Move bpf_helpers and add BPF_CORE_READ macros
Subsystems (6)
v3 cover Move bpf_helpers and add BPF_CORE_READ macros
This patch set makes bpf_helpers.h and bpf_endian.h a part of libbpf itself for consumption by user BPF programs, not just selftests. It also splits off tracing helpers into bpf_tracing.h, which also becomes part of libbpf. Some of the legacy stuff (BPF_ANNOTATE_KV_PAIR, load_{byte,half,word}, bpf_map_def with unsupported fields, etc, is extracted into selftests-only bpf_legacy.h. All the selftests and samples are switched to use libbpf's headers and selftests' ones are removed.
As part of this patch set we also add BPF_CORE_READ variadic macros, that are simplifying BPF CO-RE reads, especially the ones that have to follow few pointers. E.g., what in non-BPF world (and when using BCC) would be:
int x = s->a->b.c->d; /* s, a, and b.c are pointers */
today would have to be written using explicit bpf_probe_read() calls as:
void *t; int x; bpf_probe_read(&t, sizeof(t), s->a); bpf_probe_read(&t, sizeof(t), ((struct b *)t)->b.c); bpf_probe_read(&x, sizeof(x), ((struct c *)t)->d);
This is super inconvenient and distracts from program logic a lot. Now, with added BPF_CORE_READ() macros, you can write the above as:
int x = BPF_CORE_READ(s, a, b.c, d);
Up to 9 levels of pointer chasing are supported, which should be enough for any practical purpose, hopefully, without adding too much boilerplate macro definitions (though there is admittedly some, given how variadic and recursive C macro have to be implemented).
There is also BPF_CORE_READ_INTO() variant, which relies on caller to allocate space for result:
int x; BPF_CORE_READ_INTO(&x, s, a, b.c, d);
Result of last bpf_probe_read() call in the chain of calls is the result of BPF_CORE_READ_INTO(). If any intermediate bpf_probe_read() aall fails, then all the subsequent ones will fail too, so this is sufficient to know whether overall "operation" succeeded or not. No short-circuiting of bpf_probe_read()s is done, though.
BPF_CORE_READ_STR_INTO() is added as well, which differs from BPF_CORE_READ_INTO() only in that last bpf_probe_read() call (to read final field after chasing pointers) is replaced with bpf_probe_read_str(). Result of bpf_probe_read_str() is returned as a result of BPF_CORE_READ_STR_INTO() macro itself, so that applications can track return code and/or length of read string.
Patch set outline:
- patch #1 undoes previously added GCC-specific bpf-helpers.h include;
- patch #2 splits off legacy stuff we don't want to carry over;
- patch #3 adjusts CO-RE reloc tests to avoid subsequent naming conflict with BPF_CORE_READ;
- patch #4 splits off bpf_tracing.h;
- patch #5 moves bpf_{helpers,endian,tracing}.h into libbpf and adjusts Makefiles to include libbpf for header search;
- patch #6 adds variadic BPF_CORE_READ() macro family, as described above;
- patch #7 adds tests to verify all possible levels of pointer nestedness for BPF_CORE_READ(), as well as correctness test for BPF_CORE_READ_STR_INTO().
v2->v3:
- small formatting fixes and macro () fixes (Song);
v1->v2:
- fix CO-RE reloc tests before bpf_helpers.h move (Song);
- split off legacy stuff we don't want to carry over (Daniel, Toke);
- split off bpf_tracing.h (Daniel);
- fix samples/bpf build (assuming other fixes are applied);
- switch remaining maps either to bpf_map_def_legacy or BTF-defined maps;
Andrii Nakryiko (7):
selftests/bpf: undo GCC-specific bpf_helpers.h changes selftests/bpf: samples/bpf: split off legacy stuff from bpf_helpers.h selftests/bpf: adjust CO-RE reloc tests for new bpf_core_read() macro selftests/bpf: split off tracing-only helpers into bpf_tracing.h
libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
libbpf: add BPF_CORE_READ/BPF_CORE_READ_INTO helpers selftests/bpf: add BPF_CORE_READ and BPF_CORE_READ_STR_INTO macro
tests
samples/bpf/Makefile | 2 +-
samples/bpf/hbm_kern.h | 27 +-
samples/bpf/map_perf_test_kern.c | 24 +-
samples/bpf/offwaketime_kern.c | 1 +
samples/bpf/parse_ldabs.c | 1 +
samples/bpf/sampleip_kern.c | 1 +
samples/bpf/sockex1_kern.c | 1 +
samples/bpf/sockex2_kern.c | 1 +
samples/bpf/sockex3_kern.c | 1 +
samples/bpf/spintest_kern.c | 1 +
samples/bpf/tcbpf1_kern.c | 1 +
samples/bpf/test_map_in_map_kern.c | 16 +-
samples/bpf/test_overhead_kprobe_kern.c | 1 +
samples/bpf/test_probe_write_user_kern.c | 1 +
samples/bpf/trace_event_kern.c | 1 +
samples/bpf/tracex1_kern.c | 1 +
samples/bpf/tracex2_kern.c | 1 +
samples/bpf/tracex3_kern.c | 1 +
samples/bpf/tracex4_kern.c | 1 +
samples/bpf/tracex5_kern.c | 1 +
tools/lib/bpf/Makefile | 5 +-
.../selftests => lib}/bpf/bpf_endian.h | 0
.../selftests => lib}/bpf/bpf_helpers.h | 399 +++++++-----------
tools/lib/bpf/bpf_tracing.h | 195 +++++++++
tools/testing/selftests/bpf/Makefile | 2 +-
tools/testing/selftests/bpf/bpf_legacy.h | 39 ++
.../selftests/bpf/prog_tests/core_reloc.c | 8 +-
.../selftests/bpf/progs/core_reloc_types.h | 9 +
tools/testing/selftests/bpf/progs/loop1.c | 1 +
tools/testing/selftests/bpf/progs/loop2.c | 1 +
tools/testing/selftests/bpf/progs/loop3.c | 1 +
.../testing/selftests/bpf/progs/sockopt_sk.c | 13 +-
tools/testing/selftests/bpf/progs/tcp_rtt.c | 13 +-
.../selftests/bpf/progs/test_btf_haskv.c | 1 +
.../selftests/bpf/progs/test_btf_newkv.c | 1 +
.../bpf/progs/test_core_reloc_arrays.c | 10 +-
.../bpf/progs/test_core_reloc_flavors.c | 8 +-
.../bpf/progs/test_core_reloc_ints.c | 18 +-
.../bpf/progs/test_core_reloc_kernel.c | 60 ++-
.../bpf/progs/test_core_reloc_misc.c | 8 +-
.../bpf/progs/test_core_reloc_mods.c | 18 +-
.../bpf/progs/test_core_reloc_nesting.c | 6 +-
.../bpf/progs/test_core_reloc_primitives.c | 12 +-
.../bpf/progs/test_core_reloc_ptr_as_arr.c | 4 +-
44 files changed, 592 insertions(+), 325 deletions(-)
rename tools/{testing/selftests => lib}/bpf/bpf_endian.h (100%)
rename tools/{testing/selftests => lib}/bpf/bpf_helpers.h (65%)
create mode 100644 tools/lib/bpf/bpf_tracing.h
create mode 100644 tools/testing/selftests/bpf/bpf_legacy.hSignature · 1 line
2.17.1
PATCH v3 1/7 selftests/bpf: undo GCC-specific bpf_helpers.h changes
Having GCC provide its own bpf-helper.h is not the right approach and is going to be changed. Undo bpf_helpers.h change before moving bpf_helpers.h into libbpf.
--- tools/testing/selftests/bpf/bpf_helpers.h | 8 -------- 1 file changed, 8 deletions(-)
tools/testing/selftests/bpf/bpf_helpers.h +0 -8
@@ -13,8 +13,6 @@ ##__VA_ARGS__); \ }) -#ifdef __clang__ -/* helper macro to place programs, maps, license in * different sections in elf_bpf file. Section names * are interpreted by elf_bpf loader
@@ -258,12 +256,6 @@ struct bpf_map_def {unsigned int numa_node; }; -#else - -#include <bpf-helpers.h> - -#endif -#define BPF_ANNOTATE_KV_PAIR (name, type_key, type_val) \ struct ____btf_map_##name { \ type_key key; \
Signature · 1 line
2.17.1
Re: [PATCH v3 bpf-next 1/7] selftests/bpf: undo GCC-specific bpf_helpers.h changes
Quoted 8 lines · Andrii Nakryiko
Having GCC provide its own
bpf-helper.his not the right approach and is going to be changed. Undobpf_helpers.hchange before movingbpf_helpers.hinto libbpf.Acked-by Song Liu <[email protected]>Acked-by Ilya Leoshkevich <[email protected]>Acked-by John Fastabend <[email protected]>Signed-off-by Andrii Nakryiko <[email protected]>
PATCH v3 2/7 selftests/bpf: samples/bpf: split off legacy stuff from bpf_helpers.h
In reply to Andrii NakryikoSplit off few legacy things from bpf_helpers.h into separate bpf_legacy.h file:
- load_{byte|half|word};
- remove extra inner_idx and numa_node fields from bpf_map_def and introduce bpf_map_def_legacy for use in samples;
- move BPF_ANNOTATE_KV_PAIR into bpf_legacy.h.
Adjust samples and selftests accordingly by either including bpf_legacy.h and using bpf_map_def_legacy, or switching to BTF-defined maps altogether.
