From c39028b333f3a3a765c5c0b9726b8e38aedf0ba1 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 27 Apr 2023 18:36:38 -0700 Subject: libbpf: btf_dump_type_data_check_overflow needs to consider BTF_MEMBER_BITFIELD_SIZE The btf_dump/struct_data selftest is failing with: [...] test_btf_dump_struct_data:FAIL:unexpected return value dumping fs_context unexpected unexpected return value dumping fs_context: actual -7 != expected 264 [...] The reason is in btf_dump_type_data_check_overflow(). It does not use BTF_MEMBER_BITFIELD_SIZE from the struct's member (btf_member). Instead, it is using the enum size which is 4. It had been working till the recent commit 4e04143c869c ("fs_context: drop the unused lsm_flags member") removed an integer member which also removed the 4 bytes padding at the end of the fs_context. Missing this 4 bytes padding exposed this bug. In particular, when btf_dump_type_data_check_overflow() reaches the member 'phase', -E2BIG is returned. The fix is to pass bit_sz to btf_dump_type_data_check_overflow(). In btf_dump_type_data_check_overflow(), it does a different size check when bit_sz is not zero. The current fs_context: [3600] ENUM 'fs_context_purpose' encoding=UNSIGNED size=4 vlen=3 'FS_CONTEXT_FOR_MOUNT' val=0 'FS_CONTEXT_FOR_SUBMOUNT' val=1 'FS_CONTEXT_FOR_RECONFIGURE' val=2 [3601] ENUM 'fs_context_phase' encoding=UNSIGNED size=4 vlen=7 'FS_CONTEXT_CREATE_PARAMS' val=0 'FS_CONTEXT_CREATING' val=1 'FS_CONTEXT_AWAITING_MOUNT' val=2 'FS_CONTEXT_AWAITING_RECONF' val=3 'FS_CONTEXT_RECONF_PARAMS' val=4 'FS_CONTEXT_RECONFIGURING' val=5 'FS_CONTEXT_FAILED' val=6 [3602] STRUCT 'fs_context' size=264 vlen=21 'ops' type_id=3603 bits_offset=0 'uapi_mutex' type_id=235 bits_offset=64 'fs_type' type_id=872 bits_offset=1216 'fs_private' type_id=21 bits_offset=1280 'sget_key' type_id=21 bits_offset=1344 'root' type_id=781 bits_offset=1408 'user_ns' type_id=251 bits_offset=1472 'net_ns' type_id=984 bits_offset=1536 'cred' type_id=1785 bits_offset=1600 'log' type_id=3621 bits_offset=1664 'source' type_id=42 bits_offset=1792 'security' type_id=21 bits_offset=1856 's_fs_info' type_id=21 bits_offset=1920 'sb_flags' type_id=20 bits_offset=1984 'sb_flags_mask' type_id=20 bits_offset=2016 's_iflags' type_id=20 bits_offset=2048 'purpose' type_id=3600 bits_offset=2080 bitfield_size=8 'phase' type_id=3601 bits_offset=2088 bitfield_size=8 'need_free' type_id=67 bits_offset=2096 bitfield_size=1 'global' type_id=67 bits_offset=2097 bitfield_size=1 'oldapi' type_id=67 bits_offset=2098 bitfield_size=1 Fixes: 920d16af9b42 ("libbpf: BTF dumper support for typed data") Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20230428013638.1581263-1-martin.lau@linux.dev --- tools/lib/bpf/btf_dump.c | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c index 580985ee5545..4d9f30bf7f01 100644 --- a/tools/lib/bpf/btf_dump.c +++ b/tools/lib/bpf/btf_dump.c @@ -2250,9 +2250,25 @@ static int btf_dump_type_data_check_overflow(struct btf_dump *d, const struct btf_type *t, __u32 id, const void *data, - __u8 bits_offset) + __u8 bits_offset, + __u8 bit_sz) { - __s64 size = btf__resolve_size(d->btf, id); + __s64 size; + + if (bit_sz) { + /* bits_offset is at most 7. bit_sz is at most 128. */ + __u8 nr_bytes = (bits_offset + bit_sz + 7) / 8; + + /* When bit_sz is non zero, it is called from + * btf_dump_struct_data() where it only cares about + * negative error value. + * Return nr_bytes in success case to make it + * consistent as the regular integer case below. + */ + return data + nr_bytes > d->typed_dump->data_end ? -E2BIG : nr_bytes; + } + + size = btf__resolve_size(d->btf, id); if (size < 0 || size >= INT_MAX) { pr_warn("unexpected size [%zu] for id [%u]\n", @@ -2407,7 +2423,7 @@ static int