Currently, when debugging AF_XDP workloads, one can correlate the -ENXIO
return code as the case that XSK is not in the bound state. Returning
same code from ndo_xsk_wakeup can be misleading and simply makes it
harder to follow what is going on.
Change ENXIOs in ice's ndo_xsk_wakeup() implementation to EINVALs, so
that when probing it is clear that something is wrong on the driver
side, not the xsk_{recv,send}msg.
There is a -ENETDOWN that can happen from both kernel/driver sides
though, but I don't have a correct replacement for this on one of the
sides, so let's keep it that way.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-9-maciej.fijalkowski@intel.com
When XSK pool uses need_wakeup feature, correlate -ENOBUFS that was
returned from xdp_do_redirect() with a XSK Rx queue being full. In such
case, terminate the Rx processing that is being done on the current HW
Rx ring and let the user space consume descriptors from XSK Rx queue so
that there is room that driver can use later on.
Introduce new internal return code IXGBE_XDP_EXIT that will indicate case
described above.
Note that it does not affect Tx processing that is bound to the same
NAPI context, nor the other Rx rings.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-8-maciej.fijalkowski@intel.com
When XSK pool uses need_wakeup feature, correlate -ENOBUFS that was
returned from xdp_do_redirect() with a XSK Rx queue being full. In such
case, terminate the Rx processing that is being done on the current HW
Rx ring and let the user space consume descriptors from XSK Rx queue so
that there is room that driver can use later on.
Introduce new internal return code I40E_XDP_EXIT that will indicate case
described above.
Note that it does not affect Tx processing that is bound to the same
NAPI context, nor the other Rx rings.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-7-maciej.fijalkowski@intel.com
When XSK pool uses need_wakeup feature, correlate -ENOBUFS that was
returned from xdp_do_redirect() with a XSK Rx queue being full. In such
case, terminate the Rx processing that is being done on the current HW
Rx ring and let the user space consume descriptors from XSK Rx queue so
that there is room that driver can use later on.
Introduce new internal return code ICE_XDP_EXIT that will indicate case
described above.
Note that it does not affect Tx processing that is bound to the same
NAPI context, nor the other Rx rings.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-6-maciej.fijalkowski@intel.com
ixgbe_run_xdp_zc() suggests to compiler that XDP_REDIRECT is the most
probable action returned from BPF program that AF_XDP has in its
pipeline. Let's also bring this suggestion up to the callsite of
ixgbe_run_xdp_zc() so that compiler will be able to generate more
optimized code which in turn will make branch predictor happy.
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-5-maciej.fijalkowski@intel.com
ice_run_xdp_zc() suggests to compiler that XDP_REDIRECT is the most
probable action returned from BPF program that AF_XDP has in its
pipeline. Let's also bring this suggestion up to the callsite of
ice_run_xdp_zc() so that compiler will be able to generate more
optimized code which in turn will make branch predictor happy.
Suggested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220413153015.453864-4-maciej.fijalkowski@intel.com
Inspired by patch that made xdp_do_redirect() return values for XSKMAP
more meaningful, return -ENXIO instead of -EINVAL for socket being
unbound in xsk_rcv_check() as this is the usual value that is returned
for such event. In turn, it is now possible to easily distinguish what
went wrong, which is a bit harder when for both cases checked, -EINVAL
was returned.
Return codes can be counted in a nice way via bpftrace oneliner that
Jesper has shown:
bpftrace -e 'tracepoint:xdp:xdp_redirect* {@err[-args->err] = count();}'
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/bpf/20220413153015.453864-3-maciej.fijalkowski@intel.com
The error codes returned by xdp_do_redirect() when redirecting a frame
to an AF_XDP socket has not been very useful. A driver could not
distinguish between different errors. Prior this change the following
codes where used:
Socket not bound or incorrect queue/netdev: EINVAL
XDP frame/AF_XDP buffer size mismatch: ENOSPC
Could not allocate buffer (copy mode): ENOSPC
AF_XDP Rx buffer full: ENOSPC
After this change:
Socket not bound or incorrect queue/netdev: EINVAL
XDP frame/AF_XDP buffer size mismatch: ENOSPC
Could not allocate buffer (copy mode): ENOMEM
AF_XDP Rx buffer full: ENOBUFS
An AF_XDP zero-copy driver can now potentially determine if the
failure was due to a full Rx buffer, and if so stop processing more
frames, yielding to the userland AF_XDP application.
