On 11/23/22 5:40 PM, Alexei Starovoitov wrote:
On Tue, Nov 22, 2022 at 8:54 PM Yonghong Song <yhs@xxxxxx> wrote:Add two kfunc's bpf_rcu_read_lock() and bpf_rcu_read_unlock(). These two kfunc's can be used for all program types. The following is an example about how rcu pointer are used w.r.t. bpf_rcu_read_lock()/bpf_rcu_read_unlock(). struct task_struct { ... struct task_struct *last_wakee; struct task_struct __rcu *real_parent; ... }; Let us say prog does 'task = bpf_get_current_task_btf()' to get a 'task' pointer. The basic rules are: - 'real_parent = task->real_parent' should be inside bpf_rcu_read_lock region. this is to simulate rcu_dereference() operation. The 'real_parent' is marked as MEM_RCU only if (1). task->real_parent is inside bpf_rcu_read_lock region, and (2). task is a trusted ptr. So MEM_RCU marked ptr can be 'trusted' inside the bpf_rcu_read_lock region. - 'last_wakee = real_parent->last_wakee' should be inside bpf_rcu_read_lock region since it tries to access rcu protected memory. - the ptr 'last_wakee' will be marked as PTR_UNTRUSTED since in general it is not clear whether the object pointed by 'last_wakee' is valid or not even inside bpf_rcu_read_lock region. To prevent rcu pointer leaks outside the rcu read lock region. The verifier will clear all rcu pointer register state to unknown, i.e., scalar_value, at bpf_rcu_read_unlock() kfunc call site, so later dereference becomes impossible. The current implementation does not support nested rcu read lock region in the prog. Signed-off-by: Yonghong Song <yhs@xxxxxx> --- include/linux/bpf.h | 3 + include/linux/bpf_verifier.h | 4 +- kernel/bpf/btf.c | 3 + kernel/bpf/helpers.c | 12 +++ kernel/bpf/verifier.c | 155 ++++++++++++++++++++++++++++------- 5 files changed, 147 insertions(+), 30 deletions(-) diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 43fd7eeeeabb..c6aa6912ea16 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -572,6 +572,9 @@ enum bpf_type_flag { */ PTR_TRUSTED = BIT(12 + BPF_BASE_TYPE_BITS), + /* MEM is tagged with rcu and memory access needs rcu_read_lock protection. */ + MEM_RCU = BIT(13 + BPF_BASE_TYPE_BITS), + __BPF_TYPE_FLAG_MAX, __BPF_TYPE_LAST_FLAG = __BPF_TYPE_FLAG_MAX - 1, }; diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index 545152ac136c..1f3ce54e50ed 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -344,6 +344,7 @@ struct bpf_verifier_state { u32 id; } active_lock; bool speculative; + bool active_rcu_lock; /* first and last insn idx of this verifier state */ u32 first_insn_idx; @@ -445,6 +446,7 @@ struct bpf_insn_aux_data { u32 seen; /* this insn was processed by the verifier at env->pass_cnt */ bool sanitize_stack_spill; /* subject to Spectre v4 sanitation */ bool zext_dst; /* this insn zero extends dst reg */ + bool storage_get_func_atomic; /* bpf_*_storage_get() with atomic memory alloc */ u8 alu_state; /* used in combination with alu_limit */ /* below fields are initialized once */ @@ -680,7 +682,7 @@ static inline bool bpf_prog_check_recur(const struct bpf_prog *prog) } } -#define BPF_REG_TRUSTED_MODIFIERS (MEM_ALLOC | PTR_TRUSTED) +#define BPF_REG_TRUSTED_MODIFIERS (MEM_ALLOC | MEM_RCU | PTR_TRUSTED) static inline bool bpf_type_has_unsafe_modifiers(u32 type) { diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 1a59cc7ad730..68df0df27302 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -6237,6 +6237,9 @@ static int btf_struct_walk(struct bpf_verifier_log *log, const struct btf *btf, /* check __percpu tag */ if (strcmp(tag_value, "percpu") == 0) tmp_flag = MEM_PERCPU; + /* check __rcu tag */ + if (strcmp(tag_value, "rcu") == 0) + tmp_flag = MEM_RCU; } stype = btf_type_skip_modifiers(btf, mtype->type, &id); diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index ae565b495f3d..eda9824694bf 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -1971,6 +1971,16 @@ void *bpf_rdonly_cast(void *obj__ign, u32 btf_id__k) return obj__ign; } +void bpf_rcu_read_lock(void) +{ + rcu_read_lock(); +} + +void bpf_rcu_read_unlock(void) +{ + rcu_read_unlock(); +} +I think the check from selftest: /* rcu_tag_btf_id < 0 implies rcu tag support not available in vmlinux btf */ rcu_tag_btf_id = btf__find_by_name_kind(vmlinux_btf, "rcu", BTF_KIND_TYPE_TAG); should be done by the kernel as well. And if the kernel is not compiled with clang the verifier should probably disallow bpf_rcu_read_lock/unlock kfuncs. Otherwise the same bpf prog will work differently depending on whether the kernel was compiled with gcc or clang.
