diff options
Diffstat (limited to 'tools/bpf/bpftool')
44 files changed, 8782 insertions, 1978 deletions
diff --git a/tools/bpf/bpftool/.gitignore b/tools/bpf/bpftool/.gitignore index b13926432b84..a736f64dc5dc 100644 --- a/tools/bpf/bpftool/.gitignore +++ b/tools/bpf/bpftool/.gitignore @@ -1,7 +1,10 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) *.d +/bootstrap/ /bpftool bpftool*.8 -bpf-helpers.* FEATURE-DUMP.bpftool feature libbpf +/*.skel.h +/vmlinux.h diff --git a/tools/bpf/bpftool/Documentation/Makefile b/tools/bpf/bpftool/Documentation/Makefile index 815ac9804aee..ac8487dcff1d 100644 --- a/tools/bpf/bpftool/Documentation/Makefile +++ b/tools/bpf/bpftool/Documentation/Makefile @@ -1,6 +1,5 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) include ../../../scripts/Makefile.include -include ../../../scripts/utilities.mak INSTALL ?= install RM ?= rm -f @@ -16,35 +15,43 @@ prefix ?= /usr/local mandir ?= $(prefix)/man man8dir = $(mandir)/man8 -# Load targets for building eBPF helpers man page. -include ../../Makefile.helpers - -MAN8_RST = $(filter-out $(HELPERS_RST),$(wildcard *.rst)) +MAN8_RST = $(wildcard bpftool*.rst) _DOC_MAN8 = $(patsubst %.rst,%.8,$(MAN8_RST)) DOC_MAN8 = $(addprefix $(OUTPUT),$(_DOC_MAN8)) -man: man8 helpers +man: man8 man8: $(DOC_MAN8) RST2MAN_DEP := $(shell command -v rst2man 2>/dev/null) +RST2MAN_OPTS += --verbose --strip-comments + +list_pages = $(sort $(basename $(filter-out $(1),$(MAN8_RST)))) +see_also = $(subst " ",, \ + "\n" \ + "SEE ALSO\n" \ + "========\n" \ + "\t**bpf**\ (2),\n" \ + "\t**bpf-helpers**\\ (7)" \ + $(foreach page,$(call list_pages,$(1)),",\n\t**$(page)**\\ (8)") \ + "\n") $(OUTPUT)%.8: %.rst ifndef RST2MAN_DEP $(error "rst2man not found, but required to generate man pages") endif - $(QUIET_GEN)rst2man $< > $@ + $(QUIET_GEN)( cat $< ; printf "%b" $(call see_also,$<) ) | rst2man $(RST2MAN_OPTS) > $@ -clean: helpers-clean +clean: $(call QUIET_CLEAN, Documentation) $(Q)$(RM) $(DOC_MAN8) -install: man helpers-install +install: man $(call QUIET_INSTALL, Documentation-man) $(Q)$(INSTALL) -d -m 755 $(DESTDIR)$(man8dir) $(Q)$(INSTALL) -m 644 $(DOC_MAN8) $(DESTDIR)$(man8dir) -uninstall: helpers-uninstall +uninstall: $(call QUIET_UNINST, Documentation-man) $(Q)$(RM) $(addprefix $(DESTDIR)$(man8dir)/,$(_DOC_MAN8)) $(Q)$(RMDIR) $(DESTDIR)$(man8dir) diff --git a/tools/bpf/bpftool/Documentation/bpftool-btf.rst b/tools/bpf/bpftool/Documentation/bpftool-btf.rst index 39615f8e145b..342716f74ec4 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-btf.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-btf.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-btf ================ @@ -7,12 +9,14 @@ tool for inspection of BTF data :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **btf** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] } + *OPTIONS* := { |COMMON_OPTIONS| | { **-B** | **--base-btf** } } *COMMANDS* := { **dump** | **help** } @@ -36,6 +40,11 @@ DESCRIPTION otherwise list all BTF objects currently loaded on the system. + Since Linux 5.8 bpftool is able to discover information about + processes that hold open file descriptors (FDs) against BTF + objects. On such kernels bpftool will automatically emit this + information as well. + **bpftool btf dump** *BTF_SRC* Dump BTF entries from a given *BTF_SRC*. @@ -66,26 +75,26 @@ DESCRIPTION OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. - - -d, --debug - Print all logs available from libbpf, including debug-level - information. + .. include:: common_options.rst + + -B, --base-btf *FILE* + Pass a base BTF object. Base BTF objects are typically used + with BTF objects for kernel modules. To avoid duplicating + all kernel symbols required by modules, BTF objects for + modules are "split", they are built incrementally on top of + the kernel (vmlinux) BTF object. So the base BTF reference + should usually point to the kernel BTF. + + When the main BTF object to process (for example, the + module BTF to dump) is passed as a *FILE*, bpftool attempts + to autodetect the path for the base object, and passing + this option is optional. When the main BTF object is passed + through other handles, this option becomes necessary. EXAMPLES ======== **# bpftool btf dump id 1226** + :: [1] PTR '(anon)' type_id=2 @@ -99,6 +108,7 @@ EXAMPLES This gives an example of default output for all supported BTF kinds. **$ cat prog.c** + :: struct fwd_struct; @@ -139,6 +149,7 @@ This gives an example of default output for all supported BTF kinds. } **$ bpftool btf dump file prog.o** + :: [1] PTR '(anon)' type_id=2 @@ -225,14 +236,33 @@ All the standard ways to specify map or program are supported: **# bpftool btf dump prog pinned /sys/fs/bpf/prog_name** -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-map**\ (8), - **bpftool-prog**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8) +| +| **# bpftool btf dump file /sys/kernel/btf/i2c_smbus** +| (or) +| **# I2C_SMBUS_ID=$(bpftool btf show -p | jq '.[] | select(.name=="i2c_smbus").id')** +| **# bpftool btf dump id ${I2C_SMBUS_ID} -B /sys/kernel/btf/vmlinux** + +:: + + [104848] STRUCT 'i2c_smbus_alert' size=40 vlen=2 + 'alert' type_id=393 bits_offset=0 + 'ara' type_id=56050 bits_offset=256 + [104849] STRUCT 'alert_data' size=12 vlen=3 + 'addr' type_id=16 bits_offset=0 + 'type' type_id=56053 bits_offset=32 + 'data' type_id=7 bits_offset=64 + [104850] PTR '(anon)' type_id=104848 + [104851] PTR '(anon)' type_id=104849 + [104852] FUNC 'i2c_register_spd' type_id=84745 linkage=static + [104853] FUNC 'smbalert_driver_init' type_id=1213 linkage=static + [104854] FUNC_PROTO '(anon)' ret_type_id=18 vlen=1 + 'ara' type_id=56050 + [104855] FUNC 'i2c_handle_smbus_alert' type_id=104854 linkage=static + [104856] FUNC 'smbalert_remove' type_id=104854 linkage=static + [104857] FUNC_PROTO '(anon)' ret_type_id=18 vlen=2 + 'ara' type_id=56050 + 'id' type_id=56056 + [104858] FUNC 'smbalert_probe' type_id=104857 linkage=static + [104859] FUNC 'smbalert_work' type_id=9695 linkage=static + [104860] FUNC 'smbus_alert' type_id=71367 linkage=static + [104861] FUNC 'smbus_do_alert' type_id=84827 linkage=static diff --git a/tools/bpf/bpftool/Documentation/bpftool-cgroup.rst b/tools/bpf/bpftool/Documentation/bpftool-cgroup.rst index 06a28b07787d..bd015ec9847b 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-cgroup.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-cgroup.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-cgroup ================ @@ -7,12 +9,14 @@ tool for inspection and simple manipulation of eBPF progs :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **cgroup** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-f** | **--bpffs** } } + *OPTIONS* := { |COMMON_OPTIONS| | { **-f** | **--bpffs** } } *COMMANDS* := { **show** | **list** | **tree** | **attach** | **detach** | **help** } @@ -20,17 +24,24 @@ SYNOPSIS CGROUP COMMANDS =============== -| **bpftool** **cgroup { show | list }** *CGROUP* [**effective**] +| **bpftool** **cgroup** { **show** | **list** } *CGROUP* [**effective**] | **bpftool** **cgroup tree** [*CGROUP_ROOT*] [**effective**] | **bpftool** **cgroup attach** *CGROUP* *ATTACH_TYPE* *PROG* [*ATTACH_FLAGS*] | **bpftool** **cgroup detach** *CGROUP* *ATTACH_TYPE* *PROG* | **bpftool** **cgroup help** | | *PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* } -| *ATTACH_TYPE* := { **ingress** | **egress** | **sock_create** | **sock_ops** | **device** | -| **bind4** | **bind6** | **post_bind4** | **post_bind6** | **connect4** | **connect6** | -| **sendmsg4** | **sendmsg6** | **recvmsg4** | **recvmsg6** | **sysctl** | -| **getsockopt** | **setsockopt** } +| *ATTACH_TYPE* := { **cgroup_inet_ingress** | **cgroup_inet_egress** | +| **cgroup_inet_sock_create** | **cgroup_sock_ops** | +| **cgroup_device** | **cgroup_inet4_bind** | **cgroup_inet6_bind** | +| **cgroup_inet4_post_bind** | **cgroup_inet6_post_bind** | +| **cgroup_inet4_connect** | **cgroup_inet6_connect** | +| **cgroup_inet4_getpeername** | **cgroup_inet6_getpeername** | +| **cgroup_inet4_getsockname** | **cgroup_inet6_getsockname** | +| **cgroup_udp4_sendmsg** | **cgroup_udp6_sendmsg** | +| **cgroup_udp4_recvmsg** | **cgroup_udp6_recvmsg** | +| **cgroup_sysctl** | **cgroup_getsockopt** | **cgroup_setsockopt** | +| **cgroup_inet_sock_release** } | *ATTACH_FLAGS* := { **multi** | **override** } DESCRIPTION @@ -96,12 +107,17 @@ DESCRIPTION **sendmsg6** call to sendto(2), sendmsg(2), sendmmsg(2) for an unconnected udp6 socket (since 4.18); **recvmsg4** call to recvfrom(2), recvmsg(2), recvmmsg(2) for - an unconnected udp4 socket (since 5.2); + an unconnected udp4 socket (since 5.2); **recvmsg6** call to recvfrom(2), recvmsg(2), recvmmsg(2) for - an unconnected udp6 socket (since 5.2); + an unconnected udp6 socket (since 5.2); **sysctl** sysctl access (since 5.2); **getsockopt** call to getsockopt (since 5.3); - **setsockopt** call to setsockopt (since 5.3). + **setsockopt** call to setsockopt (since 5.3); + **getpeername4** call to getpeername(2) for an inet4 socket (since 5.8); + **getpeername6** call to getpeername(2) for an inet6 socket (since 5.8); + **getsockname4** call to getsockname(2) for an inet4 socket (since 5.8); + **getsockname6** call to getsockname(2) for an inet6 socket (since 5.8). + **sock_release** closing an userspace inet socket (since 5.9). **bpftool cgroup detach** *CGROUP* *ATTACH_TYPE* *PROG* Detach *PROG* from the cgroup *CGROUP* and attach type @@ -112,26 +128,11 @@ DESCRIPTION OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. + .. include:: common_options.rst -f, --bpffs Show file names of pinned programs. - -d, --debug - Print all logs available from libbpf, including debug-level - information. - EXAMPLES ======== | @@ -154,15 +155,3 @@ EXAMPLES :: ID AttachType AttachFlags Name - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-prog**\ (8), - **bpftool-map**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) diff --git a/tools/bpf/bpftool/Documentation/bpftool-feature.rst b/tools/bpf/bpftool/Documentation/bpftool-feature.rst index 4d08f35034a2..e44039f89be7 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-feature.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-feature.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + =============== bpftool-feature =============== @@ -7,31 +9,40 @@ tool for inspection of eBPF-related parameters for Linux kernel or net device :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **feature** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] } + *OPTIONS* := { |COMMON_OPTIONS| } *COMMANDS* := { **probe** | **help** } FEATURE COMMANDS ================ -| **bpftool** **feature probe** [*COMPONENT*] [**macros** [**prefix** *PREFIX*]] +| **bpftool** **feature probe** [*COMPONENT*] [**full**] [**unprivileged**] [**macros** [**prefix** *PREFIX*]] +| **bpftool** **feature list_builtins** *GROUP* | **bpftool** **feature help** | | *COMPONENT* := { **kernel** | **dev** *NAME* } +| *GROUP* := { **prog_types** | **map_types** | **attach_types** | **link_types** | **helpers** } DESCRIPTION =========== - **bpftool feature probe** [**kernel**] [**macros** [**prefix** *PREFIX*]] + **bpftool feature probe** [**kernel**] [**full**] [**macros** [**prefix** *PREFIX*]] Probe the running kernel and dump a number of eBPF-related - parameters, such as availability of the **bpf()** system call, + parameters, such as availability of the **bpf**\ () system call, JIT status, eBPF program types availability, eBPF helper functions availability, and more. + By default, bpftool **does not run probes** for + **bpf_probe_write_user**\ () and **bpf_trace_printk**\() + helpers which print warnings to kernel logs. To enable them + and run all probes, the **full** keyword should be used. + If the **macros** keyword (but not the **-j** option) is passed, a subset of the output is dumped as a list of **#define** macros that are ready to be included in a C @@ -44,47 +55,36 @@ DESCRIPTION Keyword **kernel** can be omitted. If no probe target is specified, probing the kernel is the default behaviour. - Note that when probed, some eBPF helpers (e.g. - **bpf_trace_printk**\ () or **bpf_probe_write_user**\ ()) may - print warnings to kernel logs. - - **bpftool feature probe dev** *NAME* [**macros** [**prefix** *PREFIX*]] + When the **unprivileged** keyword is used, bpftool will dump + only the features available to a user who does not have the + **CAP_SYS_ADMIN** capability set. The features available in + that case usually represent a small subset of the parameters + supported by the system. Unprivileged users MUST use the + **unprivileged** keyword: This is to avoid misdetection if + bpftool is inadvertently run as non-root, for example. This + keyword is unavailable if bpftool was compiled without + libcap. + + **bpftool feature probe dev** *NAME* [**full**] [**macros** [**prefix** *PREFIX*]] Probe network device for supported eBPF features and dump results to the console. - The two keywords **macros** and **prefix** have the same - role as when probing the kernel. + The keywords **full**, **macros** and **prefix** have the + same role as when probing the kernel. + + **bpftool feature list_builtins** *GROUP* + List items known to bpftool. These can be BPF program types + (**prog_types**), BPF map types (**map_types**), attach types + (**attach_types**), link types (**link_types**), or BPF helper + functions (**helpers**). The command does not probe the system, but + simply lists the elements that bpftool knows from compilation time, + as provided from libbpf (for all object types) or from the BPF UAPI + header (list of helpers). This can be used in scripts to iterate over + BPF types or helpers. **bpftool feature help** Print short help message. OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. - - -d, --debug - Print all logs available from libbpf, including debug-level - information. - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-prog**\ (8), - **bpftool-map**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) + .. include:: common_options.rst diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst index 94d91322895a..68454ef28f58 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-gen.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-gen ================ @@ -7,23 +9,48 @@ tool for BPF code-generation :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **gen** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] } + *OPTIONS* := { |COMMON_OPTIONS| | { **-L** | **--use-loader** } } - *COMMAND* := { **skeleton | **help** } + *COMMAND* := { **object** | **skeleton** | **help** } GEN COMMANDS ============= -| **bpftool** **gen skeleton** *FILE* +| **bpftool** **gen object** *OUTPUT_FILE* *INPUT_FILE* [*INPUT_FILE*...] +| **bpftool** **gen skeleton** *FILE* [**name** *OBJECT_NAME*] +| **bpftool** **gen subskeleton** *FILE* [**name** *OBJECT_NAME*] +| **bpftool** **gen min_core_btf** *INPUT* *OUTPUT* *OBJECT* [*OBJECT*...] | **bpftool** **gen help** DESCRIPTION =========== + **bpftool gen object** *OUTPUT_FILE* *INPUT_FILE* [*INPUT_FILE*...] + Statically link (combine) together one or more *INPUT_FILE*'s + into a single resulting *OUTPUT_FILE*. All the files involved + are BPF ELF object files. + + The rules of BPF static linking are mostly the same as for + user-space object files, but in addition to combining data + and instruction sections, .BTF and .BTF.ext (if present in + any of the input files) data are combined together. .BTF + data is deduplicated, so all the common types across + *INPUT_FILE*'s will only be represented once in the resulting + BTF information. + + BPF static linking allows to partition BPF source code into + individually compiled files that are then linked into + a single resulting BPF object file, which can be used to + generated BPF skeleton (with **gen skeleton** command) or + passed directly into **libbpf** (using **bpf_object__open()** + family of APIs). + **bpftool gen skeleton** *FILE* Generate BPF skeleton C header file for a given *FILE*. @@ -36,12 +63,12 @@ DESCRIPTION etc. Skeleton eliminates the need to lookup mentioned components by name. Instead, if skeleton instantiation succeeds, they are populated in skeleton structure as valid - libbpf types (e.g., struct bpf_map pointer) and can be + libbpf types (e.g., **struct bpf_map** pointer) and can be passed to existing generic libbpf APIs. In addition to simple and reliable access to maps and - programs, skeleton provides a storage for BPF links (struct - bpf_link) for each BPF program within BPF object. When + programs, skeleton provides a storage for BPF links (**struct + bpf_link**) for each BPF program within BPF object. When requested, supported BPF programs will be automatically attached and resulting BPF links stored for further use by user in pre-allocated fields in skeleton struct. For BPF @@ -75,21 +102,24 @@ DESCRIPTION specific maps, programs, etc. As part of skeleton, few custom functions are generated. - Each of them is prefixed with object name, derived from - object file name. I.e., if BPF object file name is - **example.o**, BPF object name will be **example**. The - following custom functions are provided in such case: + Each of them is prefixed with object name. Object name can + either be derived from object file name, i.e., if BPF object + file name is **example.o**, BPF object name will be + **example**. Object name can be also specified explicitly + through **name** *OBJECT_NAME* parameter. The following + custom functions are provided (assuming **example** as + the object name): - **example__open** and **example__open_opts**. These functions are used to instantiate skeleton. It - corresponds to libbpf's **bpf_object__open()** API. + corresponds to libbpf's **bpf_object__open**\ () API. **_opts** variants accepts extra **bpf_object_open_opts** options. - **example__load**. This function creates maps, loads and verifies BPF programs, initializes global data maps. It corresponds to - libppf's **bpf_object__load** API. + libppf's **bpf_object__load**\ () API. - **example__open_and_load** combines **example__open** and **example__load** invocations in one commonly used @@ -121,49 +151,78 @@ DESCRIPTION (non-read-only) data from userspace, with same simplicity as for BPF side. + **bpftool gen subskeleton** *FILE* + Generate BPF subskeleton C header file for a given *FILE*. + + Subskeletons are similar to skeletons, except they do not own + the corresponding maps, programs, or global variables. They + require that the object file used to generate them is already + loaded into a *bpf_object* by some other means. + + This functionality is useful when a library is included into a + larger BPF program. A subskeleton for the library would have + access to all objects and globals defined in it, without + having to know about the larger program. + + Consequently, there are only two functions defined + for subskeletons: + + - **example__open(bpf_object\*)** + Instantiates a subskeleton from an already opened (but not + necessarily loaded) **bpf_object**. + + - **example__destroy()** + Frees the storage for the subskeleton but *does not* unload + any BPF programs or maps. + + **bpftool** **gen min_core_btf** *INPUT* *OUTPUT* *OBJECT* [*OBJECT*...] + Generate a minimum BTF file as *OUTPUT*, derived from a given + *INPUT* BTF file, containing all needed BTF types so one, or + more, given eBPF objects CO-RE relocations may be satisfied. + + When kernels aren't compiled with CONFIG_DEBUG_INFO_BTF, + libbpf, when loading an eBPF object, has to rely on external + BTF files to be able to calculate CO-RE relocations. + + Usually, an external BTF file is built from existing kernel + DWARF data using pahole. It contains all the types used by + its respective kernel image and, because of that, is big. + + The min_core_btf feature builds smaller BTF files, customized + to one or multiple eBPF objects, so they can be distributed + together with an eBPF CO-RE based application, turning the + application portable to different kernel versions. + + Check examples bellow for more information how to use it. + **bpftool gen help** Print short help message. OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, - this option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. + .. include:: common_options.rst - -d, --debug - Print all logs available from libbpf, including debug-level - information. + -L, --use-loader + For skeletons, generate a "light" skeleton (also known as "loader" + skeleton). A light skeleton contains a loader eBPF program. It does + not use the majority of the libbpf infrastructure, and does not need + libelf. EXAMPLES ======== -**$ cat example.c** +**$ cat example1.bpf.c** + :: #include <stdbool.h> #include <linux/ptrace.h> #include <linux/bpf.h> - #include "bpf_helpers.h" + #include <bpf/bpf_helpers.h> const volatile int param1 = 42; bool global_flag = true; struct { int x; } data = {}; - struct { - __uint(type, BPF_MAP_TYPE_HASH); - __uint(max_entries, 128); - __type(key, int); - __type(value, long); - } my_map SEC(".maps"); - SEC("raw_tp/sys_enter") int handle_sys_enter(struct pt_regs *ctx) { @@ -175,6 +234,21 @@ EXAMPLES return 0; } +**$ cat example2.bpf.c** + +:: + + #include <linux/ptrace.h> + #include <linux/bpf.h> + #include <bpf/bpf_helpers.h> + + struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 128); + __type(key, int); + __type(value, long); + } my_map SEC(".maps"); + SEC("raw_tp/sys_exit") int handle_sys_exit(struct pt_regs *ctx) { @@ -184,9 +258,20 @@ EXAMPLES } This is example BPF application with two BPF programs and a mix of BPF maps -and global variables. +and global variables. Source code is split across two source code files. + +**$ clang -target bpf -g example1.bpf.c -o example1.bpf.o** + +**$ clang -target bpf -g example2.bpf.c -o example2.bpf.o** + +**$ bpftool gen object example.bpf.o example1.bpf.o example2.bpf.o** + +This set of commands compiles *example1.bpf.c* and *example2.bpf.c* +individually and then statically links respective object files into the final +BPF ELF object file *example.bpf.o*. + +**$ bpftool gen skeleton example.bpf.o name example | tee example.skel.h** -**$ bpftool gen skeleton example.o** :: /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ @@ -240,7 +325,8 @@ and global variables. #endif /* __EXAMPLE_SKEL_H__ */ -**$ cat example_user.c** +**$ cat example.c** + :: #include "example.skel.h" @@ -282,7 +368,8 @@ and global variables. return err; } -**# ./example_user** +**# ./example** + :: my_map name: my_map @@ -291,15 +378,69 @@ and global variables. This is a stripped-out version of skeleton generated for above example code. -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-map**\ (8), - **bpftool-prog**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) +min_core_btf +------------ + +**$ bpftool btf dump file 5.4.0-example.btf format raw** + +:: + + [1] INT 'long unsigned int' size=8 bits_offset=0 nr_bits=64 encoding=(none) + [2] CONST '(anon)' type_id=1 + [3] VOLATILE '(anon)' type_id=1 + [4] ARRAY '(anon)' type_id=1 index_type_id=21 nr_elems=2 + [5] PTR '(anon)' type_id=8 + [6] CONST '(anon)' type_id=5 + [7] INT 'char' size=1 bits_offset=0 nr_bits=8 encoding=(none) + [8] CONST '(anon)' type_id=7 + [9] INT 'unsigned int' size=4 bits_offset=0 nr_bits=32 encoding=(none) + <long output> + +**$ bpftool btf dump file one.bpf.o format raw** + +:: + + [1] PTR '(anon)' type_id=2 + [2] STRUCT 'trace_event_raw_sys_enter' size=64 vlen=4 + 'ent' type_id=3 bits_offset=0 + 'id' type_id=7 bits_offset=64 + 'args' type_id=9 bits_offset=128 + '__data' type_id=12 bits_offset=512 + [3] STRUCT 'trace_entry' size=8 vlen=4 + 'type' type_id=4 bits_offset=0 + 'flags' type_id=5 bits_offset=16 + 'preempt_count' type_id=5 bits_offset=24 + <long output> + +**$ bpftool gen min_core_btf 5.4.0-example.btf 5.4.0-smaller.btf one.bpf.o** + +**$ bpftool btf dump file 5.4.0-smaller.btf format raw** + +:: + + [1] TYPEDEF 'pid_t' type_id=6 + [2] STRUCT 'trace_event_raw_sys_enter' size=64 vlen=1 + 'args' type_id=4 bits_offset=128 + [3] STRUCT 'task_struct' size=9216 vlen=2 + 'pid' type_id=1 bits_offset=17920 + 'real_parent' type_id=7 bits_offset=18048 + [4] ARRAY '(anon)' type_id=5 index_type_id=8 nr_elems=6 + [5] INT 'long unsigned int' size=8 bits_offset=0 nr_bits=64 encoding=(none) + [6] TYPEDEF '__kernel_pid_t' type_id=8 + [7] PTR '(anon)' type_id=3 + [8] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED + <end> + +Now, the "5.4.0-smaller.btf" file may be used by libbpf as an external BTF file +when loading the "one.bpf.o" object into the "5.4.0-example" kernel. Note that +the generated BTF file won't allow other eBPF objects to be loaded, just the +ones given to min_core_btf. + +:: + + LIBBPF_OPTS(bpf_object_open_opts, opts, .btf_custom_path = "5.4.0-smaller.btf"); + struct bpf_object *obj; + + obj = bpf_object__open_file("one.bpf.o", &opts); + + ... diff --git a/tools/bpf/bpftool/Documentation/bpftool-iter.rst b/tools/bpf/bpftool/Documentation/bpftool-iter.rst new file mode 100644 index 000000000000..84839d488621 --- /dev/null +++ b/tools/bpf/bpftool/Documentation/bpftool-iter.rst @@ -0,0 +1,76 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +============ +bpftool-iter +============ +------------------------------------------------------------------------------- +tool to create BPF iterators +------------------------------------------------------------------------------- + +:Manual section: 8 + +.. include:: substitutions.rst + +SYNOPSIS +======== + + **bpftool** [*OPTIONS*] **iter** *COMMAND* + + *OPTIONS* := { |COMMON_OPTIONS| } + + *COMMANDS* := { **pin** | **help** } + +ITER COMMANDS +=================== + +| **bpftool** **iter pin** *OBJ* *PATH* [**map** *MAP*] +| **bpftool** **iter help** +| +| *OBJ* := /a/file/of/bpf_iter_target.o +| *MAP* := { **id** *MAP_ID* | **pinned** *FILE* } + +DESCRIPTION +=========== + **bpftool iter pin** *OBJ* *PATH* [**map** *MAP*] + A bpf iterator combines a kernel iterating of + particular kernel data (e.g., tasks, bpf_maps, etc.) + and a bpf program called for each kernel data object + (e.g., one task, one bpf_map, etc.). User space can + *read* kernel iterator output through *read()* syscall. + + The *pin* command creates a bpf iterator from *OBJ*, + and pin it to *PATH*. The *PATH* should be located + in *bpffs* mount. It must not contain a dot + character ('.'), which is reserved for future extensions + of *bpffs*. + + Map element bpf iterator requires an additional parameter + *MAP* so bpf program can iterate over map elements for + that map. User can have a bpf program in kernel to run + with each map element, do checking, filtering, aggregation, + etc. without copying data to user space. + + User can then *cat PATH* to see the bpf iterator output. + + **bpftool iter help** + Print short help message. + +OPTIONS +======= + .. include:: common_options.rst + +EXAMPLES +======== +**# bpftool iter pin bpf_iter_netlink.o /sys/fs/bpf/my_netlink** + +:: + + Create a file-based bpf iterator from bpf_iter_netlink.o and pin it + to /sys/fs/bpf/my_netlink + +**# bpftool iter pin bpf_iter_hashmap.o /sys/fs/bpf/my_hashmap map id 20** + +:: + + Create a file-based bpf iterator from bpf_iter_hashmap.o and map with + id 20, and pin it to /sys/fs/bpf/my_hashmap diff --git a/tools/bpf/bpftool/Documentation/bpftool-link.rst b/tools/bpf/bpftool/Documentation/bpftool-link.rst new file mode 100644 index 000000000000..52a4eee4af54 --- /dev/null +++ b/tools/bpf/bpftool/Documentation/bpftool-link.rst @@ -0,0 +1,112 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +================ +bpftool-link +================ +------------------------------------------------------------------------------- +tool for inspection and simple manipulation of eBPF links +------------------------------------------------------------------------------- + +:Manual section: 8 + +.. include:: substitutions.rst + +SYNOPSIS +======== + + **bpftool** [*OPTIONS*] **link** *COMMAND* + + *OPTIONS* := { |COMMON_OPTIONS| | { **-f** | **--bpffs** } | { **-n** | **--nomount** } } + + *COMMANDS* := { **show** | **list** | **pin** | **help** } + +LINK COMMANDS +============= + +| **bpftool** **link { show | list }** [*LINK*] +| **bpftool** **link pin** *LINK* *FILE* +| **bpftool** **link detach** *LINK* +| **bpftool** **link help** +| +| *LINK* := { **id** *LINK_ID* | **pinned** *FILE* } + + +DESCRIPTION +=========== + **bpftool link { show | list }** [*LINK*] + Show information about active links. If *LINK* is + specified show information only about given link, + otherwise list all links currently active on the system. + + Output will start with link ID followed by link type and + zero or more named attributes, some of which depend on type + of link. + + Since Linux 5.8 bpftool is able to discover information about + processes that hold open file descriptors (FDs) against BPF + links. On such kernels bpftool will automatically emit this + information as well. + + **bpftool link pin** *LINK* *FILE* + Pin link *LINK* as *FILE*. + + Note: *FILE* must be located in *bpffs* mount. It must not + contain a dot character ('.'), which is reserved for future + extensions of *bpffs*. + + **bpftool link detach** *LINK* + Force-detach link *LINK*. BPF link and its underlying BPF + program will stay valid, but they will be detached from the + respective BPF hook and BPF link will transition into + a defunct state until last open file descriptor for that + link is closed. + + **bpftool link help** + Print short help message. + +OPTIONS +======= + .. include:: common_options.rst + + -f, --bpffs + When showing BPF links, show file names of pinned + links. + + -n, --nomount + Do not automatically attempt to mount any virtual file system + (such as tracefs or BPF virtual file system) when necessary. + +EXAMPLES +======== +**# bpftool link show** + +:: + + 10: cgroup prog 25 + cgroup_id 614 attach_type egress + pids test_progs(223) + +**# bpftool --json --pretty link show** + +:: + + [{ + "type": "cgroup", + "prog_id": 25, + "cgroup_id": 614, + "attach_type": "egress", + "pids": [{ + "pid": 223, + "comm": "test_progs" + } + ] + } + ] + +| +| **# bpftool link pin id 10 /sys/fs/bpf/link** +| **# ls -l /sys/fs/bpf/** + +:: + + -rw------- 1 root root 0 Apr 23 21:39 link diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.rst b/tools/bpf/bpftool/Documentation/bpftool-map.rst index cdeae8ae90ba..7f3b67a8b48f 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-map.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-map.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-map ================ @@ -7,23 +9,26 @@ tool for inspection and simple manipulation of eBPF maps :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **map** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-f** | **--bpffs** } } + *OPTIONS* := { |COMMON_OPTIONS| | { **-f** | **--bpffs** } | { **-n** | **--nomount** } } *COMMANDS* := - { **show** | **list** | **create** | **dump** | **update** | **lookup** | **getnext** - | **delete** | **pin** | **help** } + { **show** | **list** | **create** | **dump** | **update** | **lookup** | **getnext** | + **delete** | **pin** | **help** } MAP COMMANDS ============= -| **bpftool** **map { show | list }** [*MAP*] +| **bpftool** **map** { **show** | **list** } [*MAP*] | **bpftool** **map create** *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE* \ -| **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**dev** *NAME*] +| **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**inner_map** *MAP*] \ +| [**dev** *NAME*] | **bpftool** **map dump** *MAP* | **bpftool** **map update** *MAP* [**key** *DATA*] [**value** *VALUE*] [*UPDATE_FLAGS*] | **bpftool** **map lookup** *MAP* [**key** *DATA*] @@ -49,7 +54,8 @@ MAP COMMANDS | | **lru_percpu_hash** | **lpm_trie** | **array_of_maps** | **hash_of_maps** | | **devmap** | **devmap_hash** | **sockmap** | **cpumap** | **xskmap** | **sockhash** | | **cgroup_storage** | **reuseport_sockarray** | **percpu_cgroup_storage** -| | **queue** | **stack** } +| | **queue** | **stack** | **sk_storage** | **struct_ops** | **ringbuf** | **inode_storage** +| | **task_storage** | **bloom_filter** | **user_ringbuf** } DESCRIPTION =========== @@ -62,10 +68,27 @@ DESCRIPTION Output will start with map ID followed by map type and zero or more named attributes (depending on kernel version). - **bpftool map create** *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE* **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**dev** *NAME*] + Since Linux 5.8 bpftool is able to discover information about + processes that hold open file descriptors (FDs) against BPF + maps. On such kernels bpftool will automatically emit this + information as well. + + **bpftool map create** *FILE* **type** *TYPE* **key** *KEY_SIZE* **value** *VALUE_SIZE* **entries** *MAX_ENTRIES* **name** *NAME* [**flags** *FLAGS*] [**inner_map** *MAP*] [**dev** *NAME*] Create a new map with given parameters and pin it to *bpffs* as *FILE*. + *FLAGS* should be an integer which is the combination of + desired flags, e.g. 1024 for **BPF_F_MMAPABLE** (see bpf.h + UAPI header for existing flags). + + To create maps of type array-of-maps or hash-of-maps, the + **inner_map** keyword must be used to pass an inner map. The + kernel needs it to collect metadata related to the inner maps + that the new map will work with. + + Keyword **dev** expects a network interface name, and is used + to request hardware offload for the map. + **bpftool map dump** *MAP* Dump all entries in a given *MAP*. In case of **name**, *MAP* may match several maps which will all be dumped. @@ -78,7 +101,7 @@ DESCRIPTION exists; **noexist** update only if entry doesn't exist. If the **hex** keyword is provided in front of the bytes - sequence, the bytes are parsed as hexadeximal values, even if + sequence, the bytes are parsed as hexadecimal values, even if no "0x" prefix is added. If the keyword is not provided, then the bytes are parsed as decimal values, unless a "0x" prefix (for hexadecimal) or a "0" prefix (for octal) is provided. @@ -100,10 +123,10 @@ DESCRIPTION extensions of *bpffs*. **bpftool** **map event_pipe** *MAP* [**cpu** *N* **index** *M*] - Read events from a BPF_MAP_TYPE_PERF_EVENT_ARRAY map. + Read events from a **BPF_MAP_TYPE_PERF_EVENT_ARRAY** map. Install perf rings into a perf event array map and dump - output of any bpf_perf_event_output() call in the kernel. + output of any **bpf_perf_event_output**\ () call in the kernel. By default read the number of CPUs on the system and install perf ring for each CPU in the corresponding index in the array. @@ -116,24 +139,24 @@ DESCRIPTION receiving events if it installed its rings earlier. **bpftool map peek** *MAP* - Peek next **value** in the queue or stack. + Peek next value in the queue or stack. **bpftool map push** *MAP* **value** *VALUE* - Push **value** onto the stack. + Push *VALUE* onto the stack. **bpftool map pop** *MAP* - Pop and print **value** from the stack. + Pop and print value from the stack. **bpftool map enqueue** *MAP* **value** *VALUE* - Enqueue **value** into the queue. + Enqueue *VALUE* into the queue. **bpftool map dequeue** *MAP* - Dequeue and print **value** from the queue. + Dequeue and print value from the queue. **bpftool map freeze** *MAP* Freeze the map as read-only from user space. Entries from a frozen map can not longer be updated or deleted with the - **bpf\ ()** system call. This operation is not reversible, + **bpf**\ () system call. This operation is not reversible, and the map remains immutable from user space until its destruction. However, read and write permissions for BPF programs to the map remain unchanged. @@ -143,18 +166,7 @@ DESCRIPTION OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. + .. include:: common_options.rst -f, --bpffs Show file names of pinned maps. @@ -163,17 +175,15 @@ OPTIONS Do not automatically attempt to mount any virtual file system (such as tracefs or BPF virtual file system) when necessary. - -d, --debug - Print all logs available from libbpf, including debug-level - information. - EXAMPLES ======== **# bpftool map show** + :: 10: hash name some_map flags 0x0 - key 4B value 8B max_entries 2048 memlock 167936B + key 4B value 8B max_entries 2048 memlock 167936B + pids systemd(1) The following three commands are equivalent: @@ -190,6 +200,7 @@ The following three commands are equivalent: **# bpftool map dump id 10** + :: key: 00 01 02 03 value: 00 01 02 03 04 05 06 07 @@ -197,6 +208,7 @@ The following three commands are equivalent: Found 2 elements **# bpftool map getnext id 10 key 0 1 2 3** + :: key: @@ -263,15 +275,3 @@ would be lost as soon as bpftool exits). key: 00 00 00 00 value: 22 02 00 00 Found 1 element - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-prog**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) diff --git a/tools/bpf/bpftool/Documentation/bpftool-net.rst b/tools/bpf/bpftool/Documentation/bpftool-net.rst index 8651b00b81ea..f4e0a516335a 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-net.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-net.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-net ================ @@ -7,12 +9,14 @@ tool for inspection of netdev/tc related bpf prog attachments :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **net** *COMMAND* - *OPTIONS* := { [{ **-j** | **--json** }] [{ **-p** | **--pretty** }] } + *OPTIONS* := { |COMMON_OPTIONS| } *COMMANDS* := { **show** | **list** | **attach** | **detach** | **help** } @@ -20,7 +24,7 @@ SYNOPSIS NET COMMANDS ============ -| **bpftool** **net { show | list }** [ **dev** *NAME* ] +| **bpftool** **net** { **show** | **list** } [ **dev** *NAME* ] | **bpftool** **net attach** *ATTACH_TYPE* *PROG* **dev** *NAME* [ **overwrite** ] | **bpftool** **net detach** *ATTACH_TYPE* **dev** *NAME* | **bpftool** **net help** @@ -31,66 +35,51 @@ NET COMMANDS DESCRIPTION =========== **bpftool net { show | list }** [ **dev** *NAME* ] - List bpf program attachments in the kernel networking subsystem. - - Currently, only device driver xdp attachments and tc filter - classification/action attachments are implemented, i.e., for - program types **BPF_PROG_TYPE_SCHED_CLS**, - **BPF_PROG_TYPE_SCHED_ACT** and **BPF_PROG_TYPE_XDP**. - For programs attached to a particular cgroup, e.g., - **BPF_PROG_TYPE_CGROUP_SKB**, **BPF_PROG_TYPE_CGROUP_SOCK**, - **BPF_PROG_TYPE_SOCK_OPS** and **BPF_PROG_TYPE_CGROUP_SOCK_ADDR**, - users can use **bpftool cgroup** to dump cgroup attachments. - For sk_{filter, skb, msg, reuseport} and lwt/seg6 - bpf programs, users should consult other tools, e.g., iproute2. - - The current output will start with all xdp program attachments, followed by - all tc class/qdisc bpf program attachments. Both xdp programs and - tc programs are ordered based on ifindex number. If multiple bpf - programs attached to the same networking device through **tc filter**, - the order will be first all bpf programs attached to tc classes, then - all bpf programs attached to non clsact qdiscs, and finally all - bpf programs attached to root and clsact qdisc. + List bpf program attachments in the kernel networking subsystem. + + Currently, only device driver xdp attachments and tc filter + classification/action attachments are implemented, i.e., for + program types **BPF_PROG_TYPE_SCHED_CLS**, + **BPF_PROG_TYPE_SCHED_ACT** and **BPF_PROG_TYPE_XDP**. + For programs attached to a particular cgroup, e.g., + **BPF_PROG_TYPE_CGROUP_SKB**, **BPF_PROG_TYPE_CGROUP_SOCK**, + **BPF_PROG_TYPE_SOCK_OPS** and **BPF_PROG_TYPE_CGROUP_SOCK_ADDR**, + users can use **bpftool cgroup** to dump cgroup attachments. + For sk_{filter, skb, msg, reuseport} and lwt/seg6 + bpf programs, users should consult other tools, e.g., iproute2. + + The current output will start with all xdp program attachments, followed by + all tc class/qdisc bpf program attachments. Both xdp programs and + tc programs are ordered based on ifindex number. If multiple bpf + programs attached to the same networking device through **tc filter**, + the order will be first all bpf programs attached to tc classes, then + all bpf programs attached to non clsact qdiscs, and finally all + bpf programs attached to root and clsact qdisc. **bpftool** **net attach** *ATTACH_TYPE* *PROG* **dev** *NAME* [ **overwrite** ] - Attach bpf program *PROG* to network interface *NAME* with - type specified by *ATTACH_TYPE*. Previously attached bpf program - can be replaced by the command used with **overwrite** option. - Currently, only XDP-related modes are supported for *ATTACH_TYPE*. + Attach bpf program *PROG* to network interface *NAME* with + type specified by *ATTACH_TYPE*. Previously attached bpf program + can be replaced by the command used with **overwrite** option. + Currently, only XDP-related modes are supported for *ATTACH_TYPE*. - *ATTACH_TYPE* can be of: - **xdp** - try native XDP and fallback to generic XDP if NIC driver does not support it; - **xdpgeneric** - Generic XDP. runs at generic XDP hook when packet already enters receive path as skb; - **xdpdrv** - Native XDP. runs earliest point in driver's receive path; - **xdpoffload** - Offload XDP. runs directly on NIC on each packet reception; + *ATTACH_TYPE* can be of: + **xdp** - try native XDP and fallback to generic XDP if NIC driver does not support it; + **xdpgeneric** - Generic XDP. runs at generic XDP hook when packet already enters receive path as skb; + **xdpdrv** - Native XDP. runs earliest point in driver's receive path; + **xdpoffload** - Offload XDP. runs directly on NIC on each packet reception; **bpftool** **net detach** *ATTACH_TYPE* **dev** *NAME* - Detach bpf program attached to network interface *NAME* with - type specified by *ATTACH_TYPE*. To detach bpf program, same - *ATTACH_TYPE* previously used for attach must be specified. - Currently, only XDP-related modes are supported for *ATTACH_TYPE*. + Detach bpf program attached to network interface *NAME* with + type specified by *ATTACH_TYPE*. To detach bpf program, same + *ATTACH_TYPE* previously used for attach must be specified. + Currently, only XDP-related modes are supported for *ATTACH_TYPE*. **bpftool net help** Print short help message. OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. - - -d, --debug - Print all logs available from libbpf, including debug-level - information. + .. include:: common_options.rst EXAMPLES ======== @@ -187,16 +176,3 @@ EXAMPLES :: xdp: - - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-prog**\ (8), - **bpftool-map**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) diff --git a/tools/bpf/bpftool/Documentation/bpftool-perf.rst b/tools/bpf/bpftool/Documentation/bpftool-perf.rst index e252bd0bc434..5fea633a82f1 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-perf.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-perf.