From 3feb263bb516ee7e1da0acd22b15afbb9a7daa19 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 9 Nov 2023 16:26:36 -0800 Subject: bpf: handle ldimm64 properly in check_cfg() ldimm64 instructions are 16-byte long, and so have to be handled appropriately in check_cfg(), just like the rest of BPF verifier does. This has implications in three places: - when determining next instruction for non-jump instructions; - when determining next instruction for callback address ldimm64 instructions (in visit_func_call_insn()); - when checking for unreachable instructions, where second half of ldimm64 is expected to be unreachable; We take this also as an opportunity to report jump into the middle of ldimm64. And adjust few test_verifier tests accordingly. Acked-by: Eduard Zingerman Reported-by: Hao Sun Fixes: 475fb78fbf48 ("bpf: verifier (add branch/goto checks)") Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20231110002638.4168352-2-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 27 ++++++++++++++++++++------- 1 file changed, 20 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index bd1c42eb540f..b87715b364fd 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -15439,15 +15439,16 @@ static int visit_func_call_insn(int t, struct bpf_insn *insns, struct bpf_verifier_env *env, bool visit_callee) { - int ret; + int ret, insn_sz; - ret = push_insn(t, t + 1, FALLTHROUGH, env, false); + insn_sz = bpf_is_ldimm64(&insns[t]) ? 2 : 1; + ret = push_insn(t, t + insn_sz, FALLTHROUGH, env, false); if (ret) return ret; - mark_prune_point(env, t + 1); + mark_prune_point(env, t + insn_sz); /* when we exit from subprog, we need to record non-linear history */ - mark_jmp_point(env, t + 1); + mark_jmp_point(env, t + insn_sz); if (visit_callee) { mark_prune_point(env, t); @@ -15469,15 +15470,17 @@ static int visit_func_call_insn(int t, struct bpf_insn *insns, static int visit_insn(int t, struct bpf_verifier_env *env) { struct bpf_insn *insns = env->prog->insnsi, *insn = &insns[t]; - int ret, off; + int ret, off, insn_sz; if (bpf_pseudo_func(insn)) return visit_func_call_insn(t, insns, env, true); /* All non-branch instructions have a single fall-through edge. */ if (BPF_CLASS(insn->code) != BPF_JMP && - BPF_CLASS(insn->code) != BPF_JMP32) - return push_insn(t, t + 1, FALLTHROUGH, env, false); + BPF_CLASS(insn->code) != BPF_JMP32) { + insn_sz = bpf_is_ldimm64(insn) ? 2 : 1; + return push_insn(t, t + insn_sz, FALLTHROUGH, env, false); + } switch (BPF_OP(insn->code)) { case BPF_EXIT: @@ -15607,11 +15610,21 @@ walk_cfg: } for (i = 0; i < insn_cnt; i++) { + struct bpf_insn *insn = &env->prog->insnsi[i]; + if (insn_state[i] != EXPLORED) { verbose(env, "unreachable insn %d\n", i); ret = -EINVAL; goto err_free; } + if (bpf_is_ldimm64(insn)) { + if (insn_state[i + 1] != 0) { + verbose(env, "jump into the middle of ldimm64 insn %d\n", i); + ret = -EINVAL; + goto err_free; + } + i++; /* skip second half of ldimm64 */ + } } ret = 0; /* cfg looks good */ -- cgit v1.2.3 From 4bb7ea946a370707315ab774432963ce47291946 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 9 Nov 2023 16:26:37 -0800 Subject: bpf: fix precision backtracking instruction iteration Fix an edge case in __mark_chain_precision() which prematurely stops backtracking instructions in a state if it happens that state's first and last instruction indexes are the same. This situations doesn't necessarily mean that there were no instructions simulated in a state, but rather that we starting from the instruction, jumped around a bit, and then ended up at the same instruction before checkpointing or marking precision. To distinguish between these two possible situations, we need to consult jump history. If it's empty or contain a single record "bridging" parent state and first instruction of processed state, then we indeed backtracked all instructions in this state. But if history is not empty, we are definitely not done yet. Move this logic inside get_prev_insn_idx() to contain it more nicely. Use -ENOENT return code to denote "we are out of instructions" situation. This bug was exposed by verifier_loop1.c's bounded_recursion subtest, once the next fix in this patch set is applied. Acked-by: Eduard Zingerman Fixes: b5dc0163d8fd ("bpf: precise scalar_value tracking") Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20231110002638.4168352-3-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index b87715b364fd..484c742f733e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -3516,12 +3516,29 @@ static int push_jmp_history(struct bpf_verifier_env *env, /* Backtrack one insn at a time. If idx is not at the top of recorded * history then previous instruction came from straight line execution. + * Return -ENOENT if we exhausted all instructions within given state. + * + * It's legal to have a bit of a looping with the same starting and ending + * insn index within the same state, e.g.: 3->4->5->3, so just because current + * instruction index is the same as state's first_idx doesn't mean we are + * done. If there is still some jump history left, we should keep going. We + * need to take into account that we might have a jump history between given + * state's parent and itself, due to checkpointing. In this case, we'll have + * history entry recording a jump from last instruction of parent state and + * first instruction of given state. */ static int get_prev_insn_idx(struct bpf_verifier_state *st, int i, u32 *history) { u32 cnt = *history; + if (i == st->first_insn_idx) { + if (cnt == 0) + return -ENOENT; + if (cnt == 1 && st->jmp_history[0].idx == i) + return -ENOENT; + } + if (cnt && st->jmp_history[cnt - 1].idx == i) { i = st->jmp_history[cnt - 1].prev_idx; (*history)--; @@ -4401,10 +4418,10 @@ static int __mark_chain_precision(struct bpf_verifier_env *env, int regno) * Nothing to be tracked further in the parent state. */ return 0; - if (i == first_idx) - break; subseq_idx = i; i = get_prev_insn_idx(st, i, &history); + if (i == -ENOENT) + break; if (i >= env->prog->len) { /* This can happen if backtracking reached insn 0 * and there are still reg_mask or stack_mask -- cgit v1.2.3 From 10e14e9652bf9e8104151bfd9200433083deae3d Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Thu, 9 Nov 2023 22:14:10 -0800 Subject: bpf: fix control-flow graph checking in privileged mode When BPF program is verified in privileged mode, BPF verifier allows bounded loops. This means that from CFG point of view there are definitely some back-edges. Original commit adjusted check_cfg() logic to not detect back-edges in control flow graph if they are resulting from conditional jumps, which the idea that subsequent full BPF verification process will determine whether such loops are bounded or not, and either accept or reject the BPF program. At least that's my reading of the intent. Unfortunately, the implementation of this idea doesn't work correctly in all possible situations. Conditional jump might not result in immediate back-edge, but just a few unconditional instructions later we can arrive at back-edge. In such situations check_cfg() would reject