diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-27 10:05:51 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-27 10:05:51 +0000 |
commit | 5d1646d90e1f2cceb9f0828f4b28318cd0ec7744 (patch) | |
tree | a94efe259b9009378be6d90eb30d2b019d95c194 /tools/perf/examples/bpf | |
parent | Initial commit. (diff) | |
download | linux-5d1646d90e1f2cceb9f0828f4b28318cd0ec7744.tar.xz linux-5d1646d90e1f2cceb9f0828f4b28318cd0ec7744.zip |
Adding upstream version 5.10.209.upstream/5.10.209
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'tools/perf/examples/bpf')
-rw-r--r-- | tools/perf/examples/bpf/5sec.c | 51 | ||||
-rw-r--r-- | tools/perf/examples/bpf/augmented_raw_syscalls.c | 294 | ||||
-rw-r--r-- | tools/perf/examples/bpf/augmented_syscalls.c | 169 | ||||
-rw-r--r-- | tools/perf/examples/bpf/empty.c | 3 | ||||
-rw-r--r-- | tools/perf/examples/bpf/etcsnoop.c | 76 | ||||
-rw-r--r-- | tools/perf/examples/bpf/hello.c | 9 | ||||
-rw-r--r-- | tools/perf/examples/bpf/sys_enter_openat.c | 33 |
7 files changed, 635 insertions, 0 deletions
diff --git a/tools/perf/examples/bpf/5sec.c b/tools/perf/examples/bpf/5sec.c new file mode 100644 index 000000000..65c4ff689 --- /dev/null +++ b/tools/perf/examples/bpf/5sec.c @@ -0,0 +1,51 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + Description: + + . Disable strace like syscall tracing (--no-syscalls), or try tracing + just some (-e *sleep). + + . Attach a filter function to a kernel function, returning when it should + be considered, i.e. appear on the output. + + . Run it system wide, so that any sleep of >= 5 seconds and < than 6 + seconds gets caught. + + . Ask for callgraphs using DWARF info, so that userspace can be unwound + + . While this is running, run something like "sleep 5s". + + . If we decide to add tv_nsec as well, then it becomes: + + int probe(hrtimer_nanosleep, rqtp->tv_sec rqtp->tv_nsec)(void *ctx, int err, long sec, long nsec) + + I.e. add where it comes from (rqtp->tv_nsec) and where it will be + accessible in the function body (nsec) + + # perf trace --no-syscalls -e tools/perf/examples/bpf/5sec.c/call-graph=dwarf/ + 0.000 perf_bpf_probe:func:(ffffffff9811b5f0) tv_sec=5 + hrtimer_nanosleep ([kernel.kallsyms]) + __x64_sys_nanosleep ([kernel.kallsyms]) + do_syscall_64 ([kernel.kallsyms]) + entry_SYSCALL_64 ([kernel.kallsyms]) + __GI___nanosleep (/usr/lib64/libc-2.26.so) + rpl_nanosleep (/usr/bin/sleep) + xnanosleep (/usr/bin/sleep) + main (/usr/bin/sleep) + __libc_start_main (/usr/lib64/libc-2.26.so) + _start (/usr/bin/sleep) + ^C# + + Copyright (C) 2018 Red Hat, Inc., Arnaldo Carvalho de Melo <acme@redhat.com> +*/ + +#include <bpf/bpf.h> + +#define NSEC_PER_SEC 1000000000L + +int probe(hrtimer_nanosleep, rqtp)(void *ctx, int err, long long sec) +{ + return sec / NSEC_PER_SEC == 5ULL; +} + +license(GPL); diff --git a/tools/perf/examples/bpf/augmented_raw_syscalls.c b/tools/perf/examples/bpf/augmented_raw_syscalls.c new file mode 100644 index 000000000..b80437971 --- /dev/null +++ b/tools/perf/examples/bpf/augmented_raw_syscalls.c @@ -0,0 +1,294 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Augment the raw_syscalls tracepoints with the contents of the pointer arguments. + * + * Test it with: + * + * perf trace -e tools/perf/examples/bpf/augmented_raw_syscalls.c cat /etc/passwd > /dev/null + * + * This exactly matches what is marshalled into the raw_syscall:sys_enter + * payload expected by the 'perf trace' beautifiers. + * + * For now it just uses the existing tracepoint augmentation code in 'perf + * trace', in the next