: expected 'phy_interface_t *'
{aka 'enum <anonymous> *'} but argument is of type 'int'
Fixes:
0c65b2b90d13c1 ("net: of_get_phy_mode: Change API to solve int/unit warnings")
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Randy Dunlap [Tue, 17 Dec 2019 02:52:45 +0000 (18:52 -0800)]
net: fix kernel-doc warning in <linux/netdevice.h>
Fix missing '*' kernel-doc notation that causes this warning:
../include/linux/netdevice.h:1779: warning: bad line: spinlock
Fixes:
ab92d68fc22f ("net: core: add generic lockdep keys")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Tue, 17 Dec 2019 02:51:03 +0000 (18:51 -0800)]
net: annotate lockless accesses to sk->sk_pacing_shift
sk->sk_pacing_shift can be read and written without lock
synchronization. This patch adds annotations to
document this fact and avoid future syzbot complains.
This might also avoid unexpected false sharing
in sk_pacing_shift_update(), as the compiler
could remove the conditional check and always
write over sk->sk_pacing_shift :
if (sk->sk_pacing_shift != val)
sk->sk_pacing_shift = val;
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Hutchings [Tue, 17 Dec 2019 01:57:40 +0000 (01:57 +0000)]
net: qlogic: Fix error paths in ql_alloc_large_buffers()
ql_alloc_large_buffers() has the usual RX buffer allocation
loop where it allocates skbs and maps them for DMA. It also
treats failure as a fatal error.
There are (at least) three bugs in the error paths:
1. ql_free_large_buffers() assumes that the lrg_buf[] entry for the
first buffer that couldn't be allocated will have .skb == NULL.
But the qla_buf[] array is not zero-initialised.
2. ql_free_large_buffers() DMA-unmaps all skbs in lrg_buf[]. This is
incorrect for the last allocated skb, if DMA mapping failed.
3. Commit
1acb8f2a7a9f ("net: qlogic: Fix memory leak in
ql_alloc_large_buffers") added a direct call to dev_kfree_skb_any()
after the skb is recorded in lrg_buf[], so ql_free_large_buffers()
will double-free it.
The bugs are somewhat inter-twined, so fix them all at once:
* Clear each entry in qla_buf[] before attempting to allocate
an skb for it. This goes half-way to fixing bug 1.
* Set the .skb field only after the skb is DMA-mapped. This
fixes the rest.
Fixes:
1357bfcf7106 ("qla3xxx: Dynamically size the rx buffer queue ...")
Fixes:
0f8ab89e825f ("qla3xxx: Check return code from pci_map_single() ...")
Fixes:
1acb8f2a7a9f ("net: qlogic: Fix memory leak in ql_alloc_large_buffers")
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Marcelo Ricardo Leitner [Tue, 17 Dec 2019 01:01:16 +0000 (22:01 -0300)]
sctp: fix memleak on err handling of stream initialization
syzbot reported a memory leak when an allocation fails within
genradix_prealloc() for output streams. That's because
genradix_prealloc() leaves initialized members initialized when the
issue happens and SCTP stack will abort the current initialization but
without cleaning up such members.
The fix here is to always call genradix_free() when genradix_prealloc()
fails, for output and also input streams, as it suffers from the same
issue.
Reported-by: syzbot+772d9e36c490b18d51d1@syzkaller.appspotmail.com
Fixes:
2075e50caf5e ("sctp: convert to genradix")
Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Tested-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 17 Dec 2019 22:27:35 +0000 (14:27 -0800)]
Merge tag 'wireless-drivers-2019-12-17' of git://git./linux/kernel/git/kvalo/wireless-drivers
Kalle Valo says:
====================
wireless-drivers fixes for v5.5
First set of fixes for v5.5. Fixing security issues, some regressions
and few major bugs.
mwifiex
* security fix for handling country Information Elements (CVE-2019-14895)
* security fix for handling TDLS Information Elements
ath9k
* fix endian issue with ath9k_pci_owl_loader
mt76
* fix default mac address handling
iwlwifi
* fix merge damage which lead to firmware crashing during boot on some devices
* fix device initialisation regression on some devices
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Ciornei [Mon, 16 Dec 2019 15:32:30 +0000 (17:32 +0200)]
dpaa2-ptp: fix double free of the ptp_qoriq IRQ
Upon reusing the ptp_qoriq driver, the ptp_qoriq_free() function was
used on the remove path to free any allocated resources.
The ptp_qoriq IRQ is among these resources that are freed in
ptp_qoriq_free() even though it is also a managed one (allocated using
devm_request_threaded_irq).
Drop the resource managed version of requesting the IRQ in order to not
trigger a double free of the interrupt as below:
[ 226.731005] Trying to free already-free IRQ 126
[ 226.735533] WARNING: CPU: 6 PID: 749 at kernel/irq/manage.c:1707
__free_irq+0x9c/0x2b8
[ 226.743435] Modules linked in:
[ 226.746480] CPU: 6 PID: 749 Comm: bash Tainted: G W
5.4.0-03629-gfd7102c32b2c-dirty #912
[ 226.755857] Hardware name: NXP Layerscape LX2160ARDB (DT)
[ 226.761244] pstate:
40000085 (nZcv daIf -PAN -UAO)
[ 226.766022] pc : __free_irq+0x9c/0x2b8
[ 226.769758] lr : __free_irq+0x9c/0x2b8
[ 226.773493] sp :
ffff8000125039f0
(...)
[ 226.856275] Call trace:
[ 226.858710] __free_irq+0x9c/0x2b8
[ 226.862098] free_irq+0x30/0x70
[ 226.865229] devm_irq_release+0x14/0x20
[ 226.869054] release_nodes+0x1b0/0x220
[ 226.872790] devres_release_all+0x34/0x50
[ 226.876790] device_release_driver_internal+0x100/0x1c0
Fixes:
d346c9e86d86 ("dpaa2-ptp: reuse ptp_qoriq driver")
Cc: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Reviewed-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Tue, 17 Dec 2019 12:28:16 +0000 (13:28 +0100)]
bpf: Fix cgroup local storage prog tracking
Recently noticed that we're tracking programs related to local storage maps
through their prog pointer. This is a wrong assumption since the prog pointer
can still change throughout the verification process, for example, whenever
bpf_patch_insn_single() is called.
Therefore, the prog pointer that was assigned via bpf_cgroup_storage_assign()
is not guaranteed to be the same as we pass in bpf_cgroup_storage_release()
and the map would therefore remain in busy state forever. Fix this by using
the prog's aux pointer which is stable throughout verification and beyond.
Fixes:
de9cbbaadba5 ("bpf: introduce cgroup storage maps")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/1471c69eca3022218666f909bc927a92388fd09e.1576580332.git.daniel@iogearbox.net
David S. Miller [Tue, 17 Dec 2019 03:26:11 +0000 (19:26 -0800)]
Merge tag 'mac80211-for-net-2019-10-16' of git://git./linux/kernel/git/jberg/mac80211
Johannes Berg says:
====================
A handful of fixes:
* disable AQL on most drivers, addressing the iwlwifi issues
* fix double-free on network namespace changes
* fix TID field in frames injected through monitor interfaces
* fix ieee80211_calc_rx_airtime()
* fix NULL pointer dereference in rfkill (and remove BUG_ON)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Arnd Bergmann [Sun, 15 Dec 2019 22:12:14 +0000 (00:12 +0200)]
net: dsa: ocelot: add NET_VENDOR_MICROSEMI dependency
Selecting MSCC_OCELOT_SWITCH is not possible when NET_VENDOR_MICROSEMI
is disabled:
WARNING: unmet direct dependencies detected for MSCC_OCELOT_SWITCH
Depends on [n]: NETDEVICES [=y] && ETHERNET [=n] && NET_VENDOR_MICROSEMI [=n] && NET_SWITCHDEV [=y] && HAS_IOMEM [=y]
Selected by [m]:
- NET_DSA_MSCC_FELIX [=m] && NETDEVICES [=y] && HAVE_NET_DSA [=y] && NET_DSA [=y] && PCI [=y]
Add a Kconfig dependency on NET_VENDOR_MICROSEMI, which also implies
CONFIG_NETDEVICES.
