- 25 Oct, 2021 5 commits
-
-
Alexei Starovoitov authored
Andrii Nakryiko says: ==================== Reduce amount of waiting time when running test_progs in parallel mode (-j) by splitting bpf_verif_scale selftests into multiple tests. Previously it was structured as a test with multiple subtests, but subtests are not easily parallelizable with test_progs' infra. Also in practice each scale subtest is really an independent test with nothing shared across all substest. This patch set changes how test_progs test discovery works. Now it is possible to define multiple tests within a single source code file. One of the patches also marks tc_redirect selftests as serial, because it's extremely harmful to the test system when run in parallel mode. ==================== Acked-by:
Yucong Sun <sunyucong@gmail.com> Signed-off-by:
Alexei Starovoitov <ast@kernel.org>
-
Andrii Nakryiko authored
Instead of using subtests in bpf_verif_scale selftest, turn each scale sub-test into its own test. Each subtest is compltely independent and just reuses a bit of common test running logic, so the conversion is trivial. For convenience, keep all of BPF verifier scale tests in one file. This conversion shaves off a significant amount of time when running test_progs in parallel mode. E.g., just running scale tests (-t verif_scale): BEFORE ====== Summary: 24/0 PASSED, 0 SKIPPED, 0 FAILED real 0m22.894s user 0m0.012s sys 0m22.797s AFTER ===== Summary: 24/0 PASSED, 0 SKIPPED, 0 FAILED real 0m12.044s user 0m0.024s sys 0m27.869s Ten second saving right there. test_progs -j is not yet ready to be turned on by default, unfortunately, and some tests fail almost every time, but this is a good improvement nevertheless. Ignoring few failures, here is sequential vs parallel run times when running all tests now: SEQUENTIAL ========== Summary: 206/953 PASSED, 4 SKIPPED, 0 FAILED real 1m5.625s user 0m4.211s sys 0m31.650s PARALLEL ======== Summary: 204/952 PASSED, 4 SKIPPED, 2 FAILED real 0m35.550s user 0m4.998s sys 0m39.890s Signed-off-by:
Andrii Nakryiko <andrii@kernel.org> Signed-off-by:
-
Andrii Nakryiko authored
It seems to cause a lot of harm to kprobe/tracepoint selftests. Yucong mentioned before that it does manipulate sysfs, which might be the reason. So let's mark it as serial, though ideally it would be less intrusive on the system at test. Signed-off-by:
-
Andrii Nakryiko authored
Revamp how test discovery works for test_progs and allow multiple test entries per file. Any global void function with no arguments and serial_test_ or test_ prefix is considered a test. Signed-off-by:
-
Andrii Nakryiko authored
Ensure that all test entry points are global void functions with no input arguments. Mark few subtest entry points as static. Signed-off-by:
-
- 23 Oct, 2021 8 commits
-
-
Andrii Nakryiko authored
Original code assumed fixed and correct BTF header length. That's not always the case, though, so fix this bug with a proper additional check. And use actual header length instead of sizeof(struct btf_header) in sanity checks. Fixes: 8a138aed ("bpf: btf: Add BTF support to libbpf") Reported-by:
Evgeny Vereshchagin <evvers@ya.ru> Signed-off-by:
-
Andrii Nakryiko authored
btf_header's str_off+str_len or type_off+type_len can overflow as they are u32s. This will lead to bypassing the sanity checks during BTF parsing, resulting in crashes afterwards. Fix by using 64-bit signed integers for comparison. Fixes: d8123624 ("libbpf: Fix BTF data layout checks and allow empty BTF") Reported-by:
-
Alexei Starovoitov authored
Yonghong Song says: ==================== Latest upstream llvm-project added support for btf_decl_tag attributes for typedef declarations ([1], [2]). Similar to other btf_decl_tag cases, func/func_param/global_var/struct/union/field, btf_decl_tag with typedef declaration can carry information from kernel source to clang compiler and then to dwarf/BTF, for bpf verification or other use cases. This patch set added kernel support for BTF_KIND_DECL_TAG to typedef declaration (Patch 1). Additional selftests are added to cover unit testing, dedup, or bpf program usage of btf_decl_tag with typedef. (Patches 2, 3 and 4). The btf documentation is updated to include BTF_KIND_DECL_TAG typedef (Patch 5). [1] https://reviews.llvm.org/D110127 [2] https://reviews.llvm.org/D112259 ==================== Signed-off-by:
-
Yonghong Song authored
Add BTF_KIND_DECL_TAG typedef support in btf.rst. Signed-off-by:
Yonghong Song <yhs@fb.com> Signed-off-by:
-
Yonghong Song authored
