All bpftool commands support the options for JSON output and debug from
libbpf. In addition, some commands support additional options
corresponding to specific use cases.

The list of options described in the man pages for the different
commands are not always accurate. The messages for interactive help are
mostly limited to HELP_SPEC_OPTIONS, and are even less representative of
the actual set of options supported for the commands.

Let's update the lists:

- HELP_SPEC_OPTIONS is modified to contain the "default" options (JSON
  and debug), and to be extensible (no ending curly bracket).
- All commands use HELP_SPEC_OPTIONS in their help message, and then
  complete the list with their specific options.
- The lists of options in the man pages are updated.
- The formatting of the list for bpftool.rst is adjusted to match
  formatting for the other man pages. This is for consistency, and also
  because it will be helpful in a future patch to automatically check
  that the files are synchronised.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210730215435.7095-5-quentin@isovalent.com

Update bpftool's list of attach type names to tell it about the latest
attach types, or the "ringbuf" map. Also update the documentation, help
messages, and bash completion when relevant.

These missing items were reported by the newly added Python script used
to help maintain consistency in bpftool.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210730215435.7095-4-quentin@isovalent.com

Whenever the eBPF subsystem gains new elements, such as new program or
map types, it is necessary to update bpftool if we want it able to
handle the new items.

In addition to the main arrays containing the names of these elements in
the source code, there are also multiple locations to update:

- The help message in the do_help() functions in bpftool's source code.
- The RST documentation files.
- The bash completion file.

This has led to omissions multiple times in the past. This patch
attempts to address this issue by adding consistency checks for all
these different locations. It also verifies that the bpf_prog_type,
bpf_map_type and bpf_attach_type enums from the UAPI BPF header have all
their members present in bpftool.

The script requires no argument to run, it reads and parses the
different files to check, and prints the mismatches, if any. It
currently reports a number of missing elements, which will be fixed in a
later patch:

  $ ./test_bpftool_synctypes.py
  Comparing [...]/linux/tools/bpf/bpftool/map.c (map_type_name) and [...]/linux/tools/bpf/bpftool/bash-completion/bpftool (BPFTOOL_MAP_CREATE_TYPES): {'ringbuf'}
  Comparing BPF header (enum bpf_attach_type) and [...]/linux/tools/bpf/bpftool/common.c (attach_type_name): {'BPF_TRACE_ITER', 'BPF_XDP_DEVMAP', 'BPF_XDP', 'BPF_SK_REUSEPORT_SELECT', 'BPF_XDP_CPUMAP', 'BPF_SK_REUSEPORT_SELECT_OR_MIGRATE'}
  Comparing [...]/linux/tools/bpf/bpftool/prog.c (attach_type_strings) and [...]/linux/tools/bpf/bpftool/prog.c (do_help() ATTACH_TYPE): {'skb_verdict'}
  Comparing [...]/linux/tools/bpf/bpftool/prog.c (attach_type_strings) and [...]/linux/tools/bpf/bpftool/Documentation/bpftool-prog.rst (ATTACH_TYPE): {'skb_verdict'}
  Comparing [...]/linux/tools/bpf/bpftool/prog.c (attach_type_strings) and [...]/linux/tools/bpf/bpftool/bash-completion/bpftool (BPFTOOL_PROG_ATTACH_TYPES): {'skb_verdict'}

Note that the script does NOT check for consistency between the list of
program types that bpftool claims it accepts and the actual list of
keywords that can be used. This is because bpftool does not "see" them,
they are ELF section names parsed by libbpf. It is not hard to parse the
section_defs[] array in libbpf, but some section names are associated
with program types that bpftool cannot load at the moment. For example,
some programs require a BTF target and an attach target that bpftool
cannot handle. The script may be extended to parse the array and check
only relevant values in the future.

The script is not added to the selftests' Makefile, because doing so
would require all patches with BPF UAPI change to also update bpftool.
Instead it is to be added to the CI.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210730215435.7095-3-quentin@isovalent.com

Bash completion for bpftool gets two minor improvements in this patch.

