diff options
author | David S. Miller <davem@davemloft.net> | 2020-09-01 13:05:08 -0700 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2020-09-01 13:22:59 -0700 |
commit | 150f29f5e6ea55d8a7d368b162a4e9947a95d2f5 (patch) | |
tree | d028c9c9a7cdddcc79feb49e79fa011af029ff8c /tools/lib | |
parent | 8aa639e1483bbdc0615796801829c773724f6645 (diff) | |
parent | ebc4ecd48ca6552b223047839f66e9a9c09aea4c (diff) | |
download | linux-150f29f5e6ea55d8a7d368b162a4e9947a95d2f5.tar.bz2 |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
Daniel Borkmann says:
====================
pull-request: bpf-next 2020-09-01
The following pull-request contains BPF updates for your *net-next* tree.
There are two small conflicts when pulling, resolve as follows:
1) Merge conflict in tools/lib/bpf/libbpf.c between 88a82120282b ("libbpf: Factor
out common ELF operations and improve logging") in bpf-next and 1e891e513e16
("libbpf: Fix map index used in error message") in net-next. Resolve by taking
the hunk in bpf-next:
[...]
scn = elf_sec_by_idx(obj, obj->efile.btf_maps_shndx);
data = elf_sec_data(obj, scn);
if (!scn || !data) {
pr_warn("elf: failed to get %s map definitions for %s\n",
MAPS_ELF_SEC, obj->path);
return -EINVAL;
}
[...]
2) Merge conflict in drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c between
9647c57b11e5 ("xsk: i40e: ice: ixgbe: mlx5: Test for dma_need_sync earlier for
better performance") in bpf-next and e20f0dbf204f ("net/mlx5e: RX, Add a prefetch
command for small L1_CACHE_BYTES") in net-next. Resolve the two locations by retaining
net_prefetch() and taking xsk_buff_dma_sync_for_cpu() from bpf-next. Should look like:
[...]
xdp_set_data_meta_invalid(xdp);
xsk_buff_dma_sync_for_cpu(xdp, rq->xsk_pool);
net_prefetch(xdp->data);
[...]
We've added 133 non-merge commits during the last 14 day(s) which contain
a total of 246 files changed, 13832 insertions(+), 3105 deletions(-).
The main changes are:
1) Initial support for sleepable BPF programs along with bpf_copy_from_user() helper
for tracing to reliably access user memory, from Alexei Starovoitov.
2) Add BPF infra for writing and parsing TCP header options, from Martin KaFai Lau.
3) bpf_d_path() helper for returning full path for given 'struct path', from Jiri Olsa.
4) AF_XDP support for shared umems between devices and queues, from Magnus Karlsson.
5) Initial prep work for full BPF-to-BPF call support in libbpf, from Andrii Nakryiko.
6) Generalize bpf_sk_storage map & add local storage for inodes, from KP Singh.
7) Implement sockmap/hash updates from BPF context, from Lorenz Bauer.
8) BPF xor verification for scalar types & add BPF link iterator, from Yonghong Song.
9) Use target's prog type for BPF_PROG_TYPE_EXT prog verification, from Udip Pant.
10) Rework BPF tracing samples to use libbpf loader, from Daniel T. Lee.
11) Fix xdpsock sample to really cycle through all buffers, from Weqaar Janjua.
12) Improve type safety for tun/veth XDP frame handling, from Maciej Żenczykowski.
13) Various smaller cleanups and improvements all over the place.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'tools/lib')
-rw-r--r-- | tools/lib/bpf/Makefile | 23 | ||||
-rw-r--r-- | tools/lib/bpf/bpf.c | 3 | ||||
-rw-r--r-- | tools/lib/bpf/bpf_core_read.h | 120 | ||||
-rw-r--r-- | tools/lib/bpf/bpf_helpers.h | 3 | ||||
-rw-r--r-- | tools/lib/bpf/bpf_prog_linfo.c | 3 | ||||
-rw-r--r-- | tools/lib/bpf/bpf_tracing.h | 4 | ||||
-rw-r--r-- | tools/lib/bpf/btf.c | 31 | ||||
-rw-r--r-- | tools/lib/bpf/btf.h | 38 | ||||
-rw-r--r-- | tools/lib/bpf/btf_dump.c | 9 | ||||
-rw-r--r-- | tools/lib/bpf/hashmap.c | 3 | ||||
-rw-r--r-- | tools/lib/bpf/libbpf.c | 1727 | ||||
-rw-r--r-- | tools/lib/bpf/libbpf.h | 4 | ||||
-rw-r--r-- | tools/lib/bpf/libbpf.map | 9 | ||||
-rw-r--r-- | tools/lib/bpf/libbpf_internal.h | 138 | ||||
-rw-r--r-- | tools/lib/bpf/libbpf_probes.c | 8 | ||||
-rw-r--r-- | tools/lib/bpf/netlink.c | 128 | ||||
-rw-r--r-- | tools/lib/bpf/nlattr.c | 9 | ||||
-rw-r--r-- | tools/lib/bpf/ringbuf.c | 8 | ||||
-rw-r--r-- | tools/lib/bpf/xsk.c | 379 | ||||
-rw-r--r-- | tools/lib/bpf/xsk.h | 9 |
20 files changed, 1746 insertions, 910 deletions
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile index bf8ed134cb8a..adbe994610f2 100644 --- a/tools/lib/bpf/Makefile +++ b/tools/lib/bpf/Makefile @@ -1,6 +1,9 @@ # SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) # Most of this file is copied from tools/lib/traceevent/Makefile +RM ?= rm +srctree = $(abs_srctree) + LIBBPF_VERSION := $(shell \ grep -oE '^LIBBPF_([0-9.]+)' libbpf.map | \ sort -rV | head -n1 | cut -d'_' -f2) @@ -56,7 +59,7 @@ ifndef VERBOSE endif FEATURE_USER = .libbpf -FEATURE_TESTS = libelf libelf-mmap zlib bpf reallocarray +FEATURE_TESTS = libelf zlib bpf FEATURE_DISPLAY = libelf zlib bpf INCLUDES = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(ARCH)/include/uapi -I$(srctree)/tools/include/uapi @@ -98,16 +101,8 @@ else CFLAGS := -g -Wall endif -ifeq ($(feature-libelf-mmap), 1) - override CFLAGS += -DHAVE_LIBELF_MMAP_SUPPORT -endif - -ifeq ($(feature-reallocarray), 0) - override CFLAGS += -DCOMPAT_NEED_REALLOCARRAY -endif - # Append required CFLAGS -override CFLAGS += $(EXTRA_WARNINGS) +override CFLAGS += $(EXTRA_WARNINGS) -Wno-switch-enum override CFLAGS += -Werror -Wall override CFLAGS += -fPIC override CFLAGS += $(INCLUDES) @@ -196,7 +191,7 @@ $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED) @ln -sf $(@F) $(OUTPUT)libbpf.so.$(LIBBPF_MAJOR_VERSION) $(OUTPUT)libbpf.a: $(BPF_IN_STATIC) - $(QUIET_LINK)$(RM) $@; $(AR) rcs $@ $^ + $(QUIET_LINK)$(RM) -f $@; $(AR) rcs $@ $^ $(OUTPUT)libbpf.pc: $(QUIET_GEN)sed -e "s|@PREFIX@|$(prefix)|" \ @@ -269,10 +264,10 @@ install: install_lib install_pkgconfig install_headers ### Cleaning rules config-clean: - $(call QUIET_CLEAN, config) + $(call QUIET_CLEAN, feature-detect) $(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null -clean: +clean: config-clean $(call QUIET_CLEAN, libbpf) $(RM) -rf $(CMD_TARGETS) \ *~ .*.d .*.cmd LIBBPF-CFLAGS $(BPF_HELPER_DEFS) \ $(SHARED_OBJDIR) $(STATIC_OBJDIR) \ @@ -299,7 +294,7 @@ cscope: cscope -b -q -I $(srctree)/include -f cscope.out tags: - rm -f TAGS tags + $(RM) -f TAGS tags ls *.c *.h | xargs $(TAGS_PROG) -a # Declare the contents of the .PHONY variable as phony. We keep that diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c index 0750681057c2..82b983ff6569 100644 --- a/tools/lib/bpf/bpf.c +++ b/tools/lib/bpf/bpf.c @@ -32,9 +32,6 @@ #include "libbpf.h" #include "libbpf_internal.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - /* * When building perf, unistd.h is overridden. __NR_bpf is * required to be defined explicitly. diff --git a/tools/lib/bpf/bpf_core_read.h b/tools/lib/bpf/bpf_core_read.h index eae5cccff761..bbcefb3ff5a5 100644 --- a/tools/lib/bpf/bpf_core_read.h +++ b/tools/lib/bpf/bpf_core_read.h @@ -19,32 +19,52 @@ enum bpf_field_info_kind { BPF_FIELD_RSHIFT_U64 = 5, }; +/* second argument to __builtin_btf_type_id() built-in */ +enum bpf_type_id_kind { + BPF_TYPE_ID_LOCAL = 0, /* BTF type ID in local program */ + BPF_TYPE_ID_TARGET = 1, /* BTF type ID in target kernel */ +}; + +/* second argument to __builtin_preserve_type_info() built-in */ +enum bpf_type_info_kind { + BPF_TYPE_EXISTS = 0, /* type existence in target kernel */ + BPF_TYPE_SIZE = 1, /* type size in target kernel */ +}; + +/* second argument to __builtin_preserve_enum_value() built-in */ +enum bpf_enum_value_kind { + BPF_ENUMVAL_EXISTS = 0, /* enum value existence in kernel */ + BPF_ENUMVAL_VALUE = 1, /* enum value value relocation */ +}; + #define __CORE_RELO(src, field, info) \ __builtin_preserve_field_info((src)->field, BPF_FIELD_##info) #if __BYTE_ORDER == __LITTLE_ENDIAN #define __CORE_BITFIELD_PROBE_READ(dst, src, fld) \ - bpf_probe_read((void *)dst, \ - __CORE_RELO(src, fld, BYTE_SIZE), \ - (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET)) + bpf_probe_read_kernel( \ + (void *)dst, \ + __CORE_RELO(src, fld, BYTE_SIZE), \ + (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET)) #else /* semantics of LSHIFT_64 assumes loading values into low-ordered bytes, so * for big-endian we need to adjust destination pointer accordingly, based on * field byte size */ #define __CORE_BITFIELD_PROBE_READ(dst, src, fld) \ - bpf_probe_read((void *)dst + (8 - __CORE_RELO(src, fld, BYTE_SIZE)), \ - __CORE_RELO(src, fld, BYTE_SIZE), \ - (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET)) + bpf_probe_read_kernel( \ + (void *)dst + (8 - __CORE_RELO(src, fld, BYTE_SIZE)), \ + __CORE_RELO(src, fld, BYTE_SIZE), \ + (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET)) #endif /* * Extract bitfield, identified by s->field, and return its value as u64. * All this is done in relocatable manner, so bitfield changes such as * signedness, bit size, offset changes, this will be handled automatically. - * This version of macro is using bpf_probe_read() to read underlying integer - * storage. Macro functions as an expression and its return type is - * bpf_probe_read()'s return value: 0, on success, <0 on error. + * This version of macro is using bpf_probe_read_kernel() to read underlying + * integer storage. Macro functions as an expression and its return type is + * bpf_probe_read_kernel()'s return value: 0, on success, <0 on error. */ #define BPF_CORE_READ_BITFIELD_PROBED(s, field) ({ \ unsigned long long val = 0; \ @@ -92,15 +112,75 @@ enum bpf_field_info_kind { __builtin_preserve_field_info(field, BPF_FIELD_EXISTS) /* - * Convenience macro to get byte size of a field. Works for integers, + * Convenience macro to get the byte size of a field. Works for integers, * struct/unions, pointers, arrays, and enums. */ #define bpf_core_field_size(field) \ __builtin_preserve_field_info(field, BPF_FIELD_BYTE_SIZE) /* - * bpf_core_read() abstracts away bpf_probe_read() call and captures offset - * relocation for source address using __builtin_preserve_access_index() + * Convenience macro to get BTF type ID of a specified type, using a local BTF + * information. Return 32-bit unsigned integer with type ID from program's own + * BTF. Always succeeds. + */ +#define bpf_core_type_id_local(type) \ + __builtin_btf_type_id(*(typeof(type) *)0, BPF_TYPE_ID_LOCAL) + +/* + * Convenience macro to get BTF type ID of a target kernel's type that matches + * specified local type. + * Returns: + * - valid 32-bit unsigned type ID in kernel BTF; + * - 0, if no matching type was found in a target kernel BTF. + */ +#define bpf_core_type_id_kernel(type) \ + __builtin_btf_type_id(*(typeof(type) *)0, BPF_TYPE_ID_TARGET) + +/* + * Convenience macro to check that provided named type + * (struct/union/enum/typedef) exists in a target kernel. + * Returns: + * 1, if such type is present in target kernel's BTF; + * 0, if no matching type is found. + */ +#define bpf_core_type_exists(type) \ + __builtin_preserve_type_info(*(typeof(type) *)0, BPF_TYPE_EXISTS) + +/* + * Convenience macro to get the byte size of a provided named type + * (struct/union/enum/typedef) in a target kernel. + * Returns: + * >= 0 size (in bytes), if type is present in target kernel's BTF; + * 0, if no matching type is found. + */ +#define bpf_core_type_size(type) \ + __builtin_preserve_type_info(*(typeof(type) *)0, BPF_TYPE_SIZE) + +/* + * Convenience macro to check that provided enumerator value is defined in + * a target kernel. + * Returns: + * 1, if specified enum type and its enumerator value are present in target + * kernel's BTF; + * 0, if no matching enum and/or enum value within that enum is found. + */ +#define bpf_core_enum_value_exists(enum_type, enum_value) \ + __builtin_preserve_enum_value(*(typeof(enum_type) *)enum_value, BPF_ENUMVAL_EXISTS) + +/* + * Convenience macro to get the integer value of an enumerator value in + * a target kernel. + * Returns: + * 64-bit value, if specified enum type and its enumerator value are + * present in target kernel's BTF; + * 0, if no matching enum and/or enum value within that enum is found. + */ +#define bpf_core_enum_value(enum_type, enum_value) \ + __builtin_preserve_enum_value(*(typeof(enum_type) *)enum_value, BPF_ENUMVAL_VALUE) + +/* + * bpf_core_read() abstracts away bpf_probe_read_kernel() call and captures + * offset relocation for source address using __builtin_preserve_access_index() * built-in, provided by Clang. * * __builtin_preserve_access_index() takes as an argument an expression of @@ -115,8 +195,8 @@ enum bpf_field_info_kind { * (local) BTF, used to record relocation. */ #define bpf_core_read(dst, sz, src) \ - bpf_probe_read(dst, sz, \ - (const void *)__builtin_preserve_access_index(src)) + bpf_probe_read_kernel(dst, sz, \ + (const void *)__builtin_preserve_access_index(src)) /* * bpf_core_read_str() is a thin wrapper around bpf_probe_read_str() @@ -124,8 +204,8 @@ enum bpf_field_info_kind { * argument. */ #define bpf_core_read_str(dst, sz, src) \ - bpf_probe_read_str(dst, sz, \ - (const void *)__builtin_preserve_access_index(src)) + bpf_probe_read_kernel_str(dst, sz, \ + (const void *)__builtin_preserve_access_index(src)) #define ___concat(a, b) a ## b #define ___apply(fn, n) ___concat(fn, n) @@ -239,15 +319,17 @@ enum bpf_field_info_kind { * 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: + * CO-RE relocatable bpf_probe_read_kernel() 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. + * involved, as well as all the reads are happening through + * bpf_probe_read_kernel() 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. diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h index e9a4ecddb7a5..1106777df00b 100644 --- a/tools/lib/bpf/bpf_helpers.h +++ b/tools/lib/bpf/bpf_helpers.h @@ -32,6 +32,9 @@ #ifndef __always_inline #define __always_inline __attribute__((always_inline)) #endif +#ifndef __noinline +#define __noinline __attribute__((noinline)) +#endif #ifndef __weak #define __weak __attribute__((weak)) #endif diff --git a/tools/lib/bpf/bpf_prog_linfo.c b/tools/lib/bpf/bpf_prog_linfo.c index bafca49cb1e6..3ed1a27b5f7c 100644 --- a/tools/lib/bpf/bpf_prog_linfo.c +++ b/tools/lib/bpf/bpf_prog_linfo.c @@ -8,9 +8,6 @@ #include "libbpf.h" #include "libbpf_internal.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - struct bpf_prog_linfo { void *raw_linfo; void *raw_jited_linfo; diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h index eebf020cbe3e..f9ef37707888 100644 --- a/tools/lib/bpf/bpf_tracing.h +++ b/tools/lib/bpf/bpf_tracing.h @@ -289,9 +289,9 @@ struct pt_regs; #define BPF_KRETPROBE_READ_RET_IP BPF_KPROBE_READ_RET_IP #else #define BPF_KPROBE_READ_RET_IP(ip, ctx) \ - ({ bpf_probe_read(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); }) + ({ bpf_probe_read_kernel(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); }) #define BPF_KRETPROBE_READ_RET_IP(ip, ctx) \ - ({ bpf_probe_read(&(ip), sizeof(ip), \ + ({ bpf_probe_read_kernel(&(ip), sizeof(ip), \ (void *)(PT_REGS_FP(ctx) + sizeof(ip))); }) #endif diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c index 7dfca7016aaa..a3d259e614b0 100644 --- a/tools/lib/bpf/btf.c +++ b/tools/lib/bpf/btf.c @@ -21,9 +21,6 @@ #include "libbpf_internal.h" #include "hashmap.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - #define BTF_MAX_NR_TYPES 0x7fffffffU #define BTF_MAX_STR_OFFSET 0x7fffffffU @@ -61,7 +58,7 @@ static int btf_add_type(struct btf *btf, struct btf_type *t) expand_by = max(btf->types_size >> 2, 16U); new_size = min(BTF_MAX_NR_TYPES, btf->types_size + expand_by); - new_types = realloc(btf->types, sizeof(*new_types) * new_size); + new_types = libbpf_reallocarray(btf->types, new_size, sizeof(*new_types)); if (!new_types) return -ENOMEM; @@ -1131,14 +1128,14 @@ static int btf_ext_setup_line_info(struct btf_ext *btf_ext) return btf_ext_setup_info(btf_ext, ¶m); } -static int btf_ext_setup_field_reloc(struct btf_ext *btf_ext) +static int btf_ext_setup_core_relos(struct btf_ext *btf_ext) { struct btf_ext_sec_setup_param param = { - .off = btf_ext->hdr->field_reloc_off, - .len = btf_ext->hdr->field_reloc_len, - .min_rec_size = sizeof(struct bpf_field_reloc), - .ext_info = &btf_ext->field_reloc_info, - .desc = "field_reloc", + .off = btf_ext->hdr->core_relo_off, + .len = btf_ext->hdr->core_relo_len, + .min_rec_size = sizeof(struct bpf_core_relo), + .ext_info = &btf_ext->core_relo_info, + .desc = "core_relo", }; return btf_ext_setup_info(btf_ext, ¶m); @@ -1217,10 +1214,9 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size) if (err) goto done; - if (btf_ext->hdr->hdr_len < - offsetofend(struct btf_ext_header, field_reloc_len)) + if (btf_ext->hdr->hdr_len < offsetofend(struct btf_ext_header, core_relo_len)) goto done; - err = btf_ext_setup_field_reloc(btf_ext); + err = btf_ext_setup_core_relos(btf_ext); if (err) goto done; @@ -1575,7 +1571,7 @@ static int btf_dedup_hypot_map_add(struct btf_dedup *d, __u32 *new_list; d->hypot_cap += max((size_t)16, d->hypot_cap / 2); - new_list = realloc(d->hypot_list, sizeof(__u32) * d->hypot_cap); + new_list = libbpf_reallocarray(d->hypot_list, d->hypot_cap, sizeof(__u32)); if (!new_list) return -ENOMEM; d->hypot_list = new_list; @@ -1871,8 +1867,7 @@ static int btf_dedup_strings(struct btf_dedup *d) struct btf_str_ptr *new_ptrs; strs.cap += max(strs.cnt / 2, 16U); - new_ptrs = realloc(strs.ptrs, - sizeof(strs.ptrs[0]) * strs.cap); + new_ptrs = libbpf_reallocarray(strs.ptrs, strs.cap, sizeof(strs.ptrs[0])); if (!new_ptrs) { err = -ENOMEM; goto done; @@ -2957,8 +2952,8 @@ static int btf_dedup_compact_types(struct btf_dedup *d) d->btf->nr_types = next_type_id - 1; d->btf->types_size = d->btf->nr_types; d->btf->hdr->type_len = p - types_start; - new_types = realloc(d->btf->types, - (1 + d->btf->nr_types) * sizeof(struct btf_type *)); + new_types = libbpf_reallocarray(d->btf->types, (1 + d->btf->nr_types), + sizeof(struct btf_type *)); if (!new_types) return -ENOMEM; d->btf->types = new_types; diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h index 1ca14448df4c..91f0ad0e0325 100644 --- a/tools/lib/bpf/btf.h +++ b/tools/lib/bpf/btf.h @@ -24,44 +24,6 @@ struct btf_type; struct bpf_object; -/* - * The .BTF.ext ELF section layout defined as - * struct btf_ext_header - * func_info subsection - * - * The func_info subsection layout: - * record size for struct bpf_func_info in the func_info subsection - * struct btf_sec_func_info for section #1 - * a list of bpf_func_info records for section #1 - * where struct bpf_func_info mimics one in include/uapi/linux/bpf.h - * but may not be identical - * struct btf_sec_func_info for section #2 - * a list of bpf_func_info records for section #2 - * ...... - * - * Note that the bpf_func_info record size in .BTF.ext may not - * be the same as the one defined in include/uapi/linux/bpf.h. - * The loader should ensure that record_size meets minimum - * requirement and pass the record as is to the kernel. The - * kernel will handle the func_info properly based on its contents. - */ -struct btf_ext_header { - __u16 magic; - __u8 version; - __u8 flags; - __u32 hdr_len; - - /* All offsets are in bytes relative to the end of this header */ - __u32 func_info_off; - __u32 func_info_len; - __u32 line_info_off; - __u32 line_info_len; - - /* optional part of .BTF.ext header */ - __u32 field_reloc_off; - __u32 field_reloc_len; -}; - LIBBPF_API void btf__free(struct btf *btf); LIBBPF_API struct btf *btf__new(const void *data, __u32 size); LIBBPF_API struct btf *btf__parse(const char *path, struct btf_ext **btf_ext); diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c index 57c00fa63932..6c079b3c8679 100644 --- a/tools/lib/bpf/btf_dump.c +++ b/tools/lib/bpf/btf_dump.c @@ -19,9 +19,6 @@ #include "libbpf.h" #include "libbpf_internal.