Add documentation for BPF_MAP_TYPE_HASH including kernel version
introduced, usage and examples. Document BPF_MAP_TYPE_PERCPU_HASH,
BPF_MAP_TYPE_LRU_HASH and BPF_MAP_TYPE_LRU_PERCPU_HASH variations.
Note that this file is included in the BPF documentation by the glob in
Documentation/bpf/maps.rst
v3:
Fix typos reported by Stanislav Fomichev and Yonghong Song.
Add note about iteration and deletion as requested by Yonghong Song.
v2:
Describe memory allocation semantics as suggested by Stanislav Fomichev.
Fix u64 typo reported by Stanislav Fomichev.
Cut down usage examples to only show usage in context.
Updated patch description to follow style recommendation, reported by
Bagas Sanjaya.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20220718125847.1390-1-donald.hunter@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add test validating that libbpf adjusts (and reflects adjusted) ringbuf
size early, before bpf_object is loaded. Also make sure we can't
successfully resize ringbuf map after bpf_object is loaded.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20220715230952.2219271-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Make libbpf adjust RINGBUF map size (rounding it up to closest power-of-2
of page_size) more eagerly: during open phase when initializing the map
and on explicit calls to bpf_map__set_max_entries().
Such approach allows user to check actual size of BPF ringbuf even
before it's created in the kernel, but also it prevents various edge
case scenarios where BPF ringbuf size can get out of sync with what it
would be in kernel. One of them (reported in [0]) is during an attempt
to pin/reuse BPF ringbuf.
Move adjust_ringbuf_sz() helper closer to its first actual use. The
implementation of the helper is unchanged.
Also make detection of whether bpf_object is already loaded more robust
by checking obj->loaded explicitly, given that map->fd can be < 0 even
if bpf_object is already loaded due to ability to disable map creation
with bpf_map__set_autocreate(map, false).
[0] Closes: https://github.com/libbpf/libbpf/pull/530
Fixes: 0087a681fa ("libbpf: Automatically fix up BPF_MAP_TYPE_RINGBUF size, if necessary")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20220715230952.2219271-1-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Andrii Nakryiko says:
====================
Fix 32-bit overflow in value pointer calculations in BPF array map. And then
raise obsolete limit on array map value size. Add selftest making sure this is
working as intended.
v1->v2:
- fix broken patch #1 (no mask_index use in helper, as stated in commit
message; and add missing semicolon).
====================
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add a simple big 16MB array and validate access to the very last byte of
it to make sure that kernel supports > KMALLOC_MAX_SIZE value_size for
BPF array maps (which are backing .bss in this case).
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220715053146.1291891-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Syscall-side map_lookup_elem() and map_update_elem() used to use
kmalloc() to allocate temporary buffers of value_size, so
KMALLOC_MAX_SIZE limit on value_size made sense to prevent creation of
array map that won't be accessible through syscall interface.
But this limitation since has been lifted by relying on kvmalloc() in
syscall handling code. So remove KMALLOC_MAX_SIZE, which among other
things means that it's possible to have BPF global variable sections
(.bss, .data, .rodata) bigger than 8MB now. Keep the sanity check to
prevent trivial overflows like round_up(map->value_size, 8) and restrict
value size to <= INT_MAX (2GB).
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220715053146.1291891-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
BPF_MAP_TYPE_ARRAY is rounding value_size to closest multiple of 8 and
stores that as array->elem_size for various memory allocations and
accesses.
But the code tends to re-calculate round_up(map->value_size, 8) in
multiple places instead of using array->elem_size. Cleaning this up and
making sure we always use array->size to avoid duplication of this
(admittedly simple) logic for consistency.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220715053146.1291891-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
If BPF array map is bigger than 4GB, element pointer calculation can
overflow because both index and elem_size are u32. Fix this everywhere
by forcing 64-bit multiplication. Extract this formula into separate
small helper and use it consistently in various places.
Speculative-preventing formula utilizing index_mask trick is left as is,
but explicit u64 casts are added in both places.
Fixes: c85d69135a ("bpf: move memory size checks to bpf_map_charge_init()")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220715053146.1291891-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The vlen bits in the BTF type of kind BTF_KIND_FUNC are used to convey the
linkage information for functions. The Linux kernel only supports
linkage values of BTF_FUNC_STATIC and BTF_FUNC_GLOBAL at this time.