Move the detection of attach types for "bpftool cgroup attach" outside
of the "case/esac" bloc, where we cannot reuse our variable holding the
list of supported attach types as a pattern list. After the change, we
have only one list of cgroup attach types to update when new types are
added, instead of the former two lists.

Also rename the variables holding lists of names for program types, map
types, and attach types, to make them more unique. This can make it
slightly easier to point people to the relevant variables to update, but
the main objective here is to help run a script to check that bash
completion is up-to-date with bpftool's source code.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210730215435.7095-2-quentin@isovalent.com

As Eric noticed, __unix_dgram_recvmsg() may acquire u->iolock
too, so we have to release it before calling this function.

Fixes: 9825d866

 ("af_unix: Implement unix_dgram_bpf_recvmsg()")
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Jakub Sitnicki <jakub@cloudflare.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>

Add two new APIs: btf__load_vmlinux_btf and btf__load_module_btf.
btf__load_vmlinux_btf is just an alias to the existing API named
libbpf_find_kernel_btf, rename to be more precisely and consistent
with existing BTF APIs. btf__load_module_btf can be used to load
module BTF, add it for completeness. These two APIs are useful for
implementing tracing tools and introspection tools. This is part
of the effort towards libbpf 1.0 ([0]).

  [0] Closes: https://github.com/libbpf/libbpf/issues/280



Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210730114012.494408-1-hengqi.chen@gmail.com

Quentin Monnet says:

====================

As part of the effort to move towards a v1.0 for libbpf [0], this set
improves some confusing function names related to BTF loading from and to
the kernel:

- btf__load() becomes btf__load_into_kernel().
- btf__get_from_id becomes btf__load_from_kernel_by_id().
- A new version btf__load_from_kernel_by_id_split() extends the former to
  add support for split BTF.

The last patch is a trivial change to bpftool to add support for dumping
split BTF objects by referencing them by their id (and not only by their
BTF path).

[0] https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0#btfh-apis



v3:
- Use libbpf_err_ptr() in btf__load_from_kernel_by_id(), ERR_PTR() in
  bpftool's get_map_kv_btf().
- Move the definition of btf__load_from_kernel_by_id() closer to the
  btf__parse() group in btf.h (move the legacy function with it).
- Fix a bug on the return value in libbpf_find_prog_btf_id(), as a new
  patch.
- Move the btf__free() fixes to their own patch.
- Add "Fixes:" tags to relevant patches.

v2:
- Remove deprecation marking of legacy functions (patch 4/6 from v1).
- Make btf__load_from_kernel_by_id{,_split}() return the btf struct, adjust
  surrounding code and call btf__free() when missing.
- Add new functions to v0.5.0 API (and not v0.6.0).
====================

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>

Split BTF objects are typically BTF objects for kernel modules, which
are incrementally built on top of kernel BTF instead of redefining all
kernel symbols they need. We can use bpftool with its -B command-line
option to dump split BTF objects. It works well when the handle provided
for the BTF object to dump is a "path" to the BTF object, typically
under /sys/kernel/btf, because bpftool internally calls
btf__parse_split() which can take a "base_btf" pointer and resolve the
BTF reconstruction (although in that case, the "-B" option is
unnecessary because bpftool performs autodetection).

However, it did not work so far when passing the BTF object through its
id, because bpftool would call btf__get_from_id() which did not provide
a way to pass a "base_btf" pointer.

In other words, the following works:

    # bpftool btf dump file /sys/kernel/btf/i2c_smbus -B /sys/kernel/btf/vmlinux

But this was not possible:

    # bpftool btf dump id 6 -B /sys/kernel/btf/vmlinux

The libbpf API has recently changed, and btf__get_from_id() has been
deprecated in favour of btf__load_from_kernel_by_id() and its version
with support for split BTF, btf__load_from_kernel_by_id_split(). Let's
update bpftool to make it able to dump the BTF object in the second case
as well.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20210729162028.29512-9-quentin@isovalent.com

Add a new API function btf__load_from_kernel_by_id_split(), which takes
a pointer to a base BTF object in order to support split BTF objects
when retrieving BTF information from the kernel.