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-perf ================ @@ -7,12 +9,14 @@ tool for inspection of perf related bpf prog attachments :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **perf** *COMMAND* - *OPTIONS* := { [{ **-j** | **--json** }] [{ **-p** | **--pretty** }] } + *OPTIONS* := { |COMMON_OPTIONS| } *COMMANDS* := { **show** | **list** | **help** } @@ -20,7 +24,7 @@ SYNOPSIS PERF COMMANDS ============= -| **bpftool** **perf { show | list }** +| **bpftool** **perf** { **show** | **list** } | **bpftool** **perf help** DESCRIPTION @@ -40,22 +44,7 @@ DESCRIPTION OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. - - -d, --debug - Print all logs available from libbpf, including debug-level - information. + .. include:: common_options.rst EXAMPLES ======== @@ -78,16 +67,3 @@ EXAMPLES {"pid":21765,"fd":5,"prog_id":7,"fd_type":"kretprobe","func":"__x64_sys_nanosleep","offset":0}, \ {"pid":21767,"fd":5,"prog_id":8,"fd_type":"tracepoint","tracepoint":"sys_enter_nanosleep"}, \ {"pid":21800,"fd":5,"prog_id":9,"fd_type":"uprobe","filename":"/home/yhs/a.out","offset":1159}] - - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-prog**\ (8), - **bpftool-map**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-btf**\ (8) diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst index 64ddf8a4c518..eb1b2a254eb1 100644 --- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst +++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ bpftool-prog ================ @@ -7,29 +9,34 @@ tool for inspection and simple manipulation of eBPF progs :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== **bpftool** [*OPTIONS*] **prog** *COMMAND* - *OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-f** | **--bpffs** } } + *OPTIONS* := { |COMMON_OPTIONS| | + { **-f** | **--bpffs** } | { **-m** | **--mapcompat** } | { **-n** | **--nomount** } | + { **-L** | **--use-loader** } } *COMMANDS* := - { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load** - | **loadall** | **help** } + { **show** | **list** | **dump xlated** | **dump jited** | **pin** | **load** | + **loadall** | **help** } PROG COMMANDS ============= -| **bpftool** **prog { show | list }** [*PROG*] +| **bpftool** **prog** { **show** | **list** } [*PROG*] | **bpftool** **prog dump xlated** *PROG* [{**file** *FILE* | **opcodes** | **visual** | **linum**}] | **bpftool** **prog dump jited** *PROG* [{**file** *FILE* | **opcodes** | **linum**}] | **bpftool** **prog pin** *PROG* *FILE* -| **bpftool** **prog { load | loadall }** *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] +| **bpftool** **prog** { **load** | **loadall** } *OBJ* *PATH* [**type** *TYPE*] [**map** {**idx** *IDX* | **name** *NAME*} *MAP*] [**dev** *NAME*] [**pinmaps** *MAP_DIR*] | **bpftool** **prog attach** *PROG* *ATTACH_TYPE* [*MAP*] | **bpftool** **prog detach** *PROG* *ATTACH_TYPE* [*MAP*] | **bpftool** **prog tracelog** | **bpftool** **prog run** *PROG* **data_in** *FILE* [**data_out** *FILE* [**data_size_out** *L*]] [**ctx_in** *FILE* [**ctx_out** *FILE* [**ctx_size_out** *M*]]] [**repeat** *N*] +| **bpftool** **prog profile** *PROG* [**duration** *DURATION*] *METRICs* | **bpftool** **prog help** | | *MAP* := { **id** *MAP_ID* | **pinned** *FILE* } @@ -40,12 +47,19 @@ PROG COMMANDS | **cgroup/sock** | **cgroup/dev** | **lwt_in** | **lwt_out** | **lwt_xmit** | | **lwt_seg6local** | **sockops** | **sk_skb** | **sk_msg** | **lirc_mode2** | | **cgroup/bind4** | **cgroup/bind6** | **cgroup/post_bind4** | **cgroup/post_bind6** | -| **cgroup/connect4** | **cgroup/connect6** | **cgroup/sendmsg4** | **cgroup/sendmsg6** | +| **cgroup/connect4** | **cgroup/connect6** | **cgroup/getpeername4** | **cgroup/getpeername6** | +| **cgroup/getsockname4** | **cgroup/getsockname6** | **cgroup/sendmsg4** | **cgroup/sendmsg6** | | **cgroup/recvmsg4** | **cgroup/recvmsg6** | **cgroup/sysctl** | -| **cgroup/getsockopt** | **cgroup/setsockopt** +| **cgroup/getsockopt** | **cgroup/setsockopt** | **cgroup/sock_release** | +| **struct_ops** | **fentry** | **fexit** | **freplace** | **sk_lookup** +| } +| *ATTACH_TYPE* := { +| **sk_msg_verdict** | **sk_skb_verdict** | **sk_skb_stream_verdict** | +| **sk_skb_stream_parser** | **flow_dissector** | } -| *ATTACH_TYPE* := { -| **msg_verdict** | **stream_verdict** | **stream_parser** | **flow_dissector** +| *METRICs* := { +| **cycles** | **instructions** | **l1d_loads** | **llc_misses** | +| **itlb_misses** | **dtlb_misses** | } @@ -69,6 +83,11 @@ DESCRIPTION program run. Activation or deactivation of the feature is performed via the **kernel.bpf_stats_enabled** sysctl knob. + Since Linux 5.8 bpftool is able to discover information about + processes that hold open file descriptors (FDs) against BPF + programs. On such kernels bpftool will automatically emit this + information as well. + **bpftool prog dump xlated** *PROG* [{ **file** *FILE* | **opcodes** | **visual** | **linum** }] Dump eBPF instructions of the programs from the kernel. By default, eBPF will be disassembled and printed to standard @@ -150,7 +169,7 @@ DESCRIPTION **bpftool prog tracelog** Dump the trace pipe of the system to the console (stdout). Hit <Ctrl+C> to stop printing. BPF programs can write to this - trace pipe at runtime with the **bpf_trace_printk()** helper. + trace pipe at runtime with the **bpf_trace_printk**\ () helper. This should be used only for debugging purposes. For streaming data from BPF programs to user space, one can use perf events (see also **bpftool-map**\ (8)). @@ -188,23 +207,18 @@ DESCRIPTION not all of them can take the **ctx_in**/**ctx_out** arguments. bpftool does not perform checks on program types. + **bpftool prog profile** *PROG* [**duration** *DURATION*] *METRICs* + Profile *METRICs* for bpf program *PROG* for *DURATION* + seconds or until user hits <Ctrl+C>. *DURATION* is optional. + If *DURATION* is not specified, the profiling will run up to + **UINT_MAX** seconds. + **bpftool prog help** Print short help message. OPTIONS ======= - -h, --help - Print short generic help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. + .. include:: common_options.rst -f, --bpffs When showing BPF programs, show file names of pinned @@ -217,10 +231,19 @@ OPTIONS Do not automatically attempt to mount any virtual file system (such as tracefs or BPF virtual file system) when necessary. - -d, --debug - Print all logs available, even debug-level information. This - includes logs from libbpf as well as from the verifier, when - attempting to load programs. + -L, --use-loader + Load program as a "loader" program. This is useful to debug + the generation of such programs. When this option is in + use, bpftool attempts to load the programs from the object + file into the kernel, but does not pin them (therefore, the + *PATH* must not be provided). + + When combined with the **-d**\ \|\ **--debug** option, + additional debug messages are generated, and the execution + of the loader program will use the **bpf_trace_printk**\ () + helper to log each step of loading BTF, creating the maps, + and loading the programs (see **bpftool prog tracelog** as + a way to dump those messages). EXAMPLES ======== @@ -231,6 +254,7 @@ EXAMPLES 10: xdp name some_prog tag 005a3d2123620c8b gpl run_time_ns 81632 run_cnt 10 loaded_at 2017-09-29T20:11:00+0000 uid 0 xlated 528B jited 370B memlock 4096B map_ids 10 + pids systemd(1) **# bpftool --json --pretty prog show** @@ -250,13 +274,18 @@ EXAMPLES "bytes_jited": 370, "bytes_memlock": 4096, "map_ids": [10 + ], + "pids": [{ + "pid": 1, + "comm": "systemd" + } ] } ] | | **# bpftool prog dump xlated id 10 file /tmp/t** -| **# ls -l /tmp/t** +| **$ ls -l /tmp/t** :: @@ -310,14 +339,25 @@ EXAMPLES **# rm /sys/fs/bpf/xdp1** -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool**\ (8), - **bpftool-map**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8) +| +| **# bpftool prog profile id 337 duration 10 cycles instructions llc_misses** + +:: + + 51397 run_cnt + 40176203 cycles (83.05%) + 42518139 instructions # 1.06 insns per cycle (83.39%) + 123 llc_misses # 2.89 LLC misses per million insns (83.15%) + +| +| Output below is for the trace logs. +| Run in separate terminals: +| **# bpftool prog tracelog** +| **# bpftool prog load -L -d file.o** + +:: + + bpftool-620059 [004] d... 2634685.517903: bpf_trace_printk: btf_load size 665 r=5 + bpftool-620059 [004] d... 2634685.517912: bpf_trace_printk: map_create sample_map idx 0 type 2 value_size 4 value_btf_id 0 r=6 + bpftool-620059 [004] d... 2634685.517997: bpf_trace_printk: prog_load sample insn_cnt 13 r=7 + bpftool-620059 [004] d... 2634685.517999: bpf_trace_printk: close(5) = 0 diff --git a/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst b/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst new file mode 100644 index 000000000000..ee53a122c0c7 --- /dev/null +++ b/tools/bpf/bpftool/Documentation/bpftool-struct_ops.rst @@ -0,0 +1,88 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +================== +bpftool-struct_ops +================== +------------------------------------------------------------------------------- +tool to register/unregister/introspect BPF struct_ops +------------------------------------------------------------------------------- + +:Manual section: 8 + +.. include:: substitutions.rst + +SYNOPSIS +======== + + **bpftool** [*OPTIONS*] **struct_ops** *COMMAND* + + *OPTIONS* := { |COMMON_OPTIONS| } + + *COMMANDS* := + { **show** | **list** | **dump** | **register** | **unregister** | **help** } + +STRUCT_OPS COMMANDS +=================== + +| **bpftool** **struct_ops { show | list }** [*STRUCT_OPS_MAP*] +| **bpftool** **struct_ops dump** [*STRUCT_OPS_MAP*] +| **bpftool** **struct_ops register** *OBJ* +| **bpftool** **struct_ops unregister** *STRUCT_OPS_MAP* +| **bpftool** **struct_ops help** +| +| *STRUCT_OPS_MAP* := { **id** *STRUCT_OPS_MAP_ID* | **name** *STRUCT_OPS_MAP_NAME* } +| *OBJ* := /a/file/of/bpf_struct_ops.o + + +DESCRIPTION +=========== + **bpftool struct_ops { show | list }** [*STRUCT_OPS_MAP*] + Show brief information about the struct_ops in the system. + If *STRUCT_OPS_MAP* is specified, it shows information only + for the given struct_ops. Otherwise, it lists all struct_ops + currently existing in the system. + + Output will start with struct_ops map ID, followed by its map + name and its struct_ops's kernel type. + + **bpftool struct_ops dump** [*STRUCT_OPS_MAP*] + Dump details information about the struct_ops in the system. + If *STRUCT_OPS_MAP* is specified, it dumps information only + for the given struct_ops. Otherwise, it dumps all struct_ops + currently existing in the system. + + **bpftool struct_ops register** *OBJ* + Register bpf struct_ops from *OBJ*. All struct_ops under + the ELF section ".struct_ops" will be registered to + its kernel subsystem. + + **bpftool struct_ops unregister** *STRUCT_OPS_MAP* + Unregister the *STRUCT_OPS_MAP* from the kernel subsystem. + + **bpftool struct_ops help** + Print short help message. + +OPTIONS +======= + .. include:: common_options.rst + +EXAMPLES +======== +**# bpftool struct_ops show** + +:: + + 100: dctcp tcp_congestion_ops + 105: cubic tcp_congestion_ops + +**# bpftool struct_ops unregister id 105** + +:: + + Unregistered tcp_congestion_ops cubic id 105 + +**# bpftool struct_ops register bpf_cubic.o** + +:: + + Registered tcp_congestion_ops cubic id 110 diff --git a/tools/bpf/bpftool/Documentation/bpftool.rst b/tools/bpf/bpftool/Documentation/bpftool.rst index 34239fda69ed..6965c94dfdaf 100644 --- a/tools/bpf/bpftool/Documentation/bpftool.rst +++ b/tools/bpf/bpftool/Documentation/bpftool.rst @@ -1,3 +1,5 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + ================ BPFTOOL ================ @@ -7,6 +9,8 @@ tool for inspection and simple manipulation of eBPF programs and maps :Manual section: 8 +.. include:: substitutions.rst + SYNOPSIS ======== @@ -16,17 +20,19 @@ SYNOPSIS **bpftool** **version** - *OBJECT* := { **map** | **program** | **cgroup** | **perf** | **net** | **feature** } + *OBJECT* := { **map** | **program** | **link** | **cgroup** | **perf** | **net** | **feature** | + **btf** | **gen** | **struct_ops** | **iter** } - *OPTIONS* := { { **-V** | **--version** } | { **-h** | **--help** } - | { **-j** | **--json** } [{ **-p** | **--pretty** }] } + *OPTIONS* := { { **-V** | **--version** } | |COMMON_OPTIONS| } *MAP-COMMANDS* := - { **show** | **list** | **create** | **dump** | **update** | **lookup** | **getnext** - | **delete** | **pin** | **event_pipe** | **help** } + { **show** | **list** | **create** | **dump** | **update** | **lookup** | **getnext** | + **delete** | **pin** | **event_pipe** | **help** } - *PROG-COMMANDS* := { **show** | **list** | **dump jited** | **dump xlated** | **pin** - | **load** | **attach** | **detach** | **help** } + *PROG-COMMANDS* := { **show** | **list** | **dump jited** | **dump xlated** | **pin** | + **load** | **attach** | **detach** | **help** } + + *LINK-COMMANDS* := { **show** | **list** | **pin** | **detach** | **help** } *CGROUP-COMMANDS* := { **show** | **list** | **attach** | **detach** | **help** } @@ -36,6 +42,14 @@ SYNOPSIS *FEATURE-COMMANDS* := { **probe** | **help** } + *BTF-COMMANDS* := { **show** | **list** | **dump** | **help** } + + *GEN-COMMANDS* := { **object** | **skeleton** | **min_core_btf** | **help** } + + *STRUCT-OPS-COMMANDS* := { **show** | **list** | **dump** | **register** | **unregister** | **help** } + + *ITER-COMMANDS* := { **pin** | **help** } + DESCRIPTION =========== *bpftool* allows for inspection and simple modification of BPF objects @@ -46,18 +60,7 @@ DESCRIPTION OPTIONS ======= - -h, --help - Print short help message (similar to **bpftool help**). - - -V, --version - Print version number (similar to **bpftool version**). - - -j, --json - Generate JSON output. For commands that cannot produce JSON, this - option has no effect. - - -p, --pretty - Generate human-readable JSON output. Implies **-j**. + .. include:: common_options.rst -m, --mapcompat Allow loading maps with unknown map definitions. @@ -65,21 +68,3 @@ OPTIONS -n, --nomount Do not automatically attempt to mount any virtual file system (such as tracefs or BPF virtual file system) when necessary. - - -d, --debug - Print all logs available, even debug-level information. This - includes logs from libbpf as well as from the verifier, when - attempting to load programs. - -SEE ALSO -======== - **bpf**\ (2), - **bpf-helpers**\ (7), - **bpftool-prog**\ (8), - **bpftool-map**\ (8), - **bpftool-cgroup**\ (8), - **bpftool-feature**\ (8), - **bpftool-net**\ (8), - **bpftool-perf**\ (8), - **bpftool-btf**\ (8), - **bpftool-gen**\ (8), diff --git a/tools/bpf/bpftool/Documentation/common_options.rst b/tools/bpf/bpftool/Documentation/common_options.rst new file mode 100644 index 000000000000..4107a586b68b --- /dev/null +++ b/tools/bpf/bpftool/Documentation/common_options.rst @@ -0,0 +1,34 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +-h, --help + Print short help message (similar to **bpftool help**). + +-V, --version + Print bpftool's version number (similar to **bpftool version**), the + number of the libbpf version in use, and optional features that were + included when bpftool was compiled. Optional features include linking + against libbfd to provide the disassembler for JIT-ted programs + (**bpftool prog dump jited**) and usage of BPF skeletons (some + features like **bpftool prog profile** or showing pids associated to + BPF objects may rely on it). + +-j, --json + Generate JSON output. For commands that cannot produce JSON, this + option has no effect. + +-p, --pretty + Generate human-readable JSON output. Implies **-j**. + +-d, --debug + Print all logs available, even debug-level information. This includes + logs from libbpf as well as from the verifier, when attempting to + load programs. + +-l, --legacy + Use legacy libbpf mode which has more relaxed BPF program + requirements. By default, bpftool has more strict requirements + about section names, changes pinning logic and doesn't support + some of the older non-BTF map declarations. + + See https://github.com/libbpf/libbpf/wiki/Libbpf:-the-road-to-v1.0 + for details. diff --git a/tools/bpf/bpftool/Documentation/substitutions.rst b/tools/bpf/bpftool/Documentation/substitutions.rst new file mode 100644 index 000000000000..ccf1ffa0686c --- /dev/null +++ b/tools/bpf/bpftool/Documentation/substitutions.rst @@ -0,0 +1,3 @@ +.. SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) + +.. |COMMON_OPTIONS| replace:: { **-j** | **--json** } [{ **-p** | **--pretty** }] | { **-d** | **--debug** } | { **-l** | **--legacy** } diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile index c4e810335810..4a95c017ad4c 100644 --- a/tools/bpf/bpftool/Makefile +++ b/tools/bpf/bpftool/Makefile @@ -1,6 +1,5 @@ -# SPDX-License-Identifier: GPL-2.0-only +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) include ../../scripts/Makefile.include -include ../../scripts/utilities.mak ifeq ($(srctree),) srctree := $(patsubst %/,%,$(dir $(CURDIR))) @@ -14,40 +13,75 @@ else Q = @ endif -BPF_DIR = $(srctree)/tools/lib/bpf/ +BPF_DIR = $(srctree)/tools/lib/bpf ifneq ($(OUTPUT),) - LIBBPF_OUTPUT = $(OUTPUT)/libbpf/ - LIBBPF_PATH = $(LIBBPF_OUTPUT) + _OUTPUT := $(OUTPUT) else - LIBBPF_PATH = $(BPF_DIR) + _OUTPUT := $(CURDIR)/ endif - -LIBBPF = $(LIBBPF_PATH)libbpf.a - -BPFTOOL_VERSION := $(shell make -rR --no-print-directory -sC ../../.. kernelversion) - -$(LIBBPF): FORCE - $(if $(LIBBPF_OUTPUT),@mkdir -p $(LIBBPF_OUTPUT)) - $(Q)$(MAKE) -C $(BPF_DIR) OUTPUT=$(LIBBPF_OUTPUT) $(LIBBPF_OUTPUT)libbpf.a - -$(LIBBPF)-clean: +BOOTSTRAP_OUTPUT := $(_OUTPUT)bootstrap/ + +LIBBPF_OUTPUT := $(_OUTPUT)libbpf/ +LIBBPF_DESTDIR := $(LIBBPF_OUTPUT) +LIBBPF_INCLUDE := $(LIBBPF_DESTDIR)include +LIBBPF_HDRS_DIR := $(LIBBPF_INCLUDE)/bpf +LIBBPF := $(LIBBPF_OUTPUT)libbpf.a + +LIBBPF_BOOTSTRAP_OUTPUT := $(BOOTSTRAP_OUTPUT)libbpf/ +LIBBPF_BOOTSTRAP_DESTDIR := $(LIBBPF_BOOTSTRAP_OUTPUT) +LIBBPF_BOOTSTRAP_INCLUDE := $(LIBBPF_BOOTSTRAP_DESTDIR)include +LIBBPF_BOOTSTRAP_HDRS_DIR := $(LIBBPF_BOOTSTRAP_INCLUDE)/bpf +LIBBPF_BOOTSTRAP := $(LIBBPF_BOOTSTRAP_OUTPUT)libbpf.a + +# We need to copy hashmap.h, nlattr.h, relo_core.h and libbpf_internal.h +# which are not otherwise exported by libbpf, but still required by bpftool. +LIBBPF_INTERNAL_HDRS := $(addprefix $(LIBBPF_HDRS_DIR)/,hashmap.h nlattr.h relo_core.h libbpf_internal.h) +LIBBPF_BOOTSTRAP_INTERNAL_HDRS := $(addprefix $(LIBBPF_BOOTSTRAP_HDRS_DIR)/,hashmap.h relo_core.h libbpf_internal.h) + +$(LIBBPF_OUTPUT) $(BOOTSTRAP_OUTPUT) $(LIBBPF_BOOTSTRAP_OUTPUT) $(LIBBPF_HDRS_DIR) $(LIBBPF_BOOTSTRAP_HDRS_DIR): + $(QUIET_MKDIR)mkdir -p $@ + +$(LIBBPF): $(wildcard $(BPF_DIR)/*.[ch] $(BPF_DIR)/Makefile) | $(LIBBPF_OUTPUT) + $(Q)$(MAKE) -C $(BPF_DIR) OUTPUT=$(LIBBPF_OUTPUT) \ + DESTDIR=$(LIBBPF_DESTDIR:/=) prefix= $(LIBBPF) install_headers + +$(LIBBPF_INTERNAL_HDRS): $(LIBBPF_HDRS_DIR)/%.h: $(BPF_DIR)/%.h | $(LIBBPF_HDRS_DIR) + $(call QUIET_INSTALL, $@) + $(Q)install -m 644 -t $(LIBBPF_HDRS_DIR) $< + +$(LIBBPF_BOOTSTRAP): $(wildcard $(BPF_DIR)/*.[ch] $(BPF_DIR)/Makefile) | $(LIBBPF_BOOTSTRAP_OUTPUT) + $(Q)$(MAKE) -C $(BPF_DIR) OUTPUT=$(LIBBPF_BOOTSTRAP_OUTPUT) \ + DESTDIR=$(LIBBPF_BOOTSTRAP_DESTDIR:/=) prefix= \ + ARCH= CROSS_COMPILE= CC=$(HOSTCC) LD=$(HOSTLD) AR=$(HOSTAR) $@ install_headers + +$(LIBBPF_BOOTSTRAP_INTERNAL_HDRS): $(LIBBPF_BOOTSTRAP_HDRS_DIR)/%.h: $(BPF_DIR)/%.h | $(LIBBPF_BOOTSTRAP_HDRS_DIR) + $(call QUIET_INSTALL, $@) + $(Q)install -m 644 -t $(LIBBPF_BOOTSTRAP_HDRS_DIR) $< + +$(LIBBPF)-clean: FORCE | $(LIBBPF_OUTPUT) $(call QUIET_CLEAN, libbpf) $(Q)$(MAKE) -C $(BPF_DIR) OUTPUT=$(LIBBPF_OUTPUT) clean >/dev/null +$(LIBBPF_BOOTSTRAP)-clean: FORCE | $(LIBBPF_BOOTSTRAP_OUTPUT) + $(call QUIET_CLEAN, libbpf-bootstrap) + $(Q)$(MAKE) -C $(BPF_DIR) OUTPUT=$(LIBBPF_BOOTSTRAP_OUTPUT) clean >/dev/null + prefix ?= /usr/local bash_compdir ?= /usr/share/bash-completion/completions CFLAGS += -O2 CFLAGS += -W -Wall -Wextra -Wno-unused-parameter -Wno-missing-field-initializers -CFLAGS += $(filter-out -Wswitch-enum,$(EXTRA_WARNINGS)) +CFLAGS += $(filter-out -Wswitch-enum -Wnested-externs,$(EXTRA_WARNINGS)) CFLAGS += -DPACKAGE='"bpftool"' -D__EXPORTED_HEADERS__ \ + -I$(or $(OUTPUT),.) \ + -I$(LIBBPF_INCLUDE) \ -I$(srctree)/kernel/bpf/ \ -I$(srctree)/tools/include \ - -I$(srctree)/tools/include/uapi \ - -I$(srctree)/tools/lib \ - -I$(srctree)/tools/perf + -I$(srctree)/tools/include/uapi +ifneq ($(BPFTOOL_VERSION),) CFLAGS += -DBPFTOOL_VERSION='"$(BPFTOOL_VERSION)"' +endif ifneq ($(EXTRA_CFLAGS),) CFLAGS += $(EXTRA_CFLAGS) endif @@ -55,14 +89,15 @@ ifneq ($(EXTRA_LDFLAGS),) LDFLAGS += $(EXTRA_LDFLAGS) endif -LIBS = $(LIBBPF) -lelf -lz - INSTALL ?= install RM ?= rm -f FEATURE_USER = .bpftool -FEATURE_TESTS = libbfd disassembler-four-args reallocarray zlib -FEATURE_DISPLAY = libbfd disassembler-four-args zlib +FEATURE_TESTS = libbfd libbfd-liberty libbfd-liberty-z \ + disassembler-four-args disassembler-init-styled libcap \ + clang-bpf-co-re +FEATURE_DISPLAY = libbfd libbfd-liberty libbfd-liberty-z \ + libcap clang-bpf-co-re check_feat := 1 NON_CHECK_FEAT_TARGETS := clean uninstall doc doc-clean doc-install doc-uninstall @@ -83,9 +118,15 @@ endif ifeq ($(feature-disassembler-four-args), 1) CFLAGS += -DDISASM_FOUR_ARGS_SIGNATURE endif +ifeq ($(feature-disassembler-init-styled), 1) + CFLAGS += -DDISASM_INIT_STYLED +endif -ifeq ($(feature-reallocarray), 0) -CFLAGS += -DCOMPAT_NEED_REALLOCARRAY +LIBS = $(LIBBPF) -lelf -lz +LIBS_BOOTSTRAP = $(LIBBPF_BOOTSTRAP) -lelf -lz +ifeq ($(feature-libcap), 1) +CFLAGS += -DUSE_LIBCAP +LIBS += -lcap endif include $(wildcard $(OUTPUT)*.d) @@ -109,31 +150,95 @@ CFLAGS += -DHAVE_LIBBFD_SUPPORT SRCS += $(BFD_SRCS) endif +HOST_CFLAGS = $(subst -I$(LIBBPF_INCLUDE),-I$(LIBBPF_BOOTSTRAP_INCLUDE),\ + $(subst $(CLANG_CROSS_FLAGS),,$(CFLAGS))) + +BPFTOOL_BOOTSTRAP := $(BOOTSTRAP_OUTPUT)bpftool + +BOOTSTRAP_OBJS = $(addprefix $(BOOTSTRAP_OUTPUT),main.o common.o json_writer.o gen.o btf.o xlated_dumper.o btf_dumper.o disasm.o) +$(BOOTSTRAP_OBJS): $(LIBBPF_BOOTSTRAP) + OBJS = $(patsubst %.c,$(OUTPUT)%.o,$(SRCS)) $(OUTPUT)disasm.o +$(OBJS): $(LIBBPF) $(LIBBPF_INTERNAL_HDRS) + +VMLINUX_BTF_PATHS ?= $(if $(O),$(O)/vmlinux) \ + $(if $(KBUILD_OUTPUT),$(KBUILD_OUTPUT)/vmlinux) \ + ../../../vmlinux \ + /sys/kernel/btf/vmlinux \ + /boot/vmlinux-$(shell uname -r) +VMLINUX_BTF ?= $(abspath $(firstword $(wildcard $(VMLINUX_BTF_PATHS)))) + +bootstrap: $(BPFTOOL_BOOTSTRAP) + +ifneq ($(VMLINUX_BTF)$(VMLINUX_H),) +ifeq ($(feature-clang-bpf-co-re),1) + +BUILD_BPF_SKELS := 1 + +$(OUTPUT)vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL_BOOTSTRAP) +ifeq ($(VMLINUX_H),) + $(QUIET_GEN)$(BPFTOOL_BOOTSTRAP) btf dump file $< format c > $@ +else + $(Q)cp "$(VMLINUX_H)" $@ +endif + +$(OUTPUT)%.bpf.o: skeleton/%.bpf.c $(OUTPUT)vmlinux.h $(LIBBPF_BOOTSTRAP) + $(QUIET_CLANG)$(CLANG) \ + -I$(or $(OUTPUT),.) \ + -I$(srctree)/tools/include/uapi/ \ + -I$(LIBBPF_BOOTSTRAP_INCLUDE) \ + -g -O2 -Wall -target bpf -c $< -o $@ + $(Q)$(LLVM_STRIP) -g $@ + +$(OUTPUT)%.skel.h: $(OUTPUT)%.bpf.o $(BPFTOOL_BOOTSTRAP) + $(QUIET_GEN)$(BPFTOOL_BOOTSTRAP) gen skeleton $< > $@ + +$(OUTPUT)prog.o: $(OUTPUT)profiler.skel.h + +$(OUTPUT)pids.o: $(OUTPUT)pid_iter.skel.h + +endif +endif + +CFLAGS += $(if $(BUILD_BPF_SKELS),,-DBPFTOOL_WITHOUT_SKELETONS) + +$(BOOTSTRAP_OUTPUT)disasm.o: $(srctree)/kernel/bpf/disasm.c + $(QUIET_CC)$(HOSTCC) $(HOST_CFLAGS) -c -MMD $< -o $@ $(OUTPUT)disasm.o: $(srctree)/kernel/bpf/disasm.c - $(QUIET_CC)$(COMPILE.c) -MMD -o $@ $< + $(QUIET_CC)$(CC) $(CFLAGS) -c -MMD $< -o $@ -$(OUTPUT)feature.o: | zdep +$(BPFTOOL_BOOTSTRAP): $(BOOTSTRAP_OBJS) $(LIBBPF_BOOTSTRAP) + $(QUIET_LINK)$(HOSTCC) $(HOST_CFLAGS) $(LDFLAGS) $(BOOTSTRAP_OBJS) $(LIBS_BOOTSTRAP) -o $@ $(OUTPUT)bpftool: $(OBJS) $(LIBBPF) - $(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) -o $@ $(OBJS) $(LIBS) + $(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) $(OBJS) $(LIBS) -o $@ + +$(BOOTSTRAP_OUTPUT)%.o: %.c $(LIBBPF_BOOTSTRAP_INTERNAL_HDRS) | $(BOOTSTRAP_OUTPUT) + $(QUIET_CC)$(HOSTCC) $(HOST_CFLAGS) -c -MMD $< -o $@ $(OUTPUT)%.o: %.c - $(QUIET_CC)$(COMPILE.c) -MMD -o $@ $< + $(QUIET_CC)$(CC) $(CFLAGS) -c -MMD $< -o $@ + +feature-detect-clean: + $(call QUIET_CLEAN, feature-detect) + $(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null -clean: $(LIBBPF)-clean +clean: $(LIBBPF)-clean $(LIBBPF_BOOTSTRAP)-clean feature-detect-clean $(call QUIET_CLEAN, bpftool) $(Q)$(RM) -- $(OUTPUT)bpftool $(OUTPUT)*.o $(OUTPUT)*.d - $(Q)$(RM) -r -- $(OUTPUT)libbpf/ + $(Q)$(RM) -- $(OUTPUT)*.skel.h $(OUTPUT)vmlinux.h + $(Q)$(RM) -r -- $(LIBBPF_OUTPUT) $(BOOTSTRAP_OUTPUT) $(call QUIET_CLEAN, core-gen) $(Q)$(RM) -- $(OUTPUT)FEATURE-DUMP.bpftool $(Q)$(RM) -r -- $(OUTPUT)feature/ -install: $(OUTPUT)bpftool +install-bin: $(OUTPUT)bpftool $(call QUIET_INSTALL, bpftool) $(Q)$(INSTALL) -m 0755 -d $(DESTDIR)$(prefix)/sbin $(Q)$(INSTALL) $(OUTPUT)bpftool $(DESTDIR)$(prefix)/sbin/bpftool + +install: install-bin $(Q)$(INSTALL) -m 0755 -d $(DESTDIR)$(bash_compdir) $(Q)$(INSTALL) -m 0644 bash-completion/bpftool $(DESTDIR)$(bash_compdir) @@ -156,9 +261,7 @@ doc-uninstall: FORCE: -zdep: - @if [ "$(feature-zlib)" != "1" ]; then echo "No zlib found"; exit 1 ; fi - -.PHONY: all FORCE clean install uninstall zdep +.SECONDARY: +.PHONY: all FORCE bootstrap clean install-bin install uninstall .PHONY: doc doc-clean doc-install doc-uninstall .DEFAULT_GOAL := all diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool index 754d8395e451..dc1641e3670e 100644 --- a/tools/bpf/bpftool/bash-completion/bpftool +++ b/tools/bpf/bpftool/bash-completion/bpftool @@ -98,6 +98,12 @@ _bpftool_get_btf_ids() command sed -n 's/.*"id": \(.*\),$/\1/p' )" -- "$cur" ) ) } +_bpftool_get_link_ids() +{ + COMPREPLY+=( $( compgen -W "$( bpftool -jp link 2>&1 | \ + command sed -n 's/.*"id": \(.*\),$/\1/p' )" -- "$cur" ) ) +} + _bpftool_get_obj_map_names() { local obj @@ -254,7 +260,8 @@ _bpftool() # Deal with options if [[ ${words[cword]} == -* ]]; then - local c='--version --json --pretty --bpffs --mapcompat --debug' + local c='--version --json --pretty --bpffs --mapcompat --debug \ + --use-loader --base-btf --legacy' COMPREPLY=( $( compgen -W "$c" -- "$cur" ) ) return 0 fi @@ -272,7 +279,7 @@ _bpftool() _sysfs_get_netdevs return 0 ;; - file|pinned) + file|pinned|-B|--base-btf) _filedir return 0 ;; @@ -285,7 +292,8 @@ _bpftool() # Remove all options so completions don't have to deal with them. local i for (( i=1; i < ${#words[@]}; )); do - if [[ ${words[i]::1} == - ]]; then + if [[ ${words[i]::1} == - ]] && + [[ ${words[i]} != "-B" ]] && [[ ${words[i]} != "--base-btf" ]]; then words=( "${words[@]:0:i}" "${words[@]:i+1}" ) [[ $i -le $cword ]] && cword=$(( cword - 1 )) else @@ -337,6 +345,8 @@ _bpftool() local PROG_TYPE='id pinned tag name' local MAP_TYPE='id pinned name' + local METRIC_TYPE='cycles instructions l1d_loads llc_misses \ + itlb_misses dtlb_misses' case $command in show|list) [[ $prev != "$command" ]] && return 0 @@ -388,7 +398,7 @@ _bpftool() _bpftool_get_prog_ids ;; name) - _bpftool_get_map_names + _bpftool_get_prog_names ;; pinned) _filedir @@ -397,8 +407,10 @@ _bpftool() return 0 ;; 5) - COMPREPLY=( $( compgen -W 'msg_verdict stream_verdict \ - stream_parser flow_dissector' -- "$cur" ) ) + local BPFTOOL_PROG_ATTACH_TYPES='sk_msg_verdict \ + sk_skb_verdict sk_skb_stream_verdict sk_skb_stream_parser \ + flow_dissector' + COMPREPLY=( $( compgen -W "$BPFTOOL_PROG_ATTACH_TYPES" -- "$cur" ) ) return 0 ;; 6) @@ -457,7 +469,7 @@ _bpftool() case $prev in type) - COMPREPLY=( $( compgen -W "socket kprobe \ + local BPFTOOL_PROG_LOAD_TYPES='socket kprobe \ kretprobe classifier flow_dissector \ action tracepoint raw_tracepoint \ xdp perf_event cgroup/skb cgroup/sock \ @@ -465,12 +477,15 @@ _bpftool() lwt_seg6local sockops sk_skb sk_msg \ lirc_mode2 cgroup/bind4 cgroup/bind6 \ cgroup/connect4 cgroup/connect6 \ + cgroup/getpeername4 cgroup/getpeername6 \ + cgroup/getsockname4 cgroup/getsockname6 \ cgroup/sendmsg4 cgroup/sendmsg6 \ cgroup/recvmsg4 cgroup/recvmsg6 \ cgroup/post_bind4 cgroup/post_bind6 \ cgroup/sysctl cgroup/getsockopt \ - cgroup/setsockopt" -- \ - "$cur" ) ) + cgroup/setsockopt cgroup/sock_release struct_ops \ + fentry fexit freplace sk_lookup' + COMPREPLY=( $( compgen -W "$BPFTOOL_PROG_LOAD_TYPES" -- "$cur" ) ) return 0 ;; id) @@ -497,9 +512,51 @@ _bpftool() tracelog) return 0 ;; + profile) + case $cword in + 3) + COMPREPLY=( $( compgen -W "$PROG_TYPE" -- "$cur" ) ) + return 0 + ;; + 4) + case $prev in + id) + _bpftool_get_prog_ids + ;; + name) + _bpftool_get_prog_names + ;; + pinned) + _filedir + ;; + esac + return 0 + ;; + 5) + COMPREPLY=( $( compgen -W "$METRIC_TYPE duration" -- "$cur" ) ) + return 0 + ;; + 6) + case $prev in + duration) + return 0 + ;; + *) + COMPREPLY=( $( compgen -W "$METRIC_TYPE" -- "$cur" ) ) + return 0 + ;; + esac + return 0 + ;; + *) + COMPREPLY=( $( compgen -W "$METRIC_TYPE" -- "$cur" ) ) + return 0 + ;; + esac + ;; run) - if [[ ${#words[@]} -lt 5 ]]; then - _filedir + if [[ ${#words[@]} -eq 4 ]]; then + COMPREPLY=( $( compgen -W "$PROG_TYPE" -- "$cur" ) ) return 0 fi case $prev in @@ -507,6 +564,10 @@ _bpftool() _bpftool_get_prog_ids return 0 ;; + name) + _bpftool_get_prog_names + return 0 + ;; data_in|data_out|ctx_in|ctx_out) _filedir return 0 @@ -524,7 +585,64 @@ _bpftool() *) [[ $prev == $object ]] && \ COMPREPLY=( $( compgen -W 'dump help pin attach detach \ - load loadall show list tracelog run' -- "$cur" ) ) + load loadall show list tracelog run profile' -- "$cur" ) ) + ;; + esac + ;; + struct_ops) + local STRUCT_OPS_TYPE='id name' + case $command in + show|list|dump|unregister) + case $prev in + $command) + COMPREPLY=( $( compgen -W "$STRUCT_OPS_TYPE" -- "$cur" ) ) + ;; + id) + _bpftool_get_map_ids_for_type struct_ops + ;; + name) + _bpftool_get_map_names_for_type struct_ops + ;; + esac + return 0 + ;; + register) + _filedir + return 0 + ;; + *) + [[ $prev == $object ]] && \ + COMPREPLY=( $( compgen -W 'register unregister show list dump help' \ + -- "$cur" ) ) + ;; + esac + ;; + iter) + case $command in + pin) + case $prev in + $command) + _filedir + ;; + id) + _bpftool_get_map_ids + ;; + name) + _bpftool_get_map_names + ;; + pinned) + _filedir + ;; + *) + _bpftool_one_of_list $MAP_TYPE + ;; + esac + return 0 + ;; + *) + [[ $prev == $object ]] && \ + COMPREPLY=( $( compgen -W 'pin help' \ + -- "$cur" ) ) ;; esac ;; @@ -585,19 +703,31 @@ _bpftool() return 0 ;; type) - COMPREPLY=( $( compgen -W 'hash array prog_array \ - perf_event_array percpu_hash percpu_array \ - stack_trace cgroup_array lru_hash \ - lru_percpu_hash lpm_trie array_of_maps \ - hash_of_maps devmap devmap_hash sockmap cpumap \ - xskmap sockhash cgroup_storage reuseport_sockarray \ - percpu_cgroup_storage queue stack' -- \ - "$cur" ) ) + local BPFTOOL_MAP_CREATE_TYPES="$(bpftool feature list_builtins map_types 2>/dev/null | \ + grep -v '^unspec$')" + COMPREPLY=( $( compgen -W "$BPFTOOL_MAP_CREATE_TYPES" -- "$cur" ) ) return 0 ;; - key|value|flags|name|entries) + key|value|flags|entries) + return 0 + ;; + inner_map) + COMPREPLY=( $( compgen -W "$MAP_TYPE" -- "$cur" ) ) return 0 ;; + id) + _bpftool_get_map_ids + ;; + name) + case $pprev in + inner_map) + _bpftool_get_map_names + ;; + *) + return 0 + ;; + esac + ;; *) _bpftool_once_attr 'type' _bpftool_once_attr 'key' @@ -605,6 +735,9 @@ _bpftool() _bpftool_once_attr 'entries' _bpftool_once_attr 'name' _bpftool_once_attr 'flags' + if _bpftool_search_list 'array_of_maps' 'hash_of_maps'; then + _bpftool_once_attr 'inner_map' + fi _bpftool_once_attr 'dev' return 0 ;; @@ -712,11 +845,17 @@ _bpftool() esac ;; pin) - if [[ $prev == "$command" ]]; then - COMPREPLY=( $( compgen -W "$PROG_TYPE" -- "$cur" ) ) - else - _filedir - fi + case $prev in + $command) + COMPREPLY=( $( compgen -W "$MAP_TYPE" -- "$cur" ) ) + ;; + id) + _bpftool_get_map_ids + ;; + name) + _bpftool_get_map_names + ;; + esac return 0 ;; event_pipe) @@ -841,12 +980,41 @@ _bpftool() ;; gen) case $command in + object) + _filedir + return 0 + ;; skeleton) + case $prev in + $command) + _filedir + return 0 + ;; + *) + _bpftool_once_attr 'name' + return 0 + ;; + esac + ;; + subskeleton) + case $prev in + $command) + _filedir + return 0 + ;; + *) + _bpftool_once_attr 'name' + return 0 + ;; + esac + ;; + min_core_btf) _filedir - ;; + return 0 + ;; *) [[ $prev == $object ]] && \ - COMPREPLY=( $( compgen -W 'skeleton help' -- "$cur" ) ) + COMPREPLY=( $( compgen -W 'object skeleton subskeleton help min_core_btf' -- "$cur" ) ) ;; esac ;; @@ -864,33 +1032,33 @@ _bpftool() return 0 ;; attach|detach) - local ATTACH_TYPES='ingress egress sock_create sock_ops \ - device bind4 bind6 post_bind4 post_bind6 connect4 \ - connect6 sendmsg4 sendmsg6 recvmsg4 recvmsg6 sysctl \ - getsockopt setsockopt' + local BPFTOOL_CGROUP_ATTACH_TYPES="$(bpftool feature list_builtins attach_types 2>/dev/null | \ + grep '^cgroup_')" local ATTACH_FLAGS='multi override' local PROG_TYPE='id pinned tag name' - case $prev in - $command) - _filedir - return 0 - ;; - ingress|egress|sock_create|sock_ops|device|bind4|bind6|\ - post_bind4|post_bind6|connect4|connect6|sendmsg4|\ - sendmsg6|recvmsg4|recvmsg6|sysctl|getsockopt|\ - setsockopt) + # Check for $prev = $command first + if [ $prev = $command ]; then + _filedir + return 0 + # Then check for attach type. This is done outside of the + # "case $prev in" to avoid writing the whole list of attach + # types again as pattern to match (where we cannot reuse + # our variable). + elif [[ $BPFTOOL_CGROUP_ATTACH_TYPES =~ $prev ]]; then COMPREPLY=( $( compgen -W "$PROG_TYPE" -- \ "$cur" ) ) return 0 - ;; + fi + # case/esac for the other cases + case $prev in id) _bpftool_get_prog_ids return 0 ;; *) - if ! _bpftool_search_list "$ATTACH_TYPES"; then - COMPREPLY=( $( compgen -W "$ATTACH_TYPES" -- \ - "$cur" ) ) + if ! _bpftool_search_list "$BPFTOOL_CGROUP_ATTACH_TYPES"; then + COMPREPLY=( $( compgen -W \ + "$BPFTOOL_CGROUP_ATTACH_TYPES" -- "$cur" ) ) elif [[ "$command" == "attach" ]]; then # We have an attach type on the command line, # but it is not the previous word, or @@ -943,6 +1111,9 @@ _bpftool() id) _bpftool_get_prog_ids ;; + name) + _bpftool_get_prog_names + ;; pinned) _filedir ;; @@ -983,16 +1154,55 @@ _bpftool() probe) [[ $prev == "prefix" ]] && return 0 if _bpftool_search_list 'macros'; then - COMPREPLY+=( $( compgen -W 'prefix' -- "$cur" ) ) + _bpftool_once_attr 'prefix' else COMPREPLY+=( $( compgen -W 'macros' -- "$cur" ) ) fi _bpftool_one_of_list 'kernel dev' + _bpftool_once_attr 'full unprivileged' + return 0 + ;; + list_builtins) + [[ $prev != "$command" ]] && return 0 + COMPREPLY=( $( compgen -W 'prog_types map_types \ + attach_types link_types helpers' -- "$cur" ) ) + ;; + *) + [[ $prev == $object ]] && \ + COMPREPLY=( $( compgen -W 'help list_builtins probe' -- "$cur" ) ) + ;; + esac + ;; + link) + case $command in + show|list|pin|detach) + case $prev in + id) + _bpftool_get_link_ids + return 0 + ;; + esac + ;; + esac + + local LINK_TYPE='id pinned' + case $command in + show|list) + [[ $prev != "$command" ]] && return 0 + COMPREPLY=( $( compgen -W "$LINK_TYPE" -- "$cur" ) ) + return 0 + ;; + pin|detach) + if [[ $prev == "$command" ]]; then + COMPREPLY=( $( compgen -W "$LINK_TYPE" -- "$cur" ) ) + else + _filedir + fi return 0 ;; *) [[ $prev == $object ]] && \ - COMPREPLY=( $( compgen -W 'help probe' -- "$cur" ) ) + COMPREPLY=( $( compgen -W 'help pin show list' -- "$cur" ) ) ;; esac ;; diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c index b3745ed711ba..68a70ac03c80 100644 --- a/tools/bpf/bpftool/btf.c +++ b/tools/bpf/bpftool/btf.c @@ -8,14 +8,14 @@ #include <stdio.h> #include <string.h> #include <unistd.h> -#include <bpf/bpf.h> -#include <bpf/btf.h> -#include <bpf/libbpf.h> #include <linux/btf.h> -#include <linux/hashtable.h> #include <sys/types.h> #include <sys/stat.h> -#include <unistd.h> + +#include <bpf/bpf.h> +#include <bpf/btf.h> +#include <bpf/hashmap.h> +#include <bpf/libbpf.h> #include "json_writer.h" #include "main.h" @@ -37,16 +37,10 @@ static const char * const btf_kind_str[NR_BTF_KINDS] = { [BTF_KIND_FUNC_PROTO] = "FUNC_PROTO", [BTF_KIND_VAR] = "VAR", [BTF_KIND_DATASEC] = "DATASEC", -}; - -struct btf_attach_table { - DECLARE_HASHTABLE(table, 16); -}; - -struct btf_attach_point { - __u32 obj_id; - __u32 btf_id; - struct hlist_node hash; + [BTF_KIND_FLOAT] = "FLOAT", + [BTF_KIND_DECL_TAG] = "DECL_TAG", + [BTF_KIND_TYPE_TAG] = "TYPE_TAG", + [BTF_KIND_ENUM64] = "ENUM64", }; static const char *btf_int_enc_str(__u8 encoding) @@ -71,7 +65,9 @@ static const char *btf_var_linkage_str(__u32 linkage) case BTF_VAR_STATIC: return "static"; case BTF_VAR_GLOBAL_ALLOCATED: - return "global-alloc"; + return "global"; + case BTF_VAR_GLOBAL_EXTERN: + return "extern"; default: return "(unknown)"; } @@ -98,26 +94,28 @@ static const char *btf_str(const struct btf *btf, __u32 off) return btf__name_by_offset(btf, off) ? : "(invalid)"; } +static int btf_kind_safe(int kind) +{ + return kind <= BTF_KIND_MAX ? kind : BTF_KIND_UNKN; +} + static int dump_btf_type(const struct btf *btf, __u32 id, const struct btf_type *t) { json_writer_t *w = json_wtr; - int kind, safe_kind; - - kind = BTF_INFO_KIND(t->info); - safe_kind = kind <= BTF_KIND_MAX ? kind : BTF_KIND_UNKN; + int kind = btf_kind(t); if (json_output) { jsonw_start_object(w); jsonw_uint_field(w, "id", id); - jsonw_string_field(w, "kind", btf_kind_str[safe_kind]); + jsonw_string_field(w, "kind", btf_kind_str[btf_kind_safe(kind)]); jsonw_string_field(w, "name", btf_str(btf, t->name_off)); } else { - printf("[%u] %s '%s'", id, btf_kind_str[safe_kind], + printf("[%u] %s '%s'", id, btf_kind_str[btf_kind_safe(kind)], btf_str(btf, t->name_off)); } - switch (BTF_INFO_KIND(t->info)) { + switch (kind) { case BTF_KIND_INT: { __u32 v = *(__u32 *)(t + 1); const char *enc; @@ -141,6 +139,7 @@ static int dump_btf_type(const struct btf *btf, __u32 id, case BTF_KIND_VOLATILE: case BTF_KIND_RESTRICT: case BTF_KIND_TYPEDEF: + case BTF_KIND_TYPE_TAG: if (json_output) jsonw_uint_field(w, "type_id", t->type); else @@ -209,15 +208,18 @@ static int dump_btf_type(const struct btf *btf, __u32 id, case BTF_KIND_ENUM: { const struct btf_enum *v = (const void *)(t + 1); __u16 vlen = BTF_INFO_VLEN(t->info); + const char *encoding; int i; + encoding = btf_kflag(t) ? "SIGNED" : "UNSIGNED"; if (json_output) { + jsonw_string_field(w, "encoding", encoding); jsonw_uint_field(w, "size", t->size); jsonw_uint_field(w, "vlen", vlen); jsonw_name(w, "values"); jsonw_start_array(w); } else { - printf(" size=%u vlen=%u", t->size, vlen); + printf(" encoding=%s size=%u vlen=%u", encoding, t->size, vlen); } for (i = 0; i < vlen; i++, v++) { const char *name = btf_str(btf, v->name_off); @@ -225,10 +227,57 @@ static int dump_btf_type(const struct btf *btf, __u32 id, if (json_output) { jsonw_start_object(w); jsonw_string_field(w, "name", name); - jsonw_uint_field(w, "val", v->val); + if (btf_kflag(t)) + jsonw_int_field(w, "val", v->val); + else + jsonw_uint_field(w, "val", v->val); jsonw_end_object(w); } else { - printf("\n\t'%s' val=%u", name, v->val); + if (btf_kflag(t)) + printf("\n\t'%s' val=%d", name, v->val); + else + printf("\n\t'%s' val=%u", name, v->val); + } + } + if (json_output) + jsonw_end_array(w); + break; + } + case BTF_KIND_ENUM64: { + const struct btf_enum64 *v = btf_enum64(t); + __u16 vlen = btf_vlen(t); + const char *encoding; + int i; + + encoding = btf_kflag(t) ? "SIGNED" : "UNSIGNED"; + if (json_output) { + jsonw_string_field(w, "encoding", encoding); + jsonw_uint_field(w, "size", t->size); + jsonw_uint_field(w, "vlen", vlen); + jsonw_name(w, "values"); + jsonw_start_array(w); + } else { + printf(" encoding=%s size=%u vlen=%u", encoding, t->size, vlen); + } + for (i = 0; i < vlen; i++, v++) { + const char *name = btf_str(btf, v->name_off); + __u64 val = ((__u64)v->val_hi32 << 32) | v->val_lo32; + + if (json_output) { + jsonw_start_object(w); + jsonw_string_field(w, "name", name); + if (btf_kflag(t)) + jsonw_int_field(w, "val", val); + else + jsonw_uint_field(w, "val", val); + jsonw_end_object(w); + } else { + if (btf_kflag(t)) + printf("\n\t'%s' val=%lldLL", name, + (unsigned long long)val); + else + printf("\n\t'%s' val=%lluULL", name, + (unsigned long long)val); } } if (json_output) @@ -300,7 +349,8 @@ static int dump_btf_type(const struct btf *btf, __u32 id, break; } case BTF_KIND_DATASEC: { - const struct btf_var_secinfo *v = (const void *)(t+1); + const struct btf_var_secinfo *v = (const void *)(t + 1); + const struct btf_type *vt; __u16 vlen = BTF_INFO_VLEN(t->info); int i; @@ -322,12 +372,37 @@ static int dump_btf_type(const struct btf *btf, __u32 id, } else { printf("\n\ttype_id=%u offset=%u size=%u", v->type, v->offset, v->size); + + if (v->type < btf__type_cnt(btf)) { + vt = btf__type_by_id(btf, v->type); + printf(" (%s '%s')", + btf_kind_str[btf_kind_safe(btf_kind(vt))], + btf_str(btf, vt->name_off)); + } } } if (json_output) jsonw_end_array(w); break; } + case BTF_KIND_FLOAT: { + if (json_output) + jsonw_uint_field(w, "size", t->size); + else + printf(" size=%u", t->size); + break; + } + case BTF_KIND_DECL_TAG: { + const struct btf_decl_tag *tag = (const void *)(t + 1); + + if (json_output) { + jsonw_uint_field(w, "type_id", t->type); + jsonw_int_field(w, "component_idx", tag->component_idx); + } else { + printf(" type_id=%u component_idx=%d", t->type, tag->component_idx); + } + break; + } default: break; } @@ -358,9 +433,15 @@ static int dump_btf_raw(const struct btf *btf, dump_btf_type(btf, root_type_ids[i], t); } } else { - int cnt = btf__get_nr_types(btf); + const struct btf *base; + int cnt = btf__type_cnt(btf); + int start_id = 1; + + base = btf__base_btf(btf); + if (base) + start_id = btf__type_cnt(base); - for (i = 1; i <= cnt; i++) { + for (i = start_id; i < cnt; i++) { t = btf__type_by_id(btf, i); dump_btf_type(btf, i, t); } @@ -385,10 +466,14 @@ static int dump_btf_c(const struct btf *btf, struct btf_dump *d; int err = 0, i; - d = btf_dump__new(btf, NULL, NULL, btf_dump_printf); - if (IS_ERR(d)) - return PTR_ERR(d); + d = btf_dump__new(btf, btf_dump_printf, NULL, NULL); + err = libbpf_get_error(d); + if (err) + return err; + printf("#ifndef __VMLINUX_H__\n"); + printf("#define __VMLINUX_H__\n"); + printf("\n"); printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n"); printf("#pragma clang attribute push (__attribute__((preserve_access_index)), apply_to = record)\n"); printf("#endif\n\n"); @@ -400,9 +485,9 @@ static int dump_btf_c(const struct btf *btf, goto done; } } else { - int cnt = btf__get_nr_types(btf); + int cnt = btf__type_cnt(btf); - for (i = 1; i <= cnt; i++) { + for (i = 1; i < cnt; i++) { err = btf_dump__dump_type(d, i); if (err) goto done; @@ -412,63 +497,62 @@ static int dump_btf_c(const struct btf *btf, printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n"); printf("#pragma clang attribute pop\n"); printf("#endif\n"); + printf("\n"); + printf("#endif /* __VMLINUX_H__ */\n"); done: btf_dump__free(d); return err; } -static struct btf *btf__parse_raw(const char *file) -{ - struct btf *btf; - struct stat st; - __u8 *buf; - FILE *f; - - if (stat(file, &st)) - return NULL; +static const char sysfs_vmlinux[] = "/sys/kernel/btf/vmlinux"; - f = fopen(file, "rb"); - if (!f) - return NULL; +static struct btf *get_vmlinux_btf_from_sysfs(void) +{ + struct btf *base; - buf = malloc(st.st_size); - if (!buf) { - btf = ERR_PTR(-ENOMEM); - goto exit_close; + base = btf__parse(sysfs_vmlinux, NULL); + if (libbpf_get_error(base)) { + p_err("failed to parse vmlinux BTF at '%s': %ld\n", + sysfs_vmlinux, libbpf_get_error(base)); + base = NULL; } - if ((size_t) st.st_size != fread(buf, 1, st.st_size, f)) { - btf = ERR_PTR(-EINVAL); - goto exit_free; - } - - btf = btf__new(buf, st.st_size); - -exit_free: - free(buf); -exit_close: - fclose(f); - return btf; + return base; } -static bool is_btf_raw(const char *file) +#define BTF_NAME_BUFF_LEN 64 + +static bool btf_is_kernel_module(__u32 btf_id) { - __u16 magic = 0; - int fd, nb_read; + struct bpf_btf_info btf_info = {}; + char btf_name[BTF_NAME_BUFF_LEN]; + int btf_fd; + __u32 len; + int err; - fd = open(file, O_RDONLY); - if (fd < 0) + btf_fd = bpf_btf_get_fd_by_id(btf_id); + if (btf_fd < 0) { + p_err("can't get BTF object by id (%u): %s", btf_id, strerror(errno)); return false; + } - nb_read = read(fd, &magic, sizeof(magic)); - close(fd); - return nb_read == sizeof(magic) && magic == BTF_MAGIC; + len = sizeof(btf_info); + btf_info.name = ptr_to_u64(btf_name); + btf_info.name_len = sizeof(btf_name); + err = bpf_obj_get_info_by_fd(btf_fd, &btf_info, &len); + close(btf_fd); + if (err) { + p_err("can't get BTF (ID %u) object info: %s", btf_id, strerror(errno)); + return false; + } + + return btf_info.kernel_btf && strncmp(btf_name, "vmlinux", sizeof(btf_name)) != 0; } static int do_dump(int argc, char **argv) { - struct btf *btf = NULL; + struct btf *btf = NULL, *base = NULL; __u32 root_type_ids[2]; int root_type_cnt = 0; bool dump_c = false; @@ -482,7 +566,6 @@ static int do_dump(int argc, char **argv) return -1; } src = GET_ARG(); - if (is_prefix(src, "map")) { struct bpf_map_info info = {}; __u32 len = sizeof(info); @@ -543,16 +626,18 @@ static int do_dump(int argc, char **argv) } NEXT_ARG(); } else if (is_prefix(src, "file")) { - if (is_btf_raw(*argv)) - btf = btf__parse_raw(*argv); - else - btf = btf__parse_elf(*argv, NULL); + const char sysfs_prefix[] = "/sys/kernel/btf/"; + + if (!base_btf && + strncmp(*argv, sysfs_prefix, sizeof(sysfs_prefix) - 1) == 0 && + strcmp(*argv, sysfs_vmlinux) != 0) + base = get_vmlinux_btf_from_sysfs(); - if (IS_ERR(btf)) { - err = PTR_ERR(btf); - btf = NULL; + btf = btf__parse_split(*argv, base ?: base_btf); + err = libbpf_get_error(btf); + if (!btf) { p_err("failed to load BTF from %s: %s", - *argv, strerror(err)); + *argv, strerror(errno)); goto done; } NEXT_ARG(); @@ -567,6 +652,7 @@ static int do_dump(int argc, char **argv) NEXT_ARG(); if (argc < 1) { p_err("expecting value for 'format' option\n"); + err = -EINVAL; goto done; } if (strcmp(*argv, "c") == 0) { @@ -576,24 +662,28 @@ static int do_dump(int argc, char **argv) } else { p_err("unrecognized format specifier: '%s', possible values: raw, c", *argv); + err = -EINVAL; goto done; } NEXT_ARG(); } else { p_err("unrecognized option: '%s'", *argv); + err = -EINVAL; goto done; } } if (!btf) { - err = btf__get_from_id(btf_id, &btf); - if (err) { - p_err("get btf by id (%u): %s", btf_id, strerror(err)); - goto done; + if (!base_btf && btf_is_kernel_module(btf_id)) { + p_info("Warning: valid base BTF was not specified with -B option, falling back to standard base BTF (%s)", + sysfs_vmlinux); + base_btf = get_vmlinux_btf_from_sysfs(); } + + btf = btf__load_from_kernel_by_id_split(btf_id, base_btf); + err = libbpf_get_error(btf); if (!btf) { - err = ENOENT; - p_err("can't find btf with ID (%u)", btf_id); + p_err("get btf by id (%u): %s", btf_id, strerror(errno)); goto done; } } @@ -612,6 +702,7 @@ static int do_dump(int argc, char **argv) done: close(fd); btf__free(btf); + btf__free(base); return err; } @@ -642,21 +733,8 @@ static int btf_parse_fd(int *argc, char ***argv) return fd; } -static void delete_btf_table(struct btf_attach_table *tab) -{ - struct btf_attach_point *obj; - struct hlist_node *tmp; - - unsigned int bkt; - - hash_for_each_safe(tab->table, bkt, tmp, obj, hash) { - hash_del(&obj->hash); - free(obj); - } -} - static int -build_btf_type_table(struct btf_attach_table *tab, enum bpf_obj_type type, +build_btf_type_table(struct hashmap *tab, enum bpf_obj_type type, void *info, __u32 *len) { static const char * const names[] = { @@ -664,7 +742,6 @@ build_btf_type_table(struct btf_attach_table *tab, enum bpf_obj_type type, [BPF_OBJ_PROG] = "prog", [BPF_OBJ_MAP] = "map", }; - struct btf_attach_point *obj_node; __u32 btf_id, id = 0; int err; int fd; @@ -738,27 +815,25 @@ build_btf_type_table(struct btf_attach_table *tab, enum bpf_obj_type type, if (!btf_id) continue; - obj_node = calloc(1, sizeof(*obj_node)); - if (!obj_node) { - p_err("failed to allocate memory: %s", strerror(errno)); + err = hashmap__append(tab, u32_as_hash_field(btf_id), + u32_as_hash_field(id)); + if (err) { + p_err("failed to append entry to hashmap for BTF ID %u, object ID %u: %s", + btf_id, id, strerror(-err)); goto err_free; } - - obj_node->obj_id = id; - obj_node->btf_id = btf_id; - hash_add(tab->table, &obj_node->hash, obj_node->btf_id); } return 0; err_free: - delete_btf_table(tab); + hashmap__free(tab); return err; } static int -build_btf_tables(struct btf_attach_table *btf_prog_table, - struct btf_attach_table *btf_map_table) +build_btf_tables(struct hashmap *btf_prog_table, + struct hashmap *btf_map_table) { struct bpf_prog_info prog_info; __u32 prog_len = sizeof(prog_info); @@ -774,7 +849,7 @@ build_btf_tables(struct btf_attach_table *btf_prog_table, err = build_btf_type_table(btf_map_table, BPF_OBJ_MAP, &map_info, &map_len); if (err) { - delete_btf_table(btf_prog_table); + hashmap__free(btf_prog_table); return err; } @@ -783,38 +858,48 @@ build_btf_tables(struct btf_attach_table *btf_prog_table, static void show_btf_plain(struct bpf_btf_info *info, int fd, - struct btf_attach_table *btf_prog_table, - struct