Signed-off-by: Indu Bhagat <indu.bhagat@oracle.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/r/20220714223310.1140097-1-indu.bhagat@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
This particular ones is about having the following:
CONFIG_BPF_LSM=y
# CONFIG_CGROUP_BPF is not set
Also, add __maybe_unused to the args for the !CONFIG_NET cases.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20220714185404.3647772-1-sdf@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Andrii Nakryiko says:
====================
Add SEC("ksyscall")/SEC("kretsyscall") sections and corresponding
bpf_program__attach_ksyscall() API that simplifies tracing kernel syscalls
through kprobe mechanism. Kprobing syscalls isn't trivial due to varying
syscall handler names in the kernel and various ways syscall argument are
passed, depending on kernel architecture and configuration. SEC("ksyscall")
allows user to not care about such details and just get access to syscall
input arguments, while libbpf takes care of necessary feature detection logic.
There are still more quirks that are not straightforward to hide completely
(see comments about mmap(), clone() and compat syscalls), so in such more
advanced scenarios user might need to fall back to plain SEC("kprobe")
approach, but for absolute majority of users SEC("ksyscall") is a big
improvement.
As part of this patch set libbpf adds two more virtual __kconfig externs, in
addition to existing LINUX_KERNEL_VERSION: LINUX_HAS_BPF_COOKIE and
LINUX_HAS_SYSCALL_WRAPPER, which let's libbpf-provided BPF-side code minimize
external dependencies and assumptions and let's user-space part of libbpf to
perform all the feature detection logic. This benefits USDT support code,
which now doesn't depend on BPF CO-RE for its functionality.
v1->v2:
- normalize extern variable-related warn and debug message formats (Alan);
rfc->v1:
- drop dependency on kallsyms and speed up SYSCALL_WRAPPER detection (Alexei);
- drop dependency on /proc/config.gz in bpf_tracing.h (Yaniv);
- add doc comment and ephasize mmap(), clone() and compat quirks that are
not supported (Ilya);
- use mechanism similar to LINUX_KERNEL_VERSION to also improve USDT code.
====================
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Convert few selftest that used plain SEC("kprobe") with arch-specific
syscall wrapper prefix to ksyscall/kretsyscall and corresponding
BPF_KSYSCALL macro. test_probe_user.c is especially benefiting from this
simplification.
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220714070755.3235561-6-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add SEC("ksyscall")/SEC("ksyscall/<syscall_name>") and corresponding
kretsyscall variants (for return kprobes) to allow users to kprobe
syscall functions in kernel. These special sections allow to ignore
complexities and differences between kernel versions and host
architectures when it comes to syscall wrapper and corresponding
__<arch>_sys_<syscall> vs __se_sys_<syscall> differences, depending on
whether host kernel has CONFIG_ARCH_HAS_SYSCALL_WRAPPER (though libbpf
itself doesn't rely on /proc/config.gz for detecting this, see
BPF_KSYSCALL patch for how it's done internally).
Combined with the use of BPF_KSYSCALL() macro, this allows to just
specify intended syscall name and expected input arguments and leave
dealing with all the variations to libbpf.
In addition to SEC("ksyscall+") and SEC("kretsyscall+") add
bpf_program__attach_ksyscall() API which allows to specify syscall name
at runtime and provide associated BPF cookie value.
At the moment SEC("ksyscall") and bpf_program__attach_ksyscall() do not
handle all the calling convention quirks for mmap(), clone() and compat
syscalls. It also only attaches to "native" syscall interfaces. If host
system supports compat syscalls or defines 32-bit syscalls in 64-bit
kernel, such syscall interfaces won't be attached to by libbpf.
These limitations may or may not change in the future. Therefore it is
recommended to use SEC("kprobe") for these syscalls or if working with
compat and 32-bit interfaces is required.
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220714070755.3235561-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Improve BPF_KPROBE_SYSCALL (and rename it to shorter BPF_KSYSCALL to
match libbpf's SEC("ksyscall") section name, added in next patch) to use
__kconfig variable to determine how to properly fetch syscall arguments.
Instead of relying on hard-coded knowledge of whether kernel's
architecture uses syscall wrapper or not (which only reflects the latest
kernel versions, but is not necessarily true for older kernels and won't
necessarily hold for later kernel versions on some particular host
architecture), determine this at runtime by attempting to create
perf_event (with fallback to kprobe event creation through tracefs on
legacy kernels, just like kprobe attachment code is doing) for kernel
function that would correspond to bpf() syscall on a system that has
CONFIG_ARCH_HAS_SYSCALL_WRAPPER set (e.g., for x86-64 it would try
'__x64_sys_bpf').
If host kernel uses syscall wrapper, syscall kernel function's first
argument is a pointer to struct pt_regs that then contains syscall
arguments. In such case we need to use bpf_probe_read_kernel() to fetch
actual arguments (which we do through BPF_CORE_READ() macro) from inner
pt_regs.