Signed-off-by: Björn Töpel <bjorn@kernel.org>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Link: https://lore.kernel.org/bpf/20220413153015.453864-2-maciej.fijalkowski@intel.com
Jie Wang says:
====================
net: ethool: add support to get/set tx push by ethtool -G/g
These three patches add tx push in ring params and adapt the set and get APIs
of ring params.
====================
Link: https://lore.kernel.org/r/20220412020121.14140-1-huangguangbin2@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch adds tx push param to hns3 ring param and adapts the set and get
API of ring params. So users can set it by cmd ethtool -G and get it by cmd
ethtool -g.
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently these two checks in ethnl_set_rings are added after rtnl_lock()
which will do useless works if the request is invalid.
So this patch moves these checks before the rtnl_lock() to avoid these
costs.
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently tx push is a standard driver feature which controls use of a fast
path descriptor push. So this patch extends the ringparam APIs and data
structures to support set/get tx push by ethtool -G/g.
Signed-off-by: Jie Wang <wangjie125@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=7gVd
-----END PGP SIGNATURE-----
Merge tag 'block-5.18-2022-04-15' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
- Moving of lower_48_bits() to the block layer and a fix for the
unaligned_be48 added with that originally (Alexander, Keith)
- Fix a bad WARN_ON() for trim size checking (Ming)
- A polled IO timeout fix for null_blk (Ming)
- Silence IO error printing for dead disks (Christoph)
- Compat mode range fix (Khazhismel)
- NVMe pull request via Christoph:
- Tone down the error logging added this merge window a bit
(Chaitanya Kulkarni)
- Quirk devices with non-unique unique identifiers (Christoph)
* tag 'block-5.18-2022-04-15' of git://git.kernel.dk/linux-block:
block: don't print I/O error warning for dead disks
block/compat_ioctl: fix range check in BLKGETSIZE
nvme-pci: disable namespace identifiers for Qemu controllers
nvme-pci: disable namespace identifiers for the MAXIO MAP1002/1202
nvme: add a quirk to disable namespace identifiers
nvme: don't print verbose errors for internal passthrough requests
block: null_blk: end timed out poll request
block: fix offset/size check in bio_trim()
asm-generic: fix __get_unaligned_be48() on 32 bit platforms
block: move lower_48_bits() to block
-----BEGIN PGP SIGNATURE-----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=aTrS
-----END PGP SIGNATURE-----
Merge tag 'io_uring-5.18-2022-04-14' of git://git.kernel.dk/linux-block
Pull io_uring fixes from Jens Axboe:
- Ensure we check and -EINVAL any use of reserved or struct padding.
Although we generally always do that, it's missed in two spots for
resource updates, one for the ring fd registration from this merge
window, and one for the extended arg. Make sure we have all of them
handled. (Dylan)
- A few fixes for the deferred file assignment (me, Pavel)
- Add a feature flag for the deferred file assignment so apps can tell
we handle it correctly (me)
- Fix a small perf regression with the current file position fix in
this merge window (me)
* tag 'io_uring-5.18-2022-04-14' of git://git.kernel.dk/linux-block:
io_uring: abort file assignment prior to assigning creds
io_uring: fix poll error reporting
io_uring: fix poll file assign deadlock
io_uring: use right issue_flags for splice/tee
io_uring: verify pad field is 0 in io_get_ext_arg
io_uring: verify resv is 0 in ringfd register/unregister
io_uring: verify that resv2 is 0 in io_uring_rsrc_update2
io_uring: move io_uring_rsrc_update2 validation
io_uring: fix assign file locking issue
io_uring: stop using io_wq_work as an fd placeholder
io_uring: move apoll->events cache
io_uring: io_kiocb_update_pos() should not touch file for non -1 offset
io_uring: flag the fact that linked file assignment is sane
This Kselftest fixes update consists of a mqueue perf test memory leak
bug fix. mq_perf_tests fail to call CPU_FREE to free memory allocated
by CPU_SET.
