diff options
author | Alexei Starovoitov <ast@kernel.org> | 2019-04-27 05:05:44 +0300 |
---|---|---|
committer | Alexei Starovoitov <ast@kernel.org> | 2019-04-27 05:05:44 +0300 |
commit | 3745dc24aa7a6a45771a6b441194b857ac615cfc (patch) | |
tree | 9fe461c7f01adec95c4f66d9a4748afa3348643b /include | |
parent | 34b8ab091f9ef57a2bb3c8c8359a0a03a8abf2f9 (diff) | |
parent | e950e843367d7990b9d7ea964e3c33876d477c4b (diff) | |
download | linux-3745dc24aa7a6a45771a6b441194b857ac615cfc.tar.xz |
Merge branch 'writeable-bpf-tracepoints'
Matt Mullins says:
====================
This adds an opt-in interface for tracepoints to expose a writable context to
BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE programs that are attached, while
supporting read-only access from existing BPF_PROG_TYPE_RAW_TRACEPOINT
programs, as well as from non-BPF-based tracepoints.
The initial motivation is to support tracing that can be observed from the
remote end of an NBD socket, e.g. by adding flags to the struct nbd_request
header. Earlier attempts included adding an NBD-specific tracepoint fd, but in
code review, I was recommended to implement it more generically -- as a result,
this patchset is far simpler than my initial try.
v4->v5:
* rebased onto bpf-next/master and fixed merge conflicts
* "tools: sync bpf.h" also syncs comments that have previously changed
in bpf-next
v3->v4:
* fixed a silly copy/paste typo in include/trace/events/bpf_test_run.h
(_TRACE_NBD_H -> _TRACE_BPF_TEST_RUN_H)
* fixed incorrect/misleading wording in patch 1's commit message,
since the pointer cannot be directly dereferenced in a
BPF_PROG_TYPE_RAW_TRACEPOINT
* cleaned up the error message wording if the prog_tests fail
* Addressed feedback from Yonghong
* reject non-pointer-sized accesses to the buffer pointer
* use sizeof(struct nbd_request) as one-byte-past-the-end in
raw_tp_writable_reject_nbd_invalid.c
* use BPF_MOV64_IMM instead of BPF_LD_IMM64
v2->v3:
* Andrew addressed Josef's comments:
* C-style commenting in nbd.c
* Collapsed identical events into a single DECLARE_EVENT_CLASS.
This saves about 2kB of kernel text
v1->v2:
* add selftests
* sync tools/include/uapi/linux/bpf.h
* reject variable offset into the buffer
* add string representation of PTR_TO_TP_BUFFER to reg_type_str
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Diffstat (limited to 'include')
-rw-r--r-- | include/linux/bpf.h | 2 | ||||
-rw-r--r-- | include/linux/bpf_types.h | 1 | ||||
-rw-r--r-- | include/linux/tracepoint-defs.h | 1 | ||||
-rw-r--r-- | include/trace/bpf_probe.h | 27 | ||||
-rw-r--r-- | include/trace/events/bpf_test_run.h | 50 | ||||
-rw-r--r-- | include/trace/events/nbd.h | 107 | ||||
-rw-r--r-- | include/uapi/linux/bpf.h | 1 |
7 files changed, 187 insertions, 2 deletions
diff --git a/include/linux/bpf.h b/include/linux/bpf.h index f15432d90728..cd6341eabd74 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -272,6 +272,7 @@ enum bpf_reg_type { PTR_TO_SOCK_COMMON_OR_NULL, /* reg points to sock_common or NULL */ PTR_TO_TCP_SOCK, /* reg points to struct tcp_sock */ PTR_TO_TCP_SOCK_OR_NULL, /* reg points to struct tcp_sock or NULL */ + PTR_TO_TP_BUFFER, /* reg points to a writable raw tp's buffer */ }; /* The information passed from prog-specific *_is_valid_access @@ -361,6 +362,7 @@ struct bpf_prog_aux { u32 used_map_cnt; u32 max_ctx_offset; u32 max_pkt_offset; + u32 max_tp_access; u32 stack_depth; u32 id; u32 func_cnt; /* used by non-func prog as the number of func progs */ diff --git a/include/linux/bpf_types.h b/include/linux/bpf_types.h index d26991a16894..a10d37bce364 100644 --- a/include/linux/bpf_types.h +++ b/include/linux/bpf_types.h @@ -25,6 +25,7 @@ BPF_PROG_TYPE(BPF_PROG_TYPE_KPROBE, kprobe) BPF_PROG_TYPE(BPF_PROG_TYPE_TRACEPOINT, tracepoint) BPF_PROG_TYPE(BPF_PROG_TYPE_PERF_EVENT, perf_event) BPF_PROG_TYPE(BPF_PROG_TYPE_RAW_TRACEPOINT, raw_tracepoint) +BPF_PROG_TYPE(BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE, raw_tracepoint_writable) #endif #ifdef CONFIG_CGROUP_BPF BPF_PROG_TYPE(BPF_PROG_TYPE_CGROUP_DEVICE, cg_dev) diff --git a/include/linux/tracepoint-defs.h b/include/linux/tracepoint-defs.h index 49ba9cde7e4b..b29950a19205 100644 --- a/include/linux/tracepoint-defs.h +++ b/include/linux/tracepoint-defs.h @@ -45,6 +45,7 @@ struct bpf_raw_event_map { struct tracepoint *tp; void *bpf_func; u32 num_args; + u32 writable_size; } __aligned(32); #endif diff --git a/include/trace/bpf_probe.h b/include/trace/bpf_probe.h index 505dae0bed80..d6e556c0a085 100644 --- a/include/trace/bpf_probe.h +++ b/include/trace/bpf_probe.h @@ -69,8 +69,7 @@ __bpf_trace_##call(void *__data, proto) \ * to make sure that if the tracepoint handling changes, the * bpf probe will fail to compile unless it too is updated. */ -#undef DEFINE_EVENT -#define DEFINE_EVENT(template, call, proto, args) \ +#define __DEFINE_EVENT(template, call, proto, args, size) \ static inline void bpf_test_probe_##call(void) \ { \ check_trace_callback_type_##call(__bpf_trace_##template); \ @@ -81,12 +80,36 @@ __bpf_trace_tp_map_##call = { \ .tp = &__tracepoint_##call, \ .bpf_func = (void *)__bpf_trace_##template, \ .num_args = COUNT_ARGS(args), \ + .writable_size = size, \ }; +#define FIRST(x, ...) x + +#undef DEFINE_EVENT_WRITABLE +#define DEFINE_EVENT_WRITABLE(template, call, proto, args, size) \ +static inline void bpf_test_buffer_##call(void) \ +{ \ + /* BUILD_BUG_ON() is ignored if the code is completely eliminated, but \ + * BUILD_BUG_ON_ZERO() uses a different mechanism that is not \ + * dead-code-eliminated. \ + */ \ + FIRST(proto); \ + (void)BUILD_BUG_ON_ZERO(size != sizeof(*FIRST(args))); \ +} \ +__DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args), size) + +#undef DEFINE_EVENT +#define DEFINE_EVENT(template, call, proto, args) \ + __DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args), 0) #undef DEFINE_EVENT_PRINT #define DEFINE_EVENT_PRINT(template, name, proto, args, print) \ DEFINE_EVENT(template, name, PARAMS(proto), PARAMS(args)) #include TRACE_INCLUDE(TRACE_INCLUDE_FILE) + +#undef DEFINE_EVENT_WRITABLE +#undef __DEFINE_EVENT +#undef FIRST + #endif /* CONFIG_BPF_EVENTS */ diff --git a/include/trace/events/bpf_test_run.h b/include/trace/events/bpf_test_run.h new file mode 100644 index 000000000000..265447e3f71a --- /dev/null +++ b/include/trace/events/bpf_test_run.h @@ -0,0 +1,50 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM bpf_test_run + +#if !defined(_TRACE_BPF_TEST_RUN_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_BPF_TEST_RUN_H + +#include <linux/tracepoint.h> + +DECLARE_EVENT_CLASS(bpf_test_finish, + + TP_PROTO(int *err), + + TP_ARGS(err), + + TP_STRUCT__entry( + __field(int, err) + ), + + TP_fast_assign( + __entry->err = *err; + ), + + TP_printk("bpf_test_finish with err=%d", __entry->err) +); + +#ifdef DEFINE_EVENT_WRITABLE +#undef BPF_TEST_RUN_DEFINE_EVENT +#define BPF_TEST_RUN_DEFINE_EVENT(template, call, proto, args, size) \ + DEFINE_EVENT_WRITABLE(template, call, PARAMS(proto), \ + PARAMS(args), size) +#else +#undef BPF_TEST_RUN_DEFINE_EVENT +#define BPF_TEST_RUN_DEFINE_EVENT(template, call, proto, args, size) \ + DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args)) +#endif + +BPF_TEST_RUN_DEFINE_EVENT(bpf_test_finish, bpf_test_finish, + + TP_PROTO(int *err), + + TP_ARGS(err), + + sizeof(int) +); + +#endif + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/include/trace/events/nbd.h b/include/trace/events/nbd.h new file mode 100644 index 000000000000..9849956f34d8 --- /dev/null +++ b/include/trace/events/nbd.h @@ -0,0 +1,107 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM nbd + +#if !defined(_TRACE_NBD_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_NBD_H + +#include <linux/tracepoint.h> + +DECLARE_EVENT_CLASS(nbd_transport_event, + + TP_PROTO(struct request *req, u64 handle), + + TP_ARGS(req, handle), + + TP_STRUCT__entry( + __field(struct request *, req) + __field(u64, handle) + ), + + TP_fast_assign( + __entry->req = req; + __entry->handle = handle; + ), + + TP_printk( + "nbd transport event: request %p, handle 0x%016llx", + __entry->req, + __entry->handle + ) +); + +DEFINE_EVENT(nbd_transport_event, nbd_header_sent, + + TP_PROTO(struct request *req, u64 handle), + + TP_ARGS(req, handle) +); + +DEFINE_EVENT(nbd_transport_event, nbd_payload_sent, + + TP_PROTO(struct request *req, u64 handle), + + TP_ARGS(req, handle) +); + +DEFINE_EVENT(nbd_transport_event, nbd_header_received, + + TP_PROTO(struct request *req, u64 handle), + + TP_ARGS(req, handle) +); + +DEFINE_EVENT(nbd_transport_event, nbd_payload_received, + + TP_PROTO(struct request *req, u64 handle), + + TP_ARGS(req, handle) +); + +DECLARE_EVENT_CLASS(nbd_send_request, + + TP_PROTO(struct nbd_request *nbd_request, int index, + struct request *rq), + + TP_ARGS(nbd_request, index, rq), + + TP_STRUCT__entry( + __field(struct nbd_request *, nbd_request) + __field(u64, dev_index) + __field(struct request *, request) + ), + + TP_fast_assign( + __entry->nbd_request = 0; + __entry->dev_index = index; + __entry->request = rq; + ), + + TP_printk("nbd%lld: request %p", __entry->dev_index, __entry->request) +); + +#ifdef DEFINE_EVENT_WRITABLE +#undef NBD_DEFINE_EVENT +#define NBD_DEFINE_EVENT(template, call, proto, args, size) \ + DEFINE_EVENT_WRITABLE(template, call, PARAMS(proto), \ + PARAMS(args), size) +#else +#undef NBD_DEFINE_EVENT +#define NBD_DEFINE_EVENT(template, call, proto, args, size) \ + DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args)) +#endif + +NBD_DEFINE_EVENT(nbd_send_request, nbd_send_request, + + TP_PROTO(struct nbd_request *nbd_request, int index, + struct request *rq), + + TP_ARGS(nbd_request, index, rq), + + sizeof(struct nbd_request) +); + +#endif + +/* This part must be outside protection */ +#include <trace/define_trace.h> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index eaf2d3284248..f7fa7a34a62d 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -168,6 +168,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_SK_REUSEPORT, BPF_PROG_TYPE_FLOW_DISSECTOR, BPF_PROG_TYPE_CGROUP_SYSCTL, + BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE, }; enum bpf_attach_type { |