kernel.git/tools/bpf/bpftool/gen.c, branch linux-6.1.y

bpftool: Fix error message of strerror

2022-09-30T22:40:46Z

strerror() expects a positive errno, however variable err will never be positive when an error occurs. This causes bpftool to output too many "unknown error", even a simple "file not exist" error can not get an accurate message. This patch fixed all "strerror(err)" patterns in bpftool. Specially in btf.c#L823, hashmap__append() is an internal function of libbpf and will not change errno, so there's a little difference. Some libbpf_get_error() calls are kept for return values. Changes since v1: https://lore.kernel.org/bpf/SY4P282MB1084B61CD8671DFA395AA8579D539@SY4P282MB1084.AUSP282.PROD.OUTLOOK.COM/ Check directly for NULL values instead of calling libbpf_get_error(). Signed-off-by: Tianyi Liu Signed-off-by: Andrii Nakryiko Reviewed-by: Quentin Monnet Link: https://lore.kernel.org/bpf/SY4P282MB1084AD9CD84A920F08DF83E29D549@SY4P282MB1084.AUSP282.PROD.OUTLOOK.COM

bpftool: Don't try to return value from void function in skeleton

2022-07-29T17:43:14Z

A skeleton generated by bpftool previously contained a return followed by an expression in OBJ_NAME__detach(), which has return type void. This did not hurt, the bpf_object__detach_skeleton() called there returns void itself anyway, but led to a warning when compiling with e.g. -pedantic. Signed-off-by: Jörn-Thorben Hinz Signed-off-by: Andrii Nakryiko Reviewed-by: Quentin Monnet Link: https://lore.kernel.org/bpf/20220726133203.514087-1-jthinz@mailbox.tu-berlin.de

bpftool: Add support for KIND_RESTRICT to gen min_core_btf command

2022-07-08T12:26:39Z

This change adjusts bpftool's type marking logic, as used in conjunction with TYPE_EXISTS relocations, to correctly recognize and handle the RESTRICT BTF kind. Suggested-by: Andrii Nakryiko Signed-off-by: Daniel Müller Signed-off-by: Daniel Borkmann Acked-by: Quentin Monnet Link: https://lore.kernel.org/bpf/20220623212205.2805002-1-deso@posteo.net/T/#m4c75205145701762a4b398e0cdb911d5b5305ffc Link: https://lore.kernel.org/bpf/20220706212855.1700615-2-deso@posteo.net

bpftool: Honor BPF_CORE_TYPE_MATCHES relocation

2022-07-06T03:26:18Z

bpftool needs to know about the newly introduced BPF_CORE_TYPE_MATCHES relocation for its 'gen min_core_btf' command to work properly in the present of this relocation. Specifically, we need to make sure to mark types and fields so that they are present in the minimized BTF for "type match" checks to work out. However, contrary to the existing btfgen_record_field_relo, we need to rely on the BTF -- and not the spec -- to find fields. With this change we handle this new variant correctly. The functionality will be tested with follow on changes to BPF selftests, which already run against a minimized BTF created with bpftool. Signed-off-by: Daniel Müller Signed-off-by: Andrii Nakryiko Acked-by: Quentin Monnet Link: https://lore.kernel.org/bpf/20220628160127.607834-3-deso@posteo.net

bpftool: Add btf enum64 support

2022-06-07T17:20:43Z

Add BTF_KIND_ENUM64 support. For example, the following enum is defined in uapi bpf.h. $ cat core.c enum A { BPF_F_INDEX_MASK = 0xffffffffULL, BPF_F_CURRENT_CPU = BPF_F_INDEX_MASK, BPF_F_CTXLEN_MASK = (0xfffffULL << 32), } g; Compiled with clang -target bpf -O2 -g -c core.c Using bpftool to dump types and generate format C file: $ bpftool btf dump file core.o ... [1] ENUM64 'A' encoding=UNSIGNED size=8 vlen=3 'BPF_F_INDEX_MASK' val=4294967295ULL 'BPF_F_CURRENT_CPU' val=4294967295ULL 'BPF_F_CTXLEN_MASK' val=4503595332403200ULL $ bpftool btf dump file core.o format c ... enum A { BPF_F_INDEX_MASK = 4294967295ULL, BPF_F_CURRENT_CPU = 4294967295ULL, BPF_F_CTXLEN_MASK = 4503595332403200ULL, }; ... Note that for raw btf output, the encoding (UNSIGNED or SIGNED) is printed out as well. The 64bit value is also represented properly in BTF and C dump. Acked-by: Andrii Nakryiko Signed-off-by: Yonghong Song Link: https://lore.kernel.org/r/20220607062652.3722649-1-yhs@fb.com Signed-off-by: Alexei Starovoitov

bpftool: Check for NULL ptr of btf in codegen_asserts

2022-06-02T23:26:43Z

bpf_object__btf() can return a NULL value. If bpf_object__btf returns null, do not progress through codegen_asserts(). This avoids a null ptr dereference at the call btf__type_cnt() in the function find_type_for_map() Signed-off-by: Michael Mullin Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20220523194917.igkgorco42537arb@jup

bpftool: bpf_link_get_from_fd support for LSM programs in lskel

2022-05-10T17:42:08Z

bpf_link_get_from_fd currently returns a NULL fd for LSM programs. LSM programs are similar to tracing programs and can also use skel_raw_tracepoint_open. Signed-off-by: KP Singh Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/20220509214905.3754984-1-kpsingh@kernel.org

bpftool: Declare generator name

2022-05-10T00:42:53Z

Most code generators declare its name so did this for bfptool. Signed-off-by: Jason Wang Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20220509090247.5457-1-jasowang@redhat.com

bpftool: Explicit errno handling in skeletons

2022-03-30T12:06:59Z

Andrii noticed that since f97b8b9bd630 ("bpftool: Fix a bug in subskeleton code generation") the subskeleton code allows bpf_object__destroy_subskeleton to overwrite the errno that subskeleton__open would return with. While this is not currently an issue, let's make it future-proof. This patch explicitly tracks err in subskeleton__open and skeleton__create (i.e. calloc failure is explicitly ENOMEM) and ensures that errno is -err on the error return path. The skeleton code had to be changed since maps and progs codegen is shared with subskeletons. Fixes: f97b8b9bd630 ("bpftool: Fix a bug in subskeleton code generation") Signed-off-by: Delyan Kratunov Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/3b6bfbb770c79ae64d8de26c1c1bd9d53a4b85f8.camel@fb.com

bpftool: Fix generated code in codegen_asserts

2022-03-29T02:10:25Z

Arnaldo reported perf compilation fail with: $ make -k BUILD_BPF_SKEL=1 CORESIGHT=1 PYTHON=python3 ... In file included from util/bpf_counter.c:28: /tmp/build/perf//util/bpf_skel/bperf_leader.skel.h: In function ‘bperf_leader_bpf__assert’: /tmp/build/perf//util/bpf_skel/bperf_leader.skel.h:351:51: error: unused parameter ‘s’ [-Werror=unused-parameter] 351 | bperf_leader_bpf__assert(struct bperf_leader_bpf *s) | ~~~~~~~~~~~~~~~~~~~~~~~~~^ cc1: all warnings being treated as errors If there's nothing to generate in the new assert function, we will get unused 's' warn/error, adding 'unused' attribute to it. Fixes: 08d4dba6ae77 ("bpftool: Bpf skeletons assert type sizes") Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Jiri Olsa Signed-off-by: Alexei Starovoitov Tested-by: Arnaldo Carvalho de Melo Link: https://lore.kernel.org/bpf/20220328083703.2880079-1-jolsa@kernel.org