Merge branch 'bpf-prog-map-ID'

Martin KaFai Lau says:

====================
Introduce bpf ID

This patch series:
1) Introduce ID for both bpf_prog and bpf_map.
2) Add bpf commands to iterate the prog IDs and map
   IDs of the system.
3) Add bpf commands to get a prog/map fd from an ID
4) Add bpf command to get prog/map info from a fd.
   The prog/map info is a jump start in this patchset
   and it is not meant to be a complete list.  They can
   be extended in the future patches.

v3:
- I suspect v2 may not have applied cleanly.
  In particular, patch 1 has conflict with a recent
  change in struct bpf_prog_aux introduced at a similar time frame:
  8726679a ("bpf: teach verifier to track stack depth")
  v3 should have fixed it.

v2:
Compiler warning fixes:
- Remove lockdep_is_held() usage.  Add comment
  to explain the lock situation instead.
- Add static for idr related variables
- Add __user to the uattr param in bpf_prog_get_info_by_fd()
  and bpf_map_get_info_by_fd().
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

arch/arm64/net/bpf_jit_comp.c

@@ -900,6 +900,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	bpf_jit_binary_lock_ro(header);
 	prog->bpf_func = (void *)ctx.image;
 	prog->jited = 1;
+	prog->jited_len = image_size;
 
 out_off:
 	kfree(ctx.offset);

arch/powerpc/net/bpf_jit_comp64.c

@@ -1052,6 +1052,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
 
 	fp->bpf_func = (void *)image;
 	fp->jited = 1;
+	fp->jited_len = alloclen;
 
 	bpf_flush_icache(bpf_hdr, (u8 *)bpf_hdr + (bpf_hdr->pages * PAGE_SIZE));
 

arch/s390/net/bpf_jit_comp.c

@@ -1329,6 +1329,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
 	bpf_jit_binary_lock_ro(header);
 	fp->bpf_func = (void *) jit.prg_buf;
 	fp->jited = 1;
+	fp->jited_len = jit.size;
 free_addrs:
 	kfree(jit.addrs);
 out:

arch/sparc/net/bpf_jit_comp_64.c

@@ -1560,6 +1560,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 
 	prog->bpf_func = (void *)ctx.image;
 	prog->jited = 1;
+	prog->jited_len = image_size;
 
 out_off:
 	kfree(ctx.offset);

arch/x86/net/bpf_jit_comp.c

@@ -1167,6 +1167,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 		bpf_jit_binary_lock_ro(header);
 		prog->bpf_func = (void *)image;
 		prog->jited = 1;
+		prog->jited_len = proglen;
 	} else {
 		prog = orig_prog;
 	}

include/linux/bpf.h

@@ -46,6 +46,7 @@ struct bpf_map {
 	u32 max_entries;
 	u32 map_flags;
 	u32 pages;
+	u32 id;
 	struct user_struct *user;
 	const struct bpf_map_ops *ops;
 	struct work_struct work;
@@ -172,6 +173,7 @@ struct bpf_prog_aux {
 	u32 used_map_cnt;
 	u32 max_ctx_offset;
 	u32 stack_depth;
+	u32 id;
 	struct latch_tree_node ksym_tnode;
 	struct list_head ksym_lnode;
 	const struct bpf_verifier_ops *ops;

include/linux/filter.h

@@ -69,8 +69,6 @@ struct bpf_prog_aux;
 /* BPF program can access up to 512 bytes of stack space. */
 #define MAX_BPF_STACK	512
 
-#define BPF_TAG_SIZE	8
-
 /* Helper macros for filter block array initializers. */
 
 /* ALU ops on registers, bpf_add|sub|...: dst_reg += src_reg */
@@ -432,6 +430,7 @@ struct bpf_prog {
 	kmemcheck_bitfield_end(meta);
 	enum bpf_prog_type	type;		/* Type of BPF program */
 	u32			len;		/* Number of filter blocks */
+	u32			jited_len;	/* Size of jited insns in bytes */
 	u8			tag[BPF_TAG_SIZE];
 	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
 	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */

include/uapi/linux/bpf.h

@@ -82,6 +82,11 @@ enum bpf_cmd {
 	BPF_PROG_ATTACH,
 	BPF_PROG_DETACH,
 	BPF_PROG_TEST_RUN,
+	BPF_PROG_GET_NEXT_ID,
+	BPF_MAP_GET_NEXT_ID,
+	BPF_PROG_GET_FD_BY_ID,
+	BPF_MAP_GET_FD_BY_ID,
+	BPF_OBJ_GET_INFO_BY_FD,
 };
 