My original idea is that the same program should work (not causingverification error) across with/without rcu tag, even for clang itself as clang <= 13 won't have rcu tag support. But you are right,
having rcu tag or not in vmlinux btf actually has different verificaiton path and potentially cause runtime difference as well due to missing/having rcu semantics. So agree that let us have verification failure if rcu tag is not in vmlinux btf and bpf_rcu_read_lock/unlock is used in the program.
__diag_pop(); BTF_SET8_START(generic_btf_ids) @@ -2012,6 +2022,8 @@ BTF_ID(func, bpf_cgroup_release) BTF_SET8_START(common_btf_ids) BTF_ID_FLAGS(func, bpf_cast_to_kern_ctx) BTF_ID_FLAGS(func, bpf_rdonly_cast) +BTF_ID_FLAGS(func, bpf_rcu_read_lock) +BTF_ID_FLAGS(func, bpf_rcu_read_unlock) BTF_SET8_END(common_btf_ids) static const struct btf_kfunc_id_set common_kfunc_set = { diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 068cc885903c..f76c341fea82 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -527,6 +527,14 @@ static bool is_callback_calling_function(enum bpf_func_id func_id) func_id == BPF_FUNC_user_ringbuf_drain; } +static bool is_storage_get_function(enum bpf_func_id func_id) +{ + return func_id == BPF_FUNC_sk_storage_get || + func_id == BPF_FUNC_inode_storage_get || + func_id == BPF_FUNC_task_storage_get || + func_id == BPF_FUNC_cgrp_storage_get; +} + static bool helper_multiple_ref_obj_use(enum bpf_func_id func_id, const struct bpf_map *map) { @@ -589,11 +597,12 @@ static const char *reg_type_str(struct bpf_verifier_env *env, strncpy(postfix, "_or_null", 16); } - snprintf(prefix, sizeof(prefix), "%s%s%s%s%s%s", + snprintf(prefix, sizeof(prefix), "%s%s%s%s%s%s%s", type & MEM_RDONLY ? "rdonly_" : "", type & MEM_RINGBUF ? "ringbuf_" : "", type & MEM_USER ? "user_" : "", type & MEM_PERCPU ? "percpu_" : "", + type & MEM_RCU ? "rcu_" : "", type & PTR_UNTRUSTED ? "untrusted_" : "", type & PTR_TRUSTED ? "trusted_" : "" ); @@ -1220,6 +1229,7 @@ static int copy_verifier_state(struct bpf_verifier_state *dst_state, dst_state->frame[i] = NULL; } dst_state->speculative = src->speculative; + dst_state->active_rcu_lock = src->active_rcu_lock; dst_state->curframe = src->curframe; dst_state->active_lock.ptr = src->active_lock.ptr; dst_state->active_lock.id = src->active_lock.id; @@ -4258,6 +4268,25 @@ static bool is_flow_key_reg(struct bpf_verifier_env *env, int regno) return reg->type == PTR_TO_FLOW_KEYS; } +static bool is_trusted_reg(const struct bpf_reg_state *reg) +{ + /* A referenced register is always trusted. */ + if (reg->ref_obj_id) + return true; + + /* If a register is not referenced, it is trusted if it has either the + * MEM_ALLOC or PTR_TRUSTED type modifiers, and no others. Some of theThe comment needs to be adjusted.
ack.
