Introduce bpf ID

This patch generates an unique ID for each created bpf_map. The approach is similar to the earlier patch for bpf_prog ID.

It is worth to note that the bpf_map's ID and bpf_prog's ID are in two independent ID spaces and both have the same valid range:
[1, INT_MAX).

---
 include/linux/bpf.h  |  1 +
 kernel/bpf/syscall.c | 34 +++++++++++++++++++++++++++++++++-
 2 files changed, 34 insertions(+), 1 deletion(-)

include/linux/bpf.h +1 -0

index c5946d19f2ca..c32bace66d3d 100644
@@ -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;

kernel/bpf/syscall.c +33 -1

index 2a1b32b470f1..4c3075b5d840 100644
@@ -27,6 +27,8 @@
 DEFINE_PER_CPU(int, bpf_prog_active);
 static DEFINE_IDR(prog_idr);
 static DEFINE_SPINLOCK(prog_idr_lock);
+static DEFINE_IDR(map_idr);
+static DEFINE_SPINLOCK(map_idr_lock);
 
 int sysctl_unprivileged_bpf_disabled __read_mostly;

@@ -117,6 +119,29 @@ static void bpf_map_uncharge_memlock(struct bpf_map *map)
 	free_uid(user);
 }
 
+static int bpf_map_alloc_id(struct bpf_map *map)
+{
+	int id;
+
+	spin_lock_bh(&map_idr_lock);
+	id = idr_alloc_cyclic(&map_idr, map, 1, INT_MAX, GFP_ATOMIC);
+	if (id > 0)
+		map->id = id;
+	spin_unlock_bh(&map_idr_lock);
+
+	if (WARN_ON_ONCE(!id))
+		return -ENOSPC;
+
+	return id > 0 ? 0 : id;
+}
+
+static void bpf_map_free_id(struct bpf_map *map)
+{
+	spin_lock_bh(&map_idr_lock);
+	idr_remove(&map_idr, map->id);
+	spin_unlock_bh(&map_idr_lock);
+}
+
 /* called from workqueue */
 static void bpf_map_free_deferred(struct work_struct *work)
 {

@@ -141,6 +166,7 @@ static void bpf_map_put_uref(struct bpf_map *map)
 void bpf_map_put(struct bpf_map *map)
 {
 	if (atomic_dec_and_test(&map->refcnt)) {
+		bpf_map_free_id(map);
 		INIT_WORK(&map->work, bpf_map_free_deferred);
 		schedule_work(&map->work);
 	}

@@ -239,14 +265,20 @@ static int map_create(union bpf_attr *attr)
 	if (err)
 		goto free_map_nouncharge;
 
+	err = bpf_map_alloc_id(map);
+	if (err)
+		goto free_map;
+
 	err = bpf_map_new_fd(map);
 	if (err < 0)
 		/* failed to allocate fd */
-		goto free_map;
+		goto free_id;
 
 	trace_bpf_map_create(map, err);
 	return err;
 
+free_id:
+	bpf_map_free_id(map);
 free_map:
 	bpf_map_uncharge_memlock(map);
 free_map_nouncharge:

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:15 UTC #4 lore raw

PATCH v3 4/8 bpf: Add BPF_PROG_GET_FD_BY_ID

Add BPF_PROG_GET_FD_BY_ID command to allow user to get a fd from a bpf_prog's ID.

bpf_prog_inc_not_zero() is added and is called with prog_idr_lock held.

__bpf_prog_put() is also added which has the 'bool do_idr_lock' param to decide if the prog_idr_lock should be acquired when freeing the prog->id.

In the error path of bpf_prog_inc_not_zero(), it may have to call __bpf_prog_put(map, false) which does not need to take the prog_idr_lock when freeing the prog->id.

It is currently limited to CAP_SYS_ADMIN which we can consider to lift it in followup patches.

---
 include/uapi/linux/bpf.h |  8 +++--
 kernel/bpf/syscall.c     | 91 ++++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 87 insertions(+), 12 deletions(-)

include/uapi/linux/bpf.h +6 -2

index 629747a3f273..d70cfed19d5e 100644
@@ -84,6 +84,7 @@ enum bpf_cmd {
 	BPF_PROG_TEST_RUN,
 	BPF_PROG_GET_NEXT_ID,
 	BPF_MAP_GET_NEXT_ID,
+	BPF_PROG_GET_FD_BY_ID,
 };
 
 enum bpf_map_type {

@@ -212,8 +213,11 @@ union bpf_attr {
 		__u32		duration;
 	} test;
 
-	struct { /* anonymous struct used by BPF_*_GET_NEXT_ID */
-		__u32		start_id;
+	struct { /* anonymous struct used by BPF_*_GET_*_ID */
+		union {
+			__u32		start_id;
+			__u32		prog_id;
+		};
 		__u32		next_id;
 	};
 } __attribute__((aligned(8)));

kernel/bpf/syscall.c +81 ~~-10~~

146 lines not on this page · raw message

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:15 UTC #5 lore raw

PATCH v3 5/8 bpf: Add BPF_MAP_GET_FD_BY_ID

Add BPF_MAP_GET_FD_BY_ID command to allow user to get a fd from a bpf_map's ID.

bpf_map_inc_not_zero() is added and is called with map_idr_lock held.

__bpf_map_put() is also added which has the 'bool do_idr_lock' param to decide if the map_idr_lock should be acquired when freeing the map->id.

In the error path of bpf_map_inc_not_zero(), it may have to call __bpf_map_put(map, false) which does not need to take the map_idr_lock when freeing the map->id.

It is currently limited to CAP_SYS_ADMIN which we can consider to lift it in followup patches.

---
 include/uapi/linux/bpf.h |  2 +
 kernel/bpf/syscall.c     | 95 +++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 87 insertions(+), 10 deletions(-)

include/uapi/linux/bpf.h +2 -0

index d70cfed19d5e..dd23f47ff00c 100644
@@ -85,6 +85,7 @@ enum bpf_cmd {
 	BPF_PROG_GET_NEXT_ID,
 	BPF_MAP_GET_NEXT_ID,
 	BPF_PROG_GET_FD_BY_ID,
+	BPF_MAP_GET_FD_BY_ID,
 };
 
 enum bpf_map_type {

@@ -217,6 +218,7 @@ union bpf_attr {
 		union {
 			__u32		start_id;
 			__u32		prog_id;
+			__u32		map_id;
 		};
 		__u32		next_id;
 	};

kernel/bpf/syscall.c +85 ~~-10~~

145 lines not on this page · raw message

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:15 UTC #7 lore raw

PATCH v3 7/8 bpf: Add BPF_OBJ_GET_INFO_BY_FD

A single BPF_OBJ_GET_INFO_BY_FD cmd is used to obtain the info for both bpf_prog and bpf_map. The kernel can figure out the fd is associated with a bpf_prog or bpf_map.

The suggested struct bpf_prog_info and struct bpf_map_info are not meant to be a complete list and it is not the goal of this patch. New fields can be added in the future patch.

The focus of this patch is to create the interface, BPF_OBJ_GET_INFO_BY_FD cmd for exposing the bpf_prog's and bpf_map's info.

The obj's info, which will be extended (and get bigger) over time, is separated from the bpf_attr to avoid bloating the bpf_attr.

---
 include/linux/filter.h   |   2 -
 include/uapi/linux/bpf.h |  28 ++++++++
 kernel/bpf/syscall.c     | 163 ++++++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 174 insertions(+), 19 deletions(-)

include/linux/filter.h +0 -2

index 1e2dddf21f3b..1fa26dc562ce 100644
@@ -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 */

include/uapi/linux/bpf.h +28 -0

index dd23f47ff00c..9b2c10b45733 100644
@@ -86,6 +86,7 @@ enum bpf_cmd {
 	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 {

@@ -222,6 +223,12 @@ union bpf_attr {
 		};
 		__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:

@@ -686,4 +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__ */

kernel/bpf/syscall.c +146 ~~-17~~

181 lines not on this page · raw message

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:15 UTC #6 lore raw

PATCH v3 6/8 bpf: Add jited_len to struct bpf_prog

Add jited_len to struct bpf_prog. It will be useful for the struct bpf_prog_info which will be added in the later patch.

---
 arch/arm64/net/bpf_jit_comp.c     | 1 +
 arch/powerpc/net/bpf_jit_comp64.c | 1 +
 arch/s390/net/bpf_jit_comp.c      | 1 +
 arch/sparc/net/bpf_jit_comp_64.c  | 1 +
 arch/x86/net/bpf_jit_comp.c       | 1 +
 include/linux/filter.h            | 1 +
 6 files changed, 6 insertions(+)

arch/arm64/net/bpf_jit_comp.c +1 -0

index b1d38eeb24f6..4f95873d7142 100644
@@ -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 +1 -0

index a01366584a4b..861c5af1c9c4 100644
@@ -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 +1 -0

index 42ad3832586c..01c6fbc3e85b 100644
@@ -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 +1 -0

index 098874a81f6e..8799ae9a8788 100644
@@ -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 +1 -0

index 617eac9c4511..e1324f280e06 100644
@@ -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/filter.h +1 -0

index a20ba40fcb73..1e2dddf21f3b 100644
@@ -432,6 +432,7 @@ 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 */

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:16 UTC #0 lore raw

v3 cover 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:
8726679a0fa3 ("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().

Martin KaFai Lau (8):
bpf: Introduce bpf_prog ID
bpf: Introduce bpf_map ID
bpf: Add BPF_(PROG|MAP)_GET_NEXT_ID command
bpf: Add BPF_PROG_GET_FD_BY_ID
bpf: Add BPF_MAP_GET_FD_BY_ID
bpf: Add jited_len to struct bpf_prog
bpf: Add BPF_OBJ_GET_INFO_BY_FD
bpf: Test for bpf ID

 arch/arm64/net/bpf_jit_comp.c             |   1 +
 arch/powerpc/net/bpf_jit_comp64.c         |   1 +
 arch/s390/net/bpf_jit_comp.c              |   1 +
 arch/sparc/net/bpf_jit_comp_64.c          |   1 +
 arch/x86/net/bpf_jit_comp.c               |   1 +
 include/linux/bpf.h                       |   2 +
 include/linux/filter.h                    |   3 +-
 include/uapi/linux/bpf.h                  |  41 +++
 kernel/bpf/syscall.c                      | 433 ++++++++++++++++++++++++++++--
 tools/include/uapi/linux/bpf.h            |  41 +++
 tools/lib/bpf/bpf.c                       |  68 +++++
 tools/lib/bpf/bpf.h                       |   5 +
 tools/testing/selftests/bpf/Makefile      |   2 +-
 tools/testing/selftests/bpf/test_obj_id.c |  35 +++
 tools/testing/selftests/bpf/test_progs.c  | 191 +++++++++++++
 15 files changed, 798 insertions(+), 28 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/test_obj_id.c

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:16 UTC #3 lore raw

PATCH v3 3/8 bpf: Add BPF_(PROG|MAP)_GET_NEXT_ID command

This patch adds BPF_PROG_GET_NEXT_ID and BPF_MAP_GET_NEXT_ID to allow userspace to iterate all bpf_prog IDs and bpf_map IDs.

The API is trying to be consistent with the existing BPF_MAP_GET_NEXT_KEY.

It is currently limited to CAP_SYS_ADMIN which we can consider to lift it in followup patches.

---
 include/uapi/linux/bpf.h |  7 +++++++
 kernel/bpf/syscall.c     | 38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+)

include/uapi/linux/bpf.h +7 -0

index e78aece03628..629747a3f273 100644
@@ -82,6 +82,8 @@ enum bpf_cmd {
 	BPF_PROG_ATTACH,
 	BPF_PROG_DETACH,
 	BPF_PROG_TEST_RUN,
+	BPF_PROG_GET_NEXT_ID,
+	BPF_MAP_GET_NEXT_ID,
 };
 
 enum bpf_map_type {

@@ -209,6 +211,11 @@ union bpf_attr {
 		__u32		repeat;
 		__u32		duration;
 	} test;
+
+	struct { /* anonymous struct used by BPF_*_GET_NEXT_ID */
+		__u32		start_id;
+		__u32		next_id;
+	};
 } __attribute__((aligned(8)));
 
 /* BPF helper function descriptions:

kernel/bpf/syscall.c +38 -0

index 4c3075b5d840..2405feedb8c1 100644
@@ -166,6 +166,7 @@ static void bpf_map_put_uref(struct bpf_map *map)
 void bpf_map_put(struct bpf_map *map)
 {
 	if (atomic_dec_and_test(&map->refcnt)) {
+		/* bpf_map_free_id() must be called first */
 		bpf_map_free_id(map);
 		INIT_WORK(&map->work, bpf_map_free_deferred);
 		schedule_work(&map->work);

@@ -726,6 +727,7 @@ void bpf_prog_put(struct bpf_prog *prog)
 {
 	if (atomic_dec_and_test(&prog->aux->refcnt)) {
 		trace_bpf_prog_put_rcu(prog);
+		/* bpf_prog_free_id() must be called first */
 		bpf_prog_free_id(prog);
 		bpf_prog_kallsyms_del(prog);
 		call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu);

@@ -1069,6 +1071,34 @@ static int bpf_prog_test_run(const union bpf_attr *attr,
 	return ret;
 }
 
+#define BPF_OBJ_GET_NEXT_ID_LAST_FIELD next_id
+
+static int bpf_obj_get_next_id(const union bpf_attr *attr,
+			       union bpf_attr __user *uattr,
+			       struct idr *idr,
+			       spinlock_t *lock)
+{
+	u32 next_id = attr->start_id;
+	int err = 0;
+
+	if (CHECK_ATTR(BPF_OBJ_GET_NEXT_ID) || next_id >= INT_MAX)
+		return -EINVAL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	next_id++;
+	spin_lock_bh(lock);
+	if (!idr_get_next(idr, &next_id))
+		err = -ENOENT;
+	spin_unlock_bh(lock);
+
+	if (!err)
+		err = put_user(next_id, &uattr->next_id);
+
+	return err;
+}
+
 SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, size)
 {
 	union bpf_attr attr = {};

@@ -1146,6 +1176,14 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
 	case BPF_PROG_TEST_RUN:
 		err = bpf_prog_test_run(&attr, uattr);
 		break;
+	case BPF_PROG_GET_NEXT_ID:
+		err = bpf_obj_get_next_id(&attr, uattr,
+					  &prog_idr, &prog_idr_lock);
+		break;
+	case BPF_MAP_GET_NEXT_ID:
+		err = bpf_obj_get_next_id(&attr, uattr,
+					  &map_idr, &map_idr_lock);
+		break;
 	default:
 		err = -EINVAL;
 		break;

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:16 UTC #8 lore raw

PATCH v3 8/8 bpf: Test for bpf ID

Add test to exercise the bpf_prog/map id generation, bpf_(prog|map)_get_next_id(), bpf_(prog|map)_get_fd_by_id() and bpf_get_obj_info_by_fd().

---
 tools/include/uapi/linux/bpf.h            |  41 +++++++
 tools/lib/bpf/bpf.c                       |  68 +++++++++++
 tools/lib/bpf/bpf.h                       |   5 +
 tools/testing/selftests/bpf/Makefile      |   2 +-
 tools/testing/selftests/bpf/test_obj_id.c |  35 ++++++
 tools/testing/selftests/bpf/test_progs.c  | 191 ++++++++++++++++++++++++++++++
 6 files changed, 341 insertions(+), 1 deletion(-)
 create mode 100644 tools/testing/selftests/bpf/test_obj_id.c

tools/include/uapi/linux/bpf.h +41 -0

index e78aece03628..9b2c10b45733 100644
@@ -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 +68 -0

71 lines not on this page · raw message

tools/lib/bpf/bpf.h +5 -0

index 972bd8333eb7..16de44a14b48 100644
@@ -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 +1 -1

index f389b02d43a0..9f0e07ba5334 100644
@@ -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 new file +35 -0

new file mode 100644
index 000000000000..d8723aaf827a
@@ -0,0 +1,35 @@
+/* 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 +191 -0

215 lines not on this page · raw message

Signature · 1 line

2.9.3

Martin KaFai Lau

author

2017-06-05 19:16 UTC #1 lore raw

PATCH v3 1/8 bpf: Introduce bpf_prog ID

This patch generates an unique ID for each BPF_PROG_LOAD-ed prog. It is worth to note that each BPF_PROG_LOAD-ed prog will have a different ID even they have the same bpf instructions.

The ID is generated by the existing idr_alloc_cyclic().
The ID is ranged from [1, INT_MAX). It is allocated in cyclic manner, so an ID will get reused every 2 billion BPF_PROG_LOAD.

The bpf_prog_alloc_id() is done after bpf_prog_select_runtime() because the jit process may have allocated a new prog. Hence, we need to ensure the value of pointer 'prog' will not be changed any more before storing the prog to the prog_idr.

After bpf_prog_select_runtime(), the prog is read-only. Hence, the id is stored in 'struct bpf_prog_aux'.

---
 include/linux/bpf.h  |  1 +
 kernel/bpf/syscall.c | 40 +++++++++++++++++++++++++++++++++++++++-
 2 files changed, 40 insertions(+), 1 deletion(-)

include/linux/bpf.h +1 -0

index fcc80ca11045..c5946d19f2ca 100644
@@ -172,6 +172,7 @@ 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;

kernel/bpf/syscall.c +39 -1

index 59da103adb85..2a1b32b470f1 100644
@@ -22,8 +22,11 @@
 #include <linux/filter.h>
 #include <linux/version.h>
 #include <linux/kernel.h>
+#include <linux/idr.h>
 
 DEFINE_PER_CPU(int, bpf_prog_active);
+static DEFINE_IDR(prog_idr);
+static DEFINE_SPINLOCK(prog_idr_lock);
 
 int sysctl_unprivileged_bpf_disabled __read_mostly;

@@ -650,6 +653,34 @@ static void bpf_prog_uncharge_memlock(struct bpf_prog *prog)
 	free_uid(user);
 }
 
+static int bpf_prog_alloc_id(struct bpf_prog *prog)
+{
+	int id;
+
+	spin_lock_bh(&prog_idr_lock);
+	id = idr_alloc_cyclic(&prog_idr, prog, 1, INT_MAX, GFP_ATOMIC);
+	if (id > 0)
+		prog->aux->id = id;
+	spin_unlock_bh(&prog_idr_lock);
+
+	/* id is in [1, INT_MAX) */
+	if (WARN_ON_ONCE(!id))
+		return -ENOSPC;
+
+	return id > 0 ? 0 : id;
+}
+
+static void bpf_prog_free_id(struct bpf_prog *prog)
+{
+	/* cBPF to eBPF migrations are currently not in the idr store. */
+	if (!prog->aux->id)
+		return;
+
+	spin_lock_bh(&prog_idr_lock);
+	idr_remove(&prog_idr, prog->aux->id);
+	spin_unlock_bh(&prog_idr_lock);
+}
+
 static void __bpf_prog_put_rcu(struct rcu_head *rcu)
 {
 	struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu);

@@ -663,6 +694,7 @@ void bpf_prog_put(struct bpf_prog *prog)
 {
 	if (atomic_dec_and_test(&prog->aux->refcnt)) {
 		trace_bpf_prog_put_rcu(prog);
+		bpf_prog_free_id(prog);
 		bpf_prog_kallsyms_del(prog);
 		call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu);
 	}

@@ -857,15 +889,21 @@ static int bpf_prog_load(union bpf_attr *attr)
 	if (err < 0)
 		goto free_used_maps;
 
+	err = bpf_prog_alloc_id(prog);
+	if (err)
+		goto free_used_maps;
+
 	err = bpf_prog_new_fd(prog);
 	if (err < 0)
 		/* failed to allocate fd */
-		goto free_used_maps;
+		goto free_id;
 
 	bpf_prog_kallsyms_add(prog);
 	trace_bpf_prog_load(prog, err);
 	return err;
 
+free_id:
+	bpf_prog_free_id(prog);
 free_used_maps:
 	free_used_maps(prog->aux);
 free_prog:

Signature · 1 line

2.9.3

David Miller

maintainer

2017-06-06 19:41 UTC #9 lore raw