 enum bpf_map_type {
@@ -209,6 +214,21 @@ union bpf_attr {
 		__u32		repeat;
 		__u32		duration;
 	} test;
+
+	struct { /* anonymous struct used by BPF_*_GET_*_ID */
+		union {
+			__u32		start_id;
+			__u32		prog_id;
+			__u32		map_id;
+		};
+		__u32		next_id;
+	};
+
+	struct { /* anonymous struct used by BPF_OBJ_GET_INFO_BY_FD */
+		__u32		bpf_fd;
+		__u32		info_len;
+		__aligned_u64	info;
+	} info;
 } __attribute__((aligned(8)));
 
 /* BPF helper function descriptions:
@@ -673,4 +693,25 @@ struct xdp_md {
 	__u32 data_end;
 };
 
+#define BPF_TAG_SIZE	8
+
+struct bpf_prog_info {
+	__u32 type;
+	__u32 id;
+	__u8  tag[BPF_TAG_SIZE];
+	__u32 jited_prog_len;
+	__u32 xlated_prog_len;
+	__aligned_u64 jited_prog_insns;
+	__aligned_u64 xlated_prog_insns;
+} __attribute__((aligned(8)));
+
+struct bpf_map_info {
+	__u32 type;
+	__u32 id;
+	__u32 key_size;
+	__u32 value_size;
+	__u32 max_entries;
+	__u32 map_flags;
+} __attribute__((aligned(8)));
+
 #endif /* _UAPI__LINUX_BPF_H__ */

tools/include/uapi/linux/bpf.h

@@ -82,6 +82,11 @@ enum bpf_cmd {
 	BPF_PROG_ATTACH,
 	BPF_PROG_DETACH,
 	BPF_PROG_TEST_RUN,
+	BPF_PROG_GET_NEXT_ID,
+	BPF_MAP_GET_NEXT_ID,
+	BPF_PROG_GET_FD_BY_ID,
+	BPF_MAP_GET_FD_BY_ID,
+	BPF_OBJ_GET_INFO_BY_FD,
 };
 
 enum bpf_map_type {
@@ -209,6 +214,21 @@ union bpf_attr {
 		__u32		repeat;
 		__u32		duration;
 	} test;
+
+	struct { /* anonymous struct used by BPF_*_GET_*_ID */
+		union {
+			__u32		start_id;
+			__u32		prog_id;
+			__u32		map_id;
+		};
+		__u32		next_id;
+	};
+
+	struct { /* anonymous struct used by BPF_OBJ_GET_INFO_BY_FD */
+		__u32		bpf_fd;
+		__u32		info_len;
+		__aligned_u64	info;
+	} info;
 } __attribute__((aligned(8)));
 
 /* BPF helper function descriptions:
@@ -673,4 +693,25 @@ struct xdp_md {
 	__u32 data_end;
 };
 
+#define BPF_TAG_SIZE	8
+
+struct bpf_prog_info {
+	__u32 type;
+	__u32 id;
+	__u8  tag[BPF_TAG_SIZE];
+	__u32 jited_prog_len;
+	__u32 xlated_prog_len;
+	__aligned_u64 jited_prog_insns;
+	__aligned_u64 xlated_prog_insns;
+} __attribute__((aligned(8)));
+
+struct bpf_map_info {
+	__u32 type;
+	__u32 id;
+	__u32 key_size;
+	__u32 value_size;
+	__u32 max_entries;
+	__u32 map_flags;
+} __attribute__((aligned(8)));
+
 #endif /* _UAPI__LINUX_BPF_H__ */