-----BEGIN PGP SIGNATURE-----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=+5jP
-----END PGP SIGNATURE-----
Merge tag 'linux-kselftest-fixes-5.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest
Pull Kselftest fixes from Shuah Khan:
"A mqueue perf test memory leak bug fix.
mq_perf_tests failed to call CPU_FREE to free memory allocated by
CPU_SET"
* tag 'linux-kselftest-fixes-5.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
testing/selftests/mqueue: Fix mq_perf_tests to free the allocated cpu set
- 'perf record --per-thread' mode doesn't have the CPU mask setup, so it
can use it to figure out the number of mmaps, fix it.
- Fix segfault accessing sample_id xyarray out of bounds, noticed while
using Intel PT where we have a dummy event to capture text poke perf
metadata events and we mixup the set of CPUs specified by the user
with the all CPUs map needed for text poke.
- Fix 'perf bench numa' to check if CPU used to bind task is online.
- Fix 'perf bench numa' usage of affinity for machines with more than
1000 CPUs.
- Fix misleading add event PMU debug message, noticed while using the
'intel_pt' PMU.
- Fix error check return value of hashmap__new() in 'perf stat', it must
use IS_ERR().
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCYlis4AAKCRCyPKLppCJ+
J7LrAQCow9vLotkRNyYkdgTnGlPa76W2ZVVd52qKs2A62ifVJQD/ewZqBH/JnHf7
UPmkn7B6a4NrEcLiuRAHm1+uLid0swc=
=Nz7z
-----END PGP SIGNATURE-----
Merge tag 'perf-tools-fixes-for-v5.18-2022-04-14' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- 'perf record --per-thread' mode doesn't have the CPU mask setup, so
it can use it to figure out the number of mmaps, fix it.
- Fix segfault accessing sample_id xyarray out of bounds, noticed while
using Intel PT where we have a dummy event to capture text poke perf
metadata events and we mixup the set of CPUs specified by the user
with the all CPUs map needed for text poke.
- Fix 'perf bench numa' to check if CPU used to bind task is online.
- Fix 'perf bench numa' usage of affinity for machines with more than
1000 CPUs.
- Fix misleading add event PMU debug message, noticed while using the
'intel_pt' PMU.
- Fix error check return value of hashmap__new() in 'perf stat', it
must use IS_ERR().
* tag 'perf-tools-fixes-for-v5.18-2022-04-14' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
perf bench: Fix numa bench to fix usage of affinity for machines with #CPUs > 1K
perf bench: Fix numa testcase to check if CPU used to bind task is online
perf record: Fix per-thread option
perf tools: Fix segfault accessing sample_id xyarray
perf stat: Fix error check return value of hashmap__new(), must use IS_ERR()
perf tools: Fix misleading add event PMU debug message
Newer platforms have DSS that aren't necessarily available for both
geometry and compute, two queries will need to exist. This introduces
the first, when passing a valid engine class and engine instance in the
flags returns a topology describing geometry.
Based on past discussion, we currently only support this new query item
on Xe_HP and beyond; earlier platforms do not need to worry about
geometry and compute pipelines having access to different topology and
should continue to use the existing topology query.
v2: fix white space errors
v3: change flags from hosting 2 8 bit numbers to holding a
i915_engine_class_instance struct
v4: add error if non rcs engine passed.
v5 (by MattR):
- Improve kerneldoc and cross references to related structs/enums.
(Daniel)
- Clarify that geometry query is only supported on render engines
(Francisco)
- Clarify that the new query is only supported on Xe_HP+.
- Fix checkpatch warnings.