btf_dump_dump_type_data(struct btf_dump *d, { int size, err = 0; - size = btf_dump_type_data_check_overflow(d, t, id, data, bits_offset); + size = btf_dump_type_data_check_overflow(d, t, id, data, bits_offset, bit_sz); if (size < 0) return size; err = btf_dump_type_data_check_zero(d, t, id, data, bits_offset, bit_sz); -- cgit v1.2.3 From 7866fc6aa0deb2294eb781fc3ac94b7d4c6eb242 Mon Sep 17 00:00:00 2001 From: Kenjiro Nakayama Date: Thu, 4 May 2023 12:54:43 +0900 Subject: libbpf: Fix comment about arc and riscv arch in bpf_tracing.h To make comments about arc and riscv arch in bpf_tracing.h accurate, this patch fixes the comment about arc and adds the comment for riscv. Signed-off-by: Kenjiro Nakayama Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20230504035443.427927-1-nakayamakenjiro@gmail.com --- tools/lib/bpf/bpf_tracing.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h index 6fb3d0f9af17..be076a4041ab 100644 --- a/tools/lib/bpf/bpf_tracing.h +++ b/tools/lib/bpf/bpf_tracing.h @@ -351,6 +351,7 @@ struct pt_regs___arm64 { * https://github.com/riscv-non-isa/riscv-elf-psabi-doc/blob/master/riscv-cc.adoc#risc-v-calling-conventions */ +/* riscv provides struct user_regs_struct instead of struct pt_regs to userspace */ #define __PT_REGS_CAST(x) ((const struct user_regs_struct *)(x)) #define __PT_PARM1_REG a0 #define __PT_PARM2_REG a1 @@ -383,7 +384,7 @@ struct pt_regs___arm64 { * https://raw.githubusercontent.com/wiki/foss-for-synopsys-dwc-arc-processors/toolchain/files/ARCv2_ABI.pdf */ -/* arc provides struct user_pt_regs instead of struct pt_regs to userspace */ +/* arc provides struct user_regs_struct instead of struct pt_regs to userspace */ #define __PT_REGS_CAST(x) ((const struct user_regs_struct *)(x)) #define __PT_PARM1_REG scratch.r0 #define __PT_PARM2_REG scratch.r1 -- cgit v1.2.3 From bdeeed3498c7871c17465bb4f11d1bc67f9098af Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Mon, 8 May 2023 23:55:02 -0700 Subject: libbpf: fix offsetof() and container_of() to work with CO-RE It seems like __builtin_offset() doesn't preserve CO-RE field relocations properly. So if offsetof() macro is defined through __builtin_offset(), CO-RE-enabled BPF code using container_of() will be subtly and silently broken. To avoid this problem, redefine offsetof() and container_of() in the form that works with CO-RE relocations more reliably. Fixes: 5fbc220862fc ("tools/libpf: Add offsetof/container_of macro in bpf_helpers.h") Reported-by: Lennart Poettering Signed-off-by: Andrii Nakryiko Acked-by: Yonghong Song Link: https://lore.kernel.org/r/20230509065502.2306180-1-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- tools/lib/bpf/bpf_helpers.h | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h index 929a3baca8ef..bbab9ad9dc5a 100644 --- a/tools/lib/bpf/bpf_helpers.h +++ b/tools/lib/bpf/bpf_helpers.h @@ -77,16 +77,21 @@ /* * Helper macros to manipulate data structures */ -#ifndef offsetof -#define offsetof(TYPE, MEMBER) ((unsigned long)&((TYPE *)0)->MEMBER) -#endif -#ifndef container_of + +/* offsetof() definition that uses __builtin_offset() might not preserve field + * offset CO-RE relocation properly, so force-redefine offsetof() using + * old-school approach which works with CO-RE correctly + */ +#undef offsetof +#define offsetof(type, member) ((unsigned long)&((type *)0)->member) + +/* redefined container_of() to ensure we use the above offsetof() macro */ +#undef container_of #define container_of(ptr, type, member) \ ({ \ void *__mptr = (void *)(ptr); \ ((type *)(__mptr - offsetof(type, member))); \ }) -#endif /* * Compiler (optimization) barrier. -- cgit v1.2.3 From 10cb8622b6958c2d47961d6a42c76e6c9f1c08f3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Mon, 15 May 