Split the introductory that explain eBPF vs classic BPF and how it maps
to hardware from the instruction set specification into a standalone
document.  This duplicates a little bit of information but gives us a
useful reference for the eBPF instrution set that is not encumbered by
classic BPF.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211223101906.977624-3-hch@lst.de

Use normal RST file reference instead of linkage copied from the old filter.rst
document that does not actually work when using HTML output.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211223101906.977624-2-hch@lst.de

Jakub Kicinski says:

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

Last change in the bpf headers - disentangling BPF uapi from
netdevice.h. Both linux/bpf.h and uapi/bpf.h changes should
now rebuild ~1k objects down from the original ~18k. There's
probably more that can be done but it's diminishing returns.

Split into two patches for ease of review.
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

netns/bpf.h gets included by netdevice.h (thru net_namespace.h)
which in turn gets included in a lot of places. We should keep
netns/bpf.h as light-weight as possible.

bpf-netns.h seems to contain more implementation details than
deserves to be included in a netns header. It needs to pull in
uapi/bpf.h to get various enum types.

Move enum netns_bpf_attach_type to netns/bpf.h and invert the
dependency. This makes netns/bpf.h fit the mold of a struct
definition header more clearly, and drops the number of objects
rebuilt when uapi/bpf.h is touched from 7.7k to 1.1k.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211230012742.770642-3-kuba@kernel.org

Add missing includes unmasked by the subsequent change.

Mostly network drivers missing an include for XDP_PACKET_HEADROOM.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211230012742.770642-2-kuba@kernel.org

KP Singh says:

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

Local storage is currently unusable in sleepable helpers. One of the
important use cases of local_storage is to attach security (or
performance) contextual information to kernel objects in LSM / tracing
programs to be used later in the life-cyle of the object.

Sometimes this context can only be gathered from sleepable programs
(because it needs accesing __user pointers or helpers like
bpf_ima_inode_hash). Allowing local storage to be used from sleepable
programs allows such context to be managed with the benefits of
local_storage.

# v2 -> v3

* Fixed some RCU issues pointed by Martin
* Added Martin's ack

# v1 -> v2

* Generalize RCU checks (will send a separate patch for updating
  non local storage code where this can be used).
* Add missing RCU lock checks from v1
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Remove the spin lock logic and update the selftests to use sleepable
programs to use a mix of sleepable and non-sleepable programs. It's more
useful to test the sleepable programs since the tests don't really need
spinlocks.
Signed-off-by: KP Singh <kpsingh@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20211224152916.1550677-3-kpsingh@kernel.org

Other maps like hashmaps are already available to sleepable programs.
Sleepable BPF programs run under trace RCU. Allow task, sk and inode
storage to be used from sleepable programs. This allows sleepable and
non-sleepable programs to provide shareable annotations on kernel
objects.

Sleepable programs run in trace RCU where as non-sleepable programs run
in a normal RCU critical section i.e.  __bpf_prog_enter{_sleepable}
and __bpf_prog_exit{_sleepable}) (rcu_read_lock or rcu_read_lock_trace).

In order to make the local storage maps accessible to both sleepable
and non-sleepable programs, one needs to call both
call_rcu_tasks_trace and call_rcu to wait for both trace and classical
RCU grace periods to expire before freeing memory.

Paul's work on call_rcu_tasks_trace allows us to have per CPU queueing
for call_rcu_tasks_trace. This behaviour can be achieved by setting
rcupdate.rcu_task_enqueue_lim=<num_cpus> boot parameter.

In light of these new performance changes and to keep the local storage
code simple, avoid adding a new flag for sleepable maps / local storage
to select the RCU synchronization (trace / classical).

Also, update the dereferencing of the pointers to use
rcu_derference_check (with either the trace or normal RCU locks held)
with a common bpf_rcu_lock_held helper method.
Signed-off-by: KP Singh <kpsingh@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20211224152916.1550677-2-kpsingh@kernel.org

Without it, kernel crashes in map_get_next_key().