BPF program even in privileged mode, despite it might be bounded loop. Next patch adds one simple program demonstrating such scenario. To keep things simple, instead of trying to detect back edges in privileged mode, just assume every back edge is valid and let subsequent BPF verification prove or reject bounded loops. Note a few test changes. For unknown reason, we have a few tests that are specified to detect a back-edge in a privileged mode, but looking at their code it seems like the right outcome is passing check_cfg() and letting subsequent verification to make a decision about bounded or not bounded looping. Bounded recursion case is also interesting. The example should pass, as recursion is limited to just a few levels and so we never reach maximum number of nested frames and never exhaust maximum stack depth. But the way that max stack depth logic works today it falsely detects this as exceeding max nested frame count. This patch series doesn't attempt to fix this orthogonal problem, so we just adjust expected verifier failure. Suggested-by: Alexei Starovoitov Fixes: 2589726d12a1 ("bpf: introduce bounded loops") Reported-by: Hao Sun Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20231110061412.2995786-1-andrii@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 23 ++++++++--------------- 1 file changed, 8 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 484c742f733e..a2267d5ed14e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -15403,8 +15403,7 @@ enum { * w - next instruction * e - edge */ -static int push_insn(int t, int w, int e, struct bpf_verifier_env *env, - bool loop_ok) +static int push_insn(int t, int w, int e, struct bpf_verifier_env *env) { int *insn_stack = env->cfg.insn_stack; int *insn_state = env->cfg.insn_state; @@ -15436,7 +15435,7 @@ static int push_insn(int t, int w, int e, struct bpf_verifier_env *env, insn_stack[env->cfg.cur_stack++] = w; return KEEP_EXPLORING; } else if ((insn_state[w] & 0xF0) == DISCOVERED) { - if (loop_ok && env->bpf_capable) + if (env->bpf_capable) return DONE_EXPLORING; verbose_linfo(env, t, "%d: ", t); verbose_linfo(env, w, "%d: ", w); @@ -15459,7 +15458,7 @@ static int visit_func_call_insn(int t, struct bpf_insn *insns, int ret, insn_sz; insn_sz = bpf_is_ldimm64(&insns[t]) ? 2 : 1; - ret = push_insn(t, t + insn_sz, FALLTHROUGH, env, false); + ret = push_insn(t, t + insn_sz, FALLTHROUGH, env); if (ret) return ret; @@ -15469,12 +15468,7 @@ static int visit_func_call_insn(int t, struct bpf_insn *insns, if (visit_callee) { mark_prune_point(env, t); - ret = push_insn(t, t + insns[t].imm + 1, BRANCH, env, - /* It's ok to allow recursion from CFG point of - * view. __check_func_call() will do the actual - * check. - */ - bpf_pseudo_func(insns + t)); + ret = push_insn(t, t + insns[t].imm + 1, BRANCH, env); } return ret; } @@ -15496,7 +15490,7 @@ static int visit_insn(int t, struct bpf_verifier_env *env) if (BPF_CLASS(insn->code) != BPF_JMP && BPF_CLASS(insn->code) != BPF_JMP32) { insn_sz = bpf_is_ldimm64(insn) ? 2 : 1; - return push_insn(t, t + insn_sz, FALLTHROUGH, env, false); + return push_insn(t, t + insn_sz, FALLTHROUGH, env); } switch (BPF_OP(insn->code)) { @@ -15543,8 +15537,7 @@ static int visit_insn(int t, struct bpf_verifier_env *env) off = insn->imm; /* unconditional jump with single edge */ - ret = push_insn(t, t + off + 1, FALLTHROUGH, env, - true); + ret = push_insn(t, t + off + 1, FALLTHROUGH, env); if (ret) return ret; @@ -15557,11 +15550,11 @@ static int visit_insn(int t, struct bpf_verifier_env *env) /* conditional jump with two edges */ mark_prune_point(env, t); - ret = push_insn(t, t + 1, FALLTHROUGH, env, true); + ret = push_insn(t, t + 1, FALLTHROUGH, env); if (ret) return ret; - return push_insn(t, t + insn->off + 1, BRANCH, env, true); + return push_insn(t, t + insn->off + 1, BRANCH, env); } } -- cgit v1.2.3 From 5c0930ccaad5a74d74e8b18b648c5eb21ed2fe94 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 7 Nov 2023 15:57:13 +0100 Subject: hrtimers: Push pending hrtimers away from outgoing CPU earlier 2b8272ff4a70 ("cpu/hotplug: Prevent self deadlock on CPU hot-unplug") solved the straight forward CPU hotplug deadlock vs. the scheduler bandwidth timer. Yu discovered a more involved variant where a task which has a bandwidth timer started on the outgoing CPU holds a lock and then gets throttled. If the lock required by one of the CPU hotplug callbacks the hotplug operation deadlocks because the unthrottling timer event is not handled on the dying CPU and can only be recovered once the control CPU reaches the hotplug state which pulls the pending hrtimers from the dead CPU. Solve this by pushing the hrtimers away from the dying CPU in the dying callbacks. Nothing can queue a hrtimer on the dying CPU at that point because all other CPUs spin in stop_machine() with interrupts disabled and once the operation is finished the CPU is marked offline. Reported-by: Yu Liao Signed-off-by: Thomas Gleixner Tested-by: Liu Tie Link: https://lore.kernel.org/r/87a5rphara.ffs@tglx --- kernel/cpu.c | 8 +++++++- kernel/time/hrtimer.c | 33 ++++++++++++--------------------- 2 files changed, 19 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index 6de7c6bb74ee..2e69a1deaa31 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -2098,7 +2098,7 @@ static struct cpuhp_step cpuhp_hp_states[] = { [CPUHP_HRTIMERS_PREPARE] = { .name = "hrtimers:prepare", .startup.single = hrtimers_prepare_cpu, - .teardown.single = hrtimers_dead_cpu, + .teardown.single = NULL, }, [CPUHP_SMPCFD_PREPARE] = { .name = "smpcfd:prepare", @@ -2190,6 +2190,12 @@ static struct cpuhp_step cpuhp_hp_states[] = { .startup.single = NULL, .teardown.single = smpcfd_dying_cpu, }, + [CPUHP_AP_HRTIMERS_DYING] = { + .name = "hrtimers:dying", + .startup.single = NULL, + .teardown.single = hrtimers_cpu_dying, + }, + /* Entry state on starting. Interrupts enabled from here on. Transient * state for synchronsization */ [CPUHP_AP_ONLINE] = { diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 238262e4aba7..760793998cdd 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -2219,29 +2219,22 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, } } -int hrtimers_dead_cpu(unsigned int scpu) +int hrtimers_cpu_dying(unsigned int dying_cpu) { struct hrtimer_cpu_base *old_base, *new_base; - int i; + int i, ncpu = cpumask_first(cpu_active_mask); - BUG_ON(cpu_online(scpu)); - tick_cancel_sched_timer(scpu); + tick_cancel_sched_timer(dying_cpu); + + old_base = this_cpu_ptr(&hrtimer_bases); + new_base = &per_cpu(hrtimer_bases, ncpu); - /* - * this BH disable ensures that raise_softirq_irqoff() does - * not wakeup ksoftirqd (and acquire the pi-lock) while - * holding the cpu_base lock - */ - local_bh_disable(); - local_irq_disable(); - old_base = &per_cpu(hrtimer_bases, scpu); - new_base = this_cpu_ptr(&hrtimer_bases); /* * The caller is globally serialized and nobody else * takes two locks at once, deadlock is not possible. */ - raw_spin_lock(&new_base->lock); - raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&old_base->lock); + raw_spin_lock_nested(&new_base->lock, SINGLE_DEPTH_NESTING); for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { migrate_hrtimer_list(&old_base->clock_base[i], @@ -2252,15 +2245,13 @@ int hrtimers_dead_cpu(unsigned int scpu) * The migration might have changed the first expiring softirq * timer on this CPU. Update it. */ - hrtimer_update_softirq_timer(new_base, false); + __hrtimer_get_next_event(new_base, HRTIMER_ACTIVE_SOFT); + /* Tell the other CPU to retrigger the next event */ + smp_call_function_single(ncpu, retrigger_next_event, NULL, 0); - raw_spin_unlock(&old_base->lock); raw_spin_unlock(&new_base->lock); + raw_spin_unlock(&old_base->lock); - /* Check, if we got expired work to do */ - __hrtimer_peek_ahead_timers(); - local_irq_enable(); - local_bh_enable(); return 0; } -- cgit v1.2.3 From eab03c23c2a162085b13200d7942fc5a00b5ccc8 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Tue, 7 Nov 2023 17:05:07 +0800 Subject: sched/eevdf: Fix vruntime adjustment on reweight vruntime of the (on_rq && !0-lag) entity needs to be adjusted when it gets re-weighted, and the calculations can be simplified based on the fact that re-weight won't change the w-average of all the entities. Please check the proofs in comments. But adjusting vruntime can also cause position change in RB-tree hence require re-queue to fix up which might be costly. This might be avoided by deferring adjustment to the time the entity actually leaves tree (dequeue/pick), but that will negatively affect task selection and probably not good enough either. Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy") Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20231107090510.71322-2-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 151 ++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 128 insertions(+), 23 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 2048138ce54b..025d90925bf6 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3666,41 +3666,140 @@ static inline void dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } #endif +static void reweight_eevdf(struct cfs_rq *cfs_rq, struct sched_entity *se, + unsigned long weight) +{ + unsigned long old_weight = se->load.weight; + u64 avruntime = avg_vruntime(cfs_rq); + s64 vlag, vslice; + + /* + * VRUNTIME + * ======== + * + * COROLLARY #1: The virtual runtime of the entity needs to be + * adjusted if re-weight at !0-lag point. + * + * Proof: For contradiction assume this is not true, so we can + * re-weight without changing vruntime at !0-lag point. + * + * Weight VRuntime Avg-VRuntime + * before w v V + * after w' v' V' + * + * Since lag needs to be preserved through re-weight: + * + * lag = (V - v)*w = (V'- v')*w', where v = v' + * ==> V' = (V - v)*w/w' + v (1) + * + * Let W be the total weight of the entities before reweight, + * since V' is the new weighted average of entities: + * + * V' = (WV + w'v - wv) / (W + w' - w) (2) + * + * by using (1) & (2) we obtain: + * + * (WV + w'v - wv) / (W + w' - w) = (V - v)*w/w' + v + * ==> (WV-Wv+Wv+w'v-wv)/(W+w'-w) = (V - v)*w/w' + v + * ==> (WV - Wv)/(W + w' - w) + v = (V - v)*w/w' + v + * ==> (V - v)*W/(W + w' - w) = (V - v)*w/w' (3) + * + * Since we are doing at !0-lag point which means V != v, we + * can simplify (3): + * + * ==> W / (W + w' - w) = w / w' + * ==> Ww' = Ww + ww' - ww + * ==> W * (w' - w) = w * (w' - w) + * ==> W = w (re-weight indicates w' != w) + * + * So the cfs_rq contains only one entity, hence vruntime of + * the entity @v should always equal to the cfs_rq's weighted + * average vruntime @V, which means we will always re-weight + * at 0-lag point, thus breach assumption. Proof completed. + * + * + * COROLLARY #2: Re-weight does NOT affect weighted average + * vruntime of all the entities. + * + * Proof: According to corollary #1, Eq. (1) should be: + * + * (V - v)*w = (V' - v')*w' + * ==> v' = V' - (V - v)*w/w' (4) + * + * According to the weighted average formula, we have: + * + * V' = (WV - wv + w'v') / (W - w + w') + * = (WV - wv + w'(V' - (V - v)w/w')) / (W - w + w') + * = (WV - wv + w'V' - Vw + wv) / (W - w + w') + * = (WV + w'V' - Vw) / (W - w + w') + * + * ==> V'*(W - w + w') = WV + w'V' - Vw + * ==> V' * (W - w) = (W - w) * V (5) + * + * If the entity is the only one in the cfs_rq, then reweight + * always occurs at 0-lag point, so V won't change. Or else + * there are other entities, hence W != w, then Eq. (5) turns + * into V' = V. So V won't change in either case, proof done. + * + * + * So according to corollary #1 & #2, the effect of re-weight + * on vruntime should be: + * + * v' = V' - (V - v) * w / w' (4) + * = V - (V - v) * w / w' + * = V - vl * w / w' + * = V - vl' + */ + if (avruntime != se->vruntime) { + vlag = (s64)(avruntime - se->vruntime); + vlag = div_s64(vlag * old_weight, weight); + se->vruntime = avruntime - vlag; + } + + /* + * DEADLINE + * ======== + * + * When the weight changes, the virtual time slope changes and + * we should adjust the relative virtual deadline accordingly. + * + * d' = v' + (d - v)*w/w' + * = V' - (V - v)*w/w' + (d - v)*w/w' + * = V - (V - v)*w/w' + (d - v)*w/w' + * = V + (d - V)*w/w' + */ + vslice = (s64)(se->deadline - avruntime); + vslice = div_s64(vslice * old_weight, weight); + se->deadline = avruntime + vslice; +} + static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, unsigned long weight) { - unsigned long old_weight = se->load.weight; + bool curr = cfs_rq->curr == se; if (se->on_rq) { /* commit outstanding execution time */ - if (cfs_rq->curr == se) + if (curr) update_curr(cfs_rq); else - avg_vruntime_sub(cfs_rq, se); + __dequeue_entity(cfs_rq, se); update_load_sub(&cfs_rq->load, se->load.weight); } dequeue_load_avg(cfs_rq, se); - update_load_set(&se->load, weight); - if (!se->on_rq) { /* * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), * we need to scale se->vlag when w_i changes. */ - se->vlag = div_s64(se->vlag * old_weight, weight); + se->vlag = div_s64(se->vlag * se->load.weight, weight); } else { - s64 deadline = se->deadline - se->vruntime; - /* - * When the weight changes, the virtual time slope changes and - * we should adjust the relative virtual deadline accordingly. - */ - deadline = div_s64(deadline * old_weight, weight); - se->deadline = se->vruntime + deadline; - if (se != cfs_rq->curr) - min_deadline_cb_propagate(&se->run_node, NULL); + reweight_eevdf(cfs_rq, se, weight); } + update_load_set(&se->load, weight); + #ifdef CONFIG_SMP do { u32 divider = get_pelt_divider(&se->avg); @@ -3712,8 +3811,17 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, enqueue_load_avg(cfs_rq, se); if (se->on_rq) { update_load_add(&cfs_rq->load, se->load.weight); - if (cfs_rq->curr != se) - avg_vruntime_add(cfs_rq, se); + if (!curr) { + /* + * The entity's vruntime has been adjusted, so let's check + * whether the rq-wide min_vruntime needs updated too. Since + * the calculations above require stable min_vruntime rather + * than up-to-date one, we do the update at the end of the + * reweight process. + */ + __enqueue_entity(cfs_rq, se); + update_min_vruntime(cfs_rq); + } } } @@ -3857,14 +3965,11 @@ static void