History

bpf: implement numbers iterator

Implement the first open-coded iterator type over a range of integers. It's public API consists of: - bpf_iter_num_new() constructor, which accepts [start, end) range (that is, start is inclusive, end is exclusive). - bpf_iter_num_next() which will keep returning read-only pointer to int until the range is exhausted, at which point NULL will be returned. If bpf_iter_num_next() is kept calling after this, NULL will be persistently returned. - bpf_iter_num_destroy() destructor, which needs to be called at some point to clean up iterator state. BPF verifier enforces that iterator destructor is called at some point before BPF program exits. Note that `start = end = X` is a valid combination to setup an empty iterator. bpf_iter_num_new() will return 0 (success) for any such combination. If bpf_iter_num_new() detects invalid combination of input arguments, it returns error, resets iterator state to, effectively, empty iterator, so any subsequent call to bpf_iter_num_next() will keep returning NULL. BPF verifier has no knowledge that returned integers are in the [start, end) value range, as both `start` and `end` are not statically known and enforced: they are runtime values. While the implementation is pretty trivial, some care needs to be taken to avoid overflows and underflows. Subsequent selftests will validate correctness of [start, end) semantics, especially around extremes (INT_MIN and INT_MAX). Similarly to bpf_loop(), we enforce that no more than BPF_MAX_LOOPS can be specified. bpf_iter_num_{new,next,destroy}() is a logical evolution from bounded BPF loops and bpf_loop() helper and is the basis for implementing ergonomic BPF loops with no statically known or verified bounds. Subsequent patches implement bpf_for() macro, demonstrating how this can be wrapped into something that works and feels like a normal for() loop in C language.