--- samples/bpf/hbm_kern.h | 27 ++++++------- samples/bpf/map_perf_test_kern.c | 23 +++++------ samples/bpf/parse_ldabs.c | 1 + samples/bpf/sockex1_kern.c | 1 + samples/bpf/sockex2_kern.c | 1 + samples/bpf/sockex3_kern.c | 1 + samples/bpf/tcbpf1_kern.c | 1 + samples/bpf/test_map_in_map_kern.c | 15 +++---- tools/testing/selftests/bpf/bpf_helpers.h | 24 +----------- tools/testing/selftests/bpf/bpf_legacy.h | 39 +++++++++++++++++++ .../testing/selftests/bpf/progs/sockopt_sk.c | 13 +++---- tools/testing/selftests/bpf/progs/tcp_rtt.c | 13 +++---- .../selftests/bpf/progs/test_btf_haskv.c | 1 + .../selftests/bpf/progs/test_btf_newkv.c | 1 + 14 files changed, 91 insertions(+), 70 deletions(-) create mode 100644 tools/testing/selftests/bpf/bpf_legacy.h
samples/bpf/hbm_kern.h +12 -15
@@ -59,21 +59,18 @@#define BYTES_PER_NS (delta, rate) ((((u64)(delta)) * (rate)) >> 20)#define BYTES_TO_NS (bytes, rate) div64_u64(((u64)(bytes)) << 20, (u64)(rate)) -struct bpf_map_def SEC("maps") queue_state = { - .type = BPF_MAP_TYPE_CGROUP_STORAGE, - .key_size = sizeof(struct bpf_cgroup_storage_key), - .value_size = sizeof(struct hbm_vqueue), -}; -BPF_ANNOTATE_KV_PAIR(queue_state, struct bpf_cgroup_storage_key, - struct hbm_vqueue); - -struct bpf_map_def SEC("maps") queue_stats = { - .type = BPF_MAP_TYPE_ARRAY, - .key_size = sizeof(u32), - .value_size = sizeof(struct hbm_queue_stats), - .max_entries = 1, -}; -BPF_ANNOTATE_KV_PAIR(queue_stats, int, struct hbm_queue_stats); +struct { + __uint(type, BPF_MAP_TYPE_CGROUP_STORAGE); + __type(key, struct bpf_cgroup_storage_key); + __type(value, struct hbm_vqueue); +} queue_state SEC(".maps"); + +struct { + __uint(type, BPF_MAP_TYPE_ARRAY); + __uint(max_entries, 1); + __type(key, u32); + __type(value, struct hvm_queue_stats); +} queue_stats SEC(".maps");struct hbm_pkt_info {int cwnd;
samples/bpf/map_perf_test_kern.c +12 -11
@@ -9,25 +9,26 @@#include <linux/version.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" #define MAX_ENTRIES 1000#define MAX_NR_CPUS 1024 -struct bpf_map_def SEC("maps") hash_map = { +struct bpf_map_def_legacy SEC("maps") hash_map = { .type = BPF_MAP_TYPE_HASH, .key_size = sizeof(u32), .value_size = sizeof(long), .max_entries = MAX_ENTRIES, }; -struct bpf_map_def SEC("maps") lru_hash_map = { +struct bpf_map_def_legacy SEC("maps") lru_hash_map = { .type = BPF_MAP_TYPE_LRU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long), .max_entries = 10000, }; -struct bpf_map_def SEC("maps") nocommon_lru_hash_map = { +struct bpf_map_def_legacy SEC("maps") nocommon_lru_hash_map = { .type = BPF_MAP_TYPE_LRU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long),
@@ -35,7 +36,7 @@ struct bpf_map_def SEC("maps") nocommon_lru_hash_map = { .map_flags = BPF_F_NO_COMMON_LRU, }; -struct bpf_map_def SEC("maps") inner_lru_hash_map = { +struct bpf_map_def_legacy SEC("maps") inner_lru_hash_map = { .type = BPF_MAP_TYPE_LRU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long),
@@ -44,20 +45,20 @@ struct bpf_map_def SEC("maps") inner_lru_hash_map = { .numa_node = 0, }; -struct bpf_map_def SEC("maps") array_of_lru_hashs = { +struct bpf_map_def_legacy SEC("maps") array_of_lru_hashs = { .type = BPF_MAP_TYPE_ARRAY_OF_MAPS, .key_size = sizeof(u32), .max_entries = MAX_NR_CPUS, }; -struct bpf_map_def SEC("maps") percpu_hash_map = { +struct bpf_map_def_legacy SEC("maps") percpu_hash_map = { .type = BPF_MAP_TYPE_PERCPU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long), .max_entries = MAX_ENTRIES, }; -struct bpf_map_def SEC("maps") hash_map_alloc = { +struct bpf_map_def_legacy SEC("maps") hash_map_alloc = { .type = BPF_MAP_TYPE_HASH, .key_size = sizeof(u32), .value_size = sizeof(long),
@@ -65,7 +66,7 @@ struct bpf_map_def SEC("maps") hash_map_alloc = { .map_flags = BPF_F_NO_PREALLOC, }; -struct bpf_map_def SEC("maps") percpu_hash_map_alloc = { +struct bpf_map_def_legacy SEC("maps") percpu_hash_map_alloc = { .type = BPF_MAP_TYPE_PERCPU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long),
@@ -73,7 +74,7 @@ struct bpf_map_def SEC("maps") percpu_hash_map_alloc = { .map_flags = BPF_F_NO_PREALLOC, }; -struct bpf_map_def SEC("maps") lpm_trie_map_alloc = { +struct bpf_map_def_legacy SEC("maps") lpm_trie_map_alloc = { .type = BPF_MAP_TYPE_LPM_TRIE, .key_size = 8, .value_size = sizeof(long),
@@ -81,14 +82,14 @@ struct bpf_map_def SEC("maps") lpm_trie_map_alloc = { .map_flags = BPF_F_NO_PREALLOC, }; -struct bpf_map_def SEC("maps") array_map = { +struct bpf_map_def_legacy SEC("maps") array_map = { .type = BPF_MAP_TYPE_ARRAY, .key_size = sizeof(u32), .value_size = sizeof(long), .max_entries = MAX_ENTRIES, }; -struct bpf_map_def SEC("maps") lru_hash_lookup_map = { +struct bpf_map_def_legacy SEC("maps") lru_hash_lookup_map = { .type = BPF_MAP_TYPE_LRU_HASH, .key_size = sizeof(u32), .value_size = sizeof(long),
samples/bpf/parse_ldabs.c +1 -0
@@ -12,6 +12,7 @@#include <linux/udp.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" #define DEFAULT_PKTGEN_UDP_PORT 9#define IP_MF 0x2000
samples/bpf/sockex1_kern.c +1 -0
@@ -3,6 +3,7 @@#include <uapi/linux/if_packet.h> #include <uapi/linux/ip.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" struct bpf_map_def SEC("maps") my_map = { .type = BPF_MAP_TYPE_ARRAY,
samples/bpf/sockex2_kern.c +1 -0
@@ -1,5 +1,6 @@#include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" #include <uapi/linux/in.h> #include <uapi/linux/if.h> #include <uapi/linux/if_ether.h>
samples/bpf/sockex3_kern.c +1 -0
@@ -6,6 +6,7 @@*/ #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" #include <uapi/linux/in.h> #include <uapi/linux/if.h> #include <uapi/linux/if_ether.h>
samples/bpf/tcbpf1_kern.c +1 -0
@@ -8,6 +8,7 @@#include <uapi/linux/filter.h> #include <uapi/linux/pkt_cls.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" /* compiler workaround */ #define _htonl __builtin_bswap32
samples/bpf/test_map_in_map_kern.c +8 -7
@@ -11,11 +11,12 @@#include <uapi/linux/bpf.h> #include <uapi/linux/in6.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" #define MAX_NR_PORTS 65536/* map #0 */ -struct bpf_map_def SEC("maps") port_a = { +struct bpf_map_def_legacy SEC("maps") port_a = { .type = BPF_MAP_TYPE_ARRAY, .key_size = sizeof(u32), .value_size = sizeof(int),
@@ -23,7 +24,7 @@ struct bpf_map_def SEC("maps") port_a = { };/* map #1 */ -struct bpf_map_def SEC("maps") port_h = { +struct bpf_map_def_legacy SEC("maps") port_h = { .type = BPF_MAP_TYPE_HASH, .key_size = sizeof(u32), .value_size = sizeof(int),
@@ -31,7 +32,7 @@ struct bpf_map_def SEC("maps") port_h = { };/* map #2 */ -struct bpf_map_def SEC("maps") reg_result_h = { +struct bpf_map_def_legacy SEC("maps") reg_result_h = { .type = BPF_MAP_TYPE_HASH, .key_size = sizeof(u32), .value_size = sizeof(int),
@@ -39,7 +40,7 @@ struct bpf_map_def SEC("maps") reg_result_h = { };/* map #3 */ -struct bpf_map_def SEC("maps") inline_result_h = { +struct bpf_map_def_legacy SEC("maps") inline_result_h = { .type = BPF_MAP_TYPE_HASH, .key_size = sizeof(u32), .value_size = sizeof(int),
@@ -47,7 +48,7 @@ struct bpf_map_def SEC("maps") inline_result_h = { };/* map #4 */ /* Test case #0 */ -struct bpf_map_def SEC("maps") a_of_port_a = { +struct bpf_map_def_legacy SEC("maps") a_of_port_a = { .type = BPF_MAP_TYPE_ARRAY_OF_MAPS, .key_size = sizeof(u32), .inner_map_idx = 0, /* map_fd[0] is port_a */
@@ -55,7 +56,7 @@ struct bpf_map_def SEC("maps") a_of_port_a = { };/* map #5 */ /* Test case #1 */ -struct bpf_map_def SEC("maps") h_of_port_a = { +struct bpf_map_def_legacy SEC("maps") h_of_port_a = { .type = BPF_MAP_TYPE_HASH_OF_MAPS, .key_size = sizeof(u32), .inner_map_idx = 0, /* map_fd[0] is port_a */
@@ -63,7 +64,7 @@ struct bpf_map_def SEC("maps") h_of_port_a = { };/* map #6 */ /* Test case #2 */ -struct bpf_map_def SEC("maps") h_of_port_h = { +struct bpf_map_def_legacy SEC("maps") h_of_port_h = { .type = BPF_MAP_TYPE_HASH_OF_MAPS, .key_size = sizeof(u32), .inner_map_idx = 1, /* map_fd[1] is port_h */
tools/testing/selftests/bpf/bpf_helpers.h +1 -23
@@ -232,19 +232,8 @@ static long long (*bpf_tcp_gen_syncookie)(struct bpf_sock *sk, void *ip, int ip_len, void *tcp, int tcp_len) = (void *) BPF_FUNC_tcp_gen_syncookie; -/* llvm builtin functions that eBPF C program may use to -* emit BPF_LD_ABS and BPF_LD_IND instructions -*/ -struct sk_buff ; -unsigned long long load_byte (void *skb, -unsigned long long off) asm("llvm.bpf.load.byte" ); -unsigned long long load_half (void *skb, -unsigned long long off) asm("llvm.bpf.load.half" ); -unsigned long long load_word (void *skb, -unsigned long long off) asm("llvm.bpf.load.word" ); -/* a helper structure used by eBPF C program -* to describe map attributes to elf_bpf loader +* to describe BPF map attributes to libbpf loader */ struct bpf_map_def {unsigned int type;
@@ -252,19 +241,8 @@ struct bpf_map_def {unsigned int value_size;unsigned int max_entries;unsigned int map_flags; -unsigned int inner_map_idx; -unsigned int numa_node; }; -#define BPF_ANNOTATE_KV_PAIR (name, type_key, type_val) \ - struct ____btf_map_##name { \ - type_key key; \ - type_val value; \ - }; \ - struct ____btf_map_##name \ - __attribute__ ((section(".maps." #name), used)) \ - ____btf_map_##name = { } -static int (*bpf_skb_load_bytes)(void *ctx,int off,void *to,int len) = (void *) BPF_FUNC_skb_load_bytes; static int (*bpf_skb_load_bytes_relative)(void *ctx, int off, void *to, int len, __u32 start_header) =
tools/testing/selftests/bpf/bpf_legacy.h new file +39 -0
@@ -0,0 +1,39 @@ +/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ +#ifndef __BPF_LEGACY__ +#define __BPF_LEGACY__ + +/* +* legacy bpf_map_def with extra fields supported only by bpf_load(), do not +* use outside of samples/bpf +*/ +struct bpf_map_def_legacy { +unsigned int type; +unsigned int key_size; +unsigned int value_size; +unsigned int max_entries; +unsigned int map_flags; +unsigned int inner_map_idx; +unsigned int numa_node; +}; + +#define BPF_ANNOTATE_KV_PAIR (name, type_key, type_val) \ + struct ____btf_map_##name { \ + type_key key; \ + type_val value; \ + }; \ + struct ____btf_map_##name \ + __attribute__ ((section(".maps." #name), used)) \ + ____btf_map_##name = { } + +/* llvm builtin functions that eBPF C program may use to +* emit BPF_LD_ABS and BPF_LD_IND instructions +*/ +unsigned long long load_byte (void *skb, +unsigned long long off) asm("llvm.bpf.load.byte" ); +unsigned long long load_half (void *skb, +unsigned long long off) asm("llvm.bpf.load.half" ); +unsigned long long load_word (void *skb, +unsigned long long off) asm("llvm.bpf.load.word" ); + +#endif +
tools/testing/selftests/bpf/progs/sockopt_sk.c +6 -7
@@ -14,13 +14,12 @@ struct sockopt_sk {__u8 val; }; -struct bpf_map_def SEC("maps") socket_storage_map = { - .type = BPF_MAP_TYPE_SK_STORAGE, - .key_size = sizeof(int), - .value_size = sizeof(struct sockopt_sk), - .map_flags = BPF_F_NO_PREALLOC, -}; -BPF_ANNOTATE_KV_PAIR(socket_storage_map, int, struct sockopt_sk); +struct { + __uint(type, BPF_MAP_TYPE_SK_STORAGE); + __uint(map_flags, BPF_F_NO_PREALLOC); + __type(key, int); + __type(value, struct sockopt_sk); +} socket_storage_map SEC(".maps"); SEC("cgroup/getsockopt")int _getsockopt (struct bpf_sockopt *ctx)
tools/testing/selftests/bpf/progs/tcp_rtt.c +6 -7
@@ -13,13 +13,12 @@ struct tcp_rtt_storage {__u32 icsk_retransmits; }; -struct bpf_map_def SEC("maps") socket_storage_map = { - .type = BPF_MAP_TYPE_SK_STORAGE, - .key_size = sizeof(int), - .value_size = sizeof(struct tcp_rtt_storage), - .map_flags = BPF_F_NO_PREALLOC, -}; -BPF_ANNOTATE_KV_PAIR(socket_storage_map, int, struct tcp_rtt_storage); +struct { + __uint(type, BPF_MAP_TYPE_SK_STORAGE); + __uint(map_flags, BPF_F_NO_PREALLOC); + __type(key, int); + __type(value, struct tcp_rtt_storage); +} socket_storage_map SEC(".maps"); SEC("sockops")int _sockops (struct bpf_sock_ops *ctx)
tools/testing/selftests/bpf/progs/test_btf_haskv.c +1 -0
@@ -2,6 +2,7 @@/* Copyright (c) 2018 Facebook */ #include <linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" int _version SEC("version") = 1;
tools/testing/selftests/bpf/progs/test_btf_newkv.c +1 -0
@@ -2,6 +2,7 @@/* Copyright (c) 2018 Facebook */ #include <linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_legacy.h" int _version SEC("version") = 1;
Signature · 1 line
2.17.1
Re: [PATCH v3 bpf-next 2/7] selftests/bpf: samples/bpf: split off legacy stuff from bpf_helpers.h
Quoted 7 lines · Andrii Nakryiko
Split off few legacy things from
bpf_helpers.hinto separatebpf_legacy.hfile:
- load_{byte|half|word};
- remove extrainner_idxandnuma_nodefields frombpf_map_defand introducebpf_map_def_legacyfor use in samples;
- moveBPF_ANNOTATE_KV_PAIRintobpf_legacy.h.