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - static const char PREFIXES[] = "\t\t\t\t\t\t\t\t\t\t\t\t\t"; static const size_t PREFIX_CNT = sizeof(PREFIXES) - 1; @@ -323,8 +320,7 @@ static int btf_dump_add_emit_queue_id(struct btf_dump *d, __u32 id) if (d->emit_queue_cnt >= d->emit_queue_cap) { new_cap = max(16, d->emit_queue_cap * 3 / 2); - new_queue = realloc(d->emit_queue, - new_cap * sizeof(new_queue[0])); + new_queue = libbpf_reallocarray(d->emit_queue, new_cap, sizeof(new_queue[0])); if (!new_queue) return -ENOMEM; d->emit_queue = new_queue; @@ -1003,8 +999,7 @@ static int btf_dump_push_decl_stack_id(struct btf_dump *d, __u32 id) if (d->decl_stack_cnt >= d->decl_stack_cap) { new_cap = max(16, d->decl_stack_cap * 3 / 2); - new_stack = realloc(d->decl_stack, - new_cap * sizeof(new_stack[0])); + new_stack = libbpf_reallocarray(d->decl_stack, new_cap, sizeof(new_stack[0])); if (!new_stack) return -ENOMEM; d->decl_stack = new_stack; diff --git a/tools/lib/bpf/hashmap.c b/tools/lib/bpf/hashmap.c index a405dad068f5..3c20b126d60d 100644 --- a/tools/lib/bpf/hashmap.c +++ b/tools/lib/bpf/hashmap.c @@ -15,6 +15,9 @@ /* make sure libbpf doesn't use kernel-only integer typedefs */ #pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 +/* prevent accidental re-addition of reallocarray() */ +#pragma GCC poison reallocarray + /* start with 4 buckets */ #define HASHMAP_MIN_CAP_BITS 2 diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c index 0ad0b0491e1f..b688aadf09c5 100644 --- a/tools/lib/bpf/libbpf.c +++ b/tools/lib/bpf/libbpf.c @@ -44,7 +44,6 @@ #include <sys/vfs.h> #include <sys/utsname.h> #include <sys/resource.h> -#include <tools/libc_compat.h> #include <libelf.h> #include <gelf.h> #include <zlib.h> @@ -56,9 +55,6 @@ #include "libbpf_internal.h" #include "hashmap.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - #ifndef EM_BPF #define EM_BPF 247 #endif @@ -67,6 +63,8 @@ #define BPF_FS_MAGIC 0xcafe4a11 #endif +#define BPF_INSN_SZ (sizeof(struct bpf_insn)) + /* vsprintf() in __base_pr() uses nonliteral format string. It may break * compilation if user enables corresponding warning. Disable it explicitly. */ @@ -154,34 +152,35 @@ static void pr_perm_msg(int err) ___err; }) #endif -#ifdef HAVE_LIBELF_MMAP_SUPPORT -# define LIBBPF_ELF_C_READ_MMAP ELF_C_READ_MMAP -#else -# define LIBBPF_ELF_C_READ_MMAP ELF_C_READ -#endif - static inline __u64 ptr_to_u64(const void *ptr) { return (__u64) (unsigned long) ptr; } -struct bpf_capabilities { +enum kern_feature_id { /* v4.14: kernel support for program & map names. */ - __u32 name:1; + FEAT_PROG_NAME, /* v5.2: kernel support for global data sections. */ - __u32 global_data:1; + FEAT_GLOBAL_DATA, + /* BTF support */ + FEAT_BTF, /* BTF_KIND_FUNC and BTF_KIND_FUNC_PROTO support */ - __u32 btf_func:1; + FEAT_BTF_FUNC, /* BTF_KIND_VAR and BTF_KIND_DATASEC support */ - __u32 btf_datasec:1; - /* BPF_F_MMAPABLE is supported for arrays */ - __u32 array_mmap:1; + FEAT_BTF_DATASEC, /* BTF_FUNC_GLOBAL is supported */ - __u32 btf_func_global:1; + FEAT_BTF_GLOBAL_FUNC, + /* BPF_F_MMAPABLE is supported for arrays */ + FEAT_ARRAY_MMAP, /* kernel support for expected_attach_type in BPF_PROG_LOAD */ - __u32 exp_attach_type:1; + FEAT_EXP_ATTACH_TYPE, + /* bpf_probe_read_{kernel,user}[_str] helpers */ + FEAT_PROBE_READ_KERN, + __FEAT_CNT, }; +static bool kernel_supports(enum kern_feature_id feat_id); + enum reloc_type { RELO_LD64, RELO_CALL, @@ -209,6 +208,7 @@ struct bpf_sec_def { bool is_exp_attach_type_optional; bool is_attachable; bool is_attach_btf; + bool is_sleepable; attach_fn_t attach_fn; }; @@ -253,8 +253,6 @@ struct bpf_program { __u32 func_info_rec_size; __u32 func_info_cnt; - struct bpf_capabilities *caps; - void *line_info; __u32 line_info_rec_size; __u32 line_info_cnt; @@ -403,6 +401,7 @@ struct bpf_object { Elf_Data *rodata; Elf_Data *bss; Elf_Data *st_ops_data; + size_t shstrndx; /* section index for section name strings */ size_t strtabidx; struct { GElf_Shdr shdr; @@ -436,12 +435,18 @@ struct bpf_object { void *priv; bpf_object_clear_priv_t clear_priv; - struct bpf_capabilities caps; - char path[]; }; #define obj_elf_valid(o) ((o)->efile.elf) +static const char *elf_sym_str(const struct bpf_object *obj, size_t off); +static const char *elf_sec_str(const struct bpf_object *obj, size_t off); +static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx); +static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name); +static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr); +static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn); +static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn); + void bpf_program__unload(struct bpf_program *prog) { int i; @@ -503,7 +508,7 @@ static char *__bpf_program__pin_name(struct bpf_program *prog) } static int -bpf_program__init(void *data, size_t size, char *section_name, int idx, +bpf_program__init(void *data, size_t size, const char *section_name, int idx, struct bpf_program *prog) { const size_t bpf_insn_sz = sizeof(struct bpf_insn); @@ -552,7 +557,7 @@ errout: static int bpf_object__add_program(struct bpf_object *obj, void *data, size_t size, - char *section_name, int idx) + const char *section_name, int idx) { struct bpf_program prog, *progs; int nr_progs, err; @@ -561,11 +566,10 @@ bpf_object__add_program(struct bpf_object *obj, void *data, size_t size, if (err) return err; - prog.caps = &obj->caps; progs = obj->programs; nr_progs = obj->nr_programs; - progs = reallocarray(progs, nr_progs + 1, sizeof(progs[0])); + progs = libbpf_reallocarray(progs, nr_progs + 1, sizeof(progs[0])); if (!progs) { /* * In this case the original obj->programs @@ -578,7 +582,7 @@ bpf_object__add_program(struct bpf_object *obj, void *data, size_t size, return -ENOMEM; } - pr_debug("found program %s\n", prog.section_name); + pr_debug("elf: found program '%s'\n", prog.section_name); obj->programs = progs; obj->nr_programs = nr_progs + 1; prog.obj = obj; @@ -598,8 +602,7 @@ bpf_object__init_prog_names(struct bpf_object *obj) prog = &obj->programs[pi]; - for (si = 0; si < symbols->d_size / sizeof(GElf_Sym) && !name; - si++) { + for (si = 0; si < symbols->d_size / sizeof(GElf_Sym) && !name; si++) { GElf_Sym sym; if (!gelf_getsym(symbols, si, &sym)) @@ -609,11 +612,9 @@ bpf_object__init_prog_names(struct bpf_object *obj) if (GELF_ST_BIND(sym.st_info) != STB_GLOBAL) continue; - name = elf_strptr(obj->efile.elf, - obj->efile.strtabidx, - sym.st_name); + name = elf_sym_str(obj, sym.st_name); if (!name) { - pr_warn("failed to get sym name string for prog %s\n", + pr_warn("prog '%s': failed to get symbol name\n", prog->section_name); return -LIBBPF_ERRNO__LIBELF; } @@ -623,17 +624,14 @@ bpf_object__init_prog_names(struct bpf_object *obj) name = ".text"; if (!name) { - pr_warn("failed to find sym for prog %s\n", + pr_warn("prog '%s': failed to find program symbol\n", prog->section_name); return -EINVAL; } prog->name = strdup(name); - if (!prog->name) { - pr_warn("failed to allocate memory for prog sym %s\n", - name); + if (!prog->name) return -ENOMEM; - } } return 0; @@ -1066,13 +1064,18 @@ static void bpf_object__elf_finish(struct bpf_object *obj) obj->efile.obj_buf_sz = 0; } +/* if libelf is old and doesn't support mmap(), fall back to read() */ +#ifndef ELF_C_READ_MMAP +#define ELF_C_READ_MMAP ELF_C_READ +#endif + static int bpf_object__elf_init(struct bpf_object *obj) { int err = 0; GElf_Ehdr *ep; if (obj_elf_valid(obj)) { - pr_warn("elf init: internal error\n"); + pr_warn("elf: init internal error\n"); return -LIBBPF_ERRNO__LIBELF; } @@ -1090,31 +1093,44 @@ static int bpf_object__elf_init(struct bpf_object *obj) err = -errno; cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg)); - pr_warn("failed to open %s: %s\n", obj->path, cp); + pr_warn("elf: failed to open %s: %s\n", obj->path, cp); return err; } - obj->efile.elf = elf_begin(obj->efile.fd, - LIBBPF_ELF_C_READ_MMAP, NULL); + obj->efile.elf = elf_begin(obj->efile.fd, ELF_C_READ_MMAP, NULL); } if (!obj->efile.elf) { - pr_warn("failed to open %s as ELF file\n", obj->path); + pr_warn("elf: failed to open %s as ELF file: %s\n", obj->path, elf_errmsg(-1)); err = -LIBBPF_ERRNO__LIBELF; goto errout; } if (!gelf_getehdr(obj->efile.elf, &obj->efile.ehdr)) { - pr_warn("failed to get EHDR from %s\n", obj->path); + pr_warn("elf: failed to get ELF header from %s: %s\n", obj->path, elf_errmsg(-1)); err = -LIBBPF_ERRNO__FORMAT; goto errout; } ep = &obj->efile.ehdr; + if (elf_getshdrstrndx(obj->efile.elf, &obj->efile.shstrndx)) { + pr_warn("elf: failed to get section names section index for %s: %s\n", + obj->path, elf_errmsg(-1)); + err = -LIBBPF_ERRNO__FORMAT; + goto errout; + } + + /* Elf is corrupted/truncated, avoid calling elf_strptr. */ + if (!elf_rawdata(elf_getscn(obj->efile.elf, obj->efile.shstrndx), NULL)) { + pr_warn("elf: failed to get section names strings from %s: %s\n", + obj->path, elf_errmsg(-1)); + return -LIBBPF_ERRNO__FORMAT; + } + /* Old LLVM set e_machine to EM_NONE */ if (ep->e_type != ET_REL || (ep->e_machine && ep->e_machine != EM_BPF)) { - pr_warn("%s is not an eBPF object file\n", obj->path); + pr_warn("elf: %s is not a valid eBPF object file\n", obj->path); err = -LIBBPF_ERRNO__FORMAT; goto errout; } @@ -1136,7 +1152,7 @@ static int bpf_object__check_endianness(struct bpf_object *obj) #else # error "Unrecognized __BYTE_ORDER__" #endif - pr_warn("endianness mismatch.\n"); + pr_warn("elf: endianness mismatch in %s.\n", obj->path); return -LIBBPF_ERRNO__ENDIAN; } @@ -1171,55 +1187,10 @@ static bool bpf_map_type__is_map_in_map(enum bpf_map_type type) return false; } -static int bpf_object_search_section_size(const struct bpf_object *obj, - const char *name, size_t *d_size) -{ - const GElf_Ehdr *ep = &obj->efile.ehdr; - Elf *elf = obj->efile.elf; - Elf_Scn *scn = NULL; - int idx = 0; - - while ((scn = elf_nextscn(elf, scn)) != NULL) { - const char *sec_name; - Elf_Data *data; - GElf_Shdr sh; - - idx++; - if (gelf_getshdr(scn, &sh) != &sh) { - pr_warn("failed to get section(%d) header from %s\n", - idx, obj->path); - return -EIO; - } - - sec_name = elf_strptr(elf, ep->e_shstrndx, sh.sh_name); - if (!sec_name) { - pr_warn("failed to get section(%d) name from %s\n", - idx, obj->path); - return -EIO; - } - - if (strcmp(name, sec_name)) - continue; - - data = elf_getdata(scn, 0); - if (!data) { - pr_warn("failed to get section(%d) data from %s(%s)\n", - idx, name, obj->path); - return -EIO; - } - - *d_size = data->d_size; - return 0; - } - - return -ENOENT; -} - int bpf_object__section_size(const struct bpf_object *obj, const char *name, __u32 *size) { int ret = -ENOENT; - size_t d_size; *size = 0; if (!name) { @@ -1237,9 +1208,13 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name, if (obj->efile.st_ops_data) *size = obj->efile.st_ops_data->d_size; } else { - ret = bpf_object_search_section_size(obj, name, &d_size); - if (!ret) - *size = d_size; + Elf_Scn *scn = elf_sec_by_name(obj, name); + Elf_Data *data = elf_sec_data(obj, scn); + + if (data) { + ret = 0; /* found it */ + *size = data->d_size; + } } return *size ? 0 : ret; @@ -1264,8 +1239,7 @@ int bpf_object__variable_offset(const struct bpf_object *obj, const char *name, GELF_ST_TYPE(sym.st_info) != STT_OBJECT) continue; - sname = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name); + sname = elf_sym_str(obj, sym.st_name); if (!sname) { pr_warn("failed to get sym name string for var %s\n", name); @@ -1290,7 +1264,7 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj) return &obj->maps[obj->nr_maps++]; new_cap = max((size_t)4, obj->maps_cap * 3 / 2); - new_maps = realloc(obj->maps, new_cap * sizeof(*obj->maps)); + new_maps = libbpf_reallocarray(obj->maps, new_cap, sizeof(*obj->maps)); if (!new_maps) { pr_warn("alloc maps for object failed\n"); return ERR_PTR(-ENOMEM); @@ -1742,12 +1716,12 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) if (!symbols) return -EINVAL; - scn = elf_getscn(obj->efile.elf, obj->efile.maps_shndx); - if (scn) - data = elf_getdata(scn, NULL); + + scn = elf_sec_by_idx(obj, obj->efile.maps_shndx); + data = elf_sec_data(obj, scn); if (!scn || !data) { - pr_warn("failed to get Elf_Data from map section %d\n", - obj->efile.maps_shndx); + pr_warn("elf: failed to get legacy map definitions for %s\n", + obj->path); return -EINVAL; } @@ -1769,12 +1743,12 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) nr_maps++; } /* Assume equally sized map definitions */ - pr_debug("maps in %s: %d maps in %zd bytes\n", - obj->path, nr_maps, data->d_size); + pr_debug("elf: found %d legacy map definitions (%zd bytes) in %s\n", + nr_maps, data->d_size, obj->path); if (!data->d_size || nr_maps == 0 || (data->d_size % nr_maps) != 0) { - pr_warn("unable to determine map definition size section %s, %d maps in %zd bytes\n", - obj->path, nr_maps, data->d_size); + pr_warn("elf: unable to determine legacy map definition size in %s\n", + obj->path); return -EINVAL; } map_def_sz = data->d_size / nr_maps; @@ -1795,8 +1769,7 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict) if (IS_ERR(map)) return PTR_ERR(map); - map_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name); + map_name = elf_sym_str(obj, sym.st_name); if (!map_name) { pr_warn("failed to get map #%d name sym string for obj %s\n", i, obj->path); @@ -1884,6 +1857,29 @@ resolve_func_ptr(const struct btf *btf, __u32 id, __u32 *res_id) return btf_is_func_proto(t) ? t : NULL; } +static const char *btf_kind_str(const struct btf_type *t) +{ + switch (btf_kind(t)) { + case BTF_KIND_UNKN: return "void"; + case BTF_KIND_INT: return "int"; + case BTF_KIND_PTR: return "ptr"; + case BTF_KIND_ARRAY: return "array"; + case BTF_KIND_STRUCT: return "struct"; + case BTF_KIND_UNION: return "union"; + case BTF_KIND_ENUM: return "enum"; + case BTF_KIND_FWD: return "fwd"; + case BTF_KIND_TYPEDEF: return "typedef"; + case BTF_KIND_VOLATILE: return "volatile"; + case BTF_KIND_CONST: return "const"; + case BTF_KIND_RESTRICT: return "restrict"; + case BTF_KIND_FUNC: return "func"; + case BTF_KIND_FUNC_PROTO: return "func_proto"; + case BTF_KIND_VAR: return "var"; + case BTF_KIND_DATASEC: return "datasec"; + default: return "unknown"; + } +} + /* * Fetch integer attribute of BTF map definition. Such attributes are * represented using a pointer to an array, in which dimensionality of array @@ -1900,8 +1896,8 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf, const struct btf_type *arr_t; if (!btf_is_ptr(t)) { - pr_warn("map '%s': attr '%s': expected PTR, got %u.\n", - map_name, name, btf_kind(t)); + pr_warn("map '%s': attr '%s': expected PTR, got %s.\n", + map_name, name, btf_kind_str(t)); return false; } @@ -1912,8 +1908,8 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf, return false; } if (!btf_is_array(arr_t)) { - pr_warn("map '%s': attr '%s': expected ARRAY, got %u.\n", - map_name, name, btf_kind(arr_t)); + pr_warn("map '%s': attr '%s': expected ARRAY, got %s.\n", + map_name, name, btf_kind_str(arr_t)); return false; } arr_info = btf_array(arr_t); @@ -1924,7 +1920,7 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf, static int build_map_pin_path(struct bpf_map *map, const char *path) { char buf[PATH_MAX]; - int err, len; + int len; if (!path) path = "/sys/fs/bpf"; @@ -1935,11 +1931,7 @@ static int build_map_pin_path(struct bpf_map *map, const char *path) else if (len >= PATH_MAX) return -ENAMETOOLONG; - err = bpf_map__set_pin_path(map, buf); - if (err) - return err; - - return 0; + return bpf_map__set_pin_path(map, buf); } @@ -2007,8 +1999,8 @@ static int parse_btf_map_def(struct bpf_object *obj, return -EINVAL; } if (!btf_is_ptr(t)) { - pr_warn("map '%s': key spec is not PTR: %u.\n", - map->name, btf_kind(t)); + pr_warn("map '%s': key spec is not PTR: %s.\n", + map->name, btf_kind_str(t)); return -EINVAL; } sz = btf__resolve_size(obj->btf, t->type); @@ -2049,8 +2041,8 @@ static int parse_btf_map_def(struct bpf_object *obj, return -EINVAL; } if (!btf_is_ptr(t)) { - pr_warn("map '%s': value spec is not PTR: %u.\n", - map->name, btf_kind(t)); + pr_warn("map '%s': value spec is not PTR: %s.