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Francisco Jerez <currojerez@riseup.net>
UMD (mesa): https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14143
Testcase: igt@i915_query@test-query-geometry-subslices
Signed-off-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-4-matthew.d.roper@intel.com
Convert the comments for drm_i915_query_perf_config and
drm_i915_perf_oa_config to kerneldoc so that they will show up in the
generated documentation. Also correct a couple places that referred to
query_id when they actually meant to refer to query_item.flags.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-2-matthew.d.roper@intel.com
This structure has a great comment describing the fields, but it's not
currently in kerneldoc form and does not show up in the generated
documentation. Let's fix that and also clarify the description of what
"subslice" refers to on gen12 platforms and beyond and that "slice" is
no longer meaningful on Xe_HP and beyond.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20220414192230.749771-1-matthew.d.roper@intel.com
- tone down the error logging added this merge window a bit
(Chaitanya Kulkarni)
- quirk devices with non-unique unique identifiers (me)
-----BEGIN PGP SIGNATURE-----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=m9fO
-----END PGP SIGNATURE-----
Merge tag 'nvme-5.18-2022-04-15' of git://git.infradead.org/nvme into block-5.18
Pull NVMe fixes from Christoph:
"nvme fixes for Linux 5.18
- tone down the error logging added this merge window a bit
(Chaitanya Kulkarni)
- quirk devices with non-unique unique identifiers (me)"
* tag 'nvme-5.18-2022-04-15' of git://git.infradead.org/nvme:
nvme-pci: disable namespace identifiers for Qemu controllers
nvme-pci: disable namespace identifiers for the MAXIO MAP1002/1202
nvme: add a quirk to disable namespace identifiers
nvme: don't print verbose errors for internal passthrough requests
When a disk has been marked dead, don't print warnings for I/O errors
as they are very much expected.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220323163815.1526998-1-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
kernel ulong and compat_ulong_t may not be same width. Use type directly
to eliminate mismatches.
This would result in truncation rather than EFBIG for 32bit mode for
large disks.
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Khazhismel Kumykov <khazhy@google.com>
Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>
Link: https://lore.kernel.org/r/20220414224056.2875681-1-khazhy@google.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The displayed list of Ethernet devices in make menuconfig
has gotten out of order. This is mostly due to changes in vendor
names etc, but also because of new Microsoft entry in wrong place.
This restores so that the display is in order even if the names
of the sub directories are not.
Fixes: ca9c54d2d6 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Given a sufficiently large number of actions, while copying and
reserving memory for a new action of a new flow, if next_offset is
greater than MAX_ACTIONS_BUFSIZE, the function reserve_sfa_size() does
not return -EMSGSIZE as expected, but it allocates MAX_ACTIONS_BUFSIZE
bytes increasing actions_len by req_size. This can then lead to an OOB
write access, especially when further actions need to be copied.
Fix it by rearranging the flow action size check.
KASAN splat below:
==================================================================
BUG: KASAN: slab-out-of-bounds in reserve_sfa_size+0x1ba/0x380 [openvswitch]
Write of size 65360 at addr ffff888147e4001c by task handler15/836
CPU: 1 PID: 836 Comm: handler15 Not tainted 5.18.0-rc1+ #27
...
Call Trace:
<TASK>
dump_stack_lvl+0x45/0x5a
print_report.cold+0x5e/0x5db
? __lock_text_start+0x8/0x8
? reserve_sfa_size+0x1ba/0x380 [openvswitch]
kasan_report+0xb5/0x130
? reserve_sfa_size+0x1ba/0x380 [openvswitch]
kasan_check_range+0xf5/0x1d0
memcpy+0x39/0x60
reserve_sfa_size+0x1ba/0x380 [openvswitch]
__add_action+0x24/0x120 [openvswitch]
ovs_nla_add_action+0xe/0x20 [openvswitch]
ovs_ct_copy_action+0x29d/0x1130 [openvswitch]
? __kernel_text_address+0xe/0x30
? unwind_get_return_address+0x56/0xa0
? create_prof_cpu_mask+0x20/0x20
? ovs_ct_verify+0xf0/0xf0 [openvswitch]
? prep_compound_page+0x198/0x2a0
? __kasan_check_byte+0x10/0x40
? kasan_unpoison+0x40/0x70
? ksize+0x44/0x60
? reserve_sfa_size+0x75/0x380 [openvswitch]
__ovs_nla_copy_actions+0xc26/0x2070 [openvswitch]
? __zone_watermark_ok+0x420/0x420
? validate_set.constprop.0+0xc90/0xc90 [openvswitch]
? __alloc_pages+0x1a9/0x3e0
? __alloc_pages_slowpath.constprop.0+0x1da0/0x1da0
? unwind_next_frame+0x991/0x1e40
? __mod_node_page_state+0x99/0x120
? __mod_lruvec_page_state+0x2e3/0x470
? __kasan_kmalloc_large+0x90/0xe0
ovs_nla_copy_actions+0x1b4/0x2c0 [openvswitch]
ovs_flow_cmd_new+0x3cd/0xb10 [openvswitch]
...
