- 11 Jun, 2022 1 commit
-
-
Feng Zhou authored
This patch use head->first in pcpu_freelist_head to check freelist having free or not. If having, grab spin_lock, or check next cpu's freelist. Before patch: hash_map performance ./map_perf_test 1 0:hash_map_perf pre-alloc 1043397 events per sec ... The average of the test results is around 1050000 events per sec. hash_map the worst: no free ./run_bench_bpf_hashmap_full_update.sh Setting up benchmark 'bpf-hashmap-ful-update'... Benchmark 'bpf-hashmap-ful-update' started. 1:hash_map_full_perf 15687 events per sec ... The average of the test results is around 16000 events per sec. ftrace trace: 0) | htab_map_update_elem() { 0) | __pcpu_freelist_pop() { 0) | _raw_spin_lock() 0) | _raw_spin_unlock() 0) | ... 0) + 25.188 us | } 0) + 28.439 us | } The test machine is 16C, trying to get spin_lock 17 times, in addition to 16c, there is an extralist. after patch: hash_map performance ./map_perf_test 1 0:hash_map_perf pre-alloc 1053298 events per sec ... The average of the test results is around 1050000 events per sec. hash_map worst: no free ./run_bench_bpf_hashmap_full_update.sh Setting up benchmark 'bpf-hashmap-ful-update'... Benchmark 'bpf-hashmap-ful-update' started. 1:hash_map_full_perf 555830 events per sec ... The average of the test results is around 550000 events per sec. ftrace trace: 0) | htab_map_update_elem() { 0) | alloc_htab_elem() { 0) 0.586 us | __pcpu_freelist_pop(); 0) 0.945 us | } 0) 8.669 us | } It can be seen that after adding this patch, the map performance is almost not degraded, and when free=0, first check head->first instead of directly acquiring spin_lock. Co-developed-by:Chengming Zhou <zhouchengming@bytedance.com> Signed-off-by:
Feng Zhou <zhoufeng.zf@bytedance.com> Link: https://lore.kernel.org/r/20220610023308.93798-2-zhoufeng.zf@bytedance.comSigned-off-by:
Alexei Starovoitov <ast@kernel.org>
-
- 09 Jun, 2022 3 commits
-
-
Andrii Nakryiko authored
Fix libbpf's bpf_program__attach_uprobe() logic of determining function's *file offset* (which is what kernel is actually expecting) when attaching uprobe/uretprobe by function name. Previously calculation was determining virtual address offset relative to base load address, which (offset) is not always the same as file offset (though very frequently it is which is why this went unnoticed for a while). Fixes: 433966e3 ("libbpf: Support function name-based attach uprobes") Signed-off-by:
Andrii Nakryiko <andrii@kernel.org> Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Cc: Riham Selim <rihams@fb.com> Cc: Alan Maguire <alan.maguire@oracle.com> Link: https://lore.kernel.org/bpf/20220606220143.3796908-1-andrii@kernel.org
-
Kosuke Fujimoto authored
"BFP" should be "BPF" Signed-off-by:
Kosuke Fujimoto <fujimotokosuke0@gmail.com> Signed-off-by:
-
Shahab Vahedi authored
This change adjusts the Makefile to use "HOSTAR" as the archive tool to keep the sanity of the build process for the bootstrap part in check. For the rationale, please continue reading. When cross compiling bpftool with buildroot, it leads to an invocation like: $ AR="/path/to/buildroot/host/bin/arc-linux-gcc-ar" \ CC="/path/to/buildroot/host/bin/arc-linux-gcc" \ ... make Which in return fails while building the bootstrap section: ----------------------------------8<---------------------------------- make: Entering directory '/src/bpftool-v6.7.0/src' ... libbfd: [ on ] ... disassembler-four-args: [ on ] ... zlib: [ on ] ... libcap: [ OFF ] ... clang-bpf-co-re: [ on ] <-- triggers bootstrap . . . LINK /src/bpftool-v6.7.0/src/bootstrap/bpftool /usr/bin/ld: /src/bpftool-v6.7.0/src/bootstrap/libbpf/libbpf.a: error adding symbols: archive has no index; run ranlib to add one collect2: error: ld returned 1 exit status make: *** [Makefile:211: /src/bpftool-v6.7.0/src/bootstrap/bpftool] Error 1 make: *** Waiting for unfinished jobs.... AR /src/bpftool-v6.7.0/src/libbpf/libbpf.a make[1]: Leaving directory '/src/bpftool-v6.7.0/libbpf/src' make: Leaving directory '/src/bpftool-v6.7.0/src' ---------------------------------->8---------------------------------- This occurs because setting "AR" confuses the build process for the bootstrap section and it calls "arc-linux-gcc-ar" to create and index "libbpf.a" instead of the host "ar". Signed-off-by:Shahab Vahedi <shahab@synopsys.com> Signed-off-by:
Quentin Monnet <quentin@isovalent.com> Cc: Jean-Philippe Brucker <jean-philippe@linaro.org> Link: https://lore.kernel.org/bpf/8d297f0c-cfd0-ef6f-3970-6dddb3d9a87a@synopsys.com