2023 15:37:47 +0200 Subject: libbpf: Store zero fd to fd_array for loader kfunc relocation When moving some of the test kfuncs to bpf_testmod I hit an issue when some of the kfuncs that object uses are in module and some in vmlinux. The problem is that both vmlinux and module kfuncs get allocated btf_fd_idx index into fd_array, but we store to it the BTF fd value only for module's kfunc, not vmlinux's one because (it's zero). Then after the program is loaded we check if fd_array[btf_fd_idx] != 0 and close the fd. When the object has kfuncs from both vmlinux and module, the fd from fd_array[btf_fd_idx] from previous load will be stored in there for vmlinux's kfunc, so we close unrelated fd (of the program we just loaded in my case). Fixing this by storing zero to fd_array[btf_fd_idx] for vmlinux kfuncs, so the we won't close stale fd. Signed-off-by: Jiri Olsa Link: https://lore.kernel.org/r/20230515133756.1658301-2-jolsa@kernel.org Signed-off-by: Alexei Starovoitov --- tools/lib/bpf/gen_loader.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/gen_loader.c b/tools/lib/bpf/gen_loader.c index 83e8e3bfd8ff..cf3323fd47b8 100644 --- a/tools/lib/bpf/gen_loader.c +++ b/tools/lib/bpf/gen_loader.c @@ -703,17 +703,17 @@ static void emit_relo_kfunc_btf(struct bpf_gen *gen, struct ksym_relo_desc *relo /* obtain fd in BPF_REG_9 */ emit(gen, BPF_MOV64_REG(BPF_REG_9, BPF_REG_7)); emit(gen, BPF_ALU64_IMM(BPF_RSH, BPF_REG_9, 32)); - /* jump to fd_array store if fd denotes module BTF */ - emit(gen, BPF_JMP_IMM(BPF_JNE, BPF_REG_9, 0, 2)); - /* set the default value for off */ - emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), 0)); - /* skip BTF fd store for vmlinux BTF */ - emit(gen, BPF_JMP_IMM(BPF_JA, 0, 0, 4)); /* load fd_array slot pointer */ emit2(gen, BPF_LD_IMM64_RAW_FULL(BPF_REG_0, BPF_PSEUDO_MAP_IDX_VALUE, 0, 0, 0, blob_fd_array_off(gen, btf_fd_idx))); - /* store BTF fd in slot */ + /* store BTF fd in slot, 0 for vmlinux */ emit(gen, BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_9, 0)); + /* jump to insn[insn_idx].off store if fd denotes module BTF */ + emit(gen, BPF_JMP_IMM(BPF_JNE, BPF_REG_9, 0, 2)); + /* set the default value for off */ + emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), 0)); + /* skip BTF fd store for vmlinux BTF */ + emit(gen, BPF_JMP_IMM(BPF_JA, 0, 0, 1)); /* store index into insn[insn_idx].off */ emit(gen, BPF_ST_MEM(BPF_H, BPF_REG_8, offsetof(struct bpf_insn, off), btf_fd_idx)); log: -- cgit v1.2.3 From 2b001b94073be6097b1a4a21defc5cfbb7aa2f9f Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Tue, 23 May 2023 10:00:11 -0700 Subject: libbpf: Start v1.3 development cycle Bump libbpf.map to v1.3.0 to start a new libbpf version cycle. Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230523170013.728457-3-andrii@kernel.org --- tools/lib/bpf/libbpf.map | 3 +++ tools/lib/bpf/libbpf_version.h | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index a5aa3a383d69..9171ac89a802 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -391,3 +391,6 @@ LIBBPF_1.2.0 { bpf_map_get_info_by_fd; bpf_prog_get_info_by_fd; } LIBBPF_1.1.0; + +LIBBPF_1.3.0 { +} LIBBPF_1.2.0; diff --git a/tools/lib/bpf/libbpf_version.h b/tools/lib/bpf/libbpf_version.h index 1fd2eeac5cfc..290411ddb39e 100644 --- a/tools/lib/bpf/libbpf_version.h +++ b/tools/lib/bpf/libbpf_version.h @@ -4,6 +4,6 @@ #define __LIBBPF_VERSION_H #define LIBBPF_MAJOR_VERSION 1 -#define LIBBPF_MINOR_VERSION 2 +#define LIBBPF_MINOR_VERSION 3 #endif /* __LIBBPF_VERSION_H */ -- cgit v1.2.3 From f1674dc79fd276aeef6cad738a5f25ed2ba6d329 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Mon, 15 May 2023 17:01:58 -0700 Subject: libbpf: Add opts-based bpf_obj_pin() API and add support for path_fd Add path_fd support for bpf_obj_pin() and bpf_obj_get() operations (through their opts-based variants). This allows to take advantage of new kernel-side support for O_PATH-based pin/get location specification. Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230523170013.728457-4-andrii@kernel.org --- tools/lib/bpf/bpf.c | 17 ++++++++++++++--- tools/lib/bpf/bpf.h | 18 ++++++++++++++++-- tools/lib/bpf/libbpf.map | 2 ++ 3 files changed, 32 insertions(+), 5 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c index 128ac723c4ea..ed86b37d8024 100644 --- a/tools/lib/bpf/bpf.c +++ b/tools/lib/bpf/bpf.c @@ -572,20 +572,30 @@ int bpf_map_update_batch(int fd, const void *keys, const void *values, __u32 *co (void *)keys, (void *)values, count, opts); } -int bpf_obj_pin(int fd, const char *pathname) +int bpf_obj_pin_opts(int fd, const char *pathname, const struct bpf_obj_pin_opts *opts) { - const size_t attr_sz = offsetofend(union bpf_attr, file_flags); + const size_t attr_sz = offsetofend(union bpf_attr, path_fd); union bpf_attr attr; int ret; + if (!OPTS_VALID(opts, bpf_obj_pin_opts)) + return libbpf_err(-EINVAL); + memset(&attr, 0, attr_sz); + attr.path_fd = OPTS_GET(opts, path_fd, 0); attr.pathname = ptr_to_u64((void *)pathname); + attr.file_flags = OPTS_GET(opts, file_flags, 0); attr.bpf_fd = fd; ret = sys_bpf(BPF_OBJ_PIN, &attr, attr_sz); return libbpf_err_errno(ret); } +int bpf_obj_pin(int fd, const char *pathname) +{ + return bpf_obj_pin_opts(fd, pathname, NULL); +} + int bpf_obj_get(const char *pathname) { return bpf_obj_get_opts(pathname, NULL); @@ -593,7 +603,7 @@ int bpf_obj_get(const char *pathname) int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts) { - const size_t attr_sz = offsetofend(union bpf_attr, file_flags); + const size_t attr_sz = offsetofend(union bpf_attr, path_fd); union bpf_attr attr; int fd; @@ -601,6 +611,7 @@ int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts) return libbpf_err(-EINVAL); memset(&attr, 0, attr_sz); + attr.path_fd = OPTS_GET(opts, path_fd, 0); attr.pathname = ptr_to_u64((void *)pathname); attr.file_flags = OPTS_GET(opts, file_flags, 0); diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h index a2c091389b18..9aa0ee473754 100644 --- a/tools/lib/bpf/bpf.h +++ b/tools/lib/bpf/bpf.h @@ -284,16 +284,30 @@ LIBBPF_API int bpf_map_update_batch(int fd, const void *keys, const void *values __u32 *count, const struct bpf_map_batch_opts *opts); -struct bpf_obj_get_opts { +struct bpf_obj_pin_opts { size_t sz; /* size of this struct for forward/backward compatibility */ __u32 file_flags; + int path_fd; size_t :0; }; -#define bpf_obj_get_opts__last_field file_flags +#define bpf_obj_pin_opts__last_field path_fd LIBBPF_API int bpf_obj_pin(int fd, const char *pathname); +LIBBPF_API int bpf_obj_pin_opts(int fd, const char *pathname, + const struct bpf_obj_pin_opts *opts); + +struct bpf_obj_get_opts { + size_t sz; /* size of this struct for forward/backward compatibility */ + + __u32 file_flags; + int path_fd; + + size_t :0; +}; +#define bpf_obj_get_opts__last_field path_fd + LIBBPF_API int bpf_obj_get(const char *pathname); LIBBPF_API int bpf_obj_get_opts(const char *pathname, const struct bpf_obj_get_opts *opts); diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index 9171ac89a802..7521a2fb7626 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -393,4 +393,6 @@ LIBBPF_1.2.0 { } LIBBPF_1.1.0; LIBBPF_1.3.0 { + global: + bpf_obj_pin_opts; } LIBBPF_1.2.0; -- cgit v1.2.3 From 9d0a23313b1aa107df857e8441ea2ed74811ea17 Mon Sep 17 00:00:00 2001 From: JP Kobryn Date: Tue, 23 May 2023 17:45:36 -0700 Subject: libbpf: Add capability for resizing datasec maps This patch updates bpf_map__set_value_size() so that if the given map is memory