btf_attach_table *btf_map_table) + struct hashmap *btf_prog_table, + struct hashmap *btf_map_table) { - struct btf_attach_point *obj; + struct hashmap_entry *entry; + const char *name = u64_to_ptr(info->name); int n; printf("%u: ", info->id); + if (info->kernel_btf) + printf("name [%s] ", name); + else if (name && name[0]) + printf("name %s ", name); + else + printf("name <anon> "); printf("size %uB", info->btf_size); n = 0; - hash_for_each_possible(btf_prog_table->table, obj, hash, info->id) { - if (obj->btf_id == info->id) - printf("%s%u", n++ == 0 ? " prog_ids " : ",", - obj->obj_id); + hashmap__for_each_key_entry(btf_prog_table, entry, + u32_as_hash_field(info->id)) { + printf("%s%u", n++ == 0 ? " prog_ids " : ",", + hash_field_as_u32(entry->value)); } n = 0; - hash_for_each_possible(btf_map_table->table, obj, hash, info->id) { - if (obj->btf_id == info->id) - printf("%s%u", n++ == 0 ? " map_ids " : ",", - obj->obj_id); + hashmap__for_each_key_entry(btf_map_table, entry, + u32_as_hash_field(info->id)) { + printf("%s%u", n++ == 0 ? " map_ids " : ",", + hash_field_as_u32(entry->value)); } + emit_obj_refs_plain(refs_table, info->id, "\n\tpids "); + printf("\n"); } static void show_btf_json(struct bpf_btf_info *info, int fd, - struct btf_attach_table *btf_prog_table, - struct btf_attach_table *btf_map_table) + struct hashmap *btf_prog_table, + struct hashmap *btf_map_table) { - struct btf_attach_point *obj; + struct hashmap_entry *entry; + const char *name = u64_to_ptr(info->name); jsonw_start_object(json_wtr); /* btf object */ jsonw_uint_field(json_wtr, "id", info->id); @@ -822,37 +907,58 @@ show_btf_json(struct bpf_btf_info *info, int fd, jsonw_name(json_wtr, "prog_ids"); jsonw_start_array(json_wtr); /* prog_ids */ - hash_for_each_possible(btf_prog_table->table, obj, hash, - info->id) { - if (obj->btf_id == info->id) - jsonw_uint(json_wtr, obj->obj_id); + hashmap__for_each_key_entry(btf_prog_table, entry, + u32_as_hash_field(info->id)) { + jsonw_uint(json_wtr, hash_field_as_u32(entry->value)); } jsonw_end_array(json_wtr); /* prog_ids */ jsonw_name(json_wtr, "map_ids"); jsonw_start_array(json_wtr); /* map_ids */ - hash_for_each_possible(btf_map_table->table, obj, hash, - info->id) { - if (obj->btf_id == info->id) - jsonw_uint(json_wtr, obj->obj_id); + hashmap__for_each_key_entry(btf_map_table, entry, + u32_as_hash_field(info->id)) { + jsonw_uint(json_wtr, hash_field_as_u32(entry->value)); } jsonw_end_array(json_wtr); /* map_ids */ + + emit_obj_refs_json(refs_table, info->id, json_wtr); /* pids */ + + jsonw_bool_field(json_wtr, "kernel", info->kernel_btf); + + if (name && name[0]) + jsonw_string_field(json_wtr, "name", name); + jsonw_end_object(json_wtr); /* btf object */ } static int -show_btf(int fd, struct btf_attach_table *btf_prog_table, - struct btf_attach_table *btf_map_table) +show_btf(int fd, struct hashmap *btf_prog_table, + struct hashmap *btf_map_table) { - struct bpf_btf_info info = {}; + struct bpf_btf_info info; __u32 len = sizeof(info); + char name[64]; int err; + memset(&info, 0, sizeof(info)); err = bpf_obj_get_info_by_fd(fd, &info, &len); if (err) { p_err("can't get BTF object info: %s", strerror(errno)); return -1; } + /* if kernel support emitting BTF object name, pass name pointer */ + if (info.name_len) { + memset(&info, 0, sizeof(info)); + info.name_len = sizeof(name); + info.name = ptr_to_u64(name); + len = sizeof(info); + + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err) { + p_err("can't get BTF object info: %s", strerror(errno)); + return -1; + } + } if (json_output) show_btf_json(&info, fd, btf_prog_table, btf_map_table); @@ -864,8 +970,8 @@ show_btf(int fd, struct btf_attach_table *btf_prog_table, static int do_show(int argc, char **argv) { - struct btf_attach_table btf_prog_table; - struct btf_attach_table btf_map_table; + struct hashmap *btf_prog_table; + struct hashmap *btf_map_table; int err, fd = -1; __u32 id = 0; @@ -881,17 +987,28 @@ static int do_show(int argc, char **argv) return BAD_ARG(); } - hash_init(btf_prog_table.table); - hash_init(btf_map_table.table); - err = build_btf_tables(&btf_prog_table, &btf_map_table); + btf_prog_table = hashmap__new(hash_fn_for_key_as_id, + equal_fn_for_key_as_id, NULL); + btf_map_table = hashmap__new(hash_fn_for_key_as_id, + equal_fn_for_key_as_id, NULL); + if (IS_ERR(btf_prog_table) || IS_ERR(btf_map_table)) { + hashmap__free(btf_prog_table); + hashmap__free(btf_map_table); + if (fd >= 0) + close(fd); + p_err("failed to create hashmap for object references"); + return -1; + } + err = build_btf_tables(btf_prog_table, btf_map_table); if (err) { if (fd >= 0) close(fd); return err; } + build_obj_refs_table(&refs_table, BPF_OBJ_BTF); if (fd >= 0) { - err = show_btf(fd, &btf_prog_table, &btf_map_table); + err = show_btf(fd, btf_prog_table, btf_map_table); close(fd); goto exit_free; } @@ -923,7 +1040,7 @@ static int do_show(int argc, char **argv) break; } - err = show_btf(fd, &btf_prog_table, &btf_map_table); + err = show_btf(fd, btf_prog_table, btf_map_table); close(fd); if (err) break; @@ -933,8 +1050,9 @@ static int do_show(int argc, char **argv) jsonw_end_array(json_wtr); /* root array */ exit_free: - delete_btf_table(&btf_prog_table); - delete_btf_table(&btf_map_table); + hashmap__free(btf_prog_table); + hashmap__free(btf_map_table); + delete_obj_refs_table(refs_table); return err; } @@ -947,17 +1065,18 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s btf { show | list } [id BTF_ID]\n" - " %s btf dump BTF_SRC [format FORMAT]\n" - " %s btf help\n" + "Usage: %1$s %2$s { show | list } [id BTF_ID]\n" + " %1$s %2$s dump BTF_SRC [format FORMAT]\n" + " %1$s %2$s help\n" "\n" " BTF_SRC := { id BTF_ID | prog PROG | map MAP [{key | value | kv | all}] | file FILE }\n" " FORMAT := { raw | c }\n" " " HELP_SPEC_MAP "\n" " " HELP_SPEC_PROGRAM "\n" - " " HELP_SPEC_OPTIONS "\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-B|--base-btf} }\n" "", - bin_name, bin_name, bin_name); + bin_name, "btf"); return 0; } diff --git a/tools/bpf/bpftool/btf_dumper.c b/tools/bpf/bpftool/btf_dumper.c index 01cc52b834fa..19924b6ce796 100644 --- a/tools/bpf/bpftool/btf_dumper.c +++ b/tools/bpf/bpftool/btf_dumper.c @@ -4,11 +4,13 @@ #include <ctype.h> #include <stdio.h> /* for (FILE *) used by json_writer */ #include <string.h> +#include <unistd.h> #include <asm/byteorder.h> #include <linux/bitops.h> #include <linux/btf.h> #include <linux/err.h> #include <bpf/btf.h> +#include <bpf/bpf.h> #include "json_writer.h" #include "main.h" @@ -22,13 +24,112 @@ static int btf_dumper_do_type(const struct btf_dumper *d, __u32 type_id, __u8 bit_offset, const void *data); -static void btf_dumper_ptr(const void *data, json_writer_t *jw, - bool is_plain_text) +static int btf_dump_func(const struct btf *btf, char *func_sig, + const struct btf_type *func_proto, + const struct btf_type *func, int pos, int size); + +static int dump_prog_id_as_func_ptr(const struct btf_dumper *d, + const struct btf_type *func_proto, + __u32 prog_id) { - if (is_plain_text) - jsonw_printf(jw, "%p", *(void **)data); + const struct btf_type *func_type; + int prog_fd = -1, func_sig_len; + struct bpf_prog_info info = {}; + __u32 info_len = sizeof(info); + const char *prog_name = NULL; + struct btf *prog_btf = NULL; + struct bpf_func_info finfo; + __u32 finfo_rec_size; + char prog_str[1024]; + int err; + + /* Get the ptr's func_proto */ + func_sig_len = btf_dump_func(d->btf, prog_str, func_proto, NULL, 0, + sizeof(prog_str)); + if (func_sig_len == -1) + return -1; + + if (!prog_id) + goto print; + + /* Get the bpf_prog's name. Obtain from func_info. */ + prog_fd = bpf_prog_get_fd_by_id(prog_id); + if (prog_fd < 0) + goto print; + + err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len); + if (err) + goto print; + + if (!info.btf_id || !info.nr_func_info) + goto print; + + finfo_rec_size = info.func_info_rec_size; + memset(&info, 0, sizeof(info)); + info.nr_func_info = 1; + info.func_info_rec_size = finfo_rec_size; + info.func_info = ptr_to_u64(&finfo); + + err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len); + if (err) + goto print; + + prog_btf = btf__load_from_kernel_by_id(info.btf_id); + if (libbpf_get_error(prog_btf)) + goto print; + func_type = btf__type_by_id(prog_btf, finfo.type_id); + if (!func_type || !btf_is_func(func_type)) + goto print; + + prog_name = btf__name_by_offset(prog_btf, func_type->name_off); + +print: + if (!prog_id) + snprintf(&prog_str[func_sig_len], + sizeof(prog_str) - func_sig_len, " 0"); + else if (prog_name) + snprintf(&prog_str[func_sig_len], + sizeof(prog_str) - func_sig_len, + " %s/prog_id:%u", prog_name, prog_id); else - jsonw_printf(jw, "%lu", *(unsigned long *)data); + snprintf(&prog_str[func_sig_len], + sizeof(prog_str) - func_sig_len, + " <unknown_prog_name>/prog_id:%u", prog_id); + + prog_str[sizeof(prog_str) - 1] = '\0'; + jsonw_string(d->jw, prog_str); + btf__free(prog_btf); + if (prog_fd >= 0) + close(prog_fd); + return 0; +} + +static void btf_dumper_ptr(const struct btf_dumper *d, + const struct btf_type *t, + const void *data) +{ + unsigned long value = *(unsigned long *)data; + const struct btf_type *ptr_type; + __s32 ptr_type_id; + + if (!d->prog_id_as_func_ptr || value > UINT32_MAX) + goto print_ptr_value; + + ptr_type_id = btf__resolve_type(d->btf, t->type); + if (ptr_type_id < 0) + goto print_ptr_value; + ptr_type = btf__type_by_id(d->btf, ptr_type_id); + if (!ptr_type || !btf_is_func_proto(ptr_type)) + goto print_ptr_value; + + if (!dump_prog_id_as_func_ptr(d, ptr_type, value)) + return; + +print_ptr_value: + if (d->is_plain_text) + jsonw_printf(d->jw, "%p", (void *)value); + else + jsonw_printf(d->jw, "%lu", value); } static int btf_dumper_modifier(const struct btf_dumper *d, __u32 type_id, @@ -43,9 +144,104 @@ static int btf_dumper_modifier(const struct btf_dumper *d, __u32 type_id, return btf_dumper_do_type(d, actual_type_id, bit_offset, data); } -static void btf_dumper_enum(const void *data, json_writer_t *jw) +static int btf_dumper_enum(const struct btf_dumper *d, + const struct btf_type *t, + const void *data) { - jsonw_printf(jw, "%d", *(int *)data); + const struct btf_enum *enums = btf_enum(t); + __s64 value; + __u16 i; + + switch (t->size) { + case 8: + value = *(__s64 *)data; + break; + case 4: + value = *(__s32 *)data; + break; + case 2: + value = *(__s16 *)data; + break; + case 1: + value = *(__s8 *)data; + break; + default: + return -EINVAL; + } + + for (i = 0; i < btf_vlen(t); i++) { + if (value == enums[i].val) { + jsonw_string(d->jw, + btf__name_by_offset(d->btf, + enums[i].name_off)); + return 0; + } + } + + jsonw_int(d->jw, value); + return 0; +} + +static int btf_dumper_enum64(const struct btf_dumper *d, + const struct btf_type *t, + const void *data) +{ + const struct btf_enum64 *enums = btf_enum64(t); + __u32 val_lo32, val_hi32; + __u64 value; + __u16 i; + + value = *(__u64 *)data; + val_lo32 = (__u32)value; + val_hi32 = value >> 32; + + for (i = 0; i < btf_vlen(t); i++) { + if (val_lo32 == enums[i].val_lo32 && val_hi32 == enums[i].val_hi32) { + jsonw_string(d->jw, + btf__name_by_offset(d->btf, + enums[i].name_off)); + return 0; + } + } + + jsonw_int(d->jw, value); + return 0; +} + +static bool is_str_array(const struct btf *btf, const struct btf_array *arr, + const char *s) +{ + const struct btf_type *elem_type; + const char *end_s; + + if (!arr->nelems) + return false; + + elem_type = btf__type_by_id(btf, arr->type); + /* Not skipping typedef. typedef to char does not count as + * a string now. + */ + while (elem_type && btf_is_mod(elem_type)) + elem_type = btf__type_by_id(btf, elem_type->type); + + if (!elem_type || !btf_is_int(elem_type) || elem_type->size != 1) + return false; + + if (btf_int_encoding(elem_type) != BTF_INT_CHAR && + strcmp("char", btf__name_by_offset(btf, elem_type->name_off))) + return false; + + end_s = s + arr->nelems; + while (s < end_s) { + if (!*s) + return true; + if (*s <= 0x1f || *s >= 0x7f) + return false; + s++; + } + + /* '\0' is not found */ + return false; } static int btf_dumper_array(const struct btf_dumper *d, __u32 type_id, @@ -57,6 +253,11 @@ static int btf_dumper_array(const struct btf_dumper *d, __u32 type_id, int ret = 0; __u32 i; + if (is_str_array(d->btf, arr, data)) { + jsonw_string(d->jw, data); + return 0; + } + elem_size = btf__resolve_size(d->btf, arr->type); if (elem_size < 0) return elem_size; @@ -106,8 +307,8 @@ static void btf_int128_print(json_writer_t *jw, const void *data, } } -static void btf_int128_shift(__u64 *print_num, u16 left_shift_bits, - u16 right_shift_bits) +static void btf_int128_shift(__u64 *print_num, __u16 left_shift_bits, + __u16 right_shift_bits) { __u64 upper_num, lower_num; @@ -251,7 +452,7 @@ static int btf_dumper_int(const struct btf_type *t, __u8 bit_offset, *(char *)data); break; case BTF_INT_BOOL: - jsonw_bool(jw, *(int *)data); + jsonw_bool(jw, *(bool *)data); break; default: /* shouldn't happen */ @@ -366,10 +567,11 @@ static int btf_dumper_do_type(const struct btf_dumper *d, __u32 type_id, case BTF_KIND_ARRAY: return btf_dumper_array(d, type_id, data); case BTF_KIND_ENUM: - btf_dumper_enum(data, d->jw); - return 0; + return btf_dumper_enum(d, t, data); + case BTF_KIND_ENUM64: + return btf_dumper_enum64(d, t, data); case BTF_KIND_PTR: - btf_dumper_ptr(data, d->jw, d->is_plain_text); + btf_dumper_ptr(d, t, data); return 0; case BTF_KIND_UNKN: jsonw_printf(d->jw, "(unknown)"); @@ -414,10 +616,6 @@ int btf_dumper_type(const struct btf_dumper *d, __u32 type_id, return -1; \ } while (0) -static int btf_dump_func(const struct btf *btf, char *func_sig, - const struct btf_type *func_proto, - const struct btf_type *func, int pos, int size); - static int __btf_dumper_type_only(const struct btf *btf, __u32 type_id, char *func_sig, int pos, int size) { @@ -436,6 +634,7 @@ static int __btf_dumper_type_only(const struct btf *btf, __u32 type_id, switch (BTF_INFO_KIND(t->info)) { case BTF_KIND_INT: case BTF_KIND_TYPEDEF: + case BTF_KIND_FLOAT: BTF_PRINT_ARG("%s ", btf__name_by_offset(btf, t->name_off)); break; case BTF_KIND_STRUCT: @@ -447,6 +646,7 @@ static int __btf_dumper_type_only(const struct btf *btf, __u32 type_id, btf__name_by_offset(btf, t->name_off)); break; case BTF_KIND_ENUM: + case BTF_KIND_ENUM64: BTF_PRINT_ARG("enum %s ", btf__name_by_offset(btf, t->name_off)); break; @@ -526,8 +726,15 @@ static int btf_dump_func(const struct btf *btf, char *func_sig, BTF_PRINT_ARG(", "); if (arg->type) { BTF_PRINT_TYPE(arg->type); - BTF_PRINT_ARG("%s", - btf__name_by_offset(btf, arg->name_off)); + if (arg->name_off) + BTF_PRINT_ARG("%s", + btf__name_by_offset(btf, arg->name_off)); + else if (pos && func_sig[pos - 1] == ' ') + /* Remove unnecessary space for + * FUNC_PROTO that does not have + * arg->name_off + */ + func_sig[--pos] = '\0'; } else { BTF_PRINT_ARG("..."); } diff --git a/tools/bpf/bpftool/cfg.c b/tools/bpf/bpftool/cfg.c index 3e21f994f262..1951219a9af7 100644 --- a/tools/bpf/bpftool/cfg.c +++ b/tools/bpf/bpftool/cfg.c @@ -157,7 +157,7 @@ static bool cfg_partition_funcs(struct cfg *cfg, struct bpf_insn *cur, return false; } -static bool is_jmp_insn(u8 code) +static bool is_jmp_insn(__u8 code) { return BPF_CLASS(code) == BPF_JMP || BPF_CLASS(code) == BPF_JMP32; } @@ -176,7 +176,7 @@ static bool func_partition_bb_head(struct func_node *func) for (; cur <= end; cur++) { if (is_jmp_insn(cur->code)) { - u8 opcode = BPF_OP(cur->code); + __u8 opcode = BPF_OP(cur->code); if (opcode == BPF_EXIT || opcode == BPF_CALL) continue; diff --git a/tools/bpf/bpftool/cgroup.c b/tools/bpf/bpftool/cgroup.c index 62c6a1d7cd18..b46a998d8f8d 100644 --- a/tools/bpf/bpftool/cgroup.c +++ b/tools/bpf/bpftool/cgroup.c @@ -15,62 +15,92 @@ #include <unistd.h> #include <bpf/bpf.h> +#include <bpf/btf.h> #include "main.h" #define HELP_SPEC_ATTACH_FLAGS \ "ATTACH_FLAGS := { multi | override }" -#define HELP_SPEC_ATTACH_TYPES \ - " ATTACH_TYPE := { ingress | egress | sock_create |\n" \ - " sock_ops | device | bind4 | bind6 |\n" \ - " post_bind4 | post_bind6 | connect4 |\n" \ - " connect6 | sendmsg4 | sendmsg6 |\n" \ - " recvmsg4 | recvmsg6 | sysctl |\n" \ - " getsockopt | setsockopt }" +#define HELP_SPEC_ATTACH_TYPES \ + " ATTACH_TYPE := { cgroup_inet_ingress | cgroup_inet_egress |\n" \ + " cgroup_inet_sock_create | cgroup_sock_ops |\n" \ + " cgroup_device | cgroup_inet4_bind |\n" \ + " cgroup_inet6_bind | cgroup_inet4_post_bind |\n" \ + " cgroup_inet6_post_bind | cgroup_inet4_connect |\n" \ + " cgroup_inet6_connect | cgroup_inet4_getpeername |\n" \ + " cgroup_inet6_getpeername | cgroup_inet4_getsockname |\n" \ + " cgroup_inet6_getsockname | cgroup_udp4_sendmsg |\n" \ + " cgroup_udp6_sendmsg | cgroup_udp4_recvmsg |\n" \ + " cgroup_udp6_recvmsg | cgroup_sysctl |\n" \ + " cgroup_getsockopt | cgroup_setsockopt |\n" \ + " cgroup_inet_sock_release }" static unsigned int query_flags; - -static const char * const attach_type_strings[] = { - [BPF_CGROUP_INET_INGRESS] = "ingress", - [BPF_CGROUP_INET_EGRESS] = "egress", - [BPF_CGROUP_INET_SOCK_CREATE] = "sock_create", - [BPF_CGROUP_SOCK_OPS] = "sock_ops", - [BPF_CGROUP_DEVICE] = "device", - [BPF_CGROUP_INET4_BIND] = "bind4", - [BPF_CGROUP_INET6_BIND] = "bind6", - [BPF_CGROUP_INET4_CONNECT] = "connect4", - [BPF_CGROUP_INET6_CONNECT] = "connect6", - [BPF_CGROUP_INET4_POST_BIND] = "post_bind4", - [BPF_CGROUP_INET6_POST_BIND] = "post_bind6", - [BPF_CGROUP_UDP4_SENDMSG] = "sendmsg4", - [BPF_CGROUP_UDP6_SENDMSG] = "sendmsg6", - [BPF_CGROUP_SYSCTL] = "sysctl", - [BPF_CGROUP_UDP4_RECVMSG] = "recvmsg4", - [BPF_CGROUP_UDP6_RECVMSG] = "recvmsg6", - [BPF_CGROUP_GETSOCKOPT] = "getsockopt", - [BPF_CGROUP_SETSOCKOPT] = "setsockopt", - [__MAX_BPF_ATTACH_TYPE] = NULL, -}; +static struct btf *btf_vmlinux; +static __u32 btf_vmlinux_id; static enum bpf_attach_type parse_attach_type(const char *str) { + const char *attach_type_str; enum bpf_attach_type type; - for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) { - if (attach_type_strings[type] && - is_prefix(str, attach_type_strings[type])) + for (type = 0; ; type++) { + attach_type_str = libbpf_bpf_attach_type_str(type); + if (!attach_type_str) + break; + if (!strcmp(str, attach_type_str)) + return type; + } + + /* Also check traditionally used attach type strings. For these we keep + * allowing prefixed usage. + */ + for (type = 0; ; type++) { + attach_type_str = bpf_attach_type_input_str(type); + if (!attach_type_str) + break; + if (is_prefix(str, attach_type_str)) return type; } return __MAX_BPF_ATTACH_TYPE; } -static int show_bpf_prog(int id, const char *attach_type_str, +static void guess_vmlinux_btf_id(__u32 attach_btf_obj_id) +{ + struct bpf_btf_info btf_info = {}; + __u32 btf_len = sizeof(btf_info); + char name[16] = {}; + int err; + int fd; + + btf_info.name = ptr_to_u64(name); + btf_info.name_len = sizeof(name); + + fd = bpf_btf_get_fd_by_id(attach_btf_obj_id); + if (fd < 0) + return; + + err = bpf_obj_get_info_by_fd(fd, &btf_info, &btf_len); + if (err) + goto out; + + if (btf_info.kernel_btf && strncmp(name, "vmlinux", sizeof(name)) == 0) + btf_vmlinux_id = btf_info.id; + +out: + close(fd); +} + +static int show_bpf_prog(int id, enum bpf_attach_type attach_type, const char *attach_flags_str, int level) { + char prog_name[MAX_PROG_FULL_NAME]; + const char *attach_btf_name = NULL; struct bpf_prog_info info = {}; + const char *attach_type_str; __u32 info_len = sizeof(info); int prog_fd; @@ -83,21 +113,53 @@ static int show_bpf_prog(int id, const char *attach_type_str, return -1; } + attach_type_str = libbpf_bpf_attach_type_str(attach_type); + + if (btf_vmlinux) { + if (!btf_vmlinux_id) + guess_vmlinux_btf_id(info.attach_btf_obj_id); + + if (btf_vmlinux_id == info.attach_btf_obj_id && + info.attach_btf_id < btf__type_cnt(btf_vmlinux)) { + const struct btf_type *t = + btf__type_by_id(btf_vmlinux, info.attach_btf_id); + attach_btf_name = + btf__name_by_offset(btf_vmlinux, t->name_off); + } + } + + get_prog_full_name(&info, prog_fd, prog_name, sizeof(prog_name)); if (json_output) { jsonw_start_object(json_wtr); jsonw_uint_field(json_wtr, "id", info.id); - jsonw_string_field(json_wtr, "attach_type", - attach_type_str); - jsonw_string_field(json_wtr, "attach_flags", - attach_flags_str); - jsonw_string_field(json_wtr, "name", info.name); + if (attach_type_str) + jsonw_string_field(json_wtr, "attach_type", attach_type_str); + else + jsonw_uint_field(json_wtr, "attach_type", attach_type); + if (!(query_flags & BPF_F_QUERY_EFFECTIVE)) + jsonw_string_field(json_wtr, "attach_flags", attach_flags_str); + jsonw_string_field(json_wtr, "name", prog_name); + if (attach_btf_name) + jsonw_string_field(json_wtr, "attach_btf_name", attach_btf_name); + jsonw_uint_field(json_wtr, "attach_btf_obj_id", info.attach_btf_obj_id); + jsonw_uint_field(json_wtr, "attach_btf_id", info.attach_btf_id); jsonw_end_object(json_wtr); } else { - printf("%s%-8u %-15s %-15s %-15s\n", level ? " " : "", - info.id, - attach_type_str, - attach_flags_str, - info.name); + printf("%s%-8u ", level ? " " : "", info.id); + if (attach_type_str) + printf("%-15s", attach_type_str); + else + printf("type %-10u", attach_type); + if (query_flags & BPF_F_QUERY_EFFECTIVE) + printf(" %-15s", prog_name); + else + printf(" %-15s %-15s", attach_flags_str, prog_name); + if (attach_btf_name) + printf(" %-15s", attach_btf_name); + else if (info.attach_btf_id) + printf(" attach_btf_obj_id=%d attach_btf_id=%d", + info.attach_btf_obj_id, info.attach_btf_id); + printf("\n"); } close(prog_fd); @@ -136,47 +198,90 @@ static int cgroup_has_attached_progs(int cgroup_fd) return no_prog ? 0 : 1; } + +static int show_effective_bpf_progs(int cgroup_fd, enum bpf_attach_type type, + int level) +{ + LIBBPF_OPTS(bpf_prog_query_opts, p); + __u32 prog_ids[1024] = {0}; + __u32 iter; + int ret; + + p.query_flags = query_flags; + p.prog_cnt = ARRAY_SIZE(prog_ids); + p.prog_ids = prog_ids; + + ret = bpf_prog_query_opts(cgroup_fd, type, &p); + if (ret) + return ret; + + if (p.prog_cnt == 0) + return 0; + + for (iter = 0; iter < p.prog_cnt; iter++) + show_bpf_prog(prog_ids[iter], type, NULL, level); + + return 0; +} + static int show_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type, int level) { + LIBBPF_OPTS(bpf_prog_query_opts, p); + __u32 prog_attach_flags[1024] = {0}; const char *attach_flags_str; __u32 prog_ids[1024] = {0}; - __u32 prog_cnt, iter; - __u32 attach_flags; char buf[32]; + __u32 iter; int ret; - prog_cnt = ARRAY_SIZE(prog_ids); - ret = bpf_prog_query(cgroup_fd, type, query_flags, &attach_flags, - prog_ids, &prog_cnt); + p.query_flags = query_flags; + p.prog_cnt = ARRAY_SIZE(prog_ids); + p.prog_ids = prog_ids; + p.prog_attach_flags = prog_attach_flags; + + ret = bpf_prog_query_opts(cgroup_fd, type, &p); if (ret) return ret; - if (prog_cnt == 0) + if (p.prog_cnt == 0) return 0; - switch (attach_flags) { - case BPF_F_ALLOW_MULTI: - attach_flags_str = "multi"; - break; - case BPF_F_ALLOW_OVERRIDE: - attach_flags_str = "override"; - break; - case 0: - attach_flags_str = ""; - break; - default: - snprintf(buf, sizeof(buf), "unknown(%x)", attach_flags); - attach_flags_str = buf; - } + for (iter = 0; iter < p.prog_cnt; iter++) { + __u32 attach_flags; + + attach_flags = prog_attach_flags[iter] ?: p.attach_flags; + + switch (attach_flags) { + case BPF_F_ALLOW_MULTI: + attach_flags_str = "multi"; + break; + case BPF_F_ALLOW_OVERRIDE: + attach_flags_str = "override"; + break; + case 0: + attach_flags_str = ""; + break; + default: + snprintf(buf, sizeof(buf), "unknown(%x)", attach_flags); + attach_flags_str = buf; + } - for (iter = 0; iter < prog_cnt; iter++) - show_bpf_prog(prog_ids[iter], attach_type_strings[type], + show_bpf_prog(prog_ids[iter], type, attach_flags_str, level); + } return 0; } +static int show_bpf_progs(int cgroup_fd, enum bpf_attach_type type, + int level) +{ + return query_flags & BPF_F_QUERY_EFFECTIVE ? + show_effective_bpf_progs(cgroup_fd, type, level) : + show_attached_bpf_progs(cgroup_fd, type, level); +} + static int do_show(int argc, char **argv) { enum bpf_attach_type type; @@ -224,10 +329,13 @@ static int do_show(int argc, char **argv) if (json_output) jsonw_start_array(json_wtr); + else if (query_flags & BPF_F_QUERY_EFFECTIVE) + printf("%-8s %-15s %-15s\n", "ID", "AttachType", "Name"); else printf("%-8s %-15s %-15s %-15s\n", "ID", "AttachType", "AttachFlags", "Name"); + btf_vmlinux = libbpf_find_kernel_btf(); for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) { /* * Not all attach types may be supported, so it's expected, @@ -235,7 +343,7 @@ static int do_show(int argc, char **argv) * If we were able to get the show for at least one * attach type, let's return 0. */ - if (show_attached_bpf_progs(cgroup_fd, type, 0) == 0) + if (show_bpf_progs(cgroup_fd, type, 0) == 0) ret = 0; } @@ -291,8 +399,9 @@ static int do_show_tree_fn(const char *fpath, const struct stat *sb, printf("%s\n", fpath); } + btf_vmlinux = libbpf_find_kernel_btf(); for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) - show_attached_bpf_progs(cgroup_fd, type, ftw->level); + show_bpf_progs(cgroup_fd, type, ftw->level); if (errno == EINVAL) /* Last attach type does not support query. @@ -366,6 +475,11 @@ static int do_show_tree(int argc, char **argv) if (json_output) jsonw_start_array(json_wtr); + else if (query_flags & BPF_F_QUERY_EFFECTIVE) + printf("%s\n" + "%-8s %-15s %-15s\n", + "CgroupPath", + "ID", "AttachType", "Name"); else printf("%s\n" "%-8s %-15s %-15s %-15s\n", @@ -508,20 +622,19 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s %s { show | list } CGROUP [**effective**]\n" - " %s %s tree [CGROUP_ROOT] [**effective**]\n" - " %s %s attach CGROUP ATTACH_TYPE PROG [ATTACH_FLAGS]\n" - " %s %s detach CGROUP ATTACH_TYPE PROG\n" - " %s %s help\n" + "Usage: %1$s %2$s { show | list } CGROUP [**effective**]\n" + " %1$s %2$s tree [CGROUP_ROOT] [**effective**]\n" + " %1$s %2$s attach CGROUP ATTACH_TYPE PROG [ATTACH_FLAGS]\n" + " %1$s %2$s detach CGROUP ATTACH_TYPE PROG\n" + " %1$s %2$s help\n" "\n" HELP_SPEC_ATTACH_TYPES "\n" " " HELP_SPEC_ATTACH_FLAGS "\n" " " HELP_SPEC_PROGRAM "\n" - " " HELP_SPEC_OPTIONS "\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-f|--bpffs} }\n" "", - bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2]); + bin_name, argv[-2]); return 0; } diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c index b75b8ec5469c..0cdb4f711510 100644 --- a/tools/bpf/bpftool/common.c +++ b/tools/bpf/bpftool/common.c @@ -1,10 +1,11 @@ // SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) /* Copyright (C) 2017-2018 Netronome Systems, Inc. */ +#define _GNU_SOURCE #include <ctype.h> #include <errno.h> #include <fcntl.h> -#include <fts.h> +#include <ftw.h> #include <libgen.h> #include <mntent.h> #include <stdbool.h> @@ -12,16 +13,21 @@ #include <stdlib.h> #include <string.h> #include <unistd.h> -#include <linux/limits.h> -#include <linux/magic.h> #include <net/if.h> #include <sys/mount.h> #include <sys/resource.h> #include <sys/stat.h> #include <sys/vfs.h> +#include <linux/filter.h> +#include <linux/limits.h> +#include <linux/magic.h> +#include <linux/unistd.h> + #include <bpf/bpf.h> +#include <bpf/hashmap.h> #include <bpf/libbpf.h> /* libbpf_num_possible_cpus */ +#include <bpf/btf.h> #include "main.h" @@ -70,11 +76,73 @@ static bool is_bpffs(char *path) return (unsigned long)st_fs.f_type == BPF_FS_MAGIC; } +/* Probe whether kernel switched from memlock-based (RLIMIT_MEMLOCK) to + * memcg-based memory accounting for BPF maps and programs. This was done in + * commit 97306be45fbe ("Merge branch 'switch to memcg-based memory + * accounting'"), in Linux 5.11. + * + * Libbpf also offers to probe for memcg-based accounting vs rlimit, but does + * so by checking for the availability of a given BPF helper and this has + * failed on some kernels with backports in the past, see commit 6b4384ff1088 + * ("Revert "bpftool: Use libbpf 1.0 API mode instead of RLIMIT_MEMLOCK""). + * Instead, we can probe by lowering the process-based rlimit to 0, trying to + * load a BPF object, and resetting the rlimit. If the load succeeds then + * memcg-based accounting is supported. + * + * This would be too dangerous to do in the library, because multithreaded + * applications might attempt to load items while the rlimit is at 0. Given + * that bpftool is single-threaded, this is fine to do here. + */ +static bool known_to_need_rlimit(void) +{ + struct rlimit rlim_init, rlim_cur_zero = {}; + struct bpf_insn insns[] = { + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_EXIT_INSN(), + }; + size_t insn_cnt = ARRAY_SIZE(insns); + union bpf_attr attr; + int prog_fd, err; + + memset(&attr, 0, sizeof(attr)); + attr.prog_type = BPF_PROG_TYPE_SOCKET_FILTER; + attr.insns = ptr_to_u64(insns); + attr.insn_cnt = insn_cnt; + attr.license = ptr_to_u64("GPL"); + + if (getrlimit(RLIMIT_MEMLOCK, &rlim_init)) + return false; + + /* Drop the soft limit to zero. We maintain the hard limit to its + * current value, because lowering it would be a permanent operation + * for unprivileged users. + */ + rlim_cur_zero.rlim_max = rlim_init.rlim_max; + if (setrlimit(RLIMIT_MEMLOCK, &rlim_cur_zero)) + return false; + + /* Do not use bpf_prog_load() from libbpf here, because it calls + * bump_rlimit_memlock(), interfering with the current probe. + */ + prog_fd = syscall(__NR_bpf, BPF_PROG_LOAD, &attr, sizeof(attr)); + err = errno; + + /* reset soft rlimit to its initial value */ + setrlimit(RLIMIT_MEMLOCK, &rlim_init); + + if (prog_fd < 0) + return err == EPERM; + + close(prog_fd); + return false; +} + void set_max_rlimit(void) { struct rlimit rinf = { RLIM_INFINITY, RLIM_INFINITY }; - setrlimit(RLIMIT_MEMLOCK, &rinf); + if (known_to_need_rlimit()) + setrlimit(RLIMIT_MEMLOCK, &rinf); } static int @@ -123,24 +191,35 @@ int mount_tracefs(const char *target) return err; } -int open_obj_pinned(char *path, bool quiet) +int open_obj_pinned(const char *path, bool quiet) { - int fd; + char *pname; + int fd = -1; + + pname = strdup(path); + if (!pname) { + if (!quiet) + p_err("mem alloc failed"); + goto out_ret; + } - fd = bpf_obj_get(path); + fd = bpf_obj_get(pname); if (fd < 0) { if (!quiet) - p_err("bpf obj get (%s): %s", path, - errno == EACCES && !is_bpffs(dirname(path)) ? + p_err("bpf obj get (%s): %s", pname, + errno == EACCES && !is_bpffs(dirname(pname)) ? "directory not in bpf file system (bpffs)" : strerror(errno)); - return -1; + goto out_free; } +out_free: + free(pname); +out_ret: return fd; } -int open_obj_pinned_any(char *path, enum bpf_obj_type exp_type) +int open_obj_pinned_any(const char *path, enum bpf_obj_type exp_type) { enum bpf_obj_type type; int fd; @@ -171,6 +250,11 @@ int mount_bpffs_for_pin(const char *name) int err = 0; file = malloc(strlen(name) + 1); + if (!file) { + p_err("mem alloc failed"); + return -1; + } + strcpy(file, name); dir = dirname(file); @@ -211,39 +295,17 @@ int do_pin_fd(int fd, const char *name) return err; } -int do_pin_any(int argc, char **argv, int (*get_fd_by_id)(__u32)) +int do_pin_any(int argc, char **argv, int (*get_fd)(int *, char ***)) { - unsigned int id; - char *endptr; int err; int fd; - if (argc < 3) { - p_err("too few arguments, id ID and FILE path is required"); - return -1; - } else if (argc > 3) { - p_err("too many arguments"); - return -1; - } - - if (!is_prefix(*argv, "id")) { - p_err("expected 'id' got %s", *argv); - return -1; - } - NEXT_ARG(); - - id = strtoul(*argv, &endptr, 0); - if (*endptr) { - p_err("can't parse %s as ID", *argv); - return -1; - } - NEXT_ARG(); + if (!REQ_ARGS(3)) + return -EINVAL; - fd = get_fd_by_id(id); - if (fd < 0) { - p_err("can't open object by id (%u): %s", id, strerror(errno)); - return -1; - } + fd = get_fd(&argc, &argv); + if (fd < 0) + return fd; err = do_pin_fd(fd, *argv); @@ -257,6 +319,7 @@ const char *get_fd_type_name(enum bpf_obj_type type) [BPF_OBJ_UNKNOWN] = "unknown", [BPF_OBJ_PROG] = "prog", [BPF_OBJ_MAP] = "map", + [BPF_OBJ_LINK] = "link", }; if (type < 0 || type >= ARRAY_SIZE(names) || !names[type]) @@ -265,6 +328,49 @@ const char *get_fd_type_name(enum bpf_obj_type type) return names[type]; } +void get_prog_full_name(const struct bpf_prog_info *prog_info, int prog_fd, + char *name_buff, size_t buff_len) +{ + const char *prog_name = prog_info->name; + const struct btf_type *func_type; + const struct bpf_func_info finfo = {}; + struct bpf_prog_info info = {}; + __u32 info_len = sizeof(info); + struct btf *prog_btf = NULL; + + if (buff_len <= BPF_OBJ_NAME_LEN || + strlen(prog_info->name) < BPF_OBJ_NAME_LEN - 1) + goto copy_name; + + if (!prog_info->btf_id || prog_info->nr_func_info == 0) + goto copy_name; + + info.nr_func_info = 1; + info.func_info_rec_size = prog_info->func_info_rec_size; + if (info.func_info_rec_size > sizeof(finfo)) + info.func_info_rec_size = sizeof(finfo); + info.func_info = ptr_to_u64(&finfo); + + if (bpf_obj_get_info_by_fd(prog_fd, &info, &info_len)) + goto copy_name; + + prog_btf = btf__load_from_kernel_by_id(info.btf_id); + if (!prog_btf) + goto copy_name; + + func_type = btf__type_by_id(prog_btf, finfo.type_id); + if (!func_type || !btf_is_func(func_type)) + goto copy_name; + + prog_name = btf__name_by_offset(prog_btf, func_type->name_off); + +copy_name: + snprintf(name_buff, buff_len, "%s", prog_name); + + if (prog_btf) + btf__free(prog_btf); +} + int get_fd_type(int fd) { char path[PATH_MAX]; @@ -287,6 +393,8 @@ int get_fd_type(int fd) return BPF_OBJ_MAP; else if (strstr(buf, "bpf-prog")) return BPF_OBJ_PROG; + else if (strstr(buf, "bpf-link")) + return BPF_OBJ_LINK; return BPF_OBJ_UNKNOWN; } @@ -353,84 +461,94 @@ void print_hex_data_json(uint8_t *data, size_t len) jsonw_end_array(json_wtr); } -int build_pinned_obj_table(struct pinned_obj_table *tab, - enum bpf_obj_type type) +/* extra params for nftw cb */ +static struct hashmap *build_fn_table; +static enum bpf_obj_type build_fn_type; + +static int do_build_table_cb(const char *fpath, const struct stat *sb, + int typeflag, struct FTW *ftwbuf) { - struct bpf_prog_info pinned_info = {}; - struct pinned_obj *obj_node = NULL; + struct bpf_prog_info pinned_info; __u32 len = sizeof(pinned_info); - struct mntent *mntent = NULL; enum bpf_obj_type objtype; + int fd, err = 0; + char *path; + + if (typeflag != FTW_F) + goto out_ret; + + fd = open_obj_pinned(fpath, true); + if (fd < 0) + goto out_ret; + + objtype = get_fd_type(fd); + if (objtype != build_fn_type) + goto out_close; + + memset(&pinned_info, 0, sizeof(pinned_info)); + if (bpf_obj_get_info_by_fd(fd, &pinned_info, &len)) + goto out_close; + + path = strdup(fpath); + if (!path) { + err = -1; + goto out_close; + } + + err = hashmap__append(build_fn_table, u32_as_hash_field(pinned_info.id), path); + if (err) { + p_err("failed to append entry to hashmap for ID %u, path '%s': %s", + pinned_info.id, path, strerror(errno)); + goto out_close; + } + +out_close: + close(fd); +out_ret: + return err; +} + +int build_pinned_obj_table(struct hashmap *tab, + enum bpf_obj_type type) +{ + struct mntent *mntent = NULL; FILE *mntfile = NULL; - FTSENT *ftse = NULL; - FTS *fts = NULL; - int fd, err; + int flags = FTW_PHYS; + int nopenfd = 16; + int err = 0; mntfile = setmntent("/proc/mounts", "r"); if (!mntfile) return -1; + build_fn_table = tab; + build_fn_type = type; + while ((mntent = getmntent(mntfile))) { - char *path[] = { mntent->mnt_dir, NULL }; + char *path = mntent->mnt_dir; if (strncmp(mntent->mnt_type, "bpf", 3) != 0) continue; - - fts = fts_open(path, 0, NULL); - if (!fts) - continue; - - while ((ftse = fts_read(fts))) { - if (!(ftse->fts_info & FTS_F)) - continue; - fd = open_obj_pinned(ftse->fts_path, true); - if (fd < 0) - continue; - - objtype = get_fd_type(fd); - if (objtype != type) { - close(fd); - continue; - } - memset(&pinned_info, 0, sizeof(pinned_info)); - err = bpf_obj_get_info_by_fd(fd, &pinned_info, &len); - if (err) { - close(fd); - continue; - } - - obj_node = malloc(sizeof(*obj_node)); - if (!obj_node) { - close(fd); - fts_close(fts); - fclose(mntfile); - return -1; - } - - memset(obj_node, 0, sizeof(*obj_node)); - obj_node->id = pinned_info.id; - obj_node->path = strdup(ftse->fts_path); - hash_add(tab->table, &obj_node->hash, obj_node->id); - - close(fd); - } - fts_close(fts); + err = nftw(path, do_build_table_cb, nopenfd, flags); + if (err) + break; } fclose(mntfile); - return 0; + return err; } -void delete_pinned_obj_table(struct pinned_obj_table *tab) +void delete_pinned_obj_table(struct hashmap *map) { - struct pinned_obj *obj; - struct hlist_node *tmp; - unsigned int bkt; + struct hashmap_entry *entry; + size_t bkt; - hash_for_each_safe(tab->table, bkt, tmp, obj, hash) { - hash_del(&obj->hash); - free(obj->path); - free(obj); - } + if (!map) + return; + + hashmap__for_each_entry(map, entry, bkt) + free(entry->value); + + hashmap__free(map); } unsigned int get_page_size(void) @@ -597,3 +715,373 @@ int parse_u32_arg(int *argc, char ***argv, __u32 *val, const char *what) return 0; } + +int __printf(2, 0) +print_all_levels(__maybe_unused enum libbpf_print_level level, + const char *format, va_list args) +{ + return vfprintf(stderr, format, args); +} + +static int prog_fd_by_nametag(void *nametag, int **fds, bool tag) +{ + char prog_name[MAX_PROG_FULL_NAME]; + unsigned int id = 0; + int fd, nb_fds = 0; + void *tmp; + int err; + + while (true) { + struct bpf_prog_info info = {}; + __u32 len = sizeof(info); + + err = bpf_prog_get_next_id(id, &id); + if (err) { + if (errno != ENOENT) { + p_err("%s", strerror(errno)); + goto err_close_fds; + } + return nb_fds; + } + + fd = bpf_prog_get_fd_by_id(id); + if (fd < 0) { + p_err("can't get prog by id (%u): %s", + id, strerror(errno)); + goto err_close_fds; + } + + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err) { + p_err("can't get prog info (%u): %s", + id, strerror(errno)); + goto err_close_fd; + } + + if (tag && memcmp(nametag, info.tag, BPF_TAG_SIZE)) { + close(fd); + continue; + } + + if (!tag) { + get_prog_full_name(&info, fd, prog_name, + sizeof(prog_name)); + if (strncmp(nametag, prog_name, sizeof(prog_name))) { + close(fd); + continue; + } + } + + if (nb_fds > 0) { + tmp = realloc(*fds, (nb_fds + 1) * sizeof(int)); + if (!tmp) { + p_err("failed to realloc"); + goto err_close_fd; + } + *fds = tmp; + } + (*fds)[nb_fds++] = fd; + } + +err_close_fd: + close(fd); +err_close_fds: + while (--nb_fds >= 0) + close((*fds)[nb_fds]); + return -1; +} + +int prog_parse_fds(int *argc, char ***argv, int **fds) +{ + if (is_prefix(**argv, "id")) { + unsigned int id; + char *endptr; + + NEXT_ARGP(); + + id = strtoul(**argv, &endptr, 0); + if (*endptr) { + p_err("can't parse %s as ID", **argv); + return -1; + } + NEXT_ARGP(); + + (*fds)[0] = bpf_prog_get_fd_by_id(id); + if ((*fds)[0] < 0) { + p_err("get by id (%u): %s", id, strerror(errno)); + return -1; + } + return 1; + } else if (is_prefix(**argv, "tag")) { + unsigned char tag[BPF_TAG_SIZE]; + + NEXT_ARGP(); + + if (sscanf(**argv, BPF_TAG_FMT, tag, tag + 1, tag + 2, + tag + 3, tag + 4, tag + 5, tag + 6, tag + 7) + != BPF_TAG_SIZE) { + p_err("can't parse tag"); + return -1; + } + NEXT_ARGP(); + + return prog_fd_by_nametag(tag, fds, true); + } else if (is_prefix(**argv, "name")) { + char *name; + + NEXT_ARGP(); + + name = **argv; + if (strlen(name) > MAX_PROG_FULL_NAME - 1) { + p_err("can't parse name"); + return -1; + } + NEXT_ARGP(); + + return prog_fd_by_nametag(name, fds, false); + } else if (is_prefix(**argv, "pinned")) { + char *path; + + NEXT_ARGP(); + + path = **argv; + NEXT_ARGP(); + + (*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_PROG); + if ((*fds)[0] < 0) + return -1; + return 1; + } + + p_err("expected 'id', 'tag', 'name' or 'pinned', got: '%s'?", **argv); + return -1; +} + +int prog_parse_fd(int *argc, char ***argv) +{ + int *fds = NULL; + int nb_fds, fd; + + fds = malloc(sizeof(int)); + if (!fds) { + p_err("mem alloc failed"); + return -1; + } + nb_fds = prog_parse_fds(argc, argv, &fds); + if (nb_fds != 1) { + if (nb_fds > 1) { + p_err("several programs match this handle"); + while (nb_fds--) + close(fds[nb_fds]); + } + fd = -1; + goto exit_free; + } + + fd = fds[0]; +exit_free: + free(fds); + return fd; +} + +static int map_fd_by_name(char *name, int **fds) +{ + unsigned int id = 0; + int fd, nb_fds = 0; + void *tmp; + int err; + + while (true) { + struct bpf_map_info info = {}; + __u32 len = sizeof(info); + + err = bpf_map_get_next_id(id, &id); + if (err) { + if (errno != ENOENT) { + p_err("%s", strerror(errno)); + goto err_close_fds; + } + return nb_fds; + } + + fd = bpf_map_get_fd_by_id(id); + if (fd < 0) { + p_err("can't get map by id (%u): %s", + id, strerror(errno)); + goto err_close_fds; + } + + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err) { + p_err("can't get map info (%u): %s", + id, strerror(errno)); + goto err_close_fd; + } + + if (strncmp(name, info.name, BPF_OBJ_NAME_LEN)) { + close(fd); + continue; + } + + if (nb_fds > 0) { + tmp = realloc(*fds, (nb_fds + 1) * sizeof(int)); + if (!tmp) { + p_err("failed to realloc"); + goto err_close_fd; + } + *fds = tmp; + } + (*fds)[nb_fds++] = fd; + } + +err_close_fd: + close(fd); +err_close_fds: + while (--nb_fds >= 0) + close((*fds)[nb_fds]); + return -1; +} + +int map_parse_fds(int *argc, char ***argv, int **fds) +{ + if (is_prefix(**argv, "id")) { + unsigned int id; + char *endptr; + + NEXT_ARGP(); + + id = strtoul(**argv, &endptr, 0); + if (*endptr) { + p_err("can't parse %s as ID", **argv); + return -1; + } + NEXT_ARGP(); + + (*fds)[0] = bpf_map_get_fd_by_id(id); + if ((*fds)[0] < 0) { + p_err("get map by id (%u): %s", id, strerror(errno)); + return -1; + } + return 1; + } else if (is_prefix(**argv, "name")) { + char *name; + + NEXT_ARGP(); + + name = **argv; + if (strlen(name) > BPF_OBJ_NAME_LEN - 1) { + p_err("can't parse name"); + return -1; + } + NEXT_ARGP(); + + return map_fd_by_name(name, fds); + } else if (is_prefix(**argv, "pinned")) { + char *path; + + NEXT_ARGP(); + + path = **argv; + NEXT_ARGP(); + + (*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_MAP); + if ((*fds)[0] < 0) + return -1; + return 1; + } + + p_err("expected 'id', 'name' or 'pinned', got: '%s'?", **argv); + return -1; +} + +int map_parse_fd(int *argc, char ***argv) +{ + int *fds = NULL; + int nb_fds, fd; + + fds = malloc(sizeof(int)); + if (!fds) { + p_err("mem alloc failed"); + return -1; + } + nb_fds = map_parse_fds(argc, argv, &fds); + if (nb_fds != 1) { + if (nb_fds > 1) { + p_err("several maps match this handle"); + while (nb_fds--) + close(fds[nb_fds]); + } + fd = -1; + goto exit_free; + } + + fd = fds[0]; +exit_free: + free(fds); + return fd; +} + +int map_parse_fd_and_info(int *argc, char ***argv, void *info, __u32 *info_len) +{ + int err; + int fd; + + fd = map_parse_fd(argc, argv); + if (fd < 0) + return -1; + + err = bpf_obj_get_info_by_fd(fd, info, info_len); + if (err) { + p_err("can't get map info: %s", strerror(errno)); + close(fd); + return err; + } + + return fd; +} + +size_t hash_fn_for_key_as_id(const void *key, void *ctx) +{ + return (size_t)key; +} + +bool equal_fn_for_key_as_id(const void *k1, const void *k2, void *ctx) +{ + return k1 == k2; +} + +const char *bpf_attach_type_input_str(enum bpf_attach_type t) +{ + switch (t) { + case BPF_CGROUP_INET_INGRESS: return "ingress"; + case BPF_CGROUP_INET_EGRESS: return "egress"; + case BPF_CGROUP_INET_SOCK_CREATE: return "sock_create"; + case BPF_CGROUP_INET_SOCK_RELEASE: return "sock_release"; + case BPF_CGROUP_SOCK_OPS: return "sock_ops"; + case BPF_CGROUP_DEVICE: return "device"; + case BPF_CGROUP_INET4_BIND: return "bind4"; + case BPF_CGROUP_INET6_BIND: return "bind6"; + case BPF_CGROUP_INET4_CONNECT: return "connect4"; + case BPF_CGROUP_INET6_CONNECT: return "connect6"; + case BPF_CGROUP_INET4_POST_BIND: return "post_bind4"; + case BPF_CGROUP_INET6_POST_BIND: return "post_bind6"; + case BPF_CGROUP_INET4_GETPEERNAME: return "getpeername4"; + case BPF_CGROUP_INET6_GETPEERNAME: return "getpeername6"; + case BPF_CGROUP_INET4_GETSOCKNAME: return "getsockname4"; + case BPF_CGROUP_INET6_GETSOCKNAME: return "getsockname6"; + case BPF_CGROUP_UDP4_SENDMSG: return "sendmsg4"; + case BPF_CGROUP_UDP6_SENDMSG: return "sendmsg6"; + case BPF_CGROUP_SYSCTL: return "sysctl"; + case BPF_CGROUP_UDP4_RECVMSG: return "recvmsg4"; + case BPF_CGROUP_UDP6_RECVMSG: return "recvmsg6"; + case BPF_CGROUP_GETSOCKOPT: return "getsockopt"; + case BPF_CGROUP_SETSOCKOPT: return "setsockopt"; + case BPF_TRACE_RAW_TP: return "raw_tp"; + case BPF_TRACE_FENTRY: return "fentry"; + case BPF_TRACE_FEXIT: return "fexit"; + case BPF_MODIFY_RETURN: return "mod_ret"; + case BPF_SK_REUSEPORT_SELECT: return "sk_skb_reuseport_select"; + case BPF_SK_REUSEPORT_SELECT_OR_MIGRATE: return "sk_skb_reuseport_select_or_migrate"; + default: return libbpf_bpf_attach_type_str(t); + } +} diff --git a/tools/bpf/bpftool/feature.c b/tools/bpf/bpftool/feature.c index 941873d778d8..36cf0f1517c9 100644 --- a/tools/bpf/bpftool/feature.c +++ b/tools/bpf/bpftool/feature.c @@ -3,9 +3,13 @@ #include <ctype.h> #include <errno.h> +#include <fcntl.h> #include <string.h> #include <unistd.h> #include <net/if.h> +#ifdef USE_LIBCAP +#include <sys/capability.h> +#endif #include <sys/utsname.h> #include <sys/vfs.h> @@ -35,8 +39,18 @@ static const char * const helper_name[] = { #undef BPF_HELPER_MAKE_ENTRY +static bool full_mode; +#ifdef USE_LIBCAP +static bool run_as_unprivileged; +#endif + /* Miscellaneous utility functions */ +static bool grep(const char *buffer, const char *pattern) +{ + return !!strstr(buffer, pattern); +} + static bool check_procfs(void) { struct statfs st_fs; @@ -72,13 +86,12 @@ print_bool_feature(const char *feat_name, const char *plain_name, printf("%s is %savailable\n", plain_name, res ? "" : "NOT "); } -static void print_kernel_option(const char *name, const char *value) +static void print_kernel_option(const char *name, const char *value, + const char *define_prefix) { char *endptr; int res; - /* No support for C-style ouptut */ - if (json_output) { if (!value) { jsonw_null_field(json_wtr, name); @@ -90,6 +103,12 @@ static void print_kernel_option(const char *name, const char *value) jsonw_int_field(json_wtr, name, res); else jsonw_string_field(json_wtr, name, value); + } else if (define_prefix) { + if (value) + printf("#define %s%s %s\n", define_prefix, + name, value); + else + printf("/* %s%s is not set */\n", define_prefix, name); } else { if (value) printf("%s is set to %s\n", name, value); @@ -112,22 +131,42 @@ print_start_section(const char *json_title, const char *plain_title, } } -static void -print_end_then_start_section(const char *json_title, const char *plain_title, - const char *define_comment, - const char *define_prefix) +static void print_end_section(void) { if (json_output) jsonw_end_object(json_wtr); else printf("\n"); - - print_start_section(json_title, plain_title, define_comment, - define_prefix); } /* Probing functions */ +static int get_vendor_id(int ifindex) +{ + char ifname[IF_NAMESIZE], path[64], buf[8]; + ssize_t len; + int fd; + + if (!if_indextoname(ifindex, ifname)) + return -1; + + snprintf(path, sizeof(path), "/sys/class/net/%s/device/vendor", ifname); + + fd = open(path, O_RDONLY | O_CLOEXEC); + if (fd < 0) + return -1; + + len = read(fd, buf, sizeof(buf)); + close(fd); + if (len < 0) + return -1; + if (len >= (ssize_t)sizeof(buf)) + return -1; + buf[len] = '\0'; + + return strtol(buf, NULL, 0); +} + static int read_procfs(const char *path) { char *endptr, *line = NULL; @@ -168,7 +207,10 @@ static void probe_unprivileged_disabled(void) printf("bpf() syscall for unprivileged users is enabled\n"); break; case 1: - printf("bpf() syscall restricted to privileged users\n"); + printf("bpf() syscall restricted to privileged users (without recovery)\n"); + break; + case 2: + printf("bpf() syscall restricted to privileged users (admin can change)\n"); break; case -1: printf("Unable to retrieve required privileges for bpf() syscall\n"); @@ -313,77 +355,88 @@ static bool read_next_kernel_config_option(gzFile file, char *buf, size_t n, return false; } -static void probe_kernel_image_config(void) +static void probe_kernel_image_config(const char *define_prefix) { - static const char * const options[] = { + static const struct { + const char * const name; + bool macro_dump; + } options[] = { /* Enable BPF */ - "CONFIG_BPF", + { "CONFIG_BPF", }, /* Enable bpf() syscall */ - "CONFIG_BPF_SYSCALL", + { "CONFIG_BPF_SYSCALL", }, /* Does selected architecture support eBPF JIT compiler */ - "CONFIG_HAVE_EBPF_JIT", + { "CONFIG_HAVE_EBPF_JIT", }, /* Compile eBPF JIT compiler */ - "CONFIG_BPF_JIT", + { "CONFIG_BPF_JIT", }, /* Avoid compiling eBPF interpreter (use JIT only) */ - "CONFIG_BPF_JIT_ALWAYS_ON", + { "CONFIG_BPF_JIT_ALWAYS_ON", }, + /* Kernel BTF debug information available */ + { "CONFIG_DEBUG_INFO_BTF", }, + /* Kernel module BTF debug information available */ + { "CONFIG_DEBUG_INFO_BTF_MODULES", }, /* cgroups */ - "CONFIG_CGROUPS", + { "CONFIG_CGROUPS", }, /* BPF programs attached to cgroups */ - "CONFIG_CGROUP_BPF", + { "CONFIG_CGROUP_BPF", }, /* bpf_get_cgroup_classid() helper */ - "CONFIG_CGROUP_NET_CLASSID", + { "CONFIG_CGROUP_NET_CLASSID", }, /* bpf_skb_{,ancestor_}cgroup_id() helpers */ - "CONFIG_SOCK_CGROUP_DATA", + { "CONFIG_SOCK_CGROUP_DATA", }, /* Tracing: attach BPF to kprobes, tracepoints, etc. */ - "CONFIG_BPF_EVENTS", + { "CONFIG_BPF_EVENTS", }, /* Kprobes */ - "CONFIG_KPROBE_EVENTS", + { "CONFIG_KPROBE_EVENTS", }, /* Uprobes */ - "CONFIG_UPROBE_EVENTS", + { "CONFIG_UPROBE_EVENTS", }, /* Tracepoints */ - "CONFIG_TRACING", + { "CONFIG_TRACING", }, /* Syscall tracepoints */ - "CONFIG_FTRACE_SYSCALLS", + { "CONFIG_FTRACE_SYSCALLS", }, /* bpf_override_return() helper support for selected arch */ - "CONFIG_FUNCTION_ERROR_INJECTION", + { "CONFIG_FUNCTION_ERROR_INJECTION", }, /* bpf_override_return() helper */ - "CONFIG_BPF_KPROBE_OVERRIDE", + { "CONFIG_BPF_KPROBE_OVERRIDE", }, /* Network */ - "CONFIG_NET", + { "CONFIG_NET", }, /* AF_XDP sockets */ - "CONFIG_XDP_SOCKETS", + { "CONFIG_XDP_SOCKETS", }, /* BPF_PROG_TYPE_LWT_* and related helpers */ - "CONFIG_LWTUNNEL_BPF", + { "CONFIG_LWTUNNEL_BPF", }, /* BPF_PROG_TYPE_SCHED_ACT, TC (traffic control) actions */ - "CONFIG_NET_ACT_BPF", + { "CONFIG_NET_ACT_BPF", }, /* BPF_PROG_TYPE_SCHED_CLS, TC filters */ - "CONFIG_NET_CLS_BPF", + { "CONFIG_NET_CLS_BPF", }, /* TC clsact qdisc */ - "CONFIG_NET_CLS_ACT", + { "CONFIG_NET_CLS_ACT", }, /* Ingress filtering with TC */ - "CONFIG_NET_SCH_INGRESS", + { "CONFIG_NET_SCH_INGRESS", }, /* bpf_skb_get_xfrm_state() helper */ - "CONFIG_XFRM", + { "CONFIG_XFRM", }, /* bpf_get_route_realm() helper */ - "CONFIG_IP_ROUTE_CLASSID", + { "CONFIG_IP_ROUTE_CLASSID", }, /* BPF_PROG_TYPE_LWT_SEG6_LOCAL and related helpers */ - "CONFIG_IPV6_SEG6_BPF", + { "CONFIG_IPV6_SEG6_BPF", }, /* BPF_PROG_TYPE_LIRC_MODE2 and related helpers */ - "CONFIG_BPF_LIRC_MODE2", + { "CONFIG_BPF_LIRC_MODE2", }, /* BPF stream parser and BPF socket maps */ - "CONFIG_BPF_STREAM_PARSER", + { "CONFIG_BPF_STREAM_PARSER", }, /* xt_bpf module for passing BPF programs to netfilter */ - "CONFIG_NETFILTER_XT_MATCH_BPF", + { "CONFIG_NETFILTER_XT_MATCH_BPF", }, /* bpfilter back-end for iptables */ - "CONFIG_BPFILTER", + { "CONFIG_BPFILTER", }, /* bpftilter module with "user mode helper" */ - "CONFIG_BPFILTER_UMH", + { "CONFIG_BPFILTER_UMH", }, /* test_bpf module for BPF tests */ - "CONFIG_TEST_BPF", + { "CONFIG_TEST_BPF", }, + + /* Misc configs useful in BPF C programs */ + /* jiffies <-> sec conversion for bpf_jiffies64() helper */ + { "CONFIG_HZ", true, } }; char *values[ARRAY_SIZE(options)] = { }; struct utsname utsn; @@ -425,7 +478,8 @@ static void probe_kernel_image_config(void) while (read_next_kernel_config_option(file, buf, sizeof(buf), &value)) { for (i = 0; i < ARRAY_SIZE(options); i++) { - if (values[i] || strcmp(buf, options[i])) + if ((define_prefix && !options[i].macro_dump) || + values[i] || strcmp(buf, options[i].name)) continue; values[i] = strdup(value); @@ -437,7 +491,9 @@ end_parse: gzclose(file); for (i = 0; i < ARRAY_SIZE(options); i++) { - print_kernel_option(options[i], values[i]); + if (define_prefix && !options[i].macro_dump) + continue; + print_kernel_option(options[i].name, values[i], define_prefix); free(values[i]); } } @@ -446,7 +502,7 @@ static bool probe_bpf_syscall(const char *define_prefix) { bool res; - bpf_load_program(BPF_PROG_TYPE_UNSPEC, NULL, 0, NULL, 0, NULL, 0); + bpf_prog_load(BPF_PROG_TYPE_UNSPEC, NULL, NULL, NULL, 0, NULL); res = (errno != ENOSYS); print_bool_feature("have_bpf_syscall", @@ -457,17 +513,50 @@ static bool probe_bpf_syscall(const char *define_prefix) return res; } +static bool +probe_prog_load_ifindex(enum bpf_prog_type prog_type, + const struct bpf_insn *insns, size_t insns_cnt, + char *log_buf, size_t log_buf_sz, + __u32 ifindex) +{ + LIBBPF_OPTS(bpf_prog_load_opts, opts, + .log_buf = log_buf, + .log_size = log_buf_sz, + .log_level = log_buf ? 