tools/lib/bpf/bpf.c

@@ -257,3 +257,71 @@ int bpf_prog_test_run(int prog_fd, int repeat, void *data, __u32 size,
 		*duration = attr.test.duration;
 	return ret;
 }
+
+int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
+{
+	union bpf_attr attr;
+	int err;
+
+	bzero(&attr, sizeof(attr));
+	attr.start_id = start_id;
+
+	err = sys_bpf(BPF_PROG_GET_NEXT_ID, &attr, sizeof(attr));
+	if (!err)
+		*next_id = attr.next_id;
+
+	return err;
+}
+
+int bpf_map_get_next_id(__u32 start_id, __u32 *next_id)
+{
+	union bpf_attr attr;
+	int err;
+
+	bzero(&attr, sizeof(attr));
+	attr.start_id = start_id;
+
+	err = sys_bpf(BPF_MAP_GET_NEXT_ID, &attr, sizeof(attr));
+	if (!err)
+		*next_id = attr.next_id;
+
+	return err;
+}
+
+int bpf_prog_get_fd_by_id(__u32 id)
+{
+	union bpf_attr attr;
+
+	bzero(&attr, sizeof(attr));
+	attr.prog_id = id;
+
+	return sys_bpf(BPF_PROG_GET_FD_BY_ID, &attr, sizeof(attr));
+}
+
+int bpf_map_get_fd_by_id(__u32 id)
+{
+	union bpf_attr attr;
+
+	bzero(&attr, sizeof(attr));
+	attr.map_id = id;
+
+	return sys_bpf(BPF_MAP_GET_FD_BY_ID, &attr, sizeof(attr));
+}
+
+int bpf_obj_get_info_by_fd(int prog_fd, void *info, __u32 *info_len)
+{
+	union bpf_attr attr;
+	int err;
+
+	bzero(&attr, sizeof(attr));
+	bzero(info, *info_len);
+	attr.info.bpf_fd = prog_fd;
+	attr.info.info_len = *info_len;
+	attr.info.info = ptr_to_u64(info);
+
+	err = sys_bpf(BPF_OBJ_GET_INFO_BY_FD, &attr, sizeof(attr));
+	if (!err)
+		*info_len = attr.info.info_len;
+
+	return err;
+}

tools/lib/bpf/bpf.h

@@ -54,5 +54,10 @@ int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
 int bpf_prog_test_run(int prog_fd, int repeat, void *data, __u32 size,
 		      void *data_out, __u32 *size_out, __u32 *retval,
 		      __u32 *duration);
+int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id);
+int bpf_map_get_next_id(__u32 start_id, __u32 *next_id);
+int bpf_prog_get_fd_by_id(__u32 id);
+int bpf_map_get_fd_by_id(__u32 id);
+int bpf_obj_get_info_by_fd(int prog_fd, void *info, __u32 *info_len);
 
 #endif

tools/testing/selftests/bpf/Makefile

@@ -14,7 +14,7 @@ LDLIBS += -lcap -lelf
 TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
 	test_align
 
-TEST_GEN_FILES = test_pkt_access.o test_xdp.o test_l4lb.o test_tcp_estats.o
+TEST_GEN_FILES = test_pkt_access.o test_xdp.o test_l4lb.o test_tcp_estats.o test_obj_id.o
 
 TEST_PROGS := test_kmod.sh
 

tools/testing/selftests/bpf/test_obj_id.c

+/* Copyright (c) 2017 Facebook
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <stddef.h>
+#include <linux/bpf.h>
+#include <linux/pkt_cls.h>
+#include "bpf_helpers.h"
+
+/* It is a dumb bpf program such that it must have no
+ * issue to be loaded since testing the verifier is
+ * not the focus here.
+ */
+
+int _version SEC("version") = 1;
+
+struct bpf_map_def SEC("maps") test_map_id = {
+	.type = BPF_MAP_TYPE_ARRAY,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(__u64),
+	.max_entries = 1,
+};
+
+SEC("test_prog_id")
+int test_prog_id(struct __sk_buff *skb)
+{
+	__u32 key = 0;
+	__u64 *value;
+
+	value = bpf_map_lookup_elem(&test_map_id, &key);
+
+	return TC_ACT_OK;
+}

tools/testing/selftests/bpf/test_progs.c

@@ -22,6 +22,8 @@ typedef __u16 __sum16;
 
 #include <sys/wait.h>
 #include <sys/resource.h>
+#include <sys/types.h>
+#include <pwd.h>
 
 #include <linux/bpf.h>
 #include <linux/err.h>
@@ -70,6 +72,7 @@ static struct {
 		pass_cnt++;						\
 		printf("%s:PASS:%s %d nsec\n", __func__, tag, duration);\
 	}								\
+	__ret;								\
 })
 
 static int bpf_prog_load(const char *file, enum bpf_prog_type type,
@@ -283,6 +286,193 @@ static void test_tcp_estats(void)
 	bpf_object__close(obj);
 }
 
+static inline __u64 ptr_to_u64(const void *ptr)
+{
+	return (__u64) (unsigned long) ptr;
+}
+
+static void test_bpf_obj_id(void)
+{
+	const __u64 array_magic_value = 0xfaceb00c;
+	const __u32 array_key = 0;
+	const int nr_iters = 2;
+	const char *file = "./test_obj_id.o";
+
+	struct bpf_object *objs[nr_iters];
+	int prog_fds[nr_iters], map_fds[nr_iters];
+	/* +1 to test for the info_len returned by kernel */
+	struct bpf_prog_info prog_infos[nr_iters + 1];
+	struct bpf_map_info map_infos[nr_iters + 1];
+	char jited_insns[128], xlated_insns[128];
+	__u32 i, next_id, info_len, nr_id_found, duration = 0;
+	int err = 0;
+	__u64 array_value;
+
+	err = bpf_prog_get_fd_by_id(0);
+	CHECK(err >= 0 || errno != ENOENT,
+	      "get-fd-by-notexist-prog-id", "err %d errno %d\n", err, errno);
+
+	err = bpf_map_get_fd_by_id(0);
+	CHECK(err >= 0 || errno != ENOENT,
+	      "get-fd-by-notexist-map-id", "err %d errno %d\n", err, errno);
+
+	for (i = 0; i < nr_iters; i++)
+		objs[i] = NULL;
+
+	/* Check bpf_obj_get_info_by_fd() */
+	for (i = 0; i < nr_iters; i++) {
+		err = bpf_prog_load(file, BPF_PROG_TYPE_SOCKET_FILTER,
+				    &objs[i], &prog_fds[i]);
+		/* test_obj_id.o is a dumb prog. It should never fail
+		 * to load.
+		 */
+		assert(!err);
+
+		/* Check getting prog info */
+		info_len = sizeof(struct bpf_prog_info) * 2;
+		prog_infos[i].jited_prog_insns = ptr_to_u64(jited_insns);
+		prog_infos[i].jited_prog_len = sizeof(jited_insns);
+		prog_infos[i].xlated_prog_insns = ptr_to_u64(xlated_insns);
+		prog_infos[i].xlated_prog_len = sizeof(xlated_insns);
+		err = bpf_obj_get_info_by_fd(prog_fds[i], &prog_infos[i],
+					     &info_len);
+		if (CHECK(err ||
+			  prog_infos[i].type != BPF_PROG_TYPE_SOCKET_FILTER ||
+			  info_len != sizeof(struct bpf_prog_info) ||
+			  !prog_infos[i].jited_prog_len ||
+			  !prog_infos[i].xlated_prog_len,
+			  "get-prog-info(fd)",
+			  "err %d errno %d i %d type %d(%d) info_len %u(%lu) jited_prog_len %u xlated_prog_len %u\n",
+			  err, errno, i,
+			  prog_infos[i].type, BPF_PROG_TYPE_SOCKET_FILTER,
+			  info_len, sizeof(struct bpf_prog_info),
+			  prog_infos[i].jited_prog_len,
+			  prog_infos[i].xlated_prog_len))
+			goto done;
+
+		map_fds[i] = bpf_find_map(__func__, objs[i], "test_map_id");
+		assert(map_fds[i] >= 0);
+		err = bpf_map_update_elem(map_fds[i], &array_key,
+					  &array_magic_value, 0);
+		assert(!err);
+
+		/* Check getting map info */
+		info_len = sizeof(struct bpf_map_info) * 2;
+		err = bpf_obj_get_info_by_fd(map_fds[i], &map_infos[i],
+					     &info_len);
+		if (CHECK(err ||
+			  map_infos[i].type != BPF_MAP_TYPE_ARRAY ||
+			  map_infos[i].key_size != sizeof(__u32) ||
+			  map_infos[i].value_size != sizeof(__u64) ||
+			  map_infos[i].max_entries != 1 ||
+			  map_infos[i].map_flags != 0 ||
+			  info_len != sizeof(struct bpf_map_info),
+			  "get-map-info(fd)",
+			  "err %d errno %d type %d(%d) info_len %u(%lu) key_size %u value_size %u max_entries %u map_flags %X\n",
+			  err, errno,
+			  map_infos[i].type, BPF_MAP_TYPE_ARRAY,
+			  info_len, sizeof(struct bpf_map_info),
+			  map_infos[i].key_size,
+			  map_infos[i].value_size,
+			  map_infos[i].max_entries,
+			  map_infos[i].map_flags))
+			goto done;
+	}
+
+	/* Check bpf_prog_get_next_id() */
+	nr_id_found = 0;
+	next_id = 0;
+	while (!bpf_prog_get_next_id(next_id, &next_id)) {
+		struct bpf_prog_info prog_info;
+		int prog_fd;
+
+		info_len = sizeof(prog_info);
+
+		prog_fd = bpf_prog_get_fd_by_id(next_id);
+		if (prog_fd < 0 && errno == ENOENT)
+			/* The bpf_prog is in the dead row */
+			continue;
+		if (CHECK(prog_fd < 0, "get-prog-fd(next_id)",
+			  "prog_fd %d next_id %d errno %d\n",
+			  prog_fd, next_id, errno))
+			break;
+
+		for (i = 0; i < nr_iters; i++)
+			if (prog_infos[i].id == next_id)
+				break;
+
+		if (i == nr_iters)
+			continue;
+
+		nr_id_found++;
+
+		err = bpf_obj_get_info_by_fd(prog_fd, &prog_info, &info_len);
+		CHECK(err || info_len != sizeof(struct bpf_prog_info) ||
+		      memcmp(&prog_info, &prog_infos[i], info_len),
+		      "get-prog-info(next_id->fd)",
+		      "err %d errno %d info_len %u(%lu) memcmp %d\n",
+		      err, errno, info_len, sizeof(struct bpf_prog_info),
+		      memcmp(&prog_info, &prog_infos[i], info_len));
+
+		close(prog_fd);
+	}
+	CHECK(nr_id_found != nr_iters,
+	      "check total prog id found by get_next_id",
+	      "nr_id_found %u(%u)\n",
+	      nr_id_found, nr_iters);
+
+	/* Check bpf_map_get_next_id() */
+	nr_id_found = 0;
+	next_id = 0;
+	while (!bpf_map_get_next_id(next_id, &next_id)) {
+		struct bpf_map_info map_info;
+		int map_fd;
+
+		info_len = sizeof(map_info);
+
+		map_fd = bpf_map_get_fd_by_id(next_id);
+		if (map_fd < 0 && errno == ENOENT)
+			/* The bpf_map is in the dead row */
+			continue;
+		if (CHECK(map_fd < 0, "get-map-fd(next_id)",
+			  "map_fd %d next_id %u errno %d\n",
+			  map_fd, next_id, errno))
+			break;
+
+		for (i = 0; i < nr_iters; i++)
+			if (map_infos[i].id == next_id)
+				break;
+
+		if (i == nr_iters)
+			continue;
+
+		nr_id_found++;
+
+		err = bpf_map_lookup_elem(map_fd, &array_key, &array_value);
+		assert(!err);
+
+		err = bpf_obj_get_info_by_fd(map_fd, &map_info, &info_len);
+		CHECK(err || info_len != sizeof(struct bpf_map_info) ||
+		      memcmp(&map_info, &map_infos[i], info_len) ||
+		      array_value != array_magic_value,
+		      "check get-map-info(next_id->fd)",
+		      "err %d errno %d info_len %u(%lu) memcmp %d array_value %llu(%llu)\n",
+		      err, errno, info_len, sizeof(struct bpf_map_info),
+		      memcmp(&map_info, &map_infos[i], info_len),
+		      array_value, array_magic_value);
+
+		close(map_fd);
+	}
+	CHECK(nr_id_found != nr_iters,
+	      "check total map id found by get_next_id",
+	      "nr_id_found %u(%u)\n",
+	      nr_id_found, nr_iters);
+
+done:
+	for (i = 0; i < nr_iters; i++)
+		bpf_object__close(objs[i]);
+}
+
 int main(void)
 {
 	struct rlimit rinf = { RLIM_INFINITY, RLIM_INFINITY };
@@ -293,6 +483,7 @@ int main(void)
 	test_xdp();
 	test_l4lb();
 	test_tcp_estats();
+	test_bpf_obj_id();
 
 	printf("Summary: %d PASSED, %d FAILED\n", pass_cnt, error_cnt);
 	return 0;