Change value type in progs/tag.c to a typedef with a btf_decl_tag. With `bpftool btf dump file tag.o`, we have ... [14] TYPEDEF 'value_t' type_id=17 [15] DECL_TAG 'tag1' type_id=14 component_idx=-1 [16] DECL_TAG 'tag2' type_id=14 component_idx=-1 [17] STRUCT '(anon)' size=8 vlen=2 'a' type_id=2 bits_offset=0 'b' type_id=2 bits_offset=32 ... The btf_tag selftest also succeeded: $ ./test_progs -t tag #21 btf_tag:OK Summary: 1/0 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: -
Yonghong Song authored
Add unit tests for deduplication of BTF_KIND_DECL_TAG to typedef types. Also changed a few comments from "tag" to "decl_tag" to match BTF_KIND_DECL_TAG enum value name. Signed-off-by:
-
Yonghong Song authored
Test good and bad variants of typedef BTF_KIND_DECL_TAG encoding. Signed-off-by:
-
Yonghong Song authored
The llvm patches ([1], [2]) added support to attach btf_decl_tag attributes to typedef declarations. This patch added support in kernel. [1] https://reviews.llvm.org/D110127 [2] https://reviews.llvm.org/D112259Signed-off-by:
-
- 22 Oct, 2021 24 commits
-
-
Andrii Nakryiko authored
Stanislav Fomichev says: ==================== Commit 15669e1d ("selftests/bpf: Normalize all the rest SEC() uses") broke flow dissector tests. With the strict section names, bpftool isn't able to pin all programs of the objects (all section names are the same now). To bring it back to life let's do the following: - teach libbpf to pin by func name with LIBBPF_STRICT_SEC_NAME - enable strict mode in bpftool (breaking cli change) - fix custom flow_dissector loader to use strict mode - fix flow_dissector tests to use new pin names (func vs sec) v5: - get rid of error when retrying with '/' (Quentin Monnet) v4: - fix comment spelling (Quentin Monnet) - retry progtype without / (Quentin Monnet) v3: - clarify program pinning in LIBBPF_STRICT_SEC_NAME, for real this time (Andrii Nakryiko) - fix possible segfault in __bpf_program__pin_name (Andrii Nakryiko) v2: - add github issue (Andrii Nakryiko) - remove sec_name from bpf_program.pin_name comment (Andrii Nakryiko) - add cover letter (Andrii Nakryiko) ==================== Signed-off-by:
-
Stanislav Fomichev authored
- update custom loader to search by name, not section name - update bpftool commands to use proper pin path Signed-off-by:
Stanislav Fomichev <sdf@google.com> Signed-off-by:
-
Stanislav Fomichev authored
We can't use section name anymore because they are not unique and pinning objects with multiple programs with the same progtype/secname will fail. [0] Closes: https://github.com/libbpf/libbpf/issues/273 Fixes: 33a2c75c ("libbpf: add internal pin_name") Signed-off-by:
Quentin Monnet <quentin@isovalent.com> Link: https://lore.kernel.org/bpf/20211021214814.1236114-2-sdf@google.com
-
Quentin Monnet authored
Bpftool creates a new JSON object for writing program metadata in plain text mode, regardless of metadata being present or not. Then this writer is freed if any metadata has been found and printed, but it leaks otherwise. We cannot destroy the object unconditionally, because the destructor prints an undesirable line break. Instead, make sure the writer is created only after we have found program metadata to print. Found with valgrind. Fixes: aff52e68 ("bpftool: Support dumping metadata") Signed-off-by:
-
Andrii Nakryiko authored
Hengqi Chen says: ==================== Add btf__type_cnt() and btf__raw_data() APIs and deprecate btf__get_nr_type() and btf__get_raw_data() since the old APIs don't follow libbpf naming convention. Also update tools/selftests to use these new APIs. This is part of effort towards libbpf v1.0 v1->v2: - Update commit message, deprecate the old APIs in libbpf v0.7 (Andrii) - Separate changes in tools/ to individual patches (Andrii) ==================== Signed-off-by: -
Hengqi Chen authored
Replace the calls to btf__get_nr_types/btf__get_raw_data in selftests with new APIs btf__type_cnt/btf__raw_data. The old APIs will be deprecated in libbpf v0.7+. Signed-off-by:
Hengqi Chen <hengqi.chen@gmail.com> Signed-off-by:
-
Hengqi Chen authored
Replace the call to btf__get_nr_types with new API btf__type_cnt. The old API will be deprecated in libbpf v0.7+. No functionality change. Signed-off-by:
-
Hengqi Chen authored
Replace the call to btf__get_nr_types with new API btf__type_cnt. The old API will be deprecated in libbpf v0.7+. No functionality change. Signed-off-by:
-
Hengqi Chen authored
Replace the call to btf__get_raw_data with new API btf__raw_data. The old APIs will be deprecated in libbpf v0.7+. No functionality change. Signed-off-by:
-
Hengqi Chen authored
Add btf__type_cnt() and btf__raw_data() APIs and deprecate btf__get_nr_type() and btf__get_raw_data() since the old APIs don't follow the libbpf naming convention for getters which omit 'get' in the name (see [0]). btf__raw_data() is just an alias to the existing btf__get_raw_data(). btf__type_cnt() now returns the number of all types of the BTF object including 'void'. [0] Closes: https://github.com/libbpf/libbpf/issues/279Signed-off-by:
-
Mauricio Vásquez authored
Free btf_dedup if btf_ensure_modifiable() returns error. Fixes: 919d2b1d ("libbpf: Allow modification of BTF and add btf__add_str API") Signed-off-by:
Mauricio Vásquez <mauricio@kinvolk.io> Signed-off-by:
-
Andrii Nakryiko authored
Recent change to use tp/syscalls/sys_enter_nanosleep for perf_buffer selftests causes this selftest to fail on 4.9 kernel in libbpf CI ([0]): libbpf: prog 'handle_sys_enter': failed to attach to perf_event FD 6: Invalid argument libbpf: prog 'handle_sys_enter': failed to attach to tracepoint 'syscalls/sys_enter_nanosleep': Invalid argument It's not exactly clear why, because perf_event itself is created for this tracepoint, but I can't even compile 4.9 kernel locally, so it's hard to figure this out. If anyone has better luck and would like to help investigating this, I'd really appreciate this. For now, unblock CI by switching back to raw_syscalls/sys_enter, but reduce amount of unnecessary samples emitted by filter by process ID. Use explicit ARRAY map for that to make it work on 4.9 as well, because global data isn't yet supported there. Fixes: aa274f98 ("selftests/bpf: Fix possible/online index mismatch in perf_buffer test") Signed-off-by:
-
Andrii Nakryiko authored
Building libbpf sources out of kernel tree (in Github repo) we run into compilation error due to unknown __aligned attribute. It must be coming from some kernel header, which is not available to Github sources. Use explicit __attribute__((aligned(16))) instead. Fixes: 961632d5 ("libbpf: Fix dumping non-aligned __int128") Signed-off-by:
-
Alexei Starovoitov authored
Andrii Nakryiko says: ==================== This patch set refactors internals of libbpf to enable support for multiple custom .rodata.* and .data.* sections. Each such section is backed by its own BPF_MAP_TYPE_ARRAY, memory-mappable just like .rodata/.data. This is not extended to .bss because .bss is not a great name, it is generated by compiler with name that reflects completely irrelevant historical implementation details. Given that users have to annotate their variables with SEC(".data.my_sec") explicitly, standardizing on .rodata. and .data. prefixes makes more sense and keeps things simpler. Additionally, this patch set makes it simpler to work with those special internal maps by allowing to look them up by their full ELF section name. Patch #1 is a preparatory patch that deprecates one libbpf API and moves custom logic into libbpf.c, where it's used. This code is later refactored with the rest of libbpf.c logic to support multiple data section maps. See individual patches for all the details. For new custom "dot maps", their full ELF section names are used as the names that are sent into the kernel. Object name isn't prepended like for .data/.rodata/.bss. The reason is that with longer custom names, there isn't much space left for object name anyways. Also, if BTF is supported, btf_value_type_id points to DATASEC BTF type, which contains full original ELF name of the section, so tools like bpftool could use that to recover full name. This patch set doesn't add this logic yet, this is left for follow up patches. One interesting possibility that is now open by these changes is that it's possible to do: bpf_trace_printk("My fmt %s", sizeof("My fmt %s"), "blah"); and it will work as expected. I haven't updated libbpf-provided helpers in bpf_helpers.h for snprintf, seq_printf, and printk, because using `static const char ___fmt[] = fmt;` trick is still efficient and doesn't fill out the buffer at runtime (no copying). But we might consider updating them in the future, especially with the array check that Kumar proposed (see [0]). [0] https://lore.kernel.org/bpf/20211012041524.udytbr2xs5wid6x2@apollo.localdomain/ v1->v2: - don't prepend object name for new dot maps; - add __read_mostly example in selftests (Daniel). ==================== Signed-off-by: -
Andrii Nakryiko authored