1 : 0, + .prog_ifindex = ifindex, + ); + int fd; + + errno = 0; + fd = bpf_prog_load(prog_type, NULL, "GPL", insns, insns_cnt, &opts); + if (fd >= 0) + close(fd); + + return fd >= 0 && errno != EINVAL && errno != EOPNOTSUPP; +} + +static bool probe_prog_type_ifindex(enum bpf_prog_type prog_type, __u32 ifindex) +{ + /* nfp returns -EINVAL on exit(0) with TC offload */ + struct bpf_insn insns[2] = { + BPF_MOV64_IMM(BPF_REG_0, 2), + BPF_EXIT_INSN() + }; + + return probe_prog_load_ifindex(prog_type, insns, ARRAY_SIZE(insns), + NULL, 0, ifindex); +} + static void -probe_prog_type(enum bpf_prog_type prog_type, bool *supported_types, - const char *define_prefix, __u32 ifindex) +probe_prog_type(enum bpf_prog_type prog_type, const char *prog_type_str, + bool *supported_types, const char *define_prefix, __u32 ifindex) { char feat_name[128], plain_desc[128], define_name[128]; const char *plain_comment = "eBPF program_type "; size_t maxlen; bool res; - if (ifindex) - /* Only test offload-able program types */ + if (ifindex) { switch (prog_type) { case BPF_PROG_TYPE_SCHED_CLS: case BPF_PROG_TYPE_XDP: @@ -476,57 +565,168 @@ probe_prog_type(enum bpf_prog_type prog_type, bool *supported_types, return; } - res = bpf_probe_prog_type(prog_type, ifindex); + res = probe_prog_type_ifindex(prog_type, ifindex); + } else { + res = libbpf_probe_bpf_prog_type(prog_type, NULL) > 0; + } + +#ifdef USE_LIBCAP + /* Probe may succeed even if program load fails, for unprivileged users + * check that we did not fail because of insufficient permissions + */ + if (run_as_unprivileged && errno == EPERM) + res = false; +#endif supported_types[prog_type] |= res; maxlen = sizeof(plain_desc) - strlen(plain_comment) - 1; - if (strlen(prog_type_name[prog_type]) > maxlen) { + if (strlen(prog_type_str) > maxlen) { p_info("program type name too long"); return; } - sprintf(feat_name, "have_%s_prog_type", prog_type_name[prog_type]); - sprintf(define_name, "%s_prog_type", prog_type_name[prog_type]); + sprintf(feat_name, "have_%s_prog_type", prog_type_str); + sprintf(define_name, "%s_prog_type", prog_type_str); uppercase(define_name, sizeof(define_name)); - sprintf(plain_desc, "%s%s", plain_comment, prog_type_name[prog_type]); + sprintf(plain_desc, "%s%s", plain_comment, prog_type_str); print_bool_feature(feat_name, plain_desc, define_name, res, define_prefix); } +static bool probe_map_type_ifindex(enum bpf_map_type map_type, __u32 ifindex) +{ + LIBBPF_OPTS(bpf_map_create_opts, opts); + int key_size, value_size, max_entries; + int fd; + + opts.map_ifindex = ifindex; + + key_size = sizeof(__u32); + value_size = sizeof(__u32); + max_entries = 1; + + fd = bpf_map_create(map_type, NULL, key_size, value_size, max_entries, + &opts); + if (fd >= 0) + close(fd); + + return fd >= 0; +} + static void -probe_map_type(enum bpf_map_type map_type, const char *define_prefix, - __u32 ifindex) +probe_map_type(enum bpf_map_type map_type, char const *map_type_str, + const char *define_prefix, __u32 ifindex) { char feat_name[128], plain_desc[128], define_name[128]; const char *plain_comment = "eBPF map_type "; size_t maxlen; bool res; - res = bpf_probe_map_type(map_type, ifindex); + if (ifindex) { + switch (map_type) { + case BPF_MAP_TYPE_HASH: + case BPF_MAP_TYPE_ARRAY: + break; + default: + return; + } + + res = probe_map_type_ifindex(map_type, ifindex); + } else { + res = libbpf_probe_bpf_map_type(map_type, NULL) > 0; + } + + /* Probe result depends on the success of map creation, no additional + * check required for unprivileged users + */ maxlen = sizeof(plain_desc) - strlen(plain_comment) - 1; - if (strlen(map_type_name[map_type]) > maxlen) { + if (strlen(map_type_str) > maxlen) { p_info("map type name too long"); return; } - sprintf(feat_name, "have_%s_map_type", map_type_name[map_type]); - sprintf(define_name, "%s_map_type", map_type_name[map_type]); + sprintf(feat_name, "have_%s_map_type", map_type_str); + sprintf(define_name, "%s_map_type", map_type_str); uppercase(define_name, sizeof(define_name)); - sprintf(plain_desc, "%s%s", plain_comment, map_type_name[map_type]); + sprintf(plain_desc, "%s%s", plain_comment, map_type_str); print_bool_feature(feat_name, plain_desc, define_name, res, define_prefix); } +static bool +probe_helper_ifindex(enum bpf_func_id id, enum bpf_prog_type prog_type, + __u32 ifindex) +{ + struct bpf_insn insns[2] = { + BPF_EMIT_CALL(id), + BPF_EXIT_INSN() + }; + char buf[4096] = {}; + bool res; + + probe_prog_load_ifindex(prog_type, insns, ARRAY_SIZE(insns), buf, + sizeof(buf), ifindex); + res = !grep(buf, "invalid func ") && !grep(buf, "unknown func "); + + switch (get_vendor_id(ifindex)) { + case 0x19ee: /* Netronome specific */ + res = res && !grep(buf, "not supported by FW") && + !grep(buf, "unsupported function id"); + break; + default: + break; + } + + return res; +} + +static bool +probe_helper_for_progtype(enum bpf_prog_type prog_type, bool supported_type, + const char *define_prefix, unsigned int id, + const char *ptype_name, __u32 ifindex) +{ + bool res = false; + + if (supported_type) { + if (ifindex) + res = probe_helper_ifindex(id, prog_type, ifindex); + else + res = libbpf_probe_bpf_helper(prog_type, id, NULL) > 0; +#ifdef USE_LIBCAP + /* Probe may succeed even if program load fails, for + * unprivileged users check that we did not fail because of + * insufficient permissions + */ + if (run_as_unprivileged && errno == EPERM) + res = false; +#endif + } + + if (json_output) { + if (res) + jsonw_string(json_wtr, helper_name[id]); + } else if (define_prefix) { + printf("#define %sBPF__PROG_TYPE_%s__HELPER_%s %s\n", + define_prefix, ptype_name, helper_name[id], + res ? "1" : "0"); + } else { + if (res) + printf("\n\t- %s", helper_name[id]); + } + + return res; +} + static void -probe_helpers_for_progtype(enum bpf_prog_type prog_type, bool supported_type, +probe_helpers_for_progtype(enum bpf_prog_type prog_type, + const char *prog_type_str, bool supported_type, const char *define_prefix, __u32 ifindex) { - const char *ptype_name = prog_type_name[prog_type]; char feat_name[128]; unsigned int id; - bool res; + bool probe_res = false; if (ifindex) /* Only test helpers for offload-able program types */ @@ -539,49 +739,424 @@ probe_helpers_for_progtype(enum bpf_prog_type prog_type, bool supported_type, } if (json_output) { - sprintf(feat_name, "%s_available_helpers", ptype_name); + sprintf(feat_name, "%s_available_helpers", prog_type_str); jsonw_name(json_wtr, feat_name); jsonw_start_array(json_wtr); } else if (!define_prefix) { printf("eBPF helpers supported for program type %s:", - ptype_name); + prog_type_str); } for (id = 1; id < ARRAY_SIZE(helper_name); id++) { - if (!supported_type) - res = false; - else - res = bpf_probe_helper(id, prog_type, ifindex); - - if (json_output) { - if (res) - jsonw_string(json_wtr, helper_name[id]); - } else if (define_prefix) { - printf("#define %sBPF__PROG_TYPE_%s__HELPER_%s %s\n", - define_prefix, ptype_name, helper_name[id], - res ? "1" : "0"); - } else { - if (res) - printf("\n\t- %s", helper_name[id]); + /* Skip helper functions which emit dmesg messages when not in + * the full mode. + */ + switch (id) { + case BPF_FUNC_trace_printk: + case BPF_FUNC_trace_vprintk: + case BPF_FUNC_probe_write_user: + if (!full_mode) + continue; + /* fallthrough */ + default: + probe_res |= probe_helper_for_progtype(prog_type, supported_type, + define_prefix, id, prog_type_str, + ifindex); } } if (json_output) jsonw_end_array(json_wtr); - else if (!define_prefix) + else if (!define_prefix) { printf("\n"); + if (!probe_res) { + if (!supported_type) + printf("\tProgram type not supported\n"); + else + printf("\tCould not determine which helpers are available\n"); + } + } + + } static void -probe_large_insn_limit(const char *define_prefix, __u32 ifindex) +probe_misc_feature(struct bpf_insn *insns, size_t len, + const char *define_prefix, __u32 ifindex, + const char *feat_name, const char *plain_name, + const char *define_name) { + LIBBPF_OPTS(bpf_prog_load_opts, opts, + .prog_ifindex = ifindex, + ); bool res; + int fd; + + errno = 0; + fd = bpf_prog_load(BPF_PROG_TYPE_SOCKET_FILTER, NULL, "GPL", + insns, len, &opts); + res = fd >= 0 || !errno; + + if (fd >= 0) + close(fd); + + print_bool_feature(feat_name, plain_name, define_name, res, + define_prefix); +} + +/* + * Probe for availability of kernel commit (5.3): + * + * c04c0d2b968a ("bpf: increase complexity limit and maximum program size") + */ +static void probe_large_insn_limit(const char *define_prefix, __u32 ifindex) +{ + struct bpf_insn insns[BPF_MAXINSNS + 1]; + int i; - res = bpf_probe_large_insn_limit(ifindex); - print_bool_feature("have_large_insn_limit", + for (i = 0; i < BPF_MAXINSNS; i++) + insns[i] = BPF_MOV64_IMM(BPF_REG_0, 1); + insns[BPF_MAXINSNS] = BPF_EXIT_INSN(); + + probe_misc_feature(insns, ARRAY_SIZE(insns), + define_prefix, ifindex, + "have_large_insn_limit", "Large program size limit", - "LARGE_INSN_LIMIT", - res, define_prefix); + "LARGE_INSN_LIMIT"); +} + +/* + * Probe for bounded loop support introduced in commit 2589726d12a1 + * ("bpf: introduce bounded loops"). + */ +static void +probe_bounded_loops(const char *define_prefix, __u32 ifindex) +{ + struct bpf_insn insns[4] = { + BPF_MOV64_IMM(BPF_REG_0, 10), + BPF_ALU64_IMM(BPF_SUB, BPF_REG_0, 1), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, -2), + BPF_EXIT_INSN() + }; + + probe_misc_feature(insns, ARRAY_SIZE(insns), + define_prefix, ifindex, + "have_bounded_loops", + "Bounded loop support", + "BOUNDED_LOOPS"); +} + +/* + * Probe for the v2 instruction set extension introduced in commit 92b31a9af73b + * ("bpf: add BPF_J{LT,LE,SLT,SLE} instructions"). + */ +static void +probe_v2_isa_extension(const char *define_prefix, __u32 ifindex) +{ + struct bpf_insn insns[4] = { + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_JMP_IMM(BPF_JLT, BPF_REG_0, 0, 1), + BPF_MOV64_IMM(BPF_REG_0, 1), + BPF_EXIT_INSN() + }; + + probe_misc_feature(insns, ARRAY_SIZE(insns), + define_prefix, ifindex, + "have_v2_isa_extension", + "ISA extension v2", + "V2_ISA_EXTENSION"); +} + +/* + * Probe for the v3 instruction set extension introduced in commit 092ed0968bb6 + * ("bpf: verifier support JMP32"). + */ +static void +probe_v3_isa_extension(const char *define_prefix, __u32 ifindex) +{ + struct bpf_insn insns[4] = { + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_JMP32_IMM(BPF_JLT, BPF_REG_0, 0, 1), + BPF_MOV64_IMM(BPF_REG_0, 1), + BPF_EXIT_INSN() + }; + + probe_misc_feature(insns, ARRAY_SIZE(insns), + define_prefix, ifindex, + "have_v3_isa_extension", + "ISA extension v3", + "V3_ISA_EXTENSION"); +} + +static void +section_system_config(enum probe_component target, const char *define_prefix) +{ + switch (target) { + case COMPONENT_KERNEL: + case COMPONENT_UNSPEC: + print_start_section("system_config", + "Scanning system configuration...", + "/*** Misc kernel config items ***/", + define_prefix); + if (!define_prefix) { + if (check_procfs()) { + probe_unprivileged_disabled(); + probe_jit_enable(); + probe_jit_harden(); + probe_jit_kallsyms(); + probe_jit_limit(); + } else { + p_info("/* procfs not mounted, skipping related probes */"); + } + } + probe_kernel_image_config(define_prefix); + print_end_section(); + break; + default: + break; + } +} + +static bool section_syscall_config(const char *define_prefix) +{ + bool res; + + print_start_section("syscall_config", + "Scanning system call availability...", + "/*** System call availability ***/", + define_prefix); + res = probe_bpf_syscall(define_prefix); + print_end_section(); + + return res; +} + +static void +section_program_types(bool *supported_types, const char *define_prefix, + __u32 ifindex) +{ + unsigned int prog_type = BPF_PROG_TYPE_UNSPEC; + const char *prog_type_str; + + print_start_section("program_types", + "Scanning eBPF program types...", + "/*** eBPF program types ***/", + define_prefix); + + while (true) { + prog_type++; + prog_type_str = libbpf_bpf_prog_type_str(prog_type); + /* libbpf will return NULL for variants unknown to it. */ + if (!prog_type_str) + break; + + probe_prog_type(prog_type, prog_type_str, supported_types, define_prefix, + ifindex); + } + + print_end_section(); +} + +static void section_map_types(const char *define_prefix, __u32 ifindex) +{ + unsigned int map_type = BPF_MAP_TYPE_UNSPEC; + const char *map_type_str; + + print_start_section("map_types", + "Scanning eBPF map types...", + "/*** eBPF map types ***/", + define_prefix); + + while (true) { + map_type++; + map_type_str = libbpf_bpf_map_type_str(map_type); + /* libbpf will return NULL for variants unknown to it. */ + if (!map_type_str) + break; + + probe_map_type(map_type, map_type_str, define_prefix, ifindex); + } + + print_end_section(); +} + +static void +section_helpers(bool *supported_types, const char *define_prefix, __u32 ifindex) +{ + unsigned int prog_type = BPF_PROG_TYPE_UNSPEC; + const char *prog_type_str; + + print_start_section("helpers", + "Scanning eBPF helper functions...", + "/*** eBPF helper functions ***/", + define_prefix); + + if (define_prefix) + printf("/*\n" + " * Use %sHAVE_PROG_TYPE_HELPER(prog_type_name, helper_name)\n" + " * to determine if <helper_name> is available for <prog_type_name>,\n" + " * e.g.\n" + " * #if %sHAVE_PROG_TYPE_HELPER(xdp, bpf_redirect)\n" + " * // do stuff with this helper\n" + " * #elif\n" + " * // use a workaround\n" + " * #endif\n" + " */\n" + "#define %sHAVE_PROG_TYPE_HELPER(prog_type, helper) \\\n" + " %sBPF__PROG_TYPE_ ## prog_type ## __HELPER_ ## helper\n", + define_prefix, define_prefix, define_prefix, + define_prefix); + while (true) { + prog_type++; + prog_type_str = libbpf_bpf_prog_type_str(prog_type); + /* libbpf will return NULL for variants unknown to it. */ + if (!prog_type_str) + break; + + probe_helpers_for_progtype(prog_type, prog_type_str, + supported_types[prog_type], + define_prefix, + ifindex); + } + + print_end_section(); +} + +static void section_misc(const char *define_prefix, __u32 ifindex) +{ + print_start_section("misc", + "Scanning miscellaneous eBPF features...", + "/*** eBPF misc features ***/", + define_prefix); + probe_large_insn_limit(define_prefix, ifindex); + probe_bounded_loops(define_prefix, ifindex); + probe_v2_isa_extension(define_prefix, ifindex); + probe_v3_isa_extension(define_prefix, ifindex); + print_end_section(); +} + +#ifdef USE_LIBCAP +#define capability(c) { c, false, #c } +#define capability_msg(a, i) a[i].set ? "" : a[i].name, a[i].set ? "" : ", " +#endif + +static int handle_perms(void) +{ +#ifdef USE_LIBCAP + struct { + cap_value_t cap; + bool set; + char name[14]; /* strlen("CAP_SYS_ADMIN") */ + } bpf_caps[] = { + capability(CAP_SYS_ADMIN), +#ifdef CAP_BPF + capability(CAP_BPF), + capability(CAP_NET_ADMIN), + capability(CAP_PERFMON), +#endif + }; + cap_value_t cap_list[ARRAY_SIZE(bpf_caps)]; + unsigned int i, nb_bpf_caps = 0; + bool cap_sys_admin_only = true; + cap_flag_value_t val; + int res = -1; + cap_t caps; + + caps = cap_get_proc(); + if (!caps) { + p_err("failed to get capabilities for process: %s", + strerror(errno)); + return -1; + } + +#ifdef CAP_BPF + if (CAP_IS_SUPPORTED(CAP_BPF)) + cap_sys_admin_only = false; +#endif + + for (i = 0; i < ARRAY_SIZE(bpf_caps); i++) { + const char *cap_name = bpf_caps[i].name; + cap_value_t cap = bpf_caps[i].cap; + + if (cap_get_flag(caps, cap, CAP_EFFECTIVE, &val)) { + p_err("bug: failed to retrieve %s status: %s", cap_name, + strerror(errno)); + goto exit_free; + } + + if (val == CAP_SET) { + bpf_caps[i].set = true; + cap_list[nb_bpf_caps++] = cap; + } + + if (cap_sys_admin_only) + /* System does not know about CAP_BPF, meaning that + * CAP_SYS_ADMIN is the only capability required. We + * just checked it, break. + */ + break; + } + + if ((run_as_unprivileged && !nb_bpf_caps) || + (!run_as_unprivileged && nb_bpf_caps == ARRAY_SIZE(bpf_caps)) || + (!run_as_unprivileged && cap_sys_admin_only && nb_bpf_caps)) { + /* We are all good, exit now */ + res = 0; + goto exit_free; + } + + if (!run_as_unprivileged) { + if (cap_sys_admin_only) + p_err("missing %s, required for full feature probing; run as root or use 'unprivileged'", + bpf_caps[0].name); + else + p_err("missing %s%s%s%s%s%s%s%srequired for full feature probing; run as root or use 'unprivileged'", + capability_msg(bpf_caps, 0), +#ifdef CAP_BPF + capability_msg(bpf_caps, 1), + capability_msg(bpf_caps, 2), + capability_msg(bpf_caps, 3) +#else + "", "", "", "", "", "" +#endif /* CAP_BPF */ + ); + goto exit_free; + } + + /* if (run_as_unprivileged && nb_bpf_caps > 0), drop capabilities. */ + if (cap_set_flag(caps, CAP_EFFECTIVE, nb_bpf_caps, cap_list, + CAP_CLEAR)) { + p_err("bug: failed to clear capabilities: %s", strerror(errno)); + goto exit_free; + } + + if (cap_set_proc(caps)) { + p_err("failed to drop capabilities: %s", strerror(errno)); + goto exit_free; + } + + res = 0; + +exit_free: + if (cap_free(caps) && !res) { + p_err("failed to clear storage object for capabilities: %s", + strerror(errno)); + res = -1; + } + + return res; +#else + /* Detection assumes user has specific privileges. + * We do not use libcap so let's approximate, and restrict usage to + * root user only. + */ + if (geteuid()) { + p_err("full feature probing requires root privileges"); + return -1; + } + + return 0; +#endif /* USE_LIBCAP */ } static int do_probe(int argc, char **argv) @@ -590,17 +1165,8 @@ static int do_probe(int argc, char **argv) const char *define_prefix = NULL; bool supported_types[128] = {}; __u32 ifindex = 0; - unsigned int i; char *ifname; - /* Detection assumes user has sufficient privileges (CAP_SYS_ADMIN). - * Let's approximate, and restrict usage to root user only. - */ - if (geteuid()) { - p_err("please run this command as root user"); - return -1; - } - set_max_rlimit(); while (argc) { @@ -629,6 +1195,9 @@ static int do_probe(int argc, char **argv) strerror(errno)); return -1; } + } else if (is_prefix(*argv, "full")) { + full_mode = true; + NEXT_ARG(); } else if (is_prefix(*argv, "macros") && !define_prefix) { define_prefix = ""; NEXT_ARG(); @@ -646,6 +1215,14 @@ static int do_probe(int argc, char **argv) if (!REQ_ARGS(1)) return -1; define_prefix = GET_ARG(); + } else if (is_prefix(*argv, "unprivileged")) { +#ifdef USE_LIBCAP + run_as_unprivileged = true; + NEXT_ARG(); +#else + p_err("unprivileged run not supported, recompile bpftool with libcap"); + return -1; +#endif } else { p_err("expected no more arguments, 'kernel', 'dev', 'macros' or 'prefix', got: '%s'?", *argv); @@ -653,103 +1230,83 @@ static int do_probe(int argc, char **argv) } } + /* Full feature detection requires specific privileges. + * Let's approximate, and warn if user is not root. + */ + if (handle_perms()) + return -1; + if (json_output) { define_prefix = NULL; jsonw_start_object(json_wtr); } - switch (target) { - case COMPONENT_KERNEL: - case COMPONENT_UNSPEC: - if (define_prefix) - break; - - print_start_section("system_config", - "Scanning system configuration...", - NULL, /* define_comment never used here */ - NULL); /* define_prefix always NULL here */ - if (check_procfs()) { - probe_unprivileged_disabled(); - probe_jit_enable(); - probe_jit_harden(); - probe_jit_kallsyms(); - probe_jit_limit(); - } else { - p_info("/* procfs not mounted, skipping related probes */"); - } - probe_kernel_image_config(); - if (json_output) - jsonw_end_object(json_wtr); - else - printf("\n"); - break; - default: - break; - } - - print_start_section("syscall_config", - "Scanning system call availability...", - "/*** System call availability ***/", - define_prefix); - - if (!probe_bpf_syscall(define_prefix)) + section_system_config(target, define_prefix); + if (!section_syscall_config(define_prefix)) /* bpf() syscall unavailable, don't probe other BPF features */ goto exit_close_json; + section_program_types(supported_types, define_prefix, ifindex); + section_map_types(define_prefix, ifindex); + section_helpers(supported_types, define_prefix, ifindex); + section_misc(define_prefix, ifindex); - print_end_then_start_section("program_types", - "Scanning eBPF program types...", - "/*** eBPF program types ***/", - define_prefix); +exit_close_json: + if (json_output) + /* End root object */ + jsonw_end_object(json_wtr); - for (i = BPF_PROG_TYPE_UNSPEC + 1; i < ARRAY_SIZE(prog_type_name); i++) - probe_prog_type(i, supported_types, define_prefix, ifindex); + return 0; +} - print_end_then_start_section("map_types", - "Scanning eBPF map types...", - "/*** eBPF map types ***/", - define_prefix); +static const char *get_helper_name(unsigned int id) +{ + if (id >= ARRAY_SIZE(helper_name)) + return NULL; - for (i = BPF_MAP_TYPE_UNSPEC + 1; i < map_type_name_size; i++) - probe_map_type(i, define_prefix, ifindex); + return helper_name[id]; +} - print_end_then_start_section("helpers", - "Scanning eBPF helper functions...", - "/*** eBPF helper functions ***/", - define_prefix); +static int do_list_builtins(int argc, char **argv) +{ + const char *(*get_name)(unsigned int id); + unsigned int id = 0; + + if (argc < 1) + usage(); + + if (is_prefix(*argv, "prog_types")) { + get_name = (const char *(*)(unsigned int))libbpf_bpf_prog_type_str; + } else if (is_prefix(*argv, "map_types")) { + get_name = (const char *(*)(unsigned int))libbpf_bpf_map_type_str; + } else if (is_prefix(*argv, "attach_types")) { + get_name = (const char *(*)(unsigned int))libbpf_bpf_attach_type_str; + } else if (is_prefix(*argv, "link_types")) { + get_name = (const char *(*)(unsigned int))libbpf_bpf_link_type_str; + } else if (is_prefix(*argv, "helpers")) { + get_name = get_helper_name; + } else { + p_err("expected 'prog_types', 'map_types', 'attach_types', 'link_types' or 'helpers', got: %s", *argv); + return -1; + } - if (define_prefix) - printf("/*\n" - " * Use %sHAVE_PROG_TYPE_HELPER(prog_type_name, helper_name)\n" - " * to determine if <helper_name> is available for <prog_type_name>,\n" - " * e.g.\n" - " * #if %sHAVE_PROG_TYPE_HELPER(xdp, bpf_redirect)\n" - " * // do stuff with this helper\n" - " * #elif\n" - " * // use a workaround\n" - " * #endif\n" - " */\n" - "#define %sHAVE_PROG_TYPE_HELPER(prog_type, helper) \\\n" - " %sBPF__PROG_TYPE_ ## prog_type ## __HELPER_ ## helper\n", - define_prefix, define_prefix, define_prefix, - define_prefix); - for (i = BPF_PROG_TYPE_UNSPEC + 1; i < ARRAY_SIZE(prog_type_name); i++) - probe_helpers_for_progtype(i, supported_types[i], - define_prefix, ifindex); - - print_end_then_start_section("misc", - "Scanning miscellaneous eBPF features...", - "/*** eBPF misc features ***/", - define_prefix); - probe_large_insn_limit(define_prefix, ifindex); + if (json_output) + jsonw_start_array(json_wtr); /* root array */ -exit_close_json: - if (json_output) { - /* End current "section" of probes */ - jsonw_end_object(json_wtr); - /* End root object */ - jsonw_end_object(json_wtr); + while (true) { + const char *name; + + name = get_name(id++); + if (!name) + break; + if (json_output) + jsonw_string(json_wtr, name); + else + printf("%s\n", name); } + if (json_output) + jsonw_end_array(json_wtr); /* root array */ + return 0; } @@ -761,19 +1318,23 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s %s probe [COMPONENT] [macros [prefix PREFIX]]\n" - " %s %s help\n" + "Usage: %1$s %2$s probe [COMPONENT] [full] [unprivileged] [macros [prefix PREFIX]]\n" + " %1$s %2$s list_builtins GROUP\n" + " %1$s %2$s help\n" "\n" " COMPONENT := { kernel | dev NAME }\n" + " GROUP := { prog_types | map_types | attach_types | link_types | helpers }\n" + " " HELP_SPEC_OPTIONS " }\n" "", - bin_name, argv[-2], bin_name, argv[-2]); + bin_name, argv[-2]); return 0; } static const struct cmd cmds[] = { - { "probe", do_probe }, - { "help", do_help }, + { "probe", do_probe }, + { "list_builtins", do_list_builtins }, + { "help", do_help }, { 0 } }; diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c index f8113b3646f5..cf8b4e525c88 100644 --- a/tools/bpf/bpftool/gen.c +++ b/tools/bpf/bpftool/gen.c @@ -14,17 +14,15 @@ #include <unistd.h> #include <bpf/bpf.h> #include <bpf/libbpf.h> +#include <bpf/libbpf_internal.h> #include <sys/types.h> #include <sys/stat.h> #include <sys/mman.h> -#include <unistd.h> #include <bpf/btf.h> -#include "bpf/libbpf_internal.h" #include "json_writer.h" #include "main.h" - #define MAX_OBJ_NAME_LEN 64 static void sanitize_identifier(char *name) @@ -36,6 +34,11 @@ static void sanitize_identifier(char *name) name[i] = '_'; } +static bool str_has_prefix(const char *str, const char *prefix) +{ + return strncmp(str, prefix, strlen(prefix)) == 0; +} + static bool str_has_suffix(const char *str, const char *suffix) { size_t i, n1 = strlen(str), n2 = strlen(suffix); @@ -61,35 +64,59 @@ static void get_obj_name(char *name, const char *file) sanitize_identifier(name); } -static void get_header_guard(char *guard, const char *obj_name) +static void get_header_guard(char *guard, const char *obj_name, const char *suffix) { int i; - sprintf(guard, "__%s_SKEL_H__", obj_name); + sprintf(guard, "__%s_%s__", obj_name, suffix); for (i = 0; guard[i]; i++) guard[i] = toupper(guard[i]); } -static const char *get_map_ident(const struct bpf_map *map) +static bool get_map_ident(const struct bpf_map *map, char *buf, size_t buf_sz) { + static const char *sfxs[] = { ".data", ".rodata", ".bss", ".kconfig" }; const char *name = bpf_map__name(map); + int i, n; - if (!bpf_map__is_internal(map)) - return name; - - if (str_has_suffix(name, ".data")) - return "data"; - else if (str_has_suffix(name, ".rodata")) - return "rodata"; - else if (str_has_suffix(name, ".bss")) - return "bss"; - else if (str_has_suffix(name, ".kconfig")) - return "kconfig"; - else - return NULL; + if (!bpf_map__is_internal(map)) { + snprintf(buf, buf_sz, "%s", name); + return true; + } + + for (i = 0, n = ARRAY_SIZE(sfxs); i < n; i++) { + const char *sfx = sfxs[i], *p; + + p = strstr(name, sfx); + if (p) { + snprintf(buf, buf_sz, "%s", p + 1); + sanitize_identifier(buf); + return true; + } + } + + return false; +} + +static bool get_datasec_ident(const char *sec_name, char *buf, size_t buf_sz) +{ + static const char *pfxs[] = { ".data", ".rodata", ".bss", ".kconfig" }; + int i, n; + + for (i = 0, n = ARRAY_SIZE(pfxs); i < n; i++) { + const char *pfx = pfxs[i]; + + if (str_has_prefix(sec_name, pfx)) { + snprintf(buf, buf_sz, "%s", sec_name + 1); + sanitize_identifier(buf); + return true; + } + } + + return false; } -static void codegen_btf_dump_printf(void *ct, const char *fmt, va_list args) +static void codegen_btf_dump_printf(void *ctx, const char *fmt, va_list args) { vprintf(fmt, args); } @@ -103,20 +130,15 @@ static int codegen_datasec_def(struct bpf_object *obj, const char *sec_name = btf__name_by_offset(btf, sec->name_off); const struct btf_var_secinfo *sec_var = btf_var_secinfos(sec); int i, err, off = 0, pad_cnt = 0, vlen = btf_vlen(sec); - const char *sec_ident; - char var_ident[256]; - - if (strcmp(sec_name, ".data") == 0) - sec_ident = "data"; - else if (strcmp(sec_name, ".bss") == 0) - sec_ident = "bss"; - else if (strcmp(sec_name, ".rodata") == 0) - sec_ident = "rodata"; - else if (strcmp(sec_name, ".kconfig") == 0) - sec_ident = "kconfig"; - else + char var_ident[256], sec_ident[256]; + bool strip_mods = false; + + if (!get_datasec_ident(sec_name, sec_ident, sizeof(sec_ident))) return 0; + if (strcmp(sec_name, ".kconfig") != 0) + strip_mods = true; + printf(" struct %s__%s {\n", obj_name, sec_ident); for (i = 0; i < vlen; i++, sec_var++) { const struct btf_type *var = btf__type_by_id(btf, sec_var->type); @@ -124,16 +146,14 @@ static int codegen_datasec_def(struct bpf_object *obj, DECLARE_LIBBPF_OPTS(btf_dump_emit_type_decl_opts, opts, .field_name = var_ident, .indent_level = 2, + .strip_mods = strip_mods, ); int need_off = sec_var->offset, align_off, align; __u32 var_type_id = var->type; - const struct btf_type *t; - t = btf__type_by_id(btf, var_type_id); - while (btf_is_mod(t)) { - var_type_id = t->type; - t = btf__type_by_id(btf, var_type_id); - } + /* static variables are not exposed through BPF skeleton */ + if (btf_var(var)->linkage == BTF_VAR_STATIC) + continue; if (off > need_off) { p_err("Something is wrong for %s's variable #%d: need offset %d, already at %d.\n", @@ -147,6 +167,20 @@ static int codegen_datasec_def(struct bpf_object *obj, var_name, align); return -EINVAL; } + /* Assume 32-bit architectures when generating data section + * struct memory layout. Given bpftool can't know which target + * host architecture it's emitting skeleton for, we need to be + * conservative and assume 32-bit one to ensure enough padding + * bytes are generated for pointer and long types. This will + * still work correctly for 64-bit architectures, because in + * the worst case we'll generate unnecessary padding field, + * which on 64-bit architectures is not strictly necessary and + * would be handled by natural 8-byte alignment. But it still + * will be a correct memory layout, based on recorded offsets + * in BTF. + */ + if (align > 4) + align = 4; align_off = (off + align - 1) / align * align; if (align_off != need_off) { @@ -175,33 +209,174 @@ static int codegen_datasec_def(struct bpf_object *obj, return 0; } +static const struct btf_type *find_type_for_map(struct btf *btf, const char *map_ident) +{ + int n = btf__type_cnt(btf), i; + char sec_ident[256]; + + for (i = 1; i < n; i++) { + const struct btf_type *t = btf__type_by_id(btf, i); + const char *name; + + if (!btf_is_datasec(t)) + continue; + + name = btf__str_by_offset(btf, t->name_off); + if (!get_datasec_ident(name, sec_ident, sizeof(sec_ident))) + continue; + + if (strcmp(sec_ident, map_ident) == 0) + return t; + } + return NULL; +} + +static bool is_internal_mmapable_map(const struct bpf_map *map, char *buf, size_t sz) +{ + if (!bpf_map__is_internal(map) || !(bpf_map__map_flags(map) & BPF_F_MMAPABLE)) + return false; + + if (!get_map_ident(map, buf, sz)) + return false; + + return true; +} + static int codegen_datasecs(struct bpf_object *obj, const char *obj_name) { struct btf *btf = bpf_object__btf(obj); - int n = btf__get_nr_types(btf); struct btf_dump *d; - int i, err = 0; + struct bpf_map *map; + const struct btf_type *sec; + char map_ident[256]; + int err = 0; - d = btf_dump__new(btf, NULL, NULL, codegen_btf_dump_printf); - if (IS_ERR(d)) - return PTR_ERR(d); + d = btf_dump__new(btf, codegen_btf_dump_printf, NULL, NULL); + err = libbpf_get_error(d); + if (err) + return err; - for (i = 1; i <= n; i++) { - const struct btf_type *t = btf__type_by_id(btf, i); + bpf_object__for_each_map(map, obj) { + /* only generate definitions for memory-mapped internal maps */ + if (!is_internal_mmapable_map(map, map_ident, sizeof(map_ident))) + continue; - if (!btf_is_datasec(t)) + sec = find_type_for_map(btf, map_ident); + + /* In some cases (e.g., sections like .rodata.cst16 containing + * compiler allocated string constants only) there will be + * special internal maps with no corresponding DATASEC BTF + * type. In such case, generate empty structs for each such + * map. It will still be memory-mapped and its contents + * accessible from user-space through BPF skeleton. + */ + if (!sec) { + printf(" struct %s__%s {\n", obj_name, map_ident); + printf(" } *%s;\n", map_ident); + } else { + err = codegen_datasec_def(obj, btf, d, sec, obj_name); + if (err) + goto out; + } + } + + +out: + btf_dump__free(d); + return err; +} + +static bool btf_is_ptr_to_func_proto(const struct btf *btf, + const struct btf_type *v) +{ + return btf_is_ptr(v) && btf_is_func_proto(btf__type_by_id(btf, v->type)); +} + +static int codegen_subskel_datasecs(struct bpf_object *obj, const char *obj_name) +{ + struct btf *btf = bpf_object__btf(obj); + struct btf_dump *d; + struct bpf_map *map; + const struct btf_type *sec, *var; + const struct btf_var_secinfo *sec_var; + int i, err = 0, vlen; + char map_ident[256], sec_ident[256]; + bool strip_mods = false, needs_typeof = false; + const char *sec_name, *var_name; + __u32 var_type_id; + + d = btf_dump__new(btf, codegen_btf_dump_printf, NULL, NULL); + if (!d) + return -errno; + + bpf_object__for_each_map(map, obj) { + /* only generate definitions for memory-mapped internal maps */ + if (!is_internal_mmapable_map(map, map_ident, sizeof(map_ident))) continue; - err = codegen_datasec_def(obj, btf, d, t, obj_name); - if (err) - goto out; + sec = find_type_for_map(btf, map_ident); + if (!sec) + continue; + + sec_name = btf__name_by_offset(btf, sec->name_off); + if (!get_datasec_ident(sec_name, sec_ident, sizeof(sec_ident))) + continue; + + strip_mods = strcmp(sec_name, ".kconfig") != 0; + printf(" struct %s__%s {\n", obj_name, sec_ident); + + sec_var = btf_var_secinfos(sec); + vlen = btf_vlen(sec); + for (i = 0; i < vlen; i++, sec_var++) { + DECLARE_LIBBPF_OPTS(btf_dump_emit_type_decl_opts, opts, + .indent_level = 2, + .strip_mods = strip_mods, + /* we'll print the name separately */ + .field_name = "", + ); + + var = btf__type_by_id(btf, sec_var->type); + var_name = btf__name_by_offset(btf, var->name_off); + var_type_id = var->type; + + /* static variables are not exposed through BPF skeleton */ + if (btf_var(var)->linkage == BTF_VAR_STATIC) + continue; + + /* The datasec member has KIND_VAR but we want the + * underlying type of the variable (e.g. KIND_INT). + */ + var = skip_mods_and_typedefs(btf, var->type, NULL); + + printf("\t\t"); + /* Func and array members require special handling. + * Instead of producing `typename *var`, they produce + * `typeof(typename) *var`. This allows us to keep a + * similar syntax where the identifier is just prefixed + * by *, allowing us to ignore C declaration minutiae. + */ + needs_typeof = btf_is_array(var) || btf_is_ptr_to_func_proto(btf, var); + if (needs_typeof) + printf("typeof("); + + err = btf_dump__emit_type_decl(d, var_type_id, &opts); + if (err) + goto out; + + if (needs_typeof) + printf(")"); + + printf(" *%s;\n", var_name); + } + printf(" } %s;\n", sec_ident); } + out: btf_dump__free(d); return err; } -static int codegen(const char *template, ...) +static void codegen(const char *template, ...) { const char *src, *end; int skip_tabs = 0, n; @@ -212,7 +387,7 @@ static int codegen(const char *template, ...) n = strlen(template); s = malloc(n + 1); if (!s) - return -ENOMEM; + exit(-1); src = template; dst = s; @@ -223,9 +398,10 @@ static int codegen(const char *template, ...) } else if (c == '\n') { break; } else { - p_err("unrecognized character at pos %td in template '%s'", - src - template - 1, template); - return -EINVAL; + p_err("unrecognized character at pos %td in template '%s': '%c'", + src - template - 1, template, c); + free(s); + exit(-1); } } @@ -235,7 +411,8 @@ static int codegen(const char *template, ...) if (*src != '\t') { p_err("not enough tabs at pos %td in template '%s'", src - template - 1, template); - return -EINVAL; + free(s); + exit(-1); } } /* trim trailing whitespace */ @@ -256,19 +433,488 @@ static int codegen(const char *template, ...) va_end(args); free(s); - return n; +} + +static void print_hex(const char *data, int data_sz) +{ + int i, len; + + for (i = 0, len = 0; i < data_sz; i++) { + int w = data[i] ? 4 : 2; + + len += w; + if (len > 78) { + printf("\\\n"); + len = w; + } + if (!data[i]) + printf("\\0"); + else + printf("\\x%02x", (unsigned char)data[i]); + } +} + +static size_t bpf_map_mmap_sz(const struct bpf_map *map) +{ + long page_sz = sysconf(_SC_PAGE_SIZE); + size_t map_sz; + + map_sz = (size_t)roundup(bpf_map__value_size(map), 8) * bpf_map__max_entries(map); + map_sz = roundup(map_sz, page_sz); + return map_sz; +} + +/* Emit type size asserts for all top-level fields in memory-mapped internal maps. */ +static void codegen_asserts(struct bpf_object *obj, const char *obj_name) +{ + struct btf *btf = bpf_object__btf(obj); + struct bpf_map *map; + struct btf_var_secinfo *sec_var; + int i, vlen; + const struct btf_type *sec; + char map_ident[256], var_ident[256]; + + if (!btf) + return; + + codegen("\ + \n\ + __attribute__((unused)) static void \n\ + %1$s__assert(struct %1$s *s __attribute__((unused))) \n\ + { \n\ + #ifdef __cplusplus \n\ + #define _Static_assert static_assert \n\ + #endif \n\ + ", obj_name); + + bpf_object__for_each_map(map, obj) { + if (!is_internal_mmapable_map(map, map_ident, sizeof(map_ident))) + continue; + + sec = find_type_for_map(btf, map_ident); + if (!sec) { + /* best effort, couldn't find the type for this map */ + continue; + } + + sec_var = btf_var_secinfos(sec); + vlen = btf_vlen(sec); + + for (i = 0; i < vlen; i++, sec_var++) { + const struct btf_type *var = btf__type_by_id(btf, sec_var->type); + const char *var_name = btf__name_by_offset(btf, var->name_off); + long var_size; + + /* static variables are not exposed through BPF skeleton */ + if (btf_var(var)->linkage == BTF_VAR_STATIC) + continue; + + var_size = btf__resolve_size(btf, var->type); + if (var_size < 0) + continue; + + var_ident[0] = '\0'; + strncat(var_ident, var_name, sizeof(var_ident) - 1); + sanitize_identifier(var_ident); + + printf("\t_Static_assert(sizeof(s->%s->%s) == %ld, \"unexpected size of '%s'\");\n", + map_ident, var_ident, var_size, var_ident); + } + } + codegen("\ + \n\ + #ifdef __cplusplus \n\ + #undef _Static_assert \n\ + #endif \n\ + } \n\ + "); +} + +static void codegen_attach_detach(struct bpf_object *obj, const char *obj_name) +{ + struct bpf_program *prog; + + bpf_object__for_each_program(prog, obj) { + const char *tp_name; + + codegen("\ + \n\ + \n\ + static inline int \n\ + %1$s__%2$s__attach(struct %1$s *skel) \n\ + { \n\ + int prog_fd = skel->progs.%2$s.prog_fd; \n\ + ", obj_name, bpf_program__name(prog)); + + switch (bpf_program__type(prog)) { + case BPF_PROG_TYPE_RAW_TRACEPOINT: + tp_name = strchr(bpf_program__section_name(prog), '/') + 1; + printf("\tint fd = skel_raw_tracepoint_open(\"%s\", prog_fd);\n", tp_name); + break; + case BPF_PROG_TYPE_TRACING: + case BPF_PROG_TYPE_LSM: + if (bpf_program__expected_attach_type(prog) == BPF_TRACE_ITER) + printf("\tint fd = skel_link_create(prog_fd, 0, BPF_TRACE_ITER);\n"); + else + printf("\tint fd = skel_raw_tracepoint_open(NULL, prog_fd);\n"); + break; + default: + printf("\tint fd = ((void)prog_fd, 0); /* auto-attach not supported */\n"); + break; + } + codegen("\ + \n\ + \n\ + if (fd > 0) \n\ + skel->links.%1$s_fd = fd; \n\ + return fd; \n\ + } \n\ + ", bpf_program__name(prog)); + } + + codegen("\ + \n\ + \n\ + static inline int \n\ + %1$s__attach(struct %1$s *skel) \n\ + { \n\ + int ret = 0; \n\ + \n\ + ", obj_name); + + bpf_object__for_each_program(prog, obj) { + codegen("\ + \n\ + ret = ret < 0 ? ret : %1$s__%2$s__attach(skel); \n\ + ", obj_name, bpf_program__name(prog)); + } + + codegen("\ + \n\ + return ret < 0 ? ret : 0; \n\ + } \n\ + \n\ + static inline void \n\ + %1$s__detach(struct %1$s *skel) \n\ + { \n\ + ", obj_name); + + bpf_object__for_each_program(prog, obj) { + codegen("\ + \n\ + skel_closenz(skel->links.%1$s_fd); \n\ + ", bpf_program__name(prog)); + } + + codegen("\ + \n\ + } \n\ + "); +} + +static void codegen_destroy(struct bpf_object *obj, const char *obj_name) +{ + struct bpf_program *prog; + struct bpf_map *map; + char ident[256]; + + codegen("\ + \n\ + static void \n\ + %1$s__destroy(struct %1$s *skel) \n\ + { \n\ + if (!skel) \n\ + return; \n\ + %1$s__detach(skel); \n\ + ", + obj_name); + + bpf_object__for_each_program(prog, obj) { + codegen("\ + \n\ + skel_closenz(skel->progs.