- 05 Mar, 2021 27 commits
-
-
Lorenz Bauer authored
Share the timing / signal interruption logic between different implementations of PROG_TEST_RUN. There is a change in behaviour as well. We check the loop exit condition before checking for pending signals. This resolves an edge case where a signal arrives during the last iteration. Instead of aborting with EINTR we return the successful result to user space. Signed-off-by:
Lorenz Bauer <lmb@cloudflare.com> Signed-off-by:
Alexei Starovoitov <ast@kernel.org> Acked-by:
Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20210303101816.36774-2-lmb@cloudflare.com
-
Alexei Starovoitov authored
Joe Stringer says: ==================== The state of bpf(2) manual pages today is not exactly ideal. For the most part, it was written several years ago and has not kept up with the pace of development in the kernel tree. For instance, out of a total of ~35 commands to the BPF syscall available today, when I pull the kernel-man-pages tree today I find just 6 documented commands: The very basics of map interaction and program load. In contrast, looking at bpf-helpers(7), I am able today to run one command[0] to fetch API documentation of the very latest eBPF helpers that have been added to the kernel. This documentation is up to date because kernel maintainers enforce documenting the APIs as part of the feature submission process. As far as I can tell, we rely on manual synchronization from the kernel tree to the kernel-man-pages tree to distribute these more widely, so all locations may not be completely up to date. That said, the documentation does in fact exist in the first place which is a major initial hurdle to overcome. Given the relative success of the process around bpf-helpers(7) to encourage developers to document their user-facing changes, in this patch series I explore applying this technique to bpf(2) as well. Unfortunately, even with bpf(2) being so out-of-date, there is still a lot of content to convert over. In particular, the following aspects of the bpf syscall could also be individually be generated from separate documentation in the header: * BPF syscall commands * BPF map types * BPF program types * BPF program subtypes (aka expected_attach_type) * BPF attachment points Rather than tackle everything at once, I have focused in this series on the syscall commands, "enum bpf_cmd". This series is structured to first import what useful descriptions there are from the kernel-man-pages tree, then piece-by-piece document a few of the syscalls in more detail in cases where I could find useful documentation from the git tree or from a casual read of the code. Not all documentation is comprehensive at this point, but a basis is provided with examples that can be further enhanced with subsequent follow-up patches. Note, the series in its current state only includes documentation around the syscall commands themselves, so in the short term it doesn't allow us to automate bpf(2) generation; Only one section of the man page could be replaced. Though if there is appetite for this approach, this should be trivial to improve on, even if just by importing the remaining static text from the kernel-man-pages tree. Following that, the series enhances the python scripting around parsing the descriptions from the header files and generating dedicated ReStructured Text and troff output. Finally, to expose the new text and reduce the likelihood of having it get out of date or break the docs parser, it is added to the selftests and exposed through the kernel documentation web pages. The eventual goal of this effort would be to extend the kernel UAPI headers such that each of the categories I had listed above (commands, maps, progs, hooks) have dedicated documentation in the kernel tree, and that developers must update the comments in the headers to document the APIs prior to patch acceptance, and that we could auto-generate the latest version of the bpf(2) manual pages based on a few static description sections combined with the dynamically-generated output from the header. This patch series can also be found at the following location on GitHub: https://github.com/joestringer/linux/tree/submit/bpf-command-docs_v2 Thanks also to Quentin Monnet for initial review. [0]: make -C tools/testing/selftests/bpf docs v2: * Remove build infrastructure in favor of kernel-doc directives * Shift userspace-api docs under Documentation/userspace-api/ebpf * Fix scripts/bpf_doc.py syscall --header (throw unsupported error) * Improve .gitignore handling of newly autogenerated files ==================== Acked-by:
Jonathan Corbet <corbet@lwn.net> Signed-off-by:
-
Joe Stringer authored
Synchronize the header after all of the recent changes. Signed-off-by:
Joe Stringer <joe@cilium.io> Signed-off-by:
Quentin Monnet <quentin@isovalent.com> Acked-by:
Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210302171947.2268128-16-joe@cilium.io
-
Joe Stringer authored
Generate the syscall command reference from the UAPI header file and include it in the main bpf docs page. Signed-off-by:
-
Joe Stringer authored
Add building of the bpf(2) syscall commands documentation as part of the docs building step in the build. This allows us to pick up on potential parse errors from the docs generator script as part of selftests. The generated manual pages here are not intended for distribution, they are just a fragment that can be integrated into the other static text of bpf(2) to form the full manual page. Signed-off-by:
Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20210302171947.2268128-14-joe@cilium.io
-
Joe Stringer authored
Previously, the Makefile here was only targeting a single manual page so it just hardcoded a bunch of individual rules to specifically handle build, clean, install, uninstall for that particular page. Upcoming commits will generate manual pages for an additional section, so this commit prepares the makefile first by converting the existing targets into an evaluated set of targets based on the manual page name and section. Signed-off-by:
-
Joe Stringer authored
This logic is used for validating the manual pages from selftests, so move the infra under tools/testing/selftests/bpf/ and rely on selftests for validation rather than tying it into the bpftool build. Signed-off-by:
-
Joe Stringer authored
Add a new target to bpf_doc.py to support generating the list of syscall commands directly from the UAPI headers. Assuming that developer submissions keep the main header up to date, this should allow the man pages to be automatically generated based on the latest API changes rather than requiring someone to separately go back through the API and describe each command. Signed-off-by:
-
Joe Stringer authored
Abstract out the target parameter so that upcoming commits, more than just the existing "helpers" target can be called to generate specific portions of docs from the eBPF UAPI headers. Signed-off-by:
-
Joe Stringer authored
Based roughly on the following commits: * Commit cb4d03ab ("bpf: Add generic support for lookup batch op") * Commit 05799638 ("bpf: Add batch ops to all htab bpf map") * Commit aa2e93b8 ("bpf: Add generic support for update and delete batch ops") Signed-off-by:
Brian Vazquez <brianvv@google.com> Acked-by:
-
Joe Stringer authored
Commit 468e2f64 ("bpf: introduce BPF_PROG_QUERY command") originally introduced this, but there have been several additions since then. Unlike BPF_PROG_ATTACH, it appears that the sockmap progs are not able to be queried so far. Signed-off-by:
-
Joe Stringer authored
Based on a brief read of the corresponding source code. Signed-off-by:
-
Joe Stringer authored
Document the prog attach command in more detail, based on git commits: * commit f4324551 ("bpf: add BPF_PROG_ATTACH and BPF_PROG_DETACH commands") * commit 4f738adb ("bpf: create tcp_bpf_ulp allowing BPF to monitor socket TX/RX data") * commit f4364dcf ("media: rc: introduce BPF_PROG_LIRC_MODE2") * commit d58e468b ("flow_dissector: implements flow dissector BPF hook") Signed-off-by:
-
Joe Stringer authored
Commit b2197755 ("bpf: add support for persistent maps/progs") contains the original implementation and git logs, used as reference for this documentation. Also pull in the filename restriction as documented in commit 6d8cb045 ("bpf: comment why dots in filenames under BPF virtual FS are not allowed") Signed-off-by:
-
Joe Stringer authored
Document the meaning of the BPF_F_LOCK flag for the map lookup/update descriptions. Based on commit 96049f3a ("bpf: introduce BPF_F_LOCK flag"). Signed-off-by:
-
Joe Stringer authored
Introduce high-level descriptions of the intent and return codes of the bpf() syscall commands. Subsequent patches may further flesh out the content to provide a more useful programming reference. Signed-off-by:
-
Joe Stringer authored
