Skip to content

L
linux
Switch branch/tag
  1. 11 Oct, 2019 2 commits
  3. 10 Oct, 2019 4 commits
  5. 09 Oct, 2019 5 commits
  7. 08 Oct, 2019 14 commits
  9. 07 Oct, 2019 5 commits
  11. 06 Oct, 2019 7 commits
    • Toke Høiland-Jørgensen's avatar
      libbpf: Add cscope and tags targets to Makefile · a9eb048d
      Toke Høiland-Jørgensen authored 
      Using cscope and/or TAGS files for navigating the source code is useful.
Add simple targets to the Makefile to generate the index files for both
tools.
Signed-off-by: default avatarToke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
Tested-by: default avatarAndrii Nakryiko <andriin@fb.com>
Acked-by: default avatarAndrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191004153444.1711278-1-toke@redhat.com
      a9eb048d
    • Alexei Starovoitov's avatar
      Merge branch 'libbpf-api' · b84fbfe2
      Alexei Starovoitov authored 
      Andrii Nakryiko says:

====================
Add bpf_object__open_file() and bpf_object__open_mem() APIs that use a new
approach to providing future-proof non-ABI-breaking API changes. It relies on
APIs accepting optional self-describing "opts" struct, containing its own
size, filled out and provided by potentially outdated (as well as
newer-than-libbpf) user application. A set of internal helper macros
(OPTS_VALID, OPTS_HAS, and OPTS_GET) streamline and simplify a graceful
handling forward and backward compatibility for user applications dynamically
linked against different versions of libbpf shared library.

Users of libbpf are provided with convenience macro LIBBPF_OPTS that takes
care of populating correct structure size and zero-initializes options struct,
which helps avoid obscure issues of unitialized padding. Uninitialized padding
in a struct might turn into garbage-populated new fields understood by future
versions of libbpf.

Patch #1 removes enforcement of kern_version in libbpf and always populates
correct one on behalf of users.
Patch #2 defines necessary infrastructure for options and two new open APIs
relying on it.
Patch #3 fixes bug in bpf_object__name().
Patch #4 switches two of test_progs' tests to use new APIs as a validation
that they work as expected.

v2->v3:
- fix LIBBPF_OPTS() to ensure zero-initialization of padded bytes;
- pass through name override and relaxed maps flag for open_file() (Toke);
- fix bpf_object__name() to actually return object name;
- don't bother parsing and verifying version section (John);

v1->v2:
- use better approach for tracking last field in opts struct;
- convert few tests to new APIs for validation;
- fix bug with using offsetof(last_field) instead of offsetofend(last_field).
====================
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
      b84fbfe2
    • Andrii Nakryiko's avatar
      selftests/bpf: switch tests to new bpf_object__open_{file, mem}() APIs · 928ca75e
      Andrii Nakryiko authored 
      Verify new bpf_object__open_mem() and bpf_object__open_file() APIs work
as expected by switching test_attach_probe test to use embedded BPF
object and bpf_object__open_mem() and test_reference_tracking to
bpf_object__open_file().
Signed-off-by: default avatarAndrii Nakryiko <andriin@fb.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
      928ca75e
    • Andrii Nakryiko's avatar
      libbpf: fix bpf_object__name() to actually return object name · c9e4c301
      Andrii Nakryiko authored 
      bpf_object__name() was returning file path, not name. Fix this.
Signed-off-by: default avatarAndrii Nakryiko <andriin@fb.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
      c9e4c301
    • Andrii Nakryiko's avatar
      libbpf: add bpf_object__open_{file, mem} w/ extensible opts · 2ce8450e
      Andrii Nakryiko authored 
      Add new set of bpf_object__open APIs using new approach to optional
parameters extensibility allowing simpler ABI compatibility approach.

This patch demonstrates an approach to implementing libbpf APIs that
makes it easy to extend existing APIs with extra optional parameters in
such a way, that ABI compatibility is preserved without having to do
symbol versioning and generating lots of boilerplate code to handle it.
To facilitate succinct code for working with options, add OPTS_VALID,
OPTS_HAS, and OPTS_GET macros that hide all the NULL, size, and zero
checks.

