Skip to content

L
linux
Switch branch/tag
  1. 17 Feb, 2020 16 commits
    • Dmitry Bogdanov's avatar
      net: atlantic: fix out of range usage of active_vlans array · 5a292c89
      Dmitry Bogdanov authored 
      fix static checker warning:
 drivers/net/ethernet/aquantia/atlantic/aq_filters.c:166 aq_check_approve_fvlan()
 error: passing untrusted data to 'test_bit()'
Reported-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Fixes: 7975d2af: ("net: aquantia: add support of rx-vlan-filter offload")
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      5a292c89
    • Pavel Belous's avatar
      net: atlantic: possible fault in transition to hibernation · 52a22f4d
      Pavel Belous authored 
      during hibernation freeze, aq_nic_stop could be invoked
on a stopped device. That may cause panic on access to
not yet allocated vector/ring structures.

Add a check to stop device if it is not yet stopped.

Similiarly after freeze in hibernation thaw, aq_nic_start
could be invoked on a not initialized net device.
Result will be the same.

Add a check to start device if it is initialized.
In our case, this is the same as started.

Fixes: 8aaa112a ("net: atlantic: refactoring pm logic")
Signed-off-by: default avatarPavel Belous <pbelous@marvell.com>
Signed-off-by: default avatarNikita Danilov <ndanilov@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      52a22f4d
    • Pavel Belous's avatar
      net: atlantic: fix potential error handling · 380ec5b9
      Pavel Belous authored 
      Code inspection found that in case of mapping error we do return current
'ret' value. But beside error, it is used to count number of descriptors
allocated for the packet. In that case map_skb function could return '1'.

Changing it to return zero (number of mapped descriptors for skb)

Fixes: 018423e9 ("net: ethernet: aquantia: Add ring support code")
Signed-off-by: default avatarPavel Belous <pbelous@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      380ec5b9
    • Pavel Belous's avatar
      net: atlantic: fix use after free kasan warn · a4980919
      Pavel Belous authored 
      skb->len is used to calculate statistics after xmit invocation.

Under a stress load it may happen that skb will be xmited,
rx interrupt will come and skb will be freed, all before xmit function
is even returned.

Eventually, skb->len will access unallocated area.

Moving stats calculation into tx_clean routine.

Fixes: 018423e9 ("net: ethernet: aquantia: Add ring support code")
Reported-by: default avatarChristophe Vu-Brugier <cvubrugier@fastmail.fm>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarPavel Belous <pbelous@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      a4980919
    • Nikita Danilov's avatar
      net: atlantic: better loopback mode handling · b42726fc
      Nikita Danilov authored 
      Add checks to not enable multiple loopback modes simultaneously,
It was also discovered that for dma loopback to function correctly
promisc mode should be enabled on device.

Fixes: ea4b4d7f ("net: atlantic: loopback tests via private flags")
Signed-off-by: default avatarNikita Danilov <ndanilov@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      b42726fc
    • Egor Pomozov's avatar
      net: atlantic: ptp gpio adjustments · f08a464c
      Egor Pomozov authored 
      Clock adjustment data should be passed to FW as well, otherwise in some
cases a drift was observed when using GPIO features.
Signed-off-by: default avatarEgor Pomozov <epomozov@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      f08a464c
    • Igor Russkikh's avatar
      net: atlantic: check rpc result and wait for rpc address · e7b5f97e
      Igor Russkikh authored 
      Artificial HW reliability tests revealed a possible hangup in
the driver. Normally, when device disappears from bus, all
register reads returns 0xFFFFFFFF.

At remote procedure invocation towards FW there is a logic
where result is compared with -1 in a loop.
That caused an infinite loop if hardware due to some issues
disappears from bus.

Add extra result checks to prevent this.
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e7b5f97e
    • Dmitry Bezrukov's avatar
      net: atlantic: checksum compat issue · 15beab0a
      Dmitry Bezrukov authored 
      Yet another checksum offload compatibility issue was found.

The known issue is that AQC HW marks tcp packets with 0xFFFF checksum
as invalid (1). This is workarounded in driver, passing all the suspicious
packets up to the stack for further csum validation.

Another HW problem (2) is that it hides invalid csum of LRO aggregated
packets inside of the individual descriptors. That was workarounded
by forced scan of all LRO descriptors for checksum errors.

However the scan logic was joint for both LRO and multi-descriptor
packets (jumbos). And this causes the issue.

We have to drop LRO packets with the detected bad checksum
because of (2), but we have to pass jumbo packets to stack because of (1).