These descriptions are present in the man-pages project from the original submissions around 2015-2016. Import them so that they can be kept up to date as developers extend the bpf syscall commands. These descriptions follow the pattern used by scripts/bpf_helpers_doc.py so that we can take advantage of the parser to generate more up-to-date man page writing based upon these headers. Some minor wording adjustments were made to make the descriptions more consistent for the description / return format. Signed-off-by:
Michael Kerrisk <mtk.manpages@gmail.com>
-
Alexei Starovoitov authored
Ilya Leoshkevich says: ==================== Some BPF programs compiled on s390 fail to load, because s390 arch-specific linux headers contain float and double types. Introduce support for such types by representing them using the new BTF_KIND_FLOAT. This series deals with libbpf, bpftool, in-kernel BTF parser as well as selftests and documentation. There are also pahole and LLVM parts: * https://github.com/iii-i/dwarves/commit/btf-kind-float-v2 * https://reviews.llvm.org/D83289 but they should go in after the libbpf part is integrated. --- v0: https://lore.kernel.org/bpf/20210210030317.78820-1-iii@linux.ibm.com/ v0 -> v1: Per Andrii's suggestion, remove the unnecessary trailing u32. v1: https://lore.kernel.org/bpf/20210216011216.3168-1-iii@linux.ibm.com/ v1 -> v2: John noticed that sanitization corrupts BTF, because new and old sizes don't match. Per Yonghong's suggestion, use a modifier type (which has the same size as the float type) as a replacement. Per Yonghong's suggestion, add size and alignment checks to the kernel BTF parser. v2: https://lore.kernel.org/bpf/20210219022543.20893-1-iii@linux.ibm.com/ v2 -> v3: Based on Yonghong's suggestions: Use BTF_KIND_CONST instead of BTF_KIND_TYPEDEF and make sure that the C code generated from the sanitized BTF is well-formed; fix size calculation in tests and use NAME_TBD everywhere; limit allowed sizes to 2, 4, 8, 12 and 16 (this should also fix m68k and nds32le builds). v3: https://lore.kernel.org/bpf/20210220034959.27006-1-iii@linux.ibm.com/ v3 -> v4: More fixes for the Yonghong's findings: fix the outdated comment in bpf_object__sanitize_btf() and add the error handling there (I've decided to check uint_id and uchar_id too in order to simplify debugging); add bpftool output example; use div64_u64_rem() instead of % in order to fix the linker error. Also fix the "invalid BTF_INFO" test (new commit, #4). v4: https://lore.kernel.org/bpf/20210222214917.83629-1-iii@linux.ibm.com/ v4 -> v5: Fixes for the Andrii's findings: Use BTF_KIND_STRUCT instead of BTF_KIND_TYPEDEF for sanitization; check byte_sz in libbpf; move btf__add_float; remove relo support; add a dedup test (new commit, #7). v5: https://lore.kernel.org/bpf/20210223231459.99664-1-iii@linux.ibm.com/ v5 -> v6: Fixes for further findings by Andrii: split whitespace issue fix into a separate patch; add 12-byte float to "float test #1, well-formed". v6: https://lore.kernel.org/bpf/20210224234535.106970-1-iii@linux.ibm.com/ v6 -> v7: John suggested to add a comment explaining why sanitization does not preserve the type name, as well as what effect it has on running the code on the older kernels. Yonghong has asked to add a comment explaining why we are not checking the alignment very precisely in the kernel. John suggested to add a bpf_core_field_size test (commit #9). Based on Alexei's feedback [1] I'm proceeding with the BTF_KIND_FLOAT approach. [1] https://lore.kernel.org/bpf/CAADnVQKWPODWZ2RSJ5FJhfYpxkuV0cvSAL1O+FSr9oP1ercoBg@mail.gmail.com/ ==================== Signed-off-by:
-
Ilya Leoshkevich authored
Also document the expansion of the kind bitfield. Signed-off-by:
Ilya Leoshkevich <iii@linux.ibm.com> Signed-off-by:
-
Ilya Leoshkevich authored
Check that floats don't interfere with struct deduplication, that they are not merged with another kinds and that floats of different sizes are not merged with each other. Suggested-by:
-
Ilya Leoshkevich authored
Test the good variants as well as the potential malformed ones. Signed-off-by:
-
Ilya Leoshkevich authored
On the kernel side, introduce a new btf_kind_operations. It is similar to that of BTF_KIND_INT, however, it does not need to handle encodings and bit offsets. Do not implement printing, since the kernel does not know how to format floating-point values. Signed-off-by:
-
Ilya Leoshkevich authored
The bit being checked by this test is no longer reserved after introducing BTF_KIND_FLOAT, so use the next one instead. Signed-off-by:
-
Ilya Leoshkevich authored
Only dumping support needs to be adjusted, the code structure follows that of BTF_KIND_INT. Example plain and JSON outputs: [4] FLOAT 'float' size=4 {"id":4,"kind":"FLOAT","name":"float","size":4} Signed-off-by: -