update_cfs_group(struct sched_entity *se) #ifndef CONFIG_SMP shares = READ_ONCE(gcfs_rq->tg->shares); - - if (likely(se->load.weight == shares)) - return; #else - shares = calc_group_shares(gcfs_rq); + shares = calc_group_shares(gcfs_rq); #endif - - reweight_entity(cfs_rq_of(se), se, shares); + if (unlikely(se->load.weight != shares)) + reweight_entity(cfs_rq_of(se), se, shares); } #else /* CONFIG_FAIR_GROUP_SCHED */ -- cgit v1.2.3 From 8b39d20eceeda6c4eb23df1497f9ed2fffdc8f69 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Thu, 26 Oct 2023 12:41:14 -0400 Subject: sched: psi: fix unprivileged polling against cgroups 519fabc7aaba ("psi: remove 500ms min window size limitation for triggers") breaks unprivileged psi polling on cgroups. Historically, we had a privilege check for polling in the open() of a pressure file in /proc, but were erroneously missing it for the open() of cgroup pressure files. When unprivileged polling was introduced in d82caa273565 ("sched/psi: Allow unprivileged polling of N*2s period"), it needed to filter privileges depending on the exact polling parameters, and as such moved the CAP_SYS_RESOURCE check from the proc open() callback to psi_trigger_create(). Both the proc files as well as cgroup files go through this during write(). This implicitly added the missing check for privileges required for HT polling for cgroups. When 519fabc7aaba ("psi: remove 500ms min window size limitation for triggers") followed right after to remove further restrictions on the RT polling window, it incorrectly assumed the cgroup privilege check was still missing and added it to the cgroup open(), mirroring what we used to do for proc files in the past. As a result, unprivileged poll requests that would be supported now get rejected when opening the cgroup pressure file for writing. Remove the cgroup open() check. psi_trigger_create() handles it. Fixes: 519fabc7aaba ("psi: remove 500ms min window size limitation for triggers") Reported-by: Luca Boccassi Signed-off-by: Johannes Weiner Signed-off-by: Peter Zijlstra (Intel) Acked-by: Luca Boccassi Acked-by: Suren Baghdasaryan Cc: stable@vger.kernel.org # 6.5+ Link: https://lore.kernel.org/r/20231026164114.2488682-1-hannes@cmpxchg.org --- kernel/cgroup/cgroup.c | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 1d5b9de3b1b9..4b9ff41ca603 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -3885,14 +3885,6 @@ static __poll_t cgroup_pressure_poll(struct kernfs_open_file *of, return psi_trigger_poll(&ctx->psi.trigger, of->file, pt); } -static int cgroup_pressure_open(struct kernfs_open_file *of) -{ - if (of->file->f_mode & FMODE_WRITE && !capable(CAP_SYS_RESOURCE)) - return -EPERM; - - return 0; -} - static void cgroup_pressure_release(struct kernfs_open_file *of) { struct cgroup_file_ctx *ctx = of->priv; @@ -5299,7 +5291,6 @@ static struct cftype cgroup_psi_files[] = { { .name = "io.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_IO]), - .open = cgroup_pressure_open, .seq_show = cgroup_io_pressure_show, .write = cgroup_io_pressure_write, .poll = cgroup_pressure_poll, @@ -5308,7 +5299,6 @@ static struct cftype cgroup_psi_files[] = { { .name = "memory.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_MEM]), - .open = cgroup_pressure_open, .seq_show = cgroup_memory_pressure_show, .write = cgroup_memory_pressure_write, .poll = cgroup_pressure_poll, @@ -5317,7 +5307,6 @@ static struct cftype cgroup_psi_files[] = { { .name = "cpu.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_CPU]), - .open = cgroup_pressure_open, .seq_show = cgroup_cpu_pressure_show, .write = cgroup_cpu_pressure_write, .poll = cgroup_pressure_poll, @@ -5327,7 +5316,6 @@ static struct cftype cgroup_psi_files[] = { { .name = "irq.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_IRQ]), - .open = cgroup_pressure_open, .seq_show = cgroup_irq_pressure_show, .write = cgroup_irq_pressure_write, .poll = cgroup_pressure_poll, -- cgit v1.2.3 From 6d7e4782bcf549221b4ccfffec2cf4d1a473f1a3 Mon Sep 17 00:00:00 2001 From: Keisuke Nishimura Date: Tue, 31 Oct 2023 14:38:22 +0100 Subject: sched/fair: Fix the decision for load balance should_we_balance is called for the decision to do load-balancing. When sched ticks invoke this function, only one CPU should return true. However, in the current code, two CPUs can return true. The following situation, where b means busy and i means idle, is an example, because CPU 0 and CPU 2 return true. [0, 1] [2, 3] b b i b This fix checks if there exists an idle CPU with busy sibling(s) after looking for a CPU on an idle core. If some idle CPUs with busy siblings are found, just the first one should do load-balancing. Fixes: b1bfeab9b002 ("sched/fair: Consider the idle state of the whole core for load balance") Signed-off-by: Keisuke Nishimura Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Chen Yu Reviewed-by: Shrikanth Hegde Reviewed-by: Vincent Guittot Link: https://lkml.kernel.org/r/20231031133821.1570861-1-keisuke.nishimura@inria.fr --- kernel/sched/fair.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 025d90925bf6..d7a3c63a2171 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11184,12 +11184,16 @@ static int should_we_balance(struct lb_env *env) continue; } - /* Are we the first idle CPU? */ + /* + * Are we the first idle core in a non-SMT domain or higher, + * or the first idle CPU in a SMT domain? + */ return cpu == env->dst_cpu; } - if (idle_smt == env->dst_cpu) - return true; + /* Are we the first idle CPU with busy siblings? */ + if (idle_smt != -1) + return idle_smt == env->dst_cpu; /* Are we the first CPU of this group ? */ return group_balance_cpu(sg) == env->dst_cpu; -- cgit v1.2.3 From 969d90ec212bae4b45bf9d21d7daa30aa6cf055e Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Tue, 14 Nov 2023 17:25:48 -0500 Subject: audit: don't WARN_ON_ONCE(!current->mm) in audit_exe_compare() eBPF can end up calling into the audit code from some odd places, and some of these places don't have @current set properly so we end up tripping the `WARN_ON_ONCE(!current->mm)` near the top of `audit_exe_compare()`. While the basic `!current->mm` check is good, the `WARN_ON_ONCE()` results in some scary console messages so let's drop that and just do the regular `!current->mm` check to avoid problems. Cc: Fixes: 47846d51348d ("audit: don't take task_lock() in audit_exe_compare() code path") Reported-by: Artem Savkov Signed-off-by: Paul Moore --- kernel/audit_watch.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index 91e82e34b51e..7a98cd176a12 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -531,7 +531,7 @@ int audit_exe_compare(struct task_struct *tsk, struct audit_fsnotify_mark *mark) if (tsk != current) return 0; - if (WARN_ON_ONCE(!current->mm)) + if (!current->mm) return 0; exe_file = get_mm_exe_file(current->mm); if (!exe_file) -- cgit v1.2.3 From c9bd1568d5462f4108417518ce1af7b924acfb6f Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 14 Nov 2023 21:36:13 +0100 Subject: futex: Fix hardcoded flags Xi reported that commit 5694289ce183 ("futex: Flag conversion") broke glibc's robust futex tests. This was narrowed down to the change of FLAGS_SHARED from 0x01 to 0x10, at which point Florian noted that handle_futex_death() has a hardcoded flags argument of 1. Change this to: FLAGS_SIZE_32 | FLAGS_SHARED, matching how futex_to_flags() unconditionally sets FLAGS_SIZE_32 for all legacy futex ops. Reported-by: Xi Ruoyao Reported-by: Florian Weimer Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/20231114201402.GA25315@noisy.programming.kicks-ass.net Fixes: 5694289ce183 ("futex: Flag conversion") Cc: --- kernel/futex/core.