Convert liveness bitmask type from u64 to spis_t, doubling the number of trackable stack slots from 64 to 128 to support 4-byte granularity. Each 8-byte SPI now maps to two consecutive 4-byte sub-slots in the bitmask: spi*2 half and spi*2+1 half. In verifier.c, check_stack_write_fixed_off() now reports 4-byte aligned writes of 4-byte writes as half-slot marks and 8-byte aligned 8-byte writes as two slots. Similar logic applied in check_stack_read_fixed_off(). Queries (is_live_before) are not yet migrated to half-slot granularity. Signed-off-by: Eduard Zingerman --- include/linux/bpf_verifier.h | 4 +- kernel/bpf/liveness.c | 122 ++++++++++++++++++++++++++++--------------- kernel/bpf/verifier.c | 55 +++++++++++-------- 3 files changed, 115 insertions(+), 66 deletions(-) diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index f34e7a074a20..7b31d8024c61 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -1217,8 +1217,8 @@ s64 bpf_kfunc_stack_access_bytes(struct bpf_verifier_env *env, int bpf_stack_liveness_init(struct bpf_verifier_env *env); void bpf_stack_liveness_free(struct bpf_verifier_env *env); int bpf_update_live_stack(struct bpf_verifier_env *env); -int bpf_mark_stack_read(struct bpf_verifier_env *env, u32 frameno, u32 insn_idx, u64 mask); -void bpf_mark_stack_write(struct bpf_verifier_env *env, u32 frameno, u64 mask); +int bpf_mark_stack_read(struct bpf_verifier_env *env, u32 frameno, u32 insn_idx, spis_t mask); +void bpf_mark_stack_write(struct bpf_verifier_env *env, u32 frameno, spis_t mask); int bpf_reset_stack_write_marks(struct bpf_verifier_env *env, u32 insn_idx); int bpf_commit_stack_write_marks(struct bpf_verifier_env *env); int bpf_live_stack_query_init(struct bpf_verifier_env *env, struct bpf_verifier_state *st); diff --git a/kernel/bpf/liveness.c b/kernel/bpf/liveness.c index 998986853c61..9e36ea5f2eec 100644 --- a/kernel/bpf/liveness.c +++ b/kernel/bpf/liveness.c @@ -93,10 +93,10 @@ struct callchain { }; struct per_frame_masks { - u64 may_read; /* stack slots that may be read by this instruction */ - u64 must_write; /* stack slots written by this instruction */ - u64 must_write_acc; /* stack slots written by this instruction and its successors */ - u64 live_before; /* stack slots that may be read by this insn and its successors */ + spis_t may_read; /* stack slots that may be read by this instruction */ + spis_t must_write; /* stack slots written by this instruction */ + spis_t must_write_acc; /* stack slots written by this instruction and its successors */ + spis_t live_before; /* stack slots that may be read by this insn and its successors */ }; /* @@ -131,7 +131,7 @@ struct bpf_liveness { * Below fields are used to accumulate stack write marks for instruction at * @write_insn_idx before submitting the marks to @cur_instance. */ - u64 write_masks_acc[MAX_CALL_FRAMES]; + spis_t write_masks_acc[MAX_CALL_FRAMES]; u32 write_insn_idx; }; @@ -299,23 +299,24 @@ static int ensure_cur_instance(struct bpf_verifier_env *env) /* Accumulate may_read masks for @frame at @insn_idx */ static int mark_stack_read(struct bpf_verifier_env *env, - struct func_instance *instance, u32 frame, u32 insn_idx, u64 mask) + struct func_instance *instance, u32 frame, u32 insn_idx, spis_t mask) { struct per_frame_masks *masks; - u64 