\n", + map->name, btf_kind_str(t)); return -EINVAL; } sz = btf__resolve_size(obj->btf, t->type); @@ -2107,14 +2099,14 @@ static int parse_btf_map_def(struct bpf_object *obj, t = skip_mods_and_typedefs(obj->btf, btf_array(t)->type, NULL); if (!btf_is_ptr(t)) { - pr_warn("map '%s': map-in-map inner def is of unexpected kind %u.\n", - map->name, btf_kind(t)); + pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n", + map->name, btf_kind_str(t)); return -EINVAL; } t = skip_mods_and_typedefs(obj->btf, t->type, NULL); if (!btf_is_struct(t)) { - pr_warn("map '%s': map-in-map inner def is of unexpected kind %u.\n", - map->name, btf_kind(t)); + pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n", + map->name, btf_kind_str(t)); return -EINVAL; } @@ -2205,8 +2197,8 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj, return -EINVAL; } if (!btf_is_var(var)) { - pr_warn("map '%s': unexpected var kind %u.\n", - map_name, btf_kind(var)); + pr_warn("map '%s': unexpected var kind %s.\n", + map_name, btf_kind_str(var)); return -EINVAL; } if (var_extra->linkage != BTF_VAR_GLOBAL_ALLOCATED && @@ -2218,8 +2210,8 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj, def = skip_mods_and_typedefs(obj->btf, var->type, NULL); if (!btf_is_struct(def)) { - pr_warn("map '%s': unexpected def kind %u.\n", - map_name, btf_kind(var)); + pr_warn("map '%s': unexpected def kind %s.\n", + map_name, btf_kind_str(var)); return -EINVAL; } if (def->size > vi->size) { @@ -2259,12 +2251,11 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict, if (obj->efile.btf_maps_shndx < 0) return 0; - scn = elf_getscn(obj->efile.elf, obj->efile.btf_maps_shndx); - if (scn) - data = elf_getdata(scn, NULL); + scn = elf_sec_by_idx(obj, obj->efile.btf_maps_shndx); + data = elf_sec_data(obj, scn); if (!scn || !data) { - pr_warn("failed to get Elf_Data from map section %d (%s)\n", - obj->efile.btf_maps_shndx, MAPS_ELF_SEC); + pr_warn("elf: failed to get %s map definitions for %s\n", + MAPS_ELF_SEC, obj->path); return -EINVAL; } @@ -2322,36 +2313,28 @@ static int bpf_object__init_maps(struct bpf_object *obj, static bool section_have_execinstr(struct bpf_object *obj, int idx) { - Elf_Scn *scn; GElf_Shdr sh; - scn = elf_getscn(obj->efile.elf, idx); - if (!scn) - return false; - - if (gelf_getshdr(scn, &sh) != &sh) + if (elf_sec_hdr(obj, elf_sec_by_idx(obj, idx), &sh)) return false; - if (sh.sh_flags & SHF_EXECINSTR) - return true; - - return false; + return sh.sh_flags & SHF_EXECINSTR; } static bool btf_needs_sanitization(struct bpf_object *obj) { - bool has_func_global = obj->caps.btf_func_global; - bool has_datasec = obj->caps.btf_datasec; - bool has_func = obj->caps.btf_func; + bool has_func_global = kernel_supports(FEAT_BTF_GLOBAL_FUNC); + bool has_datasec = kernel_supports(FEAT_BTF_DATASEC); + bool has_func = kernel_supports(FEAT_BTF_FUNC); return !has_func || !has_datasec || !has_func_global; } static void bpf_object__sanitize_btf(struct bpf_object *obj, struct btf *btf) { - bool has_func_global = obj->caps.btf_func_global; - bool has_datasec = obj->caps.btf_datasec; - bool has_func = obj->caps.btf_func; + bool has_func_global = kernel_supports(FEAT_BTF_GLOBAL_FUNC); + bool has_datasec = kernel_supports(FEAT_BTF_DATASEC); + bool has_func = kernel_supports(FEAT_BTF_FUNC); struct btf_type *t; int i, j, vlen; @@ -2499,7 +2482,7 @@ static int bpf_object__load_vmlinux_btf(struct bpf_object *obj) int err; /* CO-RE relocations need kernel BTF */ - if (obj->btf_ext && obj->btf_ext->field_reloc_info.len) + if (obj->btf_ext && obj->btf_ext->core_relo_info.len) need_vmlinux_btf = true; bpf_object__for_each_program(prog, obj) { @@ -2533,6 +2516,15 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) if (!obj->btf) return 0; + if (!kernel_supports(FEAT_BTF)) { + if (kernel_needs_btf(obj)) { + err = -EOPNOTSUPP; + goto report; + } + pr_debug("Kernel doesn't support BTF, skipping uploading it.\n"); + return 0; + } + sanitize = btf_needs_sanitization(obj); if (sanitize) { const void *raw_data; @@ -2558,6 +2550,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) } btf__free(kern_btf); } +report: if (err) { btf_mandatory = kernel_needs_btf(obj); pr_warn("Error loading .BTF into kernel: %d. %s\n", err, @@ -2569,61 +2562,199 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj) return err; } +static const char *elf_sym_str(const struct bpf_object *obj, size_t off) +{ + const char *name; + + name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, off); + if (!name) { + pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n", + off, obj->path, elf_errmsg(-1)); + return NULL; + } + + return name; +} + +static const char *elf_sec_str(const struct bpf_object *obj, size_t off) +{ + const char *name; + + name = elf_strptr(obj->efile.elf, obj->efile.shstrndx, off); + if (!name) { + pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n", + off, obj->path, elf_errmsg(-1)); + return NULL; + } + + return name; +} + +static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx) +{ + Elf_Scn *scn; + + scn = elf_getscn(obj->efile.elf, idx); + if (!scn) { + pr_warn("elf: failed to get section(%zu) from %s: %s\n", + idx, obj->path, elf_errmsg(-1)); + return NULL; + } + return scn; +} + +static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name) +{ + Elf_Scn *scn = NULL; + Elf *elf = obj->efile.elf; + const char *sec_name; + + while ((scn = elf_nextscn(elf, scn)) != NULL) { + sec_name = elf_sec_name(obj, scn); + if (!sec_name) + return NULL; + + if (strcmp(sec_name, name) != 0) + continue; + + return scn; + } + return NULL; +} + +static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr) +{ + if (!scn) + return -EINVAL; + + if (gelf_getshdr(scn, hdr) != hdr) { + pr_warn("elf: failed to get section(%zu) header from %s: %s\n", + elf_ndxscn(scn), obj->path, elf_errmsg(-1)); + return -EINVAL; + } + + return 0; +} + +static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn) +{ + const char *name; + GElf_Shdr sh; + + if (!scn) + return NULL; + + if (elf_sec_hdr(obj, scn, &sh)) + return NULL; + + name = elf_sec_str(obj, sh.sh_name); + if (!name) { + pr_warn("elf: failed to get section(%zu) name from %s: %s\n", + elf_ndxscn(scn), obj->path, elf_errmsg(-1)); + return NULL; + } + + return name; +} + +static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn) +{ + Elf_Data *data; + + if (!scn) + return NULL; + + data = elf_getdata(scn, 0); + if (!data) { + pr_warn("elf: failed to get section(%zu) %s data from %s: %s\n", + elf_ndxscn(scn), elf_sec_name(obj, scn) ?: "<?>", + obj->path, elf_errmsg(-1)); + return NULL; + } + + return data; +} + +static bool is_sec_name_dwarf(const char *name) +{ + /* approximation, but the actual list is too long */ + return strncmp(name, ".debug_", sizeof(".debug_") - 1) == 0; +} + +static bool ignore_elf_section(GElf_Shdr *hdr, const char *name) +{ + /* no special handling of .strtab */ + if (hdr->sh_type == SHT_STRTAB) + return true; + + /* ignore .llvm_addrsig section as well */ + if (hdr->sh_type == 0x6FFF4C03 /* SHT_LLVM_ADDRSIG */) + return true; + + /* no subprograms will lead to an empty .text section, ignore it */ + if (hdr->sh_type == SHT_PROGBITS && hdr->sh_size == 0 && + strcmp(name, ".text") == 0) + return true; + + /* DWARF sections */ + if (is_sec_name_dwarf(name)) + return true; + + if (strncmp(name, ".rel", sizeof(".rel") - 1) == 0) { + name += sizeof(".rel") - 1; + /* DWARF section relocations */ + if (is_sec_name_dwarf(name)) + return true; + + /* .BTF and .BTF.ext don't need relocations */ + if (strcmp(name, BTF_ELF_SEC) == 0 || + strcmp(name, BTF_EXT_ELF_SEC) == 0) + return true; + } + + return false; +} + static int bpf_object__elf_collect(struct bpf_object *obj) { Elf *elf = obj->efile.elf; - GElf_Ehdr *ep = &obj->efile.ehdr; Elf_Data *btf_ext_data = NULL; Elf_Data *btf_data = NULL; Elf_Scn *scn = NULL; int idx = 0, err = 0; - /* Elf is corrupted/truncated, avoid calling elf_strptr. */ - if (!elf_rawdata(elf_getscn(elf, ep->e_shstrndx), NULL)) { - pr_warn("failed to get e_shstrndx from %s\n", obj->path); - return -LIBBPF_ERRNO__FORMAT; - } - while ((scn = elf_nextscn(elf, scn)) != NULL) { - char *name; + const char *name; GElf_Shdr sh; Elf_Data *data; idx++; - if (gelf_getshdr(scn, &sh) != &sh) { - pr_warn("failed to get section(%d) header from %s\n", - idx, obj->path); + + if (elf_sec_hdr(obj, scn, &sh)) return -LIBBPF_ERRNO__FORMAT; - } - name = elf_strptr(elf, ep->e_shstrndx, sh.sh_name); - if (!name) { - pr_warn("failed to get section(%d) name from %s\n", - idx, obj->path); + name = elf_sec_str(obj, sh.sh_name); + if (!name) return -LIBBPF_ERRNO__FORMAT; - } - data = elf_getdata(scn, 0); - if (!data) { - pr_warn("failed to get section(%d) data from %s(%s)\n", - idx, name, obj->path); + if (ignore_elf_section(&sh, name)) + continue; + + data = elf_sec_data(obj, scn); + if (!data) return -LIBBPF_ERRNO__FORMAT; - } - pr_debug("section(%d) %s, size %ld, link %d, flags %lx, type=%d\n", + + pr_debug("elf: section(%d) %s, size %ld, link %d, flags %lx, type=%d\n", idx, name, (unsigned long)data->d_size, (int)sh.sh_link, (unsigned long)sh.sh_flags, (int)sh.sh_type); if (strcmp(name, "license") == 0) { - err = bpf_object__init_license(obj, - data->d_buf, - data->d_size); + err = bpf_object__init_license(obj, data->d_buf, data->d_size); if (err) return err; } else if (strcmp(name, "version") == 0) { - err = bpf_object__init_kversion(obj, - data->d_buf, - data->d_size); + err = bpf_object__init_kversion(obj, data->d_buf, data->d_size); if (err) return err; } else if (strcmp(name, "maps") == 0) { @@ -2636,8 +2767,7 @@ static int bpf_object__elf_collect(struct bpf_object *obj) btf_ext_data = data; } else if (sh.sh_type == SHT_SYMTAB) { if (obj->efile.symbols) { - pr_warn("bpf: multiple SYMTAB in %s\n", - obj->path); + pr_warn("elf: multiple symbol tables in %s\n", obj->path); return -LIBBPF_ERRNO__FORMAT; } obj->efile.symbols = data; @@ -2650,16 +2780,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj) err = bpf_object__add_program(obj, data->d_buf, data->d_size, name, idx); - if (err) { - char errmsg[STRERR_BUFSIZE]; - char *cp; - - cp = libbpf_strerror_r(-err, errmsg, - sizeof(errmsg)); - pr_warn("failed to alloc program %s (%s): %s", - name, obj->path, cp); + if (err) return err; - } } else if (strcmp(name, DATA_SEC) == 0) { obj->efile.data = data; obj->efile.data_shndx = idx; @@ -2670,7 +2792,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj) obj->efile.st_ops_data = data; obj->efile.st_ops_shndx = idx; } else { - pr_debug("skip section(%d) %s\n", idx, name); + pr_info("elf: skipping unrecognized data section(%d) %s\n", + idx, name); } } else if (sh.sh_type == SHT_REL) { int nr_sects = obj->efile.nr_reloc_sects; @@ -2681,34 +2804,33 @@ static int bpf_object__elf_collect(struct bpf_object *obj) if (!section_have_execinstr(obj, sec) && strcmp(name, ".rel" STRUCT_OPS_SEC) && strcmp(name, ".rel" MAPS_ELF_SEC)) { - pr_debug("skip relo %s(%d) for section(%d)\n", - name, idx, sec); + pr_info("elf: skipping relo section(%d) %s for section(%d) %s\n", + idx, name, sec, + elf_sec_name(obj, elf_sec_by_idx(obj, sec)) ?: "<?>"); continue; } - sects = reallocarray(sects, nr_sects + 1, - sizeof(*obj->efile.reloc_sects)); - if (!sects) { - pr_warn("reloc_sects realloc failed\n"); + sects = libbpf_reallocarray(sects, nr_sects + 1, + sizeof(*obj->efile.reloc_sects)); + if (!sects) return -ENOMEM; - } obj->efile.reloc_sects = sects; obj->efile.nr_reloc_sects++; obj->efile.reloc_sects[nr_sects].shdr = sh; obj->efile.reloc_sects[nr_sects].data = data; - } else if (sh.sh_type == SHT_NOBITS && - strcmp(name, BSS_SEC) == 0) { + } else if (sh.sh_type == SHT_NOBITS && strcmp(name, BSS_SEC) == 0) { obj->efile.bss = data; obj->efile.bss_shndx = idx; } else { - pr_debug("skip section(%d) %s\n", idx, name); + pr_info("elf: skipping section(%d) %s (size %zu)\n", idx, name, + (size_t)sh.sh_size); } } if (!obj->efile.strtabidx || obj->efile.strtabidx > idx) { - pr_warn("Corrupted ELF file: index of strtab invalid\n"); + pr_warn("elf: symbol strings section missing or invalid in %s\n", obj->path); return -LIBBPF_ERRNO__FORMAT; } return bpf_object__init_btf(obj, btf_data, btf_ext_data); @@ -2869,14 +2991,13 @@ static int bpf_object__collect_externs(struct bpf_object *obj) if (!obj->efile.symbols) return 0; - scn = elf_getscn(obj->efile.elf, obj->efile.symbols_shndx); - if (!scn) + scn = elf_sec_by_idx(obj, obj->efile.symbols_shndx); + if (elf_sec_hdr(obj, scn, &sh)) return -LIBBPF_ERRNO__FORMAT; - if (gelf_getshdr(scn, &sh) != &sh) - return -LIBBPF_ERRNO__FORMAT; - n = sh.sh_size / sh.sh_entsize; + n = sh.sh_size / sh.sh_entsize; pr_debug("looking for externs among %d symbols...\n", n); + for (i = 0; i < n; i++) { GElf_Sym sym; @@ -2884,13 +3005,12 @@ static int bpf_object__collect_externs(struct bpf_object *obj) return -LIBBPF_ERRNO__FORMAT; if (!sym_is_extern(&sym)) continue; - ext_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name); + ext_name = elf_sym_str(obj, sym.st_name); if (!ext_name || !ext_name[0]) continue; ext = obj->externs; - ext = reallocarray(ext, obj->nr_extern + 1, sizeof(*ext)); + ext = libbpf_reallocarray(ext, obj->nr_extern + 1, sizeof(*ext)); if (!ext) return -ENOMEM; obj->externs = ext; @@ -3109,7 +3229,7 @@ bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx) static int bpf_program__record_reloc(struct bpf_program *prog, struct reloc_desc *reloc_desc, - __u32 insn_idx, const char *name, + __u32 insn_idx, const char *sym_name, const GElf_Sym *sym, const GElf_Rel *rel) { struct bpf_insn *insn = &prog->insns[insn_idx]; @@ -3117,22 +3237,25 @@ static int bpf_program__record_reloc(struct bpf_program *prog, struct bpf_object *obj = prog->obj; __u32 shdr_idx = sym->st_shndx; enum libbpf_map_type type; + const char *sym_sec_name; struct bpf_map *map; /* sub-program call relocation */ if (insn->code == (BPF_JMP | BPF_CALL)) { if (insn->src_reg != BPF_PSEUDO_CALL) { - pr_warn("incorrect bpf_call opcode\n"); + pr_warn("prog '%s': incorrect bpf_call opcode\n", prog->name); return -LIBBPF_ERRNO__RELOC; } /* text_shndx can be 0, if no default "main" program exists */ if (!shdr_idx || shdr_idx != obj->efile.text_shndx) { - pr_warn("bad call relo against section %u\n", shdr_idx); + sym_sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, shdr_idx)); + pr_warn("prog '%s': bad call relo against '%s' in section '%s'\n", + prog->name, sym_name, sym_sec_name); return -LIBBPF_ERRNO__RELOC; } - if (sym->st_value % 8) { - pr_warn("bad call relo offset: %zu\n", - (size_t)sym->st_value); + if (sym->st_value % BPF_INSN_SZ) { + pr_warn("prog '%s': bad call relo against '%s' at offset %zu\n", + prog->name, sym_name, (size_t)sym->st_value); return -LIBBPF_ERRNO__RELOC; } reloc_desc->type = RELO_CALL; @@ -3143,8 +3266,8 @@ static int bpf_program__record_reloc(struct bpf_program *prog, } if (insn->code != (BPF_LD | BPF_IMM | BPF_DW)) { - pr_warn("invalid relo for insns[%d].code 0x%x\n", - insn_idx, insn->code); + pr_warn("prog '%s': invalid relo against '%s' for insns[%d].code 0x%x\n", + prog->name, sym_name, insn_idx, insn->code); return -LIBBPF_ERRNO__RELOC; } @@ -3159,12 +3282,12 @@ static int bpf_program__record_reloc(struct bpf_program *prog, break; } if (i >= n) { - pr_warn("extern relo failed to find extern for sym %d\n", - sym_idx); + pr_warn("prog '%s': extern relo failed to find extern for '%s' (%d)\n", + prog->name, sym_name, sym_idx); return -LIBBPF_ERRNO__RELOC; } - pr_debug("found extern #%d '%s' (sym %d) for insn %u\n", - i, ext->name, ext->sym_idx, insn_idx); + pr_debug("prog '%s': found extern #%d '%s' (sym %d) for insn #%u\n", + prog->name, i, ext->name, ext->sym_idx, insn_idx); reloc_desc->type = RELO_EXTERN; reloc_desc->insn_idx = insn_idx; reloc_desc->sym_off = i; /* sym_off stores extern index */ @@ -3172,18 +3295,19 @@ static int bpf_program__record_reloc(struct bpf_program *prog, } if (!shdr_idx || shdr_idx >= SHN_LORESERVE) { - pr_warn("invalid relo for \'%s\' in special section 0x%x; forgot to initialize global var?..\n", - name, shdr_idx); + pr_warn("prog '%s': invalid relo against '%s' in special section 0x%x; forgot to initialize global var?..