Utilize libbpf's feature of allowing to lookup internal maps by their ELF section names. No need to guess or calculate the exact truncated prefix taken from the object name. Signed-off-by:
Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20211021014404.2635234-11-andrii@kernel.org
-
Andrii Nakryiko authored
Map name that's assigned to internal maps (.rodata, .data, .bss, etc) consist of a small prefix of bpf_object's name and ELF section name as a suffix. This makes it hard for users to "guess" the name to use for looking up by name with bpf_object__find_map_by_name() API. One proposal was to drop object name prefix from the map name and just use ".rodata", ".data", etc, names. One downside called out was that when multiple BPF applications are active on the host, it will be hard to distinguish between multiple instances of .rodata and know which BPF object (app) they belong to. Having few first characters, while quite limiting, still can give a bit of a clue, in general. Note, though, that btf_value_type_id for such global data maps (ARRAY) points to DATASEC type, which encodes full ELF name, so tools like bpftool can take advantage of this fact to "recover" full original name of the map. This is also the reason why for custom .data.* and .rodata.* maps libbpf uses only their ELF names and doesn't prepend object name at all. Another downside of such approach is that it is not backwards compatible and, among direct use of bpf_object__find_map_by_name() API, will break any BPF skeleton generated using bpftool that was compiled with older libbpf version. Instead of causing all this pain, libbpf will still generate map name using a combination of object name and ELF section name, but it will allow looking such maps up by their natural names, which correspond to their respective ELF section names. This means non-truncated ELF section names longer than 15 characters are going to be expected and supported. With such set up, we get the best of both worlds: leave small bits of a clue about BPF application that instantiated such maps, as well as making it easy for user apps to lookup such maps at runtime. In this sense it closes corresponding libbpf 1.0 issue ([0]). BPF skeletons will continue using full names for lookups. [0] Closes: https://github.com/libbpf/libbpf/issues/275Signed-off-by:
-
Andrii Nakryiko authored
Enhance existing selftests to demonstrate the use of custom .data/.rodata sections. Signed-off-by:
-
Andrii Nakryiko authored
Add support for having multiple .rodata and .data data sections ([0]). .rodata/.data are supported like the usual, but now also .rodata.<whatever> and .data.<whatever> are also supported. Each such section will get its own backing BPF_MAP_TYPE_ARRAY, just like .rodata and .data. Multiple .bss maps are not supported, as the whole '.bss' name is confusing and might be deprecated soon, as well as user would need to specify custom ELF section with SEC() attribute anyway, so might as well stick to just .data.* and .rodata.* convention. User-visible map name for such new maps is going to be just their ELF section names. [0] https://github.com/libbpf/libbpf/issues/274Signed-off-by:
-
Andrii Nakryiko authored
It can happen that some data sections (e.g., .rodata.cst16, containing compiler populated string constants) won't have a corresponding BTF DATASEC type. Now that libbpf supports .rodata.* and .data.* sections, situation like that will cause invalid BPF skeleton to be generated that won't compile successfully, as some parts of skeleton would assume memory-mapped struct definitions for each special data section. Fix this by generating empty struct definitions for such data sections. Signed-off-by:
-
Andrii Nakryiko authored
Remove the assumption about only single instance of each of .rodata and .data internal maps. Nothing changes for '.rodata' and '.data' maps, but new '.rodata.something' map will get 'rodata_something' section in BPF skeleton for them (as well as having struct bpf_map * field in maps section with the same field name). Signed-off-by:
-
Andrii Nakryiko authored
Remove internal libbpf assumption that there can be only one .rodata, .data, and .bss map per BPF object. To achieve that, extend and generalize the scheme that was used for keeping track of relocation ELF sections. Now each ELF section has a temporary extra index that keeps track of logical type of ELF section (relocations, data, read-only data, BSS). Switch relocation to this scheme, as well as .rodata/.data/.bss handling. We don't yet allow multiple .rodata, .data, and .bss sections, but no libbpf internal code makes an assumption that there can be only one of each and thus they can be explicitly referenced by a single index. Next patches will actually allow multiple .rodata and .data sections. Signed-off-by:
-
Andrii Nakryiko authored
Minimize the usage of class-agnostic gelf_xxx() APIs from libelf. These APIs require copying ELF data structures into local GElf_xxx structs and have a more cumbersome API. BPF ELF file is defined to be always 64-bit ELF object, even when intended to be run on 32-bit host architectures, so there is no need to do class-agnostic conversions everywhere. BPF static linker implementation within libbpf has been using Elf64-specific types since initial implementation. Add two simple helpers, elf_sym_by_idx() and elf_rel_by_idx(), for more succinct direct access to ELF symbol and relocation records within ELF data itself and switch all the GElf_xxx usage into Elf64_xxx equivalents. The only remaining place within libbpf.c that's still using gelf API is gelf_getclass(), as there doesn't seem to be a direct way to get underlying ELF bitness. No functional changes intended. Signed-off-by:
-
Andrii Nakryiko authored
Name currently anonymous internal struct that keeps ELF-related state for bpf_object. Just a bit of clean up, no functional changes. Signed-off-by:
-
Andrii Nakryiko authored
There isn't a good use case where anyone but libbpf itself needs to call btf__finalize_data(). It was implemented for internal use and it's not clear why it was made into public API in the first place. To function, it requires active ELF data, which is stored inside bpf_object for the duration of opening phase only. But the only BTF that needs bpf_object's ELF is that bpf_object's BTF itself, which libbpf fixes up automatically during bpf_object__open() operation anyways. There is no need for any additional fix up and no reasonable scenario where it's useful and appropriate. Thus, btf__finalize_data() is just an API atavism and is better removed. So this patch marks it as deprecated immediately (v0.6+) and moves the code from btf.c into libbpf.c where it's used in the context of bpf_object opening phase. Such code co-location allows to make code structure more straightforward and remove bpf_object__section_size() and bpf_object__variable_offset() internal helpers from libbpf_internal.h, making them static. Their naming is also adjusted to not create a wrong illusion that they are some sort of method of bpf_object. They are internal helpers and are called appropriately. This is part of libbpf 1.0 effort ([0]). [0] Closes: https://github.com/libbpf/libbpf/issues/276Signed-off-by:
-
- 21 Oct, 2021 3 commits
-
-
Andrii Nakryiko authored
Jiri Olsa says: ==================== hi, sending fixes for perf_buffer test on systems with offline cpus. v2: - resend due to delivery issues, no changes thanks, jirka Acked-by:
John Fastabend <john.fastabend@gmail.com> ==================== Signed-off-by:
-
Jiri Olsa authored
The perf buffer tests triggers trace with nanosleep syscall, but monitors all syscalls, which results in lot of data in the buffer and makes it harder to debug. Let's lower the trace traffic and monitor just nanosleep syscall. Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Signed-off-by:
-
Jiri Olsa authored
The perf_buffer fails on system with offline cpus: # test_progs -t perf_buffer serial_test_perf_buffer:PASS:nr_cpus 0 nsec serial_test_perf_buffer:PASS:nr_on_cpus 0 nsec serial_test_perf_buffer:PASS:skel_load 0 nsec serial_test_perf_buffer:PASS:attach_kprobe 0 nsec serial_test_perf_buffer:PASS:perf_buf__new 0 nsec serial_test_perf_buffer:PASS:epoll_fd 0 nsec skipping offline CPU #4 serial_test_perf_buffer:PASS:perf_buffer__poll 0 nsec serial_test_perf_buffer:PASS:seen_cpu_cnt 0 nsec serial_test_perf_buffer:PASS:buf_cnt 0 nsec ... serial_test_perf_buffer:PASS:fd_check 0 nsec serial_test_perf_buffer:PASS:drain_buf 0 nsec serial_test_perf_buffer:PASS:consume_buf 0 nsec serial_test_perf_buffer:FAIL:cpu_seen cpu 5 not seen #88 perf_buffer:FAIL Summary: 0/0 PASSED, 0 SKIPPED, 1 FAILED If the offline cpu is from the middle of the possible set, we get mismatch with possible and online cpu buffers. The perf buffer test calls perf_buffer__consume_buffer for all 'possible' cpus, but the library holds only 'online' cpu buffers and perf_buffer__consume_buffer returns them based on index. Adding extra (online) index to keep track of online buffers, we need the original (possible) index to trigger trace on proper cpu. Signed-off-by:
-