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/futex/core.c b/kernel/futex/core.c index 52695c59d041..dad981a865b8 100644 --- a/kernel/futex/core.c +++ b/kernel/futex/core.c @@ -700,7 +700,8 @@ retry: owner = uval & FUTEX_TID_MASK; if (pending_op && !pi && !owner) { - futex_wake(uaddr, 1, 1, FUTEX_BITSET_MATCH_ANY); + futex_wake(uaddr, FLAGS_SIZE_32 | FLAGS_SHARED, 1, + FUTEX_BITSET_MATCH_ANY); return 0; } @@ -752,8 +753,10 @@ retry: * Wake robust non-PI futexes here. The wakeup of * PI futexes happens in exit_pi_state(): */ - if (!pi && (uval & FUTEX_WAITERS)) - futex_wake(uaddr, 1, 1, FUTEX_BITSET_MATCH_ANY); + if (!pi && (uval & FUTEX_WAITERS)) { + futex_wake(uaddr, FLAGS_SIZE_32 | FLAGS_SHARED, 1, + FUTEX_BITSET_MATCH_ANY); + } return 0; } -- cgit v1.2.3 From 889c58b3155ff4c8e8671c95daef63d6fabbb6b1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 9 Jun 2023 12:34:46 +0200 Subject: perf/core: Fix cpuctx refcounting Audit of the refcounting turned up that perf_pmu_migrate_context() fails to migrate the ctx refcount. Fixes: bd2756811766 ("perf: Rewrite core context handling") Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lkml.kernel.org/r/20230612093539.085862001@infradead.org Cc: --- kernel/events/core.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 683dc086ef10..b704d83a28b2 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4828,6 +4828,11 @@ find_get_pmu_context(struct pmu *pmu, struct perf_event_context *ctx, void *task_ctx_data = NULL; if (!ctx->task) { + /* + * perf_pmu_migrate_context() / __perf_pmu_install_event() + * relies on the fact that find_get_pmu_context() cannot fail + * for CPU contexts. + */ struct perf_cpu_pmu_context *cpc; cpc = per_cpu_ptr(pmu->cpu_pmu_context, event->cpu); @@ -12889,6 +12894,9 @@ static void __perf_pmu_install_event(struct pmu *pmu, int cpu, struct perf_event *event) { struct perf_event_pmu_context *epc; + struct perf_event_context *old_ctx = event->ctx; + + get_ctx(ctx); /* normally find_get_context() */ event->cpu = cpu; epc = find_get_pmu_context(pmu, ctx, event); @@ -12897,6 +12905,11 @@ static void __perf_pmu_install_event(struct pmu *pmu, if (event->state >= PERF_EVENT_STATE_OFF) event->state = PERF_EVENT_STATE_INACTIVE; perf_install_in_context(ctx, event, cpu); + + /* + * Now that event->ctx is updated and visible, put the old ctx. + */ + put_ctx(old_ctx); } static void __perf_pmu_install(struct perf_event_context *ctx, @@ -12935,6 +12948,10 @@ void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu) struct perf_event_context *src_ctx, *dst_ctx; LIST_HEAD(events); + /* + * Since per-cpu context is persistent, no need to grab an extra + * reference. + */ src_ctx = &per_cpu_ptr(&perf_cpu_context, src_cpu)->ctx; dst_ctx = &per_cpu_ptr(&perf_cpu_context, dst_cpu)->ctx; -- cgit v1.2.3 From 1fda5bb66ad8fb24ecb3858e61a13a6548428898 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Fri, 10 Nov 2023 17:39:28 -0800 Subject: bpf: Do not allocate percpu memory at init stage Kirill Shutemov reported significant percpu memory consumption increase after booting in 288-cpu VM ([1]) due to commit 41a5db8d8161 ("bpf: Add support for non-fix-size percpu mem allocation"). The percpu memory consumption is increased from 111MB to 969MB. The number is from /proc/meminfo. I tried to reproduce the issue with my local VM which at most supports upto 255 cpus. With 252 cpus, without the above commit, the percpu memory consumption immediately after boot is 57MB while with the above commit the percpu memory consumption is 231MB. This is not good since so far percpu memory from bpf memory allocator is not widely used yet. Let us change pre-allocation in init stage to on-demand allocation when verifier detects there is a need of percpu memory for bpf program. With this change, percpu memory consumption after boot can be reduced signicantly. [1] https://lore.kernel.org/lkml/20231109154934.4saimljtqx625l3v@box.shutemov.name/ Fixes: 41a5db8d8161 ("bpf: Add support for non-fix-size percpu mem allocation") Reported-and-tested-by: Kirill A. Shutemov Signed-off-by: Yonghong Song Acked-by: Hou Tao Link: https://lore.kernel.org/r/20231111013928.948838-1-yonghong.song@linux.dev Signed-off-by: Alexei Starovoitov --- kernel/bpf/core.c | 8 +++----- kernel/bpf/verifier.c | 20 ++++++++++++++++++-- 2 files changed, 21 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 08626b519ce2..cd3afe57ece3 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -64,8 +64,8 @@ #define OFF insn->off #define IMM insn->imm -struct bpf_mem_alloc bpf_global_ma, bpf_global_percpu_ma; -bool bpf_global_ma_set, bpf_global_percpu_ma_set; +struct bpf_mem_alloc bpf_global_ma; +bool bpf_global_ma_set; /* No hurry in this branch * @@ -2934,9 +2934,7 @@ static int __init bpf_global_ma_init(void) ret = bpf_mem_alloc_init(&bpf_global_ma, 0, false); bpf_global_ma_set = !ret; - ret = bpf_mem_alloc_init(&bpf_global_percpu_ma, 0, true); - bpf_global_percpu_ma_set = !ret; - return !bpf_global_ma_set || !bpf_global_percpu_ma_set; + return ret; } late_initcall(bpf_global_ma_init); #endif diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index a2267d5ed14e..6da370a047fe 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include "disasm.h" @@ -41,6 +42,9 @@ static const struct bpf_verifier_ops * const bpf_verifier_ops[] = { #undef BPF_LINK_TYPE }; +struct bpf_mem_alloc bpf_global_percpu_ma; +static bool bpf_global_percpu_ma_set; + /* bpf_check() is a static code analyzer that walks eBPF program * instruction by instruction and updates register/stack state. * All paths of conditional branches are analyzed until 'bpf_exit' insn. @@ -336,6 +340,7 @@ struct bpf_kfunc_call_arg_meta { struct btf *btf_vmlinux; static DEFINE_MUTEX(bpf_verifier_lock); +static DEFINE_MUTEX(bpf_percpu_ma_lock); static const struct bpf_line_info * find_linfo(const struct bpf_verifier_env *env, u32 insn_off) @@ -12091,8 +12096,19 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, if (meta.func_id == special_kfunc_list[KF_bpf_obj_new_impl] && !bpf_global_ma_set) return -ENOMEM; - if (meta.func_id == special_kfunc_list[KF_bpf_percpu_obj_new_impl] && !bpf_global_percpu_ma_set) - return -ENOMEM; + if (meta.func_id == special_kfunc_list[KF_bpf_percpu_obj_new_impl]) { + if (!bpf_global_percpu_ma_set) { + mutex_lock(&bpf_percpu_ma_lock); + if (!bpf_global_percpu_ma_set) { + err = bpf_mem_alloc_init(&bpf_global_percpu_ma, 0, true); + if (!err) + bpf_global_percpu_ma_set = true; + } + mutex_unlock(&bpf_percpu_ma_lock); + if (err) + return err; + } + } if (((u64)(u32)meta.arg_constant.value) != meta.arg_constant.value) { verbose(env, "local type ID argument must be in range [0, U32_MAX]\n"); -- cgit v1.2.3 From 793838138c157d4c49f4fb744b170747e3dabf58 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Sat, 18 Nov 2023 19:33:35 +0100 Subject: prctl: Disable prctl(PR_SET_MDWE) on parisc systemd-254 tries to use prctl(PR_SET_MDWE) for it's MemoryDenyWriteExecute functionality, but fails on parisc which still needs executable stacks in certain combinations of gcc/glibc/kernel. Disable prctl(PR_SET_MDWE) by returning -EINVAL for now on parisc, until userspace has catched up. Signed-off-by: Helge Deller Co-developed-by: Linus Torvalds Reported-by: Sam James Closes: https://github.com/systemd/systemd/issues/29775 Tested-by: Sam James Link: https://lore.kernel.org/all/875y2jro9a.fsf@gentoo.org/ Cc: # v6.3+ --- kernel/sys.