\n", + prog->name, sym_name, shdr_idx); return -LIBBPF_ERRNO__RELOC; } type = bpf_object__section_to_libbpf_map_type(obj, shdr_idx); + sym_sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, shdr_idx)); /* generic map reference relocation */ if (type == LIBBPF_MAP_UNSPEC) { if (!bpf_object__shndx_is_maps(obj, shdr_idx)) { - pr_warn("bad map relo against section %u\n", - shdr_idx); + pr_warn("prog '%s': bad map relo against '%s' in section '%s'\n", + prog->name, sym_name, sym_sec_name); return -LIBBPF_ERRNO__RELOC; } for (map_idx = 0; map_idx < nr_maps; map_idx++) { @@ -3192,14 +3316,14 @@ static int bpf_program__record_reloc(struct bpf_program *prog, map->sec_idx != sym->st_shndx || map->sec_offset != sym->st_value) continue; - pr_debug("found map %zd (%s, sec %d, off %zu) for insn %u\n", - map_idx, map->name, map->sec_idx, + pr_debug("prog '%s': found map %zd (%s, sec %d, off %zu) for insn #%u\n", + prog->name, map_idx, map->name, map->sec_idx, map->sec_offset, insn_idx); break; } if (map_idx >= nr_maps) { - pr_warn("map relo failed to find map for sec %u, off %zu\n", - shdr_idx, (size_t)sym->st_value); + pr_warn("prog '%s': map relo failed to find map for section '%s', off %zu\n", + prog->name, sym_sec_name, (size_t)sym->st_value); return -LIBBPF_ERRNO__RELOC; } reloc_desc->type = RELO_LD64; @@ -3211,21 +3335,22 @@ static int bpf_program__record_reloc(struct bpf_program *prog, /* global data map relocation */ if (!bpf_object__shndx_is_data(obj, shdr_idx)) { - pr_warn("bad data relo against section %u\n", shdr_idx); + pr_warn("prog '%s': bad data relo against section '%s'\n", + prog->name, sym_sec_name); return -LIBBPF_ERRNO__RELOC; } for (map_idx = 0; map_idx < nr_maps; map_idx++) { map = &obj->maps[map_idx]; if (map->libbpf_type != type) continue; - pr_debug("found data map %zd (%s, sec %d, off %zu) for insn %u\n", - map_idx, map->name, map->sec_idx, map->sec_offset, - insn_idx); + pr_debug("prog '%s': found data map %zd (%s, sec %d, off %zu) for insn %u\n", + prog->name, map_idx, map->name, map->sec_idx, + map->sec_offset, insn_idx); break; } if (map_idx >= nr_maps) { - pr_warn("data relo failed to find map for sec %u\n", - shdr_idx); + pr_warn("prog '%s': data relo failed to find map for section '%s'\n", + prog->name, sym_sec_name); return -LIBBPF_ERRNO__RELOC; } @@ -3241,9 +3366,17 @@ bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr, Elf_Data *data, struct bpf_object *obj) { Elf_Data *symbols = obj->efile.symbols; + const char *relo_sec_name, *sec_name; + size_t sec_idx = shdr->sh_info; int err, i, nrels; - pr_debug("collecting relocating info for: '%s'\n", prog->section_name); + relo_sec_name = elf_sec_str(obj, shdr->sh_name); + sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, sec_idx)); + if (!relo_sec_name || !sec_name) + return -EINVAL; + + pr_debug("sec '%s': collecting relocation for section(%zu) '%s'\n", + relo_sec_name, sec_idx, sec_name); nrels = shdr->sh_size / shdr->sh_entsize; prog->reloc_desc = malloc(sizeof(*prog->reloc_desc) * nrels); @@ -3254,35 +3387,34 @@ bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr, prog->nr_reloc = nrels; for (i = 0; i < nrels; i++) { - const char *name; + const char *sym_name; __u32 insn_idx; GElf_Sym sym; GElf_Rel rel; if (!gelf_getrel(data, i, &rel)) { - pr_warn("relocation: failed to get %d reloc\n", i); + pr_warn("sec '%s': failed to get relo #%d\n", relo_sec_name, i); return -LIBBPF_ERRNO__FORMAT; } if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) { - pr_warn("relocation: symbol %"PRIx64" not found\n", - GELF_R_SYM(rel.r_info)); + pr_warn("sec '%s': symbol 0x%zx not found for relo #%d\n", + relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i); return -LIBBPF_ERRNO__FORMAT; } - if (rel.r_offset % sizeof(struct bpf_insn)) + if (rel.r_offset % BPF_INSN_SZ) { + pr_warn("sec '%s': invalid offset 0x%zx for relo #%d\n", + relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i); return -LIBBPF_ERRNO__FORMAT; + } - insn_idx = rel.r_offset / sizeof(struct bpf_insn); - name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name) ? : "<?>"; + insn_idx = rel.r_offset / BPF_INSN_SZ; + sym_name = elf_sym_str(obj, sym.st_name) ?: "<?>"; - pr_debug("relo for shdr %u, symb %zu, value %zu, type %d, bind %d, name %d (\'%s\'), insn %u\n", - (__u32)sym.st_shndx, (size_t)GELF_R_SYM(rel.r_info), - (size_t)sym.st_value, GELF_ST_TYPE(sym.st_info), - GELF_ST_BIND(sym.st_info), sym.st_name, name, - insn_idx); + pr_debug("sec '%s': relo #%d: insn #%u against '%s'\n", + relo_sec_name, i, insn_idx, sym_name); err = bpf_program__record_reloc(prog, &prog->reloc_desc[i], - insn_idx, name, &sym, &rel); + insn_idx, sym_name, &sym, &rel); if (err) return err; } @@ -3433,8 +3565,14 @@ bpf_object__probe_loading(struct bpf_object *obj) return 0; } -static int -bpf_object__probe_name(struct bpf_object *obj) +static int probe_fd(int fd) +{ + if (fd >= 0) + close(fd); + return fd >= 0; +} + +static int probe_kern_prog_name(void) { struct bpf_load_program_attr attr; struct bpf_insn insns[] = { @@ -3452,16 +3590,10 @@ bpf_object__probe_name(struct bpf_object *obj) attr.license = "GPL"; attr.name = "test"; ret = bpf_load_program_xattr(&attr, NULL, 0); - if (ret >= 0) { - obj->caps.name = 1; - close(ret); - } - - return 0; + return probe_fd(ret); } -static int -bpf_object__probe_global_data(struct bpf_object *obj) +static int probe_kern_global_data(void) { struct bpf_load_program_attr prg_attr; struct bpf_create_map_attr map_attr; @@ -3498,16 +3630,23 @@ bpf_object__probe_global_data(struct bpf_object *obj) prg_attr.license = "GPL"; ret = bpf_load_program_xattr(&prg_attr, NULL, 0); - if (ret >= 0) { - obj->caps.global_data = 1; - close(ret); - } - close(map); - return 0; + return probe_fd(ret); +} + +static int probe_kern_btf(void) +{ + static const char strs[] = "\0int"; + __u32 types[] = { + /* int */ + BTF_TYPE_INT_ENC(1, BTF_INT_SIGNED, 0, 32, 4), + }; + + return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types), + strs, sizeof(strs))); } -static int bpf_object__probe_btf_func(struct bpf_object *obj) +static int probe_kern_btf_func(void) { static const char strs[] = "\0int\0x\0a"; /* void x(int a) {} */ @@ -3520,20 +3659,12 @@ static int bpf_object__probe_btf_func(struct bpf_object *obj) /* FUNC x */ /* [3] */ BTF_TYPE_ENC(5, BTF_INFO_ENC(BTF_KIND_FUNC, 0, 0), 2), }; - int btf_fd; - btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types), - strs, sizeof(strs)); - if (btf_fd >= 0) { - obj->caps.btf_func = 1; - close(btf_fd); - return 1; - } - - return 0; + return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types), + strs, sizeof(strs))); } -static int bpf_object__probe_btf_func_global(struct bpf_object *obj) +static int probe_kern_btf_func_global(void) { static const char strs[] = "\0int\0x\0a"; /* static void x(int a) {} */ @@ -3546,20 +3677,12 @@ static int bpf_object__probe_btf_func_global(struct bpf_object *obj) /* FUNC x BTF_FUNC_GLOBAL */ /* [3] */ BTF_TYPE_ENC(5, BTF_INFO_ENC(BTF_KIND_FUNC, 0, BTF_FUNC_GLOBAL), 2), }; - int btf_fd; - btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types), - strs, sizeof(strs)); - if (btf_fd >= 0) { - obj->caps.btf_func_global = 1; - close(btf_fd); - return 1; - } - - return 0; + return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types), + strs, sizeof(strs))); } -static int bpf_object__probe_btf_datasec(struct bpf_object *obj) +static int probe_kern_btf_datasec(void) { static const char strs[] = "\0x\0.data"; /* static int a; */ @@ -3573,20 +3696,12 @@ static int bpf_object__probe_btf_datasec(struct bpf_object *obj) BTF_TYPE_ENC(3, BTF_INFO_ENC(BTF_KIND_DATASEC, 0, 1), 4), BTF_VAR_SECINFO_ENC(2, 0, 4), }; - int btf_fd; - - btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types), - strs, sizeof(strs)); - if (btf_fd >= 0) { - obj->caps.btf_datasec = 1; - close(btf_fd); - return 1; - } - return 0; + return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types), + strs, sizeof(strs))); } -static int bpf_object__probe_array_mmap(struct bpf_object *obj) +static int probe_kern_array_mmap(void) { struct bpf_create_map_attr attr = { .map_type = BPF_MAP_TYPE_ARRAY, @@ -3595,27 +3710,17 @@ static int bpf_object__probe_array_mmap(struct bpf_object *obj) .value_size = sizeof(int), .max_entries = 1, }; - int fd; - fd = bpf_create_map_xattr(&attr); - if (fd >= 0) { - obj->caps.array_mmap = 1; - close(fd); - return 1; - } - - return 0; + return probe_fd(bpf_create_map_xattr(&attr)); } -static int -bpf_object__probe_exp_attach_type(struct bpf_object *obj) +static int probe_kern_exp_attach_type(void) { struct bpf_load_program_attr attr; struct bpf_insn insns[] = { BPF_MOV64_IMM(BPF_REG_0, 0), BPF_EXIT_INSN(), }; - int fd; memset(&attr, 0, sizeof(attr)); /* use any valid combination of program type and (optional) @@ -3629,36 +3734,91 @@ bpf_object__probe_exp_attach_type(struct bpf_object *obj) attr.insns_cnt = ARRAY_SIZE(insns); attr.license = "GPL"; - fd = bpf_load_program_xattr(&attr, NULL, 0); - if (fd >= 0) { - obj->caps.exp_attach_type = 1; - close(fd); - return 1; - } - return 0; + return probe_fd(bpf_load_program_xattr(&attr, NULL, 0)); } -static int -bpf_object__probe_caps(struct bpf_object *obj) -{ - int (*probe_fn[])(struct bpf_object *obj) = { - bpf_object__probe_name, - bpf_object__probe_global_data, - bpf_object__probe_btf_func, - bpf_object__probe_btf_func_global, - bpf_object__probe_btf_datasec, - bpf_object__probe_array_mmap, - bpf_object__probe_exp_attach_type, +static int probe_kern_probe_read_kernel(void) +{ + struct bpf_load_program_attr attr; + struct bpf_insn insns[] = { + BPF_MOV64_REG(BPF_REG_1, BPF_REG_10), /* r1 = r10 (fp) */ + BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8), /* r1 += -8 */ + BPF_MOV64_IMM(BPF_REG_2, 8), /* r2 = 8 */ + BPF_MOV64_IMM(BPF_REG_3, 0), /* r3 = 0 */ + BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_probe_read_kernel), + BPF_EXIT_INSN(), }; - int i, ret; - for (i = 0; i < ARRAY_SIZE(probe_fn); i++) { - ret = probe_fn[i](obj); - if (ret < 0) - pr_debug("Probe #%d failed with %d.\n", i, ret); + memset(&attr, 0, sizeof(attr)); + attr.prog_type = BPF_PROG_TYPE_KPROBE; + attr.insns = insns; + attr.insns_cnt = ARRAY_SIZE(insns); + attr.license = "GPL"; + + return probe_fd(bpf_load_program_xattr(&attr, NULL, 0)); +} + +enum kern_feature_result { + FEAT_UNKNOWN = 0, + FEAT_SUPPORTED = 1, + FEAT_MISSING = 2, +}; + +typedef int (*feature_probe_fn)(void); + +static struct kern_feature_desc { + const char *desc; + feature_probe_fn probe; + enum kern_feature_result res; +} feature_probes[__FEAT_CNT] = { + [FEAT_PROG_NAME] = { + "BPF program name", probe_kern_prog_name, + }, + [FEAT_GLOBAL_DATA] = { + "global variables", probe_kern_global_data, + }, + [FEAT_BTF] = { + "minimal BTF", probe_kern_btf, + }, + [FEAT_BTF_FUNC] = { + "BTF functions", probe_kern_btf_func, + }, + [FEAT_BTF_GLOBAL_FUNC] = { + "BTF global function", probe_kern_btf_func_global, + }, + [FEAT_BTF_DATASEC] = { + "BTF data section and variable", probe_kern_btf_datasec, + }, + [FEAT_ARRAY_MMAP] = { + "ARRAY map mmap()", probe_kern_array_mmap, + }, + [FEAT_EXP_ATTACH_TYPE] = { + "BPF_PROG_LOAD expected_attach_type attribute", + probe_kern_exp_attach_type, + }, + [FEAT_PROBE_READ_KERN] = { + "bpf_probe_read_kernel() helper", probe_kern_probe_read_kernel, } +}; - return 0; +static bool kernel_supports(enum kern_feature_id feat_id) +{ + struct kern_feature_desc *feat = &feature_probes[feat_id]; + int ret; + + if (READ_ONCE(feat->res) == FEAT_UNKNOWN) { + ret = feat->probe(); + if (ret > 0) { + WRITE_ONCE(feat->res, FEAT_SUPPORTED); + } else if (ret == 0) { + WRITE_ONCE(feat->res, FEAT_MISSING); + } else { + pr_warn("Detection of kernel %s support failed: %d\n", feat->desc, ret); + WRITE_ONCE(feat->res, FEAT_MISSING); + } + } + + return READ_ONCE(feat->res) == FEAT_SUPPORTED; } static bool map_is_reuse_compat(const struct bpf_map *map, int map_fd) @@ -3760,7 +3920,7 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map) memset(&create_attr, 0, sizeof(create_attr)); - if (obj->caps.name) + if (kernel_supports(FEAT_PROG_NAME)) create_attr.name = map->name; create_attr.map_ifindex = map->map_ifindex; create_attr.map_type = def->type; @@ -4011,6 +4171,10 @@ struct bpf_core_spec { const struct btf *btf; /* high-level spec: named fields and array indices only */ struct bpf_core_accessor spec[BPF_CORE_SPEC_MAX_LEN]; + /* original unresolved (no skip_mods_or_typedefs) root type ID */ + __u32 root_type_id; + /* CO-RE relocation kind */ + enum bpf_core_relo_kind relo_kind; /* high-level spec length */ int len; /* raw, low-level spec: 1-to-1 with accessor spec string */ @@ -4041,8 +4205,66 @@ static bool is_flex_arr(const struct btf *btf, return acc->idx == btf_vlen(t) - 1; } +static const char *core_relo_kind_str(enum bpf_core_relo_kind kind) +{ + switch (kind) { + case BPF_FIELD_BYTE_OFFSET: return "byte_off"; + case BPF_FIELD_BYTE_SIZE: return "byte_sz"; + case BPF_FIELD_EXISTS: return "field_exists"; + case BPF_FIELD_SIGNED: return "signed"; + case BPF_FIELD_LSHIFT_U64: return "lshift_u64"; + case BPF_FIELD_RSHIFT_U64: return "rshift_u64"; + case BPF_TYPE_ID_LOCAL: return "local_type_id"; + case BPF_TYPE_ID_TARGET: return "target_type_id"; + case BPF_TYPE_EXISTS: return "type_exists"; + case BPF_TYPE_SIZE: return "type_size"; + case BPF_ENUMVAL_EXISTS: return "enumval_exists"; + case BPF_ENUMVAL_VALUE: return "enumval_value"; + default: return "unknown"; + } +} + +static bool core_relo_is_field_based(enum bpf_core_relo_kind kind) +{ + switch (kind) { + case BPF_FIELD_BYTE_OFFSET: + case BPF_FIELD_BYTE_SIZE: + case BPF_FIELD_EXISTS: + case BPF_FIELD_SIGNED: + case BPF_FIELD_LSHIFT_U64: + case BPF_FIELD_RSHIFT_U64: + return true; + default: + return false; + } +} + +static bool core_relo_is_type_based(enum bpf_core_relo_kind kind) +{ + switch (kind) { + case BPF_TYPE_ID_LOCAL: + case BPF_TYPE_ID_TARGET: + case BPF_TYPE_EXISTS: + case BPF_TYPE_SIZE: + return true; + default: + return false; + } +} + +static bool core_relo_is_enumval_based(enum bpf_core_relo_kind kind) +{ + switch (kind) { + case BPF_ENUMVAL_EXISTS: + case BPF_ENUMVAL_VALUE: + return true; + default: + return false; + } +} + /* - * Turn bpf_field_reloc into a low- and high-level spec representation, + * Turn bpf_core_relo into a low- and high-level spec representation, * validating correctness along the way, as well as calculating resulting * field bit offset, specified by accessor string. Low-level spec captures * every single level of nestedness, including traversing anonymous @@ -4071,10 +4293,17 @@ static bool is_flex_arr(const struct btf *btf, * - field 'a' access (corresponds to '2' in low-level spec); * - array element #3 access (corresponds to '3' in low-level spec). * + * Type-based relocations (TYPE_EXISTS/TYPE_SIZE, + * TYPE_ID_LOCAL/TYPE_ID_TARGET) don't capture any field information. Their + * spec and raw_spec are kept empty. + * + * Enum value-based relocations (ENUMVAL_EXISTS/ENUMVAL_VALUE) use access + * string to specify enumerator's value index that need to be relocated. */ -static int bpf_core_spec_parse(const struct btf *btf, +static int bpf_core_parse_spec(const struct btf *btf, __u32 type_id, const char *spec_str, + enum bpf_core_relo_kind relo_kind, struct bpf_core_spec *spec) { int access_idx, parsed_len, i; @@ -4089,6 +4318,15 @@ static int bpf_core_spec_parse(const struct btf *btf, memset(spec, 0, sizeof(*spec)); spec->btf = btf; + spec->root_type_id = type_id; + spec->relo_kind = relo_kind; + + /* type-based relocations don't have a field access string */ + if (core_relo_is_type_based(relo_kind)) { + if (strcmp(spec_str, "0")) + return -EINVAL; + return 0; + } /* parse spec_str="0:1:2:3:4" into array raw_spec=[0, 1, 2, 3, 4] */ while (*spec_str) { @@ -4105,16 +4343,28 @@ static int bpf_core_spec_parse(const struct btf *btf, if (spec->raw_len == 0) return -EINVAL; - /* first spec value is always reloc type array index */ t = skip_mods_and_typedefs(btf, type_id, &id); if (!t) return -EINVAL; access_idx = spec->raw_spec[0]; - spec->spec[0].type_id = id; - spec->spec[0].idx = access_idx; + acc = &spec->spec[0]; + acc->type_id = id; + acc->idx = access_idx; spec->len++; + if (core_relo_is_enumval_based(relo_kind)) { + if (!btf_is_enum(t) || spec->raw_len > 1 || access_idx >= btf_vlen(t)) + return -EINVAL; + + /* record enumerator name in a first accessor */ + acc->name = btf__name_by_offset(btf, btf_enum(t)[access_idx].name_off); + return 0; + } + + if (!core_relo_is_field_based(relo_kind)) + return -EINVAL; + sz = btf__resolve_size(btf, id); if (sz < 0) return sz; @@ -4172,8 +4422,8 @@ static int bpf_core_spec_parse(const struct btf *btf, return sz; spec->bit_offset += access_idx * sz * 8; } else { - pr_warn("relo for [%u] %s (at idx %d) captures type [%d] of unexpected kind %d\n", - type_id, spec_str, i, id, btf_kind(t)); + pr_warn("relo for [%u] %s (at idx %d) captures type [%d] of unexpected kind %s\n", + type_id, spec_str, i, id, btf_kind_str(t)); return -EINVAL; } } @@ -4223,16 +4473,16 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf, { size_t local_essent_len, targ_essent_len; const char *local_name, *targ_name; - const struct btf_type *t; + const struct btf_type *t, *local_t; struct ids_vec *cand_ids; __u32 *new_ids; int i, err, n; - t = btf__type_by_id(local_btf, local_type_id); - if (!t) + local_t = btf__type_by_id(local_btf, local_type_id); + if (!local_t) return ERR_PTR(-EINVAL); - local_name = btf__name_by_offset(local_btf, t->name_off); + local_name = btf__name_by_offset(local_btf, local_t->name_off); if (str_is_empty(local_name)) return ERR_PTR(-EINVAL); local_essent_len = bpf_core_essential_name_len(local_name); @@ -4244,12 +4494,11 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf, n = btf__get_nr_types(targ_btf); for (i = 1; i <= n; i++) { t = btf__type_by_id(targ_btf, i); - targ_name = btf__name_by_offset(targ_btf, t->name_off); - if (str_is_empty(targ_name)) + if (btf_kind(t) != btf_kind(local_t)) continue; - t = skip_mods_and_typedefs(targ_btf, i, NULL); - if (!btf_is_composite(t) && !btf_is_array(t)) + targ_name = btf__name_by_offset(targ_btf, t->name_off); + if (str_is_empty(targ_name)) continue; targ_essent_len = bpf_core_essential_name_len(targ_name); @@ -4257,11 +4506,12 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf, continue; if (strncmp(local_name, targ_name, local_essent_len) == 0) { - pr_debug("[%d] %s: found candidate [%d] %s\n", - local_type_id, local_name, i, targ_name); - new_ids = reallocarray(cand_ids->data, - cand_ids->len + 1, - sizeof(*cand_ids->data)); + pr_debug("CO-RE relocating [%d] %s %s: found target candidate [%d] %s %s\n", + local_type_id, btf_kind_str(local_t), + local_name, i, btf_kind_str(t), targ_name); + new_ids = libbpf_reallocarray(cand_ids->data, + cand_ids->len + 1, + sizeof(*cand_ids->data)); if (!new_ids) { err = -ENOMEM; goto err_out; @@ -4276,8 +4526,9 @@ err_out: return ERR_PTR(err); } -/* Check two types for compatibility, skipping const/volatile/restrict and - * typedefs, to ensure we are relocating compatible entities: +/* Check two types for compatibility for the purpose of field access + * relocation. const/volatile/restrict and typedefs are skipped to ensure we + * are relocating semantically compatible entities: * - any two STRUCTs/UNIONs are compatible and can be mixed; * - any two FWDs are compatible, if their names match (modulo flavor suffix); * - any two PTRs are always compatible; @@ -4432,6 +4683,100 @@ static int bpf_core_match_member(const struct btf *local_btf, return 0; } +/* Check local and target types for compatibility. This check is used for + * type-based CO-RE relocations and follow slightly different rules than + * field-based relocations. This function assumes that root types were already + * checked for name match. Beyond that initial root-level