-
- 07 Jun, 2022 18 commits
-
-
Alexei Starovoitov authored
Yonghong Song says: ==================== Currently, btf only supports upto 32bit enum value with BTF_KIND_ENUM. But in kernel, some enum has 64bit values, e.g., in uapi bpf.h, we have enum { BPF_F_INDEX_MASK = 0xffffffffULL, BPF_F_CURRENT_CPU = BPF_F_INDEX_MASK, BPF_F_CTXLEN_MASK = (0xfffffULL << 32), }; With BTF_KIND_ENUM, the value for BPF_F_CTXLEN_MASK will be encoded as 0 which is incorrect. To solve this problem, BTF_KIND_ENUM64 is proposed in this patch set to support enum 64bit values. Also, since sometimes there is a need to generate C code from btf, e.g., vmlinux.h, btf kflag support is also added for BTF_KIND_ENUM and BTF_KIND_ENUM64 to indicate signedness, helping proper value printout. Changelog: v4 -> v5: - skip newly-added enum64 C test if clang version <= 14. v3 -> v4: - rename btf_type_is_any_enum() to btf_is_any_enum() to favor consistency in libbpf. - fix sign extension issue in btf_dump_get_enum_value(). - fix BPF_CORE_FIELD_SIGNED signedness issue in bpf_core_calc_field_relo(). v2 -> v3: - Implement separate btf_equal_enum()/btf_equal_enum64() and btf_compat_enum()/btf_compat_enum64(). - Add a new enum64 placeholder type dynamicly for enum64 sanitization. - For bpftool output and unit selftest, printed out signed/unsigned encoding as well. - fix some issues with BTF_KIND_ENUM is doc and clarified sign extension rules for enum values. v1 -> v2: - Changed kflag default from signed to unsigned - Fixed sanitization issue - Broke down libbpf related patches for easier review - Added more tests - More code refactorization - Corresponding llvm patch (to support enum64) is also updated ==================== Signed-off-by: -
Yonghong Song authored
Add BTF_KIND_ENUM64 documentation in btf.rst. Also fixed a typo for section number for BTF_KIND_TYPE_TAG from 2.2.17 to 2.2.18, and fixed a type size issue for BTF_KIND_ENUM. Acked-by:
Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/r/20220607062724.3728215-1-yhs@fb.comSigned-off-by:
-
Yonghong Song authored
Add a test for enum64 value relocations. The test will be skipped if clang version is 14 or lower since enum64 is only supported from version 15. Acked-by:
-
Yonghong Song authored
Add a few unit tests for BTF_KIND_ENUM64 deduplication. Acked-by:
-
Yonghong Song authored
Add unit tests for basic BTF_KIND_ENUM64 encoding. Acked-by:
-
Yonghong Song authored
Add tests to use the new enum kflag and enum64 API functions in selftest btf_write. Acked-by:
-
Yonghong Song authored
The kflag is supported now for BTF_KIND_ENUM. So remove the test which tests verifier failure due to existence of kflag. Acked-by:
-
Yonghong Song authored
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: -
Yonghong Song authored
The enum64 relocation support is added. The bpf local type could be either enum or enum64 and the remote type could be either enum or enum64 too. The all combinations of local enum/enum64 and remote enum/enum64 are supported. Acked-by:
-
Yonghong Song authored
Add BTF_KIND_ENUM64 support for bpf linking, which is very similar to BTF_KIND_ENUM. Acked-by:
-
Yonghong Song authored
When old kernel does not support enum64 but user space btf contains non-zero enum kflag or enum64, libbpf needs to do proper sanitization so modified btf can be accepted by the kernel. Sanitization for enum kflag can be achieved by clearing the kflag bit. For enum64, the type is replaced with an union of integer member types and the integer member size must be smaller than enum64 size. If such an integer type cannot be found, a new type is created and used for union members. Acked-by:
-
Yonghong Song authored
Add enum64 btf dumping support. For long long and unsigned long long dump, suffixes 'LL' and 'ULL' are added to avoid compilation errors in some cases. Acked-by:
-
Yonghong Song authored
Add enum64 deduplication support. BTF_KIND_ENUM64 handling is very similar to BTF_KIND_ENUM. Acked-by:
-
Yonghong Song authored
Add enum64 parsing support and two new enum64 public APIs: btf__add_enum64 btf__add_enum64_value Also add support of signedness for BTF_KIND_ENUM. The BTF_KIND_ENUM API signatures are not changed. The signedness will be changed from unsigned to signed if btf__add_enum_value() finds any negative values. Acked-by:
-
Yonghong Song authored