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index 420d9cb9cc8e..e219fcfa112d 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2394,6 +2394,10 @@ static inline int prctl_set_mdwe(unsigned long bits, unsigned long arg3, if (bits & PR_MDWE_NO_INHERIT && !(bits & PR_MDWE_REFUSE_EXEC_GAIN)) return -EINVAL; + /* PARISC cannot allow mdwe as it needs writable stacks */ + if (IS_ENABLED(CONFIG_PARISC)) + return -EINVAL; + current_bits = get_current_mdwe(); if (current_bits && current_bits != bits) return -EPERM; /* Cannot unset the flags */ -- cgit v1.2.3 From 683b96f9606ab7308ffb23c46ab43cecdef8a241 Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 21 Nov 2023 04:06:54 +0200 Subject: bpf: extract __check_reg_arg() utility function Split check_reg_arg() into two utility functions: - check_reg_arg() operating on registers from current verifier state; - __check_reg_arg() operating on a specific set of registers passed as a parameter; The __check_reg_arg() function would be used by a follow-up change for callbacks handling. Acked-by: Andrii Nakryiko Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20231121020701.26440-5-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 6da370a047fe..e6e1bcfe00f5 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -3439,13 +3439,11 @@ static void mark_insn_zext(struct bpf_verifier_env *env, reg->subreg_def = DEF_NOT_SUBREG; } -static int check_reg_arg(struct bpf_verifier_env *env, u32 regno, - enum reg_arg_type t) +static int __check_reg_arg(struct bpf_verifier_env *env, struct bpf_reg_state *regs, u32 regno, + enum reg_arg_type t) { - struct bpf_verifier_state *vstate = env->cur_state; - struct bpf_func_state *state = vstate->frame[vstate->curframe]; struct bpf_insn *insn = env->prog->insnsi + env->insn_idx; - struct bpf_reg_state *reg, *regs = state->regs; + struct bpf_reg_state *reg; bool rw64; if (regno >= MAX_BPF_REG) { @@ -3486,6 +3484,15 @@ static int check_reg_arg(struct bpf_verifier_env *env, u32 regno, return 0; } +static int check_reg_arg(struct bpf_verifier_env *env, u32 regno, + enum reg_arg_type t) +{ + struct bpf_verifier_state *vstate = env->cur_state; + struct bpf_func_state *state = vstate->frame[vstate->curframe]; + + return __check_reg_arg(env, state->regs, regno, t); +} + static void mark_jmp_point(struct bpf_verifier_env *env, int idx) { env->insn_aux_data[idx].jmp_point = true; @@ -9350,7 +9357,7 @@ static void clear_caller_saved_regs(struct bpf_verifier_env *env, /* after the call registers r0 - r5 were scratched */ for (i = 0; i < CALLER_SAVED_REGS; i++) { mark_reg_not_init(env, regs, caller_saved[i]); - check_reg_arg(env, caller_saved[i], DST_OP_NO_MARK); + __check_reg_arg(env, regs, caller_saved[i], DST_OP_NO_MARK); } } -- cgit v1.2.3 From 58124a98cb8eda69d248d7f1de954c8b2767c945 Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 21 Nov 2023 04:06:55 +0200 Subject: bpf: extract setup_func_entry() utility function Move code for simulated stack frame creation to a separate utility function. This function would be used in the follow-up change for callbacks handling. Acked-by: Andrii Nakryiko Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20231121020701.26440-6-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 84 +++++++++++++++++++++++++++++---------------------- 1 file changed, 48 insertions(+), 36 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e6e1bcfe00f5..68ee4803d3a2 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9370,11 +9370,10 @@ static int set_callee_state(struct bpf_verifier_env *env, struct bpf_func_state *caller, struct bpf_func_state *callee, int insn_idx); -static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn, - int *insn_idx, int subprog, - set_callee_state_fn set_callee_state_cb) +static int setup_func_entry(struct bpf_verifier_env *env, int subprog, int callsite, + set_callee_state_fn set_callee_state_cb, + struct bpf_verifier_state *state) { - struct bpf_verifier_state *state = env->cur_state; struct bpf_func_state *caller, *callee; int err; @@ -9384,13 +9383,53 @@ static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn return -E2BIG; } - caller = state->frame[state->curframe]; if (state->frame[state->curframe + 1]) { verbose(env, "verifier bug. Frame %d already allocated\n", state->curframe + 1); return -EFAULT; } + caller = state->frame[state->curframe]; + callee = kzalloc(sizeof(*callee), GFP_KERNEL); + if (!callee) + return -ENOMEM; + state->frame[state->curframe + 1] = callee; + + /* callee cannot access r0, r6 - r9 for reading and has to write + * into its own stack before reading from it. + * callee can read/write into caller's stack + */ + init_func_state(env, callee, + /* remember the callsite, it will be used by bpf_exit */ + callsite, + state->curframe + 1 /* frameno within this callchain */, + subprog /* subprog number within this prog */); + /* Transfer references to the callee */ + err = copy_reference_state(callee, caller); + err = err ?: set_callee_state_cb(env, caller, callee, callsite); + if (err) + goto err_out; + + /* only increment it after check_reg_arg() finished */ + state->curframe++; + + return 0; + +err_out: + free_func_state(callee); + state->frame[state->curframe + 1] = NULL; + return err; +} + +static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn, + int *insn_idx, int subprog, + set_callee_state_fn set_callee_state_cb) +{ + struct bpf_verifier_state *state = env->cur_state; + struct bpf_func_state *caller, *callee; + int err; + + caller = state->frame[state->curframe]; err = btf_check_subprog_call(env, subprog, caller->regs); if (err == -EFAULT) return err; @@ -9460,35 +9499,12 @@ static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn return 0; } - callee = kzalloc(sizeof(*callee), GFP_KERNEL); - if (!callee) - return -ENOMEM; - state->frame[state->curframe + 1] = callee; - - /* callee cannot access r0, r6 - r9 for reading and has to write - * into its own stack before reading from it. - * callee can read/write into caller's stack - */ - init_func_state(env, callee, - /* remember the callsite, it will be used by bpf_exit */ - *insn_idx /* callsite */, - state->curframe + 1 /* frameno within this callchain */, - subprog /* subprog number within this prog */); - - /* Transfer