Fixes: 9330986c ("bpf: Add bloom filter map implementation")
Reported-by: TCS Robot <tcs_robot@tencent.com>
Signed-off-by: Haimin Zhang <tcs_kernel@tencent.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Joanne Koong <joannekoong@fb.com>
Link: https://lore.kernel.org/bpf/1640776802-22421-1-git-send-email-tcs.kernel@gmail.com

sock.h is pretty heavily used (5k objects rebuilt on x86 after
it's touched). We can drop the include of filter.h from it and
add a forward declaration of struct sk_filter instead.
This decreases the number of rebuilt objects when bpf.h
is touched from ~5k to ~1k.

There's a lot of missing includes this was masking. Primarily
in networking tho, this time.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Acked-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/bpf/20211229004913.513372-1-kuba@kernel.org

Ubuntu reports incorrect kernel version through uname(), which on older
kernels leads to kprobe BPF programs failing to load due to the version
check mismatch.

Accommodate Ubuntu's quirks with LINUX_VERSION_CODE by using
Ubuntu-specific /proc/version_code to fetch major/minor/patch versions
to form LINUX_VERSION_CODE.

While at it, consolide libbpf's kernel version detection code between
libbpf.c and libbpf_probes.c.

  [0] Closes: https://github.com/libbpf/libbpf/issues/421Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20211222231003.2334940-1-andrii@kernel.org

Improve bpf_tracing.h's macro definition readability by keeping them
single-line and better aligned. This makes it easier to follow all those
variadic patterns.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20211222213924.1869758-2-andrii@kernel.org

Refactor PT_REGS macros definitions in  bpf_tracing.h to avoid excessive
duplication. We currently have classic PT_REGS_xxx() and CO-RE-enabled
PT_REGS_xxx_CORE(). We are about to add also _SYSCALL variants, which
would require excessive copying of all the per-architecture definitions.

Instead, separate architecture-specific field/register names from the
final macro that utilize them. That way for upcoming _SYSCALL variants
we'll be able to just define x86_64 exception and otherwise have one
common set of _SYSCALL macro definitions common for all architectures.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Tested-by: Ilya Leoshkevich <iii@linux.ibm.com>
Acked-by: Yonghong Song <yhs@fb.com>
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Link: https://lore.kernel.org/bpf/20211222213924.1869758-1-andrii@kernel.org

Adding btf_dump__new call to test_cpp, so we can
test C++ compilation with that.
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/20211223131736.483956-2-jolsa@kernel.org

As reported in here [0], C++ compilers don't support
__builtin_types_compatible_p(), so at least don't screw up compilation
for them and let C++ users pick btf_dump__new vs
btf_dump__new_deprecated explicitly.

  [0] https://github.com/libbpf/libbpf/issues/283#issuecomment-986100727

Fixes: 6084f5dc ("libbpf: Ensure btf_dump__new() and btf_dump_opts are future-proof")
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/20211223131736.483956-1-jolsa@kernel.org

The output of bpftool prog tracelog is currently buffered, which is
inconvenient when piping the output into other commands. A simple
tracelog | grep will typically not display anything. This patch fixes it
by enabling line buffering on stdout for the whole bpftool binary.

Fixes: 30da46b5 ("tools: bpftool: add a command to dump the trace pipe")
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Paul Chaignon <paul@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20211220214528.GA11706@Mem

In an effort to avoid open-coded arithmetic in the kernel, use the
struct_size() helper instead of open-coded calculation.
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://github.com/KSPP/linux/issues/160
Link: https://lore.kernel.org/bpf/20211220113048.2859-1-xiujianfeng@huawei.com

Migration of vmtest to libbpf/ci will change the address
of INDEX in vmtest.sh, which will cause vmtest.sh to not
work due to the failure of rootfs fetching.
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Lorenzo Fontana <lorenzo.fontana@elastic.co>
Link: https://lore.kernel.org/bpf/20211220050803.2670677-1-pulehui@huawei.com

Allow passing PTR_TO_CTX, if the kfunc expects a matching struct type,
and punt to PTR_TO_MEM block if reg->type does not fall in one of
PTR_TO_BTF_ID or PTR_TO_SOCK* types. This will be used by future commits
to get access to XDP and TC PTR_TO_CTX, and pass various data (flags,
l4proto, netns_id, etc.) encoded in opts struct passed as pointer to
kfunc.