When using windows tcp partner with jumbo frames but with LSO disabled
driver discards such frames as bad checksummed. But only LRO frames
should be dropped, not jumbos.

On such a configurations tcp stream have a chance of drops and stucks.

(1) 76f254d4 ("net: aquantia: tcp checksum 0xffff being handled incorrectly")
(2) d08b9a0a ("net: aquantia: do not pass lro session with invalid tcp checksum")

Fixes: d08b9a0a ("net: aquantia: do not pass lro session with invalid tcp checksum")
Signed-off-by: default avatarDmitry Bezrukov <dbezrukov@marvell.com>
Signed-off-by: default avatarIgor Russkikh <irusskikh@marvell.com>
Signed-off-by: default avatarDmitry Bogdanov <dbogdanov@marvell.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      15beab0a
    • Hangbin Liu's avatar
      selftests: forwarding: vxlan_bridge_1d: fix tos value · 4e867c9a
      Hangbin Liu authored 
      After commit 71130f29 ("vxlan: fix tos value before xmit") we start
strict vxlan xmit tos value by RT_TOS(), which limits the tos value less
than 0x1E. With current value 0x40 the test will failed with "v1: Expected
to capture 10 packets, got 0". So let's choose a smaller tos value for
testing.

Fixes: d417ecf5 ("selftests: forwarding: vxlan_bridge_1d: Add a TOS test")
Signed-off-by: default avatarHangbin Liu <liuhangbin@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      4e867c9a
    • Eric Dumazet's avatar
      net: rtnetlink: fix bugs in rtnl_alt_ifname() · 44bfa9c5
      Eric Dumazet authored 
      Since IFLA_ALT_IFNAME is an NLA_STRING, we have no
guarantee it is nul terminated.

We should use nla_strdup() instead of kstrdup(), since this
helper will make sure not accessing out-of-bounds data.

BUG: KMSAN: uninit-value in strlen+0x5e/0xa0 lib/string.c:535
CPU: 1 PID: 19157 Comm: syz-executor.5 Not tainted 5.5.0-rc5-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x1c9/0x220 lib/dump_stack.c:118
 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:118
 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:215
 strlen+0x5e/0xa0 lib/string.c:535
 kstrdup+0x7f/0x1a0 mm/util.c:59
 rtnl_alt_ifname net/core/rtnetlink.c:3495 [inline]
 rtnl_linkprop+0x85d/0xc00 net/core/rtnetlink.c:3553
 rtnl_newlinkprop+0x9d/0xb0 net/core/rtnetlink.c:3568
 rtnetlink_rcv_msg+0x1153/0x1570 net/core/rtnetlink.c:5424
 netlink_rcv_skb+0x451/0x650 net/netlink/af_netlink.c:2477
 rtnetlink_rcv+0x50/0x60 net/core/rtnetlink.c:5442
 netlink_unicast_kernel net/netlink/af_netlink.c:1302 [inline]
 netlink_unicast+0xf9e/0x1100 net/netlink/af_netlink.c:1328
 netlink_sendmsg+0x1248/0x14d0 net/netlink/af_netlink.c:1917
 sock_sendmsg_nosec net/socket.c:639 [inline]
 sock_sendmsg net/socket.c:659 [inline]
 ____sys_sendmsg+0x12b6/0x1350 net/socket.c:2330
 ___sys_sendmsg net/socket.c:2384 [inline]
 __sys_sendmsg+0x451/0x5f0 net/socket.c:2417
 __do_sys_sendmsg net/socket.c:2426 [inline]
 __se_sys_sendmsg+0x97/0xb0 net/socket.c:2424
 __x64_sys_sendmsg+0x4a/0x70 net/socket.c:2424
 do_syscall_64+0xb8/0x160 arch/x86/entry/common.c:296
 entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x45b3b9
Code: ad b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 7b b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007ff1c7b1ac78 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007ff1c7b1b6d4 RCX: 000000000045b3b9
RDX: 0000000000000000 RSI: 0000000020000040 RDI: 0000000000000003
RBP: 000000000075bf20 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff
R13: 00000000000009cb R14: 00000000004cb3dd R15: 000000000075bf2c