Additionally, newly added libbpf APIs are encouraged to follow similar
pattern of having all mandatory parameters as formal function parameters
and always have optional (NULL-able) xxx_opts struct, which should
always have real struct size as a first field and the rest would be
optional parameters added over time, which tune the behavior of existing
API, if specified by user.
Signed-off-by: default avatarAndrii Nakryiko <andriin@fb.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
      2ce8450e
    • Andrii Nakryiko's avatar
      libbpf: stop enforcing kern_version, populate it for users · 5e61f270
      Andrii Nakryiko authored 
      Kernel version enforcement for kprobes/kretprobes was removed from
5.0 kernel in 6c4fc209 ("bpf: remove useless version check for prog load").
Since then, BPF programs were specifying SEC("version") just to please
libbpf. We should stop enforcing this in libbpf, if even kernel doesn't
care. Furthermore, libbpf now will pre-populate current kernel version
of the host system, in case we are still running on old kernel.

This patch also removes __bpf_object__open_xattr from libbpf.h, as
nothing in libbpf is relying on having it in that header. That function
was never exported as LIBBPF_API and even name suggests its internal
version. So this should be safe to remove, as it doesn't break ABI.
Signed-off-by: default avatarAndrii Nakryiko <andriin@fb.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
      5e61f270
    • Andrii Nakryiko's avatar
      libbpf: Fix BTF-defined map's __type macro handling of arrays · a53ba15d
      Andrii Nakryiko authored 
      Due to a quirky C syntax of declaring pointers to array or function
prototype, existing __type() macro doesn't work with map key/value types
that are array or function prototype. One has to create a typedef first
and use it to specify key/value type for a BPF map.  By using typeof(),
pointer to type is now handled uniformly for all kinds of types. Convert
one of self-tests as a demonstration.
Signed-off-by: default avatarAndrii Nakryiko <andriin@fb.com>
Signed-off-by: default avatarAlexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20191004040211.2434033-1-andriin@fb.com
      a53ba15d
  13. 04 Oct, 2019 2 commits
  15. 03 Oct, 2019 1 commit
    • Ivan Khoronzhuk's avatar
      selftests/bpf: Correct path to include msg + path · c5881463
      Ivan Khoronzhuk authored 
      The "path" buf is supposed to contain path + printf msg up to 24 bytes.
It will be cut anyway, but compiler generates truncation warns like:

"
samples/bpf/../../tools/testing/selftests/bpf/cgroup_helpers.c: In
function ‘setup_cgroup_environment’:
samples/bpf/../../tools/testing/selftests/bpf/cgroup_helpers.c:52:34:
warning: ‘/cgroup.controllers’ directive output may be truncated
writing 19 bytes into a region of size between 1 and 4097
[-Wformat-truncation=]
snprintf(path, sizeof(path), "%s/cgroup.controllers", cgroup_path);
				  ^~~~~~~~~~~~~~~~~~~
samples/bpf/../../tools/testing/selftests/bpf/cgroup_helpers.c:52:2:
note: ‘snprintf’ output between 20 and 4116 bytes into a destination
of size 4097
snprintf(path, sizeof(path), "%s/cgroup.controllers", cgroup_path);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
samples/bpf/../../tools/testing/selftests/bpf/cgroup_helpers.c:72:34:
warning: ‘/cgroup.subtree_control’ directive output may be truncated
writing 23 bytes into a region of size between 1 and 4097
[-Wformat-truncation=]
snprintf(path, sizeof(path), "%s/cgroup.subtree_control",
				  ^~~~~~~~~~~~~~~~~~~~~~~
cgroup_path);
samples/bpf/../../tools/testing/selftests/bpf/cgroup_helpers.c:72:2:
note: ‘snprintf’ output between 24 and 4120 bytes into a destination
of size 4097
snprintf(path, sizeof(path), "%s/cgroup.subtree_control",
cgroup_path);
"

In order to avoid warns, lets decrease buf size for cgroup workdir on
24 bytes with assumption to include also "/cgroup.subtree_control" to
the address. The cut will never happen anyway.
Signed-off-by: default avatarIvan Khoronzhuk <ivan.khoronzhuk@linaro.org>
Signed-off-by: default avatarDaniel Borkmann <daniel@iogearbox.net>
Acked-by: default avatarSong Liu <songliubraving@fb.com>
Link: https://lore.kernel.org/bpf/20191002120404.26962-3-ivan.khoronzhuk@linaro.org
      c5881463
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7