Ilya Leoshkevich authored
The logic follows that of BTF_KIND_INT most of the time. Sanitization replaces BTF_KIND_FLOATs with equally-sized empty BTF_KIND_STRUCTs on older kernels, for example, the following: [4] FLOAT 'float' size=4 becomes the following: [4] STRUCT '(anon)' size=4 vlen=0 With dwarves patch [1] and this patch, the older kernels, which were failing with the floating-point-related errors, will now start working correctly. [1] https://github.com/iii-i/dwarves/commit/btf-kind-float-v2Signed-off-by: -
Ilya Leoshkevich authored
Remove trailing space. Signed-off-by:
-
Ilya Leoshkevich authored
Add a new kind value and expand the kind bitfield. Signed-off-by:
-
- 04 Mar, 2021 2 commits
-
-
Yonghong Song authored
The original bcc pull request https://github.com/iovisor/bcc/pull/3270 exposed a verifier failure with Clang 12/13 while Clang 4 works fine. Further investigation exposed two issues: Issue 1: LLVM may generate code which uses less refined value. The issue is fixed in LLVM patch: https://reviews.llvm.org/D97479 Issue 2: Spills with initial value 0 are marked as precise which makes later state pruning less effective. This is my rough initial analysis and further investigation is needed to find how to improve verifier pruning in such cases. With the above LLVM patch, for the new loop6.c test, which has smaller loop bound compared to original test, I got: $ test_progs -s -n 10/16 ... stack depth 64 processed 390735 insns (limit 1000000) max_states_per_insn 87 total_states 8658 peak_states 964 mark_read 6 #10/16 loop6.o:OK Use the original loop bound, i.e., commenting out "#define WORKAROUND", I got: $ test_progs -s -n 10/16 ... BPF program is too large. Processed 1000001 insn stack depth 64 processed 1000001 insns (limit 1000000) max_states_per_insn 91 total_states 23176 peak_states 5069 mark_read 6 ... #10/16 loop6.o:FAIL The purpose of this patch is to provide a regression test for the above LLVM fix and also provide a test case for further analyzing the verifier pruning issue. Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Cc: Zhenwei Pi <pizhenwei@bytedance.com> Link: https://lore.kernel.org/bpf/20210226223810.236472-1-yhs@fb.com
-
Cong Wang authored
This should fix the following warning: include/linux/skbuff.h:932: warning: Function parameter or member '_sk_redir' not described in 'sk_buff' Reported-by:
Cong Wang <cong.wang@bytedance.com> Signed-off-by:
John Fastabend <john.fastabend@gmail.com> Acked-by:
-
- 03 Mar, 2021 1 commit
-
-
Andrii Nakryiko authored
Just like was done for bpftool and selftests in ec23eb70 ("tools/bpftool: Allow substituting custom vmlinux.h for the build") and ca4db638 ("selftests/bpf: Allow substituting custom vmlinux.h for selftests build"), allow to provide pre-generated vmlinux.h for runqslower build. Signed-off-by:
Martin KaFai Lau <kafai@fb.com> Acked-by:
-
- 26 Feb, 2021 10 commits
-
-
Ian Denhardt authored
... so callers can correctly detect failure. Signed-off-by:
Ian Denhardt <ian@zenhack.net> Signed-off-by:
-
Ian Denhardt authored
Per discussion at [0] this was originally introduced as a warning due to concerns about breaking existing code, but a hard error probably makes more sense, especially given that concerns about breakage were only speculation. [0] https://lore.kernel.org/bpf/c964892195a6b91d20a67691448567ef528ffa6d.camel@linux.ibm.com/T/#tSigned-off-by:
-
Alexei Starovoitov authored
Yonghong Song says: ==================== This patch set introduced bpf_for_each_map_elem() helper. The helper permits bpf program iterates through all elements for a particular map. The work originally inspired by an internal discussion where firewall rules are kept in a map and bpf prog wants to check packet 5 tuples against all rules in the map. A bounded loop can be used but it has a few drawbacks. As the loop iteration goes up, verification time goes up too. For really large maps, verification may fail. A helper which abstracts out the loop itself will not have verification time issue. A recent discussion in [1] involves to iterate all hash map elements in bpf program. Currently iterating all hashmap elements in bpf program is not easy if key space is really big. Having a helper to abstract out the loop itself is even more meaningful. The proposed helper signature looks like: long bpf_for_each_map_elem(map, callback_fn, callback_ctx, flags) where callback_fn is a static function and callback_ctx is a piece of data allocated on the caller stack which