references to the callee */ - err = copy_reference_state(callee, caller); + err = setup_func_entry(env, subprog, *insn_idx, set_callee_state_cb, state); if (err) - goto err_out; - - err = set_callee_state_cb(env, caller, callee, *insn_idx); - if (err) - goto err_out; + return err; clear_caller_saved_regs(env, caller->regs); - /* only increment it after check_reg_arg() finished */ - state->curframe++; - /* and go analyze first insn of the callee */ *insn_idx = env->subprog_info[subprog].start - 1; @@ -9496,14 +9512,10 @@ static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn verbose(env, "caller:\n"); print_verifier_state(env, caller, true); verbose(env, "callee:\n"); - print_verifier_state(env, callee, true); + print_verifier_state(env, state->frame[state->curframe], true); } - return 0; -err_out: - free_func_state(callee); - state->frame[state->curframe + 1] = NULL; - return err; + return 0; } int map_set_for_each_callback_args(struct bpf_verifier_env *env, -- cgit v1.2.3 From ab5cfac139ab8576fb54630d4cca23c3e690ee90 Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 21 Nov 2023 04:06:56 +0200 Subject: bpf: verify callbacks as if they are called unknown number of times Prior to this patch callbacks were handled as regular function calls, execution of callback body was modeled exactly once. This patch updates callbacks handling logic as follows: - introduces a function push_callback_call() that schedules callback body verification in env->head stack; - updates prepare_func_exit() to reschedule callback body verification upon BPF_EXIT; - as calls to bpf_*_iter_next(), calls to callback invoking functions are marked as checkpoints; - is_state_visited() is updated to stop callback based iteration when some identical parent state is found. Paths with callback function invoked zero times are now verified first, which leads to necessity to modify some selftests: - the following negative tests required adding release/unlock/drop calls to avoid previously masked unrelated error reports: - cb_refs.c:underflow_prog - exceptions_fail.c:reject_rbtree_add_throw - exceptions_fail.c:reject_with_cp_reference - the following precision tracking selftests needed change in expected log trace: - verifier_subprog_precision.c:callback_result_precise (note: r0 precision is no longer propagated inside callback and I think this is a correct behavior) - verifier_subprog_precision.c:parent_callee_saved_reg_precise_with_callback - verifier_subprog_precision.c:parent_stack_slot_precise_with_callback Reported-by: Andrew Werner Closes: https://lore.kernel.org/bpf/CA+vRuzPChFNXmouzGG+wsy=6eMcfr1mFG0F3g7rbg-sedGKW3w@mail.gmail.com/ Acked-by: Andrii Nakryiko Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20231121020701.26440-7-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 274 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 175 insertions(+), 99 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 68ee4803d3a2..a60dfa56ebb3 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -547,13 +547,12 @@ static bool is_dynptr_ref_function(enum bpf_func_id func_id) return func_id == BPF_FUNC_dynptr_data; } -static bool is_callback_calling_kfunc(u32 btf_id); +static bool is_sync_callback_calling_kfunc(u32 btf_id); static bool is_bpf_throw_kfunc(struct bpf_insn *insn); -static bool is_callback_calling_function(enum bpf_func_id func_id) +static bool is_sync_callback_calling_function(enum bpf_func_id func_id) { return func_id == BPF_FUNC_for_each_map_elem || - func_id == BPF_FUNC_timer_set_callback || func_id == BPF_FUNC_find_vma || func_id == BPF_FUNC_loop || func_id == BPF_FUNC_user_ringbuf_drain; @@ -564,6 +563,18 @@ static bool is_async_callback_calling_function(enum bpf_func_id func_id) return func_id == BPF_FUNC_timer_set_callback; } +static bool is_callback_calling_function(enum bpf_func_id func_id) +{ + return is_sync_callback_calling_function(func_id) || + is_async_callback_calling_function(func_id); +} + +static bool is_sync_callback_calling_insn(struct bpf_insn *insn) +{ + return (bpf_helper_call(insn) && is_sync_callback_calling_function(insn->imm)) || + (bpf_pseudo_kfunc_call(insn) && is_sync_callback_calling_kfunc(insn->imm)); +} + static bool is_storage_get_function(enum bpf_func_id func_id) { return func_id == BPF_FUNC_sk_storage_get || @@ -1808,6 +1819,7 @@ static int copy_verifier_state(struct bpf_verifier_state *dst_state, dst_state->first_insn_idx = src->first_insn_idx; dst_state->last_insn_idx = src->last_insn_idx; dst_state->dfs_depth = src->dfs_depth; + dst_state->callback_unroll_depth = src->callback_unroll_depth; dst_state->used_as_loop_entry = src->used_as_loop_entry; for (i = 0; i <= src->curframe; i++) { dst = dst_state->frame[i]; @@ -3731,6 +3743,8 @@ static void fmt_stack_mask(char *buf, ssize_t buf_sz, u64 stack_mask) } } +static bool calls_callback(struct bpf_verifier_env *env, int insn_idx); + /* For given verifier state backtrack_insn() is called from the last insn to * the first insn. Its purpose is to compute a bitmask of registers and * stack slots that needs precision in the parent verifier state. @@ -3906,16 +3920,13 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx, int subseq_idx, return -EFAULT; return 0; } - } else if ((bpf_helper_call(insn) && - is_callback_calling_function(insn->imm) && - !is_async_callback_calling_function(insn->imm)) || - (bpf_pseudo_kfunc_call(insn) && is_callback_calling_kfunc(insn->imm))) { - /* callback-calling helper or kfunc call, which means - * we are exiting from subprog, but unlike the subprog - * call handling above, we shouldn't propagate - * precision of r1-r5 (if any requested), as they are - * not actually arguments passed directly to callback - * subprogs + } else if (is_sync_callback_calling_insn(insn) && idx != subseq_idx - 1) { + /* exit from callback subprog to callback-calling helper or + * kfunc call. Use idx/subseq_idx check to discern it from + * straight line code backtracking. + * Unlike the subprog call handling above, we shouldn't + * propagate precision of r1-r5 (if any requested), as they are + * not actually arguments passed directly to callback subprogs */ if (bt_reg_mask(bt) & ~BPF_REGMASK_ARGS) { verbose(env, "BUG regs %x\n", bt_reg_mask(bt)); @@ -3950,10 +3961,18 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx, int subseq_idx, } else if (opcode == BPF_EXIT) { bool r0_precise; + /* Backtracking to a nested function call, 'idx' is a part of + * the inner frame 'subseq_idx' is a part of the outer frame. + * In case of a regular function call, instructions giving + * precision to registers R1-R5 should have been found already. + * In case of a callback, it is ok to have R1-R5 marked for + * backtracking, as these registers are set by the function + * invoking callback. + */ + if (subseq_idx >= 0 && calls_callback(env, subseq_idx)) + for (i = BPF_REG_1; i <= BPF_REG_5; i++) + bt_clear_reg(bt, i); if (bt_reg_mask(bt) & BPF_REGMASK_ARGS) { - /* if backtracing was looking for registers R1-R5 - * they should have been found already. - */ verbose(env, "BUG regs %x\n", bt_reg_mask(bt)); WARN_ONCE(1, "verifier backtracking bug"); return -EFAULT; @@ -9421,11 +9440,11 @@ err_out: return err; } -static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn, - int *insn_idx, int subprog, - set_callee_state_fn set_callee_state_cb) +static int push_callback_call(struct bpf_verifier_env *env, struct bpf_insn *insn, + int insn_idx, int subprog, + set_callee_state_fn set_callee_state_cb) { - struct bpf_verifier_state *state = env->cur_state; + struct bpf_verifier_state *state = env->cur_state, *callback_state; struct bpf_func_state *caller, *callee; int err; @@ -9433,44 +9452,22 @@ static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn err = btf_check_subprog_call(env, subprog, caller->regs); if (err == -EFAULT) return err; - if (subprog_is_global(env, subprog)) { - if (err) { - verbose(env, "Caller passes invalid args into func#%d\n", - subprog); - return err; - } else { - if (env->log.level & BPF_LOG_LEVEL) - verbose(env, - "Func#%d is global and valid. Skipping.