But if the kernel doesn't use syscall wrapper approach, input
arguments can be read from struct pt_regs directly with no probe reading.
All this feature detection is done without requiring /proc/config.gz
existence and parsing, and BPF-side helper code uses newly added
LINUX_HAS_SYSCALL_WRAPPER virtual __kconfig extern to keep in sync with
user-side feature detection of libbpf.
BPF_KSYSCALL() macro can be used both with SEC("kprobe") programs that
define syscall function explicitly (e.g., SEC("kprobe/__x64_sys_bpf"))
and SEC("ksyscall") program added in the next patch (which are the same
kprobe program with added benefit of libbpf determining correct kernel
function name automatically).
Kretprobe and kretsyscall (added in next patch) programs don't need
BPF_KSYSCALL as they don't provide access to input arguments. Normal
BPF_KRETPROBE is completely sufficient and is recommended.
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220714070755.3235561-4-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Exercise libbpf's logic for unknown __weak virtual __kconfig externs.
USDT selftests are already excercising non-weak known virtual extern
already (LINUX_HAS_BPF_COOKIE), so no need to add explicit tests for it.
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220714070755.3235561-3-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Libbpf supports single virtual __kconfig extern currently: LINUX_KERNEL_VERSION.
LINUX_KERNEL_VERSION isn't coming from /proc/kconfig.gz and is intead
customly filled out by libbpf.
This patch generalizes this approach to support more such virtual
__kconfig externs. One such extern added in this patch is
LINUX_HAS_BPF_COOKIE which is used for BPF-side USDT supporting code in
usdt.bpf.h instead of using CO-RE-based enum detection approach for
detecting bpf_get_attach_cookie() BPF helper. This allows to remove
otherwise not needed CO-RE dependency and keeps user-space and BPF-side
parts of libbpf's USDT support strictly in sync in terms of their
feature detection.
We'll use similar approach for syscall wrapper detection for
BPF_KSYSCALL() BPF-side macro in follow up patch.
Generally, currently libbpf reserves CONFIG_ prefix for Kconfig values
and LINUX_ for virtual libbpf-backed externs. In the future we might
extend the set of prefixes that are supported. This can be done without
any breaking changes, as currently any __kconfig extern with
unrecognized name is rejected.
For LINUX_xxx externs we support the normal "weak rule": if libbpf
doesn't recognize given LINUX_xxx extern but such extern is marked as
__weak, it is not rejected and defaults to zero. This follows
CONFIG_xxx handling logic and will allow BPF applications to
opportunistically use newer libbpf virtual externs without breaking on
older libbpf versions unnecessarily.
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Reviewed-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20220714070755.3235561-2-andrii@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add support for writing a custom event reader, by exposing the ring
buffer.
With the new API perf_buffer__buffer() you will get access to the
raw mmaped()'ed per-cpu underlying memory of the ring buffer.
This region contains both the perf buffer data and header
(struct perf_event_mmap_page), which manages the ring buffer
state (head/tail positions, when accessing the head/tail position
it's important to take into consideration SMP).
With this type of low level access one can implement different types of
consumers here are few simple examples where this API helps with:
1. perf_event_read_simple is allocating using malloc, perhaps you want
to handle the wrap-around in some other way.
2. Since perf buf is per-cpu then the order of the events is not
guarnteed, for example:
Given 3 events where each event has a timestamp t0 < t1 < t2,
and the events are spread on more than 1 CPU, then we can end
up with the following state in the ring buf:
CPU[0] => [t0, t2]
CPU[1] => [t1]
When you consume the events from CPU[0], you could know there is
a t1 missing, (assuming there are no drops, and your event data
contains a sequential index).
So now one can simply do the following, for CPU[0], you can store
the address of t0 and t2 in an array (without moving the tail, so
there data is not perished) then move on the CPU[1] and set the
address of t1 in the same array.
So you end up with something like:
void **arr[] = [&t0, &t1, &t2], now you can consume it orderely
and move the tails as you process in order.
3. Assuming there are multiple CPUs and we want to start draining the
messages from them, then we can "pick" with which one to start with
according to the remaining free space in the ring buffer.
Signed-off-by: Jon Doron <jond@wiz.io>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220715181122.149224-1-arilou@gmail.com
Pu Lehui says:
====================
Currently, samples/bpf, tools/runqslower and bpf/iterators use bpftool
for vmlinux.h, skeleton, and static linking only. We can use lightweight
bootstrap version of bpftool to handle these, and it will be faster.
v2:
- make libbpf and bootstrap bpftool independent. and make it simple.
v1: https://lore.kernel.org/bpf/20220712030813.865410-1-pulehui@huawei.com
====================
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
kernel/bpf/preload/iterators use bpftool for vmlinux.h, skeleton, and
static linking only. So we can use lightweight bootstrap version of
bpftool to handle these, and it will be faster.