Cc: stable@vger.kernel.org
Fixes: f28cd2af22 ("openvswitch: fix flow actions reallocation")
Signed-off-by: Paolo Valerio <pvalerio@redhat.com>
Acked-by: Eelco Chaudron <echaudro@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If register_netdev() fails , it should return error
code in octep_probe().
Fixes: 862cd659a6 ("octeon_ep: Add driver framework and device initialization")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Feng reported an skb_under_panic BUG triggered by running
test_ip6gretap() in tools/testing/selftests/bpf/test_tunnel.sh:
[ 82.492551] skbuff: skb_under_panic: text:ffffffffb268bb8e len:403 put:12 head:ffff9997c5480000 data:ffff9997c547fff8 tail:0x18b end:0x2c0 dev:ip6gretap11
<...>
[ 82.607380] Call Trace:
[ 82.609389] <TASK>
[ 82.611136] skb_push.cold.109+0x10/0x10
[ 82.614289] __gre6_xmit+0x41e/0x590
[ 82.617169] ip6gre_tunnel_xmit+0x344/0x3f0
[ 82.620526] dev_hard_start_xmit+0xf1/0x330
[ 82.623882] sch_direct_xmit+0xe4/0x250
[ 82.626961] __dev_queue_xmit+0x720/0xfe0
<...>
[ 82.633431] packet_sendmsg+0x96a/0x1cb0
[ 82.636568] sock_sendmsg+0x30/0x40
<...>
The following sequence of events caused the BUG:
1. During ip6gretap device initialization, tunnel->tun_hlen (e.g. 4) is
calculated based on old flags (see ip6gre_calc_hlen());
2. packet_snd() reserves header room for skb A, assuming
tunnel->tun_hlen is 4;
3. Later (in clsact Qdisc), the eBPF program sets a new tunnel key for
skb A using bpf_skb_set_tunnel_key() (see _ip6gretap_set_tunnel());
4. __gre6_xmit() detects the new tunnel key, and recalculates
"tun_hlen" (e.g. 12) based on new flags (e.g. TUNNEL_KEY and
TUNNEL_SEQ);
5. gre_build_header() calls skb_push() with insufficient reserved header
room, triggering the BUG.
As sugguested by Cong, fix it by moving the call to skb_cow_head() after
the recalculation of tun_hlen.