Reference: https://github.com/libbpf/libbpf/issues/314



Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20210729162028.29512-8-quentin@isovalent.com

Replace the calls to function btf__get_from_id(), which we plan to
deprecate before the library reaches v1.0, with calls to
btf__load_from_kernel_by_id() in tools/ (bpftool, perf, selftests).
Update the surrounding code accordingly (instead of passing a pointer to
the btf struct, get it as a return value from the function).

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20210729162028.29512-6-quentin@isovalent.com

Make sure to call btf__free() (and not simply free(), which does not
free all pointers stored in the struct) on pointers to struct btf
objects retrieved at various locations.

These were found while updating the calls to btf__get_from_id().

Fixes: 999d82cb ("tools/bpf: enhance test_btf file testing to test func info")
Fixes: 254471e5 ("tools/bpf: bpftool: add support for func types")
Fixes: 7b612e29 ("perf tools: Synthesize PERF_RECORD_* for loaded BPF programs")
Fixes: d56354dc ("perf tools: Save bpf_prog_info and BTF of new BPF programs")
Fixes: 47c09d6a ("bpftool: Introduce "prog profile" command")
Fixes: fa853c4b

 ("perf stat: Enable counting events for BPF programs")
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210729162028.29512-5-quentin@isovalent.com

Rename function btf__get_from_id() as btf__load_from_kernel_by_id() to
better indicate what the function does. Change the new function so that,
instead of requiring a pointer to the pointer to update and returning
with an error code, it takes a single argument (the id of the BTF
object) and returns the corresponding pointer. This is more in line with
the existing constructors.

The other tools calling the (soon-to-be) deprecated btf__get_from_id()
function will be updated in a future commit.

References:

- https://github.com/libbpf/libbpf/issues/278
- https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0#btfh-apis



Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20210729162028.29512-4-quentin@isovalent.com

As part of the effort to move towards a v1.0 for libbpf, rename
btf__load() function, used to "upload" BTF information into the kernel,
as btf__load_into_kernel(). This new name better reflects what the
function does.

References:

- https://github.com/libbpf/libbpf/issues/278
- https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0#btfh-apis



Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20210729162028.29512-3-quentin@isovalent.com

Variable "err" is initialised to -EINVAL so that this error code is
returned when something goes wrong in libbpf_find_prog_btf_id().
However, a recent change in the function made use of the variable in
such a way that it is set to 0 if retrieving linear information on the
program is successful, and this 0 value remains if we error out on
failures at later stages.

Let's fix this by setting err to -EINVAL later in the function.

Fixes: e9fc3ce9

 ("libbpf: Streamline error reporting for high-level APIs")
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210729162028.29512-2-quentin@isovalent.com

To avoid kernel build failure due to some missing .BTF-ids referenced
functions/types, the patch ([1]) tries to fill btf_id 0 for
these types.

In bpf verifier, for percpu variable and helper returning btf_id cases,
verifier already emitted proper warning with something like
  verbose(env, "Helper has invalid btf_id in R%d\n", regno);
  verbose(env, "invalid return type %d of func %s#%d\n",
          fn->ret_type, func_id_name(func_id), func_id);

But this is not the case for bpf_iter context arguments.
I hacked resolve_btfids to encode btf_id 0 for struct task_struct.
With `./test_progs -n 7/5`, I got,
  0: (79) r2 = *(u64 *)(r1 +0)
  func 'bpf_iter_task' arg0 has btf_id 29739 type STRUCT 'bpf_iter_meta'
  ; struct seq_file *seq = ctx->meta->seq;
  1: (79) r6 = *(u64 *)(r2 +0)
  ; struct task_struct *task = ctx->task;
  2: (79) r7 = *(u64 *)(r1 +8)
  ; if (task == (void *)0) {
  3: (55) if r7 != 0x0 goto pc+11
  ...
  ; BPF_SEQ_PRINTF(seq, "%8d %8d\n", task->tgid, task->pid);
  26: (61) r1 = *(u32 *)(r7 +1372)
  Type '(anon)' is not a struct

Basically, verifier will return btf_id 0 for task_struct.
Later on, when the code tries to access task->tgid, the
verifier correctly complains the type is '(anon)' and it is
not a struct. Users still need to backtrace to find out
what is going on.