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220714024612.944071-4-pulehui@huawei.com
tools/runqslower use bpftool for vmlinux.h, skeleton, and static linking
only. So we can use lightweight bootstrap version of bpftool to handle
these, and it will be faster.
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220714024612.944071-3-pulehui@huawei.com
Currently, when cross compiling bpf samples, the host side cannot
use arch-specific bpftool to generate vmlinux.h or skeleton. Since
samples/bpf use bpftool for vmlinux.h, skeleton, and static linking
only, we can use lightweight bootstrap version of bpftool to handle
these, and it's always host-native.
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220714024612.944071-2-pulehui@huawei.com
When checking with sparse, btf_show_type_value() is causing a
warning about checking integer vs NULL when the macro is passed
a pointer, due to the 'value != 0' check. Stop sparse complaining
about any type-casting by adding a cast to the typeof(value).
This fixes the following sparse warnings:
kernel/bpf/btf.c:2579:17: warning: Using plain integer as NULL pointer
kernel/bpf/btf.c:2581:17: warning: Using plain integer as NULL pointer
kernel/bpf/btf.c:3407:17: warning: Using plain integer as NULL pointer
kernel/bpf/btf.c:3758:9: warning: Using plain integer as NULL pointer
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220714100322.260467-1-ben.dooks@sifive.com
The commit 7337224fc1 ("bpf: Improve the info.func_info and info.func_info_rec_size behavior")
accidently made bpf_prog_ksym_set_name() conservative for bpf subprograms.
Fixed it so instead of "bpf_prog_tag_F" the stack traces print "bpf_prog_tag_full_subprog_name".
Fixes: 7337224fc1 ("bpf: Improve the info.func_info and info.func_info_rec_size behavior")
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220714211637.17150-1-alexei.starovoitov@gmail.com
Alexei reported crash by running test_progs -j on system
with 32 cpus.
It turned out the kprobe_multi bench test that attaches all
ftrace-able functions will race with bpf_dispatcher_update,
that calls bpf_arch_text_poke on bpf_dispatcher_xdp_func,
which is ftrace-able function.
Ftrace is not aware of this update so this will cause
ftrace_bug with:
WARNING: CPU: 6 PID: 1985 at
arch/x86/kernel/ftrace.c:94 ftrace_verify_code+0x27/0x50
...
ftrace_replace_code+0xa3/0x170
ftrace_modify_all_code+0xbd/0x150
ftrace_startup_enable+0x3f/0x50
ftrace_startup+0x98/0xf0
register_ftrace_function+0x20/0x60
register_fprobe_ips+0xbb/0xd0
bpf_kprobe_multi_link_attach+0x179/0x430
__sys_bpf+0x18a1/0x2440
...
------------[ ftrace bug ]------------
ftrace failed to modify
[<ffffffff818d9380>] bpf_dispatcher_xdp_func+0x0/0x10
actual: ffffffe9:7b:ffffff9c:77:1e
Setting ftrace call site to call ftrace function
It looks like we need some way to hide some functions
from ftrace, but meanwhile we workaround this by skipping
bpf_dispatcher_xdp_func from kprobe_multi bench test.
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220714082316.479181-1-jolsa@kernel.org
A couple of the syscalls which load values (bpf_skb_load_helper_16() and
bpf_skb_load_helper_32()) are using u16/u32 types which are triggering
warnings as they are then converted from big-endian to CPU-endian. Fix
these by making the types __be instead.
Fixes the following sparse warnings:
net/core/filter.c:246:32: warning: cast to restricted __be16
net/core/filter.c:246:32: warning: cast to restricted __be16
net/core/filter.c:246:32: warning: cast to restricted __be16
net/core/filter.c:246:32: warning: cast to restricted __be16
net/core/filter.c:273:32: warning: cast to restricted __be32
net/core/filter.c:273:32: warning: cast to restricted __be32
net/core/filter.c:273:32: warning: cast to restricted __be32
net/core/filter.c:273:32: warning: cast to restricted __be32
net/core/filter.c:273:32: warning: cast to restricted __be32
net/core/filter.c:273:32: warning: cast to restricted __be32
Signed-off-by: Ben Dooks <ben.dooks@sifive.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20220714105101.297304-1-ben.dooks@sifive.com
BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE is also tracing type, which may
cause unexpected memory allocation if we set BPF_F_NO_PREALLOC. Let's
also warn on it similar as we do in case of BPF_PROG_TYPE_RAW_TRACEPOINT.
