This commit replaces separate headers with a single vmlinux.h to
tracing programs. Thanks to that, we no longer need to define the
argument structure for tracing programs directly. For example, argument
for the sched_switch tracpepoint (sched_switch_args) can be replaced
with the vmlinux.h provided trace_event_raw_sched_switch.
Additional defines have been added to the BPF program either directly
or through the inclusion of net_shared.h. Defined values are
PERF_MAX_STACK_DEPTH, IFNAMSIZ constants and __stringify() macro. This
change enables the BPF program to access internal structures with BTF
generated "vmlinux.h" header.
Signed-off-by: Daniel T. Lee <danieltimlee@gmail.com>
Link: https://lore.kernel.org/r/20230818090119.477441-3-danieltimlee@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Currently, compiling the bpf programs will result the warning with the
ignored attribute as follows. This commit fixes the warning by adding
cf-protection option.
In file included from ./arch/x86/include/asm/linkage.h:6:
./arch/x86/include/asm/ibt.h:77:8: warning: 'nocf_check' attribute ignored; use -fcf-protection to enable the attribute [-Wignored-attributes]
extern __noendbr u64 ibt_save(bool disable);
^
./arch/x86/include/asm/ibt.h:32:34: note: expanded from macro '__noendbr'
^
Signed-off-by: Daniel T. Lee <danieltimlee@gmail.com>
Link: https://lore.kernel.org/r/20230818090119.477441-2-danieltimlee@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Hou Tao says:
====================
Remove unnecessary synchronizations in cpumap
From: Hou Tao <houtao1@huawei.com>
Hi,
This is the formal patchset to remove unnecessary synchronizations in
cpu-map after address comments and collect Rvb tags from Toke
Høiland-Jørgensen (Big thanks to Toke). Patch #1 removes the unnecessary
rcu_barrier() when freeing bpf_cpu_map_entry and replaces it by
queue_rcu_work(). Patch #2 removes the unnecessary call_rcu() and
queue_work() when destroying cpu-map and does the freeing directly.
Test the patchset by using xdp_redirect_cpu and virtio-net. Both
xdp-mode and skb-mode have been exercised and no issues were reported.
As ususal, comments and suggestions are always welcome.
Change Log:
v1:
* address comments from Toke Høiland-Jørgensen
* add Rvb tags from Toke Høiland-Jørgensen
* update outdated comment in cpu_map_delete_elem()
RFC: https://lore.kernel.org/bpf/20230728023030.1906124-1-houtao@huaweicloud.com
====================
Link: https://lore.kernel.org/r/20230816045959.358059-1-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
After synchronous_rcu(), both the dettached XDP program and
xdp_do_flush() are completed, and the only user of bpf_cpu_map_entry
will be cpu_map_kthread_run(), so instead of calling
__cpu_map_entry_replace() to stop kthread and cleanup entry after a RCU
grace period, do these things directly.
Signed-off-by: Hou Tao <houtao1@huawei.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20230816045959.358059-3-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
As for now __cpu_map_entry_replace() uses call_rcu() to wait for the
inflight xdp program to exit the RCU read critical section, and then
launch kworker cpu_map_kthread_stop() to call kthread_stop() to flush
all pending xdp frames or skbs.
But it is unnecessary to use rcu_barrier() in cpu_map_kthread_stop() to
wait for the completion of __cpu_map_entry_free(), because rcu_barrier()
will wait for all pending RCU callbacks and cpu_map_kthread_stop() only
needs to wait for the completion of a specific __cpu_map_entry_free().
So use queue_rcu_work() to replace call_rcu(), schedule_work() and
rcu_barrier(). queue_rcu_work() will queue a __cpu_map_entry_free()
kworker after a RCU grace period. Because __cpu_map_entry_free() is
running in a kworker context, so it is OK to do all of these freeing
procedures include kthread_stop() in it.
After the update, there is no need to do reference-counting for
bpf_cpu_map_entry, because bpf_cpu_map_entry is freed directly in
__cpu_map_entry_free(), so just remove it.
Signed-off-by: Hou Tao <houtao1@huawei.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Link: https://lore.kernel.org/r/20230816045959.358059-2-houtao@huaweicloud.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When building the kernel and selftest with clang compiler (llvm17 or llvm18),
I hit the following compilation failure:
In file included from progs/test_lwt_redirect.c:3:
In file included from /usr/include/linux/ip.h:21:
In file included from /usr/include/asm/byteorder.h:5:
In file included from /usr/include/linux/byteorder/little_endian.h:13:
/usr/include/linux/swab.h:136:8: error: unknown type name '__always_inline'
136 | static __always_inline unsigned long __swab(const unsigned long y)
| ^
/usr/include/linux/swab.h:171:8: error: unknown type name '__always_inline'
171 | static __always_inline __u16 __swab16p(const __u16 *p)
...
bpf_helpers.h file provided a definition for __always_inline.