Refactor btf__add_enum() function to create a separate function btf_add_enum_common() so later the common function can be used to add enum64 btf type. There is no functionality change for this patch. Acked-by:
-
Yonghong Song authored
Currently, the 64bit relocation value in the instruction is computed as follows: __u64 imm = insn[0].imm + ((__u64)insn[1].imm << 32) Suppose insn[0].imm = -1 (0xffffffff) and insn[1].imm = 1. With the above computation, insn[0].imm will first sign-extend to 64bit -1 (0xffffffffFFFFFFFF) and then add 0x1FFFFFFFF, producing incorrect value 0xFFFFFFFF. The correct value should be 0x1FFFFFFFF. Changing insn[0].imm to __u32 first will prevent 64bit sign extension and fix the issue. Merging high and low 32bit values also changed from '+' to '|' to be consistent with other similar occurences in kernel and libbpf. Acked-by:
Dave Marchevsky <davemarchevsky@fb.com> Signed-off-by:
-
Yonghong Song authored
Currently, the libbpf limits the relocation value to be 32bit since all current relocations have such a limit. But with BTF_KIND_ENUM64 support, the enum value could be 64bit. So let us permit 64bit relocation value in libbpf. Acked-by:
-
Yonghong Song authored
Currently, BTF only supports upto 32bit enum value with BTF_KIND_ENUM. But in kernel, some enum indeed has 64bit values, e.g., in uapi bpf.h, we have enum { BPF_F_INDEX_MASK = 0xffffffffULL, BPF_F_CURRENT_CPU = BPF_F_INDEX_MASK, BPF_F_CTXLEN_MASK = (0xfffffULL << 32), }; In this case, BTF_KIND_ENUM will encode the value of BPF_F_CTXLEN_MASK as 0, which certainly is incorrect. This patch added a new btf kind, BTF_KIND_ENUM64, which permits 64bit value to cover the above use case. The BTF_KIND_ENUM64 has the following three fields followed by the common type: struct bpf_enum64 { __u32 nume_off; __u32 val_lo32; __u32 val_hi32; }; Currently, btf type section has an alignment of 4 as all element types are u32. Representing the value with __u64 will introduce a pad for bpf_enum64 and may also introduce misalignment for the 64bit value. Hence, two members of val_hi32 and val_lo32 are chosen to avoid these issues. The kflag is also introduced for BTF_KIND_ENUM and BTF_KIND_ENUM64 to indicate whether the value is signed or unsigned. The kflag intends to provide consistent output of BTF C fortmat with the original source code. For example, the original BTF_KIND_ENUM bit value is 0xffffffff. The format C has two choices, printing out 0xffffffff or -1 and current libbpf prints out as unsigned value. But if the signedness is preserved in btf, the value can be printed the same as the original source code. The kflag value 0 means unsigned values, which is consistent to the default by libbpf and should also cover most cases as well. The new BTF_KIND_ENUM64 is intended to support the enum value represented as 64bit value. But it can represent all BTF_KIND_ENUM values as well. The compiler ([1]) and pahole will generate BTF_KIND_ENUM64 only if the value has to be represented with 64 bits. In addition, a static inline function btf_kind_core_compat() is introduced which will be used later when libbpf relo_core.c changed. Here the kernel shares the same relo_core.c with libbpf. [1] https://reviews.llvm.org/D124641Acked-by:
-
- 03 Jun, 2022 5 commits
-
-
Hangbin Liu authored
As subject, we only test SKB mode for xdping at present. Now add DRV mode for xdping. Signed-off-by:
Hangbin Liu <liuhangbin@gmail.com> Signed-off-by:
Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20220602032507.464453-1-liuhangbin@gmail.com
-
Yuze Chi authored
Move the correct definition from linker.c into libbpf_internal.h. Fixes: 0087a681 ("libbpf: Automatically fix up BPF_MAP_TYPE_RINGBUF size, if necessary") Reported-by:
Yuze Chi <chiyuze@google.com> Signed-off-by:
Ian Rogers <irogers@google.com> Signed-off-by:
-
Martin KaFai Lau authored
tc_redirect_dtime was reported flaky from time to time. It always fails at the udp test and complains about the bpf@tc-ingress got a skb->tstamp when handling udp packet. It is unexpected because the skb->tstamp should have been cleared when crossing different netns. The most likely cause is that the skb is actually a tcp packet from the earlier tcp test. It could be the final TCP_FIN handling. This patch tightens the skb->tstamp check in the bpf prog. It ensures the skb is the current testing traffic. First, it checks that skb matches the IPPROTO of the running test (i.e. tcp vs udp). Second, it checks the server port (dst_ns_port). The server port is unique for each test (50000 + test_enum). Also fixed a typo in test_udp_dtime(): s/P100/P101/ Fixes: c803475f ("bpf: selftests: test skb->tstamp in redirect_neigh") Reported-by:
Martin KaFai Lau <kafai@fb.com> Signed-off-by:
-
Daniel Xu authored
These checks were effectively noops b/c there's only one way these functions get called: through prog_ops dispatching. And since there's no other callers, we can be sure that `prog` is always the correct type. Signed-off-by:
Daniel Xu <dxu@dxuuu.xyz> Signed-off-by:
-
Daniel Müller authored
This change fixes a couple of typos that were encountered while studying the source code. Signed-off-by:
Daniel Müller <deso@posteo.net> Signed-off-by:
-
- 02 Jun, 2022 13 commits
-
-
Wang Yufen authored
During TCP sockmap redirect pressure test, the following warning is triggered: WARNING: CPU: 3 PID: 2145 at net/core/stream.c:205 sk_stream_kill_queues+0xbc/0xd0 CPU: 3 PID: 2145 Comm: iperf Kdump: loaded Tainted: G W 5.10.0+ #9 Call Trace: inet_csk_destroy_sock+0x55/0x110 inet_csk_listen_stop+0xbb/0x380 tcp_close+0x41b/0x480 inet_release+0x42/0x80 __sock_release+0x3d/0xa0 sock_close+0x11/0x20 __fput+0x9d/0x240 task_work_run+0x62/0x90 exit_to_user_mode_prepare+0x110/0x120 syscall_exit_to_user_mode+0x27/0x190 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The reason we observed is that: When the listener is closing, a connection may have completed the three-way handshake but not accepted, and the client has sent some packets. The child sks in accept queue release by inet_child_forget()->inet_csk_destroy_sock(), but psocks of child sks have not released. To fix, add sock_map_destroy to release psocks. Signed-off-by:
Wang Yufen <wangyufen@huawei.com> Signed-off-by:
Jakub Sitnicki <jakub@cloudflare.com> Acked-by:
John Fastabend <john.fastabend@gmail.com> Link: https://lore.kernel.org/bpf/20220524075311.649153-1-wangyufen@huawei.com
-
Lorenzo Bianconi authored
Forward the packet to the kernel if the gw router mac address is missing in to trigger ARP discovery. Fixes: 85bf1f51 ("samples: bpf: Convert xdp_router_ipv4 to XDP samples helper") Signed-off-by:
Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by:
-
Douglas Raillard authored
One strategy employed by libbpf to guess the pointer size is by finding the size of "unsigned long" type. This is achieved by looking for a type of with the expected name and checking its size. Unfortunately, the C syntax is friendlier to humans than to computers as there is some variety in how such a type can be named. Specifically, gcc and clang do not use the same names for integer types in debug info: - clang uses "unsigned long" - gcc uses "long unsigned int" Lookup all the names for such a type so that libbpf can hope to find the information it wants. Signed-off-by:Douglas Raillard <douglas.raillard@arm.com> Signed-off-by:
-
Tadeusz Struk authored
Syzbot found a Use After Free bug in compute_effective_progs(). The reproducer creates a number of BPF links, and causes a fault injected alloc to fail, while calling bpf_link_detach on them. Link detach triggers the link to be freed by bpf_link_free(), which calls __cgroup_bpf_detach() and update_effective_progs(). If the memory allocation in this function fails, the function restores the pointer to the bpf_cgroup_link on the cgroup list, but the memory gets freed just after it returns. After this, every subsequent call to update_effective_progs() causes this already deallocated pointer to be dereferenced in prog_list_length(), and triggers KASAN UAF error. To fix this issue don't preserve the pointer to the prog or link in the list, but remove it and replace it with a dummy prog without shrinking the table. The subsequent call to __cgroup_bpf_detach() or __cgroup_bpf_detach() will correct it. Fixes: af6eea57 ("bpf: Implement bpf_link-based cgroup BPF program attachment") Reported-by: <syzbot+f264bffdfbd5614f3bb2@syzkaller.appspotmail.com> Signed-off-by:
Tadeusz Struk <tadeusz.struk@linaro.org> Signed-off-by:
-
Michael Mullin authored
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 <masmullin@gmail.com> Signed-off-by:
-
Yuntao Wang authored
In the commit da00d2f1 ("bpf: Add test ops for BPF_PROG_TYPE_TRACING"), the bpf_fentry_test1 function was moved into bpf_prog_test_run_tracing(), which is the test_run function of the tracing BPF programs. Thus calling 'bpf_prog_test_run_opts(filter_fd, &topts)' will not trigger bpf_fentry_test1 function as filter_fd is a sk_filter BPF program. Fix it by replacing filter_fd with fexit_fd in the bpf_prog_test_run_opts() function. Fixes: da00d2f1 ("bpf: Add test ops for BPF_PROG_TYPE_TRACING") Signed-off-by:
Yuntao Wang <ytcoode@gmail.com> Signed-off-by:
-
Andrii Nakryiko authored
Daniel Müller says: ==================== This patch set introduces the means for querying a textual representation of the following BPF related enum types: - enum bpf_map_type - enum bpf_prog_type - enum bpf_attach_type - enum bpf_link_type To make that possible, we introduce a new public function for each of the types: libbpf_bpf_<type>_type_str. Having a way to query a textual representation has been asked for in the past (by systemd, among others). Such representations can generally be useful in tracing and logging contexts, among others. At this point, at least one client, bpftool, maintains such a mapping manually, which is prone to get out of date as new enum variants are introduced. libbpf is arguably best situated to keep this list complete and up-to-date. This patch series adds BTF based tests to ensure that exhaustiveness is upheld moving forward. The libbpf provided textual representation can be inferred from the corresponding enum variant name by removing the prefix and lowercasing the remainder. E.g., BPF_PROG_TYPE_SOCKET_FILTER -> socket_filter. Unfortunately, bpftool does not use such a programmatic approach for some of the bpf_attach_type variants. We decided in favor of changing its behavior to work with libbpf representations. However, for user inputs, specifically, we do maintain support for the traditionally used names around (please see patch "bpftool: Use libbpf_bpf_attach_type_str"). The patch series is structured as follows: - for each enumeration type in {bpf_prog_type, bpf_map_type, bpf_attach_type, bpf_link_type}: - we first introduce the corresponding public libbpf API function - we then add BTF based self-tests - we lastly adjust bpftool to use the libbpf provided functionality Signed-off-by: -
Daniel Müller authored
This change switches bpftool over to using the recently introduced libbpf_bpf_link_type_str function instead of maintaining its own string representation for the bpf_link_type enum. Signed-off-by:
-
Daniel Müller authored
This change adds a test for libbpf_bpf_link_type_str. The test retrieves all variants of the bpf_link_type enumeration using BTF and makes sure that the function under test works as expected for them. Signed-off-by:
-
Daniel Müller authored
This change introduces a new function, libbpf_bpf_link_type_str, to the public libbpf API. The function allows users to get a string representation for a bpf_link_type enum variant. Signed-off-by:
-
Daniel Müller authored
This change switches bpftool over to using the recently introduced libbpf_bpf_attach_type_str function instead of maintaining its own string representation for the bpf_attach_type enum. Note that contrary to other enum types, the variant names that bpftool maps bpf_attach_type to do not adhere a simple to follow rule. With bpf_prog_type, for example, the textual representation can easily be inferred by stripping the BPF_PROG_TYPE_ prefix and lowercasing the remaining string. bpf_attach_type violates this rule for various variants. We decided to fix up this deficiency with this change, meaning that bpftool uses the same textual representations as libbpf. Supporting tests, completion scripts, and man pages have been adjusted accordingly. However, we did add support for accepting (the now undocumented) original attach type names when they are provided by users. For the test (test_bpftool_synctypes.py), I have removed the enum representation checks, because we no longer mirror the various enum variant names in bpftool source code. For the man page, help text, and completion script checks we are now using enum definitions from uapi/linux/bpf.h as the source of truth directly. Signed-off-by:
-
Daniel Müller authored
This change adds a test for libbpf_bpf_attach_type_str. The test retrieves all variants of the bpf_attach_type enumeration using BTF and makes sure that the function under test works as expected for them. Signed-off-by:
-
Daniel Müller authored
This change introduces a new function, libbpf_bpf_attach_type_str, to the public libbpf API. The function allows users to get a string representation for a bpf_attach_type variant. Signed-off-by:
-