Depending on a vendor config violates menuconfig locality for the DSA
driver, but is the smallest compromise since all other solutions are
much more complicated (see [0]).
https://www.spinics.net/lists/netdev/msg618808.html
Fixes:
56051948773e ("net: dsa: ocelot: add driver for Felix switch family")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Navid Emamdoost [Sun, 15 Dec 2019 01:10:44 +0000 (19:10 -0600)]
net: gemini: Fix memory leak in gmac_setup_txqs
In the implementation of gmac_setup_txqs() the allocated desc_ring is
leaked if TX queue base is not aligned. Release it via
dma_free_coherent.
Fixes:
4d5ae32f5e1e ("net: ethernet: Add a driver for Gemini gigabit ethernet")
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Fri, 13 Dec 2019 20:00:27 +0000 (12:00 -0800)]
net: dsa: b53: Fix egress flooding settings
There were several issues with
53568438e381 ("net: dsa: b53: Add support for port_egress_floods callback") that resulted in breaking connectivity for standalone ports:
- both user and CPU ports must allow unicast and multicast forwarding by
default otherwise this just flat out breaks connectivity for
standalone DSA ports
- IP multicast is treated similarly as multicast, but has separate
control registers
- the UC, MC and IPMC lookup failure register offsets were wrong, and
instead used bit values that are meaningful for the
B53_IP_MULTICAST_CTRL register
Fixes:
53568438e381 ("net: dsa: b53: Add support for port_egress_floods callback")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 17 Dec 2019 00:07:12 +0000 (16:07 -0800)]
Merge branch 'vsock-fixes'
Stefano Garzarella says:
====================
vsock/virtio: fix null-pointer dereference and related precautions
This series mainly solves a possible null-pointer dereference in
virtio_transport_recv_listen() introduced with the multi-transport
support [PATCH 1].
PATCH 2 adds a WARN_ON check for the same potential issue
and a returned error in the virtio_transport_send_pkt_info() function
to avoid crashing the kernel.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Garzarella [Fri, 13 Dec 2019 18:48:01 +0000 (19:48 +0100)]
vsock/virtio: add WARN_ON check on virtio_transport_get_ops()
virtio_transport_get_ops() and virtio_transport_send_pkt_info()
can only be used on connecting/connected sockets, since a socket
assigned to a transport is required.
This patch adds a WARN_ON() on virtio_transport_get_ops() to check
this requirement, a comment and a returned error on
virtio_transport_send_pkt_info(),
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Garzarella [Fri, 13 Dec 2019 18:48:00 +0000 (19:48 +0100)]
vsock/virtio: fix null-pointer dereference in virtio_transport_recv_listen()
With multi-transport support, listener sockets are not bound to any
transport. So, calling virtio_transport_reset(), when an error
occurs, on a listener socket produces the following null-pointer
dereference:
BUG: kernel NULL pointer dereference, address:
00000000000000e8
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: 0000 [#1] SMP PTI
CPU: 0 PID: 20 Comm: kworker/0:1 Not tainted
5.5.0-rc1-ste-00003-gb4be21f316ac-dirty #56
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS ?-20190727_073836-buildvm-ppc64le-16.ppc.fedoraproject.org-3.fc31 04/01/2014
Workqueue: virtio_vsock virtio_transport_rx_work [vmw_vsock_virtio_transport]
RIP: 0010:virtio_transport_send_pkt_info+0x20/0x130 [vmw_vsock_virtio_transport_common]
Code: 1f 84 00 00 00 00 00 0f 1f 00 55 48 89 e5 41 57 41 56 41 55 49 89 f5 41 54 49 89 fc 53 48 83 ec 10 44 8b 76 20 e8 c0 ba fe ff <48> 8b 80 e8 00 00 00 e8 64 e3 7d c1 45 8b 45 00 41 8b 8c 24 d4 02
RSP: 0018:
ffffc900000b7d08 EFLAGS:
00010282
RAX:
0000000000000000 RBX:
ffff88807bf12728 RCX:
0000000000000000
RDX:
ffff88807bf12700 RSI:
ffffc900000b7d50 RDI:
ffff888035c84000
RBP:
ffffc900000b7d40 R08:
ffff888035c84000 R09:
ffffc900000b7d08
R10:
ffff8880781de800 R11:
0000000000000018 R12:
ffff888035c84000
R13:
ffffc900000b7d50 R14:
0000000000000000 R15:
ffff88807bf12724
FS:
0000000000000000(0000) GS:
ffff88807dc00000(0000) knlGS:
0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
CR2:
00000000000000e8 CR3:
00000000790f4004 CR4:
0000000000160ef0
DR0:
0000000000000000 DR1:
0000000000000000 DR2:
0000000000000000
DR3:
0000000000000000 DR6:
00000000fffe0ff0 DR7:
0000000000000400
Call Trace:
virtio_transport_reset+0x59/0x70 [vmw_vsock_virtio_transport_common]
virtio_transport_recv_pkt+0x5bb/0xe50 [vmw_vsock_virtio_transport_common]
? detach_buf_split+0xf1/0x130
virtio_transport_rx_work+0xba/0x130 [vmw_vsock_virtio_transport]
process_one_work+0x1c0/0x300
worker_thread+0x45/0x3c0
kthread+0xfc/0x130
? current_work+0x40/0x40
? kthread_park+0x90/0x90
ret_from_fork+0x35/0x40
Modules linked in: sunrpc kvm_intel kvm vmw_vsock_virtio_transport vmw_vsock_virtio_transport_common irqbypass vsock virtio_rng rng_core
CR2:
00000000000000e8
---[ end trace
e75400e2ea2fa824 ]---
This happens because virtio_transport_reset() calls
virtio_transport_send_pkt_info() that can be used only on
connecting/connected sockets.
This patch fixes the issue, using virtio_transport_reset_no_sock()
instead of virtio_transport_reset() when we are handling a listener
socket.
Fixes:
c0cfa2d8a788 ("vsock: add multi-transports support")
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann [Mon, 16 Dec 2019 16:49:00 +0000 (17:49 +0100)]
bpf: Fix missing prog untrack in release_maps
Commit
da765a2f5993 ("bpf: Add poke dependency tracking for prog array
maps") wrongly assumed that in case of prog load errors, we're cleaning
up all program tracking via bpf_free_used_maps().
However, it can happen that we're still at the point where we didn't copy
map pointers into the prog's aux section such that env->prog->aux->used_maps
is still zero, running into a UAF. In such case, the verifier has similar
release_maps() helper that drops references to used maps from its env.
Consolidate the release code into __bpf_free_used_maps() and call it from
all sides to fix it.
Fixes:
da765a2f5993 ("bpf: Add poke dependency tracking for prog array maps")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/1c2909484ca524ae9f55109b06f22b6213e76376.1576514756.git.daniel@iogearbox.net
Aditya Pakki [Sun, 15 Dec 2019 15:34:08 +0000 (09:34 -0600)]
rfkill: Fix incorrect check to avoid NULL pointer dereference
In rfkill_register, the struct rfkill pointer is first derefernced
and then checked for NULL. This patch removes the BUG_ON and returns
an error to the caller in case rfkill is NULL.
Signed-off-by: Aditya Pakki <pakki001@umn.edu>
Link: https://lore.kernel.org/r/20191215153409.21696-1-pakki001@umn.edu
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Paul Durrant [Fri, 13 Dec 2019 13:20:40 +0000 (13:20 +0000)]
xen-netback: avoid race that can lead to NULL pointer dereference
In function xenvif_disconnect_queue(), the value of queue->rx_irq is
zeroed *before* queue->task is stopped. Unfortunately that task may call
notify_remote_via_irq(queue->rx_irq) and calling that function with a
zero value results in a NULL pointer dereference in evtchn_from_irq().
This patch simply re-orders things, stopping all tasks before zero-ing the
irq values, thereby avoiding the possibility of the race.
Fixes:
2ac061ce97f4 ("xen/netback: cleanup init and deinit code")
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Acked-by: Wei Liu <wei.liu@kernel.org>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Cristian Birsan [Fri, 13 Dec 2019 16:33:11 +0000 (18:33 +0200)]
net: usb: lan78xx: Fix error message format specifier
Display the return code as decimal integer.
Fixes:
55d7de9de6c3 ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet device driver")
Signed-off-by: Cristian Birsan <cristian.birsan@microchip.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Vishal Kulkarni [Fri, 13 Dec 2019 01:09:39 +0000 (06:39 +0530)]
cxgb4: Fix kernel panic while accessing sge_info
The sge_info debugfs collects offload queue info even when offload
capability is disabled and leads to panic.