Andrii Nakryiko committed 3 years ago

Commit 6018e1f1 parent 06accc85 files · +95 −2Browse files

Unified · Split

5 files
  1. include/linux/bpf.h +6 −2
  2. include/uapi/linux/bpf.h +8 −0
  3. kernel/bpf/bpf_iter.c +70 −0
  4. kernel/bpf/helpers.c +3 −0
  5. tools/include/uapi/linux/bpf.h +8 −0
include/linux/bpf.h+6−2
@@ -1617,8 +1617,12 @@ struct bpf_array { · 1616 unmodified lines
16171617 #define BPF_COMPLEXITY_LIMIT_INSNS 1000000 /* yes. 1M insns */
16181618 #define MAX_TAIL_CALL_CNT 33
16191619
1620−/* Maximum number of loops for bpf_loop */
1621−#define BPF_MAX_LOOPS BIT(23)
1620+/* Maximum number of loops for bpf_loop and bpf_iter_num.
1621+ * It's enum to expose it (and thus make it discoverable) through BTF.
1622+ */
1623+enum {
1624+ BPF_MAX_LOOPS = 8 * 1024 * 1024,
1625+};
16221626
16231627 #define BPF_F_ACCESS_MASK (BPF_F_RDONLY | \
16241628 BPF_F_RDONLY_PROG | \
include/uapi/linux/bpf.h+8−0
@@ -7112,4 +7112,12 @@ enum { · 7111 unmodified lines
71127112 BPF_F_TIMER_ABS = (1ULL << 0),
71137113 };
71147114
7115+/* BPF numbers iterator state */
7116+struct bpf_iter_num {
7117+ /* opaque iterator state; having __u64 here allows to preserve correct
7118+ * alignment requirements in vmlinux.h, generated from BTF
7119+ */
7120+ __u64 __opaque[1];
7121+} __attribute__((aligned(8)));
7122+
71157123 #endif /* _UAPI__LINUX_BPF_H__ */
kernel/bpf/bpf_iter.c+70−0
@@ -776,3 +776,73 @@ const struct bpf_func_proto bpf_loop_proto = { · 775 unmodified lines
776776 .arg3_type = ARG_PTR_TO_STACK_OR_NULL,
777777 .arg4_type = ARG_ANYTHING,
778778 };
779+
780+struct bpf_iter_num_kern {
781+ int cur; /* current value, inclusive */
782+ int end; /* final value, exclusive */
783+} __aligned(8);
784+
785+__diag_push();
786+__diag_ignore_all("-Wmissing-prototypes",
787+ "Global functions as their definitions will be in vmlinux BTF");
788+
789+__bpf_kfunc int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end)
790+{
791+ struct bpf_iter_num_kern *s = (void *)it;
792+
793+ BUILD_BUG_ON(sizeof(struct bpf_iter_num_kern) != sizeof(struct bpf_iter_num));
794+ BUILD_BUG_ON(__alignof__(struct bpf_iter_num_kern) != __alignof__(struct bpf_iter_num));
795+
796+ BTF_TYPE_EMIT(struct btf_iter_num);
797+
798+ /* start == end is legit, it's an empty range and we'll just get NULL
799+ * on first (and any subsequent) bpf_iter_num_next() call
800+ */
801+ if (start > end) {
802+ s->cur = s->end = 0;
803+ return -EINVAL;
804+ }
805+
806+ /* avoid overflows, e.g., if start == INT_MIN and end == INT_MAX */
807+ if ((s64)end - (s64)start > BPF_MAX_LOOPS) {
808+ s->cur = s->end = 0;
809+ return -E2BIG;
810+ }
811+
812+ /* user will call bpf_iter_num_next() first,
813+ * which will set s->cur to exactly start value;
814+ * underflow shouldn't matter
815+ */
816+ s->cur = start - 1;
817+ s->end = end;
818+
819+ return 0;
820+}
821+
822+__bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it)
823+{
824+ struct bpf_iter_num_kern *s = (void *)it;
825+
826+ /* check failed initialization or if we are done (same behavior);
827+ * need to be careful about overflow, so convert to s64 for checks,
828+ * e.g., if s->cur == s->end == INT_MAX, we can't just do
829+ * s->cur + 1 >= s->end
830+ */
831+ if ((s64)(s->cur + 1) >= s->end) {
832+ s->cur = s->end = 0;
833+ return NULL;
834+ }
835+
836+ s->cur++;
837+
838+ return &s->cur;
839+}
840+
841+__bpf_kfunc void bpf_iter_num_destroy(struct bpf_iter_num *it)
842+{
843+ struct bpf_iter_num_kern *s = (void *)it;
844+
845+ s->cur = s->end = 0;
846+}
847+
848+__diag_pop();
kernel/bpf/helpers.c+3−0
@@ -2411,6 +2411,9 @@ BTF_ID_FLAGS(func, bpf_rcu_read_lock) · 2410 unmodified lines
24112411 BTF_ID_FLAGS(func, bpf_rcu_read_unlock)
24122412 BTF_ID_FLAGS(func, bpf_dynptr_slice, KF_RET_NULL)
24132413 BTF_ID_FLAGS(func, bpf_dynptr_slice_rdwr, KF_RET_NULL)
2414+BTF_ID_FLAGS(func, bpf_iter_num_new, KF_ITER_NEW)
2415+BTF_ID_FLAGS(func, bpf_iter_num_next, KF_ITER_NEXT | KF_RET_NULL)
2416+BTF_ID_FLAGS(func, bpf_iter_num_destroy, KF_ITER_DESTROY)
24142417 BTF_SET8_END(common_btf_ids)
24152418
24162419 static const struct btf_kfunc_id_set common_kfunc_set = {
tools/include/uapi/linux/bpf.h+8−0
@@ -7112,4 +7112,12 @@ enum { · 7111 unmodified lines
71127112 BPF_F_TIMER_ABS = (1ULL << 0),
71137113 };
71147114
7115+/* BPF numbers iterator state */
7116+struct bpf_iter_num {
7117+ /* opaque iterator state; having __u64 here allows to preserve correct
7118+ * alignment requirements in vmlinux.h, generated from BTF
7119+ */
7120+ __u64 __opaque[1];
7121+} __attribute__((aligned(8)));
7122+
71157123 #endif /* _UAPI__LINUX_BPF_H__ */

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t