Let us catch the invalid btf_id 0 earlier
and provide better message indicating btf_id is wrong.
The new error message looks like below:
  R1 type=ctx expected=fp
  ; struct seq_file *seq = ctx->meta->seq;
  0: (79) r2 = *(u64 *)(r1 +0)
  func 'bpf_iter_task' arg0 has btf_id 29739 type STRUCT 'bpf_iter_meta'
  ; struct seq_file *seq = ctx->meta->seq;
  1: (79) r6 = *(u64 *)(r2 +0)
  ; struct task_struct *task = ctx->task;
  2: (79) r7 = *(u64 *)(r1 +8)
  invalid btf_id for context argument offset 8
  invalid bpf_context access off=8 size=8

[1] https://lore.kernel.org/bpf/20210727132532.2473636-1-hengqi.chen@gmail.com/



Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210728183025.1461750-1-yhs@fb.com

Kernel functions referenced by .BTF_ids may be changed from global to static
and get inlined or get renamed/removed, and thus disappears from BTF.
This causes kernel build failure when resolve_btfids do id patch for symbols
in .BTF_ids in vmlinux. Update resolve_btfids to emit warning messages and
patch zero id for missing symbols instead of aborting kernel build process.

Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210727132532.2473636-2-hengqi.chen@gmail.com

Current max cgroup storage value size is 4k (PAGE_SIZE). The other local
storages accept up to 64k (BPF_LOCAL_STORAGE_MAX_VALUE_SIZE). Let's align
max cgroup value size with the other storages.

For percpu, the max is 32k (PCPU_MIN_UNIT_SIZE) because percpu
allocator is not happy about larger values.

netcnt test is extended to exercise those maximum values
(non-percpu max size is close to, but not real max).

v4:
* remove inner union (Andrii Nakryiko)
* keep net_cnt on the stack (Andrii Nakryiko)

v3:
* refine SIZEOF_BPF_LOCAL_STORAGE_ELEM comment (Yonghong Song)
* anonymous struct in percpu_net_cnt & net_cnt (Yonghong Song)
* reorder free (Yonghong Song)

v2:
* cap max_value_size instead of BUILD_BUG_ON (Martin KaFai Lau)

Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20210727222335.4029096-1-sdf@google.com

When loading in parallel multiple programs which use the same to-be
pinned map, it is possible that two instances of the loader will call
bpf_object__create_maps() at the same time. If the map doesn't exist
when both instances call bpf_object__reuse_map(), then one of the
instances will fail with EEXIST when calling bpf_map__pin().

Fix the race by retrying reusing a map if bpf_map__pin() returns
EEXIST. The fix is similar to the one in iproute2: e4c4685fd6e4 ("bpf:
Fix race condition with map pinning").

Before retrying the pinning, we don't do any special cleaning of an
internal map state. The closer code inspection revealed that it's not
required:

    - bpf_object__create_map(): map->inner_map is destroyed after a
      successful call, map->fd is closed if pinning fails.
    - bpf_object__populate_internal_map(): created map elements is
      destroyed upon close(map->fd).
    - init_map_slots(): slots are freed after their initialization.

Signed-off-by: Martynas Pumputis <m@lambda.lt>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210726152001.34845-1-m@lambda.lt

Remove the repeated word 'the' in line 48.

Signed-off-by: Jason Wang <wangborong@cdjrlc.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210727115928.74600-1-wangborong@cdjrlc.com

There are recently added xdp samples (xdp_redirect_map_multi and
xdpsock_ctrl_proc) which are not managed by .gitignore.

This commit adds these files to .gitignore.

Signed-off-by: Juhee Kang <claudiajkang@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210727041056.23455-2-claudiajkang@gmail.com

The current behavior of 'tracex7' doesn't consist with other bpf samples
tracex{1..6}. Other samples do not require any argument to run with, but
tracex7 should be run with btrfs device argument. (it should be executed
with test_override_return.sh)

Currently, tracex7 doesn't have any description about how to run this
program and raises an unexpected error. And this result might be
confusing since users might not have a hunch about how to run this
program.