Uninit was created at:
 kmsan_save_stack_with_flags mm/kmsan/kmsan.c:144 [inline]
 kmsan_internal_poison_shadow+0x66/0xd0 mm/kmsan/kmsan.c:127
 kmsan_slab_alloc+0x8a/0xe0 mm/kmsan/kmsan_hooks.c:82
 slab_alloc_node mm/slub.c:2774 [inline]
 __kmalloc_node_track_caller+0xb40/0x1200 mm/slub.c:4382
 __kmalloc_reserve net/core/skbuff.c:141 [inline]
 __alloc_skb+0x2fd/0xac0 net/core/skbuff.c:209
 alloc_skb include/linux/skbuff.h:1049 [inline]
 netlink_alloc_large_skb net/netlink/af_netlink.c:1174 [inline]
 netlink_sendmsg+0x7d3/0x14d0 net/netlink/af_netlink.c:1892
 sock_sendmsg_nosec net/socket.c:639 [inline]
 sock_sendmsg net/socket.c:659 [inline]
 ____sys_sendmsg+0x12b6/0x1350 net/socket.c:2330
 ___sys_sendmsg net/socket.c:2384 [inline]
 __sys_sendmsg+0x451/0x5f0 net/socket.c:2417
 __do_sys_sendmsg net/socket.c:2426 [inline]
 __se_sys_sendmsg+0x97/0xb0 net/socket.c:2424
 __x64_sys_sendmsg+0x4a/0x70 net/socket.c:2424
 do_syscall_64+0xb8/0x160 arch/x86/entry/common.c:296
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes: 36fbf1e5 ("net: rtnetlink: add linkprop commands to add and delete alternative ifnames")
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Cc: Jiri Pirko <jiri@mellanox.com>
Reported-by: default avatarsyzbot <syzkaller@googlegroups.com>
Reviewed-by: default avatarJiri Pirko <jiri@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      44bfa9c5
    • Alexandre Belloni's avatar
      net: macb: ensure interface is not suspended on at91rm9200 · e6a41c23
      Alexandre Belloni authored 
      Because of autosuspend, at91ether_start is called with clocks disabled.
Ensure that pm_runtime doesn't suspend the interface as soon as it is
opened as there is no pm_runtime support is the other relevant parts of the
platform support for at91rm9200.

Fixes: d54f89af ("net: macb: Add pm runtime support")
Signed-off-by: default avatarAlexandre Belloni <alexandre.belloni@bootlin.com>
Reviewed-by: default avatarClaudiu Beznea <claudiu.beznea@microchip.com>
Acked-by: default avatarNicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e6a41c23
    • Jethro Beekman's avatar
      net: fib_rules: Correctly set table field when table number exceeds 8 bits · 540e585a
      Jethro Beekman authored 
      In 709772e6, RT_TABLE_COMPAT was added to
allow legacy software to deal with routing table numbers >= 256, but the
same change to FIB rule queries was overlooked.
Signed-off-by: default avatarJethro Beekman <jethro@fortanix.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      540e585a
    • Leon Romanovsky's avatar
      net/rds: Track user mapped pages through special API · 0d4597c8
      Leon Romanovsky authored 
      Convert net/rds to use the newly introduces pin_user_pages() API,
which properly sets FOLL_PIN. Setting FOLL_PIN is now required for
code that requires tracking of pinned pages.

Note that this effectively changes the code's behavior: it now
ultimately calls set_page_dirty_lock(), instead of set_page_dirty().
This is probably more accurate.

As Christoph Hellwig put it, "set_page_dirty() is only safe if we are
dealing with a file backed page where we have reference on the inode it
hangs off." [1]

[1] https://lore.kernel.org/r/20190723153640.GB720@lst.de

Cc: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
Cc: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: default avatarLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: default avatarJohn Hubbard <jhubbard@nvidia.com>
Acked-by: default avatarSantosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      0d4597c8
    • Benjamin Poirier's avatar
      ipv6: Fix nlmsg_flags when splitting a multipath route · afecdb37
      Benjamin Poirier authored 
      When splitting an RTA_MULTIPATH request into multiple routes and adding the
second and later components, we must not simply remove NLM_F_REPLACE but
instead replace it by NLM_F_CREATE. Otherwise, it may look like the netlink
message was malformed.

For example,
	ip route add 2001:db8::1/128 dev dummy0
	ip route change 2001:db8::1/128 nexthop via fe80::30:1 dev dummy0 \
		nexthop via fe80::30:2 dev dummy0
results in the following warnings:
[ 1035.057019] IPv6: RTM_NEWROUTE with no NLM_F_CREATE or NLM_F_REPLACE
[ 1035.057517] IPv6: NLM_F_CREATE should be set when creating new route

This patch makes the nlmsg sequence look equivalent for __ip6_ins_rt() to
what it would get if the multipath route had been added in multiple netlink
operations:
	ip route add 2001:db8::1/128 dev dummy0
	ip route change 2001:db8::1/128 nexthop via fe80::30:1 dev dummy0
	ip route append 2001:db8::1/128 nexthop via fe80::30:2 dev dummy0