mapped, it will attempt to resize the mapped region. Initial contents of the mapped region are preserved. BTF is not required, but after the mapping is resized an attempt is made to adjust the associated BTF information if the following criteria is met: - BTF info is present - the map is a datasec - the final variable in the datasec is an array ... the resulting BTF info will be updated so that the final array variable is associated with a new BTF array type sized to cover the requested size. Note that the initial resizing of the memory mapped region can succeed while the subsequent BTF adjustment can fail. In this case, BTF info is dropped from the map by clearing the key and value type. Signed-off-by: JP Kobryn Signed-off-by: Andrii Nakryiko Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20230524004537.18614-2-inwardvessel@gmail.com --- tools/lib/bpf/libbpf.c | 135 +++++++++++++++++++++++++++++++++++++++++++++---- tools/lib/bpf/libbpf.h | 18 ++++++- 2 files changed, 142 insertions(+), 11 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index ad1ec893b41b..5cca00979aae 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -1500,16 +1500,36 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj) return map; } -static size_t bpf_map_mmap_sz(const struct bpf_map *map) +static size_t bpf_map_mmap_sz(unsigned int value_sz, unsigned int max_entries) { - long page_sz = sysconf(_SC_PAGE_SIZE); + const long page_sz = sysconf(_SC_PAGE_SIZE); size_t map_sz; - map_sz = (size_t)roundup(map->def.value_size, 8) * map->def.max_entries; + map_sz = (size_t)roundup(value_sz, 8) * max_entries; map_sz = roundup(map_sz, page_sz); return map_sz; } +static int bpf_map_mmap_resize(struct bpf_map *map, size_t old_sz, size_t new_sz) +{ + void *mmaped; + + if (!map->mmaped) + return -EINVAL; + + if (old_sz == new_sz) + return 0; + + mmaped = mmap(NULL, new_sz, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0); + if (mmaped == MAP_FAILED) + return -errno; + + memcpy(mmaped, map->mmaped, min(old_sz, new_sz)); + munmap(map->mmaped, old_sz); + map->mmaped = mmaped; + return 0; +} + static char *internal_map_name(struct bpf_object *obj, const char *real_name) { char map_name[BPF_OBJ_NAME_LEN], *p; @@ -1608,6 +1628,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, { struct bpf_map_def *def; struct bpf_map *map; + size_t mmap_sz; int err; map = bpf_object__add_map(obj); @@ -1642,7 +1663,8 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type, pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n", map->name, map->sec_idx, map->sec_offset, def->map_flags); - map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE, + mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + map->mmaped = mmap(NULL, mmap_sz, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0); if (map->mmaped == MAP_FAILED) { err = -errno; @@ -8294,7 +8316,10 @@ static void bpf_map__destroy(struct bpf_map *map) map->init_slots_sz = 0; if (map->mmaped) { - munmap(map->mmaped, bpf_map_mmap_sz(map)); + size_t mmap_sz; + + mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + munmap(map->mmaped, mmap_sz); map->mmaped = NULL; } @@ -9412,10 +9437,103 @@ __u32 bpf_map__value_size(const struct bpf_map *map) return map->def.value_size; } +static int map_btf_datasec_resize(struct bpf_map *map, __u32 size) +{ + struct btf *btf; + struct btf_type *datasec_type, *var_type; + struct btf_var_secinfo *var; + const struct btf_type *array_type; + const struct btf_array *array; + int vlen, element_sz; + __u32 nr_elements, new_array_id; + + /* check btf existence */ + btf = bpf_object__btf(map->obj); + if (!btf) + return -ENOENT; + + /* verify map is datasec */ + datasec_type = btf_type_by_id(btf, bpf_map__btf_value_type_id(map)); + if (!btf_is_datasec(datasec_type)) { + pr_warn("map '%s': cannot be resized, map value type is not a datasec\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify datasec has at least one var */ + vlen = btf_vlen(datasec_type); + if (vlen == 0) { + pr_warn("map '%s': cannot be resized, map value datasec is empty\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify last var in the datasec is an array */ + var = &btf_var_secinfos(datasec_type)[vlen - 1]; + var_type = btf_type_by_id(btf, var->type); + array_type = skip_mods_and_typedefs(btf, var_type->type, NULL); + if (!btf_is_array(array_type)) { + pr_warn("map '%s': cannot be resized, last var must be an array\n", + bpf_map__name(map)); + return -EINVAL; + } + + /* verify request size aligns with array */ + array = btf_array(array_type); + element_sz = btf__resolve_size(btf, array->type); + if (element_sz <= 0 || (size - var->offset) % element_sz != 0) { + pr_warn("map '%s': cannot be resized, element size (%d) doesn't align with new total size (%u)\n", + bpf_map__name(map), element_sz, size); + return -EINVAL; + } + + /* create a new array based on the existing array, but with new length */ + nr_elements = (size - var->offset) / element_sz; + new_array_id = btf__add_array(btf, array->index_type, array->type, nr_elements); + if (new_array_id < 0) + return new_array_id; + + /* adding a new btf type invalidates existing pointers to btf objects, + * so refresh pointers before proceeding + */ + datasec_type = btf_type_by_id(btf, map->btf_value_type_id); + var = &btf_var_secinfos(datasec_type)[vlen - 1]; + var_type = btf_type_by_id(btf, var->type); + + /* finally update btf info */ + datasec_type->size = size; + var->size = size - var->offset; + var_type->type = new_array_id; + + return 0; +} + int bpf_map__set_value_size(struct bpf_map *map, __u32 size) { if (map->fd >= 0) return libbpf_err(-EBUSY); + + if (map->mmaped) { + int err; + size_t mmap_old_sz, mmap_new_sz; + + mmap_old_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); + mmap_new_sz = bpf_map_mmap_sz(size, map->def.max_entries); + err = bpf_map_mmap_resize(map, mmap_old_sz, mmap_new_sz); + if (err) { + pr_warn("map '%s': failed to resize memory-mapped region: %d\n", + bpf_map__name(map), err); + return err; + } + err = map_btf_datasec_resize(map, size); + if (err && err != -ENOENT) { + pr_warn("map '%s': failed to adjust resized BTF, clearing BTF key/value info: %d\n", + bpf_map__name(map), err); + map->btf_value_type_id = 0; + map->btf_key_type_id = 0; + } + } + map->def.value_size = size; return 0; } @@ -9441,7 +9559,7 @@ int bpf_map__set_initial_value(struct bpf_map *map, return 0; } -const void *bpf_map__initial_value(struct bpf_map *map, size_t *psize) +void *bpf_map__initial_value(struct bpf_map *map, size_t *psize) { if (!map->mmaped) return NULL; @@ -12693,7 +12811,7 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s) for (i = 0; i < s->map_cnt; i++) { struct bpf_map *map = *s->maps[i].map; - size_t mmap_sz = bpf_map_mmap_sz(map); + size_t mmap_sz = bpf_map_mmap_sz(map->def.value_size, map->def.max_entries); int prot, map_fd = bpf_map__fd(map); void **mmaped = s->maps[i].mmaped; @@ -12720,8 +12838,7 @@ int bpf_object__load_skeleton(struct bpf_object_skeleton *s) * as per normal clean up procedure, so we don't need to worry * about it from skeleton's clean up perspective. */ - *mmaped = mmap(map->mmaped, mmap_sz, prot, - MAP_SHARED | MAP_FIXED, map_fd, 0); + *mmaped = mmap(map->mmaped, mmap_sz, prot, MAP_SHARED | MAP_FIXED, map_fd, 0); if (*mmaped == MAP_FAILED) { err = -errno; *mmaped = NULL; diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index 0b7362397ea3..754da73c643b 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -869,8 +869,22 @@ LIBBPF_API int bpf_map__set_numa_node(struct bpf_map *map, __u32 numa_node); /* get/set map key size */ LIBBPF_API __u32 bpf_map__key_size(const struct bpf_map *map); LIBBPF_API int bpf_map__set_key_size(struct bpf_map *map, __u32 size); -/* get/set map value size */ +/* get map value size */ LIBBPF_API __u32 bpf_map__value_size(const struct bpf_map *map); +/** + * @brief **bpf_map__set_value_size()** sets map value size. + * @param map the BPF map instance + * @return 0, on success; negative error, otherwise + * + * There is a special case for maps with associated memory-mapped regions, like + * the global data section maps (bss, data, rodata). When this function is used + * on such a map, the mapped region is resized. Afterward, an attempt is made to + * adjust the corresponding BTF info. This attempt is best-effort and can only + * succeed if the last variable of the data section map is an array. The array + * BTF type is replaced by a new BTF array type with a different length. + * Any previously existing pointers returned from bpf_map__initial_value() or + * corresponding data section skeleton pointer must be reinitialized. + */ LIBBPF_API int bpf_map__set_value_size(struct bpf_map *map, __u32 size); /* get map key/value BTF type IDs */ LIBBPF_API __u32 bpf_map__btf_key_type_id(const struct bpf_map *map); @@ -884,7 +898,7 @@ LIBBPF_API int bpf_map__set_map_extra(struct bpf_map *map, __u64 map_extra); LIBBPF_API int bpf_map__set_initial_value(struct bpf_map *map, const void *data, size_t size); -LIBBPF_API const void *bpf_map__initial_value(struct bpf_map *map, size_t *psize); +LIBBPF_API void *bpf_map__initial_value(struct bpf_map *map, size_t *psize); /** * @brief **bpf_map__is_internal()** tells the caller whether or not the -- cgit v1.2.3 From 4c857a719bf9c5ddbcf3bd92398632041bff50d0 Mon Sep 17 00:00:00 2001 From: JP Kobryn Date: Wed, 24 May 2023 17:13:23 -0700 Subject: libbpf: Change var type in datasec resize func This changes a local variable type that stores a new array id to match the return type of btf__add_array(). Signed-off-by: JP Kobryn Signed-off-by: Andrii Nakryiko Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20230525001323.8554-1-inwardvessel@gmail.com --- tools/lib/bpf/libbpf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 5cca00979aae..1ceb3a97dadc 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -9444,8 +9444,8 @@ static int map_btf_datasec_resize(struct bpf_map *map, __u32 size) struct btf_var_secinfo *var; const struct btf_type *array_type; const struct btf_array *array; - int vlen, element_sz; - __u32 nr_elements, new_array_id; + int vlen, element_sz, new_array_id; + __u32 nr_elements; /* check btf existence */ btf = bpf_object__btf(map->obj); -- cgit v1.2.3 From 59842c5451fe830737600276ba9dee4595341d77 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 25 May 2023 15:13:10 -0700 Subject: libbpf: Ensure libbpf always opens files with O_CLOEXEC Make sure that libbpf code always gets FD with O_CLOEXEC flag set, regardless if file is open through open() or fopen(). For the latter this means to add "e" to mode string, which is supported since pretty ancient glibc v2.7. Also drop the outdated TODO comment in usdt.c, which was already completed. Suggested-by: Lennart Poettering Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230525221311.2136408-1-andrii@kernel.org --- tools/lib/bpf/btf.c | 2 +- tools/lib/bpf/libbpf.c | 6 +++--- tools/lib/bpf/libbpf_probes.c | 2 +- tools/lib/bpf/usdt.c | 5 ++--- 4 files changed, 7 insertions(+), 8 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 0a2c079244b6..8484b563b53d 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -1064,7 +1064,7 @@ static struct btf *btf_parse_raw(const