Putting 'ip.h' after 'bpf_helpers.h' fixed the issue.
Fixes: 43a7c3ef8a ("selftests/bpf: Add lwt_xmit tests for BPF_REDIRECT")
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
Link: https://lore.kernel.org/r/20230818174312.1883381-1-yonghong.song@linux.dev
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
This patch adds selftests that exercise kfunc flavor relocation
functionality added in the previous patch. The actual kfunc defined
in kernel/bpf/helpers.c is:
struct task_struct *bpf_task_acquire(struct task_struct *p)
The following relocation behaviors are checked:
struct task_struct *bpf_task_acquire___one(struct task_struct *name)
* Should succeed despite differing param name
struct task_struct *bpf_task_acquire___two(struct task_struct *p, void *ctx)
* Should fail because there is no two-param bpf_task_acquire
struct task_struct *bpf_task_acquire___three(void *ctx)
* Should fail because, despite vmlinux's bpf_task_acquire having one param,
the types don't match
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Link: https://lore.kernel.org/bpf/20230817225353.2570845-2-davemarchevsky@fb.com
The function signature of kfuncs can change at any time due to their
intentional lack of stability guarantees. As kfuncs become more widely
used, BPF program writers will need facilities to support calling
different versions of a kfunc from a single BPF object. Consider this
simplified example based on a real scenario we ran into at Meta:
/* initial kfunc signature */
int some_kfunc(void *ptr)
/* Oops, we need to add some flag to modify behavior. No problem,
change the kfunc. flags = 0 retains original behavior */
int some_kfunc(void *ptr, long flags)
If the initial version of the kfunc is deployed on some portion of the
fleet and the new version on the rest, a fleetwide service that uses
some_kfunc will currently need to load different BPF programs depending
on which some_kfunc is available.
Luckily CO-RE provides a facility to solve a very similar problem,
struct definition changes, by allowing program writers to declare
my_struct___old and my_struct___new, with ___suffix being considered a
'flavor' of the non-suffixed name and being ignored by
bpf_core_type_exists and similar calls.
This patch extends the 'flavor' facility to the kfunc extern
relocation process. BPF program writers can now declare
extern int some_kfunc___old(void *ptr)
extern int some_kfunc___new(void *ptr, int flags)
then test which version of the kfunc exists with bpf_ksym_exists.
Relocation and verifier's dead code elimination will work in concert as
expected, allowing this pattern:
if (bpf_ksym_exists(some_kfunc___old))
some_kfunc___old(ptr);
else
some_kfunc___new(ptr, 0);
Signed-off-by: Dave Marchevsky <davemarchevsky@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: David Vernet <void@manifault.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230817225353.2570845-1-davemarchevsky@fb.com
If a testcase returns a wrong (unexpected) value, print the expected and
returned value in hex notation in addition to the decimal notation.
This is very useful in tests which bit-shift hex values left or right and
helped me a lot while developing the JIT compiler for the hppa architecture.
Additionally fix two typos: dowrd -> dword, tall calls -> tail calls.
Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/ZN6ZAAVoWZpsD1Jf@p100
There is no lwt test case for BPF_REROUTE yet. Add test cases for both
normal and abnormal situations. The abnormal situation is set up with an
fq qdisc on the reroute target device. Without proper fixes, overflow
this qdisc queue limit (to trigger a drop) would panic the kernel.
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/62c8ddc1e924269dcf80d2e8af1a1e632cee0b3a.1692326837.git.yan@cloudflare.com
There is no lwt_xmit test case for BPF_REDIRECT yet. Add test cases for
both normal and abnormal situations. For abnormal test cases, devices
are set down or have its carrier set down. Without proper fixes,
BPF_REDIRECT to either ingress or egress of such device would panic the
kernel.
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/96bf435243641939d9c9da329fab29cb45f7df22.1692326837.git.yan@cloudflare.com
LWTUNNEL_XMIT_CONTINUE is implicitly assumed in ip(6)_finish_output2,
such that any positive return value from a xmit hook could cause
unexpected continue behavior, despite that related skb may have been
freed. This could be error-prone for future xmit hook ops. One of the
possible errors is to return statuses of dst_output directly.
To make the code safer, redefine LWTUNNEL_XMIT_CONTINUE value to
distinguish from dst_output statuses and check the continue
condition explicitly.
Fixes: 3a0af8fd61 ("bpf: BPF for lightweight tunnel infrastructure")
Suggested-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/96b939b85eda00e8df4f7c080f770970a4c5f698.1692326837.git.yan@cloudflare.com
BPF encap ops can return different types of positive values, such like
NET_RX_DROP, NET_XMIT_CN, NETDEV_TX_BUSY, and so on, from function
skb_do_redirect and bpf_lwt_xmit_reroute. At the xmit hook, such return
values would be treated implicitly as LWTUNNEL_XMIT_CONTINUE in
ip(6)_finish_output2. When this happens, skbs that have been freed would
continue to the neighbor subsystem, causing use-after-free bug and
kernel crashes.
To fix the incorrect behavior, skb_do_redirect return values can be
simply discarded, the same as tc-egress behavior. On the other hand,
bpf_lwt_xmit_reroute returns useful errors to local senders, e.g. PMTU
information. Thus convert its return values to avoid the conflict with
LWTUNNEL_XMIT_CONTINUE.
Fixes: 3a0af8fd61 ("bpf: BPF for lightweight tunnel infrastructure")
Reported-by: Jordan Griege <jgriege@cloudflare.com>
Suggested-by: Martin KaFai Lau <martin.lau@linux.dev>
Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Yan Zhai <yan@cloudflare.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/0d2b878186cfe215fec6b45769c1cd0591d3628d.1692326837.git.yan@cloudflare.com
Add support for 32-bit offset jmp instructions. Given the arm64 direct jump
range is +-128MB, which is large enough for BPF prog, jumps beyond this range
are not supported.
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Florent Revest <revest@chromium.org>
Acked-by: Florent Revest <revest@chromium.org>
Link: https://lore.kernel.org/bpf/20230815154158.717901-6-xukuohai@huaweicloud.com
To support BPF sign-extend load instructions, add encoders for
LDRSB/LDRSH/LDRSW.
LDRSB/LDRSH/LDRSW (immediate) is encoded as follows:
3 2 2 2 2 1 0 0
0 7 6 4 2 0 5 0
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| sz|1 1 1|0|0 1|opc| imm12 | Rn | Rt |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
LDRSB/LDRSH/LDRSW (register) is encoded as follows:
3 2 2 2 2 2 1 1 1 1 0 0
0 7 6 4 2 1 6 3 2 0 5 0
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| sz|1 1 1|0|0 0|opc|1| Rm | opt |S|1 0| Rn | Rt |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
where:
- sz
indicates whether 8-bit, 16-bit or 32-bit data is to be loaded
- opc
opc[1] (bit 23) is always 1 and opc[0] == 1 indicates regsize
is 32-bit. Since BPF signed load instructions always exend the
sign bit to bit 63 regardless of whether it loads an 8-bit,
16-bit or 32-bit data. So only 64-bit register size is required.
That is, it's sufficient to set field opc fixed to 0x2.
- opt
Indicates whether to sign extend the offset register Rm and the
effective bits of Rm. We set opt to 0x7 (SXTX) since we'll use
Rm as a sgined 64-bit value in BPF.
- S
Optional only when opt field is 0x3 (LSL)
In short, the above fields are encoded to the values listed below.
sz opc opt S
LDRSB (immediate) 0x0 0x2 na na
LDRSH (immediate) 0x1 0x2 na na
LDRSW (immediate) 0x2 0x2 na na
LDRSB (register) 0x0 0x2 0x7 0
LDRSH (register) 0x1 0x2 0x7 0
LDRSW (register) 0x2 0x2 0x7 0
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Florent Revest <revest@chromium.org>
Acked-by: Florent Revest <revest@chromium.org>
Link: https://lore.kernel.org/bpf/20230815154158.717901-2-xukuohai@huaweicloud.com
Breno Leitao says:
====================
netconsole: Enable compile time configuration
Enable netconsole features to be set at compilation time. Create two
Kconfig options that allow users to set extended logs and release
prepending features at compilation time.
The first patch de-duplicates the initialization code, and the second
patch adds the support in the de-duplicated code, avoiding touching two
different functions with the same change.
====================
Link: https://lore.kernel.org/r/20230811093158.1678322-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Enable netconsole features to be set at compilation time. Create two
Kconfig options that allow users to set extended logs and release
prepending features at compilation time.
Right now, the user needs to pass command line parameters to netconsole,
such as "+"/"r" to enable extended logs and version prepending features.
With these two options, the user could set the default values for the
features at compile time, and don't need to pass it in the command line
to get them enabled, simplifying the command line.
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230811093158.1678322-3-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
De-duplicate the initialization and allocation code for struct
netconsole_target.
The same allocation and initialization code is duplicated in two
different places in the netconsole subsystem, when the netconsole target
is initialized by command line parameters (alloc_param_target()), and
dynamically by sysfs (make_netconsole_target()).
Create a helper function, and call it from the two different functions.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230811093158.1678322-2-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When building with clang 18 I see the following warning:
| drivers/net/mdio/mdio-xgene.c:338:13: warning: cast to smaller integer
| type 'enum xgene_mdio_id' from 'const void *' [-Wvoid-pointer-to-enum-cast]
| 338 | mdio_id = (enum xgene_mdio_id)of_id->data;
This is due to the fact that `of_id->data` is a void* while `enum
xgene_mdio_id` has the size of an int. This leads to truncation and
possible data loss.
Link: https://github.com/ClangBuiltLinux/linux/issues/1910
Reported-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://lore.kernel.org/r/20230815-void-drivers-net-mdio-mdio-xgene-v1-1-5304342e0659@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
François Michel says:
====================
netem: use a seeded PRNG for loss and corruption events
In order to reproduce bugs or performance evaluation of
network protocols and applications, it is useful to have
reproducible test suites and tools. This patch adds
a way to specify a PRNG seed through the
TCA_NETEM_PRNG_SEED attribute for generating netem
loss and corruption events. Initializing the qdisc
with the same seed leads to the exact same loss
and corruption patterns. If no seed is explicitly
specified, the qdisc generates a random seed using
get_random_u64().
This patch can be and has been tested using tc from
the following iproute2-next fork:
https://github.com/francoismichel/iproute2-next
For instance, setting the seed 42424242 on the loopback
with a loss rate of 10% will systematically drop the 5th,
12th and 24th packet when sending 25 packets.
====================
Link: https://lore.kernel.org/r/20230815092348.1449179-1-francois.michel@uclouvain.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use prandom_u32_state() instead of get_random_u32() to generate
the correlated loss events of netem.
Signed-off-by: François Michel <francois.michel@uclouvain.be>
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Link: https://lore.kernel.org/r/20230815092348.1449179-4-francois.michel@uclouvain.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Use prandom_u32_state() instead of get_random_u32() to generate
the random loss events of netem. The state of the prng is part
of the prng attribute of struct netem_sched_data.
Signed-off-by: François Michel <francois.michel@uclouvain.be>
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Link: https://lore.kernel.org/r/20230815092348.1449179-3-francois.michel@uclouvain.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add prng attribute to struct netem_sched_data and
allows setting the seed of the PRNG through netlink
using the new TCA_NETEM_PRNG_SEED attribute.
The PRNG attribute is not actually used yet.
Signed-off-by: François Michel <francois.michel@uclouvain.be>
Reviewed-by: Simon Horman <horms@kernel.org>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Link: https://lore.kernel.org/r/20230815092348.1449179-2-francois.michel@uclouvain.be
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
PCI core API pci_dev_id() can be used to get the BDF number for a pci
device. We don't need to compose it manually. Use pci_dev_id() to
simplify the code a little bit.
Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
Link: https://lore.kernel.org/r/20230815024248.3519068-1-zhangjialin11@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTFp0I1jqZrAX+hPRXbK58LschIgwUCZN0eNgAKCRDbK58LschI
gwhhAQCwbrEgA3LslDlk22eqyfRH04D+9d7Kc3ISQssyjlr9swD+NfwfDvYqopwj
Dp67QkHdluixf2/NMPTEvg/CA4mlmww=
=4BwF
-----END PGP SIGNATURE-----
Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says:
====================
pull-request: bpf-next 2023-08-16
We've added 17 non-merge commits during the last 6 day(s) which contain
a total of 20 files changed, 1179 insertions(+), 37 deletions(-).
The main changes are:
1) Add a BPF hook in sys_socket() to change the protocol ID
from IPPROTO_TCP to IPPROTO_MPTCP to cover migration for legacy
applications, from Geliang Tang.
2) Follow-up/fallout fix from the SO_REUSEPORT + bpf_sk_assign work
to fix a splat on non-fullsock sks in inet[6]_steal_sock,
from Lorenz Bauer.
3) Improvements to struct_ops links to avoid forcing presence of
update/validate callbacks. Also add bpf_struct_ops fields documentation,
from David Vernet.
4) Ensure libbpf sets close-on-exec flag on gzopen, from Marco Vedovati.
5) Several new tcx selftest additions and bpftool link show support for
tcx and xdp links, from Daniel Borkmann.
6) Fix a smatch warning on uninitialized symbol in
bpf_perf_link_fill_kprobe, from Yafang Shao.
7) BPF selftest fixes e.g. misplaced break in kfunc_call test,
from Yipeng Zou.
8) Small cleanup to remove unused declaration bpf_link_new_file,
from Yue Haibing.
9) Small typo fix to bpftool's perf help message, from Daniel T. Lee.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
selftests/bpf: Add mptcpify test
selftests/bpf: Fix error checks of mptcp open_and_load
selftests/bpf: Add two mptcp netns helpers
bpf: Add update_socket_protocol hook
bpftool: Implement link show support for xdp
bpftool: Implement link show support for tcx
selftests/bpf: Add selftest for fill_link_info
bpf: Fix uninitialized symbol in bpf_perf_link_fill_kprobe()
net: Fix slab-out-of-bounds in inet[6]_steal_sock
bpf: Document struct bpf_struct_ops fields
bpf: Support default .validate() and .update() behavior for struct_ops links
selftests/bpf: Add various more tcx test cases
selftests/bpf: Clean up fmod_ret in bench_rename test script
selftests/bpf: Fix repeat option when kfunc_call verification fails
libbpf: Set close-on-exec flag on gzopen
bpftool: fix perf help message
bpf: Remove unused declaration bpf_link_new_file()
====================
Link: https://lore.kernel.org/r/20230816212840.1539-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Geliang Tang says:
====================
As is described in the "How to use MPTCP?" section in MPTCP wiki [1]:
"Your app should create sockets with IPPROTO_MPTCP as the proto:
( socket(AF_INET, SOCK_STREAM, IPPROTO_MPTCP); ). Legacy apps can be
forced to create and use MPTCP sockets instead of TCP ones via the
mptcpize command bundled with the mptcpd daemon."
But the mptcpize (LD_PRELOAD technique) command has some limitations
[2]:
- it doesn't work if the application is not using libc (e.g. GoLang
apps)
- in some envs, it might not be easy to set env vars / change the way
apps are launched, e.g. on Android
- mptcpize needs to be launched with all apps that want MPTCP: we could
have more control from BPF to enable MPTCP only for some apps or all the
ones of a netns or a cgroup, etc.
- it is not in BPF, we cannot talk about it at netdev conf.
So this patchset attempts to use BPF to implement functions similer to
mptcpize.
The main idea is to add a hook in sys_socket() to change the protocol id
from IPPROTO_TCP (or 0) to IPPROTO_MPTCP.
[1]
https://github.com/multipath-tcp/mptcp_net-next/wiki
[2]
https://github.com/multipath-tcp/mptcp_net-next/issues/79
v14:
- Use getsockopt(MPTCP_INFO) to verify mptcp protocol intead of using
nstat command.
v13:
- drop "Use random netns name for mptcp" patch.
v12:
- update diag_* log of update_socket_protocol.
- add 'ip netns show' after 'ip netns del' to check if there is
a test did not clean up its netns.
- return libbpf_get_error() instead of -EIO for the error from
open_and_load().
- Use getsockopt(SOL_PROTOCOL) to verify mptcp protocol intead of
using 'ss -tOni'.
v11:
- add comments about outputs of 'ss' and 'nstat'.
- use "err = verify_mptcpify()" instead of using =+.
v10:
- drop "#ifdef CONFIG_BPF_JIT".
- include vmlinux.h and bpf_tracing_net.h to avoid defining some
macros.
- drop unneeded checks for mptcp.
v9:
- update comment for 'update_socket_protocol'.
v8:
- drop the additional checks on the 'protocol' value after the
'update_socket_protocol()' call.
v7:
- add __weak and __diag_* for update_socket_protocol.
v6:
- add update_socket_protocol.
v5:
- add bpf_mptcpify helper.
v4:
- use lsm_cgroup/socket_create
v3:
- patch 8: char cmd[128]; -> char cmd[256];
v2:
- Fix build selftests errors reported by CI
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/79
====================
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Implement a new test program mptcpify: if the family is AF_INET or
AF_INET6, the type is SOCK_STREAM, and the protocol ID is 0 or
IPPROTO_TCP, set it to IPPROTO_MPTCP. It will be hooked in
update_socket_protocol().
Extend the MPTCP test base, add a selftest test_mptcpify() for the
mptcpify case. Open and load the mptcpify test prog to mptcpify the
TCP sockets dynamically, then use start_server() and connect_to_fd()
to create a TCP socket, but actually what's created is an MPTCP
socket, which can be verified through 'getsockopt(SOL_PROTOCOL)'
and 'getsockopt(MPTCP_INFO)'.
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Link: https://lore.kernel.org/r/364e72f307e7bb38382ec7442c182d76298a9c41.1692147782.git.geliang.tang@suse.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Return libbpf_get_error(), instead of -EIO, for the error from
mptcp_sock__open_and_load().
Load success means prog_fd and map_fd are always valid. So drop these
unneeded ASSERT_GE checks for them in mptcp run_test().
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Link: https://lore.kernel.org/r/db5fcb93293df9ab173edcbaf8252465b80da6f2.1692147782.git.geliang.tang@suse.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Add two netns helpers for mptcp tests: create_netns() and
cleanup_netns(). Use them in test_base().
These new helpers will be re-used in the following commits
introducing new tests.
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Link: https://lore.kernel.org/r/7506371fb6c417b401cc9d7365fe455754f4ba3f.1692147782.git.geliang.tang@suse.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Add a hook named update_socket_protocol in __sys_socket(), for bpf
progs to attach to and update socket protocol. One user case is to
force legacy TCP apps to create and use MPTCP sockets instead of
TCP ones.
Define a fmod_ret set named bpf_mptcp_fmodret_ids, add the hook
update_socket_protocol into this set, and register it in
bpf_mptcp_kfunc_init().
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/79
Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Link: https://lore.kernel.org/r/ac84be00f97072a46f8a72b4e2be46cbb7fa5053.1692147782.git.geliang.tang@suse.com
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Add support to dump XDP link information to bpftool. This reuses the
recently added show_link_ifindex_{plain,json}(). The XDP link info only
exposes the ifindex.
Below shows an example link dump output, and a cgroup link is included
for comparison, too:
# bpftool link
[...]
10: cgroup prog 2466
cgroup_id 1 attach_type cgroup_inet6_post_bind
[...]
16: xdp prog 2477
ifindex enp5s0(3)
[...]
Equivalent json output:
# bpftool link --json
[...]
{
"id": 10,
"type": "cgroup",
"prog_id": 2466,
"cgroup_id": 1,
"attach_type": "cgroup_inet6_post_bind"
},
[...]
{
"id": 16,
"type": "xdp",
"prog_id": 2477,
"devname": "enp5s0",
"ifindex": 3
}
[...]
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Link: https://lore.kernel.org/r/20230816095651.10014-2-daniel@iogearbox.net
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Add support to dump tcx link information to bpftool. This adds a
common helper show_link_ifindex_{plain,json}() which can be reused
also for other link types. The plain text and json device output is
the same format as in bpftool net dump.
Below shows an example link dump output along with a cgroup link
for comparison:
# bpftool link
[...]
10: cgroup prog 1977
cgroup_id 1 attach_type cgroup_inet6_post_bind
[...]
13: tcx prog 2053
ifindex enp5s0(3) attach_type tcx_ingress
14: tcx prog 2080
ifindex enp5s0(3) attach_type tcx_egress
[...]
Equivalent json output:
# bpftool link --json
[...]
{
"id": 10,
"type": "cgroup",
"prog_id": 1977,
"cgroup_id": 1,
"attach_type": "cgroup_inet6_post_bind"
},
[...]
{
"id": 13,
"type": "tcx",
"prog_id": 2053,
"devname": "enp5s0",
"ifindex": 3,
"attach_type": "tcx_ingress"
},
{
"id": 14,
"type": "tcx",
"prog_id": 2080,
"devname": "enp5s0",
"ifindex": 3,
"attach_type": "tcx_egress"
}
[...]
Suggested-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin@isovalent.com>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Link: https://lore.kernel.org/r/20230816095651.10014-1-daniel@iogearbox.net
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
Add selftest for the fill_link_info of uprobe, kprobe and tracepoint.
The result:
$ tools/testing/selftests/bpf/test_progs --name=fill_link_info
#79/1 fill_link_info/kprobe_link_info:OK
#79/2 fill_link_info/kretprobe_link_info:OK
#79/3 fill_link_info/kprobe_invalid_ubuff:OK
#79/4 fill_link_info/tracepoint_link_info:OK
#79/5 fill_link_info/uprobe_link_info:OK
#79/6 fill_link_info/uretprobe_link_info:OK
#79/7 fill_link_info/kprobe_multi_link_info:OK
#79/8 fill_link_info/kretprobe_multi_link_info:OK
#79/9 fill_link_info/kprobe_multi_invalid_ubuff:OK
#79 fill_link_info:OK
Summary: 1/9 PASSED, 0 SKIPPED, 0 FAILED
The test case for kprobe_multi won't be run on aarch64, as it is not
supported.
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20230813141900.1268-3-laoar.shao@gmail.com
The commit 1b715e1b0e ("bpf: Support ->fill_link_info for perf_event") leads
to the following Smatch static checker warning:
kernel/bpf/syscall.c:3416 bpf_perf_link_fill_kprobe()
error: uninitialized symbol 'type'.
That can happens when uname is NULL. So fix it by verifying the uname when we
really need to fill it.
Fixes: 1b715e1b0e ("bpf: Support ->fill_link_info for perf_event")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yonghong.song@linux.dev>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Closes: https://lore.kernel.org/bpf/85697a7e-f897-4f74-8b43-82721bebc462@kili.mountain
Link: https://lore.kernel.org/bpf/20230813141900.1268-2-laoar.shao@gmail.com
Kui-Feng Lee says:
====================
Remove expired routes with a separated list of routes.
FIB6 GC walks trees of fib6_tables to remove expired routes. Walking a tree
can be expensive if the number of routes in a table is big, even if most of
them are permanent. Checking routes in a separated list of routes having
expiration will avoid this potential issue.
Background
==========
The size of a Linux IPv6 routing table can become a big problem if not
managed appropriately. Now, Linux has a garbage collector to remove
expired routes periodically. However, this may lead to a situation in
which the routing path is blocked for a long period due to an
excessive number of routes.
For example, years ago, there is a commit c7bb4b8903 ("ipv6: tcp:
drop silly ICMPv6 packet too big messages"). The root cause is that
malicious ICMPv6 packets were sent back for every small packet sent to
them. These packets add routes with an expiration time that prompts
the GC to periodically check all routes in the tables, including
permanent ones.
Why Route Expires
=================
Users can add IPv6 routes with an expiration time manually. However,
the Neighbor Discovery protocol may also generate routes that can
expire. For example, Router Advertisement (RA) messages may create a
default route with an expiration time. [RFC 4861] For IPv4, it is not
possible to set an expiration time for a route, and there is no RA, so
there is no need to worry about such issues.
Create Routes with Expires
==========================
You can create routes with expires with the command.
For example,
ip -6 route add 2001:b000:591::3 via fe80::5054:ff:fe12:3457 \
dev enp0s3 expires 30
The route that has been generated will be deleted automatically in 30
seconds.
GC of FIB6
==========
The function called fib6_run_gc() is responsible for performing
garbage collection (GC) for the Linux IPv6 stack. It checks for the
expiration of every route by traversing the trees of routing
tables. The time taken to traverse a routing table increases with its
size. Holding the routing table lock during traversal is particularly
undesirable. Therefore, it is preferable to keep the lock for the
shortest possible duration.
Solution
========
The cause of the issue is keeping the routing table locked during the
traversal of large trees. To solve this problem, we can create a separate
list of routes that have expiration. This will prevent GC from checking
permanent routes.
Result
======
We conducted a test to measure the execution times of fib6_gc_timer_cb()
and observed that it enhances the GC of FIB6. During the test, we added
permanent routes with the following numbers: 1000, 3000, 6000, and
9000. Additionally, we added a route with an expiration time.
Here are the average execution times for the kernel without the patch.
- 120020 ns with 1000 permanent routes
- 308920 ns with 3000 ...
- 581470 ns with 6000 ...
- 855310 ns with 9000 ...
The kernel with the patch consistently takes around 14000 ns to execute,
regardless of the number of permanent routes that are installed.
Major changes from v7:
- Fix warings raised by the patchwork.
Major changes from v6:
- Remove unnecessary check of tb6 in fib6_clean_expires_locked().
- Use ib6_clean_expires_locked() instead in fib6_purge_rt().
Major changes from v5:
- Change the order of adding new routes to the GC list and starting
GC timer.
- Remove time measurements from the test case.
- Stop forcing GC flush.
Major changes from v4:
- Detect existence of 'strace' in the test case.
Major changes from v3:
- Fix the type of arg according to feedback.
- Add 1k temporary routes and 5K permanent routes in the test case.
Measure time spending on GC with strace.
Major changes from v2:
- Remove unnecessary and incorrect sysctl restoring in the test case.
Major changes from v1:
- Moved gc_link to avoid creating a hole in fib6_info.
- Moved fib6_set_expires*() and fib6_clean_expires*() to the header
file and inlined. And removed duplicated lines.
- Added a test case.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add 1000 IPv6 routes with expiration time (w/ and w/o additional 5000
permanet routes in the background.) Wait for a few seconds to make sure
they are removed correctly.
The expected output of the test looks like the following example.
> Fib6 garbage collection test
> TEST: ipv6 route garbage collection [ OK ]
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
FIB6 GC walks trees of fib6_tables to remove expired routes. Walking a tree
can be expensive if the number of routes in a table is big, even if most of
them are permanent. Checking routes in a separated list of routes having
expiration will avoid this potential issue.
Signed-off-by: Kui-Feng Lee <thinker.li@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
On some I219 devices, ethernet cable plugging detection only works once
from PCI D3 state. Subsequent cable plugging does set PME bit correctly,
but device still doesn't get woken up.