can be accessed by the callback_fn. The callback_fn signature might be different for different maps. For example, for hash/array maps, the signature is long callback_fn(map, key, val, callback_ctx) In the rest of series, Patches 1/2/3/4 did some refactoring. Patch 5 implemented core kernel support for the helper. Patches 6 and 7 added hashmap and arraymap support. Patches 8/9 added libbpf support. Patch 10 added bpftool support. Patches 11 and 12 added selftests for hashmap and arraymap. [1]: https://lore.kernel.org/bpf/20210122205415.113822-1-xiyou.wangcong@gmail.com/ Changelogs: v4 -> v5: - rebase on top of bpf-next. v3 -> v4: - better refactoring of check_func_call(), calculate subprogno outside of __check_func_call() helper. (Andrii) - better documentation (like the list of supported maps and their callback signatures) in uapi header. (Andrii) - implement and use ASSERT_LT in selftests. (Andrii) - a few other minor changes. v2 -> v3: - add comments in retrieve_ptr_limit(), which is in sanitize_ptr_alu(), to clarify the code is not executed for PTR_TO_MAP_KEY handling, but code is manually tested. (Alexei) - require BTF for callback function. (Alexei) - simplify hashmap/arraymap callback return handling as return value [0, 1] has been enforced by the verifier. (Alexei) - also mark global subprog (if used in ld_imm64) as RELO_SUBPROG_ADDR. (Andrii) - handle the condition to mark RELO_SUBPROG_ADDR properly. (Andrii) - make bpftool subprog insn offset dumping consist with pcrel calls. (Andrii) v1 -> v2: - setup callee frame in check_helper_call() and then proceed to verify helper return value as normal (Alexei) - use meta data to keep track of map/func pointer to avoid hard coding the register number (Alexei) - verify callback_fn return value range [0, 1]. (Alexei) - add migrate_{disable, enable} to ensure percpu value is the one bpf program expects to see. (Alexei) - change bpf_for_each_map_elem() return value to the number of iterated elements. (Andrii) - Change libbpf pseudo_func relo name to RELO_SUBPROG_ADDR and use more rigid checking for the relocation. (Andrii) - Better format to print out subprog address with bpftool. (Andrii) - Use bpf_prog_test_run to trigger bpf run, instead of bpf_iter. (Andrii) - Other misc changes. ==================== Signed-off-by:
-
Yonghong Song authored
A test is added for arraymap and percpu arraymap. The test also exercises the early return for the helper which does not traverse all elements. $ ./test_progs -n 45 #45/1 hash_map:OK #45/2 array_map:OK #45 for_each:OK Summary: 1/2 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: -
Yonghong Song authored
A test case is added for hashmap and percpu hashmap. The test also exercises nested bpf_for_each_map_elem() calls like bpf_prog: bpf_for_each_map_elem(func1) func1: bpf_for_each_map_elem(func2) func2: $ ./test_progs -n 45 #45/1 hash_map:OK #45 for_each:OK Summary: 1/1 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: -
Yonghong Song authored
With later hashmap example, using bpftool xlated output may look like: int dump_task(struct bpf_iter__task * ctx): ; struct task_struct *task = ctx->task; 0: (79) r2 = *(u64 *)(r1 +8) ; if (task == (void *)0 || called > 0) ... 19: (18) r2 = subprog[+17] 30: (18) r2 = subprog[+25] ... 36: (95) exit __u64 check_hash_elem(struct bpf_map * map, __u32 * key, __u64 * val, struct callback_ctx * data): ; struct bpf_iter__task *ctx = data->ctx; 37: (79) r5 = *(u64 *)(r4 +0) ... 55: (95) exit __u64 check_percpu_elem(struct bpf_map * map, __u32 * key, __u64 * val, void * unused): ; check_percpu_elem(struct bpf_map *map, __u32 *key, __u64 *val, void *unused) 56: (bf) r6 = r3 ... 83: (18) r2 = subprog[-47] Signed-off-by: -
Yonghong Song authored
A new relocation RELO_SUBPROG_ADDR is added to capture subprog addresses loaded with ld_imm64 insns. Such ld_imm64 insns are marked with BPF_PSEUDO_FUNC and will be passed to kernel. For bpf_for_each_map_elem() case, kernel will check that the to-be-used subprog address must be a static function and replace it with proper actual jited func address. Signed-off-by:
-
Yonghong Song authored
Move function is_ldimm64() close to the beginning of libbpf.c, so it can be reused by later code and the next patch as well. There is no functionality change. Signed-off-by:
-
Yonghong Song authored
This patch added support for arraymap and percpu arraymap. Signed-off-by:
-
Yonghong Song authored
This patch added support for hashmap, percpu hashmap, lru hashmap and percpu lru hashmap. Signed-off-by:
-