For PTR_TO_MEM support, arguments are currently limited to pointer to
scalar, or pointer to struct composed of scalars. This is done so that
unsafe scenarios (like passing PTR_TO_MEM where PTR_TO_BTF_ID of
in-kernel valid structure is expected, which may have pointers) are
avoided. Since the argument checking happens basd on argument register
type, it is not easy to ascertain what the expected type is. In the
future, support for PTR_TO_MEM for kfunc can be extended to serve other
usecases. The struct type whose pointer is passed in may have maximum
nesting depth of 4, all recursively composed of scalars or struct with
scalars.

Future commits will add negative tests that check whether these
restrictions imposed for kfunc arguments are duly rejected by BPF
verifier or not.
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217015031.1278167-4-memxor@gmail.com

Hao Luo says:

====================
This patch set consists of two changes:

 - a cleanup of arg_type, ret_type and reg_type which try to make those
   types composable. (patch 1/9 - patch 6/9)
 - a bug fix that prevents bpf programs from writing kernel memory.
   (patch 7/9 - patch 9/9)

The purpose of the cleanup is to find a scalable way to express type
nullness and read-onliness. This patchset introduces two flags that
can be applied on all three types: PTR_MAYBE_NULL and MEM_RDONLY.
Previous types such as ARG_XXX_OR_NULL can now be written as

 ARG_XXX | PTR_MAYBE_NULL

Similarly, PTR_TO_RDONLY_BUF is now "PTR_TO_BUF | MEM_RDONLY".

Flags can be composed, as ARGs can be both MEM_RDONLY and MAYBE_NULL.

 ARG_PTR_TO_MEM | PTR_MAYBE_NULL | MEM_RDONLY

Based on this new composable types, patch 7/9 applies MEM_RDONLY on
PTR_TO_MEM, in order to tag the returned memory from per_cpu_ptr as
read-only. Therefore fixing a previous bug that one can leverage
per_cpu_ptr to modify kernel memory within BPF programs.

Patch 8/9 generalizes the use of MEM_RDONLY further by tagging a set of
helper arguments ARG_PTR_TO_MEM with MEM_RDONLY. Some helper functions
may override their arguments, such as bpf_d_path, bpf_snprintf. In this
patch, we narrow the ARG_PTR_TO_MEM to be compatible with only a subset
of memory types. This prevents these helpers from writing read-only
memories. For the helpers that do not write its arguments, we add tag
MEM_RDONLY to allow taking a RDONLY memory as argument.

Changes since v1:
 - use %u to print base_type(type) instead of %lu. (Andrii, patch 3/9)
 - improve reg_type_str() by appending '_or_null' and prepending 'rdonly_'.
   use preallocated buffer in 'bpf_env'.
 - unified handling of the previous XXX_OR_NULL in adjust_ptr_min_max_vals
   (Andrii, patch 4/9)
 - move PTR_TO_MAP_KEY up to PTR_TO_MAP_VALUE so that we don't have
   to change to drivers that assume the numeric values of bpf_reg.
   (patch 4/9)
 - reintroduce the typo from previous commits in fixes tags (Andrii, patch 7/9)
 - extensive comments on the reason behind folding flags in
   check_reg_type (Andrii, patch 8/9)

