bpf: Inline the numeric open-coded iterator kfuncs
Subsystems (5)
v5 cover bpf: Inline the numeric open-coded iterator kfuncs
The bpf_for(i, start, end) macro is BPF's open-coded numeric iterator. It expands into calls to three kfuncs: bpf_iter_num_new() to set the iterator up, bpf_iter_num_next() once per iteration, and bpf_iter_num_destroy() to tear it down. The verifier emits these as ordinary kfunc calls, so a bpf_for() loop pays function-call overhead on setup, teardown, and -- most importantly -- on every single iteration via bpf_iter_num_next().
All three kfuncs are tiny and only touch the 8-byte on-stack iterator state (struct bpf_iter_num_kern { int cur; int end; }). That makes them good candidates for inlining, the same way several other special kfuncs are already open-coded in bpf_fixup_kfunc_call(). This series replaces each of the three calls with an equivalent inline BPF instruction sequence:
- bpf_iter_num_new(): the end - start range check is done with 32-bit arithmetic (start <= end is checked first, so the distance fits in a u32) and range-checked against BPF_MAX_LOOPS as unsigned. This avoids the cpuv4 sign-extension insns that some JITs do not implement. Returns the same -EINVAL / -E2BIG / 0 as the kfunc.
- bpf_iter_num_next(): the hot path. cur and end are int, so the kfunc's s->cur + 1 >= s->end test is an ordinary signed 32-bit compare and the inlined code needs no sign extension.
- bpf_iter_num_destroy(): the stack slot is no longer tracked as iterator state once destroy() returns, so nothing needs to be written to it. Both the kfunc and the inlined form become a no-op, which just drops the call.
The emitted instructions are plain BPF and remain valid for the interpreter, so interpreter fallback stays correct and no jit_required marking is needed.
Benchmark (./bench -p 1 --nr_loops 1000000 {bpf-loop,bpf-for}):
+--------+---------------------+---------------------+---------------------+
| arch | bpf_loop | bpf_for non-inlined | bpf_for inlined |
+--------+---------------------+---------------------+---------------------+
| x86-64 | 4252 M/s (0.24 ns) | 3608 M/s (0.28 ns) | 7417 M/s (0.13 ns) |
+--------+---------------------+---------------------+---------------------+
| arm64 | 649 M/s (1.54 ns) | 548 M/s (1.82 ns) | 546 M/s (1.83 ns) |
+--------+---------------------+---------------------+---------------------+On x86-64 removing the per-iteration call roughly doubles bpf_for() throughput. On arm64 it is neutral, and rather than guess why this was checked with perf: inlining removes ~28% of the executed instructions (the call) but leaves the cycle count unchanged -- IPC drops from ~4.2 to ~3.0 and backend stalls rise from ~50% to ~66%. The loop is bound by the latency of the iterator's on-stack counter, not by call overhead:bpf_iter_num_next() loads s->cur from the stack, increments it and stores it back each iteration, and the next iteration's load depends on that store. The removed call instructions were executing in the shadow of that store->load stall and were never on the critical path.
A small userspace microbenchmark isolates the effect: a same-address store->load->add round-trip (the shape of the on-stack counter) costs ~6 cycles/iteration on the tested arm64 core but ~1 cycle on x86-64, where the core collapses the same-address round-trip into a register move (memory renaming / store-to-load-forwarding elimination). So on x86-64 the loop is not latency-bound and the per-iteration call dominates -- removing it is the ~2x win -- whereas on arm64 the call fits entirely inside the store->load stall the loop already has, so adding or removing it changes nothing.
bpf_loop() is shown for reference only; it is a different construct (a callback invoked per iteration) and this series does not change it. Its counter lives in a register rather than on the stack, so on arm64 it avoids the store->load latency above and is faster than bpf_for() there.