name check, names + * are completely ignored. Compatibility rules are as follows: + * - any two STRUCTs/UNIONs/FWDs/ENUMs/INTs are considered compatible, but + * kind should match for local and target types (i.e., STRUCT is not + * compatible with UNION); + * - for ENUMs, the size is ignored; + * - for INT, size and signedness are ignored; + * - for ARRAY, dimensionality is ignored, element types are checked for + * compatibility recursively; + * - CONST/VOLATILE/RESTRICT modifiers are ignored; + * - TYPEDEFs/PTRs are compatible if types they pointing to are compatible; + * - FUNC_PROTOs are compatible if they have compatible signature: same + * number of input args and compatible return and argument types. + * These rules are not set in stone and probably will be adjusted as we get + * more experience with using BPF CO-RE relocations. + */ +static int bpf_core_types_are_compat(const struct btf *local_btf, __u32 local_id, + const struct btf *targ_btf, __u32 targ_id) +{ + const struct btf_type *local_type, *targ_type; + int depth = 32; /* max recursion depth */ + + /* caller made sure that names match (ignoring flavor suffix) */ + local_type = btf__type_by_id(local_btf, local_id); + targ_type = btf__type_by_id(targ_btf, targ_id); + if (btf_kind(local_type) != btf_kind(targ_type)) + return 0; + +recur: + depth--; + if (depth < 0) + return -EINVAL; + + local_type = skip_mods_and_typedefs(local_btf, local_id, &local_id); + targ_type = skip_mods_and_typedefs(targ_btf, targ_id, &targ_id); + if (!local_type || !targ_type) + return -EINVAL; + + if (btf_kind(local_type) != btf_kind(targ_type)) + return 0; + + switch (btf_kind(local_type)) { + case BTF_KIND_UNKN: + case BTF_KIND_STRUCT: + case BTF_KIND_UNION: + case BTF_KIND_ENUM: + case BTF_KIND_FWD: + return 1; + case BTF_KIND_INT: + /* just reject deprecated bitfield-like integers; all other + * integers are by default compatible between each other + */ + return btf_int_offset(local_type) == 0 && btf_int_offset(targ_type) == 0; + case BTF_KIND_PTR: + local_id = local_type->type; + targ_id = targ_type->type; + goto recur; + case BTF_KIND_ARRAY: + local_id = btf_array(local_type)->type; + targ_id = btf_array(targ_type)->type; + goto recur; + case BTF_KIND_FUNC_PROTO: { + struct btf_param *local_p = btf_params(local_type); + struct btf_param *targ_p = btf_params(targ_type); + __u16 local_vlen = btf_vlen(local_type); + __u16 targ_vlen = btf_vlen(targ_type); + int i, err; + + if (local_vlen != targ_vlen) + return 0; + + for (i = 0; i < local_vlen; i++, local_p++, targ_p++) { + skip_mods_and_typedefs(local_btf, local_p->type, &local_id); + skip_mods_and_typedefs(targ_btf, targ_p->type, &targ_id); + err = bpf_core_types_are_compat(local_btf, local_id, targ_btf, targ_id); + if (err <= 0) + return err; + } + + /* tail recurse for return type check */ + skip_mods_and_typedefs(local_btf, local_type->type, &local_id); + skip_mods_and_typedefs(targ_btf, targ_type->type, &targ_id); + goto recur; + } + default: + pr_warn("unexpected kind %s relocated, local [%d], target [%d]\n", + btf_kind_str(local_type), local_id, targ_id); + return 0; + } +} + /* * Try to match local spec to a target type and, if successful, produce full * target spec (high-level, low-level + bit offset). @@ -4447,10 +4792,51 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec, memset(targ_spec, 0, sizeof(*targ_spec)); targ_spec->btf = targ_btf; + targ_spec->root_type_id = targ_id; + targ_spec->relo_kind = local_spec->relo_kind; + + if (core_relo_is_type_based(local_spec->relo_kind)) { + return bpf_core_types_are_compat(local_spec->btf, + local_spec->root_type_id, + targ_btf, targ_id); + } local_acc = &local_spec->spec[0]; targ_acc = &targ_spec->spec[0]; + if (core_relo_is_enumval_based(local_spec->relo_kind)) { + size_t local_essent_len, targ_essent_len; + const struct btf_enum *e; + const char *targ_name; + + /* has to resolve to an enum */ + targ_type = skip_mods_and_typedefs(targ_spec->btf, targ_id, &targ_id); + if (!btf_is_enum(targ_type)) + return 0; + + local_essent_len = bpf_core_essential_name_len(local_acc->name); + + for (i = 0, e = btf_enum(targ_type); i < btf_vlen(targ_type); i++, e++) { + targ_name = btf__name_by_offset(targ_spec->btf, e->name_off); + targ_essent_len = bpf_core_essential_name_len(targ_name); + if (targ_essent_len != local_essent_len) + continue; + if (strncmp(local_acc->name, targ_name, local_essent_len) == 0) { + targ_acc->type_id = targ_id; + targ_acc->idx = i; + targ_acc->name = targ_name; + targ_spec->len++; + targ_spec->raw_spec[targ_spec->raw_len] = targ_acc->idx; + targ_spec->raw_len++; + return 1; + } + } + return 0; + } + + if (!core_relo_is_field_based(local_spec->relo_kind)) + return -EINVAL; + for (i = 0; i < local_spec->len; i++, local_acc++, targ_acc++) { targ_type = skip_mods_and_typedefs(targ_spec->btf, targ_id, &targ_id); @@ -4507,18 +4893,29 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec, } static int bpf_core_calc_field_relo(const struct bpf_program *prog, - const struct bpf_field_reloc *relo, + const struct bpf_core_relo *relo, const struct bpf_core_spec *spec, __u32 *val, bool *validate) { - const struct bpf_core_accessor *acc = &spec->spec[spec->len - 1]; - const struct btf_type *t = btf__type_by_id(spec->btf, acc->type_id); + const struct bpf_core_accessor *acc; + const struct btf_type *t; __u32 byte_off, byte_sz, bit_off, bit_sz; const struct btf_member *m; const struct btf_type *mt; bool bitfield; __s64 sz; + if (relo->kind == BPF_FIELD_EXISTS) { + *val = spec ? 1 : 0; + return 0; + } + + if (!spec) + return -EUCLEAN; /* request instruction poisoning */ + + acc = &spec->spec[spec->len - 1]; + t = btf__type_by_id(spec->btf, acc->type_id); + /* a[n] accessor needs special handling */ if (!acc->name) { if (relo->kind == BPF_FIELD_BYTE_OFFSET) { @@ -4604,21 +5001,158 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog, break; case BPF_FIELD_EXISTS: default: - pr_warn("prog '%s': unknown relo %d at insn #%d\n", - bpf_program__title(prog, false), - relo->kind, relo->insn_off / 8); - return -EINVAL; + return -EOPNOTSUPP; + } + + return 0; +} + +static int bpf_core_calc_type_relo(const struct bpf_core_relo *relo, + const struct bpf_core_spec *spec, + __u32 *val) +{ + __s64 sz; + + /* type-based relos return zero when target type is not found */ + if (!spec) { + *val = 0; + return 0; + } + + switch (relo->kind) { + case BPF_TYPE_ID_TARGET: + *val = spec->root_type_id; + break; + case BPF_TYPE_EXISTS: + *val = 1; + break; + case BPF_TYPE_SIZE: + sz = btf__resolve_size(spec->btf, spec->root_type_id); + if (sz < 0) + return -EINVAL; + *val = sz; + break; + case BPF_TYPE_ID_LOCAL: + /* BPF_TYPE_ID_LOCAL is handled specially and shouldn't get here */ + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int bpf_core_calc_enumval_relo(const struct bpf_core_relo *relo, + const struct bpf_core_spec *spec, + __u32 *val) +{ + const struct btf_type *t; + const struct btf_enum *e; + + switch (relo->kind) { + case BPF_ENUMVAL_EXISTS: + *val = spec ? 1 : 0; + break; + case BPF_ENUMVAL_VALUE: + if (!spec) + return -EUCLEAN; /* request instruction poisoning */ + t = btf__type_by_id(spec->btf, spec->spec[0].type_id); + e = btf_enum(t) + spec->spec[0].idx; + *val = e->val; + break; + default: + return -EOPNOTSUPP; } return 0; } +struct bpf_core_relo_res +{ + /* expected value in the instruction, unless validate == false */ + __u32 orig_val; + /* new value that needs to be patched up to */ + __u32 new_val; + /* relocation unsuccessful, poison instruction, but don't fail load */ + bool poison; + /* some relocations can't be validated against orig_val */ + bool validate; +}; + +/* Calculate original and target relocation values, given local and target + * specs and relocation kind. These values are calculated for each candidate. + * If there are multiple candidates, resulting values should all be consistent + * with each other. Otherwise, libbpf will refuse to proceed due to ambiguity. + * If instruction has to be poisoned, *poison will be set to true. + */ +static int bpf_core_calc_relo(const struct bpf_program *prog, + const struct bpf_core_relo *relo, + int relo_idx, + const struct bpf_core_spec *local_spec, + const struct bpf_core_spec *targ_spec, + struct bpf_core_relo_res *res) +{ + int err = -EOPNOTSUPP; + + res->orig_val = 0; + res->new_val = 0; + res->poison = false; + res->validate = true; + + if (core_relo_is_field_based(relo->kind)) { + err = bpf_core_calc_field_relo(prog, relo, local_spec, &res->orig_val, &res->validate); + err = err ?: bpf_core_calc_field_relo(prog, relo, targ_spec, &res->new_val, NULL); + } else if (core_relo_is_type_based(relo->kind)) { + err = bpf_core_calc_type_relo(relo, local_spec, &res->orig_val); + err = err ?: bpf_core_calc_type_relo(relo, targ_spec, &res->new_val); + } else if (core_relo_is_enumval_based(relo->kind)) { + err = bpf_core_calc_enumval_relo(relo, local_spec, &res->orig_val); + err = err ?: bpf_core_calc_enumval_relo(relo, targ_spec, &res->new_val); + } + + if (err == -EUCLEAN) { + /* EUCLEAN is used to signal instruction poisoning request */ + res->poison = true; + err = 0; + } else if (err == -EOPNOTSUPP) { + /* EOPNOTSUPP means unknown/unsupported relocation */ + pr_warn("prog '%s': relo #%d: unrecognized CO-RE relocation %s (%d) at insn #%d\n", + bpf_program__title(prog, false), relo_idx, + core_relo_kind_str(relo->kind), relo->kind, relo->insn_off / 8); + } + + return err; +} + +/* + * Turn instruction for which CO_RE relocation failed into invalid one with + * distinct signature. + */ +static void bpf_core_poison_insn(struct bpf_program *prog, int relo_idx, + int insn_idx, struct bpf_insn *insn) +{ + pr_debug("prog '%s': relo #%d: substituting insn #%d w/ invalid insn\n", + bpf_program__title(prog, false), relo_idx, insn_idx); + insn->code = BPF_JMP | BPF_CALL; + insn->dst_reg = 0; + insn->src_reg = 0; + insn->off = 0; + /* if this instruction is reachable (not a dead code), + * verifier will complain with the following message: + * invalid func unknown#195896080 + */ + insn->imm = 195896080; /* => 0xbad2310 => "bad relo" */ +} + +static bool is_ldimm64(struct bpf_insn *insn) +{ + return insn->code == (BPF_LD | BPF_IMM | BPF_DW); +} + /* * Patch relocatable BPF instruction. * * Patched value is determined by relocation kind and target specification. - * For field existence relocation target spec will be NULL if field is not - * found. + * For existence relocations target spec will be NULL if field/type is not found. * Expected insn->imm value is determined using relocation kind and local * spec, and is checked before patching instruction. If actual insn->imm value * is wrong, bail out with error. @@ -4626,58 +5160,43 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog, * Currently three kinds of BPF instructions are supported: * 1. rX = <imm> (assignment with immediate operand); * 2. rX += <imm> (arithmetic operations with immediate operand); + * 3. rX = <imm64> (load with 64-bit immediate value). */ -static int bpf_core_reloc_insn(struct bpf_program *prog, - const struct bpf_field_reloc *relo, +static int bpf_core_patch_insn(struct bpf_program *prog, + const struct bpf_core_relo *relo, int relo_idx, - const struct bpf_core_spec *local_spec, - const struct bpf_core_spec *targ_spec) + const struct bpf_core_relo_res *res) { __u32 orig_val, new_val; struct bpf_insn *insn; - bool validate = true; - int insn_idx, err; + int insn_idx; __u8 class; - if (relo->insn_off % sizeof(struct bpf_insn)) + if (relo->insn_off % BPF_INSN_SZ) return -EINVAL; - insn_idx = relo->insn_off / sizeof(struct bpf_insn); + insn_idx = relo->insn_off / BPF_INSN_SZ; insn = &prog->insns[insn_idx]; class = BPF_CLASS(insn->code); - if (relo->kind == BPF_FIELD_EXISTS) { - orig_val = 1; /* can't generate EXISTS relo w/o local field */ - new_val = targ_spec ? 1 : 0; - } else if (!targ_spec) { - pr_debug("prog '%s': relo #%d: substituting insn #%d w/ invalid insn\n", - bpf_program__title(prog, false), relo_idx, insn_idx); - insn->code = BPF_JMP | BPF_CALL; - insn->dst_reg = 0; - insn->src_reg = 0; - insn->off = 0; - /* if this instruction is reachable (not a dead code), - * verifier will complain with the following message: - * invalid func unknown#195896080 + if (res->poison) { + /* poison second part of ldimm64 to avoid confusing error from + * verifier about "unknown opcode 00" */ - insn->imm = 195896080; /* => 0xbad2310 => "bad relo" */ + if (is_ldimm64(insn)) + bpf_core_poison_insn(prog, relo_idx, insn_idx + 1, insn + 1); + bpf_core_poison_insn(prog, relo_idx, insn_idx, insn); return 0; - } else { - err = bpf_core_calc_field_relo(prog, relo, local_spec, - &orig_val, &validate); - if (err) - return err; - err = bpf_core_calc_field_relo(prog, relo, targ_spec, - &new_val, NULL); - if (err) - return err; } + orig_val = res->orig_val; + new_val = res->new_val; + switch (class) { case BPF_ALU: case BPF_ALU64: if (BPF_SRC(insn->code) != BPF_K) return -EINVAL; - if (validate && insn->imm != orig_val) { + if (res->validate && insn->imm != orig_val) { pr_warn("prog '%s': relo #%d: unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n", bpf_program__title(prog, false), relo_idx, insn_idx, insn->imm, orig_val, new_val); @@ -4692,8 +5211,8 @@ static int bpf_core_reloc_insn(struct bpf_program *prog, case BPF_LDX: case BPF_ST: case BPF_STX: - if (validate && insn->off != orig_val) { - pr_warn("prog '%s': relo #%d: unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n", + if (res->validate && insn->off != orig_val) { + pr_warn("prog '%s': relo #%d: unexpected insn #%d (LDX/ST/STX) value: got %u, exp %u -> %u\n", bpf_program__title(prog, false), relo_idx, insn_idx, insn->off, orig_val, new_val); return -EINVAL; @@ -4710,8 +5229,37 @@ static int bpf_core_reloc_insn(struct bpf_program *prog, bpf_program__title(prog, false), relo_idx, insn_idx, orig_val, new_val); break; + case BPF_LD: { + __u64 imm; + + if (!is_ldimm64(insn) || + insn[0].src_reg != 0 || insn[0].off != 0 || + insn_idx + 1 >= prog->insns_cnt || + insn[1].code != 0 || insn[1].dst_reg != 0 || + insn[1].src_reg != 0 || insn[1].off != 0) { + pr_warn("prog '%s': relo #%d: insn #%d (LDIMM64) has unexpected form\n", + bpf_program__title(prog, false), relo_idx, insn_idx); + return -EINVAL; + } + + imm = insn[0].imm + ((__u64)insn[1].imm << 32); + if (res->validate && imm != orig_val) { + pr_warn("prog '%s': relo #%d: unexpected insn #%d (LDIMM64) value: got %llu, exp %u -> %u\n", + bpf_program__title(prog, false), relo_idx, + insn_idx, (unsigned long long)imm, + orig_val, new_val); + return -EINVAL; + } + + insn[0].imm = new_val; + insn[1].imm = 0; /* currently only 32-bit values are supported */ + pr_debug("prog '%s': relo #%d: patched insn #%d (LDIMM64) imm64 %llu -> %u\n", + bpf_program__title(prog, false), relo_idx, insn_idx, + (unsigned long long)imm, new_val); + break; + } default: - pr_warn("prog '%s': relo #%d: trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n", + pr_warn("prog '%s': relo #%d: trying to relocate unrecognized insn #%d, code:0x%x, src:0x%x, dst:0x%x, off:0x%x, imm:0x%x\n", bpf_program__title(prog, false), relo_idx, insn_idx, insn->code, insn->src_reg, insn->dst_reg, insn->off, insn->imm); @@ -4728,29 +5276,48 @@ static int bpf_core_reloc_insn(struct bpf_program *prog, static void bpf_core_dump_spec(int level, const struct bpf_core_spec *spec) { const struct btf_type *t; + const struct btf_enum *e; const char *s; __u32 type_id; int i; - type_id = spec->spec[0].type_id; + type_id = spec->root_type_id; t = btf__type_by_id(spec->btf, type_id); s = btf__name_by_offset(spec->btf, t->name_off); - libbpf_print(level, "[%u] %s + ", type_id, s); - for (i = 0; i < spec->raw_len; i++) - libbpf_print(level, "%d%s", spec->raw_spec[i], - i == spec->raw_len - 1 ? " => " : ":"); + libbpf_print(level, "[%u] %s %s", type_id, btf_kind_str(t), str_is_empty(s) ? "<anon>" : s); - libbpf_print(level, "%u.%u @ &x", - spec->bit_offset / 8, spec->bit_offset % 8); + if (core_relo_is_type_based(spec->relo_kind)) + return; - for (i = 0; i < spec->len; i++) { - if (spec->spec[i].name) - libbpf_print(level, ".%s", spec->spec[i].name); - else - libbpf_print(level, "[%u]", spec->spec[i].idx); + if (core_relo_is_enumval_based(spec->relo_kind)) { + t = skip_mods_and_typedefs(spec->btf, type_id, NULL); + e = btf_enum(t) + spec->raw_spec[0]; + s = btf__name_by_offset(spec->btf, e->name_off); + + libbpf_print(level, "::%s = %u", s, e->val); + return; } + if (core_relo_is_field_based(spec->relo_kind)) { + for (i = 0; i < spec->len; i++) { + if (spec->spec[i].name) + libbpf_print(level, ".%s", spec->spec[i].name); + else if (i > 0 || spec->spec[i].idx > 0) + libbpf_print(level, "[%u]", spec->spec[i].idx); + } + + libbpf_print(level, " ("); + for (i = 0; i < spec->raw_len; i++) + libbpf_print(level, "%s%d", i == 0 ? "" : ":", spec->raw_spec[i]); + + if (spec->bit_offset % 8) + libbpf_print(level, " @ offset %u.