Quoted 7 lines
Adjust samples and selftests accordingly by either including
bpf_legacy.hand usingbpf_map_def_legacy, or switching to BTF-defined maps altogether.Acked-by John Fastabend <[email protected]>Acked-by Song Liu <[email protected]>Signed-off-by Andrii Nakryiko <[email protected]>
PATCH v3 3/7 selftests/bpf: adjust CO-RE reloc tests for new bpf_core_read() macro
In reply to Andrii NakryikoTo allow adding a variadic BPF_CORE_READ macro with slightly different syntax and semantics, define CORE_READ in CO-RE reloc tests, which is a thin wrapper around low-level bpf_core_read() macro, which in turn is just a wrapper around bpf_probe_read().
--- tools/testing/selftests/bpf/bpf_helpers.h | 8 ++++---- .../bpf/progs/test_core_reloc_arrays.c | 10 ++++++---- .../bpf/progs/test_core_reloc_flavors.c | 8 +++++--- .../selftests/bpf/progs/test_core_reloc_ints.c | 18 ++++++++++-------- .../bpf/progs/test_core_reloc_kernel.c | 6 ++++-- .../selftests/bpf/progs/test_core_reloc_misc.c | 8 +++++--- .../selftests/bpf/progs/test_core_reloc_mods.c | 18 ++++++++++-------- .../bpf/progs/test_core_reloc_nesting.c | 6 ++++-- .../bpf/progs/test_core_reloc_primitives.c | 12 +++++++----- .../bpf/progs/test_core_reloc_ptr_as_arr.c | 4 +++- 10 files changed, 58 insertions(+), 40 deletions(-)
tools/testing/selftests/bpf/bpf_helpers.h +4 -4
@@ -483,7 +483,7 @@ struct pt_regs;#endif /* -* BPF_CORE_READ abstracts away bpf_probe_read() call and captures offset +* bpf_core_read() abstracts away bpf_probe_read() call and captures offset * relocation for source address using __builtin_preserve_access_index() * built-in, provided by Clang. *
@@ -498,8 +498,8 @@ struct pt_regs;* actual field offset, based on target kernel BTF type that matches original * (local) BTF, used to record relocation. */ -#define BPF_CORE_READ (dst, src) \ - bpf_probe_read((dst), sizeof(*(src)), \ - __builtin_preserve_access_index(src)) +#define bpf_core_read (dst, sz, src) \ + bpf_probe_read(dst, sz, \ + (const void *)__builtin_preserve_access_index(src))#endif
tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c +6 -4
@@ -31,6 +31,8 @@ struct core_reloc_arrays {struct core_reloc_arrays_substruct d[1 ][2 ]; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_arrays (void *ctx) {
@@ -38,16 +40,16 @@ int test_core_arrays(void *ctx)struct core_reloc_arrays_output *out = (void *)&data.out;/* in->a[2] */ -if (BPF_CORE_READ (&out->a2, &in->a[2 ])) +if (CORE_READ (&out->a2, &in->a[2 ]))return 1 ;/* in->b[1][2][3] */ -if (BPF_CORE_READ (&out->b123, &in->b[1 ][2 ][3 ])) +if (CORE_READ (&out->b123, &in->b[1 ][2 ][3 ]))return 1 ;/* in->c[1].c */ -if (BPF_CORE_READ (&out->c1c, &in->c[1 ].c)) +if (CORE_READ (&out->c1c, &in->c[1 ].c))return 1 ;/* in->d[0][0].d */ -if (BPF_CORE_READ (&out->d00d, &in->d[0 ][0 ].d)) +if (CORE_READ (&out->d00d, &in->d[0 ][0 ].d))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c +5 -3
@@ -39,6 +39,8 @@ struct core_reloc_flavors___weird { }; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_flavors (void *ctx) {
@@ -48,13 +50,13 @@ int test_core_flavors(void *ctx)struct core_reloc_flavors *out = (void *)&data.out;/* read a using weird layout */ -if (BPF_CORE_READ (&out->a, &in_weird->a)) +if (CORE_READ (&out->a, &in_weird->a))return 1 ;/* read b using reversed layout */ -if (BPF_CORE_READ (&out->b, &in_rev->b)) +if (CORE_READ (&out->b, &in_rev->b))return 1 ;/* read c using original layout */ -if (BPF_CORE_READ (&out->c, &in_orig->c)) +if (CORE_READ (&out->c, &in_orig->c))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_ints.c +10 -8
@@ -23,20 +23,22 @@ struct core_reloc_ints {int64_t s64_field; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_ints (void *ctx) {struct core_reloc_ints *in = (void *)&data.in;struct core_reloc_ints *out = (void *)&data.out; -if (BPF_CORE_READ (&out->u8_field, &in->u8_field) || -BPF_CORE_READ (&out->s8_field, &in->s8_field) || -BPF_CORE_READ (&out->u16_field, &in->u16_field) || -BPF_CORE_READ (&out->s16_field, &in->s16_field) || -BPF_CORE_READ (&out->u32_field, &in->u32_field) || -BPF_CORE_READ (&out->s32_field, &in->s32_field) || -BPF_CORE_READ (&out->u64_field, &in->u64_field) || -BPF_CORE_READ (&out->s64_field, &in->s64_field)) +if (CORE_READ (&out->u8_field, &in->u8_field) || +CORE_READ (&out->s8_field, &in->s8_field) || +CORE_READ (&out->u16_field, &in->u16_field) || +CORE_READ (&out->s16_field, &in->s16_field) || +CORE_READ (&out->u32_field, &in->u32_field) || +CORE_READ (&out->s32_field, &in->s32_field) || +CORE_READ (&out->u64_field, &in->u64_field) || +CORE_READ (&out->s64_field, &in->s64_field))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c +4 -2
@@ -17,6 +17,8 @@ struct task_struct {int tgid; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_kernel (void *ctx) {
@@ -24,8 +26,8 @@ int test_core_kernel(void *ctx)uint64_t pid_tgid =bpf_get_current_pid_tgid ();int pid, tgid; -if (BPF_CORE_READ (&pid, &task->pid) || -BPF_CORE_READ (&tgid, &task->tgid)) +if (CORE_READ (&pid, &task->pid) || +CORE_READ (&tgid, &task->tgid))return 1 ;/* validate pid + tgid matches */
tools/testing/selftests/bpf/progs/test_core_reloc_misc.c +5 -3
@@ -32,6 +32,8 @@ struct core_reloc_misc_extensible {int b; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_misc (void *ctx) {
@@ -41,15 +43,15 @@ int test_core_misc(void *ctx)struct core_reloc_misc_output *out = (void *)&data.out;/* record two different relocations with the same accessor string */ -if (BPF_CORE_READ (&out->a, &in_a->a1) ||/* accessor: 0:0 */ -BPF_CORE_READ (&out->b, &in_b->b1))/* accessor: 0:0 */ +if (CORE_READ (&out->a, &in_a->a1) ||/* accessor: 0:0 */ +CORE_READ (&out->b, &in_b->b1))/* accessor: 0:0 */ return 1 ;/* Validate relocations capture array-only accesses for structs with * fixed header, but with potentially extendable tail. This will read * first 4 bytes of 2nd element of in_ext array of potentially * variably sized struct core_reloc_misc_extensible. */ -if (BPF_CORE_READ (&out->c, &in_ext[2 ]))/* accessor: 2 */ +if (CORE_READ (&out->c, &in_ext[2 ]))/* accessor: 2 */ return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_mods.c +10 -8
@@ -41,20 +41,22 @@ struct core_reloc_mods {core_reloc_mods_substruct_t h; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_mods (void *ctx) {struct core_reloc_mods *in = (void *)&data.in;struct core_reloc_mods_output *out = (void *)&data.out; -if (BPF_CORE_READ (&out->a, &in->a) || -BPF_CORE_READ (&out->b, &in->b) || -BPF_CORE_READ (&out->c, &in->c) || -BPF_CORE_READ (&out->d, &in->d) || -BPF_CORE_READ (&out->e, &in->e[2 ]) || -BPF_CORE_READ (&out->f, &in->f[1 ]) || -BPF_CORE_READ (&out->g, &in->g.x) || -BPF_CORE_READ (&out->h, &in->h.y)) +if (CORE_READ (&out->a, &in->a) || +CORE_READ (&out->b, &in->b) || +CORE_READ (&out->c, &in->c) || +CORE_READ (&out->d, &in->d) || +CORE_READ (&out->e, &in->e[2 ]) || +CORE_READ (&out->f, &in->f[1 ]) || +CORE_READ (&out->g, &in->g.x) || +CORE_READ (&out->h, &in->h.y))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c +4 -2
@@ -30,15 +30,17 @@ struct core_reloc_nesting { } b; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_nesting (void *ctx) {struct core_reloc_nesting *in = (void *)&data.in;struct core_reloc_nesting *out = (void *)&data.out; -if (BPF_CORE_READ (&out->a.a.a, &in->a.a.a)) +if (CORE_READ (&out->a.a.a, &in->a.a.a))return 1 ; -if (BPF_CORE_READ (&out->b.b.b, &in->b.b.b)) +if (CORE_READ (&out->b.b.b, &in->b.b.b))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c +7 -5
@@ -25,17 +25,19 @@ struct core_reloc_primitives {int (*f)(const char *); }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_primitives (void *ctx) {struct core_reloc_primitives *in = (void *)&data.in;struct core_reloc_primitives *out = (void *)&data.out; -if (BPF_CORE_READ (&out->a, &in->a) || -BPF_CORE_READ (&out->b, &in->b) || -BPF_CORE_READ (&out->c, &in->c) || -BPF_CORE_READ (&out->d, &in->d) || -BPF_CORE_READ (&out->f, &in->f)) +if (CORE_READ (&out->a, &in->a) || +CORE_READ (&out->b, &in->b) || +CORE_READ (&out->c, &in->c) || +CORE_READ (&out->d, &in->d) || +CORE_READ (&out->f, &in->f))return 1 ;return 0 ;
tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c +3 -1
@@ -16,13 +16,15 @@ struct core_reloc_ptr_as_arr {int a; }; +#define CORE_READ (dst, src) bpf_core_read(dst, sizeof(*(dst)), src) + SEC("raw_tracepoint/sys_enter")int test_core_ptr_as_arr (void *ctx) {struct core_reloc_ptr_as_arr *in = (void *)&data.in;struct core_reloc_ptr_as_arr *out = (void *)&data.out; -if (BPF_CORE_READ (&out->a, &in[2 ].a)) +if (CORE_READ (&out->a, &in[2 ].a))return 1 ;return 0 ;
Signature · 1 line
2.17.1
PATCH v3 4/7 selftests/bpf: split off tracing-only helpers into bpf_tracing.h
In reply to Andrii NakryikoSplit-off PT_REGS-related helpers into bpf_tracing.h header. Adjust selftests and samples to include it where necessary.