%1$s.prog_fd); \n\ + ", bpf_program__name(prog)); + } + + bpf_object__for_each_map(map, obj) { + if (!get_map_ident(map, ident, sizeof(ident))) + continue; + if (bpf_map__is_internal(map) && + (bpf_map__map_flags(map) & BPF_F_MMAPABLE)) + printf("\tskel_free_map_data(skel->%1$s, skel->maps.%1$s.initial_value, %2$zd);\n", + ident, bpf_map_mmap_sz(map)); + codegen("\ + \n\ + skel_closenz(skel->maps.%1$s.map_fd); \n\ + ", ident); + } + codegen("\ + \n\ + skel_free(skel); \n\ + } \n\ + ", + obj_name); +} + +static int gen_trace(struct bpf_object *obj, const char *obj_name, const char *header_guard) +{ + DECLARE_LIBBPF_OPTS(gen_loader_opts, opts); + struct bpf_map *map; + char ident[256]; + int err = 0; + + err = bpf_object__gen_loader(obj, &opts); + if (err) + return err; + + err = bpf_object__load(obj); + if (err) { + p_err("failed to load object file"); + goto out; + } + /* If there was no error during load then gen_loader_opts + * are populated with the loader program. + */ + + /* finish generating 'struct skel' */ + codegen("\ + \n\ + }; \n\ + ", obj_name); + + + codegen_attach_detach(obj, obj_name); + + codegen_destroy(obj, obj_name); + + codegen("\ + \n\ + static inline struct %1$s * \n\ + %1$s__open(void) \n\ + { \n\ + struct %1$s *skel; \n\ + \n\ + skel = skel_alloc(sizeof(*skel)); \n\ + if (!skel) \n\ + goto cleanup; \n\ + skel->ctx.sz = (void *)&skel->links - (void *)skel; \n\ + ", + obj_name, opts.data_sz); + bpf_object__for_each_map(map, obj) { + const void *mmap_data = NULL; + size_t mmap_size = 0; + + if (!is_internal_mmapable_map(map, ident, sizeof(ident))) + continue; + + codegen("\ + \n\ + skel->%1$s = skel_prep_map_data((void *)\"\\ \n\ + ", ident); + mmap_data = bpf_map__initial_value(map, &mmap_size); + print_hex(mmap_data, mmap_size); + codegen("\ + \n\ + \", %1$zd, %2$zd); \n\ + if (!skel->%3$s) \n\ + goto cleanup; \n\ + skel->maps.%3$s.initial_value = (__u64) (long) skel->%3$s;\n\ + ", bpf_map_mmap_sz(map), mmap_size, ident); + } + codegen("\ + \n\ + return skel; \n\ + cleanup: \n\ + %1$s__destroy(skel); \n\ + return NULL; \n\ + } \n\ + \n\ + static inline int \n\ + %1$s__load(struct %1$s *skel) \n\ + { \n\ + struct bpf_load_and_run_opts opts = {}; \n\ + int err; \n\ + \n\ + opts.ctx = (struct bpf_loader_ctx *)skel; \n\ + opts.data_sz = %2$d; \n\ + opts.data = (void *)\"\\ \n\ + ", + obj_name, opts.data_sz); + print_hex(opts.data, opts.data_sz); + codegen("\ + \n\ + \"; \n\ + "); + + codegen("\ + \n\ + opts.insns_sz = %d; \n\ + opts.insns = (void *)\"\\ \n\ + ", + opts.insns_sz); + print_hex(opts.insns, opts.insns_sz); + codegen("\ + \n\ + \"; \n\ + err = bpf_load_and_run(&opts); \n\ + if (err < 0) \n\ + return err; \n\ + ", obj_name); + bpf_object__for_each_map(map, obj) { + const char *mmap_flags; + + if (!is_internal_mmapable_map(map, ident, sizeof(ident))) + continue; + + if (bpf_map__map_flags(map) & BPF_F_RDONLY_PROG) + mmap_flags = "PROT_READ"; + else + mmap_flags = "PROT_READ | PROT_WRITE"; + + codegen("\ + \n\ + skel->%1$s = skel_finalize_map_data(&skel->maps.%1$s.initial_value, \n\ + %2$zd, %3$s, skel->maps.%1$s.map_fd);\n\ + if (!skel->%1$s) \n\ + return -ENOMEM; \n\ + ", + ident, bpf_map_mmap_sz(map), mmap_flags); + } + codegen("\ + \n\ + return 0; \n\ + } \n\ + \n\ + static inline struct %1$s * \n\ + %1$s__open_and_load(void) \n\ + { \n\ + struct %1$s *skel; \n\ + \n\ + skel = %1$s__open(); \n\ + if (!skel) \n\ + return NULL; \n\ + if (%1$s__load(skel)) { \n\ + %1$s__destroy(skel); \n\ + return NULL; \n\ + } \n\ + return skel; \n\ + } \n\ + \n\ + ", obj_name); + + codegen_asserts(obj, obj_name); + + codegen("\ + \n\ + \n\ + #endif /* %s */ \n\ + ", + header_guard); + err = 0; +out: + return err; +} + +static void +codegen_maps_skeleton(struct bpf_object *obj, size_t map_cnt, bool mmaped) +{ + struct bpf_map *map; + char ident[256]; + size_t i; + + if (!map_cnt) + return; + + codegen("\ + \n\ + \n\ + /* maps */ \n\ + s->map_cnt = %zu; \n\ + s->map_skel_sz = sizeof(*s->maps); \n\ + s->maps = (struct bpf_map_skeleton *)calloc(s->map_cnt, s->map_skel_sz);\n\ + if (!s->maps) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ + ", + map_cnt + ); + i = 0; + bpf_object__for_each_map(map, obj) { + if (!get_map_ident(map, ident, sizeof(ident))) + continue; + + codegen("\ + \n\ + \n\ + s->maps[%zu].name = \"%s\"; \n\ + s->maps[%zu].map = &obj->maps.%s; \n\ + ", + i, bpf_map__name(map), i, ident); + /* memory-mapped internal maps */ + if (mmaped && is_internal_mmapable_map(map, ident, sizeof(ident))) { + printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n", + i, ident); + } + i++; + } +} + +static void +codegen_progs_skeleton(struct bpf_object *obj, size_t prog_cnt, bool populate_links) +{ + struct bpf_program *prog; + int i; + + if (!prog_cnt) + return; + + codegen("\ + \n\ + \n\ + /* programs */ \n\ + s->prog_cnt = %zu; \n\ + s->prog_skel_sz = sizeof(*s->progs); \n\ + s->progs = (struct bpf_prog_skeleton *)calloc(s->prog_cnt, s->prog_skel_sz);\n\ + if (!s->progs) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ + ", + prog_cnt + ); + i = 0; + bpf_object__for_each_program(prog, obj) { + codegen("\ + \n\ + \n\ + s->progs[%1$zu].name = \"%2$s\"; \n\ + s->progs[%1$zu].prog = &obj->progs.%2$s;\n\ + ", + i, bpf_program__name(prog)); + + if (populate_links) { + codegen("\ + \n\ + s->progs[%1$zu].link = &obj->links.%2$s;\n\ + ", + i, bpf_program__name(prog)); + } + i++; + } } static int do_skeleton(int argc, char **argv) { char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SKEL_H__")]; - size_t i, map_cnt = 0, prog_cnt = 0, file_sz, mmap_sz; + size_t map_cnt = 0, prog_cnt = 0, file_sz, mmap_sz; DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts); - char obj_name[MAX_OBJ_NAME_LEN], *obj_data; + char obj_name[MAX_OBJ_NAME_LEN] = "", *obj_data; struct bpf_object *obj = NULL; - const char *file, *ident; + const char *file; + char ident[256]; struct bpf_program *prog; - int fd, len, err = -1; + int fd, err = -1; struct bpf_map *map; struct btf *btf; struct stat st; @@ -279,6 +925,28 @@ static int do_skeleton(int argc, char **argv) } file = GET_ARG(); + while (argc) { + if (!REQ_ARGS(2)) + return -1; + + if (is_prefix(*argv, "name")) { + NEXT_ARG(); + + if (obj_name[0] != '\0') { + p_err("object name already specified"); + return -1; + } + + strncpy(obj_name, *argv, MAX_OBJ_NAME_LEN - 1); + obj_name[MAX_OBJ_NAME_LEN - 1] = '\0'; + } else { + p_err("unknown arg %s", *argv); + return -1; + } + + NEXT_ARG(); + } + if (argc) { p_err("extra unknown arguments"); return -1; @@ -301,18 +969,25 @@ static int do_skeleton(int argc, char **argv) p_err("failed to mmap() %s: %s", file, strerror(errno)); goto out; } - get_obj_name(obj_name, file); + if (obj_name[0] == '\0') + get_obj_name(obj_name, file); opts.object_name = obj_name; + if (verifier_logs) + /* log_level1 + log_level2 + stats, but not stable UAPI */ + opts.kernel_log_level = 1 + 2 + 4; obj = bpf_object__open_mem(obj_data, file_sz, &opts); - if (IS_ERR(obj)) { + err = libbpf_get_error(obj); + if (err) { + char err_buf[256]; + + libbpf_strerror(err, err_buf, sizeof(err_buf)); + p_err("failed to open BPF object file: %s", err_buf); obj = NULL; - p_err("failed to open BPF object file: %ld", PTR_ERR(obj)); goto out; } bpf_object__for_each_map(map, obj) { - ident = get_map_ident(map); - if (!ident) { + if (!get_map_ident(map, ident, sizeof(ident))) { p_err("ignoring unrecognized internal map '%s'...", bpf_map__name(map)); continue; @@ -323,15 +998,32 @@ static int do_skeleton(int argc, char **argv) prog_cnt++; } - get_header_guard(header_guard, obj_name); - codegen("\ + get_header_guard(header_guard, obj_name, "SKEL_H"); + if (use_loader) { + codegen("\ + \n\ + /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ \n\ + /* THIS FILE IS AUTOGENERATED BY BPFTOOL! */ \n\ + #ifndef %2$s \n\ + #define %2$s \n\ + \n\ + #include <bpf/skel_internal.h> \n\ + \n\ + struct %1$s { \n\ + struct bpf_loader_ctx ctx; \n\ + ", + obj_name, header_guard + ); + } else { + codegen("\ \n\ /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ \n\ \n\ - /* THIS FILE IS AUTOGENERATED! */ \n\ + /* THIS FILE IS AUTOGENERATED BY BPFTOOL! */ \n\ #ifndef %2$s \n\ #define %2$s \n\ \n\ + #include <errno.h> \n\ #include <stdlib.h> \n\ #include <bpf/libbpf.h> \n\ \n\ @@ -340,15 +1032,18 @@ static int do_skeleton(int argc, char **argv) struct bpf_object *obj; \n\ ", obj_name, header_guard - ); + ); + } if (map_cnt) { printf("\tstruct {\n"); bpf_object__for_each_map(map, obj) { - ident = get_map_ident(map); - if (!ident) + if (!get_map_ident(map, ident, sizeof(ident))) continue; - printf("\t\tstruct bpf_map *%s;\n", ident); + if (use_loader) + printf("\t\tstruct bpf_map_desc %s;\n", ident); + else + printf("\t\tstruct bpf_map *%s;\n", ident); } printf("\t} maps;\n"); } @@ -356,14 +1051,22 @@ static int do_skeleton(int argc, char **argv) if (prog_cnt) { printf("\tstruct {\n"); bpf_object__for_each_program(prog, obj) { - printf("\t\tstruct bpf_program *%s;\n", - bpf_program__name(prog)); + if (use_loader) + printf("\t\tstruct bpf_prog_desc %s;\n", + bpf_program__name(prog)); + else + printf("\t\tstruct bpf_program *%s;\n", + bpf_program__name(prog)); } printf("\t} progs;\n"); printf("\tstruct {\n"); bpf_object__for_each_program(prog, obj) { - printf("\t\tstruct bpf_link *%s;\n", - bpf_program__name(prog)); + if (use_loader) + printf("\t\tint %s_fd;\n", + bpf_program__name(prog)); + else + printf("\t\tstruct bpf_link *%s;\n", + bpf_program__name(prog)); } printf("\t} links;\n"); } @@ -374,9 +1077,23 @@ static int do_skeleton(int argc, char **argv) if (err) goto out; } + if (use_loader) { + err = gen_trace(obj, obj_name, header_guard); + goto out; + } codegen("\ \n\ + \n\ + #ifdef __cplusplus \n\ + static inline struct %1$s *open(const struct bpf_object_open_opts *opts = nullptr);\n\ + static inline struct %1$s *open_and_load(); \n\ + static inline int load(struct %1$s *skel); \n\ + static inline int attach(struct %1$s *skel); \n\ + static inline void detach(struct %1$s *skel); \n\ + static inline void destroy(struct %1$s *skel); \n\ + static inline const void *elf_bytes(size_t *sz); \n\ + #endif /* __cplusplus */ \n\ }; \n\ \n\ static void \n\ @@ -396,18 +1113,26 @@ static int do_skeleton(int argc, char **argv) %1$s__open_opts(const struct bpf_object_open_opts *opts) \n\ { \n\ struct %1$s *obj; \n\ + int err; \n\ \n\ - obj = (typeof(obj))calloc(1, sizeof(*obj)); \n\ - if (!obj) \n\ + obj = (struct %1$s *)calloc(1, sizeof(*obj)); \n\ + if (!obj) { \n\ + errno = ENOMEM; \n\ return NULL; \n\ - if (%1$s__create_skeleton(obj)) \n\ - goto err; \n\ - if (bpf_object__open_skeleton(obj->skeleton, opts)) \n\ - goto err; \n\ + } \n\ + \n\ + err = %1$s__create_skeleton(obj); \n\ + if (err) \n\ + goto err_out; \n\ + \n\ + err = bpf_object__open_skeleton(obj->skeleton, opts);\n\ + if (err) \n\ + goto err_out; \n\ \n\ return obj; \n\ - err: \n\ + err_out: \n\ %1$s__destroy(obj); \n\ + errno = -err; \n\ return NULL; \n\ } \n\ \n\ @@ -427,12 +1152,15 @@ static int do_skeleton(int argc, char **argv) %1$s__open_and_load(void) \n\ { \n\ struct %1$s *obj; \n\ + int err; \n\ \n\ obj = %1$s__open(); \n\ if (!obj) \n\ return NULL; \n\ - if (%1$s__load(obj)) { \n\ + err = %1$s__load(obj); \n\ + if (err) { \n\ %1$s__destroy(obj); \n\ + errno = -err; \n\ return NULL; \n\ } \n\ return obj; \n\ @@ -447,7 +1175,7 @@ static int do_skeleton(int argc, char **argv) static inline void \n\ %1$s__detach(struct %1$s *obj) \n\ { \n\ - return bpf_object__detach_skeleton(obj->skeleton); \n\ + bpf_object__detach_skeleton(obj->skeleton); \n\ } \n\ ", obj_name @@ -456,15 +1184,19 @@ static int do_skeleton(int argc, char **argv) codegen("\ \n\ \n\ + static inline const void *%1$s__elf_bytes(size_t *sz); \n\ + \n\ static inline int \n\ %1$s__create_skeleton(struct %1$s *obj) \n\ { \n\ struct bpf_object_skeleton *s; \n\ + int err; \n\ \n\ - s = (typeof(s))calloc(1, sizeof(*s)); \n\ - if (!s) \n\ - return -1; \n\ - obj->skeleton = s; \n\ + s = (struct bpf_object_skeleton *)calloc(1, sizeof(*s));\n\ + if (!s) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ \n\ s->sz = sizeof(*s); \n\ s->name = \"%1$s\"; \n\ @@ -472,104 +1204,363 @@ static int do_skeleton(int argc, char **argv) ", obj_name ); - if (map_cnt) { - codegen("\ - \n\ + + codegen_maps_skeleton(obj, map_cnt, true /*mmaped*/); + codegen_progs_skeleton(obj, prog_cnt, true /*populate_links*/); + + codegen("\ + \n\ \n\ - /* maps */ \n\ - s->map_cnt = %zu; \n\ - s->map_skel_sz = sizeof(*s->maps); \n\ - s->maps = (typeof(s->maps))calloc(s->map_cnt, s->map_skel_sz);\n\ - if (!s->maps) \n\ - goto err; \n\ - ", - map_cnt - ); - i = 0; - bpf_object__for_each_map(map, obj) { - ident = get_map_ident(map); + s->data = (void *)%2$s__elf_bytes(&s->data_sz); \n\ + \n\ + obj->skeleton = s; \n\ + return 0; \n\ + err: \n\ + bpf_object__destroy_skeleton(s); \n\ + return err; \n\ + } \n\ + \n\ + static inline const void *%2$s__elf_bytes(size_t *sz) \n\ + { \n\ + *sz = %1$d; \n\ + return (const void *)\"\\ \n\ + " + , file_sz, obj_name); - if (!ident) - continue; + /* embed contents of BPF object file */ + print_hex(obj_data, file_sz); - codegen("\ - \n\ + codegen("\ + \n\ + \"; \n\ + } \n\ \n\ - s->maps[%zu].name = \"%s\"; \n\ - s->maps[%zu].map = &obj->maps.%s; \n\ - ", - i, bpf_map__name(map), i, ident); - /* memory-mapped internal maps */ - if (bpf_map__is_internal(map) && - (bpf_map__def(map)->map_flags & BPF_F_MMAPABLE)) { - printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n", - i, ident); + #ifdef __cplusplus \n\ + struct %1$s *%1$s::open(const struct bpf_object_open_opts *opts) { return %1$s__open_opts(opts); }\n\ + struct %1$s *%1$s::open_and_load() { return %1$s__open_and_load(); } \n\ + int %1$s::load(struct %1$s *skel) { return %1$s__load(skel); } \n\ + int %1$s::attach(struct %1$s *skel) { return %1$s__attach(skel); } \n\ + void %1$s::detach(struct %1$s *skel) { %1$s__detach(skel); } \n\ + void %1$s::destroy(struct %1$s *skel) { %1$s__destroy(skel); } \n\ + const void *%1$s::elf_bytes(size_t *sz) { return %1$s__elf_bytes(sz); } \n\ + #endif /* __cplusplus */ \n\ + \n\ + ", + obj_name); + + codegen_asserts(obj, obj_name); + + codegen("\ + \n\ + \n\ + #endif /* %1$s */ \n\ + ", + header_guard); + err = 0; +out: + bpf_object__close(obj); + if (obj_data) + munmap(obj_data, mmap_sz); + close(fd); + return err; +} + +/* Subskeletons are like skeletons, except they don't own the bpf_object, + * associated maps, links, etc. Instead, they know about the existence of + * variables, maps, programs and are able to find their locations + * _at runtime_ from an already loaded bpf_object. + * + * This allows for library-like BPF objects to have userspace counterparts + * with access to their own items without having to know anything about the + * final BPF object that the library was linked into. + */ +static int do_subskeleton(int argc, char **argv) +{ + char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SUBSKEL_H__")]; + size_t i, len, file_sz, map_cnt = 0, prog_cnt = 0, mmap_sz, var_cnt = 0, var_idx = 0; + DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts); + char obj_name[MAX_OBJ_NAME_LEN] = "", *obj_data; + struct bpf_object *obj = NULL; + const char *file, *var_name; + char ident[256]; + int fd, err = -1, map_type_id; + const struct bpf_map *map; + struct bpf_program *prog; + struct btf *btf; + const struct btf_type *map_type, *var_type; + const struct btf_var_secinfo *var; + struct stat st; + + if (!REQ_ARGS(1)) { + usage(); + return -1; + } + file = GET_ARG(); + + while (argc) { + if (!REQ_ARGS(2)) + return -1; + + if (is_prefix(*argv, "name")) { + NEXT_ARG(); + + if (obj_name[0] != '\0') { + p_err("object name already specified"); + return -1; } - i++; + + strncpy(obj_name, *argv, MAX_OBJ_NAME_LEN - 1); + obj_name[MAX_OBJ_NAME_LEN - 1] = '\0'; + } else { + p_err("unknown arg %s", *argv); + return -1; } + + NEXT_ARG(); } - if (prog_cnt) { - codegen("\ - \n\ + + if (argc) { + p_err("extra unknown arguments"); + return -1; + } + + if (use_loader) { + p_err("cannot use loader for subskeletons"); + return -1; + } + + if (stat(file, &st)) { + p_err("failed to stat() %s: %s", file, strerror(errno)); + return -1; + } + file_sz = st.st_size; + mmap_sz = roundup(file_sz, sysconf(_SC_PAGE_SIZE)); + fd = open(file, O_RDONLY); + if (fd < 0) { + p_err("failed to open() %s: %s", file, strerror(errno)); + return -1; + } + obj_data = mmap(NULL, mmap_sz, PROT_READ, MAP_PRIVATE, fd, 0); + if (obj_data == MAP_FAILED) { + obj_data = NULL; + p_err("failed to mmap() %s: %s", file, strerror(errno)); + goto out; + } + if (obj_name[0] == '\0') + get_obj_name(obj_name, file); + + /* The empty object name allows us to use bpf_map__name and produce + * ELF section names out of it. (".data" instead of "obj.data") + */ + opts.object_name = ""; + obj = bpf_object__open_mem(obj_data, file_sz, &opts); + if (!obj) { + char err_buf[256]; + + libbpf_strerror(errno, err_buf, sizeof(err_buf)); + p_err("failed to open BPF object file: %s", err_buf); + obj = NULL; + goto out; + } + + btf = bpf_object__btf(obj); + if (!btf) { + err = -1; + p_err("need btf type information for %s", obj_name); + goto out; + } + + bpf_object__for_each_program(prog, obj) { + prog_cnt++; + } + + /* First, count how many variables we have to find. + * We need this in advance so the subskel can allocate the right + * amount of storage. + */ + bpf_object__for_each_map(map, obj) { + if (!get_map_ident(map, ident, sizeof(ident))) + continue; + + /* Also count all maps that have a name */ + map_cnt++; + + if (!is_internal_mmapable_map(map, ident, sizeof(ident))) + continue; + + map_type_id = bpf_map__btf_value_type_id(map); + if (map_type_id <= 0) { + err = map_type_id; + goto out; + } + map_type = btf__type_by_id(btf, map_type_id); + + var = btf_var_secinfos(map_type); + len = btf_vlen(map_type); + for (i = 0; i < len; i++, var++) { + var_type = btf__type_by_id(btf, var->type); + + if (btf_var(var_type)->linkage == BTF_VAR_STATIC) + continue; + + var_cnt++; + } + } + + get_header_guard(header_guard, obj_name, "SUBSKEL_H"); + codegen("\ + \n\ + /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ \n\ \n\ - /* programs */ \n\ - s->prog_cnt = %zu; \n\ - s->prog_skel_sz = sizeof(*s->progs); \n\ - s->progs = (typeof(s->progs))calloc(s->prog_cnt, s->prog_skel_sz);\n\ - if (!s->progs) \n\ - goto err; \n\ - ", - prog_cnt - ); - i = 0; - bpf_object__for_each_program(prog, obj) { - codegen("\ - \n\ + /* THIS FILE IS AUTOGENERATED! */ \n\ + #ifndef %2$s \n\ + #define %2$s \n\ \n\ - s->progs[%1$zu].name = \"%2$s\"; \n\ - s->progs[%1$zu].prog = &obj->progs.%2$s;\n\ - s->progs[%1$zu].link = &obj->links.%2$s;\n\ - ", - i, bpf_program__name(prog)); - i++; + #include <errno.h> \n\ + #include <stdlib.h> \n\ + #include <bpf/libbpf.h> \n\ + \n\ + struct %1$s { \n\ + struct bpf_object *obj; \n\ + struct bpf_object_subskeleton *subskel; \n\ + ", obj_name, header_guard); + + if (map_cnt) { + printf("\tstruct {\n"); + bpf_object__for_each_map(map, obj) { + if (!get_map_ident(map, ident, sizeof(ident))) + continue; + printf("\t\tstruct bpf_map *%s;\n", ident); + } + printf("\t} maps;\n"); + } + + if (prog_cnt) { + printf("\tstruct {\n"); + bpf_object__for_each_program(prog, obj) { + printf("\t\tstruct bpf_program *%s;\n", + bpf_program__name(prog)); } + printf("\t} progs;\n"); } + + err = codegen_subskel_datasecs(obj, obj_name); + if (err) + goto out; + + /* emit code that will allocate enough storage for all symbols */ codegen("\ \n\ \n\ - s->data_sz = %d; \n\ - s->data = (void *)\"\\ \n\ + #ifdef __cplusplus \n\ + static inline struct %1$s *open(const struct bpf_object *src);\n\ + static inline void destroy(struct %1$s *skel); \n\ + #endif /* __cplusplus */ \n\ + }; \n\ + \n\ + static inline void \n\ + %1$s__destroy(struct %1$s *skel) \n\ + { \n\ + if (!skel) \n\ + return; \n\ + if (skel->subskel) \n\ + bpf_object__destroy_subskeleton(skel->subskel);\n\ + free(skel); \n\ + } \n\ + \n\ + static inline struct %1$s * \n\ + %1$s__open(const struct bpf_object *src) \n\ + { \n\ + struct %1$s *obj; \n\ + struct bpf_object_subskeleton *s; \n\ + int err; \n\ + \n\ + obj = (struct %1$s *)calloc(1, sizeof(*obj)); \n\ + if (!obj) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ + s = (struct bpf_object_subskeleton *)calloc(1, sizeof(*s));\n\ + if (!s) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ + s->sz = sizeof(*s); \n\ + s->obj = src; \n\ + s->var_skel_sz = sizeof(*s->vars); \n\ + obj->subskel = s; \n\ + \n\ + /* vars */ \n\ + s->var_cnt = %2$d; \n\ + s->vars = (struct bpf_var_skeleton *)calloc(%2$d, sizeof(*s->vars));\n\ + if (!s->vars) { \n\ + err = -ENOMEM; \n\ + goto err; \n\ + } \n\ ", - file_sz); + obj_name, var_cnt + ); - /* embed contents of BPF object file */ - for (i = 0, len = 0; i < file_sz; i++) { - int w = obj_data[i] ? 4 : 2; + /* walk through each symbol and emit the runtime representation */ + bpf_object__for_each_map(map, obj) { + if (!is_internal_mmapable_map(map, ident, sizeof(ident))) + continue; - len += w; - if (len > 78) { - printf("\\\n"); - len = w; + map_type_id = bpf_map__btf_value_type_id(map); + if (map_type_id <= 0) + /* skip over internal maps with no type*/ + continue; + + map_type = btf__type_by_id(btf, map_type_id); + var = btf_var_secinfos(map_type); + len = btf_vlen(map_type); + for (i = 0; i < len; i++, var++) { + var_type = btf__type_by_id(btf, var->type); + var_name = btf__name_by_offset(btf, var_type->name_off); + + if (btf_var(var_type)->linkage == BTF_VAR_STATIC) + continue; + + /* Note that we use the dot prefix in .data as the + * field access operator i.e. maps%s becomes maps.data + */ + codegen("\ + \n\ + \n\ + s->vars[%3$d].name = \"%1$s\"; \n\ + s->vars[%3$d].map = &obj->maps.%2$s; \n\ + s->vars[%3$d].addr = (void **) &obj->%2$s.%1$s;\n\ + ", var_name, ident, var_idx); + + var_idx++; } - if (!obj_data[i]) - printf("\\0"); - else - printf("\\x%02x", (unsigned char)obj_data[i]); } + codegen_maps_skeleton(obj, map_cnt, false /*mmaped*/); + codegen_progs_skeleton(obj, prog_cnt, false /*links*/); + codegen("\ \n\ - \"; \n\ \n\ - return 0; \n\ + err = bpf_object__open_subskeleton(s); \n\ + if (err) \n\ + goto err; \n\ + \n\ + return obj; \n\ err: \n\ - bpf_object__destroy_skeleton(s); \n\ - return -1; \n\ + %1$s__destroy(obj); \n\ + errno = -err; \n\ + return NULL; \n\ } \n\ \n\ - #endif /* %s */ \n\ + #ifdef __cplusplus \n\ + struct %1$s *%1$s::open(const struct bpf_object *src) { return %1$s__open(src); }\n\ + void %1$s::destroy(struct %1$s *skel) { %1$s__destroy(skel); }\n\ + #endif /* __cplusplus */ \n\ + \n\ + #endif /* %2$s */ \n\ ", - header_guard); + obj_name, header_guard); err = 0; out: bpf_object__close(obj); @@ -579,6 +1570,47 @@ out: return err; } +static int do_object(int argc, char **argv) +{ + struct bpf_linker *linker; + const char *output_file, *file; + int err = 0; + + if (!REQ_ARGS(2)) { + usage(); + return -1; + } + + output_file = GET_ARG(); + + linker = bpf_linker__new(output_file, NULL); + if (!linker) { + p_err("failed to create BPF linker instance"); + return -1; + } + + while (argc) { + file = GET_ARG(); + + err = bpf_linker__add_file(linker, file, NULL); + if (err) { + p_err("failed to link '%s': %s (%d)", file, strerror(errno), errno); + goto out; + } + } + + err = bpf_linker__finalize(linker); + if (err) { + p_err("failed to finalize ELF file: %s (%d)", strerror(errno), errno); + goto out; + } + + err = 0; +out: + bpf_linker__free(linker); + return err; +} + static int do_help(int argc, char **argv) { if (json_output) { @@ -587,19 +1619,718 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %1$s gen skeleton FILE\n" - " %1$s gen help\n" + "Usage: %1$s %2$s object OUTPUT_FILE INPUT_FILE [INPUT_FILE...]\n" + " %1$s %2$s skeleton FILE [name OBJECT_NAME]\n" + " %1$s %2$s subskeleton FILE [name OBJECT_NAME]\n" + " %1$s %2$s min_core_btf INPUT OUTPUT OBJECT [OBJECT...]\n" + " %1$s %2$s help\n" "\n" - " " HELP_SPEC_OPTIONS "\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-L|--use-loader} }\n" "", - bin_name); + bin_name, "gen"); return 0; } +static int btf_save_raw(const struct btf *btf, const char *path) +{ + const void *data; + FILE *f = NULL; + __u32 data_sz; + int err = 0; + + data = btf__raw_data(btf, &data_sz); + if (!data) + return -ENOMEM; + + f = fopen(path, "wb"); + if (!f) + return -errno; + + if (fwrite(data, 1, data_sz, f) != data_sz) + err = -errno; + + fclose(f); + return err; +} + +struct btfgen_info { + struct btf *src_btf; + struct btf *marked_btf; /* btf structure used to mark used types */ +}; + +static size_t btfgen_hash_fn(const void *key, void *ctx) +{ + return (size_t)key; +} + +static bool btfgen_equal_fn(const void *k1, const void *k2, void *ctx) +{ + return k1 == k2; +} + +static void *u32_as_hash_key(__u32 x) +{ + return (void *)(uintptr_t)x; +} + +static void btfgen_free_info(struct btfgen_info *info) +{ + if (!info) + return; + + btf__free(info->src_btf); + btf__free(info->marked_btf); + + free(info); +} + +static struct btfgen_info * +btfgen_new_info(const char *targ_btf_path) +{ + struct btfgen_info *info; + int err; + + info = calloc(1, sizeof(*info)); + if (!info) + return NULL; + + info->src_btf = btf__parse(targ_btf_path, NULL); + if (!info->src_btf) { + err = -errno; + p_err("failed parsing '%s' BTF file: %s", targ_btf_path, strerror(errno)); + goto err_out; + } + + info->marked_btf = btf__parse(targ_btf_path, NULL); + if (!info->marked_btf) { + err = -errno; + p_err("failed parsing '%s' BTF file: %s", targ_btf_path, strerror(errno)); + goto err_out; + } + + return info; + +err_out: + btfgen_free_info(info); + errno = -err; + return NULL; +} + +#define MARKED UINT32_MAX + +static void btfgen_mark_member(struct btfgen_info *info, int type_id, int idx) +{ + const struct btf_type *t = btf__type_by_id(info->marked_btf, type_id); + struct btf_member *m = btf_members(t) + idx; + + m->name_off = MARKED; +} + +static int +btfgen_mark_type(struct btfgen_info *info, unsigned int type_id, bool follow_pointers) +{ + const struct btf_type *btf_type = btf__type_by_id(info->src_btf, type_id); + struct btf_type *cloned_type; + struct btf_param *param; + struct btf_array *array; + int err, i; + + if (type_id == 0) + return 0; + + /* mark type on cloned BTF as used */ + cloned_type = (struct btf_type *) btf__type_by_id(info->marked_btf, type_id); + cloned_type->name_off = MARKED; + + /* recursively mark other types needed by it */ + switch (btf_kind(btf_type)) { + case BTF_KIND_UNKN: + case BTF_KIND_INT: + case BTF_KIND_FLOAT: + case BTF_KIND_ENUM: + case BTF_KIND_ENUM64: + case BTF_KIND_STRUCT: + case BTF_KIND_UNION: + break; + case BTF_KIND_PTR: + if (follow_pointers) { + err = btfgen_mark_type(info, btf_type->type, follow_pointers); + if (err) + return err; + } + break; + case BTF_KIND_CONST: + case BTF_KIND_RESTRICT: + case BTF_KIND_VOLATILE: + case BTF_KIND_TYPEDEF: + err = btfgen_mark_type(info, btf_type->type, follow_pointers); + if (err) + return err; + break; + case BTF_KIND_ARRAY: + array = btf_array(btf_type); + + /* mark array type */ + err = btfgen_mark_type(info, array->type, follow_pointers); + /* mark array's index type */ + err = err ? : btfgen_mark_type(info, array->index_type, follow_pointers); + if (err) + return err; + break; + case BTF_KIND_FUNC_PROTO: + /* mark ret type */ + err = btfgen_mark_type(info, btf_type->type, follow_pointers); + if (err) + return err; + + /* mark parameters types */ + param = btf_params(btf_type); + for (i = 0; i < btf_vlen(btf_type); i++) { + err = btfgen_mark_type(info, param->type, follow_pointers); + if (err) + return err; + param++; + } + break; + /* tells if some other type needs to be handled */ + default: + p_err("unsupported kind: %s (%d)", btf_kind_str(btf_type), type_id); + return -EINVAL; + } + + return 0; +} + +static int btfgen_record_field_relo(struct btfgen_info *info, struct bpf_core_spec *targ_spec) +{ + struct btf *btf = info->src_btf; + const struct btf_type *btf_type; + struct btf_member *btf_member; + struct btf_array *array; + unsigned int type_id = targ_spec->root_type_id; + int idx, err; + + /* mark root type */ + btf_type = btf__type_by_id(btf, type_id); + err = btfgen_mark_type(info, type_id, false); + if (err) + return err; + + /* mark types for complex types (arrays, unions, structures) */ + for (int i = 1; i < targ_spec->raw_len; i++) { + /* skip typedefs and mods */ + while (btf_is_mod(btf_type) || btf_is_typedef(btf_type)) { + type_id = btf_type->type; + btf_type = btf__type_by_id(btf, type_id); + } + + switch (btf_kind(btf_type)) { + case BTF_KIND_STRUCT: + case BTF_KIND_UNION: + idx = targ_spec->raw_spec[i]; + btf_member = btf_members(btf_type) + idx; + + /* mark member */ + btfgen_mark_member(info, type_id, idx); + + /* mark member's type */ + type_id = btf_member->type; + btf_type = btf__type_by_id(btf, type_id); + err = btfgen_mark_type(info, type_id, false); + if (err) + return err; + break; + case BTF_KIND_ARRAY: + array = btf_array(btf_type); + type_id = array->type; + btf_type = btf__type_by_id(btf, type_id); + break; + default: + p_err("unsupported kind: %s (%d)", + btf_kind_str(btf_type), btf_type->type); + return -EINVAL; + } + } + + return 0; +} + +/* Mark types, members, and member types. Compared to btfgen_record_field_relo, + * this function does not rely on the target spec for inferring members, but + * uses the associated BTF. + * + * The `behind_ptr` argument is used to stop marking of composite types reached + * through a pointer. This way, we can keep BTF size in check while providing + * reasonable match semantics. + */ +static int btfgen_mark_type_match(struct btfgen_info *info, __u32 type_id, bool behind_ptr) +{ + const struct btf_type *btf_type; + struct btf *btf = info->src_btf; + struct btf_type *cloned_type; + int i, err; + + if (type_id == 0) + return 0; + + btf_type = btf__type_by_id(btf, type_id); + /* mark type on cloned BTF as used */ + cloned_type = (struct btf_type *)btf__type_by_id(info->marked_btf, type_id); + cloned_type->name_off = MARKED; + + switch (btf_kind(btf_type)) { + case BTF_KIND_UNKN: + case BTF_KIND_INT: + case BTF_KIND_FLOAT: + case BTF_KIND_ENUM: + case BTF_KIND_ENUM64: + break; + case BTF_KIND_STRUCT: + case BTF_KIND_UNION: { + struct btf_member *m = btf_members(btf_type); + __u16 vlen = btf_vlen(btf_type); + + if (behind_ptr) + break; + + for (i = 0; i < vlen; i++, m++) { + /* mark member */ + btfgen_mark_member(info, type_id, i); + + /* mark member's type */ + err = btfgen_mark_type_match(info, m->type, false); + if (err) + return err; + } + break; + } + case BTF_KIND_CONST: + case BTF_KIND_FWD: + case BTF_KIND_RESTRICT: + case BTF_KIND_TYPEDEF: + case BTF_KIND_VOLATILE: + return btfgen_mark_type_match(info, btf_type->type, behind_ptr); + case BTF_KIND_PTR: + return btfgen_mark_type_match(info, btf_type->type, true); + case BTF_KIND_ARRAY: { + struct btf_array *array; + + array = btf_array(btf_type); + /* mark array type */ + err = btfgen_mark_type_match(info, array->type, false); + /* mark array's index type */ + err = err ? : btfgen_mark_type_match(info, array->index_type, false); + if (err) + return err; + break; + } + case BTF_KIND_FUNC_PROTO: { + __u16 vlen = btf_vlen(btf_type); + struct btf_param *param; + + /* mark ret type */ + err = btfgen_mark_type_match(info, btf_type->type, false); + if (err) + return err; + + /* mark parameters types */ + param = btf_params(btf_type); + for (i = 0; i < vlen; i++) { + err = btfgen_mark_type_match(info, param->type, false); + if (err) + return err; + param++; + } + break; + } + /* tells if some other type needs to be handled */ + default: + p_err("unsupported kind: %s (%d)", btf_kind_str(btf_type), type_id); + return -EINVAL; + } + + return 0; +} + +/* Mark types, members, and member types. Compared to btfgen_record_field_relo, + * this function does not rely on the target spec for inferring members, but + * uses the associated BTF. + */ +static int btfgen_record_type_match_relo(struct btfgen_info *info, struct bpf_core_spec *targ_spec) +{ + return btfgen_mark_type_match(info, targ_spec->root_type_id, false); +} + +static int btfgen_record_type_relo(struct btfgen_info *info, struct bpf_core_spec *targ_spec) +{ + return btfgen_mark_type(info, targ_spec->root_type_id, true); +} + +static int btfgen_record_enumval_relo(struct btfgen_info *info, struct bpf_core_spec *targ_spec) +{ + return btfgen_mark_type(info, targ_spec->root_type_id, false); +} + +static int btfgen_record_reloc(struct btfgen_info *info, struct bpf_core_spec *res) +{ + switch (res->relo_kind) { + case BPF_CORE_FIELD_BYTE_OFFSET: + case BPF_CORE_FIELD_BYTE_SIZE: + case BPF_CORE_FIELD_EXISTS: + case BPF_CORE_FIELD_SIGNED: + case BPF_CORE_FIELD_LSHIFT_U64: + case BPF_CORE_FIELD_RSHIFT_U64: + return btfgen_record_field_relo(info, res); + case BPF_CORE_TYPE_ID_LOCAL: /* BPF_CORE_TYPE_ID_LOCAL doesn't require kernel BTF */ + return 0; + case BPF_CORE_TYPE_ID_TARGET: + case BPF_CORE_TYPE_EXISTS: + case BPF_CORE_TYPE_SIZE: + return btfgen_record_type_relo(info, res); + case BPF_CORE_TYPE_MATCHES: + return btfgen_record_type_match_relo(info, res); + case BPF_CORE_ENUMVAL_EXISTS: + case BPF_CORE_ENUMVAL_VALUE: + return btfgen_record_enumval_relo(info, res); + default: + return -EINVAL; + } +} + +static struct bpf_core_cand_list * +btfgen_find_cands(const struct btf *local_btf, const struct btf *targ_btf, __u32 local_id) +{ + const struct btf_type *local_type; + struct bpf_core_cand_list *cands = NULL; + struct bpf_core_cand local_cand = {}; + size_t local_essent_len; + const char *local_name; + int err; + + local_cand.btf = local_btf; + local_cand.id = local_id; + + local_type = btf__type_by_id(local_btf, local_id); + if (!local_type) { + err = -EINVAL; + goto err_out; + } + + local_name = btf__name_by_offset(local_btf, local_type->name_off); + if (!local_name) { + err = -EINVAL; + goto err_out; + } + local_essent_len = bpf_core_essential_name_len(local_name); + + cands = calloc(1, sizeof(*cands)); + if (!cands) + return NULL; + + err = bpf_core_add_cands(&local_cand, local_essent_len, targ_btf, "vmlinux", 1, cands); + if (err) + goto err_out; + + return cands; + +err_out: + bpf_core_free_cands(cands); + errno = -err; + return NULL; +} + +/* Record relocation information for a single BPF object */ +static int btfgen_record_obj(struct btfgen_info *info, const char *obj_path) +{ + const struct btf_ext_info_sec *sec; + const struct bpf_core_relo *relo; + const struct btf_ext_info *seg; + struct hashmap_entry *entry; + struct hashmap *cand_cache = NULL; + struct btf_ext *btf_ext = NULL; + unsigned int relo_idx; + struct btf *btf = NULL; + size_t i; + int err; + + btf = btf__parse(obj_path, &btf_ext); + if (!btf) { + err = -errno; + p_err("failed to parse BPF object '%s': %s", obj_path, strerror(errno)); + return err; + } + + if (!btf_ext) { + p_err("failed to parse BPF object '%s': section %s not found", + obj_path, BTF_EXT_ELF_SEC); + err = -EINVAL; + goto out; + } + + if (btf_ext->core_relo_info.len == 0) { + err = 0; + goto out; + } + + cand_cache = hashmap__new(btfgen_hash_fn, btfgen_equal_fn, NULL); + if (IS_ERR(cand_cache)) { + err = PTR_ERR(cand_cache); + goto out; + } + + seg = &btf_ext->core_relo_info; + for_each_btf_ext_sec(seg, sec) { + for_each_btf_ext_rec(seg, sec, relo_idx, relo) { + struct bpf_core_spec specs_scratch[3] = {}; + struct bpf_core_relo_res targ_res = {}; + struct bpf_core_cand_list *cands = NULL; + const void *type_key = u32_as_hash_key(relo->type_id); + const char *sec_name = btf__name_by_offset(btf, sec->sec_name_off); + + if (relo->kind != BPF_CORE_TYPE_ID_LOCAL && + !hashmap__find(cand_cache, type_key, (void **)&cands)) { + cands = btfgen_find_cands(btf, info->src_btf, relo->type_id); + if (!cands) { + err = -errno; + goto out; + } + + err = hashmap__set(cand_cache, type_key, cands, NULL, NULL); + if (err) + goto out; + } + + err = bpf_core_calc_relo_insn(sec_name, relo, relo_idx, btf, cands, + specs_scratch, &targ_res); + if (err) + goto out; + + /* specs_scratch[2] is the target spec */ + err = btfgen_record_reloc(info, &specs_scratch[2]); + if (err) + goto out; + } + } + +out: + btf__free(btf); + btf_ext__free(btf_ext); + + if (!IS_ERR_OR_NULL(cand_cache)) { + hashmap__for_each_entry(cand_cache, entry, i) { + bpf_core_free_cands(entry->value); + } + hashmap__free(cand_cache); + } + + return err; +} + +static int btfgen_remap_id(__u32 *type_id, void *ctx) +{ + unsigned int *ids = ctx; + + *type_id = ids[*type_id]; + + return 0; +} + +/* Generate BTF from relocation information previously recorded */ +static struct btf *btfgen_get_btf(struct btfgen_info *info) +{ + struct btf *btf_new = NULL; + unsigned int *ids = NULL; + unsigned int i, n = btf__type_cnt(info->marked_btf); + int err = 0; + + btf_new = btf__new_empty(); + if (!btf_new) { + err = -errno; + goto err_out; + } + + ids = calloc(n, sizeof(*ids)); + if (!ids) { + err = -errno; + goto err_out; + } + + /* first pass: add all marked types to btf_new and add their new ids to the ids map */ + for (i = 1; i < n; i++) { + const struct btf_type *cloned_type, *type; + const char *name; + int new_id; + + cloned_type = btf__type_by_id(info->marked_btf, i); + + if (cloned_type->name_off != MARKED) + continue; + + type = btf__type_by_id(info->src_btf, i); + + /* add members for struct and union */ + if (btf_is_composite(type)) { + struct btf_member *cloned_m, *m; + unsigned short vlen; + int idx_src; + + name = btf__str_by_offset(info->src_btf, type->name_off); + + if (btf_is_struct(type)) + err = btf__add_struct(btf_new, name, type->size); + else + err = btf__add_union(btf_new, name, type->size); + + if (err < 0) + goto err_out; + new_id = err; + + cloned_m = btf_members(cloned_type); + m = btf_members(type); + vlen = btf_vlen(cloned_type); + for (idx_src = 0; idx_src < vlen; idx_src++, cloned_m++, m++) { + /* add only members that are marked as used */ + if (cloned_m->name_off != MARKED) + continue; + + name = btf__str_by_offset(info->src_btf, m->name_off); + err = btf__add_field(btf_new, name, m->type, + btf_member_bit_offset(cloned_type, idx_src), + btf_member_bitfield_size(cloned_type, idx_src)); + if (err < 0) + goto err_out; + } + } else { + err = btf__add_type(btf_new, info->src_btf, type); + if (err < 0) + goto err_out; + new_id = err; + } + + /* add ID mapping */ + ids[i] = new_id; + } + + /* second pass: fix up type ids */ + for (i = 1; i < btf__type_cnt(btf_new); i++) { + struct btf_type *btf_type = (struct btf_type *) btf__type_by_id(btf_new, i); + + err = btf_type_visit_type_ids(btf_type, btfgen_remap_id, ids); + if (err) + goto err_out; + } + + free(ids); + return btf_new; + +err_out: + btf__free(btf_new); + free(ids); + errno = -err; + return NULL; +} + +/* Create minimized BTF file for a set of BPF objects. + * + * The BTFGen algorithm is divided in two main parts: (1) collect the + * BTF types that are involved in relocations and (2) generate the BTF + * object using the collected types. + * + * In order to collect the types involved in the relocations, we parse + * the BTF and BTF.ext sections of the BPF objects and use + * bpf_core_calc_relo_insn() to get the target specification, this + * indicates how the types and fields are used in a relocation. + * + * Types are recorded in different ways according to the kind of the + * relocation. For field-based relocations only the members that are + * actually used are saved in order to reduce the size of the generated + * BTF file. For type-based relocations empty struct / unions are + * generated and for enum-based relocations the whole type is saved. + * + * The second part of the algorithm generates the BTF object. It creates + * an empty BTF object and fills it with the types recorded in the + * previous step. This function takes care of only adding the structure + * and union members that were marked as used and it also fixes up the + * type IDs on the generated BTF object. + */ +static int minimize_btf(const char *src_btf, const char *dst_btf, const char *objspaths[]) +{ + struct btfgen_info *info; + struct btf *btf_new = NULL; + int err, i; + + info = btfgen_new_info(src_btf); + if (!info) { + err = -errno; + p_err("failed to allocate info structure: %s", strerror(errno)); + goto out; + } + + for (i = 0; objspaths[i] != NULL; i++) { + err = btfgen_record_obj(info, objspaths[i]); + if (err) { + p_err("error recording relocations for %s: %s", objspaths[i], + strerror(errno)); + goto out; + } + } + + btf_new = btfgen_get_btf(info); + if (!btf_new) { + err = -errno; + p_err("error generating BTF: %s", strerror(errno)); + goto out; + } + + err = btf_save_raw(btf_new, dst_btf); + if (err) { + p_err("error saving btf file: %s", strerror(errno)); + goto out; + } + +out: + btf__free(btf_new); + btfgen_free_info(info); + + return err; +} + +static int do_min_core_btf(int argc, char **argv) +{ + const char *input, *output, **objs; + int i, err; + + if (!REQ_ARGS(3)) { + usage(); + return -1; + } + + input = GET_ARG(); + output = GET_ARG(); + + objs = (const char **) calloc(argc + 1, sizeof(*objs)); + if (!objs) { + p_err("failed to allocate array for object names"); + return -ENOMEM; + } + + i = 0; + while (argc) + objs[i++] = GET_ARG(); + + err = minimize_btf(input, output, objs); + free(objs); + return err; +} + static const struct cmd cmds[] = { - { "skeleton", do_skeleton }, - { "help", do_help }, + { "object", do_object }, + { "skeleton", do_skeleton }, + { "subskeleton", do_subskeleton }, + { "min_core_btf", do_min_core_btf}, + { "help", do_help }, { 0 } }; diff --git a/tools/bpf/bpftool/iter.c b/tools/bpf/bpftool/iter.c new file mode 100644 index 000000000000..f88fdc820d23 --- /dev/null +++ b/tools/bpf/bpftool/iter.c @@ -0,0 +1,119 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +// Copyright (C) 2020 Facebook + +#define _GNU_SOURCE +#include <unistd.h> +#include <linux/err.h> +#include <bpf/libbpf.h> + +#include "main.h" + +static int do_pin(int argc, char **argv) +{ + DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, iter_opts); + union bpf_iter_link_info linfo; + const char *objfile, *path; + struct bpf_program *prog; + struct bpf_object *obj; + struct bpf_link *link; + int err = -1, map_fd = -1; + + if (!REQ_ARGS(2)) + usage(); + + objfile = GET_ARG(); + path = GET_ARG(); + + /* optional arguments */ + if (argc) { + if (is_prefix(*argv, "map")) { + NEXT_ARG(); + + if (!REQ_ARGS(2)) { + p_err("incorrect map spec"); + return -1; + } + + map_fd = map_parse_fd(&argc, &argv); + if (map_fd < 0) + return -1; + + memset(&linfo, 0, sizeof(linfo)); + linfo.map.map_fd = map_fd; + iter_opts.link_info = &linfo; + iter_opts.link_info_len = sizeof(linfo); + } + } + + obj = bpf_object__open(objfile); + err = libbpf_get_error(obj); + if (err) { + p_err("can't open objfile %s", objfile); + goto close_map_fd; + } + + err = bpf_object__load(obj); + if (err) { + p_err("can't load objfile %s", objfile); + goto close_obj; + } + + prog = bpf_object__next_program(obj, NULL); + if (!prog) { + p_err("can't find bpf program in objfile %s", objfile); + goto close_obj; + } + + link = bpf_program__attach_iter(prog, &iter_opts); + err = libbpf_get_error(link); + if (err) { + p_err("attach_iter failed for program %s", + bpf_program__name(prog)); + goto close_obj; + } + + err = mount_bpffs_for_pin(path); + if (err) + goto close_link; + + err = bpf_link__pin(link, path); + if (err) { + p_err("pin_iter failed for program %s to path %s", + bpf_program__name(prog), path); + goto close_link; + } + +close_link: + bpf_link__destroy(link); +close_obj: + bpf_object__close(obj); +close_map_fd: + if (map_fd >= 0) + close(map_fd); + return err; +} + +static int do_help(int argc, char **argv) +{ + fprintf(stderr, + "Usage: %1$s %2$s pin OBJ PATH [map MAP]\n" + " %1$s %2$s help\n" + "\n" + " " HELP_SPEC_MAP "\n" + " " HELP_SPEC_OPTIONS " }\n" + "", + bin_name, "iter"); + + return 0; +} + +static const struct cmd cmds[] = { + { "help", do_help }, + { "pin", do_pin }, + { 0 } +}; + +int do_iter(int argc, char **argv) +{ + return cmd_select(cmds, argc, argv, do_help); +} diff --git a/tools/bpf/bpftool/jit_disasm.c b/tools/bpf/bpftool/jit_disasm.c index f7f5885aa3ba..aaf99a0168c9 100644 --- a/tools/bpf/bpftool/jit_disasm.c +++ b/tools/bpf/bpftool/jit_disasm.c @@ -15,7 +15,6 @@ #include <stdio.h> #include <stdarg.h> #include <stdint.h> -#include <stdio.h> #include <stdlib.h> #include <assert.h> #include <unistd.h> @@ -25,6 +24,7 @@ #include <sys/stat.h> #include <limits.h> #include <bpf/libbpf.h> +#include <tools/dis-asm-compat.h> #include "json_writer.h" #include "main.h" @@ -40,15 +40,14 @@ static void get_exec_path(char *tpath, size_t size) } static int oper_count; -static int fprintf_json(void *out, const char *fmt, ...) +static int printf_json(void *out, const char *fmt, va_list ap) { - va_list ap; char *s; + int err; - va_start(ap, fmt); - if (vasprintf(&s, fmt, ap) < 0) + err = vasprintf(&s, fmt, ap); + if (err < 0) return -1; - va_end(ap); if (!oper_count) { int i; @@ -72,6 +71,32 @@ static int fprintf_json(void *out, const char *fmt, ...) return 0; } +static int fprintf_json(void *out, const char *fmt, ...) +{ + va_list ap; + int r; + + va_start(ap, fmt); + r = printf_json(out, fmt, ap); + va_end(ap); + + return r; +} + +static int fprintf_json_styled(void *out, + enum disassembler_style style __maybe_unused, + const char *fmt, ...) +{ + va_list ap; + int r; + + va_start(ap, fmt); + r = printf_json(out, fmt, ap); + va_end(ap); + + return r; +} + void disasm_print_insn(unsigned char *image, ssize_t len, int opcodes, const char *arch, const char *disassembler_options, const struct btf *btf, @@ -98,11 +123,13 @@ void disasm_print_insn(unsigned char *image, ssize_t len, int opcodes, assert(bfd_check_format(bfdf, bfd_object)); if (json_output) - init_disassemble_info(&info, stdout, - (fprintf_ftype) fprintf_json); + init_disassemble_info_compat(&info, stdout, + (fprintf_ftype) fprintf_json, + fprintf_json_styled); else - init_disassemble_info(&info, stdout, - (fprintf_ftype) fprintf); + init_disassemble_info_compat(&info, stdout, + (fprintf_ftype) fprintf, + fprintf_styled); /* Update architecture info for offload. */ if (arch) { diff --git a/tools/bpf/bpftool/json_writer.c