+ * other type modifiers may be safe, but we elect to take an opt-in + * approach here as some (e.g. PTR_UNTRUSTED and PTR_MAYBE_NULL) are + * not. + * + * Eventually, we should make PTR_TRUSTED the single source of truth + * for whether a register is trusted. + */ + return type_flag(reg->type) & BPF_REG_TRUSTED_MODIFIERS && + !bpf_type_has_unsafe_modifiers(reg->type); +} + static int check_pkt_ptr_alignment(struct bpf_verifier_env *env, const struct bpf_reg_state *reg, int off, int size, bool strict) @@ -4737,9 +4766,29 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env, if (type_flag(reg->type) & PTR_UNTRUSTED) flag |= PTR_UNTRUSTED; - /* Any pointer obtained from walking a trusted pointer is no longer trusted. */ + /* By default any pointer obtained from walking a trusted pointer is + * no longer trusted except the rcu case below. + */ flag &= ~PTR_TRUSTED; + if (flag & MEM_RCU) { + /* Mark value register as MEM_RCU only if it is protected by + * bpf_rcu_read_lock() and the ptr reg is trusted. MEM_RCU + * itself can already indicate trustedness inside the rcu + * read lock region. But Mark it as PTR_TRUSTED as well + * similar to MEM_ALLOC.'similar to MEM_ALLOC' part is not true yet. Let's not get ahead of ourselves :)
ack.
+ */ + if (!env->cur_state->active_rcu_lock || !is_trusted_reg(reg)) + flag &= ~MEM_RCU; + else + flag |= PTR_TRUSTED; + } else if (reg->type & MEM_RCU) { + /* ptr (reg) is marked as MEM_RCU, but value reg is not marked + * as MEM_RCU. Mark the value reg as PTR_UNTRUSTED conservatively. + */ + flag |= PTR_UNTRUSTED;The part about 'value reg' doesn't look correct. This part of the code has no idea about 'value reg' yet. We just checked 'flag & MEM_RCU', so it's the flag that doesn't have MEM_RCU set which means that the field of the structure we're dereferencing doesn't have __rcu tag. I think it's better to adjust this comment.
ack.
+ } + if (atype == BPF_READ && value_regno >= 0) mark_btf_ld_reg(env, regs, value_regno, ret, reg->btf, btf_id, flag); @@ -5897,6 +5946,7 @@ static const struct bpf_reg_types btf_ptr_types = { .types = { PTR_TO_BTF_ID, PTR_TO_BTF_ID | PTR_TRUSTED, + PTR_TO_BTF_ID | MEM_RCU | PTR_TRUSTED, }, }; static const struct bpf_reg_types percpu_btf_ptr_types = { @@ -6075,6 +6125,7 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env, case PTR_TO_BTF_ID: case PTR_TO_BTF_ID | MEM_ALLOC: case PTR_TO_BTF_ID | PTR_TRUSTED: + case PTR_TO_BTF_ID | MEM_RCU | PTR_TRUSTED: case PTR_TO_BTF_ID | MEM_ALLOC | PTR_TRUSTED: /* When referenced PTR_TO_BTF_ID is passed to release function, * it's fixed offset must be 0. In the other cases, fixed offset @@ -7539,6 +7590,17 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn return err; } + if (env->cur_state->active_rcu_lock) { + if (fn->might_sleep) { + verbose(env, "sleepable helper %s#%din rcu_read_lock region\n", + func_id_name(func_id), func_id); + return -EINVAL; + } + + if (env->prog->aux->sleepable && is_storage_get_function(func_id)) + env->insn_aux_data[insn_idx].storage_get_func_atomic = true; + } + meta.func_id = func_id; /* check args */ for (i = 0; i < MAX_BPF_FUNC_REG_ARGS; i++) { @@ -7966,25 +8028,6 @@ static bool