--- samples/bpf/map_perf_test_kern.c | 1 + samples/bpf/offwaketime_kern.c | 1 + samples/bpf/sampleip_kern.c | 1 + samples/bpf/spintest_kern.c | 1 + samples/bpf/test_map_in_map_kern.c | 1 + samples/bpf/test_overhead_kprobe_kern.c | 1 + samples/bpf/test_probe_write_user_kern.c | 1 + samples/bpf/trace_event_kern.c | 1 + samples/bpf/tracex1_kern.c | 1 + samples/bpf/tracex2_kern.c | 1 + samples/bpf/tracex3_kern.c | 1 + samples/bpf/tracex4_kern.c | 1 + samples/bpf/tracex5_kern.c | 1 + tools/testing/selftests/bpf/bpf_helpers.h | 210 ++-------------------- tools/testing/selftests/bpf/bpf_tracing.h | 195 ++++++++++++++++++++ tools/testing/selftests/bpf/progs/loop1.c | 1 + tools/testing/selftests/bpf/progs/loop2.c | 1 + tools/testing/selftests/bpf/progs/loop3.c | 1 + 18 files changed, 221 insertions(+), 200 deletions(-) create mode 100644 tools/testing/selftests/bpf/bpf_tracing.h
samples/bpf/map_perf_test_kern.c +1 -0
@@ -10,6 +10,7 @@#include <uapi/linux/bpf.h> #include "bpf_helpers.h" #include "bpf_legacy.h" +#include "bpf_tracing.h" #define MAX_ENTRIES 1000#define MAX_NR_CPUS 1024
samples/bpf/offwaketime_kern.c +1 -0
@@ -6,6 +6,7 @@*/ #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" #include <uapi/linux/ptrace.h> #include <uapi/linux/perf_event.h> #include <linux/version.h>
samples/bpf/sampleip_kern.c +1 -0
@@ -9,6 +9,7 @@#include <uapi/linux/bpf.h> #include <uapi/linux/bpf_perf_event.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" #define MAX_IPS 8192
samples/bpf/spintest_kern.c +1 -0
@@ -10,6 +10,7 @@#include <uapi/linux/bpf.h> #include <uapi/linux/perf_event.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct bpf_map_def SEC("maps") my_map = { .type = BPF_MAP_TYPE_HASH,
samples/bpf/test_map_in_map_kern.c +1 -0
@@ -12,6 +12,7 @@#include <uapi/linux/in6.h> #include "bpf_helpers.h" #include "bpf_legacy.h" +#include "bpf_tracing.h" #define MAX_NR_PORTS 65536
samples/bpf/test_overhead_kprobe_kern.c +1 -0
@@ -8,6 +8,7 @@#include <linux/ptrace.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" #define _ (P ) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
samples/bpf/test_probe_write_user_kern.c +1 -0
@@ -9,6 +9,7 @@#include <uapi/linux/bpf.h> #include <linux/version.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct bpf_map_def SEC("maps") dnat_map = { .type = BPF_MAP_TYPE_HASH,
samples/bpf/trace_event_kern.c +1 -0
@@ -10,6 +10,7 @@#include <uapi/linux/bpf_perf_event.h> #include <uapi/linux/perf_event.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct key_t {char comm[TASK_COMM_LEN ];
samples/bpf/tracex1_kern.c +1 -0
@@ -9,6 +9,7 @@#include <uapi/linux/bpf.h> #include <linux/version.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" #define _ (P ) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
samples/bpf/tracex2_kern.c +1 -0
@@ -9,6 +9,7 @@#include <linux/version.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct bpf_map_def SEC("maps") my_map = { .type = BPF_MAP_TYPE_HASH,
samples/bpf/tracex3_kern.c +1 -0
@@ -9,6 +9,7 @@#include <linux/version.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct bpf_map_def SEC("maps") my_map = { .type = BPF_MAP_TYPE_HASH,
samples/bpf/tracex4_kern.c +1 -0
@@ -8,6 +8,7 @@#include <linux/version.h> #include <uapi/linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" struct pair {u64 val;
samples/bpf/tracex5_kern.c +1 -0
@@ -11,6 +11,7 @@#include <uapi/linux/unistd.h> #include "syscall_nrs.h" #include "bpf_helpers.h" +#include "bpf_tracing.h" #define PROG (F ) SEC("kprobe/"__stringify(F)) int bpf_func_##F
tools/testing/selftests/bpf/bpf_helpers.h +10 -200
222 lines not on this page · raw message
tools/testing/selftests/bpf/bpf_tracing.h new file +195 -0
195 lines not on this page · raw message
tools/testing/selftests/bpf/progs/loop1.c +1 -0
@@ -7,6 +7,7 @@#include <stdbool.h> #include <linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" char _license[] SEC("license") = "GPL";
tools/testing/selftests/bpf/progs/loop2.c +1 -0
@@ -7,6 +7,7 @@#include <stdbool.h> #include <linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" char _license[] SEC("license") = "GPL";
tools/testing/selftests/bpf/progs/loop3.c +1 -0
@@ -7,6 +7,7 @@#include <stdbool.h> #include <linux/bpf.h> #include "bpf_helpers.h" +#include "bpf_tracing.h" char _license[] SEC("license") = "GPL";
Signature · 1 line
2.17.1
Re: [PATCH v3 bpf-next 4/7] selftests/bpf: split off tracing-only helpers into bpf_tracing.h
Quoted 6 lines · Andrii Nakryiko
Split-off PT_REGS-related helpers into
bpf_tracing.hheader. Adjust selftests and samples to include it where necessary.Acked-by John Fastabend <[email protected]>Acked-by Song Liu <[email protected]>Signed-off-by Andrii Nakryiko <[email protected]>
PATCH v3 5/7 libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
In reply to Andrii NakryikoMove bpf_helpers.h, bpf_tracing.h, and bpf_endian.h into libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.
---
samples/bpf/Makefile | 2 +-
tools/lib/bpf/Makefile | 5 ++++-
tools/{testing/selftests => lib}/bpf/bpf_endian.h | 0
tools/{testing/selftests => lib}/bpf/bpf_helpers.h | 0
tools/{testing/selftests => lib}/bpf/bpf_tracing.h | 0
tools/testing/selftests/bpf/Makefile | 2 +-
6 files changed, 6 insertions(+), 3 deletions(-)
rename tools/{testing/selftests => lib}/bpf/bpf_endian.h (100%)
rename tools/{testing/selftests => lib}/bpf/bpf_helpers.h (100%)
rename tools/{testing/selftests => lib}/bpf/bpf_tracing.h (100%)samples/bpf/Makefile +1 -1
@@ -283,7 +283,7 @@ $(obj)/hbm_edt_kern.o: $(src)/hbm.h $(src)/hbm_kern.h $(obj)/%.o: $(src)/%.c @echo " CLANG-bpf " $@ $(Q)$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \ - -I$(srctree)/tools/testing/selftests/bpf/ \ + -I$(srctree)/tools/testing/selftests/bpf/ -I$(srctree)/tools/lib/bpf/ \ -D__KERNEL__ -D__BPF_TRACING__ -Wno-unused-value -Wno-pointer-sign \ -D__TARGET_ARCH_$(SRCARCH) -Wno-compare-distinct-pointer-types \ -Wno-gnu-variable-sized-type-not-at-end \
tools/lib/bpf/Makefile +4 -1
@@ -240,7 +240,10 @@ install_headers: $(call do_install,libbpf.h,$(prefix)/include/bpf,644); \ $(call do_install,btf.h,$(prefix)/include/bpf,644); \ $(call do_install,libbpf_util.h,$(prefix)/include/bpf,644); \ - $(call do_install,xsk.h,$(prefix)/include/bpf,644); + $(call do_install,xsk.h,$(prefix)/include/bpf,644); \ + $(call do_install,bpf_helpers.h,$(prefix)/include/bpf,644); \ + $(call do_install,bpf_tracing.h,$(prefix)/include/bpf,644); \ + $(call do_install,bpf_endian.h,$(prefix)/include/bpf,644); install_pkgconfig: $(PC_FILE) $(call QUIET_INSTALL, $(PC_FILE)) \
tools/lib/bpf/bpf_endian.h renamed from tools/testing/selftests/bpf/bpf_endian.h +0 -0
tools/lib/bpf/bpf_helpers.h renamed from tools/testing/selftests/bpf/bpf_helpers.h +0 -0
tools/lib/bpf/bpf_tracing.h renamed from tools/testing/selftests/bpf/bpf_tracing.h +0 -0
tools/testing/selftests/bpf/Makefile +1 -1
@@ -148,7 +148,7 @@ $(shell $(1) -v -E - </dev/null 2>&1 \ endef CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG)) BPF_CFLAGS = -I. -I./include/uapi -I../../../include/uapi \ - -I$(OUTPUT)/../usr/include -D__TARGET_ARCH_$(SRCARCH) + -I$(BPFDIR) -I$(OUTPUT)/../usr/include -D__TARGET_ARCH_$(SRCARCH) CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \ -Wno-compare-distinct-pointer-types
Signature · 1 line
2.17.1
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 6 lines · Andrii Nakryiko
Move
bpf_helpers.h,bpf_tracing.h, andbpf_endian.hinto libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.Acked-by Song Liu <[email protected]>Signed-off-by Andrii Nakryiko <[email protected]>
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
In reply to Andrii NakryikoQuoted 3 lines · Andrii Nakryiko
Move
bpf_helpers.h,bpf_tracing.h, andbpf_endian.hinto libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.