Changelog:
v4: https://lore.kernel.org/all/[email protected]/
Changes in v5:
- Inline the new()/next()/destroy() sequences directly in bpf_fixup_kfunc_call() instead of via helper functions (Andrii Nakryiko)
- Trim the code comments; keep the explanation in the bpf_iter.c kfuncs and leave only brief comments at the inline sites, and shorten the bpf_iter_num_destroy() kfunc to /* no-op */ (Andrii Nakryiko)
- Reword the patch 1 comment so it no longer forward-references the inlined bpf_iter_num_next(), which is only added later in the series (bpf-ci)
- Switch the bpf_iter_num_next() comment to the networking multi-line style
v3: https://lore.kernel.org/all/[email protected]/
Changes in v4:
- Drop the "elide range checks for constant bounds" patch (Andrii Nakryiko)
- bpf_iter_num_new(): range-check the distance against BPF_MAX_LOOPS with an unsigned compare (Andrii Nakryiko)
- bpf_iter_num_destroy(): make it a no-op in both the kfunc and the inlined form instead of zeroing the iterator state (Andrii Nakryiko)
- New patch: fix the misleading overflow comment in bpf_iter_num_next() and drop the redundant (s64) cast; the int wraparound is intentional and load-bearing (Andrii Nakryiko)
- bpf_for benchmark: nr_loops is int, matching what bpf_for() expects (Andrii Nakryiko)
- Corroborate the arm64/x86 benchmark difference with perf counters and a store-to-load-forwarding microbenchmark (Kumar Kartikeya Dwivedi, Andrii Nakryiko)
v2: https://lore.kernel.org/bpf/[email protected]/
Changes in v3:
- Elide the range checks in bpf_iter_num_new() when start and end are constant, marking the registers precise so paths reaching the call with different constants are not pruned (Eduard Zingerman)
- Add __xlated selftests pinning the inlined new()/next()/destroy() shapes (Eduard Zingerman)
- Use the insn_buf[i++] idiom in the inline helpers (Eduard Zingerman)
- Pick up Acked-by on patch 3
v1: https://lore.kernel.org/all/[email protected]/
Changes in v2:
- Don't emit sign-extending (movsx) moves; some JITs (e.g. x86-32, mips32, sparc64) decode them as a plain move and would miscompile the range check
Puranjay Mohan (6):
bpf: Correct the overflow check comment in bpf_iter_num_next()
bpf: Inline bpf_iter_num_new() kfunc
bpf: Inline bpf_iter_num_next() kfunc
bpf: Inline bpf_iter_num_destroy() as a no-op selftests/bpf: Verify inlined numeric iterator shape with __xlated selftests/bpf: Add bpf_for() benchmark
kernel/bpf/bpf_iter.c | 17 ++- kernel/bpf/verifier.c | 45 ++++++++ tools/testing/selftests/bpf/Makefile | 2 + tools/testing/selftests/bpf/bench.c | 4 + .../selftests/bpf/benchs/bench_bpf_for.c | 104 ++++++++++++++++++ .../selftests/bpf/benchs/run_bench_bpf_for.sh | 15 +++ .../selftests/bpf/progs/bpf_for_bench.c | 32 ++++++ tools/testing/selftests/bpf/progs/iters.c | 83 ++++++++++++++ 8 files changed, 292 insertions(+), 10 deletions(-) create mode 100644 tools/testing/selftests/bpf/benchs/bench_bpf_for.c create mode 100755 tools/testing/selftests/bpf/benchs/run_bench_bpf_for.sh create mode 100644 tools/testing/selftests/bpf/progs/bpf_for_bench.c
base-commit: 7f333f85f83da9a20a60a1d8bf518c4ae0818e5a
Signature · 1 line
2.53.0-Meta
PATCH v5 1/6 bpf: Correct the overflow check comment in bpf_iter_num_next()
The comment on the s->cur + 1 >= s->end check claims the (s64) cast is needed to avoid overflow when s->cur == s->end == INT_MAX. It isn't:s->cur + 1 is computed in int and wraps before the cast, so the cast changes nothing (INT_MAX + 1 compares the same either way).
The wraparound is the point. bpf_iter_num_new() sets s->cur = start - 1, which wraps to INT_MAX for start == INT_MIN, and the wrapping s->cur + 1 brings it back to start. (s64)s->cur + 1 would instead break iterators starting at INT_MIN.
Drop the cast and reword the comment. No functional change; the wrap is well-defined under -fno-strict-overflow.