is_kfunc_arg_kptr_get(struct bpf_kfunc_call_arg_meta *meta, int arg) return arg == 0 && (meta->kfunc_flags & KF_KPTR_GET); } -static bool is_trusted_reg(const struct bpf_reg_state *reg) -{ - /* A referenced register is always trusted. */ - if (reg->ref_obj_id) - return true; - - /* If a register is not referenced, it is trusted if it has either the - * MEM_ALLOC or PTR_TRUSTED type modifiers, and no others. Some of the - * other type modifiers may be safe, but we elect to take an opt-in - * approach here as some (e.g. PTR_UNTRUSTED and PTR_MAYBE_NULL) are - * not. - * - * Eventually, we should make PTR_TRUSTED the single source of truth - * for whether a register is trusted. - */ - return type_flag(reg->type) & BPF_REG_TRUSTED_MODIFIERS && - !bpf_type_has_unsafe_modifiers(reg->type); -} - static bool __kfunc_param_match_suffix(const struct btf *btf, const struct btf_param *arg, const char *suffix) @@ -8163,6 +8206,8 @@ enum special_kfunc_type { KF_bpf_list_pop_back, KF_bpf_cast_to_kern_ctx, KF_bpf_rdonly_cast, + KF_bpf_rcu_read_lock, + KF_bpf_rcu_read_unlock, }; BTF_SET_START(special_kfunc_set) @@ -8185,6 +8230,18 @@ BTF_ID(func, bpf_list_pop_front) BTF_ID(func, bpf_list_pop_back) BTF_ID(func, bpf_cast_to_kern_ctx) BTF_ID(func, bpf_rdonly_cast) +BTF_ID(func, bpf_rcu_read_lock) +BTF_ID(func, bpf_rcu_read_unlock) + +static bool is_kfunc_bpf_rcu_read_lock(struct bpf_kfunc_call_arg_meta *meta) +{ + return meta->func_id == special_kfunc_list[KF_bpf_rcu_read_lock]; +} + +static bool is_kfunc_bpf_rcu_read_unlock(struct bpf_kfunc_call_arg_meta *meta) +{ + return meta->func_id == special_kfunc_list[KF_bpf_rcu_read_unlock]; +} static enum kfunc_ptr_arg_type get_kfunc_ptr_arg_type(struct bpf_verifier_env *env, @@ -8817,6 +8874,7 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, const struct btf_type *t, *func, *func_proto, *ptr_type; struct bpf_reg_state *regs = cur_regs(env); const char *func_name, *ptr_type_name; + bool sleepable, rcu_lock, rcu_unlock; struct bpf_kfunc_call_arg_meta meta; u32 i, nargs, func_id, ptr_type_id; int err, insn_idx = *insn_idx_p; @@ -8858,11 +8916,38 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, return -EACCES; } - if (is_kfunc_sleepable(&meta) && !env->prog->aux->sleepable) { + sleepable = is_kfunc_sleepable(&meta); + if (sleepable && !env->prog->aux->sleepable) { verbose(env, "program must be sleepable to call sleepable kfunc %s\n", func_name); return -EACCES; } + rcu_lock = is_kfunc_bpf_rcu_read_lock(&meta); + rcu_unlock = is_kfunc_bpf_rcu_read_unlock(&meta); + if (env->cur_state->active_rcu_lock) { + struct bpf_func_state *state; + struct bpf_reg_state *reg; + + if (rcu_lock) { + verbose(env, "nested rcu read lock (kernel function %s)\n", func_name); + return -EINVAL; + } else if (rcu_unlock) { + bpf_for_each_reg_in_vstate(env->cur_state, state, reg, ({ + if (reg->type & MEM_RCU) + __mark_reg_unknown(env, reg); + }));That feels too drastic. rcu_unlock will mark all pointers as scalar, but the prog can still do bpf_rdonly_cast and read them. Why force the prog to jump through such hoops? Are we trying to prevent some kind of programming mistake? Maybe clear MEM_RCU flag here and add PTR_UNTRUSTED instead?