There are side effects to bringing bpf_helpers.h into libbpf if this gets propagated to the github sync.
bpf_helpers.h references BPF_FUNC_* which are defined in the uapi/linux/bpf.h header. That is a kernel version dependent api file which means attempts to use newer libbpf with older kernel headers is going to throw errors when compiling bpf programs -- bpf_helpers.h will contain undefined BPF_FUNC references.
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 6 lines · David Ahern
Quoted 3 lines · Andrii Nakryiko
Move
bpf_helpers.h,bpf_tracing.h, andbpf_endian.hinto libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.
Quoted 8 lines
There are side effects to bringing
bpf_helpers.hinto libbpf if this gets propagated to the github sync.
bpf_helpers.hreferences BPF_FUNC_* which are defined in theuapi/linux/bpf.hheader. That is a kernel version dependent api file which means attempts to use newer libbpf with older kernel headers is going to throw errors when compiling bpf programs --bpf_helpers.hwill contain undefinedBPF_FUNCreferences.
That's true, but I'm wondering if maintaining a copy of that enum in bpf_helpers.h itself is a good answer here?
bpf_helpers.h will be most probably used with BPF CO-RE and auto-generated vmlinux.h with all the enums and types. In that case, you'll probably want to use vmlinux.h for one of the latest kernels anyways.
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once bpf_helpers.h is in libbpf. If bpf_helpers.h doesn't work for someone, it's no worse than it is today when users don't have bpf_helpers.h at all.
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 16 lines · Andrii Nakryiko
Quoted 6 lines · David Ahern
Quoted 3 lines · Andrii Nakryiko
Move
bpf_helpers.h,bpf_tracing.h, andbpf_endian.hinto libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.Quoted 8 lines
There are side effects to bringing
bpf_helpers.hinto libbpf if this gets propagated to the github sync.
bpf_helpers.hreferences BPF_FUNC_* which are defined in theuapi/linux/bpf.hheader. That is a kernel version dependent api file which means attempts to use newer libbpf with older kernel headers is going to throw errors when compiling bpf programs --bpf_helpers.hwill contain undefinedBPF_FUNCreferences.
Quoted 7 lines
That's true, but I'm wondering if maintaining a copy of that enum in
bpf_helpers.hitself is a good answer here?
bpf_helpers.hwill be most probably used with BPF CO-RE and auto-generatedvmlinux.hwith all the enums and types. In that case, you'll probably want to usevmlinux.hfor one of the latest kernels anyways.
I'm not following you; my interpretation of your comment seems like you are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Quoted 7 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.
If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's bpf_helpers.h over whatever they are using today.
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 26 lines · David Ahern
Quoted 16 lines · Andrii Nakryiko
Quoted 6 lines · David Ahern
Quoted 3 lines · Andrii Nakryiko
Move
bpf_helpers.h,bpf_tracing.h, andbpf_endian.hinto libbpf. Ensure they are installed along the other libbpf headers. Also, adjust selftests and samples include path to include libbpf now.Quoted 8 lines
There are side effects to bringing
bpf_helpers.hinto libbpf if this gets propagated to the github sync.
bpf_helpers.hreferences BPF_FUNC_* which are defined in theuapi/linux/bpf.hheader. That is a kernel version dependent api file which means attempts to use newer libbpf with older kernel headers is going to throw errors when compiling bpf programs --bpf_helpers.hwill contain undefinedBPF_FUNCreferences.Quoted 7 lines
That's true, but I'm wondering if maintaining a copy of that enum in
bpf_helpers.hitself is a good answer here?
bpf_helpers.hwill be most probably used with BPF CO-RE and auto-generatedvmlinux.hwith all the enums and types. In that case, you'll probably want to usevmlinux.hfor one of the latest kernels anyways.
Quoted 5 lines
I'm not following you; my interpretation of your comment seems like you are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include bpf_helpers.h?
What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single vmlinux.h header with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.
So what I was pointing out there was that this vmlinux.h would be ideally generated from latest kernel and not having latest BPF_FUNC_xxx shouldn't be a problem. But see below about situation being worse.
Quoted 9 lines
Quoted 8 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.
Quoted 3 lines
If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.
Today bpf_helpers.h don't exist for users or am I missing something? bpf_helpers.h right now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. If bpf_helpers.h doesn't work for some specific use case, just don't use it (yet?).
I'm still failing to see how it's worse than situation today.
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.
Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.
Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.
Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
distro can package bpf/btf uapi headers into libbpf package. Users linking with libbpf.a/libbpf.so can use bpf/btf.h with include path pointing to libbpf dev package include directory. Could this work?
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 10 lines · Yonghong Song
Quoted 9 lines
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
Quoted 4 lines
distro can package bpf/btf uapi headers into libbpf package. Users linking with libbpf.a/libbpf.so can use
bpf/btf.hwith include path pointing to libbpf dev package include directory. Could this work?
IMHO that'd be the first thing to try.
Andrii, your option (c) also seems to me like a pretty natural fit, although it'd be a little strange to have code depending on the kernel version in tree :S
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
In reply to Yonghong SongQuoted 4 lines · Yonghong Song
distro can package bpf/btf uapi headers into libbpf package. Users linking with libbpf.a/libbpf.so can use
bpf/btf.hwith include path pointing to libbpf dev package include directory. Could this work?
I don't think it would. Distros have often a policy against bundling files that are available from one package (in this case, kernel-headers or similar) in a different package (libbpf).
The correct way is making the libbpf package depend on a particular version of kernel-headers (or newer). As I said, I don't see a problem here. It's not a special situation, it's just usual dependencies.
Jiri
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 7 lines · Jiri Benc
Quoted 4 lines · Yonghong Song
distro can package bpf/btf uapi headers into libbpf package. Users linking with libbpf.a/libbpf.so can use
bpf/btf.hwith include path pointing to libbpf dev package include directory. Could this work?
Quoted 8 lines
I don't think it would. Distros have often a policy against bundling files that are available from one package (in this case, kernel-headers or similar) in a different package (libbpf).
The correct way is making the libbpf package depend on a particular version of kernel-headers (or newer). As I said, I don't see a problem here. It's not a special situation, it's just usual dependencies.
We ended up switching to auto-generating BPF helpers from UAPI headers w/ hardcoding BPF_FUNC_xxx values in bpf_helpers.h. So there is now no dependency on any specific kernel version there.
Quoted 1 line
Jiri
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
In reply to Jakub KicinskiQuoted 58 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
1. About "adding bpf_helpers.h will make things worse". I categorically disagree, bpf_helpers.h doesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.
Also, I think we should have informal notion of "experimental" features and APIs which we add to get real-world experience of using it, before we crystalize it to something that we have to maintain backwards compatibility and never-breaking user experience for. Its sometimes impossible to finesse all the tiny issues before we get prolonged enough experience w/ real applications in a real set up. If we are going to wait to resolve all the tiny possible problems, we'll cripple outselves very significantly. I think bpf_helpers.h move walls into this "experimental" thing, which we obviously will try to "stabilize" ASAP, but we need it to be part of libbpf first to start using it in production. So in that sense it's a huge improvement already and I think we should land it as is and keep improving it, not stall progress right now.
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those BPF_FUNC_ constants. Super painful and not nice, but will work.
b. copy/paste enum bpf_func_id definition into bpf_helpers.h itself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict with vmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).
c. BPF UAPI header has __BPF_FUNC_MAPPER macro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/ vmlinux.h there is no such macro, so this should work as well.
I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Opinions?
Daniel Borkmann
maintainerRe: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 67 lines · Andrii Nakryiko
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
Quoted 7 lines
1. About "adding
bpf_helpers.hwill make things worse". I categorically disagree,bpf_helpers.hdoesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.
Right, so apps tend to ship their own uapi bpf.h header and helper signatures to avoid these issues. But question becomes once they start using soley bpf_helper.h (also in non-tracing context which is very reasonable to assume), then things might break with the patch as-is once they have a newer libbpf with more signatures than their linux/bpf.h defines (and yes, pulling from GH will have this problem), so we'd need to have an answer to that in order to avoid breaking compilation.
[...]
Quoted 18 lines
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those
BPF_FUNC_constants. Super painful and not nice, but will work.b. copy/paste enum
bpf_func_iddefinition intobpf_helpers.hitself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict withvmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).c. BPF UAPI header has
__BPF_FUNC_MAPPERmacro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/vmlinux.hthere is no such macro, so this should work as well.
Quoted 8 lines
I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Opinions?
Was thinking about something like c) as well. So I tried to do a quick hack. Here is how it could work, but it needs a small change in the __BPF_FUNC_MAPPER(), at least I didn't find an immediate way around it:
static void (*__unspec)(void);
static void *(*__map_lookup_elem)(void *map, const void *key);
static int (*__map_update_elem)(void *map, const void *key, const void *value, unsigned long long flags); static int (*__map_delete_elem)(void *map, const void *key);
static int (*__bpf_probe_read)(void *dst, int size, const void *unsafe_ptr);
#define __BPF_FUNC_MAPPER(FN, DELIM) \
FN(unspec) DELIM \
FN(map_lookup_elem) DELIM \
FN(map_update_elem) DELIM \ FN(map_delete_elem) DELIM
#define __BPF_ENUM_FN(x) BPF_FUNC_ ## x
enum bpf_func_id {
#define COM ,
__BPF_FUNC_MAPPER(__BPF_ENUM_FN, COM)
__BPF_FUNC_MAX_ID,
};#undef __BPF_ENUM_FN
#define __BPF_ASSIGN_FN(x) typeof(__ ## x) x = (void *)BPF_FUNC_ ## x #define SEM ;__BPF_FUNC_MAPPER(__BPF_ASSIGN_FN, SEM)
#undef __BPF_ASSIGN_FN
int main(void)
{
return(unsigned long)map_lookup_elem(0, 0);
}It does seem to eat it:
$ gcc foo.c $
In short, libbpf's bpf_helper.h would define signatures (or better: types) as above prefixed with __. The __BPF_FUNC_MAPPER() and the enum bpf_func_id {} is the one in the uapi header, so out of our control from libbpf pov. But aside from the signatures, the bpf_helper.h would have __BPF_ASSIGN_FN() and this would avoid the breakage issue. The only thing libbpf needs to have in the bpf_helper.h are the /latest/ signatures with /matching/ names.
Right now in the __BPF_FUNC_MAPPER() there is this comma, we probably need to make this a parameter, at least from a quick glance I didn't see a way to work around it.
Cheers,
Daniel
Daniel Borkmann
maintainerRe: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 159 lines · Daniel Borkmann
Quoted 67 lines · Andrii Nakryiko
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
Quoted 7 lines
1. About "adding
bpf_helpers.hwill make things worse". I categorically disagree,bpf_helpers.hdoesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.Right, so apps tend to ship their own uapi
bpf.hheader and helper signatures to avoid these issues. But question becomes once they start using soleybpf_helper.h(also in non-tracing context which is very reasonable to assume), then things might break with the patch as-is once they have a newer libbpf with more signatures than theirlinux/bpf.hdefines (and yes, pulling from GH will have this problem), so we'd need to have an answer to that in order to avoid breaking compilation.[...]