b/tools/bpf/bpftool/json_writer.c index 86501cd3c763..7fea83bedf48 100644 --- a/tools/bpf/bpftool/json_writer.c +++ b/tools/bpf/bpftool/json_writer.c @@ -119,6 +119,12 @@ void jsonw_pretty(json_writer_t *self, bool on) self->pretty = on; } +void jsonw_reset(json_writer_t *self) +{ + assert(self->depth == 0); + self->sep = '\0'; +} + /* Basic blocks */ static void jsonw_begin(json_writer_t *self, int c) { diff --git a/tools/bpf/bpftool/json_writer.h b/tools/bpf/bpftool/json_writer.h index 35cf1f00f96c..8ace65cdb92f 100644 --- a/tools/bpf/bpftool/json_writer.h +++ b/tools/bpf/bpftool/json_writer.h @@ -27,6 +27,9 @@ void jsonw_destroy(json_writer_t **self_p); /* Cause output to have pretty whitespace */ void jsonw_pretty(json_writer_t *self, bool on); +/* Reset separator to create new JSON */ +void jsonw_reset(json_writer_t *self); + /* Add property name */ void jsonw_name(json_writer_t *self, const char *name); diff --git a/tools/bpf/bpftool/link.c b/tools/bpf/bpftool/link.c new file mode 100644 index 000000000000..2863639706dd --- /dev/null +++ b/tools/bpf/bpftool/link.c @@ -0,0 +1,495 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +/* Copyright (C) 2020 Facebook */ + +#include <errno.h> +#include <linux/err.h> +#include <net/if.h> +#include <stdio.h> +#include <unistd.h> + +#include <bpf/bpf.h> +#include <bpf/hashmap.h> + +#include "json_writer.h" +#include "main.h" + +static struct hashmap *link_table; + +static int link_parse_fd(int *argc, char ***argv) +{ + int fd; + + if (is_prefix(**argv, "id")) { + unsigned int id; + char *endptr; + + NEXT_ARGP(); + + id = strtoul(**argv, &endptr, 0); + if (*endptr) { + p_err("can't parse %s as ID", **argv); + return -1; + } + NEXT_ARGP(); + + fd = bpf_link_get_fd_by_id(id); + if (fd < 0) + p_err("failed to get link with ID %d: %s", id, strerror(errno)); + return fd; + } else if (is_prefix(**argv, "pinned")) { + char *path; + + NEXT_ARGP(); + + path = **argv; + NEXT_ARGP(); + + return open_obj_pinned_any(path, BPF_OBJ_LINK); + } + + p_err("expected 'id' or 'pinned', got: '%s'?", **argv); + return -1; +} + +static void +show_link_header_json(struct bpf_link_info *info, json_writer_t *wtr) +{ + const char *link_type_str; + + jsonw_uint_field(wtr, "id", info->id); + link_type_str = libbpf_bpf_link_type_str(info->type); + if (link_type_str) + jsonw_string_field(wtr, "type", link_type_str); + else + jsonw_uint_field(wtr, "type", info->type); + + jsonw_uint_field(json_wtr, "prog_id", info->prog_id); +} + +static void show_link_attach_type_json(__u32 attach_type, json_writer_t *wtr) +{ + const char *attach_type_str; + + attach_type_str = libbpf_bpf_attach_type_str(attach_type); + if (attach_type_str) + jsonw_string_field(wtr, "attach_type", attach_type_str); + else + jsonw_uint_field(wtr, "attach_type", attach_type); +} + +static bool is_iter_map_target(const char *target_name) +{ + return strcmp(target_name, "bpf_map_elem") == 0 || + strcmp(target_name, "bpf_sk_storage_map") == 0; +} + +static bool is_iter_cgroup_target(const char *target_name) +{ + return strcmp(target_name, "cgroup") == 0; +} + +static const char *cgroup_order_string(__u32 order) +{ + switch (order) { + case BPF_CGROUP_ITER_ORDER_UNSPEC: + return "order_unspec"; + case BPF_CGROUP_ITER_SELF_ONLY: + return "self_only"; + case BPF_CGROUP_ITER_DESCENDANTS_PRE: + return "descendants_pre"; + case BPF_CGROUP_ITER_DESCENDANTS_POST: + return "descendants_post"; + case BPF_CGROUP_ITER_ANCESTORS_UP: + return "ancestors_up"; + default: /* won't happen */ + return "unknown"; + } +} + +static bool is_iter_task_target(const char *target_name) +{ + return strcmp(target_name, "task") == 0 || + strcmp(target_name, "task_file") == 0 || + strcmp(target_name, "task_vma") == 0; +} + +static void show_iter_json(struct bpf_link_info *info, json_writer_t *wtr) +{ + const char *target_name = u64_to_ptr(info->iter.target_name); + + jsonw_string_field(wtr, "target_name", target_name); + + if (is_iter_map_target(target_name)) + jsonw_uint_field(wtr, "map_id", info->iter.map.map_id); + else if (is_iter_task_target(target_name)) { + if (info->iter.task.tid) + jsonw_uint_field(wtr, "tid", info->iter.task.tid); + else if (info->iter.task.pid) + jsonw_uint_field(wtr, "pid", info->iter.task.pid); + } + + if (is_iter_cgroup_target(target_name)) { + jsonw_lluint_field(wtr, "cgroup_id", info->iter.cgroup.cgroup_id); + jsonw_string_field(wtr, "order", + cgroup_order_string(info->iter.cgroup.order)); + } +} + +static int get_prog_info(int prog_id, struct bpf_prog_info *info) +{ + __u32 len = sizeof(*info); + int err, prog_fd; + + prog_fd = bpf_prog_get_fd_by_id(prog_id); + if (prog_fd < 0) + return prog_fd; + + memset(info, 0, sizeof(*info)); + err = bpf_obj_get_info_by_fd(prog_fd, info, &len); + if (err) + p_err("can't get prog info: %s", strerror(errno)); + close(prog_fd); + return err; +} + +static int show_link_close_json(int fd, struct bpf_link_info *info) +{ + struct bpf_prog_info prog_info; + const char *prog_type_str; + int err; + + jsonw_start_object(json_wtr); + + show_link_header_json(info, json_wtr); + + switch (info->type) { + case BPF_LINK_TYPE_RAW_TRACEPOINT: + jsonw_string_field(json_wtr, "tp_name", + u64_to_ptr(info->raw_tracepoint.tp_name)); + break; + case BPF_LINK_TYPE_TRACING: + err = get_prog_info(info->prog_id, &prog_info); + if (err) + return err; + + prog_type_str = libbpf_bpf_prog_type_str(prog_info.type); + /* libbpf will return NULL for variants unknown to it. */ + if (prog_type_str) + jsonw_string_field(json_wtr, "prog_type", prog_type_str); + else + jsonw_uint_field(json_wtr, "prog_type", prog_info.type); + + show_link_attach_type_json(info->tracing.attach_type, + json_wtr); + break; + case BPF_LINK_TYPE_CGROUP: + jsonw_lluint_field(json_wtr, "cgroup_id", + info->cgroup.cgroup_id); + show_link_attach_type_json(info->cgroup.attach_type, json_wtr); + break; + case BPF_LINK_TYPE_ITER: + show_iter_json(info, json_wtr); + break; + case BPF_LINK_TYPE_NETNS: + jsonw_uint_field(json_wtr, "netns_ino", + info->netns.netns_ino); + show_link_attach_type_json(info->netns.attach_type, json_wtr); + break; + default: + break; + } + + if (!hashmap__empty(link_table)) { + struct hashmap_entry *entry; + + jsonw_name(json_wtr, "pinned"); + jsonw_start_array(json_wtr); + hashmap__for_each_key_entry(link_table, entry, + u32_as_hash_field(info->id)) + jsonw_string(json_wtr, entry->value); + jsonw_end_array(json_wtr); + } + + emit_obj_refs_json(refs_table, info->id, json_wtr); + + jsonw_end_object(json_wtr); + + return 0; +} + +static void show_link_header_plain(struct bpf_link_info *info) +{ + const char *link_type_str; + + printf("%u: ", info->id); + link_type_str = libbpf_bpf_link_type_str(info->type); + if (link_type_str) + printf("%s ", link_type_str); + else + printf("type %u ", info->type); + + printf("prog %u ", info->prog_id); +} + +static void show_link_attach_type_plain(__u32 attach_type) +{ + const char *attach_type_str; + + attach_type_str = libbpf_bpf_attach_type_str(attach_type); + if (attach_type_str) + printf("attach_type %s ", attach_type_str); + else + printf("attach_type %u ", attach_type); +} + +static void show_iter_plain(struct bpf_link_info *info) +{ + const char *target_name = u64_to_ptr(info->iter.target_name); + + printf("target_name %s ", target_name); + + if (is_iter_map_target(target_name)) + printf("map_id %u ", info->iter.map.map_id); + else if (is_iter_task_target(target_name)) { + if (info->iter.task.tid) + printf("tid %u ", info->iter.task.tid); + else if (info->iter.task.pid) + printf("pid %u ", info->iter.task.pid); + } + + if (is_iter_cgroup_target(target_name)) { + printf("cgroup_id %llu ", info->iter.cgroup.cgroup_id); + printf("order %s ", + cgroup_order_string(info->iter.cgroup.order)); + } +} + +static int show_link_close_plain(int fd, struct bpf_link_info *info) +{ + struct bpf_prog_info prog_info; + const char *prog_type_str; + int err; + + show_link_header_plain(info); + + switch (info->type) { + case BPF_LINK_TYPE_RAW_TRACEPOINT: + printf("\n\ttp '%s' ", + (const char *)u64_to_ptr(info->raw_tracepoint.tp_name)); + break; + case BPF_LINK_TYPE_TRACING: + err = get_prog_info(info->prog_id, &prog_info); + if (err) + return err; + + prog_type_str = libbpf_bpf_prog_type_str(prog_info.type); + /* libbpf will return NULL for variants unknown to it. */ + if (prog_type_str) + printf("\n\tprog_type %s ", prog_type_str); + else + printf("\n\tprog_type %u ", prog_info.type); + + show_link_attach_type_plain(info->tracing.attach_type); + break; + case BPF_LINK_TYPE_CGROUP: + printf("\n\tcgroup_id %zu ", (size_t)info->cgroup.cgroup_id); + show_link_attach_type_plain(info->cgroup.attach_type); + break; + case BPF_LINK_TYPE_ITER: + show_iter_plain(info); + break; + case BPF_LINK_TYPE_NETNS: + printf("\n\tnetns_ino %u ", info->netns.netns_ino); + show_link_attach_type_plain(info->netns.attach_type); + break; + default: + break; + } + + if (!hashmap__empty(link_table)) { + struct hashmap_entry *entry; + + hashmap__for_each_key_entry(link_table, entry, + u32_as_hash_field(info->id)) + printf("\n\tpinned %s", (char *)entry->value); + } + emit_obj_refs_plain(refs_table, info->id, "\n\tpids "); + + printf("\n"); + + return 0; +} + +static int do_show_link(int fd) +{ + struct bpf_link_info info; + __u32 len = sizeof(info); + char buf[256]; + int err; + + memset(&info, 0, sizeof(info)); +again: + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err) { + p_err("can't get link info: %s", + strerror(errno)); + close(fd); + return err; + } + if (info.type == BPF_LINK_TYPE_RAW_TRACEPOINT && + !info.raw_tracepoint.tp_name) { + info.raw_tracepoint.tp_name = (unsigned long)&buf; + info.raw_tracepoint.tp_name_len = sizeof(buf); + goto again; + } + if (info.type == BPF_LINK_TYPE_ITER && + !info.iter.target_name) { + info.iter.target_name = (unsigned long)&buf; + info.iter.target_name_len = sizeof(buf); + goto again; + } + + if (json_output) + show_link_close_json(fd, &info); + else + show_link_close_plain(fd, &info); + + close(fd); + return 0; +} + +static int do_show(int argc, char **argv) +{ + __u32 id = 0; + int err, fd; + + if (show_pinned) { + link_table = hashmap__new(hash_fn_for_key_as_id, + equal_fn_for_key_as_id, NULL); + if (IS_ERR(link_table)) { + p_err("failed to create hashmap for pinned paths"); + return -1; + } + build_pinned_obj_table(link_table, BPF_OBJ_LINK); + } + build_obj_refs_table(&refs_table, BPF_OBJ_LINK); + + if (argc == 2) { + fd = link_parse_fd(&argc, &argv); + if (fd < 0) + return fd; + return do_show_link(fd); + } + + if (argc) + return BAD_ARG(); + + if (json_output) + jsonw_start_array(json_wtr); + while (true) { + err = bpf_link_get_next_id(id, &id); + if (err) { + if (errno == ENOENT) + break; + p_err("can't get next link: %s%s", strerror(errno), + errno == EINVAL ? " -- kernel too old?" : ""); + break; + } + + fd = bpf_link_get_fd_by_id(id); + if (fd < 0) { + if (errno == ENOENT) + continue; + p_err("can't get link by id (%u): %s", + id, strerror(errno)); + break; + } + + err = do_show_link(fd); + if (err) + break; + } + if (json_output) + jsonw_end_array(json_wtr); + + delete_obj_refs_table(refs_table); + + if (show_pinned) + delete_pinned_obj_table(link_table); + + return errno == ENOENT ? 0 : -1; +} + +static int do_pin(int argc, char **argv) +{ + int err; + + err = do_pin_any(argc, argv, link_parse_fd); + if (!err && json_output) + jsonw_null(json_wtr); + return err; +} + +static int do_detach(int argc, char **argv) +{ + int err, fd; + + if (argc != 2) { + p_err("link specifier is invalid or missing\n"); + return 1; + } + + fd = link_parse_fd(&argc, &argv); + if (fd < 0) + return 1; + + err = bpf_link_detach(fd); + if (err) + err = -errno; + close(fd); + if (err) { + p_err("failed link detach: %s", strerror(-err)); + return 1; + } + + if (json_output) + jsonw_null(json_wtr); + + return 0; +} + +static int do_help(int argc, char **argv) +{ + if (json_output) { + jsonw_null(json_wtr); + return 0; + } + + fprintf(stderr, + "Usage: %1$s %2$s { show | list } [LINK]\n" + " %1$s %2$s pin LINK FILE\n" + " %1$s %2$s detach LINK\n" + " %1$s %2$s help\n" + "\n" + " " HELP_SPEC_LINK "\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-f|--bpffs} | {-n|--nomount} }\n" + "", + bin_name, argv[-2]); + + return 0; +} + +static const struct cmd cmds[] = { + { "show", do_show }, + { "list", do_show }, + { "help", do_help }, + { "pin", do_pin }, + { "detach", do_detach }, + { 0 } +}; + +int do_link(int argc, char **argv) +{ + return cmd_select(cmds, argc, argv, do_help); +} diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c index 6d41bbfc6459..ccd7457f92bf 100644 --- a/tools/bpf/bpftool/main.c +++ b/tools/bpf/bpftool/main.c @@ -10,6 +10,8 @@ #include <string.h> #include <bpf/bpf.h> +#include <bpf/btf.h> +#include <bpf/hashmap.h> #include <bpf/libbpf.h> #include "main.h" @@ -28,8 +30,10 @@ bool show_pinned; bool block_mount; bool verifier_logs; bool relaxed_maps; -struct pinned_obj_table prog_table; -struct pinned_obj_table map_table; +bool use_loader; +bool legacy_libbpf; +struct btf *base_btf; +struct hashmap *refs_table; static void __noreturn clean_and_exit(int i) { @@ -58,34 +62,87 @@ static int do_help(int argc, char **argv) " %s batch file FILE\n" " %s version\n" "\n" - " OBJECT := { prog | map | cgroup | perf | net | feature | btf | gen }\n" - " " HELP_SPEC_OPTIONS "\n" + " OBJECT := { prog | map | link | cgroup | perf | net | feature | btf | gen | struct_ops | iter }\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-V|--version} }\n" "", bin_name, bin_name, bin_name); return 0; } +#ifndef BPFTOOL_VERSION +/* bpftool's major and minor version numbers are aligned on libbpf's. There is + * an offset of 6 for the version number, because bpftool's version was higher + * than libbpf's when we adopted this scheme. The patch number remains at 0 + * for now. Set BPFTOOL_VERSION to override. + */ +#define BPFTOOL_MAJOR_VERSION (LIBBPF_MAJOR_VERSION + 6) +#define BPFTOOL_MINOR_VERSION LIBBPF_MINOR_VERSION +#define BPFTOOL_PATCH_VERSION 0 +#endif + static int do_version(int argc, char **argv) { +#ifdef HAVE_LIBBFD_SUPPORT + const bool has_libbfd = true; +#else + const bool has_libbfd = false; +#endif +#ifdef BPFTOOL_WITHOUT_SKELETONS + const bool has_skeletons = false; +#else + const bool has_skeletons = true; +#endif + if (json_output) { - jsonw_start_object(json_wtr); + jsonw_start_object(json_wtr); /* root object */ + jsonw_name(json_wtr, "version"); +#ifdef BPFTOOL_VERSION jsonw_printf(json_wtr, "\"%s\"", BPFTOOL_VERSION); - jsonw_end_object(json_wtr); +#else + jsonw_printf(json_wtr, "\"%d.%d.%d\"", BPFTOOL_MAJOR_VERSION, + BPFTOOL_MINOR_VERSION, BPFTOOL_PATCH_VERSION); +#endif + jsonw_name(json_wtr, "libbpf_version"); + jsonw_printf(json_wtr, "\"%d.%d\"", + libbpf_major_version(), libbpf_minor_version()); + + jsonw_name(json_wtr, "features"); + jsonw_start_object(json_wtr); /* features */ + jsonw_bool_field(json_wtr, "libbfd", has_libbfd); + jsonw_bool_field(json_wtr, "libbpf_strict", !legacy_libbpf); + jsonw_bool_field(json_wtr, "skeletons", has_skeletons); + jsonw_end_object(json_wtr); /* features */ + + jsonw_end_object(json_wtr); /* root object */ } else { + unsigned int nb_features = 0; + +#ifdef BPFTOOL_VERSION printf("%s v%s\n", bin_name, BPFTOOL_VERSION); +#else + printf("%s v%d.%d.%d\n", bin_name, BPFTOOL_MAJOR_VERSION, + BPFTOOL_MINOR_VERSION, BPFTOOL_PATCH_VERSION); +#endif + printf("using libbpf %s\n", libbpf_version_string()); + printf("features:"); + if (has_libbfd) { + printf(" libbfd"); + nb_features++; + } + if (!legacy_libbpf) { + printf("%s libbpf_strict", nb_features++ ? "," : ""); + nb_features++; + } + if (has_skeletons) + printf("%s skeletons", nb_features++ ? "," : ""); + printf("\n"); } return 0; } -static int __printf(2, 0) -print_all_levels(__maybe_unused enum libbpf_print_level level, - const char *format, va_list args) -{ - return vfprintf(stderr, format, args); -} - int cmd_select(const struct cmd *cmds, int argc, char **argv, int (*help)(int argc, char **argv)) { @@ -98,9 +155,16 @@ int cmd_select(const struct cmd *cmds, int argc, char **argv, if (argc < 1 && cmds[0].func) return cmds[0].func(argc, argv); - for (i = 0; cmds[i].func; i++) - if (is_prefix(*argv, cmds[i].cmd)) + for (i = 0; cmds[i].cmd; i++) { + if (is_prefix(*argv, cmds[i].cmd)) { + if (!cmds[i].func) { + p_err("command '%s' is not supported in bootstrap mode", + cmds[i].cmd); + return -1; + } return cmds[i].func(argc - 1, argv + 1); + } + } help(argc - 1, argv + 1); @@ -222,12 +286,15 @@ static const struct cmd cmds[] = { { "batch", do_batch }, { "prog", do_prog }, { "map", do_map }, + { "link", do_link }, { "cgroup", do_cgroup }, { "perf", do_perf }, { "net", do_net }, { "feature", do_feature }, { "btf", do_btf }, { "gen", do_gen }, + { "struct_ops", do_struct_ops }, + { "iter", do_iter }, { "version", do_version }, { 0 } }; @@ -240,7 +307,7 @@ static int do_batch(int argc, char **argv) int n_argc; FILE *fp; char *cp; - int err; + int err = 0; int i; if (argc < 2) { @@ -304,8 +371,10 @@ static int do_batch(int argc, char **argv) n_argc = make_args(buf, n_argv, BATCH_ARG_NB_MAX, lines); if (!n_argc) continue; - if (n_argc < 0) + if (n_argc < 0) { + err = n_argc; goto err_close; + } if (json_output) { jsonw_start_object(json_wtr); @@ -334,7 +403,6 @@ static int do_batch(int argc, char **argv) } else { if (!json_output) printf("processed %d commands\n", lines); - err = 0; } err_close: if (fp != stdin) @@ -357,10 +425,26 @@ int main(int argc, char **argv) { "mapcompat", no_argument, NULL, 'm' }, { "nomount", no_argument, NULL, 'n' }, { "debug", no_argument, NULL, 'd' }, + { "use-loader", no_argument, NULL, 'L' }, + { "base-btf", required_argument, NULL, 'B' }, + { "legacy", no_argument, NULL, 'l' }, { 0 } }; + bool version_requested = false; int opt, ret; + setlinebuf(stdout); + +#ifdef USE_LIBCAP + /* Libcap < 2.63 hooks before main() to compute the number of + * capabilities of the running kernel, and doing so it calls prctl() + * which may fail and set errno to non-zero. + * Let's reset errno to make sure this does not interfere with the + * batch mode. + */ + errno = 0; +#endif + last_do_help = do_help; pretty_output = false; json_output = false; @@ -368,15 +452,13 @@ int main(int argc, char **argv) block_mount = false; bin_name = argv[0]; - hash_init(prog_table.table); - hash_init(map_table.table); - opterr = 0; - while ((opt = getopt_long(argc, argv, "Vhpjfmnd", + while ((opt = getopt_long(argc, argv, "VhpjfLmndB:l", options, NULL)) >= 0) { switch (opt) { case 'V': - return do_version(argc, argv); + version_requested = true; + break; case 'h': return do_help(argc, argv); case 'p': @@ -406,6 +488,21 @@ int main(int argc, char **argv) libbpf_set_print(print_all_levels); verifier_logs = true; break; + case 'B': + base_btf = btf__parse(optarg, NULL); + if (libbpf_get_error(base_btf)) { + p_err("failed to parse base BTF at '%s': %ld\n", + optarg, libbpf_get_error(base_btf)); + base_btf = NULL; + return -1; + } + break; + case 'L': + use_loader = true; + break; + case 'l': + legacy_libbpf = true; + break; default: p_err("unrecognized option '%s'", argv[optind - 1]); if (json_output) @@ -415,20 +512,28 @@ int main(int argc, char **argv) } } + if (!legacy_libbpf) { + /* Allow legacy map definitions for skeleton generation. + * It will still be rejected if users use LIBBPF_STRICT_ALL + * mode for loading generated skeleton. + */ + libbpf_set_strict_mode(LIBBPF_STRICT_ALL & ~LIBBPF_STRICT_MAP_DEFINITIONS); + } + argc -= optind; argv += optind; if (argc < 0) usage(); + if (version_requested) + return do_version(argc, argv); + ret = cmd_select(cmds, argc, argv, do_help); if (json_output) jsonw_destroy(&json_wtr); - if (show_pinned) { - delete_pinned_obj_table(&prog_table); - delete_pinned_obj_table(&map_table); - } + btf__free(base_btf); return ret; } diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h index 4e75b58d3989..5e5060c2ac04 100644 --- a/tools/bpf/bpftool/main.h +++ b/tools/bpf/bpftool/main.h @@ -8,15 +8,28 @@ #undef GCC_VERSION #include <stdbool.h> #include <stdio.h> +#include <stdlib.h> #include <linux/bpf.h> #include <linux/compiler.h> #include <linux/kernel.h> -#include <linux/hashtable.h> -#include <tools/libc_compat.h> + +#include <bpf/hashmap.h> +#include <bpf/libbpf.h> #include "json_writer.h" -#define ptr_to_u64(ptr) ((__u64)(unsigned long)(ptr)) +/* Make sure we do not use kernel-only integer typedefs */ +#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64 + +static inline __u64 ptr_to_u64(const void *ptr) +{ + return (__u64)(unsigned long)ptr; +} + +static inline void *u64_to_ptr(__u64 ptr) +{ + return (void *)(unsigned long)ptr; +} #define NEXT_ARG() ({ argc--; argv++; if (argc < 0) usage(); }) #define NEXT_ARGP() ({ (*argc)--; (*argv)++; if (*argc < 0) usage(); }) @@ -44,47 +57,19 @@ #define HELP_SPEC_PROGRAM \ "PROG := { id PROG_ID | pinned FILE | tag PROG_TAG | name PROG_NAME }" #define HELP_SPEC_OPTIONS \ - "OPTIONS := { {-j|--json} [{-p|--pretty}] | {-f|--bpffs} |\n" \ - "\t {-m|--mapcompat} | {-n|--nomount} }" + "OPTIONS := { {-j|--json} [{-p|--pretty}] | {-d|--debug} | {-l|--legacy}" #define HELP_SPEC_MAP \ "MAP := { id MAP_ID | pinned FILE | name MAP_NAME }" +#define HELP_SPEC_LINK \ + "LINK := { id LINK_ID | pinned FILE }" -static const char * const prog_type_name[] = { - [BPF_PROG_TYPE_UNSPEC] = "unspec", - [BPF_PROG_TYPE_SOCKET_FILTER] = "socket_filter", - [BPF_PROG_TYPE_KPROBE] = "kprobe", - [BPF_PROG_TYPE_SCHED_CLS] = "sched_cls", - [BPF_PROG_TYPE_SCHED_ACT] = "sched_act", - [BPF_PROG_TYPE_TRACEPOINT] = "tracepoint", - [BPF_PROG_TYPE_XDP] = "xdp", - [BPF_PROG_TYPE_PERF_EVENT] = "perf_event", - [BPF_PROG_TYPE_CGROUP_SKB] = "cgroup_skb", - [BPF_PROG_TYPE_CGROUP_SOCK] = "cgroup_sock", - [BPF_PROG_TYPE_LWT_IN] = "lwt_in", - [BPF_PROG_TYPE_LWT_OUT] = "lwt_out", - [BPF_PROG_TYPE_LWT_XMIT] = "lwt_xmit", - [BPF_PROG_TYPE_SOCK_OPS] = "sock_ops", - [BPF_PROG_TYPE_SK_SKB] = "sk_skb", - [BPF_PROG_TYPE_CGROUP_DEVICE] = "cgroup_device", - [BPF_PROG_TYPE_SK_MSG] = "sk_msg", - [BPF_PROG_TYPE_RAW_TRACEPOINT] = "raw_tracepoint", - [BPF_PROG_TYPE_CGROUP_SOCK_ADDR] = "cgroup_sock_addr", - [BPF_PROG_TYPE_LWT_SEG6LOCAL] = "lwt_seg6local", - [BPF_PROG_TYPE_LIRC_MODE2] = "lirc_mode2", - [BPF_PROG_TYPE_SK_REUSEPORT] = "sk_reuseport", - [BPF_PROG_TYPE_FLOW_DISSECTOR] = "flow_dissector", - [BPF_PROG_TYPE_CGROUP_SYSCTL] = "cgroup_sysctl", - [BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE] = "raw_tracepoint_writable", - [BPF_PROG_TYPE_CGROUP_SOCKOPT] = "cgroup_sockopt", -}; - -extern const char * const map_type_name[]; -extern const size_t map_type_name_size; - +/* keep in sync with the definition in skeleton/pid_iter.bpf.c */ enum bpf_obj_type { BPF_OBJ_UNKNOWN, BPF_OBJ_PROG, BPF_OBJ_MAP, + BPF_OBJ_LINK, + BPF_OBJ_BTF, }; extern const char *bin_name; @@ -92,11 +77,14 @@ extern const char *bin_name; extern json_writer_t *json_wtr; extern bool json_output; extern bool show_pinned; +extern bool show_pids; extern bool block_mount; extern bool verifier_logs; extern bool relaxed_maps; -extern struct pinned_obj_table prog_table; -extern struct pinned_obj_table map_table; +extern bool use_loader; +extern bool legacy_libbpf; +extern struct btf *base_btf; +extern struct hashmap *refs_table; void __printf(1, 2) p_err(const char *fmt, ...); void __printf(1, 2) p_info(const char *fmt, ...); @@ -110,22 +98,31 @@ void set_max_rlimit(void); int mount_tracefs(const char *target); -struct pinned_obj_table { - DECLARE_HASHTABLE(table, 16); +struct obj_ref { + int pid; + char comm[16]; }; -struct pinned_obj { - __u32 id; - char *path; - struct hlist_node hash; +struct obj_refs { + int ref_cnt; + bool has_bpf_cookie; + struct obj_ref *refs; + __u64 bpf_cookie; }; struct btf; struct bpf_line_info; -int build_pinned_obj_table(struct pinned_obj_table *table, +int build_pinned_obj_table(struct hashmap *table, enum bpf_obj_type type); -void delete_pinned_obj_table(struct pinned_obj_table *tab); +void delete_pinned_obj_table(struct hashmap *table); +__weak int build_obj_refs_table(struct hashmap **table, + enum bpf_obj_type type); +__weak void delete_obj_refs_table(struct hashmap *table); +__weak void emit_obj_refs_json(struct hashmap *table, __u32 id, + json_writer_t *json_wtr); +__weak void emit_obj_refs_plain(struct hashmap *table, __u32 id, + const char *prefix); void print_dev_plain(__u32 ifindex, __u64 ns_dev, __u64 ns_inode); void print_dev_json(__u32 ifindex, __u64 ns_dev, __u64 ns_inode); @@ -137,29 +134,41 @@ struct cmd { int cmd_select(const struct cmd *cmds, int argc, char **argv, int (*help)(int argc, char **argv)); +#define MAX_PROG_FULL_NAME 128 +void get_prog_full_name(const struct bpf_prog_info *prog_info, int prog_fd, + char *name_buff, size_t buff_len); + int get_fd_type(int fd); const char *get_fd_type_name(enum bpf_obj_type type); char *get_fdinfo(int fd, const char *key); -int open_obj_pinned(char *path, bool quiet); -int open_obj_pinned_any(char *path, enum bpf_obj_type exp_type); +int open_obj_pinned(const char *path, bool quiet); +int open_obj_pinned_any(const char *path, enum bpf_obj_type exp_type); int mount_bpffs_for_pin(const char *name); -int do_pin_any(int argc, char **argv, int (*get_fd_by_id)(__u32)); +int do_pin_any(int argc, char **argv, int (*get_fd_by_id)(int *, char ***)); int do_pin_fd(int fd, const char *name); -int do_prog(int argc, char **arg); -int do_map(int argc, char **arg); -int do_event_pipe(int argc, char **argv); -int do_cgroup(int argc, char **arg); -int do_perf(int argc, char **arg); -int do_net(int argc, char **arg); -int do_tracelog(int argc, char **arg); -int do_feature(int argc, char **argv); -int do_btf(int argc, char **argv); +/* commands available in bootstrap mode */ int do_gen(int argc, char **argv); +int do_btf(int argc, char **argv); + +/* non-bootstrap only commands */ +int do_prog(int argc, char **arg) __weak; +int do_map(int argc, char **arg) __weak; +int do_link(int argc, char **arg) __weak; +int do_event_pipe(int argc, char **argv) __weak; +int do_cgroup(int argc, char **arg) __weak; +int do_perf(int argc, char **arg) __weak; +int do_net(int argc, char **arg) __weak; +int do_tracelog(int argc, char **arg) __weak; +int do_feature(int argc, char **argv) __weak; +int do_struct_ops(int argc, char **argv) __weak; +int do_iter(int argc, char **argv) __weak; int parse_u32_arg(int *argc, char ***argv, __u32 *val, const char *what); int prog_parse_fd(int *argc, char ***argv); +int prog_parse_fds(int *argc, char ***argv, int **fds); int map_parse_fd(int *argc, char ***argv); +int map_parse_fds(int *argc, char ***argv, int **fds); int map_parse_fd_and_info(int *argc, char ***argv, void *info, __u32 *info_len); struct bpf_prog_linfo; @@ -200,6 +209,7 @@ struct btf_dumper { const struct btf *btf; json_writer_t *jw; bool is_plain_text; + bool prog_id_as_func_ptr; }; /* btf_dumper_type - print data along with type information @@ -226,4 +236,40 @@ struct tcmsg; int do_xdp_dump(struct ifinfomsg *ifinfo, struct nlattr **tb); int do_filter_dump(struct tcmsg *ifinfo, struct nlattr **tb, const char *kind, const char *devname, int ifindex); + +int print_all_levels(__maybe_unused enum libbpf_print_level level, + const char *format, va_list args); + +size_t hash_fn_for_key_as_id(const void *key, void *ctx); +bool equal_fn_for_key_as_id(const void *k1, const void *k2, void *ctx); + +/* bpf_attach_type_input_str - convert the provided attach type value into a + * textual representation that we accept for input purposes. + * + * This function is similar in nature to libbpf_bpf_attach_type_str, but + * recognizes some attach type names that have been used by the program in the + * past and which do not follow the string inference scheme that libbpf uses. + * These textual representations should only be used for user input. + * + * @t: The attach type + * Returns a pointer to a static string identifying the attach type. NULL is + * returned for unknown bpf_attach_type values. + */ +const char *bpf_attach_type_input_str(enum bpf_attach_type t); + +static inline void *u32_as_hash_field(__u32 x) +{ + return (void *)(uintptr_t)x; +} + +static inline __u32 hash_field_as_u32(const void *x) +{ + return (__u32)(uintptr_t)x; +} + +static inline bool hashmap__empty(struct hashmap *map) +{ + return map ? hashmap__size(map) == 0 : true; +} + #endif diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c index e6c85680b34d..9a6ca9f31133 100644 --- a/tools/bpf/bpftool/map.c +++ b/tools/bpf/bpftool/map.c @@ -17,41 +17,12 @@ #include <bpf/bpf.h> #include <bpf/btf.h> +#include <bpf/hashmap.h> #include "json_writer.h" #include "main.h" -const char * const map_type_name[] = { - [BPF_MAP_TYPE_UNSPEC] = "unspec", - [BPF_MAP_TYPE_HASH] = "hash", - [BPF_MAP_TYPE_ARRAY] = "array", - [BPF_MAP_TYPE_PROG_ARRAY] = "prog_array", - [BPF_MAP_TYPE_PERF_EVENT_ARRAY] = "perf_event_array", - [BPF_MAP_TYPE_PERCPU_HASH] = "percpu_hash", - [BPF_MAP_TYPE_PERCPU_ARRAY] = "percpu_array", - [BPF_MAP_TYPE_STACK_TRACE] = "stack_trace", - [BPF_MAP_TYPE_CGROUP_ARRAY] = "cgroup_array", - [BPF_MAP_TYPE_LRU_HASH] = "lru_hash", - [BPF_MAP_TYPE_LRU_PERCPU_HASH] = "lru_percpu_hash", - [BPF_MAP_TYPE_LPM_TRIE] = "lpm_trie", - [BPF_MAP_TYPE_ARRAY_OF_MAPS] = "array_of_maps", - [BPF_MAP_TYPE_HASH_OF_MAPS] = "hash_of_maps", - [BPF_MAP_TYPE_DEVMAP] = "devmap", - [BPF_MAP_TYPE_DEVMAP_HASH] = "devmap_hash", - [BPF_MAP_TYPE_SOCKMAP] = "sockmap", - [BPF_MAP_TYPE_CPUMAP] = "cpumap", - [BPF_MAP_TYPE_XSKMAP] = "xskmap", - [BPF_MAP_TYPE_SOCKHASH] = "sockhash", - [BPF_MAP_TYPE_CGROUP_STORAGE] = "cgroup_storage", - [BPF_MAP_TYPE_REUSEPORT_SOCKARRAY] = "reuseport_sockarray", - [BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE] = "percpu_cgroup_storage", - [BPF_MAP_TYPE_QUEUE] = "queue", - [BPF_MAP_TYPE_STACK] = "stack", - [BPF_MAP_TYPE_SK_STORAGE] = "sk_storage", - [BPF_MAP_TYPE_STRUCT_OPS] = "struct_ops", -}; - -const size_t map_type_name_size = ARRAY_SIZE(map_type_name); +static struct hashmap *map_table; static bool map_is_per_cpu(__u32 type) { @@ -74,12 +45,18 @@ static bool map_is_map_of_progs(__u32 type) static int map_type_from_str(const char *type) { + const char *map_type_str; unsigned int i; - for (i = 0; i < ARRAY_SIZE(map_type_name); i++) + for (i = 0; ; i++) { + map_type_str = libbpf_bpf_map_type_str(i); + if (!map_type_str) + break; + /* Don't allow prefixing in case of possible future shadowing */ - if (map_type_name[i] && !strcmp(map_type_name[i], type)) + if (!strcmp(map_type_str, type)) return i; + } return -1; } @@ -92,168 +69,12 @@ static void *alloc_value(struct bpf_map_info *info) return malloc(info->value_size); } -static int map_fd_by_name(char *name, int **fds) -{ - unsigned int id = 0; - int fd, nb_fds = 0; - void *tmp; - int err; - - while (true) { - struct bpf_map_info info = {}; - __u32 len = sizeof(info); - - err = bpf_map_get_next_id(id, &id); - if (err) { - if (errno != ENOENT) { - p_err("%s", strerror(errno)); - goto err_close_fds; - } - return nb_fds; - } - - fd = bpf_map_get_fd_by_id(id); - if (fd < 0) { - p_err("can't get map by id (%u): %s", - id, strerror(errno)); - goto err_close_fds; - } - - err = bpf_obj_get_info_by_fd(fd, &info, &len); - if (err) { - p_err("can't get map info (%u): %s", - id, strerror(errno)); - goto err_close_fd; - } - - if (strncmp(name, info.name, BPF_OBJ_NAME_LEN)) { - close(fd); - continue; - } - - if (nb_fds > 0) { - tmp = realloc(*fds, (nb_fds + 1) * sizeof(int)); - if (!tmp) { - p_err("failed to realloc"); - goto err_close_fd; - } - *fds = tmp; - } - (*fds)[nb_fds++] = fd; - } - -err_close_fd: - close(fd); -err_close_fds: - while (--nb_fds >= 0) - close((*fds)[nb_fds]); - return -1; -} - -static int map_parse_fds(int *argc, char ***argv, int **fds) -{ - if (is_prefix(**argv, "id")) { - unsigned int id; - char *endptr; - - NEXT_ARGP(); - - id = strtoul(**argv, &endptr, 0); - if (*endptr) { - p_err("can't parse %s as ID", **argv); - return -1; - } - NEXT_ARGP(); - - (*fds)[0] = bpf_map_get_fd_by_id(id); - if ((*fds)[0] < 0) { - p_err("get map by id (%u): %s", id, strerror(errno)); - return -1; - } - return 1; - } else if (is_prefix(**argv, "name")) { - char *name; - - NEXT_ARGP(); - - name = **argv; - if (strlen(name) > BPF_OBJ_NAME_LEN - 1) { - p_err("can't parse name"); - return -1; - } - NEXT_ARGP(); - - return map_fd_by_name(name, fds); - } else if (is_prefix(**argv, "pinned")) { - char *path; - - NEXT_ARGP(); - - path = **argv; - NEXT_ARGP(); - - (*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_MAP); - if ((*fds)[0] < 0) - return -1; - return 1; - } - - p_err("expected 'id', 'name' or 'pinned', got: '%s'?", **argv); - return -1; -} - -int map_parse_fd(int *argc, char ***argv) -{ - int *fds = NULL; - int nb_fds, fd; - - fds = malloc(sizeof(int)); - if (!fds) { - p_err("mem alloc failed"); - return -1; - } - nb_fds = map_parse_fds(argc, argv, &fds); - if (nb_fds != 1) { - if (nb_fds > 1) { - p_err("several maps match this handle"); - while (nb_fds--) - close(fds[nb_fds]); - } - fd = -1; - goto exit_free; - } - - fd = fds[0]; -exit_free: - free(fds); - return fd; -} - -int map_parse_fd_and_info(int *argc, char ***argv, void *info, __u32 *info_len) -{ - int err; - int fd; - - fd = map_parse_fd(argc, argv); - if (fd < 0) - return -1; - - err = bpf_obj_get_info_by_fd(fd, info, info_len); - if (err) { - p_err("can't get map info: %s", strerror(errno)); - close(fd); - return err; - } - - return fd; -} - static int do_dump_btf(const struct btf_dumper *d, struct bpf_map_info *map_info, void *key, void *value) { __u32 value_id; - int ret; + int ret = 0; /* start of key-value pair */ jsonw_start_object(d->jw); @@ -367,8 +188,9 @@ static void print_entry_json(struct bpf_map_info *info, unsigned char *key, jsonw_end_object(json_wtr); } -static void print_entry_error(struct bpf_map_info *info, unsigned char *key, - const char *error_msg) +static void +print_entry_error_msg(struct bpf_map_info *info, unsigned char *key, + const char *error_msg) { int msg_size = strlen(error_msg); bool single_line, break_names; @@ -386,6 +208,40 @@ static void print_entry_error(struct bpf_map_info *info, unsigned char *key, printf("\n"); } +static void +print_entry_error(struct bpf_map_info *map_info, void *key, int lookup_errno) +{ + /* For prog_array maps or arrays of maps, failure to lookup the value + * means there is no entry for that key. Do not print an error message + * in that case. + */ + if ((map_is_map_of_maps(map_info->type) || + map_is_map_of_progs(map_info->type)) && lookup_errno == ENOENT) + return; + + if (json_output) { + jsonw_start_object(json_wtr); /* entry */ + jsonw_name(json_wtr, "key"); + print_hex_data_json(key, map_info->key_size); + jsonw_name(json_wtr, "value"); + jsonw_start_object(json_wtr); /* error */ + jsonw_string_field(json_wtr, "error", strerror(lookup_errno)); + jsonw_end_object(json_wtr); /* error */ + jsonw_end_object(json_wtr); /* entry */ + } else { + const char *msg = NULL; + + if (lookup_errno == ENOENT) + msg = "<no entry>"; + else if (lookup_errno == ENOSPC && + map_info->type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) + msg = "<cannot read>"; + + print_entry_error_msg(map_info, key, + msg ? : strerror(lookup_errno)); + } +} + static void print_entry_plain(struct bpf_map_info *info, unsigned char *key, unsigned char *value) { @@ -586,9 +442,12 @@ static int parse_elem(char **argv, struct bpf_map_info *info, static void show_map_header_json(struct bpf_map_info *info, json_writer_t *wtr) { + const char *map_type_str; + jsonw_uint_field(wtr, "id", info->id); - if (info->type < ARRAY_SIZE(map_type_name)) - jsonw_string_field(wtr, "type", map_type_name[info->type]); + map_type_str = libbpf_bpf_map_type_str(info->type); + if (map_type_str) + jsonw_string_field(wtr, "type", map_type_str); else jsonw_uint_field(wtr, "type", info->type); @@ -618,7 +477,7 @@ static int show_map_close_json(int fd, struct bpf_map_info *info) jsonw_uint_field(json_wtr, "max_entries", info->max_entries); if (memlock) - jsonw_int_field(json_wtr, "bytes_memlock", atoi(memlock)); + jsonw_int_field(json_wtr, "bytes_memlock", atoll(memlock)); free(memlock); if (info->type == BPF_MAP_TYPE_PROG_ARRAY) { @@ -627,10 +486,12 @@ static int show_map_close_json(int fd, struct bpf_map_info *info) if (owner_prog_type) { unsigned int prog_type = atoi(owner_prog_type); + const char *prog_type_str; - if (prog_type < ARRAY_SIZE(prog_type_name)) + prog_type_str = libbpf_bpf_prog_type_str(prog_type); + if (prog_type_str) jsonw_string_field(json_wtr, "owner_prog_type", - prog_type_name[prog_type]); + prog_type_str); else jsonw_uint_field(json_wtr, "owner_prog_type", prog_type); @@ -653,18 +514,19 @@ static int show_map_close_json(int fd, struct bpf_map_info *info) if (info->btf_id) jsonw_int_field(json_wtr, "btf_id", info->btf_id); - if (!hash_empty(map_table.table)) { - struct pinned_obj *obj; + if (!hashmap__empty(map_table)) { + struct hashmap_entry *entry; jsonw_name(json_wtr, "pinned"); jsonw_start_array(json_wtr); - hash_for_each_possible(map_table.table, obj, hash, info->id) { - if (obj->id == info->id) - jsonw_string(json_wtr, obj->path); - } + hashmap__for_each_key_entry(map_table, entry, + u32_as_hash_field(info->id)) + jsonw_string(json_wtr, entry->value); jsonw_end_array(json_wtr); } + emit_obj_refs_json(refs_table, info->id, json_wtr); + jsonw_end_object(json_wtr); return 0; @@ -672,9 +534,13 @@ static int show_map_close_json(int fd, struct bpf_map_info *info) static void show_map_header_plain(struct bpf_map_info *info) { + const char *map_type_str; + printf("%u: ", info->id); - if (info->type < ARRAY_SIZE(map_type_name)) - printf("%s ", map_type_name[info->type]); + + map_type_str = libbpf_bpf_map_type_str(info->type); + if (map_type_str) + printf("%s ", map_type_str); else printf("type %u ", info->type); @@ -710,10 +576,11 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info) printf("\n\t"); if (owner_prog_type) { unsigned int prog_type = atoi(owner_prog_type); + const char *prog_type_str; - if (prog_type < ARRAY_SIZE(prog_type_name)) - printf("owner_prog_type %s ", - prog_type_name[prog_type]); + prog_type_str = libbpf_bpf_prog_type_str(prog_type); + if (prog_type_str) + printf("owner_prog_type %s ", prog_type_str); else printf("owner_prog_type %d ", prog_type); } @@ -726,25 +593,21 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info) } close(fd); - if (!hash_empty(map_table.table)) { - struct pinned_obj *obj; + if (!hashmap__empty(map_table)) { + struct hashmap_entry *entry; - hash_for_each_possible(map_table.table, obj, hash, info->id) { - if (obj->id == info->id) - printf("\n\tpinned %s", obj->path); - } + hashmap__for_each_key_entry(map_table, entry, + u32_as_hash_field(info->id)) + printf("\n\tpinned %s", (char *)entry->value); } - printf("\n"); if (frozen_str) { frozen = atoi(frozen_str); free(frozen_str); } - if (!info->btf_id && !frozen) - return 0; - - printf("\t"); + if (info->btf_id || frozen) + printf("\n\t"); if (info->btf_id) printf("btf_id %d", info->btf_id); @@ -752,6 +615,8 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info) if (frozen) printf("%sfrozen", info->btf_id ? " " : ""); + emit_obj_refs_plain(refs_table, info->id, "\n\tpids "); + printf("\n"); return 0; } @@ -808,8 +673,16 @@ static int do_show(int argc, char **argv) int err; int fd; - if (show_pinned) - build_pinned_obj_table(&map_table, BPF_OBJ_MAP); + if (show_pinned) { + map_table = hashmap__new(hash_fn_for_key_as_id, + equal_fn_for_key_as_id, NULL); + if (IS_ERR(map_table)) { + p_err("failed to create hashmap for pinned paths"); + return -1; + } + build_pinned_obj_table(map_table, BPF_OBJ_MAP); + } + build_obj_refs_table(&refs_table, BPF_OBJ_MAP); if (argc == 2) return do_show_subset(argc, argv); @@ -853,6 +726,11 @@ static int do_show(int argc, char **argv) if (json_output) jsonw_end_array(json_wtr); + delete_obj_refs_table(refs_table); + + if (show_pinned) + delete_pinned_obj_table(map_table); + return errno == ENOENT ? 0 : -1; } @@ -860,56 +738,23 @@ static int dump_map_elem(int fd, void *key, void *value, struct bpf_map_info *map_info, struct btf *btf, json_writer_t *btf_wtr) { - int num_elems = 0; - int lookup_errno; - - if (!bpf_map_lookup_elem(fd, key, value)) { - if (json_output) { - print_entry_json(map_info, key, value, btf); - } else { - if (btf) { - struct btf_dumper d = { - .btf = btf, - .jw = btf_wtr, - .is_plain_text = true, - }; - - do_dump_btf(&d, map_info, key, value); - } else { - print_entry_plain(map_info, key, value); - } - num_elems++; - } - return num_elems; + if (bpf_map_lookup_elem(fd, key, value)) { + print_entry_error(map_info, key, errno); + return -1; } - /* lookup error handling */ - lookup_errno = errno; - - if (map_is_map_of_maps(map_info->type) || - map_is_map_of_progs(map_info->type)) - return 0; - if (json_output) { - jsonw_start_object(json_wtr); - jsonw_name(json_wtr, "key"); - print_hex_data_json(key, map_info->key_size); - jsonw_name(json_wtr, "value"); - jsonw_start_object(json_wtr); - jsonw_string_field(json_wtr, "error", strerror(lookup_errno)); - jsonw_end_object(json_wtr); - jsonw_end_object(json_wtr); - } else { - const char *msg = NULL; - - if (lookup_errno == ENOENT) - msg = "<no entry>"; - else if (lookup_errno == ENOSPC && - map_info->type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY) - msg = "<cannot read>"; + print_entry_json(map_info, key, value, btf); + } else if (btf) { + struct btf_dumper d = { + .btf = btf, + .jw = btf_wtr, + .is_plain_text = true, + }; - print_entry_error(map_info, key, - msg ? : strerror(lookup_errno)); + do_dump_btf(&d, map_info, key, value); + } else { + print_entry_plain(map_info, key, value); } return 0; @@ -937,39 +782,41 @@ static int maps_have_btf(int *fds, int nb_fds) static struct btf *btf_vmlinux; -static struct btf *get_map_kv_btf(const struct bpf_map_info *info) +static int get_map_kv_btf(const struct bpf_map_info *info, struct btf **btf) { - struct btf *btf = NULL; + int err = 0; if (info->btf_vmlinux_value_type_id) { if (!btf_vmlinux) { btf_vmlinux = libbpf_find_kernel_btf(); - if (IS_ERR(btf_vmlinux)) + err = libbpf_get_error(btf_vmlinux); + if (err) { p_err("failed to get kernel btf"); + return err; + } } - return btf_vmlinux; + *btf = btf_vmlinux; } else if (info->btf_value_type_id) { - int err; - - err = btf__get_from_id(info->btf_id, &btf); - if (err || !btf) { + *btf = btf__load_from_kernel_by_id(info->btf_id); + err = libbpf_get_error(*btf); + if (err) p_err("failed to get btf"); - btf = err ? ERR_PTR(err) : ERR_PTR(-ESRCH); - } + } else { + *btf = NULL; } - return btf; + return err; } static void free_map_kv_btf(struct btf *btf) { - if (!IS_ERR(btf) && btf != btf_vmlinux) + if (!libbpf_get_error(btf) && btf != btf_vmlinux) btf__free(btf); } static void free_btf_vmlinux(void) { - if (!IS_ERR(btf_vmlinux)) + if (!libbpf_get_error(btf_vmlinux)) btf__free(btf_vmlinux); } @@ -993,9 +840,8 @@ map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr, prev_key = NULL; if (wtr) { - btf = get_map_kv_btf(info); - if (IS_ERR(btf)) { - err = PTR_ERR(btf); + err = get_map_kv_btf(info, &btf); + if (err) { goto exit_free; } @@ -1010,9 +856,13 @@ map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr, } if (info->type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY && - info->value_size != 8) + info->value_size != 8) { + const char *map_type_str; + + map_type_str = libbpf_bpf_map_type_str(info->type); p_info("Warning: cannot read values from %s map with value_size != 8", - map_type_name[info->type]); + map_type_str); + } while (true) { err = bpf_map_get_next_key(fd, prev_key, key); if (err) { @@ -1020,7 +870,8 @@ map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr, err = 0; break; } - num_elems += dump_map_elem(fd, key, value, info, btf, wtr); + if (!dump_map_elem(fd, key, value, info, btf, wtr)) + num_elems++; prev_key = key; } @@ -1182,14 +1033,10 @@ static void print_key_value(struct bpf_map_info *info, void *key, void *value) { json_writer_t *btf_wtr; - struct btf *btf = NULL; - int err; + struct btf *btf; - err = btf__get_from_id(info->btf_id, &btf); - if (err) { - p_err("failed to get btf"); + if (get_map_kv_btf(info, &btf)) return; - } if (json_output) { print_entry_json(info, key, value, btf); @@ -1384,7 +1231,7 @@ static int do_pin(int argc, char **argv) { int err; - err = do_pin_any(argc, argv, bpf_map_get_fd_by_id); + err = do_pin_any(argc, argv, map_parse_fd); if (!err && json_output) jsonw_null(json_wtr); return err; @@ -1392,9 +1239,12 @@ static int do_pin(int argc, char **argv) static int do_create(int argc, char **argv) { - struct bpf_create_map_attr attr = { NULL, }; + LIBBPF_OPTS(bpf_map_create_opts, attr); + enum bpf_map_type map_type = BPF_MAP_TYPE_UNSPEC; + __u32 key_size = 0, value_size = 0, max_entries = 0; + const char *map_name = NULL; const char *pinfile; - int err, fd; + int err = -1, fd; if (!REQ_ARGS(7)) return -1; @@ -1407,78 +1257,96 @@ static int do_create(int argc, char **argv) if (is_prefix(*argv, "type")) { NEXT_ARG(); - if (attr.map_type) { + if (map_type) { p_err("map type already specified"); - return -1; + goto exit; } - attr.map_type = map_type_from_str(*argv); - if ((int)attr.map_type < 0) { + map_type = map_type_from_str(*argv); + if ((int)map_type < 0) { p_err("unrecognized map type: %s", *argv); - return -1; + goto exit; } NEXT_ARG(); } else if (is_prefix(*argv, "name")) { NEXT_ARG(); - attr.name = GET_ARG(); + map_name = GET_ARG(); } else if (is_prefix(*argv, "key")) { - if (parse_u32_arg(&argc, &argv, &attr.key_size, + if (parse_u32_arg(&argc, &argv, &key_size, "key size")) - return -1; + goto exit; } else if (is_prefix(*argv, "value")) { - if (parse_u32_arg(&argc, &argv, &attr.value_size, + if (parse_u32_arg(&argc, &argv, &value_size, "value size")) - return -1; + goto exit; } else if (is_prefix(*argv, "entries")) { - if (parse_u32_arg(&argc, &argv, &attr.max_entries, + if (parse_u32_arg(&argc, &argv, &max_entries, "max entries")) - return -1; + goto exit; } else if (is_prefix(*argv, "flags")) { if (parse_u32_arg(&argc, &argv, &attr.map_flags, "flags")) - return -1; + goto exit; } else if (is_prefix(*argv, "dev")) { NEXT_ARG(); if (attr.map_ifindex) { p_err("offload device already specified"); - return -1; + goto exit; } attr.map_ifindex = if_nametoindex(*argv); if (!attr.map_ifindex) { p_err("unrecognized netdevice '%s': %s", *argv, strerror(errno)); - return -1; + goto exit; } NEXT_ARG(); + } else if (is_prefix(*argv, "inner_map")) { + struct bpf_map_info info = {}; + __u32 len = sizeof(info); + int inner_map_fd; + + NEXT_ARG(); + if (!REQ_ARGS(2)) + usage(); + inner_map_fd = map_parse_fd_and_info(&argc, &argv, + &info, &len); + if (inner_map_fd < 0) + return -1; + attr.inner_map_fd = inner_map_fd; } else { p_err("unknown arg %s", *argv); - return -1; + goto exit; } } - if (!attr.name) { + if (!map_name) { p_err("map name not specified"); - return -1; + goto exit; } set_max_rlimit(); - fd = bpf_create_map_xattr(&attr); + fd = bpf_map_create(map_type, map_name, key_size, value_size, max_entries, &attr); if (fd < 0) { p_err("map create failed: %s", strerror(errno)); - return -1; + goto exit; } err = do_pin_fd(fd, pinfile); close(fd); if (err) - return err; + goto exit; if (json_output) jsonw_null(json_wtr); - return 0; + +exit: + if (attr.inner_map_fd > 0) + close(attr.inner_map_fd); + + return err; } static int do_pop_dequeue(int argc, char **argv) @@ -1561,24 +1429,24 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s %s { show | list } [MAP]\n" - " %s %s create FILE type TYPE key KEY_SIZE value VALUE_SIZE \\\n" - " entries MAX_ENTRIES name NAME [flags FLAGS] \\\n" - " [dev NAME]\n" - " %s %s dump MAP\n" - " %s %s update MAP [key DATA] [value VALUE] [UPDATE_FLAGS]\n" - " %s %s lookup MAP [key DATA]\n" - " %s %s getnext MAP [key DATA]\n" - " %s %s delete MAP key DATA\n" - " %s %s pin MAP FILE\n" - " %s %s event_pipe MAP [cpu N index M]\n" - " %s %s peek MAP\n" - " %s %s push MAP value VALUE\n" - " %s %s pop MAP\n" - " %s %s enqueue MAP value VALUE\n" - " %s %s dequeue MAP\n" - " %s %s freeze MAP\n" - " %s %s help\n" + "Usage: %1$s %2$s { show | list } [MAP]\n" + " %1$s %2$s create FILE type TYPE key KEY_SIZE value VALUE_SIZE \\\n" + " entries MAX_ENTRIES name NAME [flags FLAGS] \\\n" + " [inner_map MAP] [dev NAME]\n" + " %1$s %2$s dump MAP\n" + " %1$s %2$s update MAP [key DATA] [value VALUE] [UPDATE_FLAGS]\n" + " %1$s %2$s lookup MAP [key DATA]\n" + " %1$s %2$s getnext MAP [key DATA]\n" + " %1$s %2$s delete MAP key DATA\n" + " %1$s %2$s pin MAP FILE\n" + " %1$s %2$s event_pipe MAP [cpu N index M]\n" + " %1$s %2$s peek MAP\n" + " %1$s %2$s push MAP value VALUE\n" + " %1$s %2$s pop MAP\n" + " %1$s %2$s enqueue MAP value VALUE\n" + " %1$s %2$s dequeue MAP\n" + " %1$s %2$s freeze MAP\n" + " %1$s %2$s help\n" "\n" " " HELP_SPEC_MAP "\n" " DATA := { [hex] BYTES }\n" @@ -1589,14 +1457,12 @@ static int do_help(int argc, char **argv) " percpu_array | stack_trace | cgroup_array | lru_hash |\n" " lru_percpu_hash | lpm_trie | array_of_maps | hash_of_maps |\n" " devmap | devmap_hash | sockmap | cpumap | xskmap | sockhash |\n" - " cgroup_storage | reuseport_sockarray | percpu_cgroup_storage }\n" - " " HELP_SPEC_OPTIONS "\n" + " cgroup_storage | reuseport_sockarray | percpu_cgroup_storage |\n" + " queue | stack | sk_storage | struct_ops | ringbuf | inode_storage |\n" + " task_storage | bloom_filter | user_ringbuf }\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-f|--bpffs} | {-n|--nomount} }\n" "", - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2]); return 0; diff --git a/tools/bpf/bpftool/map_perf_ring.c b/tools/bpf/bpftool/map_perf_ring.c index d9b29c17fbb8..21d7d447e1f3 100644 --- a/tools/bpf/bpftool/map_perf_ring.c +++ b/tools/bpf/bpftool/map_perf_ring.c @@ -22,7 +22,6 @@ #include <sys/syscall.h> #include <bpf/bpf.h> -#include <perf-sys.h> #include "main.h" @@ -30,16 +29,9 @@ static volatile bool stop; -struct event_ring_info { - int fd; - int key; - unsigned int cpu; - void *mem; -}; - struct perf_event_sample { struct perf_event_header header; - u64 time; + __u64 time; __u32 size; unsigned char data[]; }; @@ -125,7 +117,7 @@ int do_event_pipe(int argc, char **argv) .wakeup_events = 1, }; struct bpf_map_info map_info = {}; - struct perf_buffer_raw_opts opts = {}; + LIBBPF_OPTS(perf_buffer_raw_opts, opts); struct event_pipe_ctx ctx = { .all_cpus = true, .cpu = -1, @@ -191,18 +183,14 @@ int do_event_pipe(int argc, char **argv) ctx.idx = 0; } - opts.attr = &perf_attr; - opts.event_cb = print_bpf_output; - opts.ctx = &ctx; opts.cpu_cnt = ctx.all_cpus ? 