new_may_read; + spis_t new_may_read; masks = alloc_frame_masks(env, instance, frame, insn_idx); if (IS_ERR(masks)) return PTR_ERR(masks); - new_may_read = masks->may_read | mask; - if (new_may_read != masks->may_read && - ((new_may_read | masks->live_before) != masks->live_before)) + new_may_read = spis_or(masks->may_read, mask); + if (!spis_equal(new_may_read, masks->may_read) && + !spis_equal(spis_or(new_may_read, masks->live_before), + masks->live_before)) instance->updated = true; - masks->may_read |= mask; + masks->may_read = spis_or(masks->may_read, mask); return 0; } -int bpf_mark_stack_read(struct bpf_verifier_env *env, u32 frame, u32 insn_idx, u64 mask) +int bpf_mark_stack_read(struct bpf_verifier_env *env, u32 frame, u32 insn_idx, spis_t mask) { int err; @@ -332,7 +333,7 @@ static void reset_stack_write_marks(struct bpf_verifier_env *env, liveness->write_insn_idx = insn_idx; for (i = 0; i <= instance->callchain.curframe; i++) - liveness->write_masks_acc[i] = 0; + liveness->write_masks_acc[i] = SPIS_ZERO; } int bpf_reset_stack_write_marks(struct bpf_verifier_env *env, u32 insn_idx) @@ -348,18 +349,18 @@ int bpf_reset_stack_write_marks(struct bpf_verifier_env *env, u32 insn_idx) return 0; } -void bpf_mark_stack_write(struct bpf_verifier_env *env, u32 frame, u64 mask) +void bpf_mark_stack_write(struct bpf_verifier_env *env, u32 frame, spis_t mask) { - env->liveness->write_masks_acc[frame] |= mask; + env->liveness->write_masks_acc[frame] = spis_or(env->liveness->write_masks_acc[frame], mask); } static int commit_stack_write_marks(struct bpf_verifier_env *env, struct func_instance *instance) { struct bpf_liveness *liveness = env->liveness; - u32 idx, frame, curframe, old_must_write; + u32 idx, frame, curframe; struct per_frame_masks *masks; - u64 mask; + spis_t mask, old_must_write, dropped; if (!instance) return 0; @@ -369,7 +370,7 @@ static int commit_stack_write_marks(struct bpf_verifier_env *env, for (frame = 0; frame <= curframe; frame++) { mask = liveness->write_masks_acc[frame]; /* avoid allocating frames for zero masks */ - if (mask == 0 && !instance->must_write_set[idx]) + if (spis_is_zero(mask) && !instance->must_write_set[idx]) continue; masks = alloc_frame_masks(env, instance, frame, liveness->write_insn_idx); if (IS_ERR(masks)) @@ -380,12 +381,14 @@ static int commit_stack_write_marks(struct bpf_verifier_env *env, * to @mask. Otherwise take intersection with the previous value. */ if (instance->must_write_set[idx]) - mask &= old_must_write; - if (old_must_write != mask) { + mask = spis_and(mask, old_must_write); + if (!spis_equal(old_must_write, mask)) { masks->must_write = mask; instance->updated = true; } - if (old_must_write & ~mask) + /* dropped = old_must_write & ~mask */ + dropped = spis_and(old_must_write, spis_not(mask)); + if (!spis_is_zero(dropped)) instance->must_write_dropped = true; } instance->must_write_set[idx] = true; @@ -415,22 +418,52 @@ static char *fmt_callchain(struct bpf_verifier_env *env, struct callchain *callc return env->tmp_str_buf; } +/* + * When both halves of an 8-byte SPI are set, print as "-8","-16",... + * When only one half is set, print as "-4h","-8h",... + */ +static void bpf_fmt_spis_mask(char *buf, ssize_t buf_sz, spis_t spis) +{ + bool first = true; + int spi, n; + + buf[0] = '\0'; + + for (spi = 0; spi < STACK_SLOTS / 2 && buf_sz > 0; spi++) { + bool lo = spis_test_bit(spis, spi * 2); + bool hi = spis_test_bit(spis, spi * 2 + 1); + + if (!lo && !hi) + continue; + n = snprintf(buf, buf_sz, "%s%d%s", + first ? "" : ",", + -(spi + 1) * BPF_REG_SIZE + (lo && !hi ? BPF_HALF_REG_SIZE : 0), + lo && hi ? "" : "h"); + first = false; + buf += n; + buf_sz -= n; + } +} + static void log_mask_change(struct bpf_verifier_env *env, struct callchain *callchain, - char *pfx, u32 frame, u32 insn_idx, u64 old, u64 new) + char *pfx, u32 frame, u32 insn_idx, + spis_t old, spis_t new) { - u64 changed_bits = old ^ new; - u64 new_ones = new & changed_bits; - u64 new_zeros = ~new & changed_bits; + spis_t changed_bits, new_ones, new_zeros; + + changed_bits = spis_xor(old, new); + new_ones = spis_and(new, changed_bits); + new_zeros = spis_and(spis_not(new), changed_bits); - if (!changed_bits) + if (spis_is_zero(changed_bits)) return; bpf_log(&env->log, "%s frame %d insn %d ", fmt_callchain(env, callchain), frame, insn_idx); - if (new_ones) { - bpf_fmt_stack_mask(env->tmp_str_buf, sizeof(env->tmp_str_buf), new_ones); + if (!spis_is_zero(new_ones)) { + bpf_fmt_spis_mask(env->tmp_str_buf, sizeof(env->tmp_str_buf), new_ones); bpf_log(&env->log, "+%s %s ", pfx, env->tmp_str_buf); } - if (new_zeros) { - bpf_fmt_stack_mask(env->tmp_str_buf, sizeof(env->tmp_str_buf), new_zeros); + if (!spis_is_zero(new_zeros)) { + bpf_fmt_spis_mask(env->tmp_str_buf, sizeof(env->tmp_str_buf), new_zeros); bpf_log(&env->log, "-%s %s", pfx, env->tmp_str_buf); } bpf_log(&env->log, "\n"); @@ -562,7 +595,7 @@ static inline bool update_insn(struct bpf_verifier_env *env, struct func_instance *instance, u32 frame, u32 insn_idx) { struct bpf_insn_aux_data *aux = env->insn_aux_data; - u64 new_before, new_after, must_write_acc; + spis_t new_before, new_after, must_write_acc; struct per_frame_masks *insn, *succ_insn; struct bpf_iarray *succ; u32 s; @@ -574,28 +607,30 @@ static inline bool update_insn(struct bpf_verifier_env *env, changed = false; insn = get_frame_masks(instance, frame, insn_idx); - new_before = 0; - new_after = 0; + new_before = SPIS_ZERO; + new_after = SPIS_ZERO; /* * New "must_write_acc" is an intersection of all "must_write_acc" * of successors plus all "must_write" slots of instruction itself. */ - must_write_acc = U64_MAX; + must_write_acc = SPIS_ALL; for (s = 0; s < succ->cnt; ++s) { succ_insn = get_frame_masks(instance, frame, succ->items[s]); - new_after |= succ_insn->live_before; - must_write_acc &= succ_insn->must_write_acc; + new_after = spis_or(new_after, succ_insn->live_before); + must_write_acc = spis_and(must_write_acc, succ_insn->must_write_acc); } - must_write_acc |= insn->must_write; + must_write_acc = spis_or(must_write_acc, insn->must_write); /* * New "live_before" is a union of all "live_before" of successors * minus slots written by instruction plus slots read by instruction. + * new_before = (new_after & ~insn->must_write) | insn->may_read */ - new_before = (new_after & ~insn->must_write) | insn->may_read; - changed |= new_before != insn->live_before; - changed |= must_write_acc != insn->must_write_acc; + new_before = spis_or(spis_and(new_after, spis_not(insn->must_write)), + insn->may_read); + changed |= !spis_equal(new_before, insn->live_before); + changed |= !spis_equal(must_write_acc, insn->must_write_acc); if (unlikely(env->log.level & BPF_LOG_LEVEL2) && - (insn->may_read || insn->must_write || + (!spis_is_zero(insn->may_read) || !spis_is_zero(insn->must_write) || insn_idx == callchain_subprog_start(&instance->callchain) || aux[insn_idx].prune_point)) { log_mask_change(env, &instance->callchain, "live", @@ -631,7 +666,7 @@ static int update_instance(struct bpf_verifier_env *env, struct func_instance *i for (i = 0; i < instance->insn_cnt; i++) { insn = get_frame_masks(instance, frame, this_subprog_start + i); - insn->must_write_acc = 0; + insn->must_write_acc = SPIS_ZERO; } } } @@ -702,7 +737,8 @@ static bool is_live_before(struct func_instance *instance, u32 insn_idx, u32 fra struct per_frame_masks *masks; masks = get_frame_masks(instance, frameno, insn_idx); - return masks && (masks->live_before & BIT(spi)); + return masks && (spis_test_bit(masks->live_before, spi * 2) || + spis_test_bit(masks->live_before, spi * 2 + 1)); } int bpf_live_stack_query_init(struct bpf_verifier_env *env, struct bpf_verifier_state *st) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index cd2a56308e59..ee3945a919e3 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -828,7 +828,8 @@ static int mark_stack_slots_dynptr(struct bpf_verifier_env *env, struct bpf_reg_ state->stack[spi - 1].spilled_ptr.ref_obj_id = id; } - bpf_mark_stack_write(env, state->frameno, BIT(spi - 1) | BIT(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi - 1)); return 0; } @@ -845,7 +846,8 @@ static void invalidate_dynptr(struct bpf_verifier_env *env, struct bpf_func_stat __mark_reg_not_init(env, &state->stack[spi].spilled_ptr); __mark_reg_not_init(env, &state->stack[spi - 1].spilled_ptr); - bpf_mark_stack_write(env, state->frameno, BIT(spi - 1) | BIT(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi - 1)); } static int unmark_stack_slots_dynptr(struct bpf_verifier_env *env, struct bpf_reg_state *reg) @@ -963,7 +965,8 @@ static int destroy_if_dynptr_stack_slot(struct bpf_verifier_env *env, __mark_reg_not_init(env, &state->stack[spi].spilled_ptr); __mark_reg_not_init(env, &state->stack[spi - 1].spilled_ptr); - bpf_mark_stack_write(env, state->frameno, BIT(spi - 1) | BIT(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi - 1)); return 0; } @@ -1090,7 +1093,7 @@ static int mark_stack_slots_iter(struct bpf_verifier_env *env, for (j = 0; j < BPF_REG_SIZE; j++) slot->slot_type[j] = STACK_ITER; - bpf_mark_stack_write(env, state->frameno, BIT(spi - i)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi - i)); mark_stack_slot_scratched(env, spi - i); } @@ -1119,7 +1122,7 @@ static int unmark_stack_slots_iter(struct bpf_verifier_env *env, for (j = 0; j < BPF_REG_SIZE; j++) slot->slot_type[j] = STACK_INVALID; - bpf_mark_stack_write(env, state->frameno, BIT(spi - i)); + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi - i)); mark_stack_slot_scratched(env, spi - i); } @@ -1209,7 +1212,7 @@ static int mark_stack_slot_irq_flag(struct bpf_verifier_env *env, slot = &state->stack[spi]; st = &slot->spilled_ptr; - bpf_mark_stack_write(env, reg->frameno, BIT(spi)); + bpf_mark_stack_write(env, reg->frameno, spis_single_slot(spi)); __mark_reg_known_zero(st); st->type = PTR_TO_STACK; /* we don't have dedicated