Quoted 18 lines
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those
BPF_FUNC_constants. Super painful and not nice, but will work.b. copy/paste enum
bpf_func_iddefinition intobpf_helpers.hitself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict withvmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).c. BPF UAPI header has
__BPF_FUNC_MAPPERmacro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/vmlinux.hthere is no such macro, so this should work as well.Quoted 8 lines
I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Opinions?
Was thinking about something like c) as well. So I tried to do a quick hack. Here is how it could work, but it needs a small change in the
__BPF_FUNC_MAPPER(), at least I didn't find an immediate way around it:static void (*__unspec)(void);
static void *(*__map_lookup_elem)(void *map, const void *key);
static int (*__map_update_elem)(void *map, const void *key, const void *value, unsigned long long flags); static int (*__map_delete_elem)(void *map, const void *key);
static int (*__bpf_probe_read)(void *dst, int size, const void*unsafe_ptr);#define __BPF_FUNC_MAPPER(FN, DELIM) \ FN(unspec) DELIM \ FN(map_lookup_elem) DELIM \ FN(map_update_elem) DELIM \
FN(map_delete_elem)DELIM#define
__BPF_ENUM_FN(x)BPF_FUNC_## x
enumbpf_func_id{#define COM , __BPF_FUNC_MAPPER(__BPF_ENUM_FN, COM) __BPF_FUNC_MAX_ID, };#undef
__BPF_ENUM_FN#define
__BPF_ASSIGN_FN(x)typeof(__ ## x)x = (void *)BPF_FUNC_ ## x #define SEM ;__BPF_FUNC_MAPPER(__BPF_ASSIGN_FN, SEM)
#undef__BPF_ASSIGN_FNint main(void)
{ return(unsigned long)map_lookup_elem(0, 0); }It does seem to eat it:
$ gcc foo.c $In short, libbpf's
bpf_helper.hwould define signatures (or better: types) as above prefixed with __. The__BPF_FUNC_MAPPER()and the enumbpf_func_id{} is the one in the uapi header, so out of our control from libbpf pov. But aside from the signatures, thebpf_helper.hwould have__BPF_ASSIGN_FN()and this would avoid the breakage issue. The only thing libbpf needs to have in thebpf_helper.hare the /latest/ signatures with /matching/ names.
Quoted 3 lines
Right now in the
__BPF_FUNC_MAPPER()there is this comma, we probably need to make this a parameter, at least from a quick glance I didn't see a way to work around it.
Small kernel side change would look like below, seems useful regardless, imho. Perhaps that approach where you'd only need signatures like above would also resolve the 'concern' from gcc folks where they wanted a special helper attribute thingy to achieve the same.
From e89cd0d873ae07e82526ab573746212a5e1b69cb Mon Sep 17 00:00:00 2001
Message-Id: <e89cd0d873ae07e82526ab573746212a5e1b69cb.1570225802.git.daniel@iogearbox.net>
From: Daniel Borkmann <[email protected]>
Date: Fri, 4 Oct 2019 23:46:25 +0200
Subject: [PATCH] bpf: Allow __BPF_FUNC_MAPPER to be more customizable
--- include/uapi/linux/bpf.h | 228 ++++++++++++++++++++------------------- 1 file changed, 116 insertions(+), 112 deletions(-)
include/uapi/linux/bpf.h +116 -112
@@ -2751,118 +2751,122 @@ union bpf_attr {* * **-EPROTONOSUPPORT** IP packet version is not 4 or 6 */ -#define __BPF_FUNC_MAPPER (FN ) \ - FN(unspec), \ - FN(map_lookup_elem), \ - FN(map_update_elem), \ - FN(map_delete_elem), \ - FN(probe_read), \ - FN(ktime_get_ns), \ - FN(trace_printk), \ - FN(get_prandom_u32), \ - FN(get_smp_processor_id), \ - FN(skb_store_bytes), \ - FN(l3_csum_replace), \ - FN(l4_csum_replace), \ - FN(tail_call), \ - FN(clone_redirect), \ - FN(get_current_pid_tgid), \ - FN(get_current_uid_gid), \ - FN(get_current_comm), \ - FN(get_cgroup_classid), \ - FN(skb_vlan_push), \ - FN(skb_vlan_pop), \ - FN(skb_get_tunnel_key), \ - FN(skb_set_tunnel_key), \ - FN(perf_event_read), \ - FN(redirect), \ - FN(get_route_realm), \ - FN(perf_event_output), \ - FN(skb_load_bytes), \ - FN(get_stackid), \ - FN(csum_diff), \ - FN(skb_get_tunnel_opt), \ - FN(skb_set_tunnel_opt), \ - FN(skb_change_proto), \ - FN(skb_change_type), \ - FN(skb_under_cgroup), \ - FN(get_hash_recalc), \ - FN(get_current_task), \ - FN(probe_write_user), \ - FN(current_task_under_cgroup), \ - FN(skb_change_tail), \ - FN(skb_pull_data), \ - FN(csum_update), \ - FN(set_hash_invalid), \ - FN(get_numa_node_id), \ - FN(skb_change_head), \ - FN(xdp_adjust_head), \ - FN(probe_read_str), \ - FN(get_socket_cookie), \ - FN(get_socket_uid), \ - FN(set_hash), \ - FN(setsockopt), \ - FN(skb_adjust_room), \ - FN(redirect_map), \ - FN(sk_redirect_map), \ - FN(sock_map_update), \ - FN(xdp_adjust_meta), \ - FN(perf_event_read_value), \ - FN(perf_prog_read_value), \ - FN(getsockopt), \ - FN(override_return), \ - FN(sock_ops_cb_flags_set), \ - FN(msg_redirect_map), \ - FN(msg_apply_bytes), \ - FN(msg_cork_bytes), \ - FN(msg_pull_data), \ - FN(bind), \ - FN(xdp_adjust_tail), \ - FN(skb_get_xfrm_state), \ - FN(get_stack), \ - FN(skb_load_bytes_relative), \ - FN(fib_lookup), \ - FN(sock_hash_update), \ - FN(msg_redirect_hash), \ - FN(sk_redirect_hash), \ - FN(lwt_push_encap), \ - FN(lwt_seg6_store_bytes), \ - FN(lwt_seg6_adjust_srh), \ - FN(lwt_seg6_action), \ - FN(rc_repeat), \ - FN(rc_keydown), \ - FN(skb_cgroup_id), \ - FN(get_current_cgroup_id), \ - FN(get_local_storage), \ - FN(sk_select_reuseport), \ - FN(skb_ancestor_cgroup_id), \ - FN(sk_lookup_tcp), \ - FN(sk_lookup_udp), \ - FN(sk_release), \ - FN(map_push_elem), \ - FN(map_pop_elem), \ - FN(map_peek_elem), \ - FN(msg_push_data), \ - FN(msg_pop_data), \ - FN(rc_pointer_rel), \ - FN(spin_lock), \ - FN(spin_unlock), \ - FN(sk_fullsock), \ - FN(tcp_sock), \ - FN(skb_ecn_set_ce), \ - FN(get_listener_sock), \ - FN(skc_lookup_tcp), \ - FN(tcp_check_syncookie), \ - FN(sysctl_get_name), \ - FN(sysctl_get_current_value), \ - FN(sysctl_get_new_value), \ - FN(sysctl_set_new_value), \ - FN(strtol), \ - FN(strtoul), \ - FN(sk_storage_get), \ - FN(sk_storage_delete), \ - FN(send_signal), \ - FN(tcp_gen_syncookie), +#define ____BPF_FUNC_MAPPER (FN ,DL ) \ + FN(unspec) DL \ + FN(map_lookup_elem) DL \ + FN(map_update_elem) DL \ + FN(map_delete_elem) DL \ + FN(probe_read) DL \ + FN(ktime_get_ns) DL \ + FN(trace_printk) DL \ + FN(get_prandom_u32) DL \ + FN(get_smp_processor_id) DL \ + FN(skb_store_bytes) DL \ + FN(l3_csum_replace) DL \ + FN(l4_csum_replace) DL \ + FN(tail_call) DL \ + FN(clone_redirect) DL \ + FN(get_current_pid_tgid) DL \ + FN(get_current_uid_gid) DL \ + FN(get_current_comm) DL \ + FN(get_cgroup_classid) DL \ + FN(skb_vlan_push) DL \ + FN(skb_vlan_pop) DL \ + FN(skb_get_tunnel_key) DL \ + FN(skb_set_tunnel_key) DL \ + FN(perf_event_read) DL \ + FN(redirect) DL \ + FN(get_route_realm) DL \ + FN(perf_event_output) DL \ + FN(skb_load_bytes) DL \ + FN(get_stackid) DL \ + FN(csum_diff) DL \ + FN(skb_get_tunnel_opt) DL \ + FN(skb_set_tunnel_opt) DL \ + FN(skb_change_proto) DL \ + FN(skb_change_type) DL \ + FN(skb_under_cgroup) DL \ + FN(get_hash_recalc) DL \ + FN(get_current_task) DL \ + FN(probe_write_user) DL \ + FN(current_task_under_cgroup) DL\ + FN(skb_change_tail) DL \ + FN(skb_pull_data) DL \ + FN(csum_update) DL \ + FN(set_hash_invalid) DL \ + FN(get_numa_node_id) DL \ + FN(skb_change_head) DL \ + FN(xdp_adjust_head) DL \ + FN(probe_read_str) DL \ + FN(get_socket_cookie) DL \ + FN(get_socket_uid) DL \ + FN(set_hash) DL \ + FN(setsockopt) DL \ + FN(skb_adjust_room) DL \ + FN(redirect_map) DL \ + FN(sk_redirect_map) DL \ + FN(sock_map_update) DL \ + FN(xdp_adjust_meta) DL \ + FN(perf_event_read_value) DL \ + FN(perf_prog_read_value) DL \ + FN(getsockopt) DL \ + FN(override_return) DL \ + FN(sock_ops_cb_flags_set) DL \ + FN(msg_redirect_map) DL \ + FN(msg_apply_bytes) DL \ + FN(msg_cork_bytes) DL \ + FN(msg_pull_data) DL \ + FN(bind) DL \ + FN(xdp_adjust_tail) DL \ + FN(skb_get_xfrm_state) DL \ + FN(get_stack) DL \ + FN(skb_load_bytes_relative) DL \ + FN(fib_lookup) DL \ + FN(sock_hash_update) DL \ + FN(msg_redirect_hash) DL \ + FN(sk_redirect_hash) DL \ + FN(lwt_push_encap) DL \ + FN(lwt_seg6_store_bytes) DL \ + FN(lwt_seg6_adjust_srh) DL \ + FN(lwt_seg6_action) DL \ + FN(rc_repeat) DL \ + FN(rc_keydown) DL \ + FN(skb_cgroup_id) DL \ + FN(get_current_cgroup_id) DL \ + FN(get_local_storage) DL \ + FN(sk_select_reuseport) DL \ + FN(skb_ancestor_cgroup_id) DL \ + FN(sk_lookup_tcp) DL \ + FN(sk_lookup_udp) DL \ + FN(sk_release) DL \ + FN(map_push_elem) DL \ + FN(map_pop_elem) DL \ + FN(map_peek_elem) DL \ + FN(msg_push_data) DL \ + FN(msg_pop_data) DL \ + FN(rc_pointer_rel) DL \ + FN(spin_lock) DL \ + FN(spin_unlock) DL \ + FN(sk_fullsock) DL \ + FN(tcp_sock) DL \ + FN(skb_ecn_set_ce) DL \ + FN(get_listener_sock) DL \ + FN(skc_lookup_tcp) DL \ + FN(tcp_check_syncookie) DL \ + FN(sysctl_get_name) DL \ + FN(sysctl_get_current_value) DL \ + FN(sysctl_get_new_value) DL \ + FN(sysctl_set_new_value) DL \ + FN(strtol) DL \ + FN(strtoul) DL \ + FN(sk_storage_get) DL \ + FN(sk_storage_delete) DL \ + FN(send_signal) DL \ + FN(tcp_gen_syncookie) DL + +#define ____BPF_DELIM_COMMA , +#define __BPF_FUNC_MAPPER (FN ) \ + ____BPF_FUNC_MAPPER(FN, ____BPF_DELIM_COMMA)/* integer value in 'imm' field of BPF_CALL instruction selects which helper * function eBPF program intends to call
Signature · 1 line
2.21.0
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 113 lines · Daniel Borkmann
Quoted 111 lines · Daniel Borkmann
Quoted 67 lines · Andrii Nakryiko
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
Quoted 7 lines
1. About "adding
bpf_helpers.hwill make things worse". I categorically disagree,bpf_helpers.hdoesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.Right, so apps tend to ship their own uapi
bpf.hheader and helper signatures to avoid these issues. But question becomes once they start using soleybpf_helper.h(also in non-tracing context which is very reasonable to assume), then things might break with the patch as-is once they have a newer libbpf with more signatures than theirlinux/bpf.hdefines (and yes, pulling from GH will have this problem), so we'd need to have an answer to that in order to avoid breaking compilation.[...]