Changes since RFC v2:
 - renamed BPF_BASE_TYPE to a more succinct name base_type and move its
   definition to bpf_verifier.h. Same for BPF_TYPE_FLAG. (Alexei)
 - made checking MEM_RDONLY in check_reg_type() universal (Alexei)
 - ran through majority of test_progs and fixed bugs in RFC v2:
   - fixed incorrect BPF_BASE_TYPE_MASK. The high bit of GENMASK should
     be BITS - 1, rather than BITS. patch 1/9.
   - fixed incorrect conditions when checking ARG_PTR_TO_MAP_VALUE in
     check_func_arg(). See patch 2/9.
   - fixed a bug where PTR_TO_BTF_ID may be combined with MEM_RDONLY,
     causing the check in check_mem_access() to fall through to the
     'else' branch. See check_helper_call() in patch 7/9.
 - fixed build failure on netronome driver. Entries in bpf_reg_type have
   been ordered. patch 4/9.
 - fixed build warnings of using '%d' to print base_type. patch 4/9
 - unify arg_type_may_be_null() and reg_type_may_be_null() into a single
   type_may_be_null().

Previous versions:

 v1:
   https://lwn.net/Articles/877938/

 RFC v2:
   https://lwn.net/Articles/877171/

 RFC v1:
   https://lore.kernel.org/bpf/20211109003052.3499225-1-haoluo@google.com/T/
   https://lore.kernel.org/bpf/20211109021624.1140446-8-haoluo@google.com/T/
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

This test verifies that a ksym of non-struct can not be directly
updated.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-10-haoluo@google.com

Some helper functions may modify its arguments, for example,
bpf_d_path, bpf_get_stack etc. Previously, their argument types
were marked as ARG_PTR_TO_MEM, which is compatible with read-only
mem types, such as PTR_TO_RDONLY_BUF. Therefore it's legitimate,
but technically incorrect, to modify a read-only memory by passing
it into one of such helper functions.

This patch tags the bpf_args compatible with immutable memory with
MEM_RDONLY flag. The arguments that don't have this flag will be
only compatible with mutable memory types, preventing the helper
from modifying a read-only memory. The bpf_args that have
MEM_RDONLY are compatible with both mutable memory and immutable
memory.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-9-haoluo@google.com

Tag the return type of {per, this}_cpu_ptr with RDONLY_MEM. The
returned value of this pair of helpers is kernel object, which
can not be updated by bpf programs. Previously these two helpers
return PTR_OT_MEM for kernel objects of scalar type, which allows
one to directly modify the memory. Now with RDONLY_MEM tagging,
the verifier will reject programs that write into RDONLY_MEM.

Fixes: 63d9b80d ("bpf: Introducte bpf_this_cpu_ptr()")
Fixes: eaa6bcb7 ("bpf: Introduce bpf_per_cpu_ptr()")
Fixes: 4976b718 ("bpf: Introduce pseudo_btf_id")
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-8-haoluo@google.com

Remove PTR_TO_MEM_OR_NULL and replace it with PTR_TO_MEM combined with
flag PTR_MAYBE_NULL.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-7-haoluo@google.com

This patch introduce a flag MEM_RDONLY to tag a reg value
pointing to read-only memory. It makes the following changes:

1. PTR_TO_RDWR_BUF -> PTR_TO_BUF
2. PTR_TO_RDONLY_BUF -> PTR_TO_BUF | MEM_RDONLY
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-6-haoluo@google.com

We have introduced a new type to make bpf_reg composable, by
allocating bits in the type to represent flags.

One of the flags is PTR_MAYBE_NULL which indicates a pointer
may be NULL. This patch switches the qualified reg_types to
use this flag. The reg_types changed in this patch include:

1. PTR_TO_MAP_VALUE_OR_NULL
2. PTR_TO_SOCKET_OR_NULL
3. PTR_TO_SOCK_COMMON_OR_NULL
4. PTR_TO_TCP_SOCK_OR_NULL
5. PTR_TO_BTF_ID_OR_NULL
6. PTR_TO_MEM_OR_NULL
7. PTR_TO_RDONLY_BUF_OR_NULL
8. PTR_TO_RDWR_BUF_OR_NULL
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/r/20211217003152.48334-5-haoluo@google.com

We have introduced a new type to make bpf_ret composable, by
reserving high bits to represent flags.