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220713160936.57488-1-laoar.shao@gmail.com
When application runs in busy poll mode and does not receive a single
packet but only sends them, it is currently impossible to get into
napi_busy_loop() as napi_id is only marked on Rx side in xsk_rcv_check().
In there, napi_id is being taken from xdp_rxq_info carried by xdp_buff.
From Tx perspective, we do not have access to it. What we have handy is
the xsk pool.
Xsk pool works on a pool of internal xdp_buff wrappers called xdp_buff_xsk.
AF_XDP ZC enabled drivers call xp_set_rxq_info() so each of xdp_buff_xsk
has a valid pointer to xdp_rxq_info of underlying queue. Therefore, on Tx
side, napi_id can be pulled from xs->pool->heads[0].xdp.rxq->napi_id. Hide
this pointer chase under helper function, xsk_pool_get_napi_id().
Do this only for sockets working in ZC mode as otherwise rxq pointers would
not be initialized.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Link: https://lore.kernel.org/bpf/20220707130842.49408-1-maciej.fijalkowski@intel.com
When building with clang + CONFIG_CFI_CLANG=y, the following error
occurs at link time:
ld.lld: error: undefined symbol: dummy_tramp
dummy_tramp is declared globally in C but its definition in inline
assembly does not use .global, which prevents clang from properly
resolving the references to it when creating the CFI jump tables.
Mark dummy_tramp as global so that the reference can be properly
resolved.
Fixes: b2ad54e153 ("bpf, arm64: Implement bpf_arch_text_poke() for arm64")
Suggested-by: Sami Tolvanen <samitolvanen@google.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Link: https://github.com/ClangBuiltLinux/linux/issues/1661
Link: https://lore.kernel.org/bpf/20220713173503.3889486-1-nathan@kernel.org
Return boolean values ("true" or "false") instead of 1 or 0 from bool
functions. This fixes the following warnings from coccicheck:
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:407:9-10: WARNING:
return of 0/1 in function 'decap_v4' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:389:9-10: WARNING:
return of 0/1 in function 'decap_v6' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:290:9-10: WARNING:
return of 0/1 in function 'encap_v6' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:264:9-10: WARNING:
return of 0/1 in function 'parse_tcp' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:242:9-10: WARNING:
return of 0/1 in function 'parse_udp' with return type bool
Generated by: scripts/coccinelle/misc/boolreturn.cocci
Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Linkui Xiao <xiaolinkui@kylinos.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/bpf/20220714015647.25074-1-xiaolinkui@kylinos.cn
BPF map name is limited to BPF_OBJ_NAME_LEN.
A map name is defined as being longer than BPF_OBJ_NAME_LEN,
it will be truncated to BPF_OBJ_NAME_LEN when a userspace program
calls libbpf to create the map. A pinned map also generates a path
in the /sys. If the previous program wanted to reuse the map，
it can not get bpf_map by name, because the name of the map is only
partially the same as the name which get from pinned path.
The syscall information below show that map name "process_pinned_map"
is truncated to "process_pinned_".
bpf(BPF_OBJ_GET, {pathname="/sys/fs/bpf/process_pinned_map",
bpf_fd=0, file_flags=0}, 144) = -1 ENOENT (No such file or directory)
bpf(BPF_MAP_CREATE, {map_type=BPF_MAP_TYPE_HASH, key_size=4,
value_size=4,max_entries=1024, map_flags=0, inner_map_fd=0,
map_name="process_pinned_",map_ifindex=0, btf_fd=3, btf_key_type_id=6,
btf_value_type_id=10,btf_vmlinux_value_type_id=0}, 72) = 4
This patch check that if the name of pinned map are the same as the
actual name for the first (BPF_OBJ_NAME_LEN - 1),
bpf map still uses the name which is included in bpf object.
Fixes: 26736eb9a4 ("tools: libbpf: allow map reuse")
Signed-off-by: Anquan Wu <leiqi96@hotmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/OSZP286MB1725CEA1C95C5CB8E7CCC53FB8869@OSZP286MB1725.JPNP286.PROD.OUTLOOK.COM
The ARRAY_SIZE macro is more compact and more formal in linux source.
Signed-off-by: Linkui Xiao <xiaolinkui@kylinos.cn>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20220712072302.13761-1-xiaolinkui@kylinos.cn
This patch does two things:
1. For matching against the arg type, the match should be against the
base type of the arg type, since the arg type can have different
bpf_type_flags set on it.