Quoted 25 lines
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those
BPF_FUNC_constants. Super painful and not nice, but will work.b. copy/paste enum
bpf_func_iddefinition intobpf_helpers.hitself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict withvmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).c. BPF UAPI header has
__BPF_FUNC_MAPPERmacro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/vmlinux.hthere is no such macro, so this should work as well.I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
Opinions?
Was thinking about something like c) as well. So I tried to do a quick hack. Here is how it could work, but it needs a small change in the
__BPF_FUNC_MAPPER(), at least I didn't find an immediate way around it:
Well, we are stuck with this comma, so rather than have to support two bpf.h headers, I'd solve the problem for existing one. It's annoying, but you can do it with having "/* <your macro> /*" in each FN macro and then before you apply everything you add /* and after all the applications you add */.
I'm going to prototype something like what you have below, but will see we I can minimize amount of extra declarations we need. Do you think this needs to be done as part of this patch set, or I can defer that to a follow up patch?
Quoted 7 lines
Quoted 7 lines
static void (*__unspec)(void);
static void *(*__map_lookup_elem)(void *map, const void *key);
static int (*__map_update_elem)(void *map, const void *key, const void *value, unsigned long long flags); static int (*__map_delete_elem)(void *map, const void *key);
static int (*__bpf_probe_read)(void *dst, int size, const void*unsafe_ptr);
[...]
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 117 lines · Andrii Nakryiko
Quoted 115 lines · Daniel Borkmann
Quoted 113 lines · Daniel Borkmann
Quoted 67 lines · Andrii Nakryiko
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
Quoted 7 lines
1. About "adding
bpf_helpers.hwill make things worse". I categorically disagree,bpf_helpers.hdoesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.Right, so apps tend to ship their own uapi
bpf.hheader and helper signatures to avoid these issues. But question becomes once they start using soleybpf_helper.h(also in non-tracing context which is very reasonable to assume), then things might break with the patch as-is once they have a newer libbpf with more signatures than theirlinux/bpf.hdefines (and yes, pulling from GH will have this problem), so we'd need to have an answer to that in order to avoid breaking compilation.[...]
Quoted 26 lines
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those
BPF_FUNC_constants. Super painful and not nice, but will work.b. copy/paste enum
bpf_func_iddefinition intobpf_helpers.hitself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict withvmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).c. BPF UAPI header has
__BPF_FUNC_MAPPERmacro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/vmlinux.hthere is no such macro, so this should work as well.I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Opinions?
Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
Was thinking about something like c) as well. So I tried to do a quick hack. Here is how it could work, but it needs a small change in the
__BPF_FUNC_MAPPER(), at least I didn't find an immediate way around it:Well, we are stuck with this comma, so rather than have to support two
bpf.hheaders, I'd solve the problem for existing one. It's annoying, but you can do it with having "/* <your macro> /*" in each FN macro
err, was supposed to be "*/ (close previous comment) <your macro> /* (open next comment".
Quoted 17 lines
and then before you apply everything you add /* and after all the applications you add */.
I'm going to prototype something like what you have below, but will see we I can minimize amount of extra declarations we need. Do you think this needs to be done as part of this patch set, or I can defer that to a follow up patch?
Quoted 7 lines
Quoted 7 lines
static void (*__unspec)(void);
static void *(*__map_lookup_elem)(void *map, const void *key);
static int (*__map_update_elem)(void *map, const void *key, const void *value, unsigned long long flags); static int (*__map_delete_elem)(void *map, const void *key);
static int (*__bpf_probe_read)(void *dst, int size, const void*unsafe_ptr);[...]
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
Quoted 124 lines · Andrii Nakryiko
Quoted 121 lines · Andrii Nakryiko
Quoted 118 lines · Daniel Borkmann
Quoted 116 lines · Daniel Borkmann
Quoted 67 lines · Andrii Nakryiko
Quoted 57 lines · Jakub Kicinski
Quoted 23 lines · Andrii Nakryiko
Quoted 5 lines · David Ahern
Quoted 1 line
I'm not following you; my interpretation of your comment seems like you
are making huge assumptions.
I build bpf programs for specific kernel versions using the devel packages for the specific kernel of interest.
Sure, and you can keep doing that, just don't include
bpf_helpers.h?What I was saying, though, especially having in mind tracing BPF programs that need to inspect kernel structures, is that it's quite impractical to have to build many different versions of BPF programs for each supported kernel version and distribute them in binary form. So people usually use BCC and do compilation on-the-fly using BCC's embedded Clang.
BPF CO-RE is providing an alternative, which will allow to pre-compile your program once for many different kernels you might be running your program on. There is tooling that eliminates the need for system headers. Instead we pre-generate a single
vmlinux.hheader with all the types/enums/etc, that are then used w/ BPF CO-RE to build portable BPF programs capable of working on multiple kernel versions.Quoted 4 lines
So what I was pointing out there was that this
vmlinux.hwould be ideally generated from latest kernel and not having latestBPF_FUNC_xxxshouldn't be a problem. But see below about situation being worse.Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
Quoted 11 lines
Quoted 10 lines
Quoted 6 lines
Nevertheless, it is a problem and thanks for bringing it up! I'd say for now we should still go ahead with this move and try to solve with issue once
bpf_helpers.his in libbpf. Ifbpf_helpers.hdoesn't work for someone, it's no worse than it is today when users don't havebpf_helpers.hat all.If this syncs to the github libbpf, it will be worse than today in the sense of compile failures if someone's header file ordering picks libbpf's
bpf_helpers.hover whatever they are using today.Quoted 7 lines
Today
bpf_helpers.hdon't exist for users or am I missing something?bpf_helpers.hright now are purely for selftests. But they are really useful outside that context, so I'm making it available for everyone by distributing with libbpf sources. Ifbpf_helpers.hdoesn't work for some specific use case, just don't use it (yet?).I'm still failing to see how it's worse than situation today.
Having a header which works today, but may not work tomorrow is going to be pretty bad user experience :( No matter how many warnings you put in the source people will get caught off guard by this :(
Quoted 5 lines
If you define the current state as "users can use all features of libbpf and nothing should break on libbpf update" (which is in my understanding a goal of the project, we bent over backwards trying to not break things) then adding this header will in fact make things worse. The statement in quotes would no longer be true, no?
So there are few things here.
Quoted 7 lines
1. About "adding
bpf_helpers.hwill make things worse". I categorically disagree,bpf_helpers.hdoesn't exist in user land at all and it's sorely missing. So adding it is strictly better experience already. Right now people have to re-declare those helper signatures and do all kinds of unnecessary hackery just to be able to use BPF stuff, and they still can run into the same problem with having too old kernel headers.Right, so apps tend to ship their own uapi
bpf.hheader and helper signatures to avoid these issues. But question becomes once they start using soleybpf_helper.h(also in non-tracing context which is very reasonable to assume), then things might break with the patch as-is once they have a newer libbpf with more signatures than theirlinux/bpf.hdefines (and yes, pulling from GH will have this problem), so we'd need to have an answer to that in order to avoid breaking compilation.[...]
Quoted 18 lines
2. As to the problem of running bleeding-edge libbpf against older kernel. There are few possible solutions:
a. we hard-code all those
BPF_FUNC_constants. Super painful and not nice, but will work.b. copy/paste enum
bpf_func_iddefinition intobpf_helpers.hitself and try to keep it in sync with UAPI. Apart from obvious redundancy that involves, we also will need to make sure this doesn't conflict withvmlinux.h, so enum name should be different and each value should be different (which means some wort of additional prefix or suffix).c. BPF UAPI header has
__BPF_FUNC_MAPPERmacro "iterating" over all defined values for a particular kernel version. We can use that and additional macro trickery to conditionally define helpers. Again, we need to keep in mind that w/vmlinux.hthere is no such macro, so this should work as well.Quoted 8 lines
I'm happy to hear opinions about these choices (maybe there are some other I missed), but in any case I'd like to do it in a follow up patch and land this one as is. It has already quite a lot packed in it. I personally lean towards c) as it will have a benefit of not declaring helpers that are not supported by kernel we are compiling against, even though it requires additional macro trickery.
Opinions?
Was thinking about something like c) as well. So I tried to do a quick hack. Here is how it could work, but it needs a small change in the
__BPF_FUNC_MAPPER(), at least I didn't find an immediate way around it:
Quoted 6 lines
Quoted 3 lines
Well, we are stuck with this comma, so rather than have to support two
bpf.hheaders, I'd solve the problem for existing one. It's annoying, but you can do it with having "/* <your macro> /*" in each FN macroerr, was supposed to be "*/ (close previous comment) <your macro> /* (open next comment".
That won't work because macro can't emit comments... There must be some other creative way to "screen" that comma out, but I'm too tired to figure this out at the moment.
But more importantly, if we go this way, we'll have to work with bpf.h headers from existing kernels, without Daniel's patch, which makes Daniel's patch useless for this.
Another alternative that came up in offline discussions was to just auto-generate these helper definitions from bpf.h itself as part of libbpf build. We can also preserve documentation comments along the way. I want to try that as well and bypass this whole macro ugliness. If we don't manage to do this in time for next release, we can always temporarily hard-code all the constants and not depend on bpf.h. How's that for a plan?
Quoted 19 lines
Quoted 19 lines
and then before you apply everything you add /* and after all the applications you add */.
I'm going to prototype something like what you have below, but will see we I can minimize amount of extra declarations we need. Do you think this needs to be done as part of this patch set, or I can defer that to a follow up patch?
Quoted 7 lines
Quoted 7 lines
static void (*__unspec)(void);
static void *(*__map_lookup_elem)(void *map, const void *key);
static int (*__map_update_elem)(void *map, const void *key, const void *value, unsigned long long flags); static int (*__map_delete_elem)(void *map, const void *key);
static int (*__bpf_probe_read)(void *dst, int size, const void*unsafe_ptr);[...]