0 : 1; opts.cpus = &ctx.cpu; opts.map_keys = &ctx.idx; - - pb = perf_buffer__new_raw(map_fd, MMAP_PAGE_CNT, &opts); - err = libbpf_get_error(pb); - if (err) { + pb = perf_buffer__new_raw(map_fd, MMAP_PAGE_CNT, &perf_attr, + print_bpf_output, &ctx, &opts); + if (!pb) { p_err("failed to create perf buffer: %s (%d)", - strerror(err), err); + strerror(errno), errno); goto err_close_map; } @@ -217,7 +205,7 @@ int do_event_pipe(int argc, char **argv) err = perf_buffer__poll(pb, 200); if (err < 0 && err != -EINTR) { p_err("perf buffer polling failed: %s (%d)", - strerror(err), err); + strerror(errno), errno); goto err_close_pb; } } diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c index c5e3895b7c8b..526a332c48e6 100644 --- a/tools/bpf/bpftool/net.c +++ b/tools/bpf/bpftool/net.c @@ -6,22 +6,26 @@ #include <fcntl.h> #include <stdlib.h> #include <string.h> +#include <time.h> #include <unistd.h> #include <bpf/bpf.h> #include <bpf/libbpf.h> #include <net/if.h> -#include <linux/if.h> #include <linux/rtnetlink.h> +#include <linux/socket.h> #include <linux/tc_act/tc_bpf.h> #include <sys/socket.h> #include <sys/stat.h> #include <sys/types.h> #include "bpf/nlattr.h" -#include "bpf/libbpf_internal.h" #include "main.h" #include "netlink_dumper.h" +#ifndef SOL_NETLINK +#define SOL_NETLINK 270 +#endif + struct ip_devname_ifindex { char devname[64]; int ifindex; @@ -85,6 +89,266 @@ static enum net_attach_type parse_attach_type(const char *str) return net_attach_type_size; } +typedef int (*dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb); + +typedef int (*__dump_nlmsg_t)(struct nlmsghdr *nlmsg, dump_nlmsg_t, void *cookie); + +static int netlink_open(__u32 *nl_pid) +{ + struct sockaddr_nl sa; + socklen_t addrlen; + int one = 1, ret; + int sock; + + memset(&sa, 0, sizeof(sa)); + sa.nl_family = AF_NETLINK; + + sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE); + if (sock < 0) + return -errno; + + if (setsockopt(sock, SOL_NETLINK, NETLINK_EXT_ACK, + &one, sizeof(one)) < 0) { + p_err("Netlink error reporting not supported"); + } + + if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) { + ret = -errno; + goto cleanup; + } + + addrlen = sizeof(sa); + if (getsockname(sock, (struct sockaddr *)&sa, &addrlen) < 0) { + ret = -errno; + goto cleanup; + } + + if (addrlen != sizeof(sa)) { + ret = -LIBBPF_ERRNO__INTERNAL; + goto cleanup; + } + + *nl_pid = sa.nl_pid; + return sock; + +cleanup: + close(sock); + return ret; +} + +static int netlink_recv(int sock, __u32 nl_pid, __u32 seq, + __dump_nlmsg_t _fn, dump_nlmsg_t fn, + void *cookie) +{ + bool multipart = true; + struct nlmsgerr *err; + struct nlmsghdr *nh; + char buf[4096]; + int len, ret; + + while (multipart) { + multipart = false; + len = recv(sock, buf, sizeof(buf), 0); + if (len < 0) { + ret = -errno; + goto done; + } + + if (len == 0) + break; + + for (nh = (struct nlmsghdr *)buf; NLMSG_OK(nh, (unsigned int)len); + nh = NLMSG_NEXT(nh, len)) { + if (nh->nlmsg_pid != nl_pid) { + ret = -LIBBPF_ERRNO__WRNGPID; + goto done; + } + if (nh->nlmsg_seq != seq) { + ret = -LIBBPF_ERRNO__INVSEQ; + goto done; + } + if (nh->nlmsg_flags & NLM_F_MULTI) + multipart = true; + switch (nh->nlmsg_type) { + case NLMSG_ERROR: + err = (struct nlmsgerr *)NLMSG_DATA(nh); + if (!err->error) + continue; + ret = err->error; + libbpf_nla_dump_errormsg(nh); + goto done; + case NLMSG_DONE: + return 0; + default: + break; + } + if (_fn) { + ret = _fn(nh, fn, cookie); + if (ret) + return ret; + } + } + } + ret = 0; +done: + return ret; +} + +static int __dump_class_nlmsg(struct nlmsghdr *nlh, + 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); +} + +static int netlink_get_class(int sock, unsigned int nl_pid, int ifindex, + 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 netlink_recv(sock, nl_pid, seq, __dump_class_nlmsg, + dump_class_nlmsg, cookie); +} + +static int __dump_qdisc_nlmsg(struct nlmsghdr *nlh, + 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); +} + +static int netlink_get_qdisc(int sock, unsigned int nl_pid, int ifindex, + 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 netlink_recv(sock, nl_pid, seq, __dump_qdisc_nlmsg, + dump_qdisc_nlmsg, cookie); +} + +static int __dump_filter_nlmsg(struct nlmsghdr *nlh, + 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); +} + +static int netlink_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle, + 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 netlink_recv(sock, nl_pid, seq, __dump_filter_nlmsg, + dump_filter_nlmsg, cookie); +} + +static int __dump_link_nlmsg(struct nlmsghdr *nlh, + dump_nlmsg_t dump_link_nlmsg, void *cookie) +{ + struct nlattr *tb[IFLA_MAX + 1], *attr; + struct ifinfomsg *ifi = NLMSG_DATA(nlh); + int len; + + len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*ifi)); + attr = (struct nlattr *) ((void *) ifi + NLMSG_ALIGN(sizeof(*ifi))); + if (libbpf_nla_parse(tb, IFLA_MAX, attr, len, NULL) != 0) + return -LIBBPF_ERRNO__NLPARSE; + + return dump_link_nlmsg(cookie, ifi, tb); +} + +static int netlink_get_link(int sock, unsigned int nl_pid, + dump_nlmsg_t dump_link_nlmsg, void *cookie) +{ + struct { + struct nlmsghdr nlh; + struct ifinfomsg ifm; + } req = { + .nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg)), + .nlh.nlmsg_type = RTM_GETLINK, + .nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST, + .ifm.ifi_family = AF_PACKET, + }; + int seq = time(NULL); + + req.nlh.nlmsg_seq = seq; + if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0) + return -errno; + + return netlink_recv(sock, nl_pid, seq, __dump_link_nlmsg, + dump_link_nlmsg, cookie); +} + static int dump_link_nlmsg(void *cookie, void *msg, struct nlattr **tb) { struct bpf_netdev_t *netinfo = cookie; @@ -168,14 +432,14 @@ static int show_dev_tc_bpf(int sock, unsigned int nl_pid, tcinfo.array_len = 0; tcinfo.is_qdisc = false; - ret = libbpf_nl_get_class(sock, nl_pid, dev->ifindex, - dump_class_qdisc_nlmsg, &tcinfo); + ret = netlink_get_class(sock, nl_pid, dev->ifindex, + dump_class_qdisc_nlmsg, &tcinfo); if (ret) goto out; tcinfo.is_qdisc = true; - ret = libbpf_nl_get_qdisc(sock, nl_pid, dev->ifindex, - dump_class_qdisc_nlmsg, &tcinfo); + ret = netlink_get_qdisc(sock, nl_pid, dev->ifindex, + dump_class_qdisc_nlmsg, &tcinfo); if (ret) goto out; @@ -183,9 +447,9 @@ static int show_dev_tc_bpf(int sock, unsigned int nl_pid, filter_info.ifindex = dev->ifindex; for (i = 0; i < tcinfo.used_len; i++) { filter_info.kind = tcinfo.handle_array[i].kind; - ret = libbpf_nl_get_filter(sock, nl_pid, dev->ifindex, - tcinfo.handle_array[i].handle, - dump_filter_nlmsg, &filter_info); + ret = netlink_get_filter(sock, nl_pid, dev->ifindex, + tcinfo.handle_array[i].handle, + dump_filter_nlmsg, &filter_info); if (ret) goto out; } @@ -193,22 +457,22 @@ static int show_dev_tc_bpf(int sock, unsigned int nl_pid, /* root, ingress and egress handle */ handle = TC_H_ROOT; filter_info.kind = "root"; - ret = libbpf_nl_get_filter(sock, nl_pid, dev->ifindex, handle, - dump_filter_nlmsg, &filter_info); + ret = netlink_get_filter(sock, nl_pid, dev->ifindex, handle, + dump_filter_nlmsg, &filter_info); if (ret) goto out; handle = TC_H_MAKE(TC_H_CLSACT, TC_H_MIN_INGRESS); filter_info.kind = "clsact/ingress"; - ret = libbpf_nl_get_filter(sock, nl_pid, dev->ifindex, handle, - dump_filter_nlmsg, &filter_info); + ret = netlink_get_filter(sock, nl_pid, dev->ifindex, handle, + dump_filter_nlmsg, &filter_info); if (ret) goto out; handle = TC_H_MAKE(TC_H_CLSACT, TC_H_MIN_EGRESS); filter_info.kind = "clsact/egress"; - ret = libbpf_nl_get_filter(sock, nl_pid, dev->ifindex, handle, - dump_filter_nlmsg, &filter_info); + ret = netlink_get_filter(sock, nl_pid, dev->ifindex, handle, + dump_filter_nlmsg, &filter_info); if (ret) goto out; @@ -287,7 +551,7 @@ static int do_attach_detach_xdp(int progfd, enum net_attach_type attach_type, if (attach_type == NET_ATTACH_TYPE_XDP_OFFLOAD) flags |= XDP_FLAGS_HW_MODE; - return bpf_set_link_xdp_fd(ifindex, progfd, flags); + return bpf_xdp_attach(ifindex, progfd, flags, NULL); } static int do_attach(int argc, char **argv) @@ -313,8 +577,8 @@ static int do_attach(int argc, char **argv) ifindex = net_parse_dev(&argc, &argv); if (ifindex < 1) { - close(progfd); - return -EINVAL; + err = -EINVAL; + goto cleanup; } if (argc) { @@ -322,8 +586,8 @@ static int do_attach(int argc, char **argv) overwrite = true; } else { p_err("expected 'overwrite', got: '%s'?", *argv); - close(progfd); - return -EINVAL; + err = -EINVAL; + goto cleanup; } } @@ -331,17 +595,17 @@ static int do_attach(int argc, char **argv) if (is_prefix("xdp", attach_type_strings[attach_type])) err = do_attach_detach_xdp(progfd, attach_type, ifindex, overwrite); - - if (err < 0) { + if (err) { p_err("interface %s attach failed: %s", attach_type_strings[attach_type], strerror(-err)); - return err; + goto cleanup; } if (json_output) jsonw_null(json_wtr); - - return 0; +cleanup: + close(progfd); + return err; } static int do_detach(int argc, char **argv) @@ -386,7 +650,7 @@ static int do_show(int argc, char **argv) struct bpf_attach_info attach_info = {}; int i, sock, ret, filter_idx = -1; struct bpf_netdev_t dev_array; - unsigned int nl_pid; + unsigned int nl_pid = 0; char err_buf[256]; if (argc == 2) { @@ -401,7 +665,7 @@ static int do_show(int argc, char **argv) if (ret) return -1; - sock = libbpf_netlink_open(&nl_pid); + sock = netlink_open(&nl_pid); if (sock < 0) { fprintf(stderr, "failed to open netlink sock\n"); return -1; @@ -416,7 +680,7 @@ static int do_show(int argc, char **argv) jsonw_start_array(json_wtr); NET_START_OBJECT; NET_START_ARRAY("xdp", "%s:\n"); - ret = libbpf_nl_get_link(sock, nl_pid, dump_link_nlmsg, &dev_array); + ret = netlink_get_link(sock, nl_pid, dump_link_nlmsg, &dev_array); NET_END_ARRAY("\n"); if (!ret) { @@ -458,20 +722,21 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s %s { show | list } [dev <devname>]\n" - " %s %s attach ATTACH_TYPE PROG dev <devname> [ overwrite ]\n" - " %s %s detach ATTACH_TYPE dev <devname>\n" - " %s %s help\n" + "Usage: %1$s %2$s { show | list } [dev <devname>]\n" + " %1$s %2$s attach ATTACH_TYPE PROG dev <devname> [ overwrite ]\n" + " %1$s %2$s detach ATTACH_TYPE dev <devname>\n" + " %1$s %2$s help\n" "\n" " " HELP_SPEC_PROGRAM "\n" " ATTACH_TYPE := { xdp | xdpgeneric | xdpdrv | xdpoffload }\n" + " " HELP_SPEC_OPTIONS " }\n" "\n" "Note: Only xdp and tc attachments are supported now.\n" " For progs attached to cgroups, use \"bpftool cgroup\"\n" " to dump program attachments. For program types\n" " sk_{filter,skb,msg,reuseport} and lwt/seg6, please\n" - " consult iproute2.\n", - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], + " consult iproute2.\n" + "", bin_name, argv[-2]); return 0; diff --git a/tools/bpf/bpftool/perf.c b/tools/bpf/bpftool/perf.c index 3341aa14acda..226ec2c39052 100644 --- a/tools/bpf/bpftool/perf.c +++ b/tools/bpf/bpftool/perf.c @@ -11,7 +11,7 @@ #include <sys/stat.h> #include <sys/types.h> #include <unistd.h> -#include <ftw.h> +#include <dirent.h> #include <bpf/bpf.h> @@ -147,81 +147,83 @@ static void print_perf_plain(int pid, int fd, __u32 prog_id, __u32 fd_type, } } -static int show_proc(const char *fpath, const struct stat *sb, - int tflag, struct FTW *ftwbuf) +static int show_proc(void) { + struct dirent *proc_de, *pid_fd_de; __u64 probe_offset, probe_addr; __u32 len, prog_id, fd_type; - int err, pid = 0, fd = 0; + DIR *proc, *pid_fd; + int err, pid, fd; const char *pch; char buf[4096]; - /* prefix always /proc */ - pch = fpath + 5; - if (*pch == '\0') - return 0; + proc = opendir("/proc"); + if (!proc) + return -1; - /* pid should be all numbers */ - pch++; - while (isdigit(*pch)) { - pid = pid * 10 + *pch - '0'; - pch++; - } - if (*pch == '\0') - return 0; - if (*pch != '/') - return FTW_SKIP_SUBTREE; - - /* check /proc/<pid>/fd directory */ - pch++; - if (strncmp(pch, "fd", 2)) - return FTW_SKIP_SUBTREE; - pch += 2; - if (*pch == '\0') - return 0; - if (*pch != '/') - return FTW_SKIP_SUBTREE; - - /* check /proc/<pid>/fd/<fd_num> */ - pch++; - while (isdigit(*pch)) { - fd = fd * 10 + *pch - '0'; - pch++; - } - if (*pch != '\0') - return FTW_SKIP_SUBTREE; + while ((proc_de = readdir(proc))) { + pid = 0; + pch = proc_de->d_name; - /* query (pid, fd) for potential perf events */ - len = sizeof(buf); - err = bpf_task_fd_query(pid, fd, 0, buf, &len, &prog_id, &fd_type, - &probe_offset, &probe_addr); - if (err < 0) - return 0; + /* pid should be all numbers */ + while (isdigit(*pch)) { + pid = pid * 10 + *pch - '0'; + pch++; + } + if (*pch != '\0') + continue; - if (json_output) - print_perf_json(pid, fd, prog_id, fd_type, buf, probe_offset, - probe_addr); - else - print_perf_plain(pid, fd, prog_id, fd_type, buf, probe_offset, - probe_addr); + err = snprintf(buf, sizeof(buf), "/proc/%s/fd", proc_de->d_name); + if (err < 0 || err >= (int)sizeof(buf)) + continue; + + pid_fd = opendir(buf); + if (!pid_fd) + continue; + while ((pid_fd_de = readdir(pid_fd))) { + fd = 0; + pch = pid_fd_de->d_name; + + /* fd should be all numbers */ + while (isdigit(*pch)) { + fd = fd * 10 + *pch - '0'; + pch++; + } + if (*pch != '\0') + continue; + + /* query (pid, fd) for potential perf events */ + len = sizeof(buf); + err = bpf_task_fd_query(pid, fd, 0, buf, &len, + &prog_id, &fd_type, + &probe_offset, &probe_addr); + if (err < 0) + continue; + + if (json_output) + print_perf_json(pid, fd, prog_id, fd_type, buf, + probe_offset, probe_addr); + else + print_perf_plain(pid, fd, prog_id, fd_type, buf, + probe_offset, probe_addr); + } + closedir(pid_fd); + } + closedir(proc); return 0; } static int do_show(int argc, char **argv) { - int flags = FTW_ACTIONRETVAL | FTW_PHYS; - int err = 0, nopenfd = 16; + int err; if (!has_perf_query_support()) return -1; if (json_output) jsonw_start_array(json_wtr); - if (nftw("/proc", show_proc, nopenfd, flags) == -1) { - p_err("%s", strerror(errno)); - err = -1; - } + err = show_proc(); if (json_output) jsonw_end_array(json_wtr); @@ -231,7 +233,10 @@ static int do_show(int argc, char **argv) static int do_help(int argc, char **argv) { fprintf(stderr, - "Usage: %s %s { show | list | help }\n" + "Usage: %1$s %2$s { show | list }\n" + " %1$s %2$s help }\n" + "\n" + " " HELP_SPEC_OPTIONS " }\n" "", bin_name, argv[-2]); diff --git a/tools/bpf/bpftool/pids.c b/tools/bpf/bpftool/pids.c new file mode 100644 index 000000000000..bb6c969a114a --- /dev/null +++ b/tools/bpf/bpftool/pids.c @@ -0,0 +1,256 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +/* Copyright (C) 2020 Facebook */ +#include <errno.h> +#include <linux/err.h> +#include <stdbool.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> + +#include <bpf/bpf.h> +#include <bpf/hashmap.h> + +#include "main.h" +#include "skeleton/pid_iter.h" + +#ifdef BPFTOOL_WITHOUT_SKELETONS + +int build_obj_refs_table(struct hashmap **map, enum bpf_obj_type type) +{ + return -ENOTSUP; +} +void delete_obj_refs_table(struct hashmap *map) {} +void emit_obj_refs_plain(struct hashmap *map, __u32 id, const char *prefix) {} +void emit_obj_refs_json(struct hashmap *map, __u32 id, json_writer_t *json_writer) {} + +#else /* BPFTOOL_WITHOUT_SKELETONS */ + +#include "pid_iter.skel.h" + +static void add_ref(struct hashmap *map, struct pid_iter_entry *e) +{ + struct hashmap_entry *entry; + struct obj_refs *refs; + struct obj_ref *ref; + int err, i; + void *tmp; + + hashmap__for_each_key_entry(map, entry, u32_as_hash_field(e->id)) { + refs = entry->value; + + for (i = 0; i < refs->ref_cnt; i++) { + if (refs->refs[i].pid == e->pid) + return; + } + + tmp = realloc(refs->refs, (refs->ref_cnt + 1) * sizeof(*ref)); + if (!tmp) { + p_err("failed to re-alloc memory for ID %u, PID %d, COMM %s...", + e->id, e->pid, e->comm); + return; + } + refs->refs = tmp; + ref = &refs->refs[refs->ref_cnt]; + ref->pid = e->pid; + memcpy(ref->comm, e->comm, sizeof(ref->comm)); + refs->ref_cnt++; + + return; + } + + /* new ref */ + refs = calloc(1, sizeof(*refs)); + if (!refs) { + p_err("failed to alloc memory for ID %u, PID %d, COMM %s...", + e->id, e->pid, e->comm); + return; + } + + refs->refs = malloc(sizeof(*refs->refs)); + if (!refs->refs) { + free(refs); + p_err("failed to alloc memory for ID %u, PID %d, COMM %s...", + e->id, e->pid, e->comm); + return; + } + ref = &refs->refs[0]; + ref->pid = e->pid; + memcpy(ref->comm, e->comm, sizeof(ref->comm)); + refs->ref_cnt = 1; + refs->has_bpf_cookie = e->has_bpf_cookie; + refs->bpf_cookie = e->bpf_cookie; + + err = hashmap__append(map, u32_as_hash_field(e->id), refs); + if (err) + p_err("failed to append entry to hashmap for ID %u: %s", + e->id, strerror(errno)); +} + +static int __printf(2, 0) +libbpf_print_none(__maybe_unused enum libbpf_print_level level, + __maybe_unused const char *format, + __maybe_unused va_list args) +{ + return 0; +} + +int build_obj_refs_table(struct hashmap **map, enum bpf_obj_type type) +{ + struct pid_iter_entry *e; + char buf[4096 / sizeof(*e) * sizeof(*e)]; + struct pid_iter_bpf *skel; + int err, ret, fd = -1, i; + libbpf_print_fn_t default_print; + + *map = hashmap__new(hash_fn_for_key_as_id, equal_fn_for_key_as_id, NULL); + if (IS_ERR(*map)) { + p_err("failed to create hashmap for PID references"); + return -1; + } + set_max_rlimit(); + + skel = pid_iter_bpf__open(); + if (!skel) { + p_err("failed to open PID iterator skeleton"); + return -1; + } + + skel->rodata->obj_type = type; + + /* we don't want output polluted with libbpf errors if bpf_iter is not + * supported + */ + default_print = libbpf_set_print(libbpf_print_none); + err = pid_iter_bpf__load(skel); + libbpf_set_print(default_print); + if (err) { + /* too bad, kernel doesn't support BPF iterators yet */ + err = 0; + goto out; + } + err = pid_iter_bpf__attach(skel); + if (err) { + /* if we loaded above successfully, attach has to succeed */ + p_err("failed to attach PID iterator: %d", err); + goto out; + } + + fd = bpf_iter_create(bpf_link__fd(skel->links.iter)); + if (fd < 0) { + err = -errno; + p_err("failed to create PID iterator session: %d", err); + goto out; + } + + while (true) { + ret = read(fd, buf, sizeof(buf)); + if (ret < 0) { + if (errno == EAGAIN) + continue; + err = -errno; + p_err("failed to read PID iterator output: %d", err); + goto out; + } + if (ret == 0) + break; + if (ret % sizeof(*e)) { + err = -EINVAL; + p_err("invalid PID iterator output format"); + goto out; + } + ret /= sizeof(*e); + + e = (void *)buf; + for (i = 0; i < ret; i++, e++) { + add_ref(*map, e); + } + } + err = 0; +out: + if (fd >= 0) + close(fd); + pid_iter_bpf__destroy(skel); + return err; +} + +void delete_obj_refs_table(struct hashmap *map) +{ + struct hashmap_entry *entry; + size_t bkt; + + if (!map) + return; + + hashmap__for_each_entry(map, entry, bkt) { + struct obj_refs *refs = entry->value; + + free(refs->refs); + free(refs); + } + + hashmap__free(map); +} + +void emit_obj_refs_json(struct hashmap *map, __u32 id, + json_writer_t *json_writer) +{ + struct hashmap_entry *entry; + + if (hashmap__empty(map)) + return; + + hashmap__for_each_key_entry(map, entry, u32_as_hash_field(id)) { + struct obj_refs *refs = entry->value; + int i; + + if (refs->ref_cnt == 0) + break; + + if (refs->has_bpf_cookie) + jsonw_lluint_field(json_writer, "bpf_cookie", refs->bpf_cookie); + + jsonw_name(json_writer, "pids"); + jsonw_start_array(json_writer); + for (i = 0; i < refs->ref_cnt; i++) { + struct obj_ref *ref = &refs->refs[i]; + + jsonw_start_object(json_writer); + jsonw_int_field(json_writer, "pid", ref->pid); + jsonw_string_field(json_writer, "comm", ref->comm); + jsonw_end_object(json_writer); + } + jsonw_end_array(json_writer); + break; + } +} + +void emit_obj_refs_plain(struct hashmap *map, __u32 id, const char *prefix) +{ + struct hashmap_entry *entry; + + if (hashmap__empty(map)) + return; + + hashmap__for_each_key_entry(map, entry, u32_as_hash_field(id)) { + struct obj_refs *refs = entry->value; + int i; + + if (refs->ref_cnt == 0) + break; + + if (refs->has_bpf_cookie) + printf("\n\tbpf_cookie %llu", (unsigned long long) refs->bpf_cookie); + + printf("%s", prefix); + for (i = 0; i < refs->ref_cnt; i++) { + struct obj_ref *ref = &refs->refs[i]; + + printf("%s%s(%d)", i == 0 ? "" : ", ", ref->comm, ref->pid); + } + break; + } +} + + +#endif diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c index b352ab041160..c81362a001ba 100644 --- a/tools/bpf/bpftool/prog.c +++ b/tools/bpf/bpftool/prog.c @@ -4,6 +4,7 @@ #define _GNU_SOURCE #include <errno.h> #include <fcntl.h> +#include <signal.h> #include <stdarg.h> #include <stdio.h> #include <stdlib.h> @@ -11,38 +12,70 @@ #include <time.h> #include <unistd.h> #include <net/if.h> +#include <sys/ioctl.h> #include <sys/types.h> #include <sys/stat.h> +#include <sys/syscall.h> +#include <dirent.h> #include <linux/err.h> +#include <linux/perf_event.h> #include <linux/sizes.h> #include <bpf/bpf.h> #include <bpf/btf.h> +#include <bpf/hashmap.h> #include <bpf/libbpf.h> +#include <bpf/libbpf_internal.h> +#include <bpf/skel_internal.h> #include "cfg.h" #include "main.h" #include "xlated_dumper.h" +#define BPF_METADATA_PREFIX "bpf_metadata_" +#define BPF_METADATA_PREFIX_LEN (sizeof(BPF_METADATA_PREFIX) - 1) + enum dump_mode { DUMP_JITED, DUMP_XLATED, }; +static const bool attach_types[] = { + [BPF_SK_SKB_STREAM_PARSER] = true, + [BPF_SK_SKB_STREAM_VERDICT] = true, + [BPF_SK_SKB_VERDICT] = true, + [BPF_SK_MSG_VERDICT] = true, + [BPF_FLOW_DISSECTOR] = true, + [__MAX_BPF_ATTACH_TYPE] = false, +}; + +/* Textual representations traditionally used by the program and kept around + * for the sake of backwards compatibility. + */ static const char * const attach_type_strings[] = { [BPF_SK_SKB_STREAM_PARSER] = "stream_parser", [BPF_SK_SKB_STREAM_VERDICT] = "stream_verdict", + [BPF_SK_SKB_VERDICT] = "skb_verdict", [BPF_SK_MSG_VERDICT] = "msg_verdict", - [BPF_FLOW_DISSECTOR] = "flow_dissector", [__MAX_BPF_ATTACH_TYPE] = NULL, }; +static struct hashmap *prog_table; + static enum bpf_attach_type parse_attach_type(const char *str) { enum bpf_attach_type type; for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) { + if (attach_types[type]) { + const char *attach_type_str; + + attach_type_str = libbpf_bpf_attach_type_str(type); + if (!strcmp(str, attach_type_str)) + return type; + } + if (attach_type_strings[type] && is_prefix(str, attach_type_strings[type])) return type; @@ -51,6 +84,76 @@ static enum bpf_attach_type parse_attach_type(const char *str) return __MAX_BPF_ATTACH_TYPE; } +static int prep_prog_info(struct bpf_prog_info *const info, enum dump_mode mode, + void **info_data, size_t *const info_data_sz) +{ + struct bpf_prog_info holder = {}; + size_t needed = 0; + void *ptr; + + if (mode == DUMP_JITED) { + holder.jited_prog_len = info->jited_prog_len; + needed += info->jited_prog_len; + } else { + holder.xlated_prog_len = info->xlated_prog_len; + needed += info->xlated_prog_len; + } + + holder.nr_jited_ksyms = info->nr_jited_ksyms; + needed += info->nr_jited_ksyms * sizeof(__u64); + + holder.nr_jited_func_lens = info->nr_jited_func_lens; + needed += info->nr_jited_func_lens * sizeof(__u32); + + holder.nr_func_info = info->nr_func_info; + holder.func_info_rec_size = info->func_info_rec_size; + needed += info->nr_func_info * info->func_info_rec_size; + + holder.nr_line_info = info->nr_line_info; + holder.line_info_rec_size = info->line_info_rec_size; + needed += info->nr_line_info * info->line_info_rec_size; + + holder.nr_jited_line_info = info->nr_jited_line_info; + holder.jited_line_info_rec_size = info->jited_line_info_rec_size; + needed += info->nr_jited_line_info * info->jited_line_info_rec_size; + + if (needed > *info_data_sz) { + ptr = realloc(*info_data, needed); + if (!ptr) + return -1; + + *info_data = ptr; + *info_data_sz = needed; + } + ptr = *info_data; + + if (mode == DUMP_JITED) { + holder.jited_prog_insns = ptr_to_u64(ptr); + ptr += holder.jited_prog_len; + } else { + holder.xlated_prog_insns = ptr_to_u64(ptr); + ptr += holder.xlated_prog_len; + } + + holder.jited_ksyms = ptr_to_u64(ptr); + ptr += holder.nr_jited_ksyms * sizeof(__u64); + + holder.jited_func_lens = ptr_to_u64(ptr); + ptr += holder.nr_jited_func_lens * sizeof(__u32); + + holder.func_info = ptr_to_u64(ptr); + ptr += holder.nr_func_info * holder.func_info_rec_size; + + holder.line_info = ptr_to_u64(ptr); + ptr += holder.nr_line_info * holder.line_info_rec_size; + + holder.jited_line_info = ptr_to_u64(ptr); + ptr += holder.nr_jited_line_info * holder.jited_line_info_rec_size; + + *info = holder; + return 0; +} + static void print_boot_time(__u64 nsecs, char *buf, unsigned int size) { struct timespec real_time_ts, boot_time_ts; @@ -82,198 +185,246 @@ static void print_boot_time(__u64 nsecs, char *buf, unsigned int size) strftime(buf, size, "%FT%T%z", &load_tm); } -static int prog_fd_by_nametag(void *nametag, int **fds, bool tag) +static void show_prog_maps(int fd, __u32 num_maps) { - unsigned int id = 0; - int fd, nb_fds = 0; - void *tmp; + struct bpf_prog_info info = {}; + __u32 len = sizeof(info); + __u32 map_ids[num_maps]; + unsigned int i; int err; - while (true) { - struct bpf_prog_info info = {}; - __u32 len = sizeof(info); + info.nr_map_ids = num_maps; + info.map_ids = ptr_to_u64(map_ids); - err = bpf_prog_get_next_id(id, &id); - if (err) { - if (errno != ENOENT) { - p_err("%s", strerror(errno)); - goto err_close_fds; - } - return nb_fds; - } + err = bpf_obj_get_info_by_fd(fd, &info, &len); + if (err || !info.nr_map_ids) + return; - fd = bpf_prog_get_fd_by_id(id); - if (fd < 0) { - p_err("can't get prog by id (%u): %s", - id, strerror(errno)); - goto err_close_fds; - } + if (json_output) { + jsonw_name(json_wtr, "map_ids"); + jsonw_start_array(json_wtr); + for (i = 0; i < info.nr_map_ids; i++) + jsonw_uint(json_wtr, map_ids[i]); + jsonw_end_array(json_wtr); + } else { + printf(" map_ids "); + for (i = 0; i < info.nr_map_ids; i++) + printf("%u%s", map_ids[i], + i == info.nr_map_ids - 1 ? "" : ","); + } +} - err = bpf_obj_get_info_by_fd(fd, &info, &len); - if (err) { - p_err("can't get prog info (%u): %s", - id, strerror(errno)); - goto err_close_fd; +static void *find_metadata(int prog_fd, struct bpf_map_info *map_info) +{ + struct bpf_prog_info prog_info; + __u32 prog_info_len; + __u32 map_info_len; + void *value = NULL; + __u32 *map_ids; + int nr_maps; + int key = 0; + int map_fd; + int ret; + __u32 i; + + memset(&prog_info, 0, sizeof(prog_info)); + prog_info_len = sizeof(prog_info); + ret = bpf_obj_get_info_by_fd(prog_fd, &prog_info, &prog_info_len); + if (ret) + return NULL; + + if (!prog_info.nr_map_ids) + return NULL; + + map_ids = calloc(prog_info.nr_map_ids, sizeof(__u32)); + if (!map_ids) + return NULL; + + nr_maps = prog_info.nr_map_ids; + memset(&prog_info, 0, sizeof(prog_info)); + prog_info.nr_map_ids = nr_maps; + prog_info.map_ids = ptr_to_u64(map_ids); + prog_info_len = sizeof(prog_info); + + ret = bpf_obj_get_info_by_fd(prog_fd, &prog_info, &prog_info_len); + if (ret) + goto free_map_ids; + + for (i = 0; i < prog_info.nr_map_ids; i++) { + map_fd = bpf_map_get_fd_by_id(map_ids[i]); + if (map_fd < 0) + goto free_map_ids; + + memset(map_info, 0, sizeof(*map_info)); + map_info_len = sizeof(*map_info); + ret = bpf_obj_get_info_by_fd(map_fd, map_info, &map_info_len); + if (ret < 0) { + close(map_fd); + goto free_map_ids; } - if ((tag && memcmp(nametag, info.tag, BPF_TAG_SIZE)) || - (!tag && strncmp(nametag, info.name, BPF_OBJ_NAME_LEN))) { - close(fd); + if (map_info->type != BPF_MAP_TYPE_ARRAY || + map_info->key_size != sizeof(int) || + map_info->max_entries != 1 || + !map_info->btf_value_type_id || + !strstr(map_info->name, ".rodata")) { + close(map_fd); continue; } - if (nb_fds > 0) { - tmp = realloc(*fds, (nb_fds + 1) * sizeof(int)); - if (!tmp) { - p_err("failed to realloc"); - goto err_close_fd; - } - *fds = tmp; + value = malloc(map_info->value_size); + if (!value) { + close(map_fd); + goto free_map_ids; + } + + if (bpf_map_lookup_elem(map_fd, &key, value)) { + close(map_fd); + free(value); + value = NULL; + goto free_map_ids; } - (*fds)[nb_fds++] = fd; + + close(map_fd); + break; } -err_close_fd: - close(fd); -err_close_fds: - while (--nb_fds >= 0) - close((*fds)[nb_fds]); - return -1; +free_map_ids: + free(map_ids); + return value; } -static int prog_parse_fds(int *argc, char ***argv, int **fds) +static bool has_metadata_prefix(const char *s) { - if (is_prefix(**argv, "id")) { - unsigned int id; - char *endptr; - - NEXT_ARGP(); + return strncmp(s, BPF_METADATA_PREFIX, BPF_METADATA_PREFIX_LEN) == 0; +} - id = strtoul(**argv, &endptr, 0); - if (*endptr) { - p_err("can't parse %s as ID", **argv); - return -1; - } - NEXT_ARGP(); +static void show_prog_metadata(int fd, __u32 num_maps) +{ + const struct btf_type *t_datasec, *t_var; + struct bpf_map_info map_info; + struct btf_var_secinfo *vsi; + bool printed_header = false; + unsigned int i, vlen; + void *value = NULL; + const char *name; + struct btf *btf; + int err; - (*fds)[0] = bpf_prog_get_fd_by_id(id); - if ((*fds)[0] < 0) { - p_err("get by id (%u): %s", id, strerror(errno)); - return -1; - } - return 1; - } else if (is_prefix(**argv, "tag")) { - unsigned char tag[BPF_TAG_SIZE]; + if (!num_maps) + return; - NEXT_ARGP(); + memset(&map_info, 0, sizeof(map_info)); + value = find_metadata(fd, &map_info); + if (!value) + return; - if (sscanf(**argv, BPF_TAG_FMT, tag, tag + 1, tag + 2, - tag + 3, tag + 4, tag + 5, tag + 6, tag + 7) - != BPF_TAG_SIZE) { - p_err("can't parse tag"); - return -1; - } - NEXT_ARGP(); + btf = btf__load_from_kernel_by_id(map_info.btf_id); + if (libbpf_get_error(btf)) + goto out_free; - return prog_fd_by_nametag(tag, fds, true); - } else if (is_prefix(**argv, "name")) { - char *name; + t_datasec = btf__type_by_id(btf, map_info.btf_value_type_id); + if (!btf_is_datasec(t_datasec)) + goto out_free; - NEXT_ARGP(); + vlen = btf_vlen(t_datasec); + vsi = btf_var_secinfos(t_datasec); - name = **argv; - if (strlen(name) > BPF_OBJ_NAME_LEN - 1) { - p_err("can't parse name"); - return -1; - } - NEXT_ARGP(); + /* We don't proceed to check the kinds of the elements of the DATASEC. + * The verifier enforces them to be BTF_KIND_VAR. + */ - return prog_fd_by_nametag(name, fds, false); - } else if (is_prefix(**argv, "pinned")) { - char *path; + if (json_output) { + struct btf_dumper d = { + .btf = btf, + .jw = json_wtr, + .is_plain_text = false, + }; - NEXT_ARGP(); + for (i = 0; i < vlen; i++, vsi++) { + t_var = btf__type_by_id(btf, vsi->type); + name = btf__name_by_offset(btf, t_var->name_off); - path = **argv; - NEXT_ARGP(); + if (!has_metadata_prefix(name)) + continue; - (*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_PROG); - if ((*fds)[0] < 0) - return -1; - return 1; - } + if (!printed_header) { + jsonw_name(json_wtr, "metadata"); + jsonw_start_object(json_wtr); + printed_header = true; + } - p_err("expected 'id', 'tag', 'name' or 'pinned', got: '%s'?", **argv); - return -1; -} + jsonw_name(json_wtr, name + BPF_METADATA_PREFIX_LEN); + err = btf_dumper_type(&d, t_var->type, value + vsi->offset); + if (err) { + p_err("btf dump failed: %d", err); + break; + } + } + if (printed_header) + jsonw_end_object(json_wtr); + } else { + json_writer_t *btf_wtr; + struct btf_dumper d = { + .btf = btf, + .is_plain_text = true, + }; -int prog_parse_fd(int *argc, char ***argv) -{ - int *fds = NULL; - int nb_fds, fd; + for (i = 0; i < vlen; i++, vsi++) { + t_var = btf__type_by_id(btf, vsi->type); + name = btf__name_by_offset(btf, t_var->name_off); - fds = malloc(sizeof(int)); - if (!fds) { - p_err("mem alloc failed"); - return -1; - } - nb_fds = prog_parse_fds(argc, argv, &fds); - if (nb_fds != 1) { - if (nb_fds > 1) { - p_err("several programs match this handle"); - while (nb_fds--) - close(fds[nb_fds]); - } - fd = -1; - goto exit_free; - } + if (!has_metadata_prefix(name)) + continue; - fd = fds[0]; -exit_free: - free(fds); - return fd; -} + if (!printed_header) { + printf("\tmetadata:"); -static void show_prog_maps(int fd, u32 num_maps) -{ - struct bpf_prog_info info = {}; - __u32 len = sizeof(info); - __u32 map_ids[num_maps]; - unsigned int i; - int err; + btf_wtr = jsonw_new(stdout); + if (!btf_wtr) { + p_err("jsonw alloc failed"); + goto out_free; + } + d.jw = btf_wtr, - info.nr_map_ids = num_maps; - info.map_ids = ptr_to_u64(map_ids); + printed_header = true; + } - err = bpf_obj_get_info_by_fd(fd, &info, &len); - if (err || !info.nr_map_ids) - return; + printf("\n\t\t%s = ", name + BPF_METADATA_PREFIX_LEN); - if (json_output) { - jsonw_name(json_wtr, "map_ids"); - jsonw_start_array(json_wtr); - for (i = 0; i < info.nr_map_ids; i++) - jsonw_uint(json_wtr, map_ids[i]); - jsonw_end_array(json_wtr); - } else { - printf(" map_ids "); - for (i = 0; i < info.nr_map_ids; i++) - printf("%u%s", map_ids[i], - i == info.nr_map_ids - 1 ? "" : ","); + jsonw_reset(btf_wtr); + err = btf_dumper_type(&d, t_var->type, value + vsi->offset); + if (err) { + p_err("btf dump failed: %d", err); + break; + } + } + if (printed_header) + jsonw_destroy(&btf_wtr); } + +out_free: + btf__free(btf); + free(value); } -static void print_prog_header_json(struct bpf_prog_info *info) +static void print_prog_header_json(struct bpf_prog_info *info, int fd) { + const char *prog_type_str; + char prog_name[MAX_PROG_FULL_NAME]; + jsonw_uint_field(json_wtr, "id", info->id); - if (info->type < ARRAY_SIZE(prog_type_name)) - jsonw_string_field(json_wtr, "type", - prog_type_name[info->type]); + prog_type_str = libbpf_bpf_prog_type_str(info->type); + + if (prog_type_str) + jsonw_string_field(json_wtr, "type", prog_type_str); else jsonw_uint_field(json_wtr, "type", info->type); - if (*info->name) - jsonw_string_field(json_wtr, "name", info->name); + if (*info->name) { + get_prog_full_name(info, fd, prog_name, sizeof(prog_name)); + jsonw_string_field(json_wtr, "name", prog_name); + } jsonw_name(json_wtr, "tag"); jsonw_printf(json_wtr, "\"" BPF_TAG_FMT "\"", @@ -285,6 +436,8 @@ static void print_prog_header_json(struct bpf_prog_info *info) jsonw_uint_field(json_wtr, "run_time_ns", info->run_time_ns); jsonw_uint_field(json_wtr, "run_cnt", info->run_cnt); } + if (info->recursion_misses) + jsonw_uint_field(json_wtr, "recursion_misses", info->recursion_misses); } static void print_prog_json(struct bpf_prog_info *info, int fd) @@ -292,7 +445,7 @@ static void print_prog_json(struct bpf_prog_info *info, int fd) char *memlock; jsonw_start_object(json_wtr); - print_prog_header_json(info); + print_prog_header_json(info, fd); print_dev_json(info->ifindex, info->netns_dev, info->netns_ino); if (info->load_time) { @@ -317,7 +470,7 @@ static void print_prog_json(struct bpf_prog_info *info, int fd) memlock = get_fdinfo(fd, "memlock"); if (memlock) - jsonw_int_field(json_wtr, "bytes_memlock", atoi(memlock)); + jsonw_int_field(json_wtr, "bytes_memlock", atoll(memlock)); free(memlock); if (info->nr_map_ids) @@ -326,31 +479,40 @@ static void print_prog_json(struct bpf_prog_info *info, int fd) if (info->btf_id) jsonw_int_field(json_wtr, "btf_id", info->btf_id); - if (!hash_empty(prog_table.table)) { - struct pinned_obj *obj; + if (!hashmap__empty(prog_table)) { + struct hashmap_entry *entry; jsonw_name(json_wtr, "pinned"); jsonw_start_array(json_wtr); - hash_for_each_possible(prog_table.table, obj, hash, info->id) { - if (obj->id == info->id) - jsonw_string(json_wtr, obj->path); - } + hashmap__for_each_key_entry(prog_table, entry, + u32_as_hash_field(info->id)) + jsonw_string(json_wtr, entry->value); jsonw_end_array(json_wtr); } + emit_obj_refs_json(refs_table, info->id, json_wtr); + + show_prog_metadata(fd, info->nr_map_ids); + jsonw_end_object(json_wtr); } -static void print_prog_header_plain(struct bpf_prog_info *info) +static void print_prog_header_plain(struct bpf_prog_info *info, int fd) { + const char *prog_type_str; + char prog_name[MAX_PROG_FULL_NAME]; + printf("%u: ", info->id); - if (info->type < ARRAY_SIZE(prog_type_name)) - printf("%s ", prog_type_name[info->type]); + prog_type_str = libbpf_bpf_prog_type_str(info->type); + if (prog_type_str) + printf("%s ", prog_type_str); else printf("type %u ", info->type); - if (*info->name) - printf("name %s ", info->name); + if (*info->name) { + get_prog_full_name(info, fd, prog_name, sizeof(prog_name)); + printf("name %s ", prog_name); + } printf("tag "); fprint_hex(stdout, info->tag, BPF_TAG_SIZE, ""); @@ -359,6 +521,8 @@ static void print_prog_header_plain(struct bpf_prog_info *info) if (info->run_time_ns) printf(" run_time_ns %lld run_cnt %lld", info->run_time_ns, info->run_cnt); + if (info->recursion_misses) + printf(" recursion_misses %lld", info->recursion_misses); printf("\n"); } @@ -366,7 +530,7 @@ static void print_prog_plain(struct bpf_prog_info *info, int fd) { char *memlock; - print_prog_header_plain(info); + print_prog_header_plain(info, fd); if (info->load_time) { char buf[32]; @@ -392,19 +556,22 @@ static void print_prog_plain(struct bpf_prog_info *info, int fd) if (info->nr_map_ids) show_prog_maps(fd, info->nr_map_ids); - if (!hash_empty(prog_table.table)) { - struct pinned_obj *obj; + if (!hashmap__empty(prog_table)) { + struct hashmap_entry *entry; - hash_for_each_possible(prog_table.table, obj, hash, info->id) { - if (obj->id == info->id) - printf("\n\tpinned %s", obj->path); - } + hashmap__for_each_key_entry(prog_table, entry, + u32_as_hash_field(info->id)) + printf("\n\tpinned %s", (char *)entry->value); } if (info->btf_id) printf("\n\tbtf_id %d", info->btf_id); + emit_obj_refs_plain(refs_table, info->id, "\n\tpids "); + printf("\n"); + + show_prog_metadata(fd, info->nr_map_ids); } static int show_prog(int fd) @@ -467,8 +634,16 @@ static int do_show(int argc, char **argv) int err; int fd; - if (show_pinned) - build_pinned_obj_table(&prog_table, BPF_OBJ_PROG); + if (show_pinned) { + prog_table = hashmap__new(hash_fn_for_key_as_id, + equal_fn_for_key_as_id, NULL); + if (IS_ERR(prog_table)) { + p_err("failed to create hashmap for pinned paths"); + return -1; + } + build_pinned_obj_table(prog_table, BPF_OBJ_PROG); + } + build_obj_refs_table(&refs_table, BPF_OBJ_PROG); if (argc == 2) return do_show_subset(argc, argv); @@ -510,6 +685,11 @@ static int do_show(int argc, char **argv) if (json_output) jsonw_end_array(json_wtr); + delete_obj_refs_table(refs_table); + + if (show_pinned) + delete_pinned_obj_table(prog_table); + return err; } @@ -525,31 +705,34 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, char func_sig[1024]; unsigned char *buf; __u32 member_len; + int fd, err = -1; ssize_t n; - int fd; if (mode == DUMP_JITED) { if (info->jited_prog_len == 0 || !info->jited_prog_insns) { p_info("no instructions returned"); return -1; } - buf = (unsigned char *)(info->jited_prog_insns); + buf = u64_to_ptr(info->jited_prog_insns); member_len = info->jited_prog_len; } else { /* DUMP_XLATED */ if (info->xlated_prog_len == 0 || !info->xlated_prog_insns) { p_err("error retrieving insn dump: kernel.kptr_restrict set?"); return -1; } - buf = (unsigned char *)info->xlated_prog_insns; + buf = u64_to_ptr(info->xlated_prog_insns); member_len = info->xlated_prog_len; } - if (info->btf_id && btf__get_from_id(info->btf_id, &btf)) { - p_err("failed to get btf"); - return -1; + if (info->btf_id) { + btf = btf__load_from_kernel_by_id(info->btf_id); + if (libbpf_get_error(btf)) { + p_err("failed to get btf"); + return -1; + } } - func_info = (void *)info->func_info; + func_info = u64_to_ptr(info->func_info); if (info->nr_line_info) { prog_linfo = bpf_prog_linfo__new(info); @@ -562,15 +745,15 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, if (fd < 0) { p_err("can't open file %s: %s", filepath, strerror(errno)); - return -1; + goto exit_free; } n = write(fd, buf, member_len); close(fd); - if (n != member_len) { + if (n != (ssize_t)member_len) { p_err("error writing output file: %s", n < 0 ? strerror(errno) : "short write"); - return -1; + goto exit_free; } if (json_output) @@ -584,7 +767,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, info->netns_ino, &disasm_opt); if (!name) - return -1; + goto exit_free; } if (info->nr_jited_func_lens && info->jited_func_lens) { @@ -597,13 +780,13 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, __u32 i; if (info->nr_jited_ksyms) { kernel_syms_load(&dd); - ksyms = (__u64 *) info->jited_ksyms; + ksyms = u64_to_ptr(info->jited_ksyms); } if (json_output) jsonw_start_array(json_wtr); - lens = (__u32 *) info->jited_func_lens; + lens = u64_to_ptr(info->jited_func_lens); for (i = 0; i < info->nr_jited_func_lens; i++) { if (ksyms) { sym = kernel_syms_search(&dd, ksyms[i]); @@ -664,7 +847,7 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, } else { kernel_syms_load(&dd); dd.nr_jited_ksyms = info->nr_jited_ksyms; - dd.jited_ksyms = (__u64 *) info->jited_ksyms; + dd.jited_ksyms = u64_to_ptr(info->jited_ksyms); dd.btf = btf; dd.func_info = func_info; dd.finfo_rec_size = info->func_info_rec_size; @@ -679,21 +862,28 @@ prog_dump(struct bpf_prog_info *info, enum dump_mode mode, kernel_syms_destroy(&dd); } - return 0; + err = 0; + +exit_free: + btf__free(btf); + bpf_prog_linfo__free(prog_linfo); + return err; } static int do_dump(int argc, char **argv) { - struct bpf_prog_info_linear *info_linear; + struct bpf_prog_info info; + __u32 info_len = sizeof(info); + size_t info_data_sz = 0; + void *info_data = NULL; char *filepath = NULL; bool opcodes = false; bool visual = false; enum dump_mode mode; bool linum = false; - int *fds = NULL; int nb_fds, i = 0; + int *fds = NULL; int err = -1; - __u64 arrays; if (is_prefix(*argv, "jited")) { if (disasm_init()) @@ -753,43 +943,44 @@ static int do_dump(int argc, char **argv) goto exit_close; } - if (mode == DUMP_JITED) - arrays = 1UL << BPF_PROG_INFO_JITED_INSNS; - else - arrays = 1UL << BPF_PROG_INFO_XLATED_INSNS; - - arrays |= 1UL << BPF_PROG_INFO_JITED_KSYMS; - arrays |= 1UL << BPF_PROG_INFO_JITED_FUNC_LENS; - arrays |= 1UL << BPF_PROG_INFO_FUNC_INFO; - arrays |= 1UL << BPF_PROG_INFO_LINE_INFO; - arrays |= 1UL << BPF_PROG_INFO_JITED_LINE_INFO; - if (json_output && nb_fds > 1) jsonw_start_array(json_wtr); /* root array */ for (i = 0; i < nb_fds; i++) { - info_linear = bpf_program__get_prog_info_linear(fds[i], arrays); - if (IS_ERR_OR_NULL(info_linear)) { + memset(&info, 0, sizeof(info)); + + err = bpf_obj_get_info_by_fd(fds[i], &info, &info_len); + if (err) { + p_err("can't get prog info: %s", strerror(errno)); + break; + } + + err = prep_prog_info(&info, mode, &info_data, &info_data_sz); + if (err) { + p_err("can't grow prog info_data"); + break; + } + + err = bpf_obj_get_info_by_fd(fds[i], &info, &info_len); + if (err) { p_err("can't get prog info: %s", strerror(errno)); break; } if (json_output && nb_fds > 1) { jsonw_start_object(json_wtr); /* prog object */ - print_prog_header_json(&info_linear->info); + print_prog_header_json(&info, fds[i]); jsonw_name(json_wtr, "insns"); } else if (nb_fds > 1) { - print_prog_header_plain(&info_linear->info); + print_prog_header_plain(&info, fds[i]); } - err = prog_dump(&info_linear->info, mode, filepath, opcodes, - visual, linum); + err = prog_dump(&info, mode, filepath, opcodes, visual, linum); if (json_output && nb_fds > 1) jsonw_end_object(json_wtr); /* prog object */ else if (i != nb_fds - 1 && nb_fds > 1) printf("\n"); - free(info_linear); if (err) break; close(fds[i]); @@ -801,6 +992,7 @@ exit_close: for (; i < nb_fds; i++) close(fds[i]); exit_free: + free(info_data); free(fds); return err; } @@ -809,7 +1001,7 @@ static int do_pin(int argc, char **argv) { int err; - err = do_pin_any(argc, argv, bpf_prog_get_fd_by_id); + err = do_pin_any(argc, argv, prog_parse_fd); if (!err && json_output) jsonw_null(json_wtr); return err; @@ -846,7 +1038,7 @@ static int parse_attach_detach_args(int argc, char **argv, int *progfd, } if (*attach_type == BPF_FLOW_DISSECTOR) { - *mapfd = -1; + *mapfd = 0; return 0; } @@ -1068,12 +1260,12 @@ static int do_run(int argc, char **argv) { char *data_fname_in = NULL, *data_fname_out = NULL; char *ctx_fname_in = NULL, *ctx_fname_out = NULL; - struct bpf_prog_test_run_attr test_attr = {0}; const unsigned int default_size = SZ_32K; void *data_in = NULL, *data_out = NULL; void *ctx_in = NULL, *ctx_out = NULL; unsigned int repeat = 1; int fd, err; + LIBBPF_OPTS(bpf_test_run_opts, test_attr); if (!REQ_ARGS(4)) return -1; @@ -1191,14 +1383,13 @@ static int do_run(int argc, char **argv) goto free_ctx_in; } - test_attr.prog_fd = fd; test_attr.repeat = repeat; test_attr.data_in = data_in; test_attr.data_out = data_out; test_attr.ctx_in = ctx_in; test_attr.ctx_out = ctx_out; - err = bpf_prog_test_run_xattr(&test_attr); + err = bpf_prog_test_run_opts(fd, &test_attr); if (err) { p_err("failed to run program: %s", strerror(errno)); goto free_ctx_out; @@ -1243,13 +1434,31 @@ free_data_in: return err; } +static int +get_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type, + enum bpf_attach_type *expected_attach_type) +{ + libbpf_print_fn_t print_backup; + int ret; + + ret = libbpf_prog_type_by_name(name, prog_type, expected_attach_type); + if (!ret) + return ret; + + /* libbpf_prog_type_by_name() failed, let's re-run with debug level */ + print_backup = libbpf_set_print(print_all_levels); + ret = libbpf_prog_type_by_name(name, prog_type, expected_attach_type); + libbpf_set_print(print_backup); + + return ret; +} + static int load_with_options(int argc, char **argv, bool first_prog_only) { enum bpf_prog_type common_prog_type = BPF_PROG_TYPE_UNSPEC; DECLARE_LIBBPF_OPTS(bpf_object_open_opts, open_opts, .relaxed_maps = relaxed_maps, ); - struct bpf_object_load_attr load_attr = { 0 }; enum bpf_attach_type expected_attach_type; struct map_replace *map_replace = NULL; struct bpf_program *prog = NULL, *pos; @@ -1271,8 +1480,6 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) while (argc) { if (is_prefix(*argv, "type")) { - char *type; - NEXT_ARG(); if (common_prog_type != BPF_PROG_TYPE_UNSPEC) { @@ -1282,21 +1489,26 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) if (!REQ_ARGS(1)) goto err_free_reuse_maps; - /* Put a '/' at the end of type to appease libbpf */ - type = malloc(strlen(*argv) + 2); - if (!type) { - p_err("mem alloc failed"); - goto err_free_reuse_maps; - } - *type = 0; - strcat(type, *argv); - strcat(type, "/"); - - err = libbpf_prog_type_by_name(type, &common_prog_type, + err = libbpf_prog_type_by_name(*argv, &common_prog_type, &expected_attach_type); - free(type); - if (err < 0) - goto err_free_reuse_maps; + if (err < 0) { + /* Put a '/' at the end of type to appease libbpf */ + char *type = malloc(strlen(*argv) + 2); + + if (!type) { + p_err("mem alloc failed"); + goto err_free_reuse_maps; + } + *type = 0; + strcat(type, *argv); + strcat(type, "/"); + + err = get_prog_type_by_name(type, &common_prog_type, + &expected_attach_type); + free(type); + if (err < 0) + goto err_free_reuse_maps; + } NEXT_ARG(); } else if (is_prefix(*argv, "map")) { @@ -1334,9 +1546,9 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) if (fd < 0) goto err_free_reuse_maps; - new_map_replace = reallocarray(map_replace, - old_map_fds + 1, - sizeof(*map_replace)); + new_map_replace = libbpf_reallocarray(map_replace, + old_map_fds + 1, + sizeof(*map_replace)); if (!new_map_replace) { p_err("mem alloc failed"); goto err_free_reuse_maps; @@ -1380,8 +1592,12 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) set_max_rlimit(); + if (verifier_logs) + /* log_level1 + log_level2 + stats, but not stable UAPI */ + open_opts.kernel_log_level = 1 + 2 + 4; + obj = bpf_object__open_file(file, &open_opts); - if (IS_ERR_OR_NULL(obj)) { + if (libbpf_get_error(obj)) { p_err("failed to open object file"); goto err_free_reuse_maps; } @@ -1390,10 +1606,10 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) enum bpf_prog_type prog_type = common_prog_type; if (prog_type == BPF_PROG_TYPE_UNSPEC) { - const char *sec_name = bpf_program__title(pos, false); + const char *sec_name = bpf_program__section_name(pos); - err = libbpf_prog_type_by_name(sec_name, &prog_type, - &expected_attach_type); + err = get_prog_type_by_name(sec_name, &prog_type, + &expected_attach_type); if (err < 0) goto err_close_obj; } @@ -1434,7 +1650,7 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) j = 0; idx = 0; bpf_object__for_each_map(map, obj) { - if (!bpf_map__is_offload_neutral(map)) + if (bpf_map__type(map) != BPF_MAP_TYPE_PERF_EVENT_ARRAY) bpf_map__set_ifindex(map, ifindex); if (j < old_map_fds && idx == map_replace[j].idx) { @@ -1459,12 +1675,7 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) goto err_close_obj; } - load_attr.obj = obj; - if (verifier_logs) - /* log_level1 + log_level2 + stats, but not stable UAPI */ - load_attr.log_level = 1 + 2 + 4; - - err = bpf_object__load_xattr(&load_attr); + err = bpf_object__load(obj); if (err) { p_err("failed to load object file"); goto err_close_obj; @@ -1475,7 +1686,7 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) goto err_close_obj; if (first_prog_only) { - prog = bpf_program__next(NULL, obj); + prog = bpf_object__next_program(obj, NULL); if (!prog) { p_err("object file doesn't contain any bpf program"); goto err_close_obj; @@ -1484,7 +1695,7 @@ static int load_with_options(int argc, char **argv, bool first_prog_only) err = bpf_obj_pin(bpf_program__fd(prog), pinfile); if (err) { p_err("failed to pin program %s", - bpf_program__title(prog, false)); + bpf_program__section_name(prog)); goto err_close_obj; } } else { @@ -1519,6 +1730,11 @@ err_unpin: else bpf_object__unpin_programs(obj, pinfile); err_close_obj: + if (!legacy_libbpf) { + p_info("Warning: bpftool is now running in libbpf strict mode and has more stringent requirements about BPF programs.\n" + "If it used to work for this object file but now doesn't, see --legacy option for more details.\n"); + } + bpf_object__close(obj); err_free_reuse_maps: for (i = 0; i < old_map_fds; i++) @@ -1527,8 +1743,110 @@ err_free_reuse_maps: return -1; } +static int count_open_fds(void) +{ + DIR *dp = opendir("/proc/self/fd"); + struct dirent *de; + int cnt = -3; + + if (!dp) + return -1; + + while ((de = readdir(dp))) + cnt++; + + closedir(dp); + return cnt; +} + +static int try_loader(struct gen_loader_opts *gen) +{ + struct bpf_load_and_run_opts opts = {}; + struct bpf_loader_ctx *ctx; + int ctx_sz = sizeof(*ctx) + 64 * max(sizeof(struct bpf_map_desc), + sizeof(struct bpf_prog_desc)); + int log_buf_sz = (1u << 24) - 1; + int err, fds_before, fd_delta; + char *log_buf = NULL; + + ctx = alloca(ctx_sz); + memset(ctx, 0, ctx_sz); + ctx->sz = ctx_sz; + if (verifier_logs) { + ctx->log_level = 1 + 2 + 4; + ctx->log_size = log_buf_sz; + log_buf = malloc(log_buf_sz); + if (!log_buf) + return -ENOMEM; + ctx->log_buf = (long) log_buf; + } + opts.ctx = ctx; + opts.data = gen->data; + opts.data_sz = gen->data_sz; + opts.insns = gen->insns; + opts.insns_sz = gen->insns_sz; + fds_before = count_open_fds(); + err = bpf_load_and_run(&opts); + fd_delta = count_open_fds() - fds_before; + if (err < 0 || verifier_logs) { + fprintf(stderr, "err %d\n%s\n%s", err, opts.errstr, log_buf); + if (fd_delta && err < 0) + fprintf(stderr, "loader prog leaked %d FDs\n", + fd_delta); + } + free(log_buf); + return err; +} + +static int do_loader(int argc, char **argv) +{ + DECLARE_LIBBPF_OPTS(bpf_object_open_opts, open_opts); + DECLARE_LIBBPF_OPTS(gen_loader_opts, gen); + struct bpf_object *obj; + const char *file; + int err = 0; + + if (!REQ_ARGS(1)) + return -1; + file = GET_ARG(); + + if (verifier_logs) + /* log_level1 + log_level2 + stats, but not stable UAPI */ + open_opts.kernel_log_level = 1 + 2 + 4; + + obj = bpf_object__open_file(file, &open_opts); + if (libbpf_get_error(obj)) { + p_err("failed to open object file"); + goto err_close_obj; + } + + err = bpf_object__gen_loader(obj, &gen); + if (err) + goto err_close_obj; + + err = bpf_object__load(obj); + if (err) { + p_err("failed to load object file"); + goto err_close_obj; + } + + if (verifier_logs) { + struct dump_data dd = {}; + + kernel_syms_load(&dd); + dump_xlated_plain(&dd, (void *)gen.insns, gen.insns_sz, false, false); + kernel_syms_destroy(&dd); + } + err = try_loader(&gen); +err_close_obj: + bpf_object__close(obj); + return err; +} + static int do_load(int argc, char **argv) { + if (use_loader) + return do_loader(argc, argv); return load_with_options(argc, argv, true); } @@ -1537,6 +1855,473 @@ static int do_loadall(int argc, char **argv) return load_with_options(argc, argv, false); } +#ifdef BPFTOOL_WITHOUT_SKELETONS + +static int do_profile(int argc, char **argv) +{ + p_err("bpftool prog profile command is not supported. Please build bpftool with clang >= 10.0.0"); + return 0; +} + +#else /* BPFTOOL_WITHOUT_SKELETONS */ + +#include "profiler.skel.h" + +struct profile_metric { + const char *name; + struct bpf_perf_event_value val; + struct perf_event_attr attr; + bool selected; + + /* calculate ratios like instructions per cycle */ + const int ratio_metric; /* 0 for N/A, 1 for index 0 (cycles) */ + const char *ratio_desc; + const float ratio_mul; +} metrics[] = { + { + .name = "cycles", + .attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .exclude_user = 1, + }, + }, + { + .name = "instructions", + .attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_INSTRUCTIONS, + .exclude_user = 1, + }, + .ratio_metric = 1, + .ratio_desc = "insns per cycle", + .ratio_mul = 1.0, + }, + { + .name = "l1d_loads", + .attr = { + .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16), + .exclude_user = 1, + }, + }, + { + .name = "llc_misses", + .attr = { + .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_LL | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16), + .exclude_user = 1 + }, + .ratio_metric = 2, + .ratio_desc = "LLC misses per million insns", + .ratio_mul = 1e6, + }, + { + .name = "itlb_misses", + .attr = { + .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_ITLB | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16), + .exclude_user = 1 + }, + .ratio_metric = 2, + .ratio_desc = "itlb misses per million insns", + .ratio_mul = 1e6, + }, + { + .name = "dtlb_misses", + .attr = { + .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_DTLB | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16), + .exclude_user = 1 + }, + .ratio_metric = 2, + .ratio_desc = "dtlb misses per million insns", + .ratio_mul = 1e6, + }, +}; + +static __u64 profile_total_count; + +#define MAX_NUM_PROFILE_METRICS 4 + +static int profile_parse_metrics(int argc, char **argv) +{ + unsigned int metric_cnt; + int selected_cnt = 0; + unsigned int i; + + metric_cnt = ARRAY_SIZE(metrics); + + while (argc > 0) { + for (i = 0; i < metric_cnt; i++) { + if (is_prefix(argv[0], metrics[i].name)) { + if (!metrics[i].selected) + selected_cnt++; + metrics[i].selected = true; + break; + } + } + if (i == metric_cnt) { + p_err("unknown metric %s", argv[0]); + return -1; + } + NEXT_ARG(); + } + if (selected_cnt > MAX_NUM_PROFILE_METRICS) { + p_err("too many (%d) metrics, please specify no more than %d metrics at at time", + selected_cnt, MAX_NUM_PROFILE_METRICS); + return -1; + } + return selected_cnt; +} + +static void profile_read_values(struct profiler_bpf *obj) +{ + __u32 m, cpu, num_cpu = obj->rodata->num_cpu; + int reading_map_fd, count_map_fd; + __u64 counts[num_cpu]; + __u32 key = 0; + int err; + + reading_map_fd = bpf_map__fd(obj->maps.accum_readings); + count_map_fd = bpf_map__fd(obj->maps.counts); + if (reading_map_fd < 0 || count_map_fd < 0) { + p_err("failed to get fd for map"); + return; + } + + err = bpf_map_lookup_elem(count_map_fd, &key, counts); + if (err) { + p_err("failed to read count_map: %s", strerror(errno)); + return; + } + + profile_total_count = 0; + for (cpu = 0; cpu < num_cpu; cpu++) + profile_total_count += counts[cpu]; + + for (m = 0; m < ARRAY_SIZE(metrics); m++) { + struct bpf_perf_event_value values[num_cpu]; + + if (!metrics[m].selected) + continue; + + err = bpf_map_lookup_elem(reading_map_fd, &key, values); + if (err) { + p_err("failed to read reading_map: %s", + strerror(errno)); + return; + } + for (cpu = 0; cpu < num_cpu; cpu++) { + metrics[m].val.counter += values[cpu].counter; + metrics[m].val.enabled += values[cpu].enabled; + metrics[m].val.running += values[cpu].running; + } + key++; + } +} + +static void profile_print_readings_json(void) +{ + __u32 m; + + jsonw_start_array(json_wtr); + for (m = 0; m < ARRAY_SIZE(metrics); m++) { + if (!metrics[m].selected) + continue; + jsonw_start_object(json_wtr); + jsonw_string_field(json_wtr, "metric", metrics[m].name); + jsonw_lluint_field(json_wtr, "run_cnt", profile_total_count); + jsonw_lluint_field(json_wtr, "value", metrics[m].val.counter); + jsonw_lluint_field(json_wtr, "enabled", metrics[m].val.enabled); + jsonw_lluint_field(json_wtr, "running", metrics[m].val.running); + + jsonw_end_object(json_wtr); + } + jsonw_end_array(json_wtr); +} + +static void profile_print_readings_plain(void) +{ + __u32 m; + + printf("\n%18llu %-20s\n", profile_total_count, "run_cnt"); + for (m = 0; m < ARRAY_SIZE(metrics); m++) { + struct bpf_perf_event_value *val = &metrics[m].val; + int r; + + if (!metrics[m].selected) + continue; + printf("%18llu %-20s", val->counter, metrics[m].name); + + r = metrics[m].ratio_metric - 1; + if (r >= 0 && metrics[r].selected && + metrics[r].val.counter > 0) { + printf("# %8.2f %-30s", + val->counter * metrics[m].ratio_mul / + metrics[r].val.counter, + metrics[m].ratio_desc); + } else { + printf("%-41s", ""); + } + + if (val->enabled > val->running) + printf("(%4.2f%%)", + val->running * 100.0 / val->enabled); + printf("\n"); + } +} + +static void profile_print_readings(void) +{ + if (json_output) + profile_print_readings_json(); + else + profile_print_readings_plain(); +} + +static char *profile_target_name(int tgt_fd) +{ + struct bpf_func_info func_info; + struct bpf_prog_info info = {}; + __u32 info_len = sizeof(info); + const struct btf_type *t; + __u32 func_info_rec_size; + struct btf *btf = NULL; + char *name = NULL; + int err; + + err = bpf_obj_get_info_by_fd(tgt_fd, &info, &info_len); + if (err) { + p_err("failed to bpf_obj_get_info_by_fd for prog FD %d", tgt_fd); + goto out; + } + + if (info.btf_id == 0) { + p_err("prog FD %d doesn't have valid btf", tgt_fd); + goto out; + } + + func_info_rec_size = info.func_info_rec_size; + if (info.nr_func_info == 0) { + p_err("bpf_obj_get_info_by_fd for prog FD %d found 0 func_info", tgt_fd); + goto out; + } + + memset(&info, 0, sizeof(info)); + info.nr_func_info = 1; + info.func_info_rec_size = func_info_rec_size; + info.func_info = ptr_to_u64(&func_info); + + err = bpf_obj_get_info_by_fd(tgt_fd, &info, &info_len); + if (err) { + p_err("failed to get func_info for prog FD %d", tgt_fd); + goto out; + } + + btf = btf__load_from_kernel_by_id(info.btf_id); + if (libbpf_get_error(btf)) { + p_err("failed to load btf for prog FD %d", tgt_fd); + goto out; + } + + t = btf__type_by_id(btf, func_info.type_id); + if (!t) { + p_err("btf %d doesn't have type %d", + info.btf_id, func_info.type_id); + goto out; + } + name = strdup(btf__name_by_offset(btf, t->name_off)); +out: + btf__free(btf); + return name; +} + +static struct profiler_bpf *profile_obj; +static int profile_tgt_fd = -1; +static char *profile_tgt_name; +static int *profile_perf_events; +static int profile_perf_event_cnt; + +static void profile_close_perf_events(struct profiler_bpf *obj) +{ + int i; + + for (i = profile_perf_event_cnt - 1; i >= 0; i--) + close(profile_perf_events[i]); + + free(profile_perf_events); + profile_perf_event_cnt = 0; +} + +static int profile_open_perf_events(struct profiler_bpf *obj) +{ + unsigned int cpu, m; + int map_fd, pmu_fd; + + profile_perf_events = calloc( + sizeof(int), obj->rodata->num_cpu * obj->rodata->num_metric); + if (!profile_perf_events) { + p_err("failed to allocate memory for perf_event array: %s", + strerror(errno)); + return -1; + } + map_fd = bpf_map__fd(obj->maps.events); + if (map_fd < 0) { + p_err("failed to get fd for events map"); + return -1; + } + + for (m = 0; m < ARRAY_SIZE(metrics); m++) { + if (!metrics[m].selected) + continue; + for (cpu = 0; cpu < obj->rodata->num_cpu; cpu++) { + pmu_fd = syscall(__NR_perf_event_open, &metrics[m].attr, + -1/*pid*/, cpu, -1/*group_fd*/, 0); + if (pmu_fd < 0 || + bpf_map_update_elem(map_fd, &profile_perf_event_cnt, + &pmu_fd, BPF_ANY) || + ioctl(pmu_fd, PERF_EVENT_IOC_ENABLE, 0)) { + p_err("failed to create event %s on cpu %d", + metrics[m].name, cpu); + return -1; + } + profile_perf_events[profile_perf_event_cnt++] = pmu_fd; + } + } + return 0; +} + +static void profile_print_and_cleanup(void) +{ + profile_close_perf_events(profile_obj); + profile_read_values(profile_obj); + profile_print_readings(); + profiler_bpf__destroy(profile_obj); + + close(profile_tgt_fd); + free(profile_tgt_name); +} + +static void int_exit(int signo) +{ + profile_print_and_cleanup(); + exit(0); +} + +static int do_profile(int argc, char **argv) +{ + int num_metric, num_cpu, err = -1; + struct bpf_program *prog; + unsigned long duration; + char *endptr; + + /* we at least need two args for the prog and one metric */ + if (!REQ_ARGS(3)) + return -EINVAL; + + /* parse target fd */ + profile_tgt_fd = prog_parse_fd(&argc, &argv); + if (profile_tgt_fd < 0) { + p_err("failed to parse fd"); + return -1; + } + + /* parse profiling optional duration */ + if (argc > 2 && is_prefix(argv[0], "duration")) { + NEXT_ARG(); + duration = strtoul(*argv, &endptr, 0); + if (*endptr) + usage(); + NEXT_ARG(); + } else { + duration = UINT_MAX; + } + + num_metric = profile_parse_metrics(argc, argv); + if (num_metric <= 0) + goto out; + + num_cpu = libbpf_num_possible_cpus(); + if (num_cpu <= 0) { + p_err("failed to identify number of CPUs"); + goto out; + } + + profile_obj = profiler_bpf__open(); + if (!profile_obj) { + p_err("failed to open and/or load BPF object"); + goto out; + } + + profile_obj->rodata->num_cpu = num_cpu; + profile_obj->rodata->num_metric = num_metric; + + /* adjust map sizes */ + bpf_map__set_max_entries(profile_obj->maps.events, num_metric * num_cpu); + bpf_map__set_max_entries(profile_obj->maps.fentry_readings, num_metric); + bpf_map__set_max_entries(profile_obj->maps.accum_readings, num_metric); + bpf_map__set_max_entries(profile_obj->maps.counts, 1); + + /* change target name */ + profile_tgt_name = profile_target_name(profile_tgt_fd); + if (!profile_tgt_name) + goto out; + + bpf_object__for_each_program(prog, profile_obj->obj) { + err = bpf_program__set_attach_target(prog, profile_tgt_fd, + profile_tgt_name); + if (err) { + p_err("failed to set attach target\n"); + goto out; + } + } + + set_max_rlimit(); + err = profiler_bpf__load(profile_obj); + if (err) { + p_err("failed to load profile_obj"); + goto out; + } + + err = profile_open_perf_events(profile_obj); + if (err) + goto out; + + err = profiler_bpf__attach(profile_obj); + if (err) { + p_err("failed to attach profile_obj"); + goto out; + } + signal(SIGINT, int_exit); + + sleep(duration); + profile_print_and_cleanup(); + return 0; + +out: + profile_close_perf_events(profile_obj); + if (profile_obj) + profiler_bpf__destroy(profile_obj); + close(profile_tgt_fd); + free(profile_tgt_name); + return err; +} + +#endif /* BPFTOOL_WITHOUT_SKELETONS */ + static int do_help(int argc, char **argv) { if (json_output) { @@ -1545,23 +2330,24 @@ static int do_help(int argc, char **argv) } fprintf(stderr, - "Usage: %s %s { show | list } [PROG]\n" - " %s %s dump xlated PROG [{ file FILE | opcodes | visual | linum }]\n" - " %s %s dump jited PROG [{ file FILE | opcodes | linum }]\n" - " %s %s pin PROG FILE\n" - " %s %s { load | loadall } OBJ PATH \\\n" + "Usage: %1$s %2$s { show | list } [PROG]\n" + " %1$s %2$s dump xlated PROG [{ file FILE | opcodes | visual | linum }]\n" + " %1$s %2$s dump jited PROG [{ file FILE | opcodes | linum }]\n" + " %1$s %2$s pin PROG FILE\n" + " %1$s %2$s { load | loadall } OBJ PATH \\\n" " [type TYPE] [dev NAME] \\\n" " [map { idx IDX | name NAME } MAP]\\\n" " [pinmaps MAP_DIR]\n" - " %s %s attach PROG ATTACH_TYPE [MAP]\n" - " %s %s detach PROG ATTACH_TYPE [MAP]\n" - " %s %s run PROG \\\n" + " %1$s %2$s attach PROG ATTACH_TYPE [MAP]\n" + " %1$s %2$s detach PROG ATTACH_TYPE [MAP]\n" + " %1$s %2$s run PROG \\\n" " data_in FILE \\\n" " [data_out FILE [data_size_out L]] \\\n" " [ctx_in FILE [ctx_out FILE [ctx_size_out M]]] \\\n" " [repeat N]\n" - " %s %s tracelog\n" - " %s %s help\n" + " %1$s %2$s profile PROG [duration DURATION] METRICs\n" + " %1$s %2$s tracelog\n" + " %1$s %2$s help\n" "\n" " " HELP_SPEC_MAP "\n" " " HELP_SPEC_PROGRAM "\n" @@ -1572,16 +2358,18 @@ static int do_help(int argc, char **argv) " sk_reuseport | flow_dissector | cgroup/sysctl |\n" " cgroup/bind4 | cgroup/bind6 | cgroup/post_bind4 |\n" " cgroup/post_bind6 | cgroup/connect4 | cgroup/connect6 |\n" - " cgroup/sendmsg4 | cgroup/sendmsg6 | cgroup/recvmsg4 |\n" - " cgroup/recvmsg6 | cgroup/getsockopt |\n" - " cgroup/setsockopt }\n" - " ATTACH_TYPE := { msg_verdict | stream_verdict | stream_parser |\n" - " flow_dissector }\n" - " " HELP_SPEC_OPTIONS "\n" + " cgroup/getpeername4 | cgroup/getpeername6 |\n" + " cgroup/getsockname4 | cgroup/getsockname6 | cgroup/sendmsg4 |\n" + " cgroup/sendmsg6 | cgroup/recvmsg4 | cgroup/recvmsg6 |\n" + " cgroup/getsockopt | cgroup/setsockopt | cgroup/sock_release |\n" + " struct_ops | fentry | fexit | freplace | sk_lookup }\n" + " ATTACH_TYPE := { sk_msg_verdict | sk_skb_verdict | sk_skb_stream_verdict |\n" + " sk_skb_stream_parser | flow_dissector }\n" + " METRIC := { cycles | instructions | l1d_loads | llc_misses | itlb_misses | dtlb_misses }\n" + " " HELP_SPEC_OPTIONS " |\n" + " {-f|--bpffs} | {-m|--mapcompat} | {-n|--nomount} |\n" + " {-L|--use-loader} }\n" "", - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], - bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2], bin_name, argv[-2]); return 0; @@ -1599,6 +2387,7 @@ static const struct cmd cmds[] = { { "detach", do_detach }, { "tracelog", do_tracelog }, { "run", do_run }, + { "profile", do_profile }, { 0 } }; diff --git a/tools/bpf/bpftool/skeleton/pid_iter.bpf.c b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c new file mode 100644 index 000000000000..eb05ea53afb1 --- /dev/null +++ b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c @@ -0,0 +1,103 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +/* Copyright (c) 2020 Facebook */ +#include <vmlinux.h> +#include <bpf/bpf_helpers.h> +#include <bpf/bpf_core_read.h> +#include <bpf/bpf_tracing.h> +#include "pid_iter.h" + +/* keep in sync with the definition in main.h */ +enum bpf_obj_type { + BPF_OBJ_UNKNOWN, + BPF_OBJ_PROG, + BPF_OBJ_MAP, + BPF_OBJ_LINK, + BPF_OBJ_BTF, +}; + +extern const void bpf_link_fops __ksym; +extern const void bpf_map_fops __ksym; +extern const void bpf_prog_fops __ksym; +extern const void btf_fops __ksym; + +const volatile enum bpf_obj_type obj_type = BPF_OBJ_UNKNOWN; + +static __always_inline __u32 get_obj_id(void *ent, enum bpf_obj_type type) +{ + switch (type) { + case BPF_OBJ_PROG: + return BPF_CORE_READ((struct bpf_prog *)ent, aux, id); + case BPF_OBJ_MAP: + return BPF_CORE_READ((struct bpf_map *)ent, id); + case BPF_OBJ_BTF: + return BPF_CORE_READ((struct btf *)ent, id); + case BPF_OBJ_LINK: + return BPF_CORE_READ((struct bpf_link *)ent, id); + default: + return 0; + } +} + +/* could be used only with BPF_LINK_TYPE_PERF_EVENT links */ +static __u64 get_bpf_cookie(struct bpf_link *link) +{ + struct bpf_perf_link *perf_link; + struct perf_event *event; + + perf_link = container_of(link, struct bpf_perf_link, link); + event = BPF_CORE_READ(perf_link, perf_file, private_data); + return BPF_CORE_READ(event, bpf_cookie); +} + +SEC("iter/task_file") +int iter(struct bpf_iter__task_file *ctx) +{ + struct file *file = ctx->file; + struct task_struct *task = ctx->task; + struct pid_iter_entry e; + const void *fops; + + if (!file || !task) + return 0; + + switch (obj_type) { + case BPF_OBJ_PROG: + fops = &bpf_prog_fops; + break; + case BPF_OBJ_MAP: + fops = &bpf_map_fops; + break; + case BPF_OBJ_BTF: + fops = &btf_fops; + break; + case BPF_OBJ_LINK: + fops = &bpf_link_fops; + break; + default: + return 0; + } + + if (file->f_op != fops) + return 0; + + __builtin_memset(&e, 0, sizeof(e)); + e.pid = task->tgid; + e.id = get_obj_id(file->private_data, obj_type); + + if (obj_type == BPF_OBJ_LINK) { + struct bpf_link *link = (struct bpf_link *) file->private_data; + + if (BPF_CORE_READ(link, type) == BPF_LINK_TYPE_PERF_EVENT) { + e.has_bpf_cookie = true; + e.bpf_cookie = get_bpf_cookie(link); + } + } + + bpf_probe_read_kernel_str(&e.comm, sizeof(e.comm), + task->group_leader->comm); + bpf_seq_write(ctx->meta->seq, &e, sizeof(e)); + + return 0; +} + +char LICENSE[] SEC("license") = "Dual BSD/GPL"; diff --git a/tools/bpf/bpftool/skeleton/pid_iter.h b/tools/bpf/bpftool/skeleton/pid_iter.h new file mode 100644 index 000000000000..bbb570d4cca6 --- /dev/null +++ b/tools/bpf/bpftool/skeleton/pid_iter.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ +/* Copyright (c) 2020 Facebook */ +#ifndef __PID_ITER_H +#define __PID_ITER_H + +struct pid_iter_entry { + __u32 id; + int pid; + __u64 bpf_cookie; + bool has_bpf_cookie; + char comm[16]; +}; + +#endif diff --git a/tools/bpf/bpftool/skeleton/profiler.bpf.c b/tools/bpf/bpftool/skeleton/profiler.bpf.c new file mode 100644 index 000000000000..ce5b65e07ab1 --- /dev/null +++ b/tools/bpf/bpftool/skeleton/profiler.bpf.c @@ -0,0 +1,118 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +// Copyright (c) 2020 Facebook +#include <vmlinux.h> +#include <bpf/bpf_helpers.h> +#include <bpf/bpf_tracing.h> + +/* map of perf event fds, num_cpu * num_metric entries */ +struct { + __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); + __uint(key_size, sizeof(u32)); + __uint(value_size, sizeof(int)); +} events SEC(".maps"); + +/* readings at fentry */ +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(key_size, sizeof(u32)); + __uint(value_size, sizeof(struct bpf_perf_event_value)); +} fentry_readings SEC(".maps"); + +/* accumulated readings */ +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(key_size, sizeof(u32)); + __uint(value_size, sizeof(struct bpf_perf_event_value)); +} accum_readings SEC(".maps"); + +/* sample counts, one per cpu */ +struct { + __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY); + __uint(key_size, sizeof(u32)); + __uint(value_size, sizeof(u64)); +} counts SEC(".maps"); + +const volatile __u32 num_cpu = 1; +const volatile __u32 num_metric = 1; +#define MAX_NUM_MATRICS 4 + +SEC("fentry/XXX") +int BPF_PROG(fentry_XXX) +{ + struct bpf_perf_event_value *ptrs[MAX_NUM_MATRICS]; + u32 key = bpf_get_smp_processor_id(); + u32 i; + + /* look up before reading, to reduce error */ + for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) { + u32 flag = i; + + ptrs[i] = bpf_map_lookup_elem(&fentry_readings, &flag); + if (!ptrs[i]) + return 0; + } + + for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) { + struct bpf_perf_event_value reading; + int err; + + err = bpf_perf_event_read_value(&events, key, &reading, + sizeof(reading)); + if (err) + return 0; + *(ptrs[i]) = reading; + key += num_cpu; + } + + return 0; +} + +static inline void +fexit_update_maps(u32 id, struct bpf_perf_event_value *after) +{ + struct bpf_perf_event_value *before, diff; + + before = bpf_map_lookup_elem(&fentry_readings, &id); + /* only account samples with a valid fentry_reading */ + if (before && before->counter) { + struct bpf_perf_event_value *accum; + + diff.counter = after->counter - before->counter; + diff.enabled = after->enabled - before->enabled; + diff.running = after->running - before->running; + + accum = bpf_map_lookup_elem(&accum_readings, &id); + if (accum) { + accum->counter += diff.counter; + accum->enabled += diff.enabled; + accum->running += diff.running; + } + } +} + +SEC("fexit/XXX") +int BPF_PROG(fexit_XXX) +{ + struct bpf_perf_event_value readings[MAX_NUM_MATRICS]; + u32 cpu = bpf_get_smp_processor_id(); + u32 i, zero = 0; + int err; + u64 *count; + + /* read all events before updating the maps, to reduce error */ + for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) { + err = bpf_perf_event_read_value(&events, cpu + i * num_cpu, + readings + i, sizeof(*readings)); + if (err) + return 0; + } + count = bpf_map_lookup_elem(&counts, &zero); + if (count) { + *count += 1; + for (i = 0; i < num_metric && i < MAX_NUM_MATRICS; i++) + fexit_update_maps(i, &readings[i]); + } + return 0; +} + +char LICENSE[] SEC("license") = "Dual BSD/GPL"; diff --git a/tools/bpf/bpftool/struct_ops.c b/tools/bpf/bpftool/struct_ops.c new file mode 100644 index 000000000000..e08a6ff2866c --- /dev/null +++ b/tools/bpf/bpftool/struct_ops.c @@ -0,0 +1,600 @@ +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +/* Copyright (C) 2020 Facebook */ + +#include <errno.h> +#include <stdio.h> +#include <unistd.h> + +#include <linux/err.h> + +#include <bpf/bpf.h> +#include <bpf/btf.h> +#include <bpf/libbpf.h> + +#include "json_writer.h" +#include "main.h" + +#define STRUCT_OPS_VALUE_PREFIX "bpf_struct_ops_" + +static const struct btf_type *map_info_type; +static __u32 map_info_alloc_len; +static struct btf *btf_vmlinux; +static __s32 map_info_type_id; + +struct res { + unsigned int nr_maps; + unsigned int nr_errs; +}; + +static const struct btf *get_btf_vmlinux(void) +{ + if (btf_vmlinux) + return btf_vmlinux; + + btf_vmlinux = libbpf_find_kernel_btf(); + if (libbpf_get_error(btf_vmlinux)) + p_err("struct_ops requires kernel CONFIG_DEBUG_INFO_BTF=y"); + + return btf_vmlinux; +} + +static const char *get_kern_struct_ops_name(const struct bpf_map_info *info) +{ + const struct btf *kern_btf; + const struct btf_type *t; + const char *st_ops_name; + + kern_btf = get_btf_vmlinux(); + if (libbpf_get_error(kern_btf)) + return "<btf_vmlinux_not_found>"; + + t = btf__type_by_id(kern_btf, info->btf_vmlinux_value_type_id); + st_ops_name = btf__name_by_offset(kern_btf, t->name_off); + st_ops_name += strlen(STRUCT_OPS_VALUE_PREFIX); + + return st_ops_name; +} + +static __s32 get_map_info_type_id(void) +{ + const struct btf *kern_btf; + + if (map_info_type_id) + return map_info_type_id; + + kern_btf = get_btf_vmlinux(); + if (libbpf_get_error(kern_btf)) { + map_info_type_id = PTR_ERR(kern_btf); + return map_info_type_id; + } + + map_info_type_id = btf__find_by_name_kind(kern_btf, "bpf_map_info", + BTF_KIND_STRUCT); + if (map_info_type_id < 0) { + p_err("can't find bpf_map_info from btf_vmlinux"); + return map_info_type_id; + } + map_info_type = btf__type_by_id(kern_btf, map_info_type_id); + + /* Ensure map_info_alloc() has at least what the bpftool needs */ + map_info_alloc_len = map_info_type->size; + if (map_info_alloc_len < sizeof(struct bpf_map_info)) + map_info_alloc_len = sizeof(struct bpf_map_info); + + return map_info_type_id; +} + +/* If the subcmd needs to print out the bpf_map_info, + * it should always call map_info_alloc to allocate + * a bpf_map_info object instead of allocating it + * on the stack. + * + * map_info_alloc() will take the running kernel's btf + * into account. i.e. it will consider the + * sizeof(struct bpf_map_info) of the running kernel. + * + * It will enable the "struct_ops" cmd to print the latest + * "struct bpf_map_info". + * + * [ Recall that "struct_ops" requires the kernel's btf to + * be available ] + */ +static struct bpf_map_info *map_info_alloc(__u32 *alloc_len) +{ + struct bpf_map_info *info; + + if (get_map_info_type_id() < 0) + return NULL; + + info = calloc(1, map_info_alloc_len); + if (!info) + p_err("mem alloc failed"); + else + *alloc_len = map_info_alloc_len; + + return info; +} + +/* It iterates all struct_ops maps of the system. + * It returns the fd in "*res_fd" and map_info in "*info". + * In the very first iteration, info->id should be 0. + * An optional map "*name" filter can be specified. + * The filter can be made more flexible in the future. + * e.g. filter by kernel-struct-ops-name, regex-name, glob-name, ...etc. + * + * Return value: + * 1: A struct_ops map found. It is returned in "*res_fd" and "*info". + * The caller can continue to call get_next in the future. + * 0: No struct_ops map is returned. + * All struct_ops map has been found. + * -1: Error and the caller should abort the iteration. + */ +static int get_next_struct_ops_map(const char *name, int *res_fd, + struct bpf_map_info *info, __u32 info_len) +{ + __u32 id = info->id; + int err, fd; + + while (true) { + err = bpf_map_get_next_id(id, &id); + if (err) { + if (errno == ENOENT) + return 0; + p_err("can't get next map: %s", strerror(errno)); + return -1; + } + + fd = bpf_map_get_fd_by_id(id); + if (fd < 0) { + if (errno == ENOENT) + continue; + p_err("can't get map by id (%u): %s", + id, strerror(errno)); + return -1; + } + + err = bpf_obj_get_info_by_fd(fd, info, &info_len); + if (err) { + p_err("can't get map info: %s", strerror(errno)); + close(fd); + return -1; + } + + if (info->type == BPF_MAP_TYPE_STRUCT_OPS && + (!name || !strcmp(name, info->name))) { + *res_fd = fd; + return 1; + } + close(fd); + } +} + +static int cmd_retval(const struct res *res, bool must_have_one_map) +{ + if (res->nr_errs || (!res->nr_maps && must_have_one_map)) + return -1; + + return 0; +} + +/* "data" is the work_func private storage */ +typedef int (*work_func)(int fd, const struct bpf_map_info *info, void *data, + struct json_writer *wtr); + +/* Find all struct_ops map in the system. + * Filter out by "name" (if specified). + * Then call "func(fd, info, data, wtr)" on each struct_ops map found. + */ +static struct res do_search(const char *name, work_func func, void *data, + struct json_writer *wtr) +{ + struct bpf_map_info *info; + struct res res = {}; + __u32 info_len; + int fd, err; + + info = map_info_alloc(&info_len); + if (!info) { + res.nr_errs++; + return res; + } + + if (wtr) + jsonw_start_array(wtr); + while ((err = get_next_struct_ops_map(name, &fd, info, info_len)) == 1) { + res.nr_maps++; + err = func(fd, info, data, wtr); + if (err) + res.nr_errs++; + close(fd); + } + if (wtr) + jsonw_end_array(wtr); + + if (err) + res.nr_errs++; + + if (!wtr && name && !res.nr_errs && !res.nr_maps) + /* It is not printing empty []. + * Thus, needs to specifically say nothing found + * for "name" here. + */ + p_err("no struct_ops found for %s", name); + else if (!wtr && json_output && !res.nr_errs) + /* The "func()" above is not writing any json (i.e. !wtr + * test here). + * + * However, "-j" is enabled and there is no errs here, + * so call json_null() as the current convention of + * other cmds. + */ + jsonw_null(json_wtr); + + free(info); + return res; +} + +static struct res do_one_id(const char *id_str, work_func func, void *data, + struct json_writer *wtr) +{ + struct bpf_map_info *info; + struct res res = {}; + unsigned long id; + __u32 info_len; + char *endptr; + int fd; + + id = strtoul(id_str, &endptr, 0); + if (*endptr || !id || id > UINT32_MAX) { + p_err("invalid id %s", id_str); + res.nr_errs++; + return res; + } + + fd = bpf_map_get_fd_by_id(id); + if (fd < 0) { + p_err("can't get map by id (%lu): %s", id, strerror(errno)); + res.nr_errs++; + return res; + } + + info = map_info_alloc(&info_len); + if (!info) { + res.nr_errs++; + goto done; + } + + if (bpf_obj_get_info_by_fd(fd, info, &info_len)) { + p_err("can't get map info: %s", strerror(errno)); + res.nr_errs++; + goto done; + } + + if (info->type != BPF_MAP_TYPE_STRUCT_OPS) { + p_err("%s id %u is not a struct_ops map", info->name, info->id); + res.nr_errs++; + goto done; + } + + res.nr_maps++; + + if (func(fd, info, data, wtr)) + res.nr_errs++; + else if (!wtr && json_output) + /* The "func()" above is not writing any json (i.e. !wtr + * test here). + * + * However, "-j" is enabled and there is no errs here, + * so call json_null() as the current convention of + * other cmds. + */ + jsonw_null(json_wtr); + +done: + free(info); + close(fd); + + return res; +} + +static struct res do_work_on_struct_ops(const char *search_type, + const char *search_term, + work_func func, void *data, + struct json_writer *wtr) +{ + if (search_type) { + if (is_prefix(search_type, "id")) + return do_one_id(search_term, func, data, wtr); + else if (!is_prefix(search_type, "name")) + usage(); + } + + return do_search(search_term, func, data, wtr); +} + +static int __do_show(int fd, const struct bpf_map_info *info, void *data, + struct json_writer *wtr) +{ + if (wtr) { + jsonw_start_object(wtr); + jsonw_uint_field(wtr, "id", info->id); + jsonw_string_field(wtr, "name", info->name); + jsonw_string_field(wtr, "kernel_struct_ops", + get_kern_struct_ops_name(info)); + jsonw_end_object(wtr); + } else { + printf("%u: %-15s %-32s\n", info->id, info->name, + get_kern_struct_ops_name(info)); + } + + return 0; +} + +static int do_show(int argc, char **argv) +{ + const char *search_type = NULL, *search_term = NULL; + struct res res; + + if (argc && argc != 2) + usage(); + + if (argc == 2) { + search_type = GET_ARG(); + search_term = GET_ARG(); + } + + res = do_work_on_struct_ops(search_type, search_term, __do_show, + NULL, json_wtr); + + return cmd_retval(&res, !!search_term); +} + +static int __do_dump(int fd, const struct bpf_map_info *info, void *data, + struct json_writer *wtr) +{ + struct btf_dumper *d = (struct btf_dumper *)data; + const struct btf_type *struct_ops_type; + const struct btf *kern_btf = d->btf; + const char *struct_ops_name; + int zero = 0; + void *value; + + /* note: d->jw == wtr */ + + kern_btf = d->btf; + + /* The kernel supporting BPF_MAP_TYPE_STRUCT_OPS must have + * btf_vmlinux_value_type_id. + */ + struct_ops_type = btf__type_by_id(kern_btf, + info->btf_vmlinux_value_type_id); + struct_ops_name = btf__name_by_offset(kern_btf, + struct_ops_type->name_off); + value = calloc(1, info->value_size); + if (!value) { + p_err("mem alloc failed"); + return -1; + } + + if (bpf_map_lookup_elem(fd, &zero, value)) { + p_err("can't lookup struct_ops map %s id %u", + info->name, info->id); + free(value); + return -1; + } + + jsonw_start_object(wtr); + jsonw_name(wtr, "bpf_map_info"); + btf_dumper_type(d, map_info_type_id, (void *)info); + jsonw_end_object(wtr); + + jsonw_start_object(wtr); + jsonw_name(wtr, struct_ops_name); + btf_dumper_type(d, info->btf_vmlinux_value_type_id, value); + jsonw_end_object(wtr); + + free(value); + + return 0; +} + +static int do_dump(int argc, char **argv) +{ + const char *search_type = NULL, *search_term = NULL; + json_writer_t *wtr = json_wtr; + const struct btf *kern_btf; + struct btf_dumper d = {}; + struct res res; + + if (argc && argc != 2) + usage(); + + if (argc == 2) { + search_type = GET_ARG(); + search_term = GET_ARG(); + } + + kern_btf = get_btf_vmlinux(); + if (libbpf_get_error(kern_btf)) + return -1; + + if (!json_output) { + wtr = jsonw_new(stdout); + if (!wtr) { + p_err("can't create json writer"); + return -1; + } + jsonw_pretty(wtr, true); + } + + d.btf = kern_btf; + d.jw = wtr; + d.is_plain_text = !json_output; + d.prog_id_as_func_ptr = true; + + res = do_work_on_struct_ops(search_type, search_term, __do_dump, &d, + wtr); + + if (!json_output) + jsonw_destroy(&wtr); + + return cmd_retval(&res, !!search_term); +} + +static int __do_unregister(int fd, const struct bpf_map_info *info, void *data, + struct json_writer *wtr) +{ + int zero = 0; + + if (bpf_map_delete_elem(fd, &zero)) { + p_err("can't unload %s %s id %u: %s", + get_kern_struct_ops_name(info), info->name, + info->id, strerror(errno)); + return -1; + } + + p_info("Unregistered %s %s id %u", + get_kern_struct_ops_name(info), info->name, + info->id); + + return 0; +} + +static int do_unregister(int argc, char **argv) +{ + const char *search_type, *search_term; + struct res res; + + if (argc != 2) + usage(); + + search_type = GET_ARG(); + search_term = GET_ARG(); + + res = do_work_on_struct_ops(search_type, search_term, + __do_unregister, NULL, NULL); + + return cmd_retval(&res, true); +} + +static int do_register(int argc, char **argv) +{ + LIBBPF_OPTS(bpf_object_open_opts, open_opts); + struct bpf_map_info info = {}; + __u32 info_len = sizeof(info); + int nr_errs = 0, nr_maps = 0; + struct bpf_object *obj; + struct bpf_link *link; + struct bpf_map *map; + const char *file; + + if (argc != 1) + usage(); + + file = GET_ARG(); + + if (verifier_logs) + /* log_level1 + log_level2 + stats, but not stable UAPI */ + open_opts.kernel_log_level = 1 + 2 + 4; + + obj = bpf_object__open_file(file, &open_opts); + if (libbpf_get_error(obj)) + return -1; + + set_max_rlimit(); + + if (bpf_object__load(obj)) { + bpf_object__close(obj); + return -1; + } + + bpf_object__for_each_map(map, obj) { + if (bpf_map__type(map) != BPF_MAP_TYPE_STRUCT_OPS) + continue; + + link = bpf_map__attach_struct_ops(map); + if (libbpf_get_error(link)) { + p_err("can't register struct_ops %s: %s", + bpf_map__name(map), + strerror(-PTR_ERR(link))); + nr_errs++; + continue; + } + nr_maps++; + + bpf_link__disconnect(link); + bpf_link__destroy(link); + + if (!bpf_obj_get_info_by_fd(bpf_map__fd(map), &info, + &info_len)) + p_info("Registered %s %s id %u", + get_kern_struct_ops_name(&info), + bpf_map__name(map), + info.id); + else + /* Not p_err. The struct_ops was attached + * successfully. + */ + p_info("Registered %s but can't find id: %s", + bpf_map__name(map), strerror(errno)); + } + + bpf_object__close(obj); + + if (nr_errs) + return -1; + + if (!nr_maps) { + p_err("no struct_ops found in %s", file); + return -1; + } + + if (json_output) + jsonw_null(json_wtr); + + return 0; +} + +static int do_help(int argc, char **argv) +{ + if (json_output) { + jsonw_null(json_wtr); + return 0; + } + + fprintf(stderr, + "Usage: %1$s %2$s { show | list } [STRUCT_OPS_MAP]\n" + " %1$s %2$s dump [STRUCT_OPS_MAP]\n" + " %1$s %2$s register OBJ\n" + " %1$s %2$s unregister STRUCT_OPS_MAP\n" + " %1$s %2$s help\n" + "\n" + " STRUCT_OPS_MAP := [ id STRUCT_OPS_MAP_ID | name STRUCT_OPS_MAP_NAME ]\n" + " " HELP_SPEC_OPTIONS " }\n" + "", + bin_name, argv[-2]); + + return 0; +} + +static const struct cmd cmds[] = { + { "show", do_show }, + { "list", do_show }, + { "register", do_register }, + { "unregister", do_unregister }, + { "dump", do_dump }, + { "help", do_help }, + { 0 } +}; + +int do_struct_ops(int argc, char **argv) +{ + int err; + + err = cmd_select(cmds, argc, argv, do_help); + + if (!libbpf_get_error(btf_vmlinux)) + btf__free(btf_vmlinux); + + return err; +} diff --git a/tools/bpf/bpftool/tracelog.c b/tools/bpf/bpftool/tracelog.c index e80a5c79b38f..bf1f02212797 100644 --- a/tools/bpf/bpftool/tracelog.c +++ b/tools/bpf/bpftool/tracelog.c @@ -9,7 +9,7 @@ #include <string.h> #include <unistd.h> #include <linux/magic.h> -#include <sys/fcntl.h> +#include <fcntl.h> #include <sys/vfs.h> #include "main.h" diff --git a/tools/bpf/bpftool/xlated_dumper.c b/tools/bpf/bpftool/xlated_dumper.c index 8608cd68cdd0..2d9cd6a7b3c8 100644 --- a/tools/bpf/bpftool/xlated_dumper.c +++ b/tools/bpf/bpftool/xlated_dumper.c @@ -8,6 +8,7 @@ #include <string.h> #include <sys/types.h> #include <bpf/libbpf.h> +#include <bpf/libbpf_internal.h> #include "disasm.h" #include "json_writer.h" @@ -32,8 +33,8 @@ void kernel_syms_load(struct dump_data *dd) return; while (fgets(buff, sizeof(buff), fp)) { - tmp = reallocarray(dd->sym_mapping, dd->sym_count + 1, - sizeof(*dd->sym_mapping)); + tmp = libbpf_reallocarray(dd->sym_mapping, dd->sym_count + 1, + sizeof(*dd->sym_mapping)); if (!tmp) { out: free(dd->sym_mapping); @@ -196,6 +197,12 @@ static const char *print_imm(void *private_data, else if (insn->src_reg == BPF_PSEUDO_MAP_VALUE) snprintf(dd->scratch_buff, sizeof(dd->scratch_buff), "map[id:%u][0]+%u", insn->imm, (insn + 1)->imm); + else if (insn->src_reg == BPF_PSEUDO_MAP_IDX_VALUE) + snprintf(dd->scratch_buff, sizeof(dd->scratch_buff), + "map[idx:%u]+%u", insn->imm, (insn + 1)->imm); + else if (insn->src_reg == BPF_PSEUDO_FUNC) + snprintf(dd->scratch_buff, sizeof(dd->scratch_buff), + "subprog[%+d]", insn->imm); else snprintf(dd->scratch_buff, sizeof(dd->scratch_buff), "0x%llx", (unsigned long long)full_imm); |