One of the flag is PTR_MAYBE_NULL, which indicates a pointer
may be NULL. When applying this flag to ret_types, it means
the returned value could be a NULL pointer. This patch
switches the qualified arg_types to use this flag.
The ret_types changed in this patch include:

1. RET_PTR_TO_MAP_VALUE_OR_NULL
2. RET_PTR_TO_SOCKET_OR_NULL
3. RET_PTR_TO_TCP_SOCK_OR_NULL
4. RET_PTR_TO_SOCK_COMMON_OR_NULL
5. RET_PTR_TO_ALLOC_MEM_OR_NULL
6. RET_PTR_TO_MEM_OR_BTF_ID_OR_NULL
7. RET_PTR_TO_BTF_ID_OR_NULL

This patch doesn't eliminate the use of these names, instead
it makes them aliases to 'RET_PTR_TO_XXX | PTR_MAYBE_NULL'.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-4-haoluo@google.com

We have introduced a new type to make bpf_arg composable, by
reserving high bits of bpf_arg to represent flags of a type.

One of the flags is PTR_MAYBE_NULL which indicates a pointer
may be NULL. When applying this flag to an arg_type, it means
the arg can take NULL pointer. This patch switches the
qualified arg_types to use this flag. The arg_types changed
in this patch include:

1. ARG_PTR_TO_MAP_VALUE_OR_NULL
2. ARG_PTR_TO_MEM_OR_NULL
3. ARG_PTR_TO_CTX_OR_NULL
4. ARG_PTR_TO_SOCKET_OR_NULL
5. ARG_PTR_TO_ALLOC_MEM_OR_NULL
6. ARG_PTR_TO_STACK_OR_NULL

This patch does not eliminate the use of these arg_types, instead
it makes them an alias to the 'ARG_XXX | PTR_MAYBE_NULL'.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-3-haoluo@google.com

There are some common properties shared between bpf reg, ret and arg
values. For instance, a value may be a NULL pointer, or a pointer to
a read-only memory. Previously, to express these properties, enumeration
was used. For example, in order to test whether a reg value can be NULL,
reg_type_may_be_null() simply enumerates all types that are possibly
NULL. The problem of this approach is that it's not scalable and causes
a lot of duplication. These properties can be combined, for example, a
type could be either MAYBE_NULL or RDONLY, or both.

This patch series rewrites the layout of reg_type, arg_type and
ret_type, so that common properties can be extracted and represented as
composable flag. For example, one can write

 ARG_PTR_TO_MEM | PTR_MAYBE_NULL

which is equivalent to the previous

 ARG_PTR_TO_MEM_OR_NULL

The type ARG_PTR_TO_MEM are called "base type" in this patch. Base
types can be extended with flags. A flag occupies the higher bits while
base types sits in the lower bits.

This patch in particular sets up a set of macro for this purpose. The
following patches will rewrite arg_types, ret_types and reg_types
respectively.
Signed-off-by: Hao Luo <haoluo@google.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211217003152.48334-2-haoluo@google.com

Reimplement bpf_probe_large_insn_limit() in bpftool, as that libbpf API
is scheduled for deprecation in v0.8.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/bpf/20211217171202.3352835-4-andrii@kernel.org

Add selftests for prog/map/prog+helper feature probing APIs. Prog and
map selftests are designed in such a way that they will always test all
the possible prog/map types, based on running kernel's vmlinux BTF enum
definition. This way we'll always be sure that when adding new BPF
program types or map types, libbpf will be always updated accordingly to
be able to feature-detect them.

BPF prog_helper selftest will have to be manually extended with
interesting and important prog+helper combinations, it's easy, but can't
be completely automated.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Link: https://lore.kernel.org/bpf/20211217171202.3352835-3-andrii@kernel.org

Create three extensible alternatives to inconsistently named
feature-probing APIs:

  - libbpf_probe_bpf_prog_type() instead of bpf_probe_prog_type();
  - libbpf_probe_bpf_map_type() instead of bpf_probe_map_type();
  - libbpf_probe_bpf_helper() instead of bpf_probe_helper().