    // Current behavior
    # ./tracex7
    sh: 1: Syntax error: word unexpected (expecting ")")
    // Fixed behavior
    # ./tracex7
    ERROR: Run with the btrfs device argument!

In order to fix this error, this commit adds logic to report a message
and exit when running this program with a missing argument.

Additionally in test_override_return.sh, there is a problem with
multiple directory(tmpmnt) creation. So in this commit adds a line with
removing the directory with every execution.

Signed-off-by: Juhee Kang <claudiajkang@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210727041056.23455-1-claudiajkang@gmail.com

In the tc_redirect test only use ping6 if it's available and
otherwise fall back to using "ping -6".

Signed-off-by: Jussi Maki <joamaki@gmail.com>

Alexei Starovoitov says:

====================

From: Alexei Starovoitov <ast@kernel.org>

Split CO-RE processing logic from libbpf into separate file
with an interface that doesn't dependend on libbpf internal details.
As the next step relo_core.c will be compiled with libbpf and with the kernel.
The _internal_ interface between libbpf/CO-RE and kernel/CO-RE will be:
int bpf_core_apply_relo_insn(const char *prog_name, struct bpf_insn *insn,
			     int insn_idx,
			     const struct bpf_core_relo *relo,
			     int relo_idx,
			     const struct btf *local_btf,
			     struct bpf_core_cand_list *cands);
where bpf_core_relo and bpf_core_cand_list are simple types
prepared by kernel and libbpf.

Though diff stat shows a lot of lines inserted/deleted they are moved lines.
Pls review with diff.colorMoved.
====================

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>

Move CO-RE logic into separate file.
The internal interface between libbpf and CO-RE is through
bpf_core_apply_relo_insn() function and few structs defined in relo_core.h.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721000822.40958-5-alexei.starovoitov@gmail.com

In order to make a clean split of CO-RE logic move its types
into independent header file.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721000822.40958-4-alexei.starovoitov@gmail.com

bpf_core_apply_relo() doesn't need to know bpf_program internals
and hashmap details.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721000822.40958-3-alexei.starovoitov@gmail.com

CO-RE processing functions don't need to know 'struct bpf_program' details.
Cleanup the layering to eventually be able to move CO-RE logic into a separate file.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721000822.40958-2-alexei.starovoitov@gmail.com

Each test case can have a set of sub-tests, where each sub-test can
run the cBPF/eBPF test snippet with its own data_size and expected
result. Before, the end of the sub-test array was indicated by both
data_size and result being zero. However, most or all of the internal
eBPF tests has a data_size of zero already. When such a test also had
an expected value of zero, the test was never run but reported as
PASS anyway.

Now the test runner always runs the first sub-test, regardless of the
data_size and result values. The sub-test array zero-termination only
applies for any additional sub-tests.

There are other ways fix it of course, but this solution at least
removes the surprise of eBPF tests with a zero result always succeeding.

Signed-off-by: Johan Almbladh <johan.almbladh@anyfinetworks.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721103822.3755111-1-johan.almbladh@anyfinetworks.com

This test now operates on DW as stated instead of W, which was
already covered by another test.

Signed-off-by: Johan Almbladh <johan.almbladh@anyfinetworks.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210721104058.3755254-1-johan.almbladh@anyfinetworks.com

Add a list of vmtest script dependencies to make it easier for new
contributors to get going.

Signed-off-by: Evgeniy Litvinenko <evgeniyl@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210723223645.907802-1-evgeniyl@fb.com

Add bpf_map__pin_path, so that the inconsistently named
bpf_map__get_pin_path can be deprecated later. This is part of the
effort towards libbpf v1.0: https://github.com/libbpf/libbpf/issues/307



Also, add a selftest for the new function.

Signed-off-by: Evgeniy Litvinenko <evgeniyl@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210723221511.803683-1-evgeniyl@fb.com

Martin KaFai says:

====================

This set is to allow bpf tcp iter to call bpf_(get|set)sockopt.