Re: [PATCH v3 bpf-next 5/7] libbpf: move bpf_{helpers,endian,tracing}.h into libbpf
In reply to Jakub KicinskiQuoted 4 lines · Jakub Kicinski
Surely for distroes tho - they would have kernel headers matching the kernel release they ship. If parts of libbpf from GH only work with the latest kernel, distroes should ship libbpf from the kernel source, rather than GH.
I don't see a problem here for distros. Distros control both the kernel and libbpf, there's no problem in keeping those in sync.
Packaging libbpf from the kernel source would not help anything - updating libbpf would still require a new kernel. There's no difference between updating libbpf from a github repo or from the kernel source, as far as dependencies are concerned.
Jiri
PATCH v3 6/7 libbpf: add BPF_CORE_READ/BPF_CORE_READ_INTO helpers
In reply to Andrii NakryikoAdd few macros simplifying BCC-like multi-level probe reads, while also emitting CO-RE relocations for each read.
--- tools/lib/bpf/bpf_helpers.h | 153 +++++++++++++++++++++++++++++++++++- 1 file changed, 150 insertions(+), 3 deletions(-)
tools/lib/bpf/bpf_helpers.h +150 -3
@@ -5,7 +5,7 @@#define __uint (name, val) int (*name)[val]#define __type (name, val) val *name -/* helper macro to print out debug messages */ +/* Helper macro to print out debug messages */ #define bpf_printk (fmt, ...) \ ({ \ char ____fmt[] = fmt; \
@@ -13,12 +13,17 @@ ##__VA_ARGS__); \ }) -/* helper macro to place programs, maps, license in +/* +* Helper macro to place programs, maps, license in * different sections in elf_bpf file. Section names * are interpreted by elf_bpf loader */ #define SEC (NAME ) __attribute__((section(NAME), used)) +#ifndef __always_inline +#define __always_inline __attribute__((always_inline)) +#endif +/* helper functions called from eBPF programs written in C */ static void *(*bpf_map_lookup_elem)(void *map,const void *key) = (void *) BPF_FUNC_map_lookup_elem;
@@ -281,7 +286,8 @@ static int (*bpf_skb_adjust_room)(void *ctx, __s32 len_diff, __u32 mode, unsigned long long flags) = (void *) BPF_FUNC_skb_adjust_room; -/* a helper structure used by eBPF C program +/* +* Helper structure used by eBPF C program * to describe BPF map attributes to libbpf loader */ struct bpf_map_def {
@@ -312,4 +318,145 @@ struct bpf_map_def {bpf_probe_read (dst, sz, \ (const void *)__builtin_preserve_access_index (src)) +/* +* bpf_core_read_str() is a thin wrapper around bpf_probe_read_str() +* additionally emitting BPF CO-RE field relocation for specified source +* argument. + */ +#define bpf_core_read_str (dst, sz, src) \ + bpf_probe_read_str(dst, sz, \ + (const void *)__builtin_preserve_access_index(src)) + +#define ___concat (a, b) a ## b +#define ___apply (fn, n) ___concat(fn, n) +#define ___nth (_1, _2, _3, _4, _5, _6, _7, _8, _9, _10, __11,N , ...) N + +/* +* return number of provided arguments; used for switch-based variadic macro +* definitions (see ___last, ___arrow, etc below) +*/ +#define ___narg (...) ___nth(_, ##__VA_ARGS__, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0) +/* +* return 0 if no arguments are passed, N - otherwise; used for +* recursively-defined macros to specify termination (0) case, and generic +* (N) case (e.g., ___read_ptrs, ___core_read) +*/ +#define ___empty (...) ___nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0) + +#define ___last1 (x) x +#define ___last2 (a, x) x +#define ___last3 (a, b, x) x +#define ___last4 (a, b, c, x) x +#define ___last5 (a, b, c, d, x) x +#define ___last6 (a, b, c, d, e, x) x +#define ___last7 (a, b, c, d, e, f, x) x +#define ___last8 (a, b, c, d, e, f, g, x) x +#define ___last9 (a, b, c, d, e, f, g, h, x) x +#define ___last10 (a, b, c, d, e, f, g, h, i, x) x +#define ___last (...) ___apply(___last, ___narg(__VA_ARGS__))(__VA_ARGS__) + +#define ___nolast2 (a, _) a +#define ___nolast3 (a, b, _) a, b +#define ___nolast4 (a, b, c, _) a, b, c +#define ___nolast5 (a, b, c, d, _) a, b, c, d +#define ___nolast6 (a, b, c, d, e, _) a, b, c, d, e +#define ___nolast7 (a, b, c, d, e, f, _) a, b, c, d, e, f +#define ___nolast8 (a, b, c, d, e, f, g, _) a, b, c, d, e, f, g +#define ___nolast9 (a, b, c, d, e, f, g, h, _) a, b, c, d, e, f, g, h +#define ___nolast10 (a, b, c, d, e, f, g, h, i, _) a, b, c, d, e, f, g, h, i +#define ___nolast (...) ___apply(___nolast, ___narg(__VA_ARGS__))(__VA_ARGS__) + +#define ___arrow1 (a) a +#define ___arrow2 (a, b) a->b +#define ___arrow3 (a, b, c) a->b->c +#define ___arrow4 (a, b, c, d) a->b->c->d +#define ___arrow5 (a, b, c, d, e) a->b->c->d->e +#define ___arrow6 (a, b, c, d, e, f) a->b->c->d->e->f +#define ___arrow7 (a, b, c, d, e, f, g) a->b->c->d->e->f->g +#define ___arrow8 (a, b, c, d, e, f, g, h) a->b->c->d->e->f->g->h +#define ___arrow9 (a, b, c, d, e, f, g, h, i) a->b->c->d->e->f->g->h->i +#define ___arrow10 (a, b, c, d, e, f, g, h, i, j) a->b->c->d->e->f->g->h->i->j +#define ___arrow (...) ___apply(___arrow, ___narg(__VA_ARGS__))(__VA_ARGS__) + +#define ___type (...) typeof(___arrow(__VA_ARGS__)) + +#define ___read (read_fn, dst, src_type, src, accessor) \ + read_fn((void *)(dst), sizeof(*(dst)), &((src_type)(src))->accessor) + +/* "recursively" read a sequence of inner pointers using local __t var */ +#define ___rd_last (...) \ + ___read(bpf_core_read, &__t, \ + ___type(___nolast(__VA_ARGS__)), __t, ___last(__VA_ARGS__)); +#define ___rd_p0 (src) const void *__t = src; +#define ___rd_p1 (...) ___rd_p0(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p2 (...) ___rd_p1(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p3 (...) ___rd_p2(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p4 (...) ___rd_p3(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p5 (...) ___rd_p4(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p6 (...) ___rd_p5(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p7 (...) ___rd_p6(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p8 (...) ___rd_p7(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___rd_p9 (...) ___rd_p8(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__) +#define ___read_ptrs (src, ...) \ + ___apply(___rd_p, ___narg(__VA_ARGS__))(src, __VA_ARGS__) + +#define ___core_read0 (fn, dst, src, a) \ + ___read(fn, dst, ___type(src), src, a); +#define ___core_readN (fn, dst, src, ...) \ + ___read_ptrs(src, ___nolast(__VA_ARGS__)) \ + ___read(fn, dst, ___type(src, ___nolast(__VA_ARGS__)), __t, \ + ___last(__VA_ARGS__)); +#define ___core_read (fn, dst, src, a, ...) \ + ___apply(___core_read, ___empty(__VA_ARGS__))(fn, dst, \ + src, a, ##__VA_ARGS__) + +/* +* BPF_CORE_READ_INTO() is a more performance-conscious variant of +* BPF_CORE_READ(), in which final field is read into user-provided storage. +* See BPF_CORE_READ() below for more details on general usage. +*/ +#define BPF_CORE_READ_INTO (dst, src, a, ...) \ + ({ \ + ___core_read(bpf_core_read, dst, src, a, ##__VA_ARGS__) \ + }) + +/* +* BPF_CORE_READ_STR_INTO() does same "pointer chasing" as +* BPF_CORE_READ() for intermediate pointers, but then executes (and returns +* corresponding error code) bpf_core_read_str() for final string read. +*/ +#define BPF_CORE_READ_STR_INTO (dst, src, a, ...) \ + ({ \ + ___core_read(bpf_core_read_str, dst, src, a, ##__VA_ARGS__) \ + }) + +/* +* BPF_CORE_READ() is used to simplify BPF CO-RE relocatable read, especially +* when there are few pointer chasing steps. +* E.g., what in non-BPF world (or in BPF w/ BCC) would be something like: +* int x = s->a.b.c->d.e->f->g; +* can be succinctly achieved using BPF_CORE_READ as: +* int x = BPF_CORE_READ(s, a.b.c, d.e, f, g); +* +* BPF_CORE_READ will decompose above statement into 4 bpf_core_read (BPF +* CO-RE relocatable bpf_probe_read() wrapper) calls, logically equivalent to: +* 1. const void *__t = s->a.b.c; +* 2. __t = __t->d.e; +* 3. __t = __t->f; +* 4. return __t->g; +* +* Equivalence is logical, because there is a heavy type casting/preservation +* involved, as well as all the reads are happening through bpf_probe_read() +* calls using __builtin_preserve_access_index() to emit CO-RE relocations. +* +* N.B. Only up to 9 "field accessors" are supported, which should be more +* than enough for any practical purpose. +*/ +#define BPF_CORE_READ (src, a, ...) \ + ({ \ + ___type(src, a, ##__VA_ARGS__) __r; \ + BPF_CORE_READ_INTO(&__r, src, a, ##__VA_ARGS__); \ + __r; \ + }) +#endif
Signature · 1 line
2.17.1
PATCH v3 7/7 selftests/bpf: add BPF_CORE_READ and BPF_CORE_READ_STR_INTO macro tests
In reply to Andrii NakryikoValidate BPF_CORE_READ correctness and handling of up to 9 levels of nestedness using cyclic task->(group_leader->)*->tgid chains.
Also add a test of maximum-dpeth BPF_CORE_READ_STR_INTO() macro.
--- .../selftests/bpf/prog_tests/core_reloc.c | 8 ++- .../selftests/bpf/progs/core_reloc_types.h | 9 ++++ .../bpf/progs/test_core_reloc_kernel.c | 54 ++++++++++++++++++- 3 files changed, 68 insertions(+), 3 deletions(-)
tools/testing/selftests/bpf/prog_tests/core_reloc.c +6 -2
@@ -193,8 +193,12 @@ static struct core_reloc_test_case test_cases[] = { .btf_src_file = NULL, /* load from /lib/modules/$(uname -r) */ .input = "", .input_len = 0, - .output = "\1", /* true */ - .output_len = 1, + .output = STRUCT_TO_CHAR_PTR(core_reloc_kernel_output) { + .valid = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, }, + .comm ="test_progs\0\0\0\0\0" , + .comm_len = 11, + }, + .output_len = sizeof(struct core_reloc_kernel_output), },/* validate BPF program can use multiple flavors to match against
tools/testing/selftests/bpf/progs/core_reloc_types.h +9 -0
@@ -1,5 +1,14 @@#include <stdint.h> #include <stdbool.h> +/* +* KERNEL +*/ + +struct core_reloc_kernel_output { +int valid[10 ]; +char comm[16 ]; +int comm_len; +};/* * FLAVORS
tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c +53 -1
76 lines not on this page · raw message
Signature · 1 line
2.17.1