csets we'll hook up these with the sys_enter/sys_exit + * code that will combine entry/exit in a strace like way. + */ + +#include <unistd.h> +#include <linux/limits.h> +#include <linux/socket.h> +#include <pid_filter.h> + +/* bpf-output associated map */ +bpf_map(__augmented_syscalls__, PERF_EVENT_ARRAY, int, u32, __NR_CPUS__); + +/* + * string_args_len: one per syscall arg, 0 means not a string or don't copy it, + * PATH_MAX for copying everything, any other value to limit + * it a la 'strace -s strsize'. + */ +struct syscall { + bool enabled; + u16 string_args_len[6]; +}; + +bpf_map(syscalls, ARRAY, int, struct syscall, 512); + +/* + * What to augment at entry? + * + * Pointer arg payloads (filenames, etc) passed from userspace to the kernel + */ +bpf_map(syscalls_sys_enter, PROG_ARRAY, u32, u32, 512); + +/* + * What to augment at exit? + * + * Pointer arg payloads returned from the kernel (struct stat, etc) to userspace. + */ +bpf_map(syscalls_sys_exit, PROG_ARRAY, u32, u32, 512); + +struct syscall_enter_args { + unsigned long long common_tp_fields; + long syscall_nr; + unsigned long args[6]; +}; + +struct syscall_exit_args { + unsigned long long common_tp_fields; + long syscall_nr; + long ret; +}; + +struct augmented_arg { + unsigned int size; + int err; + char value[PATH_MAX]; +}; + +pid_filter(pids_filtered); + +struct augmented_args_payload { + struct syscall_enter_args args; + union { + struct { + struct augmented_arg arg, arg2; + }; + struct sockaddr_storage saddr; + }; +}; + +// We need more tmp space than the BPF stack can give us +bpf_map(augmented_args_tmp, PERCPU_ARRAY, int, struct augmented_args_payload, 1); + +static inline struct augmented_args_payload *augmented_args_payload(void) +{ + int key = 0; + return bpf_map_lookup_elem(&augmented_args_tmp, &key); +} + +static inline int augmented__output(void *ctx, struct augmented_args_payload *args, int len) +{ + /* If perf_event_output fails, return non-zero so that it gets recorded unaugmented */ + return perf_event_output(ctx, &__augmented_syscalls__, BPF_F_CURRENT_CPU, args, len); +} + +static inline +unsigned int augmented_arg__read_str(struct augmented_arg *augmented_arg, const void *arg, unsigned int arg_len) +{ + unsigned int augmented_len = sizeof(*augmented_arg); + int string_len = probe_read_str(&augmented_arg->value, arg_len, arg); + + augmented_arg->size = augmented_arg->err = 0; + /* + * probe_read_str may return < 0, e.g. -EFAULT + * So we leave that in the augmented_arg->size that userspace will + */ + if (string_len > 0) { + augmented_len -= sizeof(augmented_arg->value) - string_len; + augmented_len &= sizeof(augmented_arg->value) - 1; + augmented_arg->size = string_len; + } else { + /* + * So that username notice the error while still being able + * to skip this augmented arg record + */ + augmented_arg->err = string_len; + augmented_len = offsetof(struct augmented_arg, value); + } + + return augmented_len; +} + +SEC("!raw_syscalls:unaugmented") +int syscall_unaugmented(struct syscall_enter_args *args) +{ + return 1; +} + +/* + * These will be tail_called from SEC("raw_syscalls:sys_enter"), so will find in + * augmented_args_tmp what was read by that raw_syscalls:sys_enter and go + * on from there, reading the first syscall arg as a string, i.e. open's + * filename. + */ +SEC("!syscalls:sys_enter_connect") +int sys_enter_connect(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *sockaddr_arg = (const void *)args->args[1]; + unsigned int socklen = args->args[2]; + unsigned int len = sizeof(augmented_args->args); + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + if (socklen > sizeof(augmented_args->saddr)) + socklen = sizeof(augmented_args->saddr); + + probe_read(&augmented_args->saddr, socklen, sockaddr_arg); + + return augmented__output(args, augmented_args, len + socklen); +} + +SEC("!syscalls:sys_enter_sendto") +int sys_enter_sendto(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *sockaddr_arg = (const void *)args->args[4]; + unsigned int socklen = args->args[5]; + unsigned int len = sizeof(augmented_args->args); + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + if (socklen > sizeof(augmented_args->saddr)) + socklen = sizeof(augmented_args->saddr); + + probe_read(&augmented_args->saddr, socklen, sockaddr_arg); + + return augmented__output(args, augmented_args, len + socklen); +} + +SEC("!syscalls:sys_enter_open") +int sys_enter_open(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *filename_arg = (const void *)args->args[0]; + unsigned int len = sizeof(augmented_args->args); + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + len += augmented_arg__read_str(&augmented_args->arg, filename_arg, sizeof(augmented_args->arg.value)); + + return augmented__output(args, augmented_args, len); +} + +SEC("!syscalls:sys_enter_openat") +int sys_enter_openat(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *filename_arg = (const void *)args->args[1]; + unsigned int len = sizeof(augmented_args->args); + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + len += augmented_arg__read_str(&augmented_args->arg, filename_arg, sizeof(augmented_args->arg.value)); + + return augmented__output(args, augmented_args, len); +} + +SEC("!syscalls:sys_enter_rename") +int sys_enter_rename(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *oldpath_arg = (const void *)args->args[0], + *newpath_arg = (const void *)args->args[1]; + unsigned int len = sizeof(augmented_args->args), oldpath_len; + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + oldpath_len = augmented_arg__read_str(&augmented_args->arg, oldpath_arg, sizeof(augmented_args->arg.value)); + len += oldpath_len + augmented_arg__read_str((void *)(&augmented_args->arg) + oldpath_len, newpath_arg, sizeof(augmented_args->arg.value)); + + return augmented__output(args, augmented_args, len); +} + +SEC("!syscalls:sys_enter_renameat") +int sys_enter_renameat(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args = augmented_args_payload(); + const void *oldpath_arg = (const void *)args->args[1], + *newpath_arg = (const void *)args->args[3]; + unsigned int len = sizeof(augmented_args->args), oldpath_len; + + if (augmented_args == NULL) + return 1; /* Failure: don't filter */ + + oldpath_len = augmented_arg__read_str(&augmented_args->arg, oldpath_arg, sizeof(augmented_args->arg.value)); + len += oldpath_len + augmented_arg__read_str((void *)(&augmented_args->arg) + oldpath_len, newpath_arg, sizeof(augmented_args->arg.value)); + + return augmented__output(args, augmented_args, len); +} + +SEC("raw_syscalls:sys_enter") +int sys_enter(struct syscall_enter_args *args) +{ + struct augmented_args_payload *augmented_args; + /* + * We start len, the amount of data that will be in the perf ring + * buffer, if this is not filtered out by one of pid_filter__has(), + * syscall->enabled, etc, with the non-augmented raw syscall payload, + * i.e. sizeof(augmented_args->args). + * + * We'll add to this as we add augmented syscalls right after that + * initial, non-augmented raw_syscalls:sys_enter payload. + */ + unsigned int len = sizeof(augmented_args->args); + struct syscall *syscall; + + if (pid_filter__has(&pids_filtered, getpid())) + return 0; + + augmented_args = augmented_args_payload(); + if (augmented_args == NULL) + return 1; + + probe_read(&augmented_args->args, sizeof(augmented_args->args), args); + + /* + * Jump to syscall specific augmenter, even if the default one, + * "!raw_syscalls:unaugmented" that will just return 1 to return the + * unagmented tracepoint payload. + */ + bpf_tail_call(args, &syscalls_sys_enter, augmented_args->args.syscall_nr); + + // If not found on the PROG_ARRAY syscalls map, then we're filtering it: + return 0; +} + +SEC("raw_syscalls:sys_exit") +int sys_exit(struct syscall_exit_args *args) +{ + struct syscall_exit_args exit_args; + + if (pid_filter__has(&pids_filtered, getpid())) + return 0; + + probe_read(&exit_args, sizeof(exit_args), args); + /* + * Jump to syscall specific return augmenter, even if the default one, + * "!raw_syscalls:unaugmented" that will just return 1 to return the + * unagmented tracepoint payload. + */ + bpf_tail_call(args, &syscalls_sys_exit, exit_args.syscall_nr); + /* + * If not found on the PROG_ARRAY syscalls map, then we're filtering it: + */ + return 0; +} + +license(GPL); diff --git a/tools/perf/examples/bpf/augmented_syscalls.c b/tools/perf/examples/bpf/augmented_syscalls.c new file mode 100644 index 000000000..524fdb853 --- /dev/null +++ b/tools/perf/examples/bpf/augmented_syscalls.c @@ -0,0 +1,169 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Augment syscalls with the contents of the pointer arguments. + * + * Test it with: + * + * perf trace -e tools/perf/examples/bpf/augmented_syscalls.c cat /etc/passwd > /dev/null + * + * It'll catch some openat syscalls related to the dynamic linked and + * the last one should be the one for '/etc/passwd'. + * + * This matches what is marshalled into the raw_syscall:sys_enter payload + * expected by the 'perf trace' beautifiers, and can be used by them, that will + * check if perf_sample->raw_data is more than what is expected for each + * syscalls:sys_{enter,exit}_SYSCALL tracepoint, uing the extra data as the + * contents of pointer arguments. + */ + +#include <stdio.h> +#include <linux/socket.h> + +/* bpf-output associated map */ +bpf_map(__augmented_syscalls__, PERF_EVENT_ARRAY, int, u32, __NR_CPUS__); + +struct syscall_exit_args { + unsigned long long common_tp_fields; + long syscall_nr; + long ret; +}; + +struct augmented_filename { + unsigned int size; + int reserved; + char value[256]; +}; + +#define augmented_filename_syscall(syscall) \ +struct augmented_enter_##syscall##_args { \ + struct syscall_enter_##syscall##_args args; \ + struct augmented_filename filename; \ +}; \ +int syscall_enter(syscall)(struct syscall_enter_##syscall##_args *args) \ +{ \ + struct augmented_enter_##syscall##_args augmented_args = { .filename.reserved = 0, }; \ + unsigned int len = sizeof(augmented_args); \ + probe_read(&augmented_args.args, sizeof(augmented_args.args), args); \ + augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, \ + sizeof(augmented_args.filename.value), \ + args->filename_ptr); \ + if (augmented_args.filename.size < sizeof(augmented_args.filename.value)) { \ + len -= sizeof(augmented_args.filename.value) - augmented_args.filename.size; \ + len &= sizeof(augmented_args.filename.value) - 1; \ + } \ + /* If perf_event_output fails, return non-zero so that it gets recorded unaugmented */ \ + return perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, \ + &augmented_args, len); \ +} \ +int syscall_exit(syscall)(struct syscall_exit_args *args) \ +{ \ + return 1; /* 0 as soon as we start copying data returned by the kernel, e.g. 'read' */ \ +} + +struct syscall_enter_openat_args { + unsigned long long common_tp_fields; + long syscall_nr; + long dfd; + char *filename_ptr; + long flags; + long mode; +}; + +augmented_filename_syscall(openat); + +struct syscall_enter_open_args { + unsigned long long common_tp_fields; + long syscall_nr; + char *filename_ptr; + long flags; + long mode; +}; + +augmented_filename_syscall(open); + +struct syscall_enter_inotify_add_watch_args { + unsigned long long common_tp_fields; + long syscall_nr; + long fd; + char *filename_ptr; + long mask; +}; + +augmented_filename_syscall(inotify_add_watch); + +struct statbuf; + +struct syscall_enter_newstat_args { + unsigned long long common_tp_fields; + long syscall_nr; + char *filename_ptr; + struct stat *statbuf; +}; + +augmented_filename_syscall(newstat); + +#ifndef _K_SS_MAXSIZE +#define _K_SS_MAXSIZE 128 +#endif + +#define augmented_sockaddr_syscall(syscall) \ +struct augmented_enter_##syscall##_args { \ + struct syscall_enter_##syscall##_args args; \ + struct sockaddr_storage addr; \ +}; \ +int syscall_enter(syscall)(struct syscall_enter_##syscall##_args *args) \ +{ \ + struct augmented_enter_##syscall##_args augmented_args; \ + unsigned long addrlen = sizeof(augmented_args.addr); \ + probe_read(&augmented_args.args, sizeof(augmented_args.args), args); \ +/* FIXME_CLANG_OPTIMIZATION_THAT_ACCESSES_USER_CONTROLLED_ADDRLEN_DESPITE_THIS_CHECK */ \ +/* if (addrlen > augmented_args.args.addrlen) */ \ +/* addrlen = augmented_args.args.addrlen; */ \ +/* */ \ + probe_read(&augmented_args.addr, addrlen, args->addr_ptr); \ + /* If perf_event_output fails, return non-zero so that it gets recorded unaugmented */ \ + return perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, \ + &augmented_args, \ + sizeof(augmented_args) - sizeof(augmented_args.addr) + addrlen);\ +} \ +int syscall_exit(syscall)(struct syscall_exit_args *args) \ +{ \ + return 1; /* 0 as soon as we start copying data returned by the kernel, e.g. 'read' */ \ +} + +struct sockaddr; + +struct syscall_enter_bind_args { + unsigned long long common_tp_fields; + long syscall_nr; + long fd; + struct sockaddr *addr_ptr; + unsigned long addrlen; +}; + +augmented_sockaddr_syscall(bind); + +struct syscall_enter_connect_args { + unsigned long long common_tp_fields; + long syscall_nr; + long fd; + struct sockaddr *addr_ptr; + unsigned long addrlen; +}; + +augmented_sockaddr_syscall(connect); + +struct syscall_enter_sendto_args { + unsigned long long common_tp_fields; + long syscall_nr; + long fd; + void *buff; + long len; + unsigned long flags; + struct sockaddr *addr_ptr; + long addr_len; +}; + +augmented_sockaddr_syscall(sendto); + +license(GPL); diff --git a/tools/perf/examples/bpf/empty.c b/tools/perf/examples/bpf/empty.c new file mode 100644 index 000000000..7d7fb0c9f --- /dev/null +++ b/tools/perf/examples/bpf/empty.c @@ -0,0 +1,3 @@ +#include <bpf/bpf.h> + +license(GPL); diff --git a/tools/perf/examples/bpf/etcsnoop.c b/tools/perf/examples/bpf/etcsnoop.c new file mode 100644 index 000000000..e81b53534 --- /dev/null +++ b/tools/perf/examples/bpf/etcsnoop.c @@ -0,0 +1,76 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Augment the filename syscalls with the contents of the filename pointer argument + * filtering only those that do not start with /etc/. + * + * Test it with: + * + * perf trace -e tools/perf/examples/bpf/augmented_syscalls.c cat /etc/passwd > /dev/null + * + * It'll catch some openat syscalls related to the dynamic linked and + * the last one should be the one for '/etc/passwd'. + * + * This matches what is marshalled into the raw_syscall:sys_enter payload + * expected by the 'perf trace' beautifiers, and can be used by them unmodified, + * which will be done as that feature is implemented in the next csets, for now + * it will appear in a dump done by the default tracepoint handler in 'perf trace', + * that uses bpf_output__fprintf() to just dump those contents, as done with + * the bpf-output event associated with the __bpf_output__ map declared in + * tools/perf/include/bpf/stdio.h. + */ + +#include <stdio.h> + +/* bpf-output associated map */ +bpf_map(__augmented_syscalls__, PERF_EVENT_ARRAY, int, u32, __NR_CPUS__); + +struct augmented_filename { + int size; + int reserved; + char value[64]; +}; + +#define augmented_filename_syscall_enter(syscall) \ +struct augmented_enter_##syscall##_args { \ + struct syscall_enter_##syscall##_args args; \ + struct augmented_filename filename; \ +}; \ +int syscall_enter(syscall)(struct syscall_enter_##syscall##_args *args) \ +{ \ + char etc[6] = "/etc/"; \ + struct augmented_enter_##syscall##_args augmented_args = { .filename.reserved = 0, }; \ + probe_read(&augmented_args.args, sizeof(augmented_args.args), args); \ + augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, \ + sizeof(augmented_args.filename.value), \ + args->filename_ptr); \ + if (__builtin_memcmp(augmented_args.filename.value, etc, 4) != 0) \ + return 0; \ + /* If perf_event_output fails, return non-zero so that it gets recorded unaugmented */ \ + return perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, \ + &augmented_args, \ + (sizeof(augmented_args) - sizeof(augmented_args.filename.value) + \ + augmented_args.filename.size)); \ +} + +struct syscall_enter_openat_args { + unsigned long long common_tp_fields; + long syscall_nr; + long dfd; + char *filename_ptr; + long flags; + long mode; +}; + +augmented_filename_syscall_enter(openat); + +struct syscall_enter_open_args { + unsigned long long common_tp_fields; + long syscall_nr; + char *filename_ptr; + long flags; + long mode; +}; + +augmented_filename_syscall_enter(open); + +license(GPL); diff --git a/tools/perf/examples/bpf/hello.c b/tools/perf/examples/bpf/hello.c new file mode 100644 index 000000000..cf3c2fdc7 --- /dev/null +++ b/tools/perf/examples/bpf/hello.c @@ -0,0 +1,9 @@ +#include <stdio.h> + +int syscall_enter(openat)(void *args) +{ + puts("Hello, world\n"); + return 0; +} + +license(GPL); diff --git a/tools/perf/examples/bpf/sys_enter_openat.c b/tools/perf/examples/bpf/sys_enter_openat.c new file mode 100644 index 000000000..c4481c390 --- /dev/null +++ b/tools/perf/examples/bpf/sys_enter_openat.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Hook into 'openat' syscall entry tracepoint + * + * Test it with: + * + * perf trace -e tools/perf/examples/bpf/sys_enter_openat.c cat /etc/passwd > /dev/null + * + * It'll catch some openat syscalls related to the dynamic linked and + * the last one should be the one for '/etc/passwd'. + * + * The syscall_enter_openat_args can be used to get the syscall fields + * and use them for filtering calls, i.e. use in expressions for + * the return value. + */ + +#include <bpf/bpf.h> + +struct syscall_enter_openat_args { + unsigned long long unused; + long syscall_nr; + long dfd; + char *filename_ptr; + long flags; + long mode; +}; + +int syscall_enter(openat)(struct syscall_enter_openat_args *args) +{ + return 1; +} + +license(GPL); |