Reproducer:
OBJ=$LINUX/tools/testing/selftests/bpf/test_tunnel_kern.o
ip netns add at_ns0
ip link add veth0 type veth peer name veth1
ip link set veth0 netns at_ns0
ip netns exec at_ns0 ip addr add 172.16.1.100/24 dev veth0
ip netns exec at_ns0 ip link set dev veth0 up
ip link set dev veth1 up mtu 1500
ip addr add dev veth1 172.16.1.200/24
ip netns exec at_ns0 ip addr add ::11/96 dev veth0
ip netns exec at_ns0 ip link set dev veth0 up
ip addr add dev veth1 ::22/96
ip link set dev veth1 up
ip netns exec at_ns0 \
ip link add dev ip6gretap00 type ip6gretap seq flowlabel 0xbcdef key 2 \
local ::11 remote ::22
ip netns exec at_ns0 ip addr add dev ip6gretap00 10.1.1.100/24
ip netns exec at_ns0 ip addr add dev ip6gretap00 fc80::100/96
ip netns exec at_ns0 ip link set dev ip6gretap00 up
ip link add dev ip6gretap11 type ip6gretap external
ip addr add dev ip6gretap11 10.1.1.200/24
ip addr add dev ip6gretap11 fc80::200/24
ip link set dev ip6gretap11 up
tc qdisc add dev ip6gretap11 clsact
tc filter add dev ip6gretap11 egress bpf da obj $OBJ sec ip6gretap_set_tunnel
tc filter add dev ip6gretap11 ingress bpf da obj $OBJ sec ip6gretap_get_tunnel
ping6 -c 3 -w 10 -q ::11
Fixes: 6712abc168 ("ip6_gre: add ip6 gre and gretap collect_md mode")
Reported-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Co-developed-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do not update tunnel->tun_hlen in data plane code. Use a local variable
instead, just like "tunnel_hlen" in net/ipv4/ip_gre.c:gre_fb_xmit().
Co-developed-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Radhey Shyam Pandey says:
====================
net: emaclite: Trivial code cleanup
This patchset fix coding style issues, remove BUFFER_ALIGN
macro and also update copyright text.
I have to resend as earlier series didn't reach mailing list
due to some configuration issue.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
BUFFER_ALIGN macro is used to calculate the number of bytes
required for the next alignment. Instead of this, we can directly
use the skb_reserve(skb, NET_IP_ALIGN) to make the protocol header
buffer aligned on at least a 4-byte boundary, where the NET_IP_ALIGN
is by default defined as 2. So removing the BUFFER_ALIGN and its
related defines which it can be done by the skb_reserve() itself.
Signed-off-by: Shravya Kumbham <shravya.kumbham@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based on recommended guidance Copyright term should be also present in
front of (c). That's why aligned driver to match this pattern.
It helps automated tools with source code scanning.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make coding style changes to fix checkpatch script warnings.
There is no functional change. Fixes below check and warnings-
CHECK: Blank lines aren't necessary after an open brace '{'
CHECK: spinlock_t definition without comment
CHECK: Please don't use multiple blank lines
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
CHECK: braces {} should be used on all arms of this statement
CHECK: Unbalanced braces around else statement
CHECK: Alignment should match open parenthesis
WARNING: Missing a blank line after declarations
Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-04-14
This series contains updates to ice driver only.
Maciej adjusts implementation in __ice_alloc_rx_bufs_zc() for when
ice_fill_rx_descs() does not return the entire buffer request and fixes a
return value for !CONFIG_NET_SWITCHDEV configuration which was preventing
VF creation.
Wojciech prevents eswitch transmit when VFs are being removed which was
causing NULL pointer dereference.
Jianglei Nie fixes a memory leak on error path of getting OROM data.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Steffen Klassert says:
====================
pull request (net): ipsec 2022-04-14
1) Fix the output interface for VRF cases in xfrm_dst_lookup.
From David Ahern.
2) Fix write out of bounds by doing COW on esp output when the
packet size is larger than a page.
From Sabrina Dubroca.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Using pm_runtime_resume_and_get() to replace pm_runtime_get_sync and
pm_runtime_put_noidle. This change is just to simplify the code, no
actual functional changes.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
packet_sock xmit could be dev_queue_xmit, which also returns negative
errors. So only checking positive errors is not enough, or userspace
sendmsg may return success while packet is not send out.
Move the net_xmit_errno() assignment in the braces as checkpatch.pl said
do not use assignment in if condition.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a spelling mistake in a dev_info message. Fix it.
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since commit e5d5aadcf3 ("net/smc: fix sk_refcnt underflow on linkdown
and fallback"), for a fallback connection, __smc_release() does not call
sock_put() if its state is already SMC_CLOSED.
When calling smc_shutdown() after falling back, its state is set to
SMC_CLOSED but does not call sock_put(), so this patch calls it.
Reported-and-tested-by: syzbot+6e29a053eb165bd50de5@syzkaller.appspotmail.com
Fixes: e5d5aadcf3 ("net/smc: fix sk_refcnt underflow on linkdown and fallback")
Signed-off-by: Tony Lu <tonylu@linux.alibaba.com>
Acked-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ido Schimmel says:
====================
mlxsw: Preparations for line cards support
Currently, mlxsw registers thermal zones as well as hwmon entries for
objects such as transceiver modules and gearboxes. In upcoming modular
systems, these objects are no longer found on the main board (i.e., slot
0), but on plug-able line cards. This patchset prepares mlxsw for such
systems in terms of hwmon, thermal and cable access support.
Patches #1-#3 gradually prepare mlxsw for transceiver modules access
support for line cards by splitting some of the internal structures and
some APIs.
Patches #4-#5 gradually prepare mlxsw for hwmon support for line cards
by splitting some of the internal structures and augmenting them with a
slot index.
Patches #6-#7 do the same for thermal zones.
Patch #8 selects cooling device for binding to a thermal zone by exact
name match to prevent binding to non-relevant devices.
Patch #9 replaces internal define for thermal zone name length with a
common define.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace internal define 'MLXSW_THERMAL_ZONE_MAX_NAME' by common
'THERMAL_NAME_LENGTH'.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Modular system supports additional cooling devices "mlxreg_fan1",
"mlxreg_fan2", etcetera. Thermal zones in "mlxsw" driver should be
bound to the same device as before called "mlxreg_fan". Used exact
match for cooling device name to avoid binding to new additional
cooling devices.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add prefix "lc#n" to thermal zones associated with the thermal objects
found on line cards.
For example thermal zone for module #9 located at line card #7 will
have type:
mlxsw-lc7-module9.
And thermal zone for gearbox #3 located at line card #5 will have type:
mlxsw-lc5-gearbox3.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce intermediate level for thermal zones areas.
Currently all thermal zones are associated with thermal objects located
within the main board. Such objects are created during driver
initialization and removed during driver de-initialization.
For line cards in modular system the thermal zones are to be associated
with the specific line card. They should be created whenever new line
card is available (inserted, validated, powered and enabled) and
removed, when line card is getting unavailable.
The thermal objects found on the line card #n are accessed by setting
slot index to #n, while for access to objects found on the main board
slot index should be set to default value zero.
Each thermal area contains the set of thermal zones associated with
particular slot index.
Thus introduction of thermal zone areas allows to use the same APIs for
the main board and line cards, by adding slot index argument.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add 'slot' parameter to 'mlxsw_hwmon_dev' structure. Use this parameter
in mlxsw_reg_mtmp_pack(), mlxsw_reg_mtbr_pack(), mlxsw_reg_mgpir_pack()
and mlxsw_reg_mtmp_slot_index_set() routines.
For main board it'll always be zero, for line cards it'll be set to
the physical slot number in modular systems.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, mlxsw supports a single hwmon device and registers it with
attributes corresponding to the various objects found on the main
board such as fans and gearboxes.
Line cards can have the same objects, but unlike the main board they
can be added and removed while the system is running. The various
hwmon objects found on these line cards should be created when the
line card becomes available and destroyed when the line card becomes
unavailable.
The above can be achieved by representing each line card as a
different hwmon device and registering / unregistering it when the
line card becomes available / unavailable.
Prepare for multi hwmon device support by splitting
'struct mlxsw_hwmon' into 'struct mlxsw_hwmon' and
'struct mlxsw_hwmon_dev'. The first will hold information relevant to
all hwmon devices, whereas the second will hold per-hwmon device
information.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use a separate function for enablement of port module events such
plug/unplug and temperature threshold crossing. The motivation is to
reuse the function for line cards.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The port module core is tasked with module operations such as setting
power mode policy and reset. The per-module information is currently
stored in one large array suited for non-modular systems where only the
main board is present (i.e., slot index 0).
As a preparation for line cards support, allocate a per line card array
according to the queried number of slots in the system. For each line
card, allocate a module array according to the queried maximum number of
modules per-slot.
Signed-off-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