\n", - subprog); - clear_caller_saved_regs(env, caller->regs); - - /* All global functions return a 64-bit SCALAR_VALUE */ - mark_reg_unknown(env, caller->regs, BPF_REG_0); - caller->regs[BPF_REG_0].subreg_def = DEF_NOT_SUBREG; - - /* continue with next insn after call */ - return 0; - } - } /* set_callee_state is used for direct subprog calls, but we are * interested in validating only BPF helpers that can call subprogs as * callbacks */ - if (set_callee_state_cb != set_callee_state) { - env->subprog_info[subprog].is_cb = true; - if (bpf_pseudo_kfunc_call(insn) && - !is_callback_calling_kfunc(insn->imm)) { - verbose(env, "verifier bug: kfunc %s#%d not marked as callback-calling\n", - func_id_name(insn->imm), insn->imm); - return -EFAULT; - } else if (!bpf_pseudo_kfunc_call(insn) && - !is_callback_calling_function(insn->imm)) { /* helper */ - verbose(env, "verifier bug: helper %s#%d not marked as callback-calling\n", - func_id_name(insn->imm), insn->imm); - return -EFAULT; - } + env->subprog_info[subprog].is_cb = true; + if (bpf_pseudo_kfunc_call(insn) && + !is_sync_callback_calling_kfunc(insn->imm)) { + verbose(env, "verifier bug: kfunc %s#%d not marked as callback-calling\n", + func_id_name(insn->imm), insn->imm); + return -EFAULT; + } else if (!bpf_pseudo_kfunc_call(insn) && + !is_callback_calling_function(insn->imm)) { /* helper */ + verbose(env, "verifier bug: helper %s#%d not marked as callback-calling\n", + func_id_name(insn->imm), insn->imm); + return -EFAULT; } if (insn->code == (BPF_JMP | BPF_CALL) && @@ -9481,25 +9478,76 @@ static int __check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn /* there is no real recursion here. timer callbacks are async */ env->subprog_info[subprog].is_async_cb = true; async_cb = push_async_cb(env, env->subprog_info[subprog].start, - *insn_idx, subprog); + insn_idx, subprog); if (!async_cb) return -EFAULT; callee = async_cb->frame[0]; callee->async_entry_cnt = caller->async_entry_cnt + 1; /* Convert bpf_timer_set_callback() args into timer callback args */ - err = set_callee_state_cb(env, caller, callee, *insn_idx); + err = set_callee_state_cb(env, caller, callee, insn_idx); if (err) return err; + return 0; + } + + /* for callback functions enqueue entry to callback and + * proceed with next instruction within current frame. + */ + callback_state = push_stack(env, env->subprog_info[subprog].start, insn_idx, false); + if (!callback_state) + return -ENOMEM; + + err = setup_func_entry(env, subprog, insn_idx, set_callee_state_cb, + callback_state); + if (err) + return err; + + callback_state->callback_unroll_depth++; + return 0; +} + +static int check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn, + int *insn_idx) +{ + struct bpf_verifier_state *state = env->cur_state; + struct bpf_func_state *caller; + int err, subprog, target_insn; + + target_insn = *insn_idx + insn->imm + 1; + subprog = find_subprog(env, target_insn); + if (subprog < 0) { + verbose(env, "verifier bug. No program starts at insn %d\n", target_insn); + return -EFAULT; + } + + caller = state->frame[state->curframe]; + err = btf_check_subprog_call(env, subprog, caller->regs); + if (err == -EFAULT) + return err; + if (subprog_is_global(env, subprog)) { + if (err) { + verbose(env, "Caller passes invalid args into func#%d\n", subprog); + return err; + } + + if (env->log.level & BPF_LOG_LEVEL) + verbose(env, "Func#%d is global and valid. Skipping.\n", subprog); clear_caller_saved_regs(env, caller->regs); + + /* All global functions return a 64-bit SCALAR_VALUE */ mark_reg_unknown(env, caller->regs, BPF_REG_0); caller->regs[BPF_REG_0].subreg_def = DEF_NOT_SUBREG; + /* continue with next insn after call */ return 0; } - err = setup_func_entry(env, subprog, *insn_idx, set_callee_state_cb, state); + /* for regular function entry setup new frame and continue + * from that frame. + */ + err = setup_func_entry(env, subprog, *insn_idx, set_callee_state, state); if (err) return err; @@ -9559,22 +9607,6 @@ static int set_callee_state(struct bpf_verifier_env *env, return 0; } -static int check_func_call(struct bpf_verifier_env *env, struct bpf_insn *insn, - int *insn_idx) -{ - int subprog, target_insn; - - target_insn = *insn_idx + insn->imm + 1; - subprog = find_subprog(env, target_insn); - if (subprog < 0) { - verbose(env, "verifier bug. No program starts at insn %d\n", - target_insn); - return -EFAULT; - } - - return __check_func_call(env, insn, insn_idx, subprog, set_callee_state); -} - static int set_map_elem_callback_state(struct bpf_verifier_env *env, struct bpf_func_state *caller, struct bpf_func_state *callee, @@ -9798,6 +9830,11 @@ static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) verbose_invalid_scalar(env, r0, &range, "callback return", "R0"); return -EINVAL; } + if (!calls_callback(env, callee->callsite)) { + verbose(env, "BUG: in callback at %d, callsite %d !calls_callback\n", + *insn_idx, callee->callsite); + return -EFAULT; + } } else { /* return to the caller whatever r0 had in the callee */ caller->regs[BPF_REG_0] = *r0; @@ -9815,7 +9852,15 @@ static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) return err; } - *insn_idx = callee->callsite + 1; + /* for callbacks like bpf_loop or bpf_for_each_map_elem go back to callsite, + * there function call logic would reschedule callback visit. If iteration + * converges is_state_visited() would prune that visit eventually. + */ + if (callee->in_callback_fn) + *insn_idx = callee->callsite; + else + *insn_idx = callee->callsite + 1; + if (env->log.level & BPF_LOG_LEVEL) { verbose(env, "returning from callee:\n"); print_verifier_state(env, callee, true); @@ -10228,24 +10273,24 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn } break; case BPF_FUNC_for_each_map_elem: - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_map_elem_callback_state); + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_map_elem_callback_state); break; case BPF_FUNC_timer_set_callback: - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_timer_callback_state); + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_timer_callback_state); