Quentin Monnet says:
====================
Hi,
This set is an update to the documentation for the BPF helper functions in
the UAPI header bpf.h, used to generate the bpf-helpers(7) man page.
First patch contains fixes to the current documentation. The second patch
adds documentation for the two helpers bpf_spin_lock() and
bpf_spin_unlock(). The last patch simply reports the changes to the version
of that header in tools/.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Synchronise the bpf.h header under tools, to report the latest fixes and
additions to the documentation for the BPF helpers.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add documentation for the BPF spinlock-related helpers to the doc in
bpf.h. I added the constraints and restrictions coming with the use of
spinlocks for BPF: not all of it is directly related to the use of the
helper, but I thought it would be nice for users to find them in the man
page.
This list of restrictions is nearly a verbatim copy of the list in
Alexei's commit log for those helpers.
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Another round of minor fixes for the documentation of the BPF helpers
located in the UAPI bpf.h header file. Changes include:
- Moving around description of some helpers, to keep the descriptions in
the same order as helpers are declared (bpf_map_push_elem(), leftover
from commit 90b1023f68 ("bpf: fix documentation for eBPF helpers"),
bpf_rc_keydown(), and bpf_skb_ancestor_cgroup_id()).
- Fixing typos ("contex" -> "context").
- Harmonising return types ("void* " -> "void *", "uint64_t" -> "u64").
- Addition of the "bpf_" prefix to bpf_get_storage().
- Light additions of RST markup on some keywords.
- Empty line deletion between description and return value for
bpf_tcp_sock().
- Edit for the description for bpf_skb_ecn_set_ce() (capital letters,
acronym expansion, no effect if ECT not set, more details on return
value).
Signed-off-by: Quentin Monnet <quentin.monnet@netronome.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Andrii Nakryiko says:
====================
This patchset adds ability to resolve forward-declared enums into their
corresponding full enum definition types during type deduplication,
eliminating one of the reasons for having duplicated graphs of types.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch adds test verifying new btf_dedup logic of resolving
forward-declared enums.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
GCC and clang support enum forward declarations as an extension. Such
forward-declared enums will be represented as normal BTF_KIND_ENUM types with
vlen=0. This patch adds ability to resolve such enums to their corresponding
fully defined enums. This helps to avoid duplicated BTF type graphs which only
differ by some types referencing forward-declared enum vs full enum.
One such example in kernel is enum irqchip_irq_state, defined in
include/linux/interrupt.h and forward-declared in include/linux/irq.h. This
causes entire struct task_struct and all referenced types to be duplicated in
btf_dedup output. This patch eliminates such duplication cases.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Martin KaFai Lau says:
====================
This set addresses issue about accessing invalid
ptr returned from bpf_tcp_sock() and bpf_sk_fullsock()
after bpf_sk_release().
v4:
- Tried the one "id" approach. It does not work well and the reason is in
the Patch 1 commit message.
- Rename refcount_id to ref_obj_id.
- With ref_obj_id, resetting reg->id to 0 is fine in mark_ptr_or_null_reg()
because ref_obj_id is passed to release_reference() instead of reg->id.
- Also reset reg->ref_obj_id in mark_ptr_or_null_reg() when is_null == true
- sk_to_full_sk() is removed from bpf_sk_fullsock() and bpf_tcp_sock().
- bpf_get_listener_sock() is added to do sk_to_full_sk() in Patch 2.
- If tp is from bpf_tcp_sock(sk) and sk is a refcounted ptr,
bpf_sk_release(tp) is also allowed.
v3:
- reset reg->refcount_id for the is_null case in mark_ptr_or_null_reg()
v2:
- Remove refcount_id arg from release_reference() because
id == refcount_id
- Add a WARN_ON_ONCE to mark_ptr_or_null_regs() to catch
an internal verifier bug.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This patch adds an example in using the new helper
bpf_get_listener_sock().
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Adding verifier tests to ensure the ptr returned from bpf_tcp_sock() and
bpf_sk_fullsock() cannot be accessed after bpf_sk_release() is called.
A few of the tests are derived from a reproducer test by Lorenz Bauer.
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add a new helper "struct bpf_sock *bpf_get_listener_sock(struct bpf_sock *sk)"
which returns a bpf_sock in TCP_LISTEN state. It will trace back to
the listener sk from a request_sock if possible. It returns NULL
for all other cases.
No reference is taken because the helper ensures the sk is
in SOCK_RCU_FREE (where the TCP_LISTEN sock should be in).
Hence, bpf_sk_release() is unnecessary and the verifier does not
allow bpf_sk_release(listen_sk) to be called either.
The following is also allowed because the bpf_prog is run under
rcu_read_lock():
sk = bpf_sk_lookup_tcp();
/* if (!sk) { ... } */
listen_sk = bpf_get_listener_sock(sk);
/* if (!listen_sk) { ... } */
bpf_sk_release(sk);
src_port = listen_sk->src_port; /* Allowed */
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Lorenz Bauer [thanks!] reported that a ptr returned by bpf_tcp_sock(sk)
can still be accessed after bpf_sk_release(sk).
Both bpf_tcp_sock() and bpf_sk_fullsock() have the same issue.
This patch addresses them together.
A simple reproducer looks like this:
sk = bpf_sk_lookup_tcp();
/* if (!sk) ... */
tp = bpf_tcp_sock(sk);
/* if (!tp) ... */
bpf_sk_release(sk);
snd_cwnd = tp->snd_cwnd; /* oops! The verifier does not complain. */
The problem is the verifier did not scrub the register's states of
the tcp_sock ptr (tp) after bpf_sk_release(sk).
[ Note that when calling bpf_tcp_sock(sk), the sk is not always
refcount-acquired. e.g. bpf_tcp_sock(skb->sk). The verifier works
fine for this case. ]
Currently, the verifier does not track if a helper's return ptr (in REG_0)
is "carry"-ing one of its argument's refcount status. To carry this info,
the reg1->id needs to be stored in reg0.
One approach was tried, like "reg0->id = reg1->id", when calling
"bpf_tcp_sock()". The main idea was to avoid adding another "ref_obj_id"
for the same reg. However, overlapping the NULL marking and ref
tracking purpose in one "id" does not work well:
ref_sk = bpf_sk_lookup_tcp();
fullsock = bpf_sk_fullsock(ref_sk);
tp = bpf_tcp_sock(ref_sk);
if (!fullsock) {
bpf_sk_release(ref_sk);
return 0;
}
/* fullsock_reg->id is marked for NOT-NULL.
* Same for tp_reg->id because they have the same id.
*/
/* oops. verifier did not complain about the missing !tp check */
snd_cwnd = tp->snd_cwnd;
Hence, a new "ref_obj_id" is needed in "struct bpf_reg_state".
With a new ref_obj_id, when bpf_sk_release(sk) is called, the verifier can
scrub all reg states which has a ref_obj_id match. It is done with the
changes in release_reg_references() in this patch.
While fixing it, sk_to_full_sk() is removed from bpf_tcp_sock() and
bpf_sk_fullsock() to avoid these helpers from returning
another ptr. It will make bpf_sk_release(tp) possible:
sk = bpf_sk_lookup_tcp();
/* if (!sk) ... */
tp = bpf_tcp_sock(sk);
/* if (!tp) ... */
bpf_sk_release(tp);
A separate helper "bpf_get_listener_sock()" will be added in a later
patch to do sk_to_full_sk().
Misc change notes:
- To allow bpf_sk_release(tp), the arg of bpf_sk_release() is changed
from ARG_PTR_TO_SOCKET to ARG_PTR_TO_SOCK_COMMON. ARG_PTR_TO_SOCKET
is removed from bpf.h since no helper is using it.
- arg_type_is_refcounted() is renamed to arg_type_may_be_refcounted()
because ARG_PTR_TO_SOCK_COMMON is the only one and skb->sk is not
refcounted. All bpf_sk_release(), bpf_sk_fullsock() and bpf_tcp_sock()
take ARG_PTR_TO_SOCK_COMMON.
- check_refcount_ok() ensures is_acquire_function() cannot take
arg_type_may_be_refcounted() as its argument.
- The check_func_arg() can only allow one refcount-ed arg. It is
guaranteed by check_refcount_ok() which ensures at most one arg can be
refcounted. Hence, it is a verifier internal error if >1 refcount arg
found in check_func_arg().
- In release_reference(), release_reference_state() is called
first to ensure a match on "reg->ref_obj_id" can be found before
scrubbing the reg states with release_reg_references().
- reg_is_refcounted() is no longer needed.
1. In mark_ptr_or_null_regs(), its usage is replaced by
"ref_obj_id && ref_obj_id == id" because,
when is_null == true, release_reference_state() should only be
called on the ref_obj_id obtained by a acquire helper (i.e.
is_acquire_function() == true). Otherwise, the following
would happen:
sk = bpf_sk_lookup_tcp();
/* if (!sk) { ... } */
fullsock = bpf_sk_fullsock(sk);
if (!fullsock) {
/*
* release_reference_state(fullsock_reg->ref_obj_id)
* where fullsock_reg->ref_obj_id == sk_reg->ref_obj_id.
*
* Hence, the following bpf_sk_release(sk) will fail
* because the ref state has already been released in the
* earlier release_reference_state(fullsock_reg->ref_obj_id).
*/
bpf_sk_release(sk);
}
2. In release_reg_references(), the current reg_is_refcounted() call
is unnecessary because the id check is enough.
- The type_is_refcounted() and type_is_refcounted_or_null()
are no longer needed also because reg_is_refcounted() is removed.
Fixes: 655a51e536 ("bpf: Add struct bpf_tcp_sock and BPF_FUNC_tcp_sock")
Reported-by: Lorenz Bauer <lmb@cloudflare.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
In xsk_socket__create(), the libbpf_flags field was not checked for
setting currently unused/unknown flags. This patch fixes that by
returning -EINVAL if the user has set any flag that is not in use at
this point in time.
Fixes: 1cad078842 ("libbpf: add support for using AF_XDP sockets")
Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Reviewed-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
The test_progs subtests, test_spin_lock() and test_map_lock(),
requires BTF present to run successfully.
Currently, when BTF failed to load, test_progs will segfault,
$ ./test_progs
...
12: (bf) r1 = r8
13: (85) call bpf_spin_lock#93
map 'hash_map' has to have BTF in order to use bpf_spin_lock
libbpf: -- END LOG --
libbpf: failed to load program 'map_lock_demo'
libbpf: failed to load object './test_map_lock.o'
test_map_lock:bpf_prog_load errno 13
Segmentation fault
The segfault is caused by uninitialized variable "obj", which
is used in bpf_object__close(obj), when bpf prog failed to load.
Initializing variable "obj" to NULL in two occasions fixed the problem.
$ ./test_progs
...
Summary: 219 PASSED, 2 FAILED
Fixes: b4d4556c32 ("selftests/bpf: add bpf_spin_lock verifier tests")
Fixes: ba72a7b4ba ("selftests/bpf: test for BPF_F_LOCK")
Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
This patch splits and cleans up error handling logic for loading BTF data.
Previously, if BTF data was parsed successfully, but failed to load into
kernel, we'd report nonsensical error code, instead of error returned from
btf__load(). Now btf__new() and btf__load() are handled separately with proper
cleanup and warning reporting.
Fixes: d29d87f7e6 ("btf: separate btf creation and loading")
Reported-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Same reasons than the ones explained in commit 4179cb5a4c
("vxlan: test dev->flags & IFF_UP before calling netif_rx()")
netif_rx() or gro_cells_receive() must be called under a strict contract.
At device dismantle phase, core networking clears IFF_UP
and flush_all_backlogs() is called after rcu grace period
to make sure no incoming packet might be in a cpu backlog
and still referencing the device.
A similar protocol is used for gro_cells infrastructure, as
gro_cells_destroy() will be called only after a full rcu
grace period is observed after IFF_UP has been cleared.
Most drivers call netif_rx() from their interrupt handler,
and since the interrupts are disabled at device dismantle,
netif_rx() does not have to check dev->flags & IFF_UP
Virtual drivers do not have this guarantee, and must
therefore make the check themselves.
Otherwise we risk use-after-free and/or crashes.
Fixes: d342894c5d ("vxlan: virtual extensible lan")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In case ioremap fails, the fix returns -ENOMEM to avoid NULL
pointer dereference.
Signed-off-by: Kangjie Lu <kjlu@umn.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
The commit bfcb79fca1
("PCI/ERR: Run error recovery callbacks for all affected devices")
affected the non-fatal error recovery logic for the HNS and RDMA devices.
This is because each HNS PF under PCIe bus receive callbacks
from the AER driver when an error is reported for one of the PF.
This causes unwanted PF resets because
the HNS decides which PF to reset based on the reset type set.
The HNS error handling code sets the reset type based on the hw error
type detected.
This patch provides fix for the above issue for the recovery of
the hw errors in the HNS and RDMA devices.
This patch needs backporting to the kernel v5.0+
Fixes: 332fbf5765 ("net: hns3: add handling of hw ras errors using new set of commands")
Reported-by: Xiaofei Tan <tanxiaofei@huawei.com>
Signed-off-by: Shiju Jose <shiju.jose@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As Eric Dumazet said, "We do not have a way to tell if the req was ever
inserted in a hash table, so better play safe.".
Let's remove this comment, so that nobody will be tempted to drop the
WARN_ON_ONCE() line.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When building with -Wsometimes-uninitialized, Clang warns:
drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c:111:2: error: variable
'ns' is used uninitialized whenever 'if' condition is false
[-Werror,-Wsometimes-uninitialized]
drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c:111:2: error: variable
'ns' is used uninitialized whenever '&&' condition is false
[-Werror,-Wsometimes-uninitialized]
Clang is concerned with the use of stmmac_do_void_callback (which
stmmac_get_systime wraps), as it may fail to initialize these values if
the if condition was ever false (meaning the callback doesn't exist).
It's not wrong because the callback is what initializes ns. While it's
unlikely that the callback is going to disappear at some point and make
that condition false, we can easily avoid this warning by zero
initializing the variable.
Link: https://github.com/ClangBuiltLinux/linux/issues/384
Fixes: df10317085 ("net: stmmac: Avoid sometimes uninitialized Clang warnings")
Suggested-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Björn Töpel
Alexei Starovoitov
Quentin Monnet
Andrii Nakryiko
Martin KaFai Lau
Magnus Karlsson
Yonghong Song
Eric Dumazet
Kangjie Lu
Shiju Jose
Guillaume Nault
Nathan Chancellor