%u)", + spec->bit_offset / 8, spec->bit_offset % 8); + else + libbpf_print(level, " @ offset %u)", spec->bit_offset / 8); + return; + } } static size_t bpf_core_hash_fn(const void *key, void *ctx) @@ -4814,22 +5381,23 @@ static void *u32_as_hash_key(__u32 x) * CPU-wise compared to prebuilding a map from all local type names to * a list of candidate type names. It's also sped up by caching resolved * list of matching candidates per each local "root" type ID, that has at - * least one bpf_field_reloc associated with it. This list is shared + * least one bpf_core_relo associated with it. This list is shared * between multiple relocations for the same type ID and is updated as some * of the candidates are pruned due to structural incompatibility. */ -static int bpf_core_reloc_field(struct bpf_program *prog, - const struct bpf_field_reloc *relo, - int relo_idx, - const struct btf *local_btf, - const struct btf *targ_btf, - struct hashmap *cand_cache) +static int bpf_core_apply_relo(struct bpf_program *prog, + const struct bpf_core_relo *relo, + int relo_idx, + const struct btf *local_btf, + const struct btf *targ_btf, + struct hashmap *cand_cache) { const char *prog_name = bpf_program__title(prog, false); - struct bpf_core_spec local_spec, cand_spec, targ_spec; + struct bpf_core_spec local_spec, cand_spec, targ_spec = {}; const void *type_key = u32_as_hash_key(relo->type_id); - const struct btf_type *local_type, *cand_type; - const char *local_name, *cand_name; + struct bpf_core_relo_res cand_res, targ_res; + const struct btf_type *local_type; + const char *local_name; struct ids_vec *cand_ids; __u32 local_id, cand_id; const char *spec_str; @@ -4841,32 +5409,49 @@ static int bpf_core_reloc_field(struct bpf_program *prog, return -EINVAL; local_name = btf__name_by_offset(local_btf, local_type->name_off); - if (str_is_empty(local_name)) + if (!local_name) return -EINVAL; spec_str = btf__name_by_offset(local_btf, relo->access_str_off); if (str_is_empty(spec_str)) return -EINVAL; - err = bpf_core_spec_parse(local_btf, local_id, spec_str, &local_spec); + err = bpf_core_parse_spec(local_btf, local_id, spec_str, relo->kind, &local_spec); if (err) { - pr_warn("prog '%s': relo #%d: parsing [%d] %s + %s failed: %d\n", - prog_name, relo_idx, local_id, local_name, spec_str, - err); + pr_warn("prog '%s': relo #%d: parsing [%d] %s %s + %s failed: %d\n", + prog_name, relo_idx, local_id, btf_kind_str(local_type), + str_is_empty(local_name) ? "<anon>" : local_name, + spec_str, err); return -EINVAL; } - pr_debug("prog '%s': relo #%d: kind %d, spec is ", prog_name, relo_idx, - relo->kind); + pr_debug("prog '%s': relo #%d: kind <%s> (%d), spec is ", prog_name, + relo_idx, core_relo_kind_str(relo->kind), relo->kind); bpf_core_dump_spec(LIBBPF_DEBUG, &local_spec); libbpf_print(LIBBPF_DEBUG, "\n"); + /* TYPE_ID_LOCAL relo is special and doesn't need candidate search */ + if (relo->kind == BPF_TYPE_ID_LOCAL) { + targ_res.validate = true; + targ_res.poison = false; + targ_res.orig_val = local_spec.root_type_id; + targ_res.new_val = local_spec.root_type_id; + goto patch_insn; + } + + /* libbpf doesn't support candidate search for anonymous types */ + if (str_is_empty(spec_str)) { + pr_warn("prog '%s': relo #%d: <%s> (%d) relocation doesn't support anonymous types\n", + prog_name, relo_idx, core_relo_kind_str(relo->kind), relo->kind); + return -EOPNOTSUPP; + } + if (!hashmap__find(cand_cache, type_key, (void **)&cand_ids)) { cand_ids = bpf_core_find_cands(local_btf, local_id, targ_btf); if (IS_ERR(cand_ids)) { - pr_warn("prog '%s': relo #%d: target candidate search failed for [%d] %s: %ld", - prog_name, relo_idx, local_id, local_name, - PTR_ERR(cand_ids)); + pr_warn("prog '%s': relo #%d: target candidate search failed for [%d] %s %s: %ld", + prog_name, relo_idx, local_id, btf_kind_str(local_type), + local_name, PTR_ERR(cand_ids)); return PTR_ERR(cand_ids); } err = hashmap__set(cand_cache, type_key, cand_ids, NULL, NULL); @@ -4878,36 +5463,51 @@ static int bpf_core_reloc_field(struct bpf_program *prog, for (i = 0, j = 0; i < cand_ids->len; i++) { cand_id = cand_ids->data[i]; - cand_type = btf__type_by_id(targ_btf, cand_id); - cand_name = btf__name_by_offset(targ_btf, cand_type->name_off); - - err = bpf_core_spec_match(&local_spec, targ_btf, - cand_id, &cand_spec); - pr_debug("prog '%s': relo #%d: matching candidate #%d %s against spec ", - prog_name, relo_idx, i, cand_name); - bpf_core_dump_spec(LIBBPF_DEBUG, &cand_spec); - libbpf_print(LIBBPF_DEBUG, ": %d\n", err); + err = bpf_core_spec_match(&local_spec, targ_btf, cand_id, &cand_spec); if (err < 0) { - pr_warn("prog '%s': relo #%d: matching error: %d\n", - prog_name, relo_idx, err); + pr_warn("prog '%s': relo #%d: error matching candidate #%d ", + prog_name, relo_idx, i); + bpf_core_dump_spec(LIBBPF_WARN, &cand_spec); + libbpf_print(LIBBPF_WARN, ": %d\n", err); return err; } + + pr_debug("prog '%s': relo #%d: %s candidate #%d ", prog_name, + relo_idx, err == 0 ? "non-matching" : "matching", i); + bpf_core_dump_spec(LIBBPF_DEBUG, &cand_spec); + libbpf_print(LIBBPF_DEBUG, "\n"); + if (err == 0) continue; + err = bpf_core_calc_relo(prog, relo, relo_idx, &local_spec, &cand_spec, &cand_res); + if (err) + return err; + if (j == 0) { + targ_res = cand_res; targ_spec = cand_spec; } else if (cand_spec.bit_offset != targ_spec.bit_offset) { - /* if there are many candidates, they should all - * resolve to the same bit offset + /* if there are many field relo candidates, they + * should all resolve to the same bit offset */ - pr_warn("prog '%s': relo #%d: offset ambiguity: %u != %u\n", + pr_warn("prog '%s': relo #%d: field offset ambiguity: %u != %u\n", prog_name, relo_idx, cand_spec.bit_offset, targ_spec.bit_offset); return -EINVAL; + } else if (cand_res.poison != targ_res.poison || cand_res.new_val != targ_res.new_val) { + /* all candidates should result in the same relocation + * decision and value, otherwise it's dangerous to + * proceed due to ambiguity + */ + pr_warn("prog '%s': relo #%d: relocation decision ambiguity: %s %u != %s %u\n", + prog_name, relo_idx, + cand_res.poison ? "failure" : "success", cand_res.new_val, + targ_res.poison ? "failure" : "success", targ_res.new_val); + return -EINVAL; } - cand_ids->data[j++] = cand_spec.spec[0].type_id; + cand_ids->data[j++] = cand_spec.root_type_id; } /* @@ -4926,19 +5526,25 @@ static int bpf_core_reloc_field(struct bpf_program *prog, * as well as expected case, depending whether instruction w/ * relocation is guarded in some way that makes it unreachable (dead * code) if relocation can't be resolved. This is handled in - * bpf_core_reloc_insn() uniformly by replacing that instruction with + * bpf_core_patch_insn() uniformly by replacing that instruction with * BPF helper call insn (using invalid helper ID). If that instruction * is indeed unreachable, then it will be ignored and eliminated by * verifier. If it was an error, then verifier will complain and point * to a specific instruction number in its log. */ - if (j == 0) - pr_debug("prog '%s': relo #%d: no matching targets found for [%d] %s + %s\n", - prog_name, relo_idx, local_id, local_name, spec_str); + if (j == 0) { + pr_debug("prog '%s': relo #%d: no matching targets found\n", + prog_name, relo_idx); - /* bpf_core_reloc_insn should know how to handle missing targ_spec */ - err = bpf_core_reloc_insn(prog, relo, relo_idx, &local_spec, - j ? &targ_spec : NULL); + /* calculate single target relo result explicitly */ + err = bpf_core_calc_relo(prog, relo, relo_idx, &local_spec, NULL, &targ_res); + if (err) + return err; + } + +patch_insn: + /* bpf_core_patch_insn() should know how to handle missing targ_spec */ + err = bpf_core_patch_insn(prog, relo, relo_idx, &targ_res); if (err) { pr_warn("prog '%s': relo #%d: failed to patch insn at offset %d: %d\n", prog_name, relo_idx, relo->insn_off, err); @@ -4949,10 +5555,10 @@ static int bpf_core_reloc_field(struct bpf_program *prog, } static int -bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path) +bpf_object__relocate_core(struct bpf_object *obj, const char *targ_btf_path) { const struct btf_ext_info_sec *sec; - const struct bpf_field_reloc *rec; + const struct bpf_core_relo *rec; const struct btf_ext_info *seg; struct hashmap_entry *entry; struct hashmap *cand_cache = NULL; @@ -4961,6 +5567,9 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path) const char *sec_name; int i, err = 0; + if (obj->btf_ext->core_relo_info.len == 0) + return 0; + if (targ_btf_path) targ_btf = btf__parse_elf(targ_btf_path, NULL); else @@ -4976,7 +5585,7 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path) goto out; } - seg = &obj->btf_ext->field_reloc_info; + seg = &obj->btf_ext->core_relo_info; for_each_btf_ext_sec(seg, sec) { sec_name = btf__name_by_offset(obj->btf, sec->sec_name_off); if (str_is_empty(sec_name)) { @@ -4997,15 +5606,15 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path) goto out; } - pr_debug("prog '%s': performing %d CO-RE offset relocs\n", + pr_debug("sec '%s': found %d CO-RE relocations\n", sec_name, sec->num_info); for_each_btf_ext_rec(seg, sec, i, rec) { - err = bpf_core_reloc_field(prog, rec, i, obj->btf, - targ_btf, cand_cache); + err = bpf_core_apply_relo(prog, rec, i, obj->btf, + targ_btf, cand_cache); if (err) { pr_warn("prog '%s': relo #%d: failed to relocate: %d\n", - sec_name, i, err); + prog->name, i, err); goto out; } } @@ -5025,17 +5634,6 @@ out: } static int -bpf_object__relocate_core(struct bpf_object *obj, const char *targ_btf_path) -{ - int err = 0; - - if (obj->btf_ext->field_reloc_info.len) - err = bpf_core_reloc_fields(obj, targ_btf_path); - - return err; -} - -static int bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj, struct reloc_desc *relo) { @@ -5051,7 +5649,7 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj, return -LIBBPF_ERRNO__RELOC; } new_cnt = prog->insns_cnt + text->insns_cnt; - new_insn = reallocarray(prog->insns, new_cnt, sizeof(*insn)); + new_insn = libbpf_reallocarray(prog->insns, new_cnt, sizeof(*insn)); if (!new_insn) { pr_warn("oom in prog realloc\n"); return -ENOMEM; @@ -5136,7 +5734,8 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj) return err; break; default: - pr_warn("relo #%d: bad relo type %d\n", i, relo->type); + pr_warn("prog '%s': relo #%d: bad relo type %d\n", + prog->name, i, relo->type); return -EINVAL; } } @@ -5171,7 +5770,8 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path) err = bpf_program__relocate(prog, obj); if (err) { - pr_warn("failed to relocate '%s'\n", prog->section_name); + pr_warn("prog '%s': failed to relocate data references: %d\n", + prog->name, err); return err; } break; @@ -5186,7 +5786,8 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path) err = bpf_program__relocate(prog, obj); if (err) { - pr_warn("failed to relocate '%s'\n", prog->section_name); + pr_warn("prog '%s': failed to relocate calls: %d\n", + prog->name, err); return err; } } @@ -5230,8 +5831,7 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, i, (size_t)GELF_R_SYM(rel.r_info)); return -LIBBPF_ERRNO__FORMAT; } - name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name) ? : "<?>"; + name = elf_sym_str(obj, sym.st_name) ?: "<?>"; if (sym.st_shndx != obj->efile.btf_maps_shndx) { pr_warn(".maps relo #%d: '%s' isn't a BTF-defined map\n", i, name); @@ -5293,7 +5893,7 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj, moff /= bpf_ptr_sz; if (moff >= map->init_slots_sz) { new_sz = moff + 1; - tmp = realloc(map->init_slots, new_sz * host_ptr_sz); + tmp = libbpf_reallocarray(map->init_slots, new_sz, host_ptr_sz); if (!tmp) return -ENOMEM; map->init_slots = tmp; @@ -5348,6 +5948,51 @@ static int bpf_object__collect_reloc(struct bpf_object *obj) return 0; } +static bool insn_is_helper_call(struct bpf_insn *insn, enum bpf_func_id *func_id) +{ + if (BPF_CLASS(insn->code) == BPF_JMP && + BPF_OP(insn->code) == BPF_CALL && + BPF_SRC(insn->code) == BPF_K && + insn->src_reg == 0 && + insn->dst_reg == 0) { + *func_id = insn->imm; + return true; + } + return false; +} + +static int bpf_object__sanitize_prog(struct bpf_object* obj, struct bpf_program *prog) +{ + struct bpf_insn *insn = prog->insns; + enum bpf_func_id func_id; + int i; + + for (i = 0; i < prog->insns_cnt; i++, insn++) { + if (!insn_is_helper_call(insn, &func_id)) + continue; + + /* on kernels that don't yet support + * bpf_probe_read_{kernel,user}[_str] helpers, fall back + * to bpf_probe_read() which works well for old kernels + */ + switch (func_id) { + case BPF_FUNC_probe_read_kernel: + case BPF_FUNC_probe_read_user: + if (!kernel_supports(FEAT_PROBE_READ_KERN)) + insn->imm = BPF_FUNC_probe_read; + break; + case BPF_FUNC_probe_read_kernel_str: + case BPF_FUNC_probe_read_user_str: + if (!kernel_supports(FEAT_PROBE_READ_KERN)) + insn->imm = BPF_FUNC_probe_read_str; + break; + default: + break; + } + } + return 0; +} + static int load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt, char *license, __u32 kern_version, int *pfd) @@ -5364,12 +6009,12 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt, memset(&load_attr, 0, sizeof(struct bpf_load_program_attr)); load_attr.prog_type = prog->type; /* old kernels might not support specifying expected_attach_type */ - if (!prog->caps->exp_attach_type && prog->sec_def && + if (!kernel_supports(FEAT_EXP_ATTACH_TYPE) && prog->sec_def && prog->sec_def->is_exp_attach_type_optional) load_attr.expected_attach_type = 0; else load_attr.expected_attach_type = prog->expected_attach_type; - if (prog->caps->name) + if (kernel_supports(FEAT_PROG_NAME)) load_attr.name = prog->name; load_attr.insns = insns; load_attr.insns_cnt = insns_cnt; @@ -5387,7 +6032,7 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt, } /* specify func_info/line_info only if kernel supports them */ btf_fd = bpf_object__btf_fd(prog->obj); - if (btf_fd >= 0 && prog->obj->caps.btf_func) { + if (btf_fd >= 0 && kernel_supports(FEAT_BTF_FUNC)) { load_attr.prog_btf_fd = btf_fd; load_attr.func_info = prog->func_info; load_attr.func_info_rec_size = prog->func_info_rec_size; @@ -5425,7 +6070,7 @@ retry_load: free(log_buf); goto retry_load; } - ret = -errno; + ret = errno ? -errno : -LIBBPF_ERRNO__LOAD; cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg)); pr_warn("load bpf program failed: %s\n", cp); pr_perm_msg(ret); @@ -5564,11 +6209,17 @@ bpf_object__load_progs(struct bpf_object *obj, int log_level) for (i = 0; i < obj->nr_programs; i++) { prog = &obj->programs[i]; + err = bpf_object__sanitize_prog(obj, prog); + if (err) + return err; + } + + for (i = 0; i < obj->nr_programs; i++) { + prog = &obj->programs[i]; if (bpf_program__is_function_storage(prog, obj)) continue; if (!prog->load) { - pr_debug("prog '%s'('%s'): skipped loading\n", - prog->name, prog->section_name); + pr_debug("prog '%s': skipped loading\n", prog->name); continue; } prog->log_level |= log_level; @@ -5641,6 +6292,8 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz, /* couldn't guess, but user might manually specify */ continue; + if (prog->sec_def->is_sleepable) + prog->prog_flags |= BPF_F_SLEEPABLE; bpf_program__set_type(prog, prog->sec_def->prog_type); bpf_program__set_expected_attach_type(prog, prog->sec_def->expected_attach_type); @@ -5750,11 +6403,11 @@ static int bpf_object__sanitize_maps(struct bpf_object *obj) bpf_object__for_each_map(m, obj) { if (!bpf_map__is_internal(m)) continue; - if (!obj->caps.global_data) { + if (!kernel_supports(FEAT_GLOBAL_DATA)) { pr_warn("kernel doesn't support global data\n"); return -ENOTSUP; } - if (!obj->caps.array_mmap) + if (!kernel_supports(FEAT_ARRAY_MMAP)) m->def.map_flags ^= BPF_F_MMAPABLE; } @@ -5904,7 +6557,6 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr) } err = bpf_object__probe_loading(obj); - err = err ? : bpf_object__probe_caps(obj); err = err ? : bpf_object__resolve_externs(obj, obj->kconfig); err = err ? : bpf_object__sanitize_and_load_btf(obj); err = err ? : bpf_object__sanitize_maps(obj); @@ -6713,7 +7365,7 @@ int bpf_program__fd(const struct bpf_program *prog) size_t bpf_program__size(const struct bpf_program *prog) { - return prog->insns_cnt * sizeof(struct bpf_insn); + return prog->insns_cnt * BPF_INSN_SZ; } int bpf_program__set_prep(struct bpf_program *prog, int nr_instances, @@ -6910,6 +7562,21 @@ static const struct bpf_sec_def section_defs[] = { .expected_attach_type = BPF_TRACE_FEXIT, .is_attach_btf = true, .attach_fn = attach_trace), + SEC_DEF("fentry.s/", TRACING, + .expected_attach_type = BPF_TRACE_FENTRY, + .is_attach_btf = true, + .is_sleepable = true, + .attach_fn = attach_trace), + SEC_DEF("fmod_ret.s/", TRACING, + .expected_attach_type = BPF_MODIFY_RETURN, + .is_attach_btf = true, + .is_sleepable = true, + .attach_fn = attach_trace), + SEC_DEF("fexit.s/", TRACING, + .expected_attach_type = BPF_TRACE_FEXIT, + .is_attach_btf = true, + .is_sleepable = true, + .attach_fn = attach_trace), SEC_DEF("freplace/", EXT, .is_attach_btf = true, .attach_fn = attach_trace), @@ -6917,6 +7584,11 @@ static const struct bpf_sec_def section_defs[] = { .is_attach_btf = true, .expected_attach_type = BPF_LSM_MAC, .attach_fn = attach_lsm), + SEC_DEF("lsm.s/", LSM, + .is_attach_btf = true, + .is_sleepable = true, + .expected_attach_type = BPF_LSM_MAC, + .attach_fn = attach_lsm), SEC_DEF("iter/", TRACING, .expected_attach_type = BPF_TRACE_ITER, .is_attach_btf = true, @@ -7122,8 +7794,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj, return -LIBBPF_ERRNO__FORMAT; } - name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, - sym.st_name) ? : "<?>"; + name = elf_sym_str(obj, sym.st_name) ?: "<?