The original idea is to prevent rcu pointer from leaking out of rcu read lock region. The goal is to ensure rcu common practice. Maybe this isindeed too strict. As you suggested, the rcu pointer can be marked as PTR_UNTRUSTED so it can still be used outside rcu read lock region
but not able to pass to helper/kfunc.
+ env->cur_state->active_rcu_lock = false; + } else if (sleepable) { + verbose(env, "kernel func %s is sleepable within rcu_read_lock region\n", func_name); + return -EACCES; + } + } else if (rcu_lock) { + env->cur_state->active_rcu_lock = true; + } else if (rcu_unlock) { + verbose(env, "unmatched rcu read unlock (kernel function %s)\n", func_name); + return -EINVAL; + } + /* Check the arguments */ err = check_kfunc_args(env, &meta); if (err < 0) @@ -11754,6 +11839,11 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn) return -EINVAL; } + if (env->cur_state->active_rcu_lock) { + verbose(env, "BPF_LD_[ABS|IND] cannot be used inside bpf_rcu_read_lock-ed region\n"); + return -EINVAL; + } + if (regs[ctx_reg].type != PTR_TO_CTX) { verbose(env, "at the time of BPF_LD_ABS|IND R6 != pointer to skb\n"); @@ -13019,6 +13109,9 @@ static bool states_equal(struct bpf_verifier_env *env, old->active_lock.id != cur->active_lock.id) return false; + if (old->active_rcu_lock != cur->active_rcu_lock) + return false; + /* for states to be equal callsites have to be the same * and all frame states need to be equivalent */ @@ -13706,6 +13799,11 @@ static int do_check(struct bpf_verifier_env *env) return -EINVAL; } + if (env->cur_state->active_rcu_lock) { + verbose(env, "bpf_rcu_read_unlock is missing\n"); + return -EINVAL; + } + /* We must do check_reference_leak here before * prepare_func_exit to handle the case when * state->curframe > 0, it may be a callback @@ -14802,6 +14900,7 @@ static int convert_ctx_accesses(struct bpf_verifier_env *env) case PTR_TO_BTF_ID: case PTR_TO_BTF_ID | PTR_UNTRUSTED: case PTR_TO_BTF_ID | PTR_TRUSTED: + case PTR_TO_BTF_ID | MEM_RCU | PTR_TRUSTED: /* PTR_TO_BTF_ID | MEM_ALLOC always has a valid lifetime, unlike * PTR_TO_BTF_ID, and an active ref_obj_id, but the same cannot * be said once it is marked PTR_UNTRUSTED, hence we must handlewait a sec. Why are we converting PTR_TRUSTED, MEM_RCU, MEM_ALLOC pointers into BPF_PROBE_MEM ? The mistake slipped in earlier, but let's fix it first. BPF_REG_TRUSTED_MODIFIERS should stay as normal LDX.
Good point, let me fix this.
@@ -15494,14 +15593,12 @@ static int do_misc_fixups(struct bpf_verifier_env *env) goto patch_call_imm; } - if (insn->imm == BPF_FUNC_task_storage_get || - insn->imm == BPF_FUNC_sk_storage_get || - insn->imm == BPF_FUNC_inode_storage_get || - insn->imm == BPF_FUNC_cgrp_storage_get) { - if (env->prog->aux->sleepable) - insn_buf[0] = BPF_MOV64_IMM(BPF_REG_5, (__force __s32)GFP_KERNEL); - else + if (is_storage_get_function(insn->imm)) { + if (!env->prog->aux->sleepable || + env->insn_aux_data[i + delta].storage_get_func_atomic) insn_buf[0] = BPF_MOV64_IMM(BPF_REG_5, (__force __s32)GFP_ATOMIC); + else + insn_buf[0] = BPF_MOV64_IMM(BPF_REG_5, (__force __s32)GFP_KERNEL); insn_buf[1] = *insn; cnt = 2; -- 2.30.2