reg type */ st->ref_obj_id = id; @@ -1265,7 +1268,7 @@ static int unmark_stack_slot_irq_flag(struct bpf_verifier_env *env, struct bpf_r __mark_reg_not_init(env, st); - bpf_mark_stack_write(env, reg->frameno, BIT(spi)); + bpf_mark_stack_write(env, reg->frameno, spis_single_slot(spi)); for (i = 0; i < BPF_REG_SIZE; i++) slot->slot_type[i] = STACK_INVALID; @@ -3846,7 +3849,8 @@ static int mark_stack_slot_obj_read(struct bpf_verifier_env *env, struct bpf_reg int err, i; for (i = 0; i < nr_slots; i++) { - err = bpf_mark_stack_read(env, reg->frameno, env->insn_idx, BIT(spi - i)); + err = bpf_mark_stack_read(env, reg->frameno, env->insn_idx, + spis_single_slot(spi - i)); if (err) return err; mark_stack_slot_scratched(env, spi - i); @@ -5398,17 +5402,15 @@ static int check_stack_write_fixed_off(struct bpf_verifier_env *env, if (err) return err; - if (!(off % BPF_REG_SIZE) && size == BPF_REG_SIZE) { - /* only mark the slot as written if all 8 bytes were written - * otherwise read propagation may incorrectly stop too soon - * when stack slots are partially written. - * This heuristic means that read propagation will be - * conservative, since it will add reg_live_read marks - * to stack slots all the way to first state when programs - * writes+reads less than 8 bytes - */ - bpf_mark_stack_write(env, state->frameno, BIT(spi)); - } + if (!(off % BPF_REG_SIZE) && size == BPF_REG_SIZE) + /* 8-byte aligned, 8-byte write */ + bpf_mark_stack_write(env, state->frameno, spis_single_slot(spi)); + else if (!(off % BPF_REG_SIZE) && size == BPF_HALF_REG_SIZE) + /* 8-byte aligned, 4-byte write */ + bpf_mark_stack_write(env, state->frameno, spis_one_bit(spi * 2 + 1)); + else if (!(off % BPF_HALF_REG_SIZE) && size == BPF_HALF_REG_SIZE) + /* 4-byte aligned, 4-byte write */ + bpf_mark_stack_write(env, state->frameno, spis_one_bit(spi * 2)); check_fastcall_stack_contract(env, state, insn_idx, off); mark_stack_slot_scratched(env, spi); @@ -5666,6 +5668,7 @@ static int check_stack_read_fixed_off(struct bpf_verifier_env *env, struct bpf_reg_state *reg; u8 *stype, type; int insn_flags = insn_stack_access_flags(reg_state->frameno, spi); + spis_t mask; int err; stype = reg_state->stack[spi].slot_type; @@ -5673,7 +5676,16 @@ static int check_stack_read_fixed_off(struct bpf_verifier_env *env, mark_stack_slot_scratched(env, spi); check_fastcall_stack_contract(env, state, env->insn_idx, off); - err = bpf_mark_stack_read(env, reg_state->frameno, env->insn_idx, BIT(spi)); + if (!(off % BPF_REG_SIZE) && size == BPF_HALF_REG_SIZE) + /* 8-byte aligned, 4-byte read */ + mask = spis_one_bit(spi * 2 + 1); + else if (!(off % BPF_HALF_REG_SIZE) && size == BPF_HALF_REG_SIZE) + /* 4-byte aligned, 4-byte read */ + mask = spis_one_bit(spi * 2); + else + mask = spis_single_slot(spi); + + err = bpf_mark_stack_read(env, reg_state->frameno, env->insn_idx, mask); if (err) return err; @@ -8470,7 +8482,8 @@ static int check_stack_range_initialized( /* reading any byte out of 8-byte 'spill_slot' will cause * the whole slot to be marked as 'read' */ - err = bpf_mark_stack_read(env, reg->frameno, env->insn_idx, BIT(spi)); + err = bpf_mark_stack_read(env, reg->frameno, env->insn_idx, + spis_single_slot(spi)); if (err) return err; /* We do not call bpf_mark_stack_write(), as we can not -- 2.53.0