Set up return values such that libbpf can report errors (e.g., if some
combination of input arguments isn't possible to validate, etc), in
addition to whether the feature is supported (return value 1) or not
supported (return value 0).

Also schedule deprecation of those three APIs. Also schedule deprecation
of bpf_probe_large_insn_limit().

Also fix all the existing detection logic for various program and map
types that never worked:

  - BPF_PROG_TYPE_LIRC_MODE2;
  - BPF_PROG_TYPE_TRACING;
  - BPF_PROG_TYPE_LSM;
  - BPF_PROG_TYPE_EXT;
  - BPF_PROG_TYPE_SYSCALL;
  - BPF_PROG_TYPE_STRUCT_OPS;
  - BPF_MAP_TYPE_STRUCT_OPS;
  - BPF_MAP_TYPE_BLOOM_FILTER.

Above prog/map types needed special setups and detection logic to work.
Subsequent patch adds selftests that will make sure that all the
detection logic keeps working for all current and future program and map
types, avoiding otherwise inevitable bit rot.

  [0] Closes: https://github.com/libbpf/libbpf/issues/312Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Dave Marchevsky <davemarchevsky@fb.com>
Cc: Julia Kartseva <hex@fb.com>
Link: https://lore.kernel.org/bpf/20211217171202.3352835-2-andrii@kernel.org

Backtracking information is very verbose, don't print it in log
level 1 to improve readability.
Signed-off-by: Christy Lee <christylee@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20211216213358.3374427-4-christylee@fb.com

Make the verifier logs more readable, print the verifier states
on the corresponding instruction line. If the previous line was
not a bpf instruction, then print the verifier states on its own
line.

Before:

Validating test_pkt_access_subprog3() func#3...
86: R1=invP(id=0) R2=ctx(id=0,off=0,imm=0) R10=fp0
; int test_pkt_access_subprog3(int val, struct __sk_buff *skb)
86: (bf) r6 = r2
87: R2=ctx(id=0,off=0,imm=0) R6_w=ctx(id=0,off=0,imm=0)
87: (bc) w7 = w1
88: R1=invP(id=0) R7_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
; return get_skb_len(skb) * get_skb_ifindex(val, skb, get_constant(123));
88: (bf) r1 = r6
89: R1_w=ctx(id=0,off=0,imm=0) R6_w=ctx(id=0,off=0,imm=0)
89: (85) call pc+9
Func#4 is global and valid. Skipping.
90: R0_w=invP(id=0)
90: (bc) w8 = w0
91: R0_w=invP(id=0) R8_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
; return get_skb_len(skb) * get_skb_ifindex(val, skb, get_constant(123));
91: (b7) r1 = 123
92: R1_w=invP123
92: (85) call pc+65
Func#5 is global and valid. Skipping.
93: R0=invP(id=0)

After:

86: R1=invP(id=0) R2=ctx(id=0,off=0,imm=0) R10=fp0
; int test_pkt_access_subprog3(int val, struct __sk_buff *skb)
86: (bf) r6 = r2                      ; R2=ctx(id=0,off=0,imm=0) R6_w=ctx(id=0,off=0,imm=0)
87: (bc) w7 = w1                      ; R1=invP(id=0) R7_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
; return get_skb_len(skb) * get_skb_ifindex(val, skb, get_constant(123));
88: (bf) r1 = r6                      ; R1_w=ctx(id=0,off=0,imm=0) R6_w=ctx(id=0,off=0,imm=0)
89: (85) call pc+9
Func#4 is global and valid. Skipping.
90: R0_w=invP(id=0)
90: (bc) w8 = w0                      ; R0_w=invP(id=0) R8_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
; return get_skb_len(skb) * get_skb_ifindex(val, skb, get_constant(123));
91: (b7) r1 = 123                     ; R1_w=invP123
92: (85) call pc+65
Func#5 is global and valid. Skipping.
93: R0=invP(id=0)
Signed-off-by: Christy Lee <christylee@fb.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

When printing verifier state for any log level, print full verifier
state only on function calls or on errors. Otherwise, only print the
registers and stack slots that were accessed.