With bpf-tcp-cc, new algo rollout happens more often.  Instead of
restarting the applications to pick up the new tcp-cc, this set
allows the bpf tcp iter to call bpf_(get|set)sockopt(TCP_CONGESTION).
It is not limited to TCP_CONGESTION, the bpf tcp iter can call
bpf_(get|set)sockopt() with other options.  The bpf tcp iter can read
into all the fields of a tcp_sock, so there is a lot of flexibility
to select the desired sk to do setsockopt(), e.g. it can test for
TCP_LISTEN only and leave the established connections untouched,
or check the addr/port, or check the current tcp-cc name, ...etc.

Patch 1-4 are some cleanup and prep work in the tcp and bpf seq_file.

Patch 5 is to have the tcp seq_file iterate on the
port+addr lhash2 instead of the port only listening_hash.

Patch 6 is to have the bpf tcp iter doing batching which
then allows lock_sock.  lock_sock is needed for setsockopt.

Patch 7 allows the bpf tcp iter to call bpf_(get|set)sockopt.

v2:
- Use __GFP_NOWARN in patch 6
- Add bpf_getsockopt() in patch 7 to give a symmetrical user experience.
  selftest in patch 8 is changed to also cover bpf_getsockopt().
- Remove CAP_NET_ADMIN check in patch 7. Tracing bpf prog has already
  required CAP_SYS_ADMIN or CAP_PERFMON.
- Move some def macros to bpf_tracing_net.h in patch 8
====================

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>

This patch adds tests for the batching and bpf_(get|set)sockopt in
bpf tcp iter.

It first creates:
a) 1 non SO_REUSEPORT listener in lhash2.
b) 256 passive and active fds connected to the listener in (a).
c) 256 SO_REUSEPORT listeners in one of the lhash2 bucket.

The test sets all listeners and connections to bpf_cubic before
running the bpf iter.

The bpf iter then calls setsockopt(TCP_CONGESTION) to switch
each listener and connection from bpf_cubic to bpf_dctcp.

The bpf iter has a random_retry mode such that it can return EAGAIN
to the usespace in the middle of a batch.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200625.1036874-1-kafai@fb.com

This patch allows bpf tcp iter to call bpf_(get|set)sockopt.
To allow a specific bpf iter (tcp here) to call a set of helpers,
get_func_proto function pointer is added to bpf_iter_reg.
The bpf iter is a tracing prog which currently requires
CAP_PERFMON or CAP_SYS_ADMIN, so this patch does not
impose other capability checks for bpf_(get|set)sockopt.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200619.1036715-1-kafai@fb.com

This patch does batching and lock_sock for the bpf tcp iter.
It does not affect the proc fs iteration.

With bpf-tcp-cc, new algo rollout happens more often.  Instead of
restarting the application to pick up the new tcp-cc, the next patch
will allow bpf iter to do setsockopt(TCP_CONGESTION).  This requires
locking the sock.

Also, unlike the proc iteration (cat /proc/net/tcp[6]), the bpf iter
can inspect all fields of a tcp_sock.  It will be useful to have a
consistent view on some of the fields (e.g. the ones reported in
tcp_get_info() that also acquires the sock lock).

Double lock: locking the bucket first and then locking the sock could
lead to deadlock.  This patch takes a batching approach similar to
inet_diag.  While holding the bucket lock, it batch a number of sockets
into an array first and then unlock the bucket.  Before doing show(),
it then calls lock_sock_fast().

In a machine with ~400k connections, the maximum number of
sk in a bucket of the established hashtable is 7.  0.02% of
the established connections fall into this bucket size.

For listen hash (port+addr lhash2), the bucket is usually very
small also except for the SO_REUSEPORT use case which the
userspace could have one SO_REUSEPORT socket per thread.

While batching is used, it can also minimize the chance of missing
sock in the setsockopt use case if the whole bucket is batched.
This patch will start with a batch array with INIT_BATCH_SZ (16)
which will be enough for the most common cases.  bpf_iter_tcp_batch()
will try to realloc to a larger array to handle exception case (e.g.
the SO_REUSEPORT case in the lhash2).

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200613.1036157-1-kafai@fb.com

This patch moves the tcp seq_file iteration on listeners
from the port only listening_hash to the port+addr lhash2.

When iterating from the bpf iter, the next patch will need to
lock the socket such that the bpf iter can call setsockopt (e.g. to
change the TCP_CONGESTION).  To avoid locking the bucket and then locking
the sock, the bpf iter will first batch some sockets from the same bucket
and then unlock the bucket.  If the bucket size is small (which
usually is), it is easier to batch the whole bucket such that it is less
likely to miss a setsockopt on a socket due to changes in the bucket.

However, the port only listening_hash could have many listeners
hashed to a bucket (e.g. many individual VIP(s):443 and also
multiple by the number of SO_REUSEPORT).  We have seen bucket size in
tens of thousands range.  Also, the chance of having changes
in some popular port buckets (e.g. 443) is also high.

The port+addr lhash2 was introduced to solve this large listener bucket
issue.  Also, the listening_hash usage has already been replaced with
lhash2 in the fast path inet[6]_lookup_listener().  This patch follows
the same direction on moving to lhash2 and iterates the lhash2
instead of listening_hash.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200606.1035783-1-kafai@fb.com

The current listening_get_next() is overloaded by passing
NULL to the 2nd arg, like listening_get_next(seq, NULL), to
mean get_first().

This patch moves some logic from the listening_get_next() into
a new function listening_get_first().  It will be equivalent
to the current established_get_first() and established_get_next()
setup.  get_first() is to find a non empty bucket and return
the first sk.  get_next() is to find the next sk of the current
bucket and then resorts to get_first() if the current bucket is
exhausted.

The next patch is to move the listener seq_file iteration from
listening_hash (port only) to lhash2 (port+addr).
Separating out listening_get_first() from listening_get_next()
here will make the following lhash2 changes cleaner and easier to
follow.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200600.1035353-1-kafai@fb.com

A following patch will create a separate struct to store extra
bpf_iter state and it will embed the existing tcp_iter_state like this:
struct bpf_tcp_iter_state {
	struct tcp_iter_state state;
	/* More bpf_iter specific states here ... */
}

As a prep work, this patch removes the
"struct tcp_seq_afinfo *bpf_seq_afinfo" where its purpose is
to tell if it is iterating from bpf_iter instead of proc fs.
Currently, if "*bpf_seq_afinfo" is not NULL, it is iterating from
bpf_iter.  The kernel should not filter by the addr family and
leave this filtering decision to the bpf prog.

Instead of adding a "*bpf_seq_afinfo" pointer, this patch uses the
"seq->op == &bpf_iter_tcp_seq_ops" test to tell if it is iterating
from the bpf iter.

The bpf_iter_(init|fini)_tcp() is left here to prepare for
the change of a following patch.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200554.1034982-1-kafai@fb.com

This patch refactors the net and family matching into
two new helpers, seq_sk_match() and seq_file_family().

seq_file_family() is in the later part of the file to prepare
the change of a following patch.

Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200548.1034629-1-kafai@fb.com

st->bucket stores the current bucket number.
st->offset stores the offset within this bucket that is the sk to be
seq_show().  Thus, st->offset only makes sense within the same
st->bucket.

These two variables are an optimization for the common no-lseek case.
When resuming the seq_file iteration (i.e. seq_start()),
tcp_seek_last_pos() tries to continue from the st->offset
at bucket st->bucket.

However, it is possible that the bucket pointed by st->bucket
has changed and st->offset may end up skipping the whole st->bucket
without finding a sk.  In this case, tcp_seek_last_pos() currently
continues to satisfy the offset condition in the next (and incorrect)
bucket.  Instead, regardless of the offset value, the first sk of the
next bucket should be returned.  Thus, "bucket == st->bucket" check is
added to tcp_seek_last_pos().

The chance of hitting this is small and the issue is a decade old,
so targeting for the next tree.

Fixes: a8b690f9

 ("tcp: Fix slowness in read /proc/net/tcp")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Acked-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20210701200541.1033917-1-kafai@fb.com