Fixes: 27596472 ("ipv6: fix ECMP route replacement")
Signed-off-by: default avatarBenjamin Poirier <bpoirier@cumulusnetworks.com>
Reviewed-by: default avatarMichal Kubecek <mkubecek@suse.cz>
Reviewed-by: default avatarDavid Ahern <dsahern@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      afecdb37
    • Benjamin Poirier's avatar
      ipv6: Fix route replacement with dev-only route · e404b8c7
      Benjamin Poirier authored 
      After commit 27596472 ("ipv6: fix ECMP route replacement") it is no
longer possible to replace an ECMP-able route by a non ECMP-able route.
For example,
	ip route add 2001:db8::1/128 via fe80::1 dev dummy0
	ip route replace 2001:db8::1/128 dev dummy0
does not work as expected.

Tweak the replacement logic so that point 3 in the log of the above commit
becomes:
3. If the new route is not ECMP-able, and no matching non-ECMP-able route
exists, replace matching ECMP-able route (if any) or add the new route.

We can now summarize the entire replace semantics to:
When doing a replace, prefer replacing a matching route of the same
"ECMP-able-ness" as the replace argument. If there is no such candidate,
fallback to the first route found.

Fixes: 27596472 ("ipv6: fix ECMP route replacement")
Signed-off-by: default avatarBenjamin Poirier <bpoirier@cumulusnetworks.com>
Reviewed-by: default avatarMichal Kubecek <mkubecek@suse.cz>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e404b8c7
    • Hangbin Liu's avatar
      selftests: forwarding: use proto icmp for {gretap, ip6gretap}_mac testing · e8023b03
      Hangbin Liu authored 
      For tc ip_proto filter, when we extract the flow via __skb_flow_dissect()
without flag FLOW_DISSECTOR_F_STOP_AT_ENCAP, we will continue extract to
the inner proto.

So for GRE + ICMP messages, we should not track GRE proto, but inner ICMP
proto.

For test mirror_gre.sh, it may make user confused if we capture ICMP
message on $h3(since the flow is GRE message). So I move the capture
dev to h3-gt{4,6}, and only capture ICMP message.

Before the fix:
]# ./mirror_gre.sh
TEST: ingress mirror to gretap (skip_hw)                            [ OK ]
TEST: egress mirror to gretap (skip_hw)                             [ OK ]
TEST: ingress mirror to ip6gretap (skip_hw)                         [ OK ]
TEST: egress mirror to ip6gretap (skip_hw)                          [ OK ]
TEST: ingress mirror to gretap: envelope MAC (skip_hw)              [FAIL]
 Expected to capture 10 packets, got 0.
TEST: egress mirror to gretap: envelope MAC (skip_hw)               [FAIL]
 Expected to capture 10 packets, got 0.
TEST: ingress mirror to ip6gretap: envelope MAC (skip_hw)           [FAIL]
 Expected to capture 10 packets, got 0.
TEST: egress mirror to ip6gretap: envelope MAC (skip_hw)            [FAIL]
 Expected to capture 10 packets, got 0.
TEST: two simultaneously configured mirrors (skip_hw)               [ OK ]
WARN: Could not test offloaded functionality

After fix:
]# ./mirror_gre.sh
TEST: ingress mirror to gretap (skip_hw)                            [ OK ]
TEST: egress mirror to gretap (skip_hw)                             [ OK ]
TEST: ingress mirror to ip6gretap (skip_hw)                         [ OK ]
TEST: egress mirror to ip6gretap (skip_hw)                          [ OK ]
TEST: ingress mirror to gretap: envelope MAC (skip_hw)              [ OK ]
TEST: egress mirror to gretap: envelope MAC (skip_hw)               [ OK ]
TEST: ingress mirror to ip6gretap: envelope MAC (skip_hw)           [ OK ]
TEST: egress mirror to ip6gretap: envelope MAC (skip_hw)            [ OK ]
TEST: two simultaneously configured mirrors (skip_hw)               [ OK ]
WARN: Could not test offloaded functionality

Fixes: ba8d3987 ("selftests: forwarding: Add test for mirror to gretap")
Signed-off-by: default avatarHangbin Liu <liuhangbin@gmail.com>
Reviewed-by: default avatarPetr Machata <pmachata@gmail.com>
Tested-by: default avatarPetr Machata <pmachata@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      e8023b03
  3. 14 Feb, 2020 24 commits
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7