Log size differences:

verif_scale_loop6 before: 234566564
verif_scale_loop6 after: 72143943
69% size reduction

kfree_skb before: 166406
kfree_skb after: 55386
69% size reduction

Before:

156: (61) r0 = *(u32 *)(r1 +0)
157: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1=ctx(id=0,off=0,imm=0) R2_w=invP0 R10=fp0 fp-8_w=00000000 fp-16_w=00\
000000 fp-24_w=00000000 fp-32_w=00000000 fp-40_w=00000000 fp-48_w=00000000 fp-56_w=00000000 fp-64_w=00000000 fp-72_w=00000000 fp-80_w=00000\
000 fp-88_w=00000000 fp-96_w=00000000 fp-104_w=00000000 fp-112_w=00000000 fp-120_w=00000000 fp-128_w=00000000 fp-136_w=00000000 fp-144_w=00\
000000 fp-152_w=00000000 fp-160_w=00000000 fp-168_w=00000000 fp-176_w=00000000 fp-184_w=00000000 fp-192_w=00000000 fp-200_w=00000000 fp-208\
_w=00000000 fp-216_w=00000000 fp-224_w=00000000 fp-232_w=00000000 fp-240_w=00000000 fp-248_w=00000000 fp-256_w=00000000 fp-264_w=00000000 f\
p-272_w=00000000 fp-280_w=00000000 fp-288_w=00000000 fp-296_w=00000000 fp-304_w=00000000 fp-312_w=00000000 fp-320_w=00000000 fp-328_w=00000\
000 fp-336_w=00000000 fp-344_w=00000000 fp-352_w=00000000 fp-360_w=00000000 fp-368_w=00000000 fp-376_w=00000000 fp-384_w=00000000 fp-392_w=\
00000000 fp-400_w=00000000 fp-408_w=00000000 fp-416_w=00000000 fp-424_w=00000000 fp-432_w=00000000 fp-440_w=00000000 fp-448_w=00000000
; return skb->len;
157: (95) exit
Func#4 is safe for any args that match its prototype
Validating get_constant() func#5...
158: R1=invP(id=0) R10=fp0
; int get_constant(long val)
158: (bf) r0 = r1
159: R0_w=invP(id=1) R1=invP(id=1) R10=fp0
; return val - 122;
159: (04) w0 += -122
160: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1=invP(id=1) R10=fp0
; return val - 122;
160: (95) exit
Func#5 is safe for any args that match its prototype
Validating get_skb_ifindex() func#6...
161: R1=invP(id=0) R2=ctx(id=0,off=0,imm=0) R3=invP(id=0) R10=fp0
; int get_skb_ifindex(int val, struct __sk_buff *skb, int var)
161: (bc) w0 = w3
162: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1=invP(id=0) R2=ctx(id=0,off=0,imm=0) R3=invP(id=0) R10=fp0

After:

156: (61) r0 = *(u32 *)(r1 +0)
157: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R1=ctx(id=0,off=0,imm=0)
; return skb->len;
157: (95) exit
Func#4 is safe for any args that match its prototype
Validating get_constant() func#5...
158: R1=invP(id=0) R10=fp0
; int get_constant(long val)
158: (bf) r0 = r1
159: R0_w=invP(id=1) R1=invP(id=1)
; return val - 122;
159: (04) w0 += -122
160: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff))
; return val - 122;
160: (95) exit
Func#5 is safe for any args that match its prototype
Validating get_skb_ifindex() func#6...
161: R1=invP(id=0) R2=ctx(id=0,off=0,imm=0) R3=invP(id=0) R10=fp0
; int get_skb_ifindex(int val, struct __sk_buff *skb, int var)
161: (bc) w0 = w3
162: R0_w=invP(id=0,umax_value=4294967295,var_off=(0x0; 0xffffffff)) R3=invP(id=0)
Signed-off-by: Christy Lee <christylee@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20211216213358.3374427-2-christylee@fb.com

Jakub Kicinski says:

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

Changes to bpf.h tend to clog up our build systems. The netdev/bpf
build bot does incremental builds to save time (reusing the build
directory to only rebuild changed objects).