--- kernel/bpf/bpf_iter.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)
kernel/bpf/bpf_iter.c +4 -5
@@ -802,12 +802,11 @@ __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) {struct bpf_iter_num_kern *s = (void *)it; -/* check failed initialization or if we are done (same behavior); -* need to be careful about overflow, so convert to s64 for checks, -* e.g., if s->cur == s->end == INT_MAX, we can't just do -* s->cur + 1 >= s->end +/* +* s->cur < s->end while iterating, else s->cur == s->end == 0; the signed +* s->cur + 1 >= s->end holds even when s->cur + 1 wraps (start == INT_MIN). */ -if ((s64)(s->cur +1 ) >= s->end) { +if (s->cur +1 >= s->end) { s->cur = s->end =0 ;return NULL ; }
Signature · 1 line
2.53.0-Meta
[email protected]
AI reviewRe: [PATCH bpf-next v5 1/6] bpf: Correct the overflow check comment in bpf_iter_num_next()
The replacement comment says "s->cur + 1 >= s->end holds even when s->cur + 1 wraps (start == INT_MIN)". Doesn't the inequality evaluate to false in the wrap case, not true?
Quoted 5 lines
kernel/bpf/bpf_iter.c +0
-0
kernel/bpf/bpf_iter.c __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) · line 807
Quoted 10 lines
kernel/bpf/bpf_iter.c +3
-4@@ -802,12 +802,11 @@ __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) {struct bpf_iter_num_kern *s = (void *)it; -/* check failed initialization or if we are done (same behavior); -* need to be careful about overflow, so convert to s64 for checks, -* e.g., if s->cur == s->end == INT_MAX, we can't just do -* s->cur + 1 >= s->end +/* +* s->cur < s->end while iterating, else s->cur == s->end == 0; the signed +* s->cur + 1 >= s->end holds even when s->cur + 1 wraps (start == INT_MIN).
^^^^^
kernel/bpf/bpf_iter.c __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) · line 812
Quoted 6 lines
kernel/bpf/bpf_iter.c +1
-1*/ -if ((s64)(s->cur +1 ) >= s->end) { +if (s->cur +1 >= s->end) { s->cur = s->end =0 ;return NULL ; }
The replacement comment says "s->cur + 1 >= s->end holds even when s->cur + 1 wraps (start == INT_MIN)". Doesn't the inequality evaluate to false in the wrap case, not true?
Looking at the code: bpf_iter_num_new() sets s->cur = start - 1, so when start == INT_MIN, s->cur wraps to INT_MAX. On the first call to bpf_iter_num_next(), s->cur + 1 wraps back to INT_MIN. For a non-empty range like the selftest bpf_for(i, INT_MIN, INT_MIN + 2) at tools/testing/selftests/bpf/progs/iters_num.c:101, the test becomes INT_MIN >= INT_MIN + 2, which is false, so the function returns the first element.
The commit message describes this correctly ("the wrapping s->cur + 1 brings it back to start"), but "X >= Y holds" reads as "the inequality is true," when what seems to be meant is that the comparison mechanism remains correct under wraparound.
Also, the first clause says "s->cur < s->end while iterating, else s->cur == s->end == 0". This doesn't cover the post-bpf_iter_num_new(), pre-first-next() state for start == INT_MIN, where s->cur == INT_MAX and s->end == INT_MIN + 2 (neither s->cur < s->end nor s->cur == s->end == 0).
Since the sole purpose of this patch is to make the comment accurate, could the wording be adjusted to avoid asserting a condition that is false in the cited case?
---
AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md
CI run summary: https://github.com/kernel-patches/bpf/actions/runs/30915992491
Re: [PATCH bpf-next v5 1/6] bpf: Correct the overflow check comment in bpf_iter_num_next()
Quoted 41 lines · On Tue, Aug 4, 2026 at 3:47 PM <[email protected]> wrote:
Quoted 6 lines
kernel/bpf/bpf_iter.c +0
-0kernel/bpf/bpf_iter.c __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) · line 807
Quoted 10 lines
kernel/bpf/bpf_iter.c +3
-4@@ -802,12 +802,11 @@ __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) {struct bpf_iter_num_kern *s = (void *)it; -/* check failed initialization or if we are done (same behavior); -* need to be careful about overflow, so convert to s64 for checks, -* e.g., if s->cur == s->end == INT_MAX, we can't just do -* s->cur + 1 >= s->end +/* +* s->cur < s->end while iterating, else s->cur == s->end == 0; the signed +* s->cur + 1 >= s->end holds even when s->cur + 1 wraps (start == INT_MIN). ^^^^^kernel/bpf/bpf_iter.c __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) · line 812
Quoted 6 lines
kernel/bpf/bpf_iter.c +1
-1*/ -if ((s64)(s->cur +1 ) >= s->end) { +if (s->cur +1 >= s->end) { s->cur = s->end =0 ;return NULL ; }The replacement comment says "
s->cur+ 1 >=s->endholds even whens->cur+ 1 wraps (start ==INT_MIN)". Doesn't the inequality evaluate to false in the wrap case, not true?Looking at the code:
bpf_iter_num_new()setss->cur= start - 1, so when start ==INT_MIN,s->curwraps toINT_MAX. On the first call tobpf_iter_num_next(),s->cur+ 1 wraps back toINT_MIN. For a non-empty range like the selftestbpf_for(i, INT_MIN, INT_MIN + 2)attools/testing/selftests/bpf/progs/iters_num.c:101, the test becomesINT_MIN>=INT_MIN+ 2, which is false, so the function returns the first element.The commit message describes this correctly ("the wrapping
s->cur+ 1 brings it back to start"), but "X >= Y holds" reads as "the inequality is true," when what seems to be meant is that the comparison mechanism remains correct under wraparound.
Quoted 8 lines
Also, the first clause says "
s->cur<s->endwhile iterating, elses->cur==s->end== 0". This doesn't cover the post-bpf_iter_num_new(), pre-first-next() state for start ==INT_MIN, wheres->cur==INT_MAXands->end==INT_MIN+ 2 (neithers->cur<s->endnors->cur==s->end== 0).Since the sole purpose of this patch is to make the comment accurate, could the wording be adjusted to avoid asserting a condition that is false in the cited case?
Andrii do you have a preference for the comment here? I will copy whatever you say verbatim to the next version.
Thanks,
Puranjay
PATCH v5 2/6 bpf: Inline bpf_iter_num_new() kfunc
In reply to Puranjay Mohanbpf_for() expands to the bpf_iter_num_{new,next,destroy}() kfuncs, which the verifier emits as regular calls. They are tiny and only touch the 8-byte on-stack iterator state, so open-code them in bpf_fixup_kfunc_call() like the other special kfuncs there.
Start with bpf_iter_num_new(): R1 points to the iterator, R2/R3 hold start/end. The inlined sequence mirrors the kfunc and returns the same -EINVAL / -E2BIG / 0.
start > end is rejected first, so end - start fits in a u32; range-check it as u32 on both sides ((u32)(end - start) in the kfunc). A movsx-based check would emit a cpuv4 instruction that some JITs (x86-32, mips32, sparc64) decode as a plain move and get wrong.
The emitted instructions are plain BPF, so the interpreter path stays correct and no jit_required marking is needed.
--- kernel/bpf/bpf_iter.c | 4 ++-- kernel/bpf/verifier.c | 24 ++++++++++++++++++++++++ 2 files changed, 26 insertions(+), 2 deletions(-)
kernel/bpf/bpf_iter.c +2 -2
@@ -782,8 +782,8 @@ __bpf_kfunc int bpf_iter_num_new(struct bpf_iter_num *it, int start, int end)return -EINVAL ; } -/* avoid overflows, e.g., if start == INT_MIN and end == INT_MAX */ -if ((s64 )end - (s64 )start >BPF_MAX_LOOPS ) { +/* start <= end here, so end - start fits in a u32 without overflow */ +if ((u32)(end - start) >BPF_MAX_LOOPS ) { s->cur = s->end =0 ;return -E2BIG ; }
kernel/bpf/verifier.c +24 -0
@@ -20006,6 +20006,30 @@ int bpf_fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, insn_buf[4 ] =BPF_ALU64_REG (BPF_SUB ,BPF_REG_0 ,BPF_REG_1 ); insn_buf[5 ] =BPF_ALU64_IMM (BPF_NEG ,BPF_REG_0 ,0 ); *cnt =6 ; + }else if (desc->func_id == special_kfunc_list[KF_bpf_iter_num_new]) { +/* inline bpf_iter_num_new(&it, start, end); R1=&it, R2=start, R3=end */ +int i =0 ; + +/* if (start > end) goto einval; */ + insn_buf[i++] =BPF_JMP32_REG (BPF_JSGT ,BPF_REG_2 ,BPF_REG_3 ,8 ); +/* r0 = (u32)end - (u32)start; if (r0 > BPF_MAX_LOOPS) goto e2big; */ + insn_buf[i++] =BPF_MOV32_REG (BPF_REG_0 ,BPF_REG_3 ); + insn_buf[i++] =BPF_ALU32_REG (BPF_SUB ,BPF_REG_0 ,BPF_REG_2 ); + insn_buf[i++] =BPF_JMP_IMM (BPF_JGT ,BPF_REG_0 ,BPF_MAX_LOOPS ,8 ); +/* s->cur = start - 1; s->end = end; return 0; */ + insn_buf[i++] =BPF_ALU32_IMM (BPF_ADD ,BPF_REG_2 ,-1 ); + insn_buf[i++] =BPF_STX_MEM (BPF_W ,BPF_REG_1 ,BPF_REG_2 ,0 ); + insn_buf[i++] =BPF_STX_MEM (BPF_W ,BPF_REG_1 ,BPF_REG_3 ,4 ); + insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 ,0 ); + insn_buf[i++] =BPF_JMP_A (5 ); +/* einval: s->cur = s->end = 0; return -EINVAL; */ + insn_buf[i++] =BPF_ST_MEM (BPF_DW ,BPF_REG_1 ,0 ,0 ); + insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 , -EINVAL ); + insn_buf[i++] =BPF_JMP_A (2 ); +/* e2big: s->cur = s->end = 0; return -E2BIG; */ + insn_buf[i++] =BPF_ST_MEM (BPF_DW ,BPF_REG_1 ,0 ,0 ); + insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 , -E2BIG ); + *cnt = i; }if (env->insn_aux_data[insn_idx].arg_prog) {
Signature · 1 line
2.53.0-Meta
PATCH v5 3/6 bpf: Inline bpf_iter_num_next() kfunc
In reply to Puranjay Mohanbpf_iter_num_next() runs on every bpf_for() iteration, so inlining it drops a call from the loop body. R1 points to the iterator; the returned pointer to s->cur is R1 itself, since s->cur is first.
s->cur and s->end are int, so the kfunc's s->cur + 1 >= s->end is a signed 32-bit compare and the inlined code needs no sign extension.
--- kernel/bpf/verifier.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
kernel/bpf/verifier.c +17 -0
@@ -20030,6 +20030,23 @@ int bpf_fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, insn_buf[i++] =BPF_ST_MEM (BPF_DW ,BPF_REG_1 ,0 ,0 ); insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 , -E2BIG ); *cnt = i; + }else if (desc->func_id == special_kfunc_list[KF_bpf_iter_num_next]) { +/* inline bpf_iter_num_next(&it); R1=&it, returns &s->cur or NULL */ +int i =0 ; + +/* r0 = s->cur + 1; if ((s32)r0 >= s->end) goto done; */ + insn_buf[i++] =BPF_LDX_MEM (BPF_W ,BPF_REG_0 ,BPF_REG_1 ,0 ); + insn_buf[i++] =BPF_ALU32_IMM (BPF_ADD ,BPF_REG_0 ,1 ); + insn_buf[i++] =BPF_LDX_MEM (BPF_W ,BPF_REG_2 ,BPF_REG_1 ,4 ); + insn_buf[i++] =BPF_JMP32_REG (BPF_JSGE ,BPF_REG_0 ,BPF_REG_2 ,3 ); +/* s->cur = r0; return &s->cur; */ + insn_buf[i++] =BPF_STX_MEM (BPF_W ,BPF_REG_1 ,BPF_REG_0 ,0 ); + insn_buf[i++] =BPF_MOV64_REG (BPF_REG_0 ,BPF_REG_1 ); + insn_buf[i++] =BPF_JMP_A (2 ); +/* done: s->cur = s->end = 0; return NULL; */ + insn_buf[i++] =BPF_ST_MEM (BPF_DW ,BPF_REG_1 ,0 ,0 ); + insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 ,0 ); + *cnt = i; }if (env->insn_aux_data[insn_idx].arg_prog) {
Signature · 1 line
2.53.0-Meta
PATCH v5 4/6 bpf: Inline bpf_iter_num_destroy() as a no-op
In reply to Puranjay MohanOnce destroy() returns the stack slot is no longer tracked as iterator state, so zeroing it is dead work. Make the kfunc a no-op and inline the call to a single BPF_JA 0 (the fixup can't drop the instruction outright, so emit a nop; the JITs elide it).
--- kernel/bpf/bpf_iter.c | 4 +--- kernel/bpf/verifier.c | 4 ++++ 2 files changed, 5 insertions(+), 3 deletions(-)
kernel/bpf/bpf_iter.c +1 -3
@@ -818,9 +818,7 @@ __bpf_kfunc int *bpf_iter_num_next(struct bpf_iter_num* it) __bpf_kfunc void bpf_iter_num_destroy(struct bpf_iter_num *it) { -struct bpf_iter_num_kern *s = (void *)it; - - s->cur = s->end =0 ; +/* no-op */ }__bpf_kfunc_end_defs ();
kernel/bpf/verifier.c +4 -0
@@ -20047,6 +20047,10 @@ int bpf_fixup_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, insn_buf[i++] =BPF_ST_MEM (BPF_DW ,BPF_REG_1 ,0 ,0 ); insn_buf[i++] =BPF_MOV64_IMM (BPF_REG_0 ,0 ); *cnt = i; + }else if (desc->func_id == special_kfunc_list[KF_bpf_iter_num_destroy]) { +/* bpf_iter_num_destroy() is a no-op; emit a nop to drop the call */ + insn_buf[0 ] =BPF_JMP_A (0 ); + *cnt =1 ; }if (env->insn_aux_data[insn_idx].arg_prog) {
Signature · 1 line
2.53.0-Meta
PATCH v5 5/6 selftests/bpf: Verify inlined numeric iterator shape with __xlated
In reply to Puranjay MohanAdd an __xlated test pinning the inlined bpf_iter_num_{new,next,destroy}() shapes. The program is __naked, so there is no compiler glue and the whole sequence is matched instruction for instruction.
Gate it to x86_64 and arm64 (bpf_jit_needs_zext() == false); elsewhere the verifier interleaves "wN = wN" zero-extensions that would not match. The inlining is arch independent, so these two are enough.
--- tools/testing/selftests/bpf/progs/iters.c | 83 +++++++++++++++++++++++ 1 file changed, 83 insertions(+)
tools/testing/selftests/bpf/progs/iters.c +83 -0
@@ -88,6 +88,89 @@ int iter_err_unsafe_asm_loop(const void *ctx)return 0 ; } +/* +* Naked function, so there is no compiler-generated glue and the whole inlined program can be +* matched. Pinned to arches whose JITs zero-extend 32-bit writes implicitly +* (bpf_jit_needs_zext() == false); on arches that need explicit zero-extension the verifier +* interleaves "wN = wN" insns and the fixed shape below would not match. The inlining itself is +* arch independent, so checking it on these arches is sufficient. +* +* bpf_iter_num_new() emits the full range check (distance computation and both the -EINVAL and +* -E2BIG error paths); bpf_iter_num_next() and bpf_iter_num_destroy() are inlined too. +*/ +SEC("raw_tp") +__arch_x86_64 +__arch_arm64 +__success +__xlated("r6 = r10") +__xlated("r6 += -8") +__xlated("call unknown") +__xlated("r3 = r0") +__xlated("r3 &= 65535") +__xlated("r1 = r6") +__xlated ("r2 = 0" ) +/* bpf_iter_num_new(&it, 0, <non-const>) with the range check kept */ +__xlated("if w2 s> w3 goto pc+8") +__xlated("w0 = w3") +__xlated("w0 -= w2") +__xlated("if r0 > 0x800000 goto pc+8") +__xlated("w2 += -1") +__xlated("*(u32 *)(r1 +0) = r2") +__xlated("*(u32 *)(r1 +4) = r3") +__xlated("r0 = 0") +__xlated("goto pc+5") +__xlated("*(u64 *)(r1 +0) = 0") +__xlated("r0 = -22") +__xlated("goto pc+2") +__xlated("*(u64 *)(r1 +0) = 0") +__xlated("r0 = -7") +__xlated ("r1 = r6" ) +/* bpf_iter_num_next(&it) */ +__xlated("r0 = *(u32 *)(r1 +0)") +__xlated("w0 += 1") +__xlated("r2 = *(u32 *)(r1 +4)") +__xlated("if w0 s>= w2 goto pc+3") +__xlated("*(u32 *)(r1 +0) = r0") +__xlated("r0 = r1") +__xlated("goto pc+2") +__xlated("*(u64 *)(r1 +0) = 0") +__xlated("r0 = 0") +__xlated("if r0 != 0x0 goto pc-11") +__xlated ("r1 = r6" ) +/* bpf_iter_num_destroy(&it) is inlined to a nop */ +__xlated("goto pc+0") +__xlated("r0 = 0") +__xlated("exit") +int __naked iter_num_new_inlined (void ) +{ + asmvolatile ( +/* r6 points to struct bpf_iter_num on the stack */ +"r6 = r10;" +"r6 += -8;" +/* non-constant end so the range checks are kept */ +"call %[bpf_get_prandom_u32];" +"r3 = r0;" +"r3 &= 0xffff;" +"r1 = r6;" +"r2 = 0;" +"call %[bpf_iter_num_new];" +"1:" +"r1 = r6;" +"call %[bpf_iter_num_next];" +"if r0 != 0 goto 1b;" +"r1 = r6;" +"call %[bpf_iter_num_destroy];" +"r0 = 0;" +"exit;" + : + : __imm(bpf_get_prandom_u32), +__imm (bpf_iter_num_new), +__imm (bpf_iter_num_next), +__imm (bpf_iter_num_destroy) + : __clobber_common, "r6" + ); +} + SEC("raw_tp") __successint iter_while_loop (const void *ctx)
Signature · 1 line
2.53.0-Meta
PATCH v5 6/6 selftests/bpf: Add bpf_for() benchmark
In reply to Puranjay MohanAdd a bpf_for() benchmark modelled on bench_bpf_loop so the per-iteration iterator cost can be measured and compared against bpf_loop. It runs an empty bpf_for(i, 0, nr_loops) loop 1000 times per trigger and accounts nr_loops hits per outer iteration:
$ ./bench -p 1 --nr_loops 1000 bpf-for
--- tools/testing/selftests/bpf/Makefile | 2 + tools/testing/selftests/bpf/bench.c | 4 + .../selftests/bpf/benchs/bench_bpf_for.c | 104 ++++++++++++++++++ .../selftests/bpf/benchs/run_bench_bpf_for.sh | 15 +++ .../selftests/bpf/progs/bpf_for_bench.c | 32 ++++++ 5 files changed, 157 insertions(+) create mode 100644 tools/testing/selftests/bpf/benchs/bench_bpf_for.c create mode 100755 tools/testing/selftests/bpf/benchs/run_bench_bpf_for.sh create mode 100644 tools/testing/selftests/bpf/progs/bpf_for_bench.c
tools/testing/selftests/bpf/Makefile +2 -0
@@ -965,6 +965,7 @@ $(OUTPUT)/bench_ringbufs.o: $(OUTPUT)/ringbuf_bench.skel.h \ $(OUTPUT)/perfbuf_bench.skel.h $(OUTPUT)/bench_bloom_filter_map.o: $(OUTPUT)/bloom_filter_bench.skel.h $(OUTPUT)/bench_bpf_loop.o: $(OUTPUT)/bpf_loop_bench.skel.h +$(OUTPUT)/bench_bpf_for.o: $(OUTPUT)/bpf_for_bench.skel.h $(OUTPUT)/bench_strncmp.o: $(OUTPUT)/strncmp_bench.skel.h $(OUTPUT)/bench_bpf_hashmap_full_update.o: $(OUTPUT)/bpf_hashmap_full_update_bench.skel.h $(OUTPUT)/bench_local_storage.o: $(OUTPUT)/local_storage_bench.skel.h
@@ -990,6 +991,7 @@ $(OUTPUT)/bench: $(OUTPUT)/bench.o \ $(OUTPUT)/bench_ringbufs.o \ $(OUTPUT)/bench_bloom_filter_map.o \ $(OUTPUT)/bench_bpf_loop.o \ + $(OUTPUT)/bench_bpf_for.o \ $(OUTPUT)/bench_strncmp.o \ $(OUTPUT)/bench_bpf_hashmap_full_update.o \ $(OUTPUT)/bench_local_storage.o \
tools/testing/selftests/bpf/bench.c +4 -0
@@ -276,6 +276,7 @@ static const struct argp_option opts[] = {extern struct argp bench_ringbufs_argp;extern struct argp bench_bloom_map_argp;extern struct argp bench_bpf_loop_argp; +extern struct argp bench_bpf_for_argp;extern struct argp bench_local_storage_argp;extern struct argp bench_local_storage_rcu_tasks_trace_argp;extern struct argp bench_strncmp_argp;
@@ -292,6 +293,7 @@ static const struct argp_child bench_parsers[] = { { &bench_ringbufs_argp, 0, "Ring buffers benchmark", 0 }, { &bench_bloom_map_argp, 0, "Bloom filter map benchmark", 0 }, { &bench_bpf_loop_argp, 0, "bpf_loop helper benchmark", 0 }, + { &bench_bpf_for_argp, 0, "bpf_for loop benchmark", 0 }, { &bench_local_storage_argp, 0, "local_storage benchmark", 0 }, { &bench_strncmp_argp, 0, "bpf_strncmp helper benchmark", 0 }, { &bench_local_storage_rcu_tasks_trace_argp, 0,
@@ -557,6 +559,7 @@ extern const struct bench bench_bloom_false_positive;extern const struct bench bench_hashmap_without_bloom;extern const struct bench bench_hashmap_with_bloom;extern const struct bench bench_bpf_loop; +extern const struct bench bench_bpf_for;extern const struct bench bench_strncmp_no_helper;extern const struct bench bench_strncmp_helper;extern const struct bench bench_bpf_hashmap_full_update;
@@ -640,6 +643,7 @@ static const struct bench *benchs[] = { &bench_hashmap_without_bloom, &bench_hashmap_with_bloom, &bench_bpf_loop, + &bench_bpf_for, &bench_strncmp_no_helper, &bench_strncmp_helper, &bench_bpf_hashmap_full_update,
tools/testing/selftests/bpf/benchs/bench_bpf_for.c new file +104 -0
104 lines not on this page · raw message
tools/testing/selftests/bpf/benchs/run_bench_bpf_for.sh new file +15 -0
@@ -0,0 +1,15 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 + +source ./benchs/run_common.sh + +set -eufo pipefail + +for t in 1 4 8 12 16; do +for i in 10 100 500 1000 5000 10000 50000 100000 500000 1000000; do +subtitle "nr_loops: $i, nr_threads: $t" + summarize_ops "bpf_for: " \ + "$($RUN_BENCH -p $t --nr_loops $i bpf-for)" + printf "\n" +done +done
tools/testing/selftests/bpf/progs/bpf_for_bench.c new file +32 -0
@@ -0,0 +1,32 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2026 Meta Platforms, Inc. and affiliates. */ + +#include "vmlinux.h" +#include <bpf/bpf_helpers.h> +#include "bpf_misc.h" + +char _license[] SEC("license") = "GPL"; + +int nr_loops; +long hits; + +static int outer_loop (__u32 index,void *data) +{ +int i; + +/* +* Empty body: the work being measured is the open-coded numeric iterator itself +* (bpf_iter_num_new/next/destroy behind bpf_for()). +*/ +bpf_for (i,0 , nr_loops) + ; +__sync_add_and_fetch (&hits, nr_loops); +return 0 ; +} + +SEC("fentry/" SYS_PREFIX "sys_getpgid") +int benchmark (void *ctx) +{ +bpf_loop (1000 , outer_loop,NULL ,0 ); +return 0 ; +}
Signature · 1 line
2.53.0-Meta
This series was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:
Hello:
This series was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:
Quoted 8 lines · On Tue, 4 Aug 2026 06:45:52 -0700 you wrote:
The
bpf_for(i, start, end)macro is BPF's open-coded numeric iterator. It expands into calls to three kfuncs:bpf_iter_num_new()to set the iterator up,bpf_iter_num_next()once per iteration, andbpf_iter_num_destroy()to tear it down. The verifier emits these as ordinary kfunc calls, so abpf_for()loop pays function-call overhead on setup, teardown, and -- most importantly -- on every single iteration viabpf_iter_num_next().[...]
Here is the summary with links:
- [bpf-next,v5,1/6] bpf: Correct the overflow check comment in bpf_iter_num_next()
https://git.kernel.org/bpf/bpf-next/c/8efd87051c3a - [bpf-next,v5,2/6] bpf: Inline bpf_iter_num_new() kfunc https://git.kernel.org/bpf/bpf-next/c/f8f2b567d560 - [bpf-next,v5,3/6] bpf: Inline bpf_iter_num_next() kfunc https://git.kernel.org/bpf/bpf-next/c/e93347704878 - [bpf-next,v5,4/6] bpf: Inline bpf_iter_num_destroy() as a no-op https://git.kernel.org/bpf/bpf-next/c/39f047682fe3 - [bpf-next,v5,5/6] selftests/bpf: Verify inlined numeric iterator shape with __xlated https://git.kernel.org/bpf/bpf-next/c/b829bc167705 - [bpf-next,v5,6/6] selftests/bpf: Add bpf_for() benchmark https://git.kernel.org/bpf/bpf-next/c/5a1de41147b0
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html