char *path, struct btf *base_btf) int err = 0; long sz; - f = fopen(path, "rb"); + f = fopen(path, "rbe"); if (!f) { err = -errno; goto err_out; diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 1ceb3a97dadc..60ef4c5e3bee 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -4351,7 +4351,7 @@ static int bpf_get_map_info_from_fdinfo(int fd, struct bpf_map_info *info) snprintf(file, sizeof(file), "/proc/%d/fdinfo/%d", getpid(), fd); memset(info, 0, sizeof(*info)); - fp = fopen(file, "r"); + fp = fopen(file, "re"); if (!fp) { err = -errno; pr_warn("failed to open %s: %d. No procfs support?\n", file, @@ -7455,7 +7455,7 @@ int libbpf_kallsyms_parse(kallsyms_cb_t cb, void *ctx) int ret, err = 0; FILE *f; - f = fopen("/proc/kallsyms", "r"); + f = fopen("/proc/kallsyms", "re"); if (!f) { err = -errno; pr_warn("failed to open /proc/kallsyms: %d\n", err); @@ -10075,7 +10075,7 @@ static int parse_uint_from_file(const char *file, const char *fmt) int err, ret; FILE *f; - f = fopen(file, "r"); + f = fopen(file, "re"); if (!f) { err = -errno; pr_debug("failed to open '%s': %s\n", file, diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c index 6065f408a59c..5f78a023b474 100644 --- a/tools/lib/bpf/libbpf_probes.c +++ b/tools/lib/bpf/libbpf_probes.c @@ -38,7 +38,7 @@ static __u32 get_ubuntu_kernel_version(void) if (faccessat(AT_FDCWD, ubuntu_kver_file, R_OK, AT_EACCESS) != 0) return 0; - f = fopen(ubuntu_kver_file, "r"); + f = fopen(ubuntu_kver_file, "re"); if (!f) return 0; diff --git a/tools/lib/bpf/usdt.c b/tools/lib/bpf/usdt.c index 086eef355ab3..f1a141555f08 100644 --- a/tools/lib/bpf/usdt.c +++ b/tools/lib/bpf/usdt.c @@ -466,7 +466,7 @@ static int parse_vma_segs(int pid, const char *lib_path, struct elf_seg **segs, proceed: sprintf(line, "/proc/%d/maps", pid); - f = fopen(line, "r"); + f = fopen(line, "re"); if (!f) { err = -errno; pr_warn("usdt: failed to open '%s' to get base addr of '%s': %d\n", @@ -954,8 +954,7 @@ struct bpf_link *usdt_manager_attach_usdt(struct usdt_manager *man, const struct spec_map_fd = bpf_map__fd(man->specs_map); ip_map_fd = bpf_map__fd(man->ip_to_spec_id_map); - /* TODO: perform path resolution similar to uprobe's */ - fd = open(path, O_RDONLY); + fd = open(path, O_RDONLY | O_CLOEXEC); if (fd < 0) { err = -errno; pr_warn("usdt: failed to open ELF binary '%s': %d\n", path, err); -- cgit v1.2.3 From 4aadd2920b81b3d7e5c8ac63c7d5d673f3c8aaeb Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 25 May 2023 15:13:11 -0700 Subject: libbpf: Ensure FD >= 3 during bpf_map__reuse_fd() Improve bpf_map__reuse_fd() logic and ensure that dup'ed map FD is "good" (>= 3) and has O_CLOEXEC flags. Use fcntl(F_DUPFD_CLOEXEC) for that, similarly to ensure_good_fd() helper we already use in low-level APIs that work with bpf() syscall. Suggested-by: Lennart Poettering Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230525221311.2136408-2-andrii@kernel.org --- tools/lib/bpf/libbpf.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'tools/lib') diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 60ef4c5e3bee..47632606b06d 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -4414,18 +4414,17 @@ int bpf_map__reuse_fd(struct bpf_map *map, int fd) if (!new_name) return libbpf_err(-errno); - new_fd = open("/", O_RDONLY | O_CLOEXEC); + /* + * Like dup(), but make sure new FD is >= 3 and has O_CLOEXEC set. + * This is similar to what we do in ensure_good_fd(), but without + * closing original FD. + */ + new_fd = fcntl(fd, F_DUPFD_CLOEXEC, 3); if (new_fd < 0) { err = -errno; goto err_free_new_name; } - new_fd = dup3(fd, new_fd, O_CLOEXEC); - if (new_fd < 0) { - err = -errno; - goto err_close_new_fd; - } - err = zclose(map->fd); if (err) { err = -errno; -- cgit v1.2.3