>"; map = find_struct_ops_map_by_offset(obj, rel.r_offset); if (!map) { pr_warn("struct_ops reloc: cannot find map at rel.r_offset %zu\n", @@ -7640,7 +8311,7 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr, prog->prog_ifindex = attr->ifindex; prog->log_level = attr->log_level; - prog->prog_flags = attr->prog_flags; + prog->prog_flags |= attr->prog_flags; if (!first_prog) first_prog = prog; } @@ -8594,7 +9265,7 @@ struct perf_buffer *perf_buffer__new(int map_fd, size_t page_cnt, struct perf_buffer_params p = {}; struct perf_event_attr attr = { 0, }; - attr.config = PERF_COUNT_SW_BPF_OUTPUT, + attr.config = PERF_COUNT_SW_BPF_OUTPUT; attr.type = PERF_TYPE_SOFTWARE; attr.sample_type = PERF_SAMPLE_RAW; attr.sample_period = 1; @@ -8832,6 +9503,11 @@ static int perf_buffer__process_records(struct perf_buffer *pb, return 0; } +int perf_buffer__epoll_fd(const struct perf_buffer *pb) +{ + return pb->epoll_fd; +} + int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms) { int i, cnt, err; @@ -8849,6 +9525,55 @@ int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms) return cnt < 0 ? -errno : cnt; } +/* Return number of PERF_EVENT_ARRAY map slots set up by this perf_buffer + * manager. + */ +size_t perf_buffer__buffer_cnt(const struct perf_buffer *pb) +{ + return pb->cpu_cnt; +} + +/* + * Return perf_event FD of a ring buffer in *buf_idx* slot of + * PERF_EVENT_ARRAY BPF map. This FD can be polled for new data using + * select()/poll()/epoll() Linux syscalls. + */ +int perf_buffer__buffer_fd(const struct perf_buffer *pb, size_t buf_idx) +{ + struct perf_cpu_buf *cpu_buf; + + if (buf_idx >= pb->cpu_cnt) + return -EINVAL; + + cpu_buf = pb->cpu_bufs[buf_idx]; + if (!cpu_buf) + return -ENOENT; + + return cpu_buf->fd; +} + +/* + * Consume data from perf ring buffer corresponding to slot *buf_idx* in + * PERF_EVENT_ARRAY BPF map without waiting/polling. If there is no data to + * consume, do nothing and return success. + * Returns: + * - 0 on success; + * - <0 on failure. + */ +int perf_buffer__consume_buffer(struct perf_buffer *pb, size_t buf_idx) +{ + struct perf_cpu_buf *cpu_buf; + + if (buf_idx >= pb->cpu_cnt) + return -EINVAL; + + cpu_buf = pb->cpu_bufs[buf_idx]; + if (!cpu_buf) + return -ENOENT; + + return perf_buffer__process_records(pb, cpu_buf); +} + int perf_buffer__consume(struct perf_buffer *pb) { int i, err; @@ -8861,7 +9586,7 @@ int perf_buffer__consume(struct perf_buffer *pb) err = perf_buffer__process_records(pb, cpu_buf); if (err) { - pr_warn("error while processing records: %d\n", err); + pr_warn("perf_buffer: failed to process records in buffer #%d: %d\n", i, err); return err; } } diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h index 5ecb4069a9f0..308e0ded8f14 100644 --- a/tools/lib/bpf/libbpf.h +++ b/tools/lib/bpf/libbpf.h @@ -588,8 +588,12 @@ perf_buffer__new_raw(int map_fd, size_t page_cnt, const struct perf_buffer_raw_opts *opts); LIBBPF_API void perf_buffer__free(struct perf_buffer *pb); +LIBBPF_API int perf_buffer__epoll_fd(const struct perf_buffer *pb); LIBBPF_API int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms); LIBBPF_API int perf_buffer__consume(struct perf_buffer *pb); +LIBBPF_API int perf_buffer__consume_buffer(struct perf_buffer *pb, size_t buf_idx); +LIBBPF_API size_t perf_buffer__buffer_cnt(const struct perf_buffer *pb); +LIBBPF_API int perf_buffer__buffer_fd(const struct perf_buffer *pb, size_t buf_idx); typedef enum bpf_perf_event_ret (*bpf_perf_event_print_t)(struct perf_event_header *hdr, diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map index e35bd6cdbdbf..3fedcdc4ae2f 100644 --- a/tools/lib/bpf/libbpf.map +++ b/tools/lib/bpf/libbpf.map @@ -299,3 +299,12 @@ LIBBPF_0.1.0 { btf__set_fd; btf__set_pointer_size; } LIBBPF_0.0.9; + +LIBBPF_0.2.0 { + global: + perf_buffer__buffer_cnt; + perf_buffer__buffer_fd; + perf_buffer__epoll_fd; + perf_buffer__consume_buffer; + xsk_socket__create_shared; +} LIBBPF_0.1.0; diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h index 50d70e90d5f1..4d1c366fca2c 100644 --- a/tools/lib/bpf/libbpf_internal.h +++ b/tools/lib/bpf/libbpf_internal.h @@ -9,6 +9,15 @@ #ifndef __LIBBPF_LIBBPF_INTERNAL_H #define __LIBBPF_LIBBPF_INTERNAL_H +#include <stdlib.h> +#include <limits.h> + +/* make sure libbpf doesn't use kernel-only integer typedefs */ +#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 + +/* prevent accidental re-addition of reallocarray() */ +#pragma GCC poison reallocarray + #include "libbpf.h" #define BTF_INFO_ENC(kind, kind_flag, vlen) \ @@ -23,6 +32,12 @@ #define BTF_PARAM_ENC(name, type) (name), (type) #define BTF_VAR_SECINFO_ENC(type, offset, size) (type), (offset), (size) +#ifndef likely +#define likely(x) __builtin_expect(!!(x), 1) +#endif +#ifndef unlikely +#define unlikely(x) __builtin_expect(!!(x), 0) +#endif #ifndef min # define min(x, y) ((x) < (y) ? (x) : (y)) #endif @@ -63,6 +78,33 @@ do { \ #define pr_info(fmt, ...) __pr(LIBBPF_INFO, fmt, ##__VA_ARGS__) #define pr_debug(fmt, ...) __pr(LIBBPF_DEBUG, fmt, ##__VA_ARGS__) +#ifndef __has_builtin +#define __has_builtin(x) 0 +#endif +/* + * Re-implement glibc's reallocarray() for libbpf internal-only use. + * reallocarray(), unfortunately, is not available in all versions of glibc, + * so requires extra feature detection and using reallocarray() stub from + * <tools/libc_compat.h> and COMPAT_NEED_REALLOCARRAY. All this complicates + * build of libbpf unnecessarily and is just a maintenance burden. Instead, + * it's trivial to implement libbpf-specific internal version and use it + * throughout libbpf. + */ +static inline void *libbpf_reallocarray(void *ptr, size_t nmemb, size_t size) +{ + size_t total; + +#if __has_builtin(__builtin_mul_overflow) + if (unlikely(__builtin_mul_overflow(nmemb, size, &total))) + return NULL; +#else + if (size == 0 || nmemb > ULONG_MAX / size) + return NULL; + total = nmemb * size; +#endif + return realloc(ptr, total); +} + static inline bool libbpf_validate_opts(const char *opts, size_t opts_sz, size_t user_sz, const char *type_name) @@ -105,18 +147,6 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name, int bpf_object__variable_offset(const struct bpf_object *obj, const char *name, __u32 *off); -struct nlattr; -typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb); -int libbpf_netlink_open(unsigned int *nl_pid); -int libbpf_nl_get_link(int sock, unsigned int nl_pid, - libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie); -int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex, - libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie); -int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex, - libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie); -int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle, - libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie); - struct btf_ext_info { /* * info points to the individual info section (e.g. func_info and @@ -138,6 +168,44 @@ struct btf_ext_info { i < (sec)->num_info; \ i++, rec = (void *)rec + (seg)->rec_size) +/* + * The .BTF.ext ELF section layout defined as + * struct btf_ext_header + * func_info subsection + * + * The func_info subsection layout: + * record size for struct bpf_func_info in the func_info subsection + * struct btf_sec_func_info for section #1 + * a list of bpf_func_info records for section #1 + * where struct bpf_func_info mimics one in include/uapi/linux/bpf.h + * but may not be identical + * struct btf_sec_func_info for section #2 + * a list of bpf_func_info records for section #2 + * ...... + * + * Note that the bpf_func_info record size in .BTF.ext may not + * be the same as the one defined in include/uapi/linux/bpf.h. + * The loader should ensure that record_size meets minimum + * requirement and pass the record as is to the kernel. The + * kernel will handle the func_info properly based on its contents. + */ +struct btf_ext_header { + __u16 magic; + __u8 version; + __u8 flags; + __u32 hdr_len; + + /* All offsets are in bytes relative to the end of this header */ + __u32 func_info_off; + __u32 func_info_len; + __u32 line_info_off; + __u32 line_info_len; + + /* optional part of .BTF.ext header */ + __u32 core_relo_off; + __u32 core_relo_len; +}; + struct btf_ext { union { struct btf_ext_header *hdr; @@ -145,7 +213,7 @@ struct btf_ext { }; struct btf_ext_info func_info; struct btf_ext_info line_info; - struct btf_ext_info field_reloc_info; + struct btf_ext_info core_relo_info; __u32 data_size; }; @@ -170,32 +238,40 @@ struct bpf_line_info_min { __u32 line_col; }; -/* bpf_field_info_kind encodes which aspect of captured field has to be - * adjusted by relocations. Currently supported values are: - * - BPF_FIELD_BYTE_OFFSET: field offset (in bytes); - * - BPF_FIELD_EXISTS: field existence (1, if field exists; 0, otherwise); +/* bpf_core_relo_kind encodes which aspect of captured field/type/enum value + * has to be adjusted by relocations. */ -enum bpf_field_info_kind { +enum bpf_core_relo_kind { BPF_FIELD_BYTE_OFFSET = 0, /* field byte offset */ - BPF_FIELD_BYTE_SIZE = 1, + BPF_FIELD_BYTE_SIZE = 1, /* field size in bytes */ BPF_FIELD_EXISTS = 2, /* field existence in target kernel */ - BPF_FIELD_SIGNED = 3, - BPF_FIELD_LSHIFT_U64 = 4, - BPF_FIELD_RSHIFT_U64 = 5, + BPF_FIELD_SIGNED = 3, /* field signedness (0 - unsigned, 1 - signed) */ + BPF_FIELD_LSHIFT_U64 = 4, /* bitfield-specific left bitshift */ + BPF_FIELD_RSHIFT_U64 = 5, /* bitfield-specific right bitshift */ + BPF_TYPE_ID_LOCAL = 6, /* type ID in local BPF object */ + BPF_TYPE_ID_TARGET = 7, /* type ID in target kernel */ + BPF_TYPE_EXISTS = 8, /* type existence in target kernel */ + BPF_TYPE_SIZE = 9, /* type size in bytes */ + BPF_ENUMVAL_EXISTS = 10, /* enum value existence in target kernel */ + BPF_ENUMVAL_VALUE = 11, /* enum value integer value */ }; -/* The minimum bpf_field_reloc checked by the loader +/* The minimum bpf_core_relo checked by the loader * - * Field relocation captures the following data: + * CO-RE relocation captures the following data: * - insn_off - instruction offset (in bytes) within a BPF program that needs * its insn->imm field to be relocated with actual field info; * - type_id - BTF type ID of the "root" (containing) entity of a relocatable - * field; + * type or field; * - access_str_off - offset into corresponding .BTF string section. String - * itself encodes an accessed field using a sequence of field and array - * indicies, separated by colon (:). It's conceptually very close to LLVM's - * getelementptr ([0]) instruction's arguments for identifying offset to - * a field. + * interpretation depends on specific relocation kind: + * - for field-based relocations, string encodes an accessed field using + * a sequence of field and array indices, separated by colon (:). It's + * conceptually very close to LLVM's getelementptr ([0]) instruction's + * arguments for identifying offset to a field. + * - for type-based relocations, strings is expected to be just "0"; + * - for enum value-based relocations, string contains an index of enum + * value within its enum type; * * Example to provide a better feel. * @@ -226,11 +302,11 @@ enum bpf_field_info_kind { * * [0] https://llvm.org/docs/LangRef.html#getelementptr-instruction */ -struct bpf_field_reloc { +struct bpf_core_relo { __u32 insn_off; __u32 type_id; __u32 access_str_off; - enum bpf_field_info_kind kind; + enum bpf_core_relo_kind kind; }; #endif /* __LIBBPF_LIBBPF_INTERNAL_H */ diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c index 5a3d3f078408..5482a9b7ae2d 100644 --- a/tools/lib/bpf/libbpf_probes.c +++ b/tools/lib/bpf/libbpf_probes.c @@ -17,9 +17,6 @@ #include "libbpf.h" #include "libbpf_internal.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - static bool grep(const char *buffer, const char *pattern) { return !!strstr(buffer, pattern); @@ -173,7 +170,7 @@ int libbpf__load_raw_btf(const char *raw_types, size_t types_len, return btf_fd; } -static int load_sk_storage_btf(void) +static int load_local_storage_btf(void) { const char strs[] = "\0bpf_spin_lock\0val\0cnt\0l"; /* struct bpf_spin_lock { @@ -232,12 +229,13 @@ bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex) key_size = 0; break; case BPF_MAP_TYPE_SK_STORAGE: + case BPF_MAP_TYPE_INODE_STORAGE: btf_key_type_id = 1; btf_value_type_id = 3; value_size = 8; max_entries = 0; map_flags = BPF_F_NO_PREALLOC; - btf_fd = load_sk_storage_btf(); + btf_fd = load_local_storage_btf(); if (btf_fd < 0) return false; break; diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c index 312f887570b2..4dd73de00b6f 100644 --- a/tools/lib/bpf/netlink.c +++ b/tools/lib/bpf/netlink.c @@ -15,13 +15,12 @@ #include "libbpf_internal.h" #include "nlattr.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - #ifndef SOL_NETLINK #define SOL_NETLINK 270 #endif +typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb); + typedef int (*__dump_nlmsg_t)(struct nlmsghdr *nlmsg, libbpf_dump_nlmsg_t, void *cookie); @@ -31,7 +30,7 @@ struct xdp_id_md { struct xdp_link_info info; }; -int libbpf_netlink_open(__u32 *nl_pid) +static int libbpf_netlink_open(__u32 *nl_pid) { struct sockaddr_nl sa; socklen_t addrlen; @@ -283,6 +282,9 @@ static int get_xdp_info(void *cookie, void *msg, struct nlattr **tb) return 0; } +static int libbpf_nl_get_link(int sock, unsigned int nl_pid, + libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie); + int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info, size_t info_size, __u32 flags) { @@ -368,121 +370,3 @@ int libbpf_nl_get_link(int sock, unsigned int nl_pid, return bpf_netlink_recv(sock, nl_pid, seq, __dump_link_nlmsg, dump_link_nlmsg, cookie); } - -static int __dump_class_nlmsg(struct nlmsghdr *nlh, - libbpf_dump_nlmsg_t dump_class_nlmsg, - void *cookie) -{ - struct nlattr *tb[TCA_MAX + 1], *attr; - struct tcmsg *t = NLMSG_DATA(nlh); - int len; - - len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t)); - attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t))); - if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0) - return -LIBBPF_ERRNO__NLPARSE; - - return dump_class_nlmsg(cookie, t, tb); -} - -int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex, - libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie) -{ - struct { - struct nlmsghdr nlh; - struct tcmsg t; - } req = { - .nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)), - .nlh.nlmsg_type = RTM_GETTCLASS, - .nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST, - .t.tcm_family = AF_UNSPEC, - .t.tcm_ifindex = ifindex, - }; - int seq = time(NULL); - - req.nlh.nlmsg_seq = seq; - if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0) - return -errno; - - return bpf_netlink_recv(sock, nl_pid, seq, __dump_class_nlmsg, - dump_class_nlmsg, cookie); -} - -static int __dump_qdisc_nlmsg(struct nlmsghdr *nlh, - libbpf_dump_nlmsg_t dump_qdisc_nlmsg, - void *cookie) -{ - struct nlattr *tb[TCA_MAX + 1], *attr; - struct tcmsg *t = NLMSG_DATA(nlh); - int len; - - len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t)); - attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t))); - if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0) - return -LIBBPF_ERRNO__NLPARSE; - - return dump_qdisc_nlmsg(cookie, t, tb); -} - -int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex, - libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie) -{ - struct { - struct nlmsghdr nlh; - struct tcmsg t; - } req = { - .nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)), - .nlh.nlmsg_type = RTM_GETQDISC, - .nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST, - .t.tcm_family = AF_UNSPEC, - .t.tcm_ifindex = ifindex, - }; - int seq = time(NULL); - - req.nlh.nlmsg_seq = seq; - if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0) - return -errno; - - return bpf_netlink_recv(sock, nl_pid, seq, __dump_qdisc_nlmsg, - dump_qdisc_nlmsg, cookie); -} - -static int __dump_filter_nlmsg(struct nlmsghdr *nlh, - libbpf_dump_nlmsg_t dump_filter_nlmsg, - void *cookie) -{ - struct nlattr *tb[TCA_MAX + 1], *attr; - struct tcmsg *t = NLMSG_DATA(nlh); - int len; - - len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t)); - attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t))); - if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0) - return -LIBBPF_ERRNO__NLPARSE; - - return dump_filter_nlmsg(cookie, t, tb); -} - -int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle, - libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie) -{ - struct { - struct nlmsghdr nlh; - struct tcmsg t; - } req = { - .nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)), - .nlh.nlmsg_type = RTM_GETTFILTER, - .nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST, - .t.tcm_family = AF_UNSPEC, - .t.tcm_ifindex = ifindex, - .t.tcm_parent = handle, - }; - int seq = time(NULL); - - req.nlh.nlmsg_seq = seq; - if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0) - return -errno; - - return bpf_netlink_recv(sock, nl_pid, seq, __dump_filter_nlmsg, - dump_filter_nlmsg, cookie); -} diff --git a/tools/lib/bpf/nlattr.c b/tools/lib/bpf/nlattr.c index 0ad41dfea8eb..b607fa9852b1 100644 --- a/tools/lib/bpf/nlattr.c +++ b/tools/lib/bpf/nlattr.c @@ -7,14 +7,11 @@ */ #include <errno.h> -#include "nlattr.h" -#include "libbpf_internal.h" -#include <linux/rtnetlink.h> #include <string.h> #include <stdio.h> - -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 +#include <linux/rtnetlink.h> +#include "nlattr.h" +#include "libbpf_internal.h" static uint16_t nla_attr_minlen[LIBBPF_NLA_TYPE_MAX+1] = { [LIBBPF_NLA_U8] = sizeof(uint8_t), diff --git a/tools/lib/bpf/ringbuf.c b/tools/lib/bpf/ringbuf.c index 4fc6c6cbb4eb..5c6522c89af1 100644 --- a/tools/lib/bpf/ringbuf.c +++ b/tools/lib/bpf/ringbuf.c @@ -16,15 +16,11 @@ #include <asm/barrier.h> #include <sys/mman.h> #include <sys/epoll.h> -#include <tools/libc_compat.h> #include "libbpf.h" #include "libbpf_internal.h" #include "bpf.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - struct ring { ring_buffer_sample_fn sample_cb; void *ctx; @@ -82,12 +78,12 @@ int ring_buffer__add(struct ring_buffer *rb, int map_fd, return -EINVAL; } - tmp = reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings)); + tmp = libbpf_reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings)); if (!tmp) return -ENOMEM; rb->rings = tmp; - tmp = reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events)); + tmp = libbpf_reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events)); if (!tmp) return -ENOMEM; rb->events = tmp; diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c index f7f4efb70a4c..49c324594792 100644 --- a/tools/lib/bpf/xsk.c +++ b/tools/lib/bpf/xsk.c @@ -20,6 +20,7 @@ #include <linux/if_ether.h> #include <linux/if_packet.h> #include <linux/if_xdp.h> +#include <linux/list.h> #include <linux/sockios.h> #include <net/if.h> #include <sys/ioctl.h> @@ -32,9 +33,6 @@ #include "libbpf_internal.h" #include "xsk.h" -/* make sure libbpf doesn't use kernel-only integer typedefs */ -#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 - #ifndef SOL_XDP #define SOL_XDP 283 #endif @@ -48,26 +46,35 @@ #endif struct xsk_umem { - struct xsk_ring_prod *fill; - struct xsk_ring_cons *comp; + struct xsk_ring_prod *fill_save; + struct xsk_ring_cons *comp_save; char *umem_area; struct xsk_umem_config config; int fd; int refcount; + struct list_head ctx_list; +}; + +struct xsk_ctx { + struct xsk_ring_prod *fill; + struct xsk_ring_cons *comp; + __u32 queue_id; + struct xsk_umem *umem; + int refcount; + int ifindex; + struct list_head list; + int prog_fd; + int xsks_map_fd; + char ifname[IFNAMSIZ]; }; struct xsk_socket { struct xsk_ring_cons *rx; struct xsk_ring_prod *tx; __u64 outstanding_tx; - struct xsk_umem *umem; + struct xsk_ctx *ctx; struct xsk_socket_config config; int fd; - int ifindex; - int prog_fd; - int xsks_map_fd; - __u32 queue_id; - char ifname[IFNAMSIZ]; }; struct xsk_nl_info { @@ -203,15 +210,73 @@ static int xsk_get_mmap_offsets(int fd, struct xdp_mmap_offsets *off) return -EINVAL; } +static int xsk_create_umem_rings(struct xsk_umem *umem, int fd, + struct xsk_ring_prod *fill, + struct xsk_ring_cons *comp) +{ + struct xdp_mmap_offsets off; + void *map; + int err; + + err = setsockopt(fd, SOL_XDP, XDP_UMEM_FILL_RING, + &umem->config.fill_size, + sizeof(umem->config.fill_size)); + if (err) + return -errno; + + err = setsockopt(fd, SOL_XDP, XDP_UMEM_COMPLETION_RING, + &umem->config.comp_size, + sizeof(umem->config.comp_size)); + if (err) + return -errno; + + err = xsk_get_mmap_offsets(fd, &off); + if (err) + return -errno; + + map = mmap(NULL, off.fr.desc + umem->config.fill_size * sizeof(__u64), + PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, fd, + XDP_UMEM_PGOFF_FILL_RING); + if (map == MAP_FAILED) + return -errno; + + fill->mask = umem->config.fill_size - 1; + fill->size = umem->config.fill_size; + fill->producer = map + off.fr.producer; + fill->consumer = map + off.fr.consumer; + fill->flags = map + off.fr.flags; + fill->ring = map + off.fr.desc; + fill->cached_cons = umem->config.fill_size; + + map = mmap(NULL, off.cr.desc + umem->config.comp_size * sizeof(__u64), + PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, fd, + XDP_UMEM_PGOFF_COMPLETION_RING); + if (map == MAP_FAILED) { + err = -errno; + goto out_mmap; + } + + comp->mask = umem->config.comp_size - 1; + comp->size = umem->config.comp_size; + comp->producer = map + off.cr.producer; + comp->consumer = map + off.cr.consumer; + comp->flags = map + off.cr.flags; + comp->ring = map + off.cr.desc; + + return 0; + +out_mmap: + munmap(map, off.fr.desc + umem->config.fill_size * sizeof(__u64)); + return err; +} + int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area, __u64 size, struct xsk_ring_prod *fill, struct xsk_ring_cons *comp, const struct xsk_umem_config *usr_config) { - struct xdp_mmap_offsets off; struct xdp_umem_reg mr; struct xsk_umem *umem; - void *map; int err; if (!umem_area || !umem_ptr || !fill || !comp) @@ -230,6 +295,7 @@ int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area, } umem->umem_area = umem_area; + INIT_LIST_HEAD(&umem->ctx_list); xsk_set_umem_config(&umem->config, usr_config); memset(&mr, 0, sizeof(mr)); @@ -244,71 +310,16 @@ int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area, err = -errno; goto out_socket; } - err = setsockopt(umem->fd, SOL_XDP, XDP_UMEM_FILL_RING, - &umem->config.fill_size, - sizeof(umem->config.fill_size)); - if (err) { - err = -errno; - goto out_socket; - } - err = setsockopt(umem->fd, SOL_XDP, XDP_UMEM_COMPLETION_RING, - &umem->config.comp_size, - sizeof(umem->config.comp_size)); - if (err) { - err = -errno; - goto out_socket; - } - err = xsk_get_mmap_offsets(umem->fd, &off); - if (err) { - err = -errno; - goto out_socket; - } - - map = mmap(NULL, off.fr.desc + umem->config.fill_size * sizeof(__u64), - PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd, - XDP_UMEM_PGOFF_FILL_RING); - if (map == MAP_FAILED) { - err = -errno; + err = xsk_create_umem_rings(umem, umem->fd, fill, comp); + if (err) goto out_socket; - } - - umem->fill = fill; - fill->mask = umem->config.fill_size - 1; - fill->size = umem->config.fill_size; - fill->producer = map + off.fr.producer; - fill->consumer = map + off.fr.consumer; - fill->flags = map + off.fr.flags; - fill->ring = map + off.fr.desc; - fill->cached_prod = *fill->producer; - /* cached_cons is "size" bigger than the real consumer pointer - * See xsk_prod_nb_free - */ - fill->cached_cons = *fill->consumer + umem->config.fill_size; - - map = mmap(NULL, off.cr.desc + umem->config.comp_size * sizeof(__u64), - PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd, - XDP_UMEM_PGOFF_COMPLETION_RING); - if (map == MAP_FAILED) { - err = -errno; - goto out_mmap; - } - - umem->comp = comp; - comp->mask = umem->config.comp_size - 1; - comp->size = umem->config.comp_size; - comp->producer = map + off.cr.producer; - comp->consumer = map + off.cr.consumer; - comp->flags = map + off.cr.flags; - comp->ring = map + off.cr.desc; - comp->cached_prod = *comp->producer; - comp->cached_cons = *comp->consumer; + umem->fill_save = fill; + umem->comp_save = comp; *umem_ptr = umem; return 0; -out_mmap: - munmap(map, off.fr.desc + umem->config.fill_size * sizeof(__u64)); out_socket: close(umem->fd); out_umem_alloc: @@ -342,6 +353,7 @@ DEFAULT_VERSION(xsk_umem__create_v0_0_4, xsk_umem__create, LIBBPF_0.0.4) static int xsk_load_xdp_prog(struct xsk_socket *xsk) { static const int log_buf_size = 16 * 1024; + struct xsk_ctx *ctx = xsk->ctx; char log_buf[log_buf_size]; int err, prog_fd; @@ -369,7 +381,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk) /* *(u32 *)(r10 - 4) = r2 */ BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_2, -4), /* r1 = xskmap[] */ - BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd), + BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd), /* r3 = XDP_PASS */ BPF_MOV64_IMM(BPF_REG_3, 2), /* call bpf_redirect_map */ @@ -381,7 +393,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk) /* r2 += -4 */ BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r1 = xskmap[] */ - BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd), + BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd), /* call bpf_map_lookup_elem */ BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem), /* r1 = r0 */ @@ -393,7 +405,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk) /* r2 = *(u32 *)(r10 - 4) */ BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_10, -4), /* r1 = xskmap[] */ - BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd), + BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd), /* r3 = 0 */ BPF_MOV64_IMM(BPF_REG_3, 0), /* call bpf_redirect_map */ @@ -411,19 +423,21 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk) return prog_fd; } - err = bpf_set_link_xdp_fd(xsk->ifindex, prog_fd, xsk->config.xdp_flags); + err = bpf_set_link_xdp_fd(xsk->ctx->ifindex, prog_fd, + xsk->config.xdp_flags); if (err) { close(prog_fd); return err; } - xsk->prog_fd = prog_fd; + ctx->prog_fd = prog_fd; return 0; } static int xsk_get_max_queues(struct xsk_socket *xsk) { struct ethtool_channels channels = { .cmd = ETHTOOL_GCHANNELS }; + struct xsk_ctx *ctx = xsk->ctx; struct ifreq ifr = {}; int fd, err, ret; @@ -432,7 +446,7 @@ static int xsk_get_max_queues(struct xsk_socket *xsk) return -errno; ifr.ifr_data = (void *)&channels; - memcpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1); + memcpy(ifr.ifr_name, ctx->ifname, IFNAMSIZ - 1); ifr.ifr_name[IFNAMSIZ - 1] = '\0'; err = ioctl(fd, SIOCETHTOOL, &ifr); if (err && errno != EOPNOTSUPP) { @@ -460,6 +474,7 @@ out: static int xsk_create_bpf_maps(struct xsk_socket *xsk) { + struct xsk_ctx *ctx = xsk->ctx; int max_queues; int fd; @@ -472,15 +487,17 @@ static int xsk_create_bpf_maps(struct xsk_socket *xsk) if (fd < 0) return fd; - xsk->xsks_map_fd = fd; + ctx->xsks_map_fd = fd; return 0; } static void xsk_delete_bpf_maps(struct xsk_socket *xsk) { - bpf_map_delete_elem(xsk->xsks_map_fd, &xsk->queue_id); - close(xsk->xsks_map_fd); + struct xsk_ctx *ctx = xsk->ctx; + + bpf_map_delete_elem(ctx->xsks_map_fd, &ctx->queue_id); + close(ctx->xsks_map_fd); } static int xsk_lookup_bpf_maps(struct xsk_socket *xsk) @@ -488,10 +505,11 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk) __u32 i, *map_ids, num_maps, prog_len = sizeof(struct bpf_prog_info); __u32 map_len = sizeof(struct bpf_map_info); struct bpf_prog_info prog_info = {}; + struct xsk_ctx *ctx = xsk->ctx; struct bpf_map_info map_info; int fd, err; - err = bpf_obj_get_info_by_fd(xsk->prog_fd, &prog_info, &prog_len); + err = bpf_obj_get_info_by_fd(ctx->prog_fd, &prog_info, &prog_len); if (err) return err; @@ -505,11 +523,11 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk) prog_info.nr_map_ids = num_maps; prog_info.map_ids = (__u64)(unsigned long)map_ids; - err = bpf_obj_get_info_by_fd(xsk->prog_fd, &prog_info, &prog_len); + err = bpf_obj_get_info_by_fd(ctx->prog_fd, &prog_info, &prog_len); if (err) goto out_map_ids; - xsk->xsks_map_fd = -1; + ctx->xsks_map_fd = -1; for (i = 0; i < prog_info.nr_map_ids; i++) { fd = bpf_map_get_fd_by_id(map_ids[i]); @@ -523,7 +541,7 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk) } if (!strcmp(map_info.name, "xsks_map")) { - xsk->xsks_map_fd = fd; + ctx->xsks_map_fd = fd; continue; } @@ -531,7 +549,7 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk) } err = 0; - if (xsk->xsks_map_fd == -1) + if (ctx->xsks_map_fd == -1) err = -ENOENT; out_map_ids: @@ -541,16 +559,19 @@ out_map_ids: static int xsk_set_bpf_maps(struct xsk_socket *xsk) { - return bpf_map_update_elem(xsk->xsks_map_fd, &xsk->queue_id, + struct xsk_ctx *ctx = xsk->ctx; + + return bpf_map_update_elem(ctx->xsks_map_fd, &ctx->queue_id, &xsk->fd, 0); } static int xsk_setup_xdp_prog(struct xsk_socket *xsk) { + struct xsk_ctx *ctx = xsk->ctx; __u32 prog_id = 0; int err; - err = bpf_get_link_xdp_id(xsk->ifindex, &prog_id, + err = bpf_get_link_xdp_id(ctx->ifindex, &prog_id, xsk->config.xdp_flags); if (err) return err; @@ -566,12 +587,12 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk) return err; } } else { - xsk->prog_fd = bpf_prog_get_fd_by_id(prog_id); - if (xsk->prog_fd < 0) + ctx->prog_fd = bpf_prog_get_fd_by_id(prog_id); + if (ctx->prog_fd < 0) return -errno; err = xsk_lookup_bpf_maps(xsk); if (err) { - close(xsk->prog_fd); + close(ctx->prog_fd); return err; } } @@ -580,25 +601,110 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk) err = xsk_set_bpf_maps(xsk); if (err) { xsk_delete_bpf_maps(xsk); - close(xsk->prog_fd); + close(ctx->prog_fd); return err; } return 0; } -int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, - __u32 queue_id, struct xsk_umem *umem, - struct xsk_ring_cons *rx, struct xsk_ring_prod *tx, - const struct xsk_socket_config *usr_config) +static struct xsk_ctx *xsk_get_ctx(struct xsk_umem *umem, int ifindex, + __u32 queue_id) +{ + struct xsk_ctx *ctx; + + if (list_empty(&umem->ctx_list)) + return NULL; + + list_for_each_entry(ctx, &umem->ctx_list, list) { + if (ctx->ifindex == ifindex && ctx->queue_id == queue_id) { + ctx->refcount++; + return ctx; + } + } + + return NULL; +} + +static void xsk_put_ctx(struct xsk_ctx *ctx) +{ + struct xsk_umem *umem = ctx->umem; + struct xdp_mmap_offsets off; + int err; + + if (--ctx->refcount == 0) { + err = xsk_get_mmap_offsets(umem->fd, &off); + if (!err) { + munmap(ctx->fill->ring - off.fr.desc, + off.fr.desc + umem->config.fill_size * + sizeof(__u64)); + munmap(ctx->comp->ring - off.cr.desc, + off.cr.desc + umem->config.comp_size * + sizeof(__u64)); + } + + list_del(&ctx->list); + free(ctx); + } +} + +static struct xsk_ctx *xsk_create_ctx(struct xsk_socket *xsk, + struct xsk_umem *umem, int ifindex, + const char *ifname, __u32 queue_id, + struct xsk_ring_prod *fill, + struct xsk_ring_cons *comp) +{ + struct xsk_ctx *ctx; + int err; + + ctx = calloc(1, sizeof(*ctx)); + if (!ctx) + return NULL; + + if (!umem->fill_save) { + err = xsk_create_umem_rings(umem, xsk->fd, fill, comp); + if (err) { + free(ctx); + return NULL; + } + } else if (umem->fill_save != fill || umem->comp_save != comp) { + /* Copy over rings to new structs. */ + memcpy(fill, umem->fill_save, sizeof(*fill)); + memcpy(comp, umem->comp_save, sizeof(*comp)); + } + + ctx->ifindex = ifindex; + ctx->refcount = 1; + ctx->umem = umem; + ctx->queue_id = queue_id; + memcpy(ctx->ifname, ifname, IFNAMSIZ - 1); + ctx->ifname[IFNAMSIZ - 1] = '\0'; + + umem->fill_save = NULL; + umem->comp_save = NULL; + ctx->fill = fill; + ctx->comp = comp; + list_add(&ctx->list, &umem->ctx_list); + return ctx; +} + +int xsk_socket__create_shared(struct xsk_socket **xsk_ptr, + const char *ifname, + __u32 queue_id, struct xsk_umem *umem, + struct xsk_ring_cons *rx, + struct xsk_ring_prod *tx, + struct xsk_ring_prod *fill, + struct xsk_ring_cons *comp, + const struct xsk_socket_config *usr_config) { void *rx_map = NULL, *tx_map = NULL; struct sockaddr_xdp sxdp = {}; struct xdp_mmap_offsets off; struct xsk_socket *xsk; - int err; + struct xsk_ctx *ctx; + int err, ifindex; - if (!umem || !xsk_ptr || !(rx || tx)) + if (!umem || !xsk_ptr || !(rx || tx) || !fill || !comp) return -EFAULT; xsk = calloc(1, sizeof(*xsk)); @@ -609,10 +715,10 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, if (err) goto out_xsk_alloc; - if (umem->refcount && - !(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) { - pr_warn("Error: shared umems not supported by libbpf supplied XDP program.\n"); - err = -EBUSY; + xsk->outstanding_tx = 0; + ifindex = if_nametoindex(ifname); + if (!ifindex) { + err = -errno; goto out_xsk_alloc; } @@ -626,16 +732,16 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, xsk->fd = umem->fd; } - xsk->outstanding_tx = 0; - xsk->queue_id = queue_id; - xsk->umem = umem; - xsk->ifindex = if_nametoindex(ifname); - if (!xsk->ifindex) { - err = -errno; - goto out_socket; + ctx = xsk_get_ctx(umem, ifindex, queue_id); + if (!ctx) { + ctx = xsk_create_ctx(xsk, umem, ifindex, ifname, queue_id, + fill, comp); + if (!ctx) { + err = -ENOMEM; + goto out_socket; + } } - memcpy(xsk->ifname, ifname, IFNAMSIZ - 1); - xsk->ifname[IFNAMSIZ - 1] = '\0'; + xsk->ctx = ctx; if (rx) { err = setsockopt(xsk->fd, SOL_XDP, XDP_RX_RING, @@ -643,7 +749,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, sizeof(xsk->config.rx_size)); if (err) { err = -errno; - goto out_socket; + goto out_put_ctx; } } if (tx) { @@ -652,14 +758,14 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, sizeof(xsk->config.tx_size)); if (err) { err = -errno; - goto out_socket; + goto out_put_ctx; } } err = xsk_get_mmap_offsets(xsk->fd, &off); if (err) { err = -errno; - goto out_socket; + goto out_put_ctx; } if (rx) { @@ -669,7 +775,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, xsk->fd, XDP_PGOFF_RX_RING); if (rx_map == MAP_FAILED) { err = -errno; - goto out_socket; + goto out_put_ctx; } rx->mask = xsk->config.rx_size - 1; @@ -708,10 +814,10 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, xsk->tx = tx; sxdp.sxdp_family = PF_XDP; - sxdp.sxdp_ifindex = xsk->ifindex; - sxdp.sxdp_queue_id = xsk->queue_id; + sxdp.sxdp_ifindex = ctx->ifindex; + sxdp.sxdp_queue_id = ctx->queue_id; if (umem->refcount > 1) { - sxdp.sxdp_flags = XDP_SHARED_UMEM; + sxdp.sxdp_flags |= XDP_SHARED_UMEM; sxdp.sxdp_shared_umem_fd = umem->fd; } else { sxdp.sxdp_flags = xsk->config.bind_flags; @@ -723,7 +829,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, goto out_mmap_tx; } - xsk->prog_fd = -1; + ctx->prog_fd = -1; if (!(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) { err = xsk_setup_xdp_prog(xsk); @@ -742,6 +848,8 @@ out_mmap_rx: if (rx) munmap(rx_map, off.rx.desc + xsk->config.rx_size * sizeof(struct xdp_desc)); +out_put_ctx: + xsk_put_ctx(ctx); out_socket: if (--umem->refcount) close(xsk->fd); @@ -750,25 +858,24 @@ out_xsk_alloc: return err; } -int xsk_umem__delete(struct xsk_umem *umem) +int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname, + __u32 queue_id, struct xsk_umem *umem, + struct xsk_ring_cons *rx, struct xsk_ring_prod *tx, + const struct xsk_socket_config *usr_config) { - struct xdp_mmap_offsets off; - int err; + return xsk_socket__create_shared(xsk_ptr, ifname, queue_id, umem, + rx, tx, umem->fill_save, + umem->comp_save, usr_config); +} +int xsk_umem__delete(struct xsk_umem *umem) +{ if (!umem) return 0; if (umem->refcount) return -EBUSY; - err = xsk_get_mmap_offsets(umem->fd, &off); - if (!err) { - munmap(umem->fill->ring - off.fr.desc, - off.fr.desc + umem->config.fill_size * sizeof(__u64)); - munmap(umem->comp->ring - off.cr.desc, - off.cr.desc + umem->config.comp_size * sizeof(__u64)); - } - close(umem->fd); free(umem); @@ -778,15 +885,16 @@ int xsk_umem__delete(struct xsk_umem *umem) void xsk_socket__delete(struct xsk_socket *xsk) { size_t desc_sz = sizeof(struct xdp_desc); + struct xsk_ctx *ctx = xsk->ctx; struct xdp_mmap_offsets off; int err; if (!xsk) return; - if (xsk->prog_fd != -1) { + if (ctx->prog_fd != -1) { xsk_delete_bpf_maps(xsk); - close(xsk->prog_fd); + close(ctx->prog_fd); } err = xsk_get_mmap_offsets(xsk->fd, &off); @@ -799,14 +907,15 @@ void xsk_socket__delete(struct xsk_socket *xsk) munmap(xsk->tx->ring - off.tx.desc, off.tx.desc + xsk->config.tx_size * desc_sz); } - } - xsk->umem->refcount--; + xsk_put_ctx(ctx); + + ctx->umem->refcount--; /* Do not close an fd that also has an associated umem connected * to it. */ - if (xsk->fd != xsk->umem->fd) + if (xsk->fd != ctx->umem->fd) close(xsk->fd); free(xsk); } diff --git a/tools/lib/bpf/xsk.h b/tools/lib/bpf/xsk.h index 584f6820a639..1069c46364ff 100644 --- a/tools/lib/bpf/xsk.h +++ b/tools/lib/bpf/xsk.h @@ -234,6 +234,15 @@ LIBBPF_API int xsk_socket__create(struct xsk_socket **xsk, struct xsk_ring_cons *rx, struct xsk_ring_prod *tx, const struct xsk_socket_config *config); +LIBBPF_API int +xsk_socket__create_shared(struct xsk_socket **xsk_ptr, + const char *ifname, + __u32 queue_id, struct xsk_umem *umem, + struct xsk_ring_cons *rx, + struct xsk_ring_prod *tx, + struct xsk_ring_prod *fill, + struct xsk_ring_cons *comp, + const struct xsk_socket_config *config); /* Returns 0 for success and -EBUSY if the umem is still in use. */ LIBBPF_API int xsk_umem__delete(struct xsk_umem *umem); |