Since I219 connects to the root complex directly, it relies on platform
firmware (ACPI) to wake it up. In this case, the GPE from _PRW only
works for first cable plugging but fails to notify the driver for
subsequent plugging events.
The issue was originally found on CNP, but the same issue can be found
on ADL too. So workaround the issue by continuing use PME poll after
first ACPI wake. As PME poll is always used, the runtime suspend
restriction for CNP can also be removed.
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Tested-by: Naama Meir <naamax.meir@linux.intel.com>
Acked-by: Sasha Neftin <sasha.neftin@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now there are two indicators of socket memory pressure sit inside
struct mem_cgroup, socket_pressure and tcpmem_pressure, indicating
memory reclaim pressure in memcg->memory and ->tcpmem respectively.
When in legacy mode (cgroupv1), the socket memory is charged into
->tcpmem which is independent of ->memory, so socket_pressure has
nothing to do with socket's pressure at all. Things could be worse
by taking socket_pressure into consideration in legacy mode, as a
pressure in ->memory can lead to premature reclamation/throttling
in socket.
While for the default mode (cgroupv2), the socket memory is charged
into ->memory, and ->tcpmem/->tcpmem_pressure are simply not used.
So {socket,tcpmem}_pressure are only used in default/legacy mode
respectively for indicating socket memory pressure. This patch fixes
the pieces of code that make mixed use of both.
Fixes: 8e8ae64524 ("mm: memcontrol: hook up vmpressure to socket pressure")
Signed-off-by: Abel Wu <wuyun.abel@bytedance.com>
Acked-by: Shakeel Butt <shakeelb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Take over maintainership of the nfp driver from Simon as he
is moving away from Corigine.
Signed-off-by: Louis Peens <louis.peens@corigine.com>
Acked-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds MQPRIO Qdisc offload in full 'channel' mode which allows
not only setting up pri:tc mapping, but also configuring TX shapers on
external port FIFOs. The K3 CPSW MQPRIO Qdisc offload is expected to work
with VLAN/priority tagged packets. Non-tagged packets have to be mapped
only to TC0.
- TX traffic classes must be rated starting from TC that has highest
priority and with no gaps
- Traffic classes are used starting from 0, that has highest priority
- min_rate defines Committed Information Rate (guaranteed)
- max_rate defines Excess Information Rate (non guaranteed) and offloaded
as (max_rate[i] - tcX_min_rate[i])
- VLAN/priority tagged packets mapped to TC0 will exit switch with VLAN tag
priority 0
The configuration example:
ethtool -L eth1 tx 5
ethtool --set-priv-flags eth1 p0-rx-ptype-rrobin off
tc qdisc add dev eth1 parent root handle 100: mqprio num_tc 3 \
map 0 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 \
queues 1@0 1@1 1@2 hw 1 mode channel \
shaper bw_rlimit min_rate 0 100mbit 200mbit max_rate 0 101mbit 202mbit
tc qdisc replace dev eth2 handle 100: parent root mqprio num_tc 1 \
map 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 queues 1@0 hw 1
ip link add link eth1 name eth1.100 type vlan id 100
ip link set eth1.100 type vlan egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
In the above example two ports share the same TX CPPI queue 0 for low
priority traffic. 3 traffic classes are defined for eth1 and mapped to:
TC0 - low priority, TX CPPI queue 0 -> ext Port 1 fifo0, no rate limit
TC1 - prio 2, TX CPPI queue 1 -> ext Port 1 fifo1, CIR=100Mbit/s, EIR=1Mbit/s
TC2 - prio 3, TX CPPI queue 2 -> ext Port 1 fifo2, CIR=200Mbit/s, EIR=2Mbit/s
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Roger Quadros <rogerq@kernel.org>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet says:
====================
inet: socket lock and data-races avoidance
In this series, I converted 20 bits in "struct inet_sock" and made
them truly atomic.
This allows to implement many IP_ socket options in a lockless
fashion (no need to acquire socket lock), and fixes data-races
that were showing up in various KCSAN reports.
I also took care of IP_TTL/IP_MINTTL, but left few other options
for another series.
v4: Rebased after recent mptcp changes.
Added Reviewed-by: tags from Simon (thanks !)
v3: fixed patch 7, feedback from build bot about ipvs set_mcast_loop()
v2: addressed a feedback from a build bot in patch 9 by removing
unused issk variable in mptcp_setsockopt_sol_ip_set_transparent()
Added Acked-by: tags from Soheil (thanks !)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
inet->min_ttl is already read with READ_ONCE().
Implementing IP_MINTTL socket option set/read
without holding the socket lock is easy.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel T. Lee