This is the rough breakdown of how many objects needs to be rebuilt
based on file touched:

kernel.h      40633
bpf.h         17881
bpf-cgroup.h  17875
skbuff.h      10696
bpf-netns.h    7604
netdevice.h    7452
filter.h       5003
sock.h         4959
tcp.h          4048

As the stats show touching bpf.h is _very_ expensive.

Bulk of the objects get rebuilt because MM includes cgroup headers.
Luckily bpf-cgroup.h does not fundamentally depend on bpf.h so we
can break that dependency and reduce the number of objects.

With the patches applied touching bpf.h causes 5019 objects to be rebuilt
(17881 / 5019 = 3.56x). That's pretty much down to filter.h plus noise.

v2:
Try to make the new headers wider in scope. Collapse bpf-link and
bpf-cgroup-types into one header, which may serve as "BPF kernel
API" header in the future if needed. Rename bpf-cgroup-storage.h
to bpf-inlines.h.

Add a fix for the s390 build issue.

v3: https://lore.kernel.org/all/20211215061916.715513-1-kuba@kernel.org/
Merge bpf-includes.h into bpf.h.

v4: https://lore.kernel.org/all/20211215181231.1053479-1-kuba@kernel.org/
Change course - break off cgroup instead of breaking off bpf.

v5:
Add forward declaration of struct bpf_prog to perf_event.h
when !CONFIG_BPF_SYSCALL (kbuild bot).
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Remove the dependency from cgroup-defs.h to bpf-cgroup.h and bpf.h.
This reduces the incremental build size of x86 allmodconfig after
bpf.h was touched from ~17k objects rebuilt to ~5k objects.
bpf.h is 2.2kLoC and is modified relatively often.

We need a new header with just the definition of struct cgroup_bpf
and enum cgroup_bpf_attach_type, this is akin to cgroup-defs.h.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/bpf/20211216025538.1649516-4-kuba@kernel.org

We're about to break the cgroup-defs.h -> bpf-cgroup.h dependency,
make sure those who actually need more than the definition of
struct cgroup_bpf include bpf-cgroup.h explicitly.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Tejun Heo <tj@kernel.org>
Link: https://lore.kernel.org/bpf/20211216025538.1649516-3-kuba@kernel.org

cgroup pulls in BPF which pulls in a lot of includes.
We're about to break that chain so fix those who were
depending on it.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20211216025538.1649516-2-kuba@kernel.org

Jean-Philippe Brucker says:

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

Since v1 [1], I added Quentin's acks and applied Andrii's suggestions:
* Pass CFLAGS to libbpf link in patch 3
* Substitute CLANG_CROSS_FLAGS whole in HOST_CFLAGS to avoid accidents,
  patch 4

Add support for cross-building BPF tools and selftests with clang, by
passing LLVM=1 or CC=clang to make, as well as CROSS_COMPILE. A single
clang toolchain can generate binaries for multiple architectures, so
instead of having prefixes such as aarch64-linux-gnu-gcc, clang uses the
-target parameter: `clang -target aarch64-linux-gnu'.

Patch 1 adds the parameter in Makefile.include so tools can easily
support this. Patch 2 prepares for the libbpf change from patch 3 (keep
building resolve_btfids's libbpf in the host arch, when cross-building
the kernel with clang). Patches 3-6 enable cross-building BPF tools with
clang.

[1] https://lore.kernel.org/bpf/20211122192019.1277299-1-jean-philippe@linaro.org/
====================
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>