[ 144.139871] CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
[ 144.139874] CR2:
0000000000000000 CR3:
000000082d456005 CR4:
00000000001606e0
[ 144.139876] Call Trace:
[ 144.139887] sge_queue_start+0x12/0x30 [cxgb4]
[ 144.139897] seq_read+0x1d4/0x3d0
[ 144.139906] full_proxy_read+0x50/0x70
[ 144.139913] vfs_read+0x89/0x140
[ 144.139916] ksys_read+0x55/0xd0
[ 144.139924] do_syscall_64+0x5b/0x1d0
[ 144.139933] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 144.139936] RIP: 0033:0x7f4b01493990
Fix this crash by skipping the offload queue access in sge_qinfo when
offload capability is disabled
Signed-off-by: Herat Ramani <herat@chelsio.com>
Signed-off-by: Vishal Kulkarni <vishal@chelsio.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Ursula Braun [Thu, 12 Dec 2019 21:35:58 +0000 (22:35 +0100)]
net/smc: add fallback check to connect()
FASTOPEN setsockopt() or sendmsg() may switch the SMC socket to fallback
mode. Once fallback mode is active, the native TCP socket functions are
called. Nevertheless there is a small race window, when FASTOPEN
setsockopt/sendmsg runs in parallel to a connect(), and switch the
socket into fallback mode before connect() takes the sock lock.
Make sure the SMC-specific connect setup is omitted in this case.
This way a syzbot-reported refcount problem is fixed, triggered by
different threads running non-blocking connect() and FASTOPEN_KEY
setsockopt.
Reported-by: syzbot+96d3f9ff6a86d37e44c8@syzkaller.appspotmail.com
Fixes:
6d6dd528d5af ("net/smc: fix refcount non-blocking connect() -part 2")
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Russell King [Fri, 13 Dec 2019 10:06:30 +0000 (10:06 +0000)]
net: phylink: fix interface passed to mac_link_up
A mismerge between the following two commits:
c678726305b9 ("net: phylink: ensure consistent phy interface mode")
27755ff88c0e ("net: phylink: Add phylink_mac_link_{up, down} wrapper functions")
resulted in the wrong interface being passed to the mac_link_up()
function. Fix this up.
Fixes:
b4b12b0d2f02 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Thadeu Lima de Souza Cascardo [Fri, 13 Dec 2019 10:39:02 +0000 (07:39 -0300)]
selftests: net: tls: remove recv_rcvbuf test
This test only works when [1] is applied, which was rejected.
Basically, the errors are reported and cleared. In this particular case of
tls sockets, following reads will block.
The test case was originally submitted with the rejected patch, but, then,
was included as part of a different patchset, possibly by mistake.
[1] https://lore.kernel.org/netdev/
20191007035323.4360-2-jakub.kicinski@netronome.com/#t
Thanks Paolo Pisati for pointing out the original patchset where this
appeared.
Fixes:
65190f77424d (selftests/tls: add a test for fragmented messages)
Reported-by: Paolo Pisati <paolo.pisati@canonical.com>
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Jakub Kicinski [Sun, 15 Dec 2019 01:16:12 +0000 (17:16 -0800)]
Merge branch 'gtp-fix-several-bugs-in-gtp-module'
Taehee Yoo says:
====================
gtp: fix several bugs in gtp module
This patchset fixes several bugs in the GTP module.
1. Do not allow adding duplicate TID and ms_addr pdp context.
In the current code, duplicate TID and ms_addr pdp context could be added.
So, RX and TX path could find correct pdp context.
2. Fix wrong condition in ->dumpit() callback.
->dumpit() callback is re-called if dump packet size is too big.
So, before return, it saves last position and then restart from
last dump position.
TID value is used to find last dump position.
GTP module allows adding zero TID value. But ->dumpit() callback ignores
zero TID value.
So, dump would not work correctly if dump packet size too big.
3. Fix use-after-free in ipv4_pdp_find().
RX and TX patch always uses gtp->tid_hash and gtp->addr_hash.
but while packet processing, these hash pointer would be freed.
So, use-after-free would occur.
4. Fix panic because of zero size hashtable
GTP hashtable size could be set by user-space.
If hashsize is set to 0, hashtable will not work and panic will occur.
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Taehee Yoo [Wed, 11 Dec 2019 08:23:48 +0000 (08:23 +0000)]
gtp: avoid zero size hashtable
GTP default hashtable size is 1024 and userspace could set specific
hashtable size with IFLA_GTP_PDP_HASHSIZE. If hashtable size is set to 0
from userspace, hashtable will not work and panic will occur.
Fixes:
459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Taehee Yoo [Wed, 11 Dec 2019 08:23:34 +0000 (08:23 +0000)]
gtp: fix an use-after-free in ipv4_pdp_find()
ipv4_pdp_find() is called in TX packet path of GTP.
ipv4_pdp_find() internally uses gtp->tid_hash to lookup pdp context.
In the current code, gtp->tid_hash and gtp->addr_hash are freed by
->dellink(), which is gtp_dellink().
But gtp_dellink() would be called while packets are processing.
So, gtp_dellink() should not free gtp->tid_hash and gtp->addr_hash.
Instead, dev->priv_destructor() would be used because this callback
is called after all packet processing safely.
Test commands:
ip link add veth1 type veth peer name veth2
ip a a 172.0.0.1/24 dev veth1
ip link set veth1 up
ip a a 172.99.0.1/32 dev lo
gtp-link add gtp1 &
gtp-tunnel add gtp1 v1 200 100 172.99.0.2 172.0.0.2
ip r a 172.99.0.2/32 dev gtp1
ip link set gtp1 mtu 1500
ip netns add ns2
ip link set veth2 netns ns2
ip netns exec ns2 ip a a 172.0.0.2/24 dev veth2
ip netns exec ns2 ip link set veth2 up
ip netns exec ns2 ip a a 172.99.0.2/32 dev lo
ip netns exec ns2 ip link set lo up
ip netns exec ns2 gtp-link add gtp2 &
ip netns exec ns2 gtp-tunnel add gtp2 v1 100 200 172.99.0.1 172.0.0.1
ip netns exec ns2 ip r a 172.99.0.1/32 dev gtp2
ip netns exec ns2 ip link set gtp2 mtu 1500
hping3 172.99.0.2 -2 --flood &
ip link del gtp1
Splat looks like:
[ 72.568081][ T1195] BUG: KASAN: use-after-free in ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.568916][ T1195] Read of size 8 at addr
ffff8880b9a35d28 by task hping3/1195
[ 72.569631][ T1195]
[ 72.569861][ T1195] CPU: 2 PID: 1195 Comm: hping3 Not tainted 5.5.0-rc1 #199
[ 72.570547][ T1195] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[ 72.571438][ T1195] Call Trace:
[ 72.571764][ T1195] dump_stack+0x96/0xdb
[ 72.572171][ T1195] ? ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.572761][ T1195] print_address_description.constprop.5+0x1be/0x360
[ 72.573400][ T1195] ? ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.573971][ T1195] ? ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.574544][ T1195] __kasan_report+0x12a/0x16f
[ 72.575014][ T1195] ? ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.575593][ T1195] kasan_report+0xe/0x20
[ 72.576004][ T1195] ipv4_pdp_find.isra.12+0x130/0x170 [gtp]
[ 72.576577][ T1195] gtp_build_skb_ip4+0x199/0x1420 [gtp]
[ ... ]
[ 72.647671][ T1195] BUG: unable to handle page fault for address:
ffff8880b9a35d28
[ 72.648512][ T1195] #PF: supervisor read access in kernel mode
[ 72.649158][ T1195] #PF: error_code(0x0000) - not-present page
[ 72.649849][ T1195] PGD
a6c01067 P4D
a6c01067 PUD
11fb07067 PMD
11f939067 PTE
800fffff465ca060
[ 72.652958][ T1195] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI
[ 72.653834][ T1195] CPU: 2 PID: 1195 Comm: hping3 Tainted: G B 5.5.0-rc1 #199
[ 72.668062][ T1195] RIP: 0010:ipv4_pdp_find.isra.12+0x86/0x170 [gtp]
[ ... ]
[ 72.679168][ T1195] Call Trace:
[ 72.679603][ T1195] gtp_build_skb_ip4+0x199/0x1420 [gtp]
[ 72.681915][ T1195] ? ipv4_pdp_find.isra.12+0x170/0x170 [gtp]
[ 72.682513][ T1195] ? lock_acquire+0x164/0x3b0
[ 72.682966][ T1195] ? gtp_dev_xmit+0x35e/0x890 [gtp]
[ 72.683481][ T1195] gtp_dev_xmit+0x3c2/0x890 [gtp]
[ ... ]
Fixes:
459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Taehee Yoo [Wed, 11 Dec 2019 08:23:17 +0000 (08:23 +0000)]
gtp: fix wrong condition in gtp_genl_dump_pdp()
gtp_genl_dump_pdp() is ->dumpit() callback of GTP module and it is used
to dump pdp contexts. it would be re-executed because of dump packet size.
If dump packet size is too big, it saves current dump pointer
(gtp interface pointer, bucket, TID value) then it restarts dump from
last pointer.
Current GTP code allows adding zero TID pdp context but dump code
ignores zero TID value. So, last dump pointer will not be found.
In addition, this patch adds missing rcu_read_lock() in
gtp_genl_dump_pdp().
Fixes:
459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Taehee Yoo [Wed, 11 Dec 2019 08:23:00 +0000 (08:23 +0000)]
gtp: do not allow adding duplicate tid and ms_addr pdp context
GTP RX packet path lookups pdp context with TID. If duplicate TID pdp
contexts are existing in the list, it couldn't select correct pdp context.
So, TID value should be unique.
GTP TX packet path lookups pdp context with ms_addr. If duplicate ms_addr pdp
contexts are existing in the list, it couldn't select correct pdp context.
So, ms_addr value should be unique.
Fixes:
459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Mahesh Bandewar [Fri, 6 Dec 2019 23:44:55 +0000 (15:44 -0800)]
bonding: fix active-backup transition after link failure
After the recent fix in commit
1899bb325149 ("bonding: fix state
transition issue in link monitoring"), the active-backup mode with
miimon initially come-up fine but after a link-failure, both members
transition into backup state.
Following steps to reproduce the scenario (eth1 and eth2 are the
slaves of the bond):
ip link set eth1 up
ip link set eth2 down
sleep 1
ip link set eth2 up
ip link set eth1 down
cat /sys/class/net/eth1/bonding_slave/state
cat /sys/class/net/eth2/bonding_slave/state
Fixes:
1899bb325149 ("bonding: fix state transition issue in link monitoring")
CC: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Mahesh Bandewar <maheshb@google.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Jakub Kicinski [Sat, 14 Dec 2019 21:01:17 +0000 (13:01 -0800)]
Merge branch 'bnx2x-bug-fixes'
Manish Chopra says:
====================
bnx2x: bug fixes
This series has two driver changes, one to fix some unexpected
hardware behaviour casued during the parity error recovery in
presence of SR-IOV VFs and another one related for fixing resource
management in the driver among the PFs configured on an engine.
Please consider applying it to "net".
V1->V2:
=======
Fix the compilation errors reported by kbuild test robot
on the patch #1 with CONFIG_BNX2X_SRIOV=n
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Manish Chopra [Wed, 11 Dec 2019 17:59:56 +0000 (09:59 -0800)]
bnx2x: Fix logic to get total no. of PFs per engine
Driver doesn't calculate total number of PFs configured on a
given engine correctly which messed up resources in the PFs
loaded on that engine, leading driver to exceed configuration
of resources (like vlan filters etc.) beyond the limit per
engine, which ended up with asserts from the firmware.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Manish Chopra [Wed, 11 Dec 2019 17:59:55 +0000 (09:59 -0800)]
bnx2x: Do not handle requests from VFs after parity
Parity error from the hardware will cause PF to lose the state
of their VFs due to PF's internal reload and hardware reset following
the parity error. Restrict any configuration request from the VFs after
the parity as it could cause unexpected hardware behavior, only way
for VFs to recover would be to trigger FLR on VFs and reload them.
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Arnd Bergmann [Wed, 11 Dec 2019 12:56:10 +0000 (13:56 +0100)]
net: ethernet: ti: build cpsw-common for switchdev
Without the common part of the driver, the new file fails to link:
drivers/net/ethernet/ti/cpsw_new.o: In function `cpsw_probe':
cpsw_new.c:(.text+0x312c): undefined reference to `ti_cm_get_macid'
Use the same Makefile hack as before, and build cpsw-common.o for
any driver that needs it.
Fixes:
ed3525eda4c4 ("net: ethernet: ti: introduce cpsw switchdev based driver part 1 - dual-emac")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Arnd Bergmann [Wed, 11 Dec 2019 12:56:09 +0000 (13:56 +0100)]
net: ethernet: ti: select PAGE_POOL for switchdev driver
The new driver misses a dependency:
drivers/net/ethernet/ti/cpsw_new.o: In function `cpsw_rx_handler':
cpsw_new.c:(.text+0x259c): undefined reference to `__page_pool_put_page'
cpsw_new.c:(.text+0x25d0): undefined reference to `page_pool_alloc_pages'
drivers/net/ethernet/ti/cpsw_priv.o: In function `cpsw_fill_rx_channels':
cpsw_priv.c:(.text+0x22d8): undefined reference to `page_pool_alloc_pages'
cpsw_priv.c:(.text+0x2420): undefined reference to `__page_pool_put_page'
drivers/net/ethernet/ti/cpsw_priv.o: In function `cpsw_create_xdp_rxqs':
cpsw_priv.c:(.text+0x2624): undefined reference to `page_pool_create'
drivers/net/ethernet/ti/cpsw_priv.o: In function `cpsw_run_xdp':
cpsw_priv.c:(.text+0x2dc8): undefined reference to `__page_pool_put_page'
Other drivers use 'select' for PAGE_POOL, so do the same here.
Fixes:
ed3525eda4c4 ("net: ethernet: ti: introduce cpsw switchdev based driver part 1 - dual-emac")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Haiyang Zhang [Wed, 11 Dec 2019 22:26:27 +0000 (14:26 -0800)]
hv_netvsc: Fix tx_table init in rndis_set_subchannel()
Host can provide send indirection table messages anytime after RSS is
enabled by calling rndis_filter_set_rss_param(). So the host provided
table values may be overwritten by the initialization in
rndis_set_subchannel().
To prevent this problem, move the tx_table initialization before calling
rndis_filter_set_rss_param().
Fixes:
a6fb6aa3cfa9 ("hv_netvsc: Set tx_table to equal weight after subchannels open")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Russell King [Tue, 10 Dec 2019 22:33:05 +0000 (22:33 +0000)]
net: marvell: mvpp2: phylink requires the link interrupt
phylink requires the MAC to report when its link status changes when
operating in inband modes. Failure to report link status changes
means that phylink has no idea when the link events happen, which
results in either the network interface's carrier remaining up or
remaining permanently down.
For example, with a fiber module, if the interface is brought up and
link is initially established, taking the link down at the far end
will cut the optical power. The SFP module's LOS asserts, we
deactivate the link, and the network interface reports no carrier.
When the far end is brought back up, the SFP module's LOS deasserts,
but the MAC may be slower to establish link. If this happens (which
in my tests is a certainty) then phylink never hears that the MAC
has established link with the far end, and the network interface is
stuck reporting no carrier. This means the interface is
non-functional.
Avoiding the link interrupt when we have phylink is basically not
an option, so remove the !port->phylink from the test.
Fixes:
4bb043262878 ("net: mvpp2: phylink support")
Tested-by: Sven Auhagen <sven.auhagen@voleatech.de>
Tested-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Jakub Kicinski [Sat, 14 Dec 2019 17:57:36 +0000 (09:57 -0800)]
Merge branch 'tcp-take-care-of-empty-skbs-in-write-queue'
Eric Dumazet says:
====================
tcp: take care of empty skbs in write queue
We understood recently that TCP sockets could have an empty
skb at the tail of the write queue, leading to various problems.
This patch series :
1) Make sure we do not send an empty packet since this
was unintended and causing crashes in old kernels.
2) Change tcp_write_queue_empty() to not be fooled by
the presence of an empty skb.
3) Fix a bug that could trigger suboptimal epoll()
application behavior under memory pressure.
====================
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Eric Dumazet [Thu, 12 Dec 2019 20:55:31 +0000 (12:55 -0800)]
tcp: refine rule to allow EPOLLOUT generation under mem pressure
At the time commit
ce5ec440994b ("tcp: ensure epoll edge trigger
wakeup when write queue is empty") was added to the kernel,
we still had a single write queue, combining rtx and write queues.
Once we moved the rtx queue into a separate rb-tree, testing
if sk_write_queue is empty has been suboptimal.
Indeed, if we have packets in the rtx queue, we probably want
to delay the EPOLLOUT generation at the time incoming packets
will free them, making room, but more importantly avoiding
flooding application with EPOLLOUT events.
Solution is to use tcp_rtx_and_write_queues_empty() helper.
Fixes:
75c119afe14f ("tcp: implement rb-tree based retransmit queue")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Jason Baron <jbaron@akamai.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Eric Dumazet [Thu, 12 Dec 2019 20:55:30 +0000 (12:55 -0800)]
tcp: refine tcp_write_queue_empty() implementation
Due to how tcp_sendmsg() is implemented, we can have an empty
skb at the tail of the write queue.
Most [1] tcp_write_queue_empty() callers want to know if there is
anything to send (payload and/or FIN)
Instead of checking if the sk_write_queue is empty, we need
to test if tp->write_seq == tp->snd_nxt
[1] tcp_send_fin() was the only caller that expected to
see if an skb was in the write queue, I have changed the code
to reuse the tcp_write_queue_tail() result.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Eric Dumazet [Thu, 12 Dec 2019 20:55:29 +0000 (12:55 -0800)]
tcp: do not send empty skb from tcp_write_xmit()
Backport of commit
fdfc5c8594c2 ("tcp: remove empty skb from
write queue in error cases") in linux-4.14 stable triggered
various bugs. One of them has been fixed in commit
ba2ddb43f270
("tcp: Don't dequeue SYN/FIN-segments from write-queue"), but
we still have crashes in some occasions.
Root-cause is that when tcp_sendmsg() has allocated a fresh
skb and could not append a fragment before being blocked
in sk_stream_wait_memory(), tcp_write_xmit() might be called
and decide to send this fresh and empty skb.
Sending an empty packet is not only silly, it might have caused
many issues we had in the past with tp->packets_out being
out of sync.
Fixes:
c65f7f00c587 ("[TCP]: Simplify SKB data portion allocation with NETIF_F_SG.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Christoph Paasch <cpaasch@apple.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Cc: Jason Baron <jbaron@akamai.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Eric Dumazet [Thu, 12 Dec 2019 18:32:13 +0000 (10:32 -0800)]
6pack,mkiss: fix possible deadlock
We got another syzbot report [1] that tells us we must use
write_lock_irq()/write_unlock_irq() to avoid possible deadlock.
[1]
WARNING: inconsistent lock state
5.5.0-rc1-syzkaller #0 Not tainted
--------------------------------
inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-R} usage.
syz-executor826/9605 [HC1[1]:SC0[0]:HE0:SE1] takes:
ffffffff8a128718 (disc_data_lock){+-..}, at: sp_get.isra.0+0x1d/0xf0 drivers/net/ppp/ppp_synctty.c:138
{HARDIRQ-ON-W} state was registered at:
lock_acquire+0x190/0x410 kernel/locking/lockdep.c:4485
__raw_write_lock_bh include/linux/rwlock_api_smp.h:203 [inline]
_raw_write_lock_bh+0x33/0x50 kernel/locking/spinlock.c:319
sixpack_close+0x1d/0x250 drivers/net/hamradio/6pack.c:657
tty_ldisc_close.isra.0+0x119/0x1a0 drivers/tty/tty_ldisc.c:489
tty_set_ldisc+0x230/0x6b0 drivers/tty/tty_ldisc.c:585
tiocsetd drivers/tty/tty_io.c:2337 [inline]
tty_ioctl+0xe8d/0x14f0 drivers/tty/tty_io.c:2597
vfs_ioctl fs/ioctl.c:47 [inline]
file_ioctl fs/ioctl.c:545 [inline]
do_vfs_ioctl+0x977/0x14e0 fs/ioctl.c:732
ksys_ioctl+0xab/0xd0 fs/ioctl.c:749
__do_sys_ioctl fs/ioctl.c:756 [inline]
__se_sys_ioctl fs/ioctl.c:754 [inline]
__x64_sys_ioctl+0x73/0xb0 fs/ioctl.c:754
do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294
entry_SYSCALL_64_after_hwframe+0x49/0xbe
irq event stamp: 3946
hardirqs last enabled at (3945): [<
ffffffff87c86e43>] __raw_spin_unlock_irq include/linux/spinlock_api_smp.h:168 [inline]
hardirqs last enabled at (3945): [<
ffffffff87c86e43>] _raw_spin_unlock_irq+0x23/0x80 kernel/locking/spinlock.c:199
hardirqs last disabled at (3946): [<
ffffffff8100675f>] trace_hardirqs_off_thunk+0x1a/0x1c arch/x86/entry/thunk_64.S:42
softirqs last enabled at (2658): [<
ffffffff86a8b4df>] spin_unlock_bh include/linux/spinlock.h:383 [inline]
softirqs last enabled at (2658): [<
ffffffff86a8b4df>] clusterip_netdev_event+0x46f/0x670 net/ipv4/netfilter/ipt_CLUSTERIP.c:222
softirqs last disabled at (2656): [<
ffffffff86a8b22b>] spin_lock_bh include/linux/spinlock.h:343 [inline]
softirqs last disabled at (2656): [<
ffffffff86a8b22b>] clusterip_netdev_event+0x1bb/0x670 net/ipv4/netfilter/ipt_CLUSTERIP.c:196
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(disc_data_lock);
<Interrupt>
lock(disc_data_lock);
*** DEADLOCK ***
5 locks held by syz-executor826/9605:
#0:
ffff8880a905e198 (&tty->legacy_mutex){+.+.}, at: tty_lock+0xc7/0x130 drivers/tty/tty_mutex.c:19
#1:
ffffffff899a56c0 (rcu_read_lock){....}, at: mutex_spin_on_owner+0x0/0x330 kernel/locking/mutex.c:413
#2:
ffff8880a496a2b0 (&(&i->lock)->rlock){-.-.}, at: spin_lock include/linux/spinlock.h:338 [inline]
#2:
ffff8880a496a2b0 (&(&i->lock)->rlock){-.-.}, at: serial8250_interrupt+0x2d/0x1a0 drivers/tty/serial/8250/8250_core.c:116
#3:
ffffffff8c104048 (&port_lock_key){-.-.}, at: serial8250_handle_irq.part.0+0x24/0x330 drivers/tty/serial/8250/8250_port.c:1823
#4:
ffff8880a905e090 (&tty->ldisc_sem){++++}, at: tty_ldisc_ref+0x22/0x90 drivers/tty/tty_ldisc.c:288
stack backtrace:
CPU: 1 PID: 9605 Comm: syz-executor826 Not tainted 5.5.0-rc1-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x197/0x210 lib/dump_stack.c:118
print_usage_bug.cold+0x327/0x378 kernel/locking/lockdep.c:3101
valid_state kernel/locking/lockdep.c:3112 [inline]
mark_lock_irq kernel/locking/lockdep.c:3309 [inline]
mark_lock+0xbb4/0x1220 kernel/locking/lockdep.c:3666
mark_usage kernel/locking/lockdep.c:3554 [inline]
__lock_acquire+0x1e55/0x4a00 kernel/locking/lockdep.c:3909
lock_acquire+0x190/0x410 kernel/locking/lockdep.c:4485
__raw_read_lock include/linux/rwlock_api_smp.h:149 [inline]
_raw_read_lock+0x32/0x50 kernel/locking/spinlock.c:223
sp_get.isra.0+0x1d/0xf0 drivers/net/ppp/ppp_synctty.c:138
sixpack_write_wakeup+0x25/0x340 drivers/net/hamradio/6pack.c:402
tty_wakeup+0xe9/0x120 drivers/tty/tty_io.c:536
tty_port_default_wakeup+0x2b/0x40 drivers/tty/tty_port.c:50
tty_port_tty_wakeup+0x57/0x70 drivers/tty/tty_port.c:387
uart_write_wakeup+0x46/0x70 drivers/tty/serial/serial_core.c:104
serial8250_tx_chars+0x495/0xaf0 drivers/tty/serial/8250/8250_port.c:1761
serial8250_handle_irq.part.0+0x2a2/0x330 drivers/tty/serial/8250/8250_port.c:1834
serial8250_handle_irq drivers/tty/serial/8250/8250_port.c:1820 [inline]
serial8250_default_handle_irq+0xc0/0x150 drivers/tty/serial/8250/8250_port.c:1850
serial8250_interrupt+0xf1/0x1a0 drivers/tty/serial/8250/8250_core.c:126
__handle_irq_event_percpu+0x15d/0x970 kernel/irq/handle.c:149
handle_irq_event_percpu+0x74/0x160 kernel/irq/handle.c:189
handle_irq_event+0xa7/0x134 kernel/irq/handle.c:206
handle_edge_irq+0x25e/0x8d0 kernel/irq/chip.c:830
generic_handle_irq_desc include/linux/irqdesc.h:156 [inline]
do_IRQ+0xde/0x280 arch/x86/kernel/irq.c:250
common_interrupt+0xf/0xf arch/x86/entry/entry_64.S:607
</IRQ>
RIP: 0010:cpu_relax arch/x86/include/asm/processor.h:685 [inline]
RIP: 0010:mutex_spin_on_owner+0x247/0x330 kernel/locking/mutex.c:579
Code: c3 be 08 00 00 00 4c 89 e7 e8 e5 06 59 00 4c 89 e0 48 c1 e8 03 42 80 3c 38 00 0f 85 e1 00 00 00 49 8b 04 24 a8 01 75 96 f3 90 <e9> 2f fe ff ff 0f 0b e8 0d 19 09 00 84 c0 0f 85 ff fd ff ff 48 c7
RSP: 0018:
ffffc90001eafa20 EFLAGS:
00000246 ORIG_RAX:
ffffffffffffffd7
RAX:
0000000000000000 RBX:
ffff88809fd9e0c0 RCX:
1ffffffff13266dd
RDX:
0000000000000000 RSI:
0000000000000008 RDI:
0000000000000000
RBP:
ffffc90001eafa60 R08:
1ffff11013d22898 R09:
ffffed1013d22899
R10:
ffffed1013d22898 R11:
ffff88809e9144c7 R12:
ffff8880a905e138
R13:
ffff88809e9144c0 R14:
0000000000000000 R15:
dffffc0000000000
mutex_optimistic_spin kernel/locking/mutex.c:673 [inline]
__mutex_lock_common kernel/locking/mutex.c:962 [inline]
__mutex_lock+0x32b/0x13c0 kernel/locking/mutex.c:1106
mutex_lock_nested+0x16/0x20 kernel/locking/mutex.c:1121
tty_lock+0xc7/0x130 drivers/tty/tty_mutex.c:19
tty_release+0xb5/0xe90 drivers/tty/tty_io.c:1665
__fput+0x2ff/0x890 fs/file_table.c:280
____fput+0x16/0x20 fs/file_table.c:313
task_work_run+0x145/0x1c0 kernel/task_work.c:113
exit_task_work include/linux/task_work.h:22 [inline]
do_exit+0x8e7/0x2ef0 kernel/exit.c:797
do_group_exit+0x135/0x360 kernel/exit.c:895
__do_sys_exit_group kernel/exit.c:906 [inline]
__se_sys_exit_group kernel/exit.c:904 [inline]
__x64_sys_exit_group+0x44/0x50 kernel/exit.c:904
do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x43fef8
Code: Bad RIP value.
RSP: 002b:
00007ffdb07d2338 EFLAGS:
00000246 ORIG_RAX:
00000000000000e7
RAX:
ffffffffffffffda RBX:
0000000000000000 RCX:
000000000043fef8
RDX:
0000000000000000 RSI:
000000000000003c RDI:
0000000000000000
RBP:
00000000004bf730 R08:
00000000000000e7 R09:
ffffffffffffffd0
R10:
00000000004002c8 R11:
0000000000000246 R12:
0000000000000001
R13:
00000000006d1180 R14:
0000000000000000 R15:
0000000000000000
Fixes:
6e4e2f811bad ("6pack,mkiss: fix lock inconsistency")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Eric Dumazet [Sat, 14 Dec 2019 02:20:41 +0000 (18:20 -0800)]
tcp/dccp: fix possible race __inet_lookup_established()
Michal Kubecek and Firo Yang did a very nice analysis of crashes
happening in __inet_lookup_established().
Since a TCP socket can go from TCP_ESTABLISH to TCP_LISTEN
(via a close()/socket()/listen() cycle) without a RCU grace period,
I should not have changed listeners linkage in their hash table.
They must use the nulls protocol (Documentation/RCU/rculist_nulls.txt),
so that a lookup can detect a socket in a hash list was moved in
another one.
Since we added code in commit
d296ba60d8e2 ("soreuseport: Resolve
merge conflict for v4/v6 ordering fix"), we have to add
hlist_nulls_add_tail_rcu() helper.
Fixes:
3b24d854cb35 ("tcp/dccp: do not touch listener sk_refcnt under synflood")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Michal Kubecek <mkubecek@suse.cz>
Reported-by: Firo Yang <firo.yang@suse.com>
Reviewed-by: Michal Kubecek <mkubecek@suse.cz>
Link: https://lore.kernel.org/netdev/20191120083919.GH27852@unicorn.suse.cz/
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Thomas Falcon [Wed, 11 Dec 2019 15:38:39 +0000 (09:38 -0600)]
net/ibmvnic: Fix typo in retry check
This conditional is missing a bang, with the intent
being to break when the retry count reaches zero.
Fixes:
476d96ca9cc5 ("ibmvnic: Bound waits for device queries")
Suggested-by: Juliet Kim <julietk@linux.vnet.ibm.com>
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Hangbin Liu [Wed, 11 Dec 2019 14:20:16 +0000 (22:20 +0800)]
ipv6/addrconf: only check invalid header values when NETLINK_F_STRICT_CHK is set
In commit
4b1373de73a3 ("net: ipv6: addr: perform strict checks also for
doit handlers") we add strict check for inet6_rtm_getaddr(). But we did
the invalid header values check before checking if NETLINK_F_STRICT_CHK
is set. This may break backwards compatibility if user already set the
ifm->ifa_prefixlen, ifm->ifa_flags, ifm->ifa_scope in their netlink code.
I didn't move the nlmsg_len check because I thought it's a valid check.
Reported-by: Jianlin Shi <jishi@redhat.com>
Fixes:
4b1373de73a3 ("net: ipv6: addr: perform strict checks also for doit handlers")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Arnd Bergmann [Tue, 10 Dec 2019 19:56:34 +0000 (20:56 +0100)]
ptp: clockmatrix: add I2C dependency
Without I2C, we get a link failure:
drivers/ptp/ptp_clockmatrix.o: In function `idtcm_xfer.isra.3':
ptp_clockmatrix.c:(.text+0xcc): undefined reference to `i2c_transfer'
drivers/ptp/ptp_clockmatrix.o: In function `idtcm_driver_init':
ptp_clockmatrix.c:(.init.text+0x14): undefined reference to `i2c_register_driver'
drivers/ptp/ptp_clockmatrix.o: In function `idtcm_driver_exit':
ptp_clockmatrix.c:(.exit.text+0x10): undefined reference to `i2c_del_driver'
Fixes:
3a6ba7dc7799 ("ptp: Add a ptp clock driver for IDT ClockMatrix.")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Vincent Cheng <vincent.cheng.xh@renesas.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Jonathan Lemon [Tue, 10 Dec 2019 16:39:46 +0000 (08:39 -0800)]
bnxt: apply computed clamp value for coalece parameter
After executing "ethtool -C eth0 rx-usecs-irq 0", the box becomes
unresponsive, likely due to interrupt livelock. It appears that
a minimum clamp value for the irq timer is computed, but is never
applied.
Fix by applying the corrected clamp value.
Fixes:
74706afa712d ("bnxt_en: Update interrupt coalescing logic.")
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Lorenz Bauer [Fri, 13 Dec 2019 18:08:17 +0000 (18:08 +0000)]
bpf: Clear skb->tstamp in bpf_redirect when necessary
Redirecting a packet from ingress to egress by using bpf_redirect
breaks if the egress interface has an fq qdisc installed. This is the same
problem as fixed in 'commit
8203e2d844d3 ("net: clear skb->tstamp in forwarding paths")
Clear skb->tstamp when redirecting into the egress path.
Fixes:
80b14dee2bea ("net: Add a new socket option for a future transmit time.")
Fixes:
fb420d5d91c1 ("tcp/fq: move back to CLOCK_MONOTONIC")
Signed-off-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/bpf/20191213180817.2510-1-lmb@cloudflare.com
Vivien Didelot [Thu, 12 Dec 2019 17:59:08 +0000 (12:59 -0500)]
mailmap: add entry for myself
I no longer work at Savoir-faire Linux but even though MAINTAINERS is
up-to-date, some emails are still sent to my old email address.
Signed-off-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Subash Abhinov Kasiviswanathan [Thu, 12 Dec 2019 04:00:15 +0000 (04:00 +0000)]
MAINTAINERS: Add maintainers for rmnet
Add myself and Sean as maintainers for rmnet driver.
Signed-off-by: Sean Tranchetti <stranche@codeaurora.org>
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Toke Høiland-Jørgensen [Thu, 12 Dec 2019 11:14:37 +0000 (12:14 +0100)]
mac80211: Turn AQL into an NL80211_EXT_FEATURE
Instead of just having an airtime flag in debugfs, turn AQL into a proper
NL80211_EXT_FEATURE, so drivers can turn it on when they are ready, and so
we also expose the presence of the feature to userspace.
This also has the effect of flipping the default, so drivers have to opt in
to using AQL instead of getting it by default with TXQs. To keep
functionality the same as pre-patch, we set this feature for ath10k (which
is where it is needed the most).
While we're at it, split out the debugfs interface so AQL gets its own
per-station debugfs file instead of using the 'airtime' file.
[Johannes:]
This effectively disables AQL for iwlwifi, where it fixes a number of
issues:
* TSO in iwlwifi is causing underflows and associated warnings in AQL
* HE (802.11ax) rates aren't reported properly so at HE rates, AQL could
never have a valid estimate (it'd use 6 Mbps instead of up to 2400!)
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20191212111437.224294-1-toke@redhat.com
Fixes:
3ace10f5b5ad ("mac80211: Implement Airtime-based Queue Limit (AQL)")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Dan Carpenter [Tue, 26 Nov 2019 12:09:39 +0000 (15:09 +0300)]
mac80211: airtime: Fix an off by one in ieee80211_calc_rx_airtime()
This code was copied from mt76 and inherited an off by one bug from
there. The > should be >= so that we don't read one element beyond
the end of the array.
Fixes:
db3e1c40cf2f ("mac80211: Import airtime calculation code from mt76")
Reported-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20191126120910.ftr4t7me3by32aiz@kili.mountain
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Stefan Bühler [Tue, 26 Nov 2019 10:05:44 +0000 (11:05 +0100)]
cfg80211: fix double-free after changing network namespace
If wdev->wext.keys was initialized it didn't get reset to NULL on
unregister (and it doesn't get set in cfg80211_init_wdev either), but
wdev is reused if unregister was triggered through
cfg80211_switch_netns.
The next unregister (for whatever reason) will try to free
wdev->wext.keys again.
Signed-off-by: Stefan Bühler <source@stbuehler.de>
Link: https://lore.kernel.org/r/20191126100543.782023-1-stefan.buehler@tik.uni-stuttgart.de
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Fredrik Olofsson [Tue, 19 Nov 2019 13:34:51 +0000 (14:34 +0100)]
mac80211: fix TID field in monitor mode transmit
Fix overwriting of the qos_ctrl.tid field for encrypted frames injected on
a monitor interface. While qos_ctrl.tid is not encrypted, it's used as an
input into the encryption algorithm so it's protected, and thus cannot be
modified after encryption. For injected frames, the encryption may already
have been done in userspace, so we cannot change any fields.
Before passing the frame to the driver, the qos_ctrl.tid field is updated
from skb->priority. Prior to
dbd50a851c50 skb->priority was updated in
ieee80211_select_queue_80211(), but this function is no longer always
called.
Update skb->priority in ieee80211_monitor_start_xmit() so that the value
is stored, and when later code 'modifies' the TID it really sets it to
the same value as before, preserving the encryption.
Fixes:
dbd50a851c50 ("mac80211: only allocate one queue when using iTXQs")
Signed-off-by: Fredrik Olofsson <fredrik.olofsson@anyfinetworks.com>
Link: https://lore.kernel.org/r/20191119133451.14711-1-fredrik.olofsson@anyfinetworks.com
[rewrite commit message based on our discussion]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Manish Chopra [Thu, 12 Dec 2019 14:49:28 +0000 (06:49 -0800)]
qede: Fix multicast mac configuration
Driver doesn't accommodate the configuration for max number
of multicast mac addresses, in such particular case it leaves
the device with improper/invalid multicast configuration state,
causing connectivity issues (in lacp bonding like scenarios).
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: Ariel Elior <aelior@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cristian Birsan [Thu, 12 Dec 2019 11:52:47 +0000 (13:52 +0200)]
net: usb: lan78xx: Fix suspend/resume PHY register access error
Lan78xx driver accesses the PHY registers through MDIO bus over USB
connection. When performing a suspend/resume, the PHY registers can be
accessed before the USB connection is resumed. This will generate an
error and will prevent the device to resume correctly.
This patch adds the dependency between the MDIO bus and USB device to
allow correct handling of suspend/resume.
Fixes:
ce85e13ad6ef ("lan78xx: Update to use phylib instead of mii_if_info.")
Signed-off-by: Cristian Birsan <cristian.birsan@microchip.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 12 Dec 2019 04:13:45 +0000 (20:13 -0800)]
Merge git://git./pub/scm/linux/kernel/git/bpf/bpf
Alexei Starovoitov says:
====================
pull-request: bpf 2019-12-11
The following pull-request contains BPF updates for your *net* tree.
We've added 8 non-merge commits during the last 1 day(s) which contain
a total of 10 files changed, 126 insertions(+), 18 deletions(-).
The main changes are:
1) Make BPF trampoline co-exist with ftrace-based tracers, from Alexei.
2) Fix build in minimal configurations, from Arnd.
3) Fix mips, riscv bpf_tail_call limit, from Paul.
4) Fix bpftool segfault, from Toke.
5) Fix samples/bpf, from Daniel.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel T. Lee [Thu, 5 Dec 2019 08:01:14 +0000 (17:01 +0900)]
samples: bpf: fix syscall_tp due to unused syscall
Currently, open() is called from the user program and it calls the syscall
'sys_openat', not the 'sys_open'. This leads to an error of the program
of user side, due to the fact that the counter maps are zero since no
function such 'sys_open' is called.
This commit adds the kernel bpf program which are attached to the
tracepoint 'sys_enter_openat' and 'sys_enter_openat'.
Fixes:
1da236b6be963 ("bpf: add a test case for syscalls/sys_{enter|exit}_* tracepoints")
Signed-off-by: Daniel T. Lee <danieltimlee@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Daniel T. Lee [Thu, 5 Dec 2019 08:01:13 +0000 (17:01 +0900)]
samples: bpf: Replace symbol compare of trace_event
Previously, when this sample is added, commit
1c47910ef8013
("samples/bpf: add perf_event+bpf example"), a symbol 'sys_read' and
'sys_write' has been used without no prefixes. But currently there are
no exact symbols with these under kallsyms and this leads to failure.
This commit changes exact compare to substring compare to keep compatible
with exact symbol or prefixed symbol.
Fixes:
1c47910ef8013 ("samples/bpf: add perf_event+bpf example")
Signed-off-by: Daniel T. Lee <danieltimlee@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191205080114.19766-2-danieltimlee@gmail.com
Alexei Starovoitov [Mon, 9 Dec 2019 00:01:14 +0000 (16:01 -0800)]
selftests/bpf: Test function_graph tracer and bpf trampoline together
Add simple test script to execute funciton graph tracer while BPF trampoline
attaches and detaches from the functions being graph traced.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191209000114.1876138-4-ast@kernel.org
Alexei Starovoitov [Mon, 9 Dec 2019 00:01:13 +0000 (16:01 -0800)]
bpf: Make BPF trampoline use register_ftrace_direct() API
Make BPF trampoline attach its generated assembly code to kernel functions via
register_ftrace_direct() API. It helps ftrace-based tracers co-exist with BPF
trampoline on the same kernel function. It also switches attaching logic from
arch specific text_poke to generic ftrace that is available on many
architectures. text_poke is still necessary for bpf-to-bpf attach and for
bpf_tail_call optimization.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20191209000114.1876138-3-ast@kernel.org
Toke Høiland-Jørgensen [Tue, 10 Dec 2019 18:14:12 +0000 (19:14 +0100)]
bpftool: Don't crash on missing jited insns or ksyms
When the kptr_restrict sysctl is set, the kernel can fail to return
jited_ksyms or jited_prog_insns, but still have positive values in
nr_jited_ksyms and jited_prog_len. This causes bpftool to crash when
trying to dump the program because it only checks the len fields not
the actual pointers to the instructions and ksyms.
Fix this by adding the missing checks.