2. Uses switch casing to improve readability + efficiency.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Acked-by: Hao Luo <haoluo@google.com>
Link: https://lore.kernel.org/r/20220712210603.123791-1-joannelkoong@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
binary_path is a required non-null parameter for bpf_program__attach_usdt
and bpf_program__attach_uprobe_opts. Check it against NULL to prevent
coredump on strchr.
Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20220712025745.2703995-1-hengqi.chen@gmail.com
GFP_ATOMIC doesn't cooperate well with memcg pressure so far, especially
if we allocate too much GFP_ATOMIC memory. For example, when we set the
memcg limit to limit a non-preallocated bpf memory, the GFP_ATOMIC can
easily break the memcg limit by force charge. So it is very dangerous to
use GFP_ATOMIC in non-preallocated case. One way to make it safe is to
remove __GFP_HIGH from GFP_ATOMIC, IOW, use (__GFP_ATOMIC |
__GFP_KSWAPD_RECLAIM) instead, then it will be limited if we allocate
too much memory. There's a plan to completely remove __GFP_ATOMIC in the
mm side[1], so let's use GFP_NOWAIT instead.
We introduced BPF_F_NO_PREALLOC is because full map pre-allocation is
too memory expensive for some cases. That means removing __GFP_HIGH
doesn't break the rule of BPF_F_NO_PREALLOC, but has the same goal with
it-avoiding issues caused by too much memory. So let's remove it.
This fix can also apply to other run-time allocations, for example, the
allocation in lpm trie, local storage and devmap. So let fix it
consistently over the bpf code
It also fixes a typo in the comment.
[1]. https://lore.kernel.org/linux-mm/163712397076.13692.4727608274002939094@noble.neil.brown.name/
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: NeilBrown <neilb@suse.de>
Signed-off-by: Yafang Shao <laoar.shao@gmail.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Link: https://lore.kernel.org/r/20220709154457.57379-2-laoar.shao@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
syzbot reported a few issues with bpf_prog_pack [1], [2]. This only happens
with multiple subprogs. In jit_subprogs(), we first call bpf_int_jit_compile()
on each sub program. And then, we call it on each sub program again. jit_data
is not freed in the first call of bpf_int_jit_compile(). Similarly we don't
call bpf_jit_binary_pack_finalize() in the first call of bpf_int_jit_compile().
If bpf_int_jit_compile() failed for one sub program, we will call
bpf_jit_binary_pack_finalize() for this sub program. However, we don't have a
chance to call it for other sub programs. Then we will hit "goto out_free" in
jit_subprogs(), and call bpf_jit_free on some subprograms that haven't got
bpf_jit_binary_pack_finalize() yet.
At this point, bpf_jit_binary_pack_free() is called and the whole 2MB page is
freed erroneously.
Fix this with a custom bpf_jit_free() for x86_64, which calls
bpf_jit_binary_pack_finalize() if necessary. Also, with custom
bpf_jit_free(), bpf_prog_aux->use_bpf_prog_pack is not needed any more,
remove it.
Fixes: 1022a5498f ("bpf, x86_64: Use bpf_jit_binary_pack_alloc")
[1] https://syzkaller.appspot.com/bug?extid=2f649ec6d2eea1495a8f
[2] https://syzkaller.appspot.com/bug?extid=87f65c75f4a72db05445
Reported-by: syzbot+2f649ec6d2eea1495a8f@syzkaller.appspotmail.com
Reported-by: syzbot+87f65c75f4a72db05445@syzkaller.appspotmail.com
Signed-off-by: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/r/20220706002612.4013790-1-song@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The memory consumed by a bpf map is always accounted to the memory
cgroup of the process which created the map. The map can outlive
the memory cgroup if it's used by processes in other cgroups or
is pinned on bpffs. In this case the map pins the original cgroup
in the dying state.
For other types of objects (slab objects, non-slab kernel allocations,
percpu objects and recently LRU pages) there is a reparenting process
implemented: on cgroup offlining charged objects are getting
reassigned to the parent cgroup. Because all charges and statistics
are fully recursive it's a fairly cheap operation.
For efficiency and consistency with other types of objects, let's do
the same for bpf maps. Fortunately thanks to the objcg API, the
required changes are minimal.
Please, note that individual allocations (slabs, percpu and large
kmallocs) already have the reparenting mechanism. This commit adds
it to the saved map->memcg pointer by replacing it to map->objcg.
Because dying cgroups are not visible for a user and all charges are
recursive, this commit doesn't bring any behavior changes for a user.
v2:
added a missing const qualifier
Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Link: https://lore.kernel.org/r/20220711162827.184743-1-roman.gushchin@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Alan Maguire says:
====================
a ksym BPF iterator would be useful as it would allow more flexible
interactions with kernel symbols than are currently supported; it could
for example create more efficient map representations for lookup,
speed up symbol resolution etc.
The idea was initially discussed here [1].
Changes since v5 [2]:
- no need to add kallsym_iter to bpf_iter.h as it has existed in kernels
for a long time so will by in vmlinux.h for older kernels too, unlike
struct bpf_iter__ksym (Yonghong, patch 2)
Changes since v4 [3]:
- add BPF_ITER_RESCHED to improve responsiveness (Hao, patch 1)
- remove pr_warn to be consistent with other iterators (Andrii, patch 1)
- add definitions to bpf_iter.h to ensure iter tests build on older
kernels (Andrii, patch 2)
Changes since v3 [4]:
- use late_initcall() to register iter; means we are both consistent
with other iters and can encapsulate all iter-specific code in
kallsyms.c in CONFIG_BPF_SYSCALL (Alexei, Yonghong, patch 1).
Changes since v2 [5]:
- set iter->show_value on initialization based on current creds
and use it in selftest to determine if we show values
(Yonghong, patches 1/2)
- inline iter registration into kallsyms_init (Yonghong, patch 1)
Changes since RFC [6]:
- change name of iterator (and associated structures/fields) to "ksym"
(Andrii, patches 1, 2)
- remove dependency on CONFIG_PROC_FS; it was used for other BPF
iterators, and I assumed it was needed because of seq ops but I
don't think it is required on digging futher (Andrii, patch 1)
[1] https://lore.kernel.org/all/YjRPZj6Z8vuLeEZo@krava/
[2] https://lore.kernel.org/bpf/1657490998-31468-1-git-send-email-alan.maguire@oracle.com/
[3] https://lore.kernel.org/bpf/1657113391-5624-1-git-send-email-alan.maguire@oracle.com/
[4] https://lore.kernel.org/bpf/1656942916-13491-1-git-send-email-alan.maguire@oracle.com
[5] https://lore.kernel.org/bpf/1656667620-18718-1-git-send-email-alan.maguire@oracle.com/
[6] https://lore.kernel.org/all/1656089118-577-1-git-send-email-alan.maguire@oracle.com/
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
add subtest verifying BPF ksym iter behaviour. The BPF ksym
iter program shows an example of dumping a format different to
/proc/kallsyms. It adds KIND and MAX_SIZE fields which represent the
kind of symbol (core kernel, module, ftrace, bpf, or kprobe) and
the maximum size the symbol can be. The latter is calculated from
the difference between current symbol value and the next symbol
value.
The key benefit for this iterator will likely be supporting in-kernel
data-gathering rather than dumping symbol details to userspace and
parsing the results.
Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/1657629105-7812-3-git-send-email-alan.maguire@oracle.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Our CI[1] reported these warnings when using Sparse:
$ touch net/mptcp/bpf.c
$ make C=1 net/mptcp/bpf.o
net/mptcp/bpf.c: note: in included file:
include/linux/bpf_verifier.h:348:26: error: dubious one-bit signed bitfield
include/linux/bpf_verifier.h:349:29: error: dubious one-bit signed bitfield
Set them as 'unsigned' to avoid warnings.
[1] https://github.com/multipath-tcp/mptcp_net-next/actions/runs/2643588487
Fixes: 1ade237119 ("bpf: Inline calls to bpf_loop when callback is known")
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220711081200.2081262-1-matthieu.baerts@tessares.net
LLVM compiler optimized out the memcpy in xdp_redirect_map_egress,
which caused the Ethernet source MAC-addr to always be zero
when enabling the devmap egress prog via cmdline --load-egress.
Issue observed with LLVM version 14.0.0
- Shipped with Fedora 36 on target: x86_64-redhat-linux-gnu.
In verbose mode print the source MAC-addr in case xdp_devmap_attached
mode is used.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/165754826292.575614.5636444052787717159.stgit@firesoul
Implement bpf_arch_text_poke() for arm64, so bpf prog or bpf trampoline
can be patched with it.
When the target address is NULL, the original instruction is patched to
a NOP.
When the target address and the source address are within the branch
range, the original instruction is patched to a bl instruction to the
target address directly.
To support attaching bpf trampoline to both regular kernel function and
bpf prog, we follow the ftrace patchsite way for bpf prog. That is, two
instructions are inserted at the beginning of bpf prog, the first one
saves the return address to x9, and the second is a nop which will be
patched to a bl instruction when a bpf trampoline is attached.
However, when a bpf trampoline is attached to bpf prog, the distance
between target address and source address may exceed 128MB, the maximum
branch range, because bpf trampoline and bpf prog are allocated
separately with vmalloc. So long jump should be handled.
When a bpf prog is constructed, a plt pointing to empty trampoline
dummy_tramp is placed at the end:
bpf_prog:
mov x9, lr
nop // patchsite
...
ret
plt:
ldr x10, target
br x10
target:
.quad dummy_tramp // plt target
This is also the state when no trampoline is attached.
When a short-jump bpf trampoline is attached, the patchsite is patched to
a bl instruction to the trampoline directly:
bpf_prog:
mov x9, lr
bl <short-jump bpf trampoline address> // patchsite
...
ret
plt:
ldr x10, target
br x10
target:
.quad dummy_tramp // plt target
When a long-jump bpf trampoline is attached, the plt target is filled with
the trampoline address and the patchsite is patched to a bl instruction to
the plt:
bpf_prog:
mov x9, lr
bl plt // patchsite
...
ret
plt:
ldr x10, target
br x10
target:
.quad <long-jump bpf trampoline address>
dummy_tramp is used to prevent another CPU from jumping to an unknown
location during the patching process, making the patching process easier.
The patching process is as follows:
1. when neither the old address or the new address is a long jump, the
patchsite is replaced with a bl to the new address, or nop if the new
address is NULL;
2. when the old address is not long jump but the new one is, the
branch target address is written to plt first, then the patchsite
is replaced with a bl instruction to the plt;
3. when the old address is long jump but the new one is not, the address
of dummy_tramp is written to plt first, then the patchsite is replaced
with a bl to the new address, or a nop if the new address is NULL;
4. when both the old address and the new address are long jump, the
new address is written to plt and the patchsite is not changed.
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jakub Sitnicki <jakub@cloudflare.com>
Reviewed-by: KP Singh <kpsingh@kernel.org>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20220711150823.2128542-4-xukuohai@huawei.com
Add LDR (literal) instruction to load data from address relative to PC.
This instruction will be used to implement long jump from bpf prog to
bpf trampoline in the follow-up patch.
The instruction encoding:
3 2 2 2 0 0
0 7 6 4 5 0
+-----+-------+---+-----+-------------------------------------+--------+
| 0 x | 0 1 1 | 0 | 0 0 | imm19 | Rt |
+-----+-------+---+-----+-------------------------------------+--------+
for 32-bit, variant x == 0; for 64-bit, x == 1.
branch_imm_common() is used to check the distance between pc and target
address, since it's reused by this patch and LDR (literal) is not a branch
instruction, rename it to label_imm_common().
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/bpf/20220711150823.2128542-3-xukuohai@huawei.com
Before generating bpf trampoline, x86 calls is_valid_bpf_tramp_flags()
to check the input flags. This check is architecture independent.
So, to be consistent with x86, arm64 should also do this check
before generating bpf trampoline.
However, the BPF_TRAMP_F_XXX flags are not used by user code and the
flags argument is almost constant at compile time, so this run time
check is a bit redundant.
Remove is_valid_bpf_tramp_flags() and add some comments to the usage of
BPF_TRAMP_F_XXX flags, as suggested by Alexei.
Signed-off-by: Xu Kuohai <xukuohai@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Acked-by: Song Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20220711150823.2128542-2-xukuohai@huawei.com
In sk_psock_skb_ingress_enqueue function, if the linear area + nr_frags +
frag_list of the SKB has NR_MSG_FRAG_IDS blocks in total, skb_to_sgvec
will return NR_MSG_FRAG_IDS, then msg->sg.end will be set to
NR_MSG_FRAG_IDS, and in addition, (NR_MSG_FRAG_IDS - 1) is set to the last
SG of msg. Recv the msg in sk_msg_recvmsg, when i is (NR_MSG_FRAG_IDS - 1),
the sk_msg_iter_var_next(i) will change i to 0 (not NR_MSG_FRAG_IDS), the
judgment condition "msg_rx->sg.start==msg_rx->sg.end" and
"i != msg_rx->sg.end" can not work.
As a result, the processed msg cannot be deleted from ingress_msg list.
But the length of all the sge of the msg has changed to 0. Then the next
recvmsg syscall will process the msg repeatedly, because the length of sge
is 0, the -EFAULT error is always returned.
Fixes: 604326b41a ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Liu Jian <liujian56@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20220628123616.186950-1-liujian56@huawei.com