Fixes:
71bb428fe2c1 ("tools: bpf: add bpftool")
Fixes:
f84192ee00b7 ("tools: bpftool: resolve calls without using imm field")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191210181412.151226-1-toke@redhat.com
Arnd Bergmann [Tue, 10 Dec 2019 20:35:46 +0000 (21:35 +0100)]
bpf: Fix build in minimal configurations, again
Building with -Werror showed another failure:
kernel/bpf/btf.c: In function 'btf_get_prog_ctx_type.isra.31':
kernel/bpf/btf.c:3508:63: error: array subscript 0 is above array bounds of 'u8[0]' {aka 'unsigned char[0]'} [-Werror=array-bounds]
ctx_type = btf_type_member(conv_struct) + bpf_ctx_convert_map[prog_type] * 2;
I don't actually understand why the array is empty, but a similar
fix has addressed a related problem, so I suppose we can do the
same thing here.
Fixes:
ce27709b8162 ("bpf: Fix build in minimal configurations")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20191210203553.2941035-1-arnd@arndb.de
Paul Chaignon [Mon, 9 Dec 2019 18:52:52 +0000 (19:52 +0100)]
bpf, mips: Limit to 33 tail calls
All BPF JIT compilers except RISC-V's and MIPS' enforce a 33-tail calls
limit at runtime. In addition, a test was recently added, in tailcalls2,
to check this limit.
This patch updates the tail call limit in MIPS' JIT compiler to allow
33 tail calls.
Fixes:
b6bd53f9c4e8 ("MIPS: Add missing file for eBPF JIT.")
Reported-by: Mahshid Khezri <khezri.mahshid@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/b8eb2caac1c25453c539248e56ca22f74b5316af.1575916815.git.paul.chaignon@gmail.com
Paul Chaignon [Mon, 9 Dec 2019 18:52:07 +0000 (19:52 +0100)]
bpf, riscv: Limit to 33 tail calls
All BPF JIT compilers except RISC-V's and MIPS' enforce a 33-tail calls
limit at runtime. In addition, a test was recently added, in tailcalls2,
to check this limit.
This patch updates the tail call limit in RISC-V's JIT compiler to allow
33 tail calls. I tested it using the above selftest on an emulated
RISCV64.
Fixes:
2353ecc6f91f ("bpf, riscv: add BPF JIT for RV64G")
Reported-by: Mahshid Khezri <khezri.mahshid@gmail.com>
Signed-off-by: Paul Chaignon <paul.chaignon@orange.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn.topel@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/966fe384383bf23a0ee1efe8d7291c78a3fb832b.1575916815.git.paul.chaignon@gmail.com
Netanel Belgazal [Tue, 10 Dec 2019 11:27:44 +0000 (11:27 +0000)]
net: ena: fix napi handler misbehavior when the napi budget is zero
In netpoll the napi handler could be called with budget equal to zero.
Current ENA napi handler doesn't take that into consideration.
The napi handler handles Rx packets in a do-while loop.
Currently, the budget check happens only after decrementing the
budget, therefore the napi handler, in rare cases, could run over
MAX_INT packets.
In addition to that, this moves all budget related variables to int
calculation and stop mixing u32 to avoid ambiguity
Fixes:
1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)")
Signed-off-by: Netanel Belgazal <netanel@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Wed, 11 Dec 2019 01:45:04 +0000 (17:45 -0800)]
Merge branch 'tipc-fix-some-issues'
Tuong Lien says:
====================
tipc: fix some issues
This series consists of some bug-fixes for TIPC.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Tuong Lien [Tue, 10 Dec 2019 08:21:05 +0000 (15:21 +0700)]
tipc: fix use-after-free in tipc_disc_rcv()
In the function 'tipc_disc_rcv()', the 'msg_peer_net_hash()' is called
to read the header data field but after the message skb has been freed,
that might result in a garbage value...
This commit fixes it by defining a new local variable to store the data
first, just like the other header fields' handling.
Fixes:
f73b12812a3d ("tipc: improve throughput between nodes in netns")
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tuong Lien [Tue, 10 Dec 2019 08:21:04 +0000 (15:21 +0700)]
tipc: fix retrans failure due to wrong destination
When a user message is sent, TIPC will check if the socket has faced a
congestion at link layer. If that happens, it will make a sleep to wait
for the congestion to disappear. This leaves a gap for other users to
take over the socket (e.g. multi threads) since the socket is released
as well. Also, in case of connectionless (e.g. SOCK_RDM), user is free
to send messages to various destinations (e.g. via 'sendto()'), then
the socket's preformatted header has to be updated correspondingly
prior to the actual payload message building.
Unfortunately, the latter action is done before the first action which
causes a condition issue that the destination of a certain message can
be modified incorrectly in the middle, leading to wrong destination
when that message is built. Consequently, when the message is sent to
the link layer, it gets stuck there forever because the peer node will
simply reject it. After a number of retransmission attempts, the link
is eventually taken down and the retransmission failure is reported.
This commit fixes the problem by rearranging the order of actions to
prevent the race condition from occurring, so the message building is
'atomic' and its header will not be modified by anyone.
Fixes:
365ad353c256 ("tipc: reduce risk of user starvation during link congestion")
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tuong Lien [Tue, 10 Dec 2019 08:21:03 +0000 (15:21 +0700)]
tipc: fix potential hanging after b/rcast changing
In commit
c55c8edafa91 ("tipc: smooth change between replicast and
broadcast"), we allow instant switching between replicast and broadcast
by sending a dummy 'SYN' packet on the last used link to synchronize
packets on the links. The 'SYN' message is an object of link congestion
also, so if that happens, a 'SOCK_WAKEUP' will be scheduled to be sent
back to the socket...
However, in that commit, we simply use the same socket 'cong_link_cnt'
counter for both the 'SYN' & normal payload message sending. Therefore,
if both the replicast & broadcast links are congested, the counter will
be not updated correctly but overwritten by the latter congestion.
Later on, when the 'SOCK_WAKEUP' messages are processed, the counter is
reduced one by one and eventually overflowed. Consequently, further
activities on the socket will only wait for the false congestion signal
to disappear but never been met.
Because sending the 'SYN' message is vital for the mechanism, it should
be done anyway. This commit fixes the issue by marking the message with
an error code e.g. 'TIPC_ERR_NO_PORT', so its sending should not face a
link congestion, there is no need to touch the socket 'cong_link_cnt'
either. In addition, in the event of any error (e.g. -ENOBUFS), we will
purge the entire payload message queue and make a return immediately.
Fixes:
c55c8edafa91 ("tipc: smooth change between replicast and broadcast")
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tuong Lien [Tue, 10 Dec 2019 08:21:02 +0000 (15:21 +0700)]
tipc: fix name table rbtree issues
The current rbtree for service ranges in the name table is built based
on the 'lower' & 'upper' range values resulting in a flaw in the rbtree
searching. Some issues have been observed in case of range overlapping:
Case #1: unable to withdraw a name entry:
After some name services are bound, all of them are withdrawn by user
but one remains in the name table forever. This corrupts the table and
that service becomes dummy i.e. no real port.
E.g.
/
{22, 22}
/
/
---> {10, 50}
/ \
/ \
{10, 30} {20, 60}
The node {10, 30} cannot be removed since the rbtree searching stops at
the node's ancestor i.e. {10, 50}, so starting from it will never reach
the finding node.
Case #2: failed to send data in some cases:
E.g. Two service ranges: {20, 60}, {10, 50} are bound. The rbtree for
this service will be one of the two cases below depending on the order
of the bindings:
{20, 60} {10, 50} <--
/ \ / \
/ \ / \
{10, 50} NIL <-- NIL {20, 60}
(a) (b)
Now, try to send some data to service {30}, there will be two results:
(a): Failed, no route to host.
(b): Ok.
The reason is that the rbtree searching will stop at the pointing node
as shown above.
Case #3: Same as case #2b above but if the data sending's scope is
local and the {10, 50} is published by a peer node, then it will result
in 'no route to host' even though the other {20, 60} is for example on
the local node which should be able to get the data.
The issues are actually due to the way we built the rbtree. This commit
fixes it by introducing an additional field to each node - named 'max',
which is the largest 'upper' of that node subtree. The 'max' value for
each subtrees will be propagated correctly whenever a node is inserted/
removed or the tree is rebalanced by the augmented rbtree callbacks.
By this way, we can change the rbtree searching appoarch to solve the
issues above. Another benefit from this is that we can now improve the
searching for a next range matching e.g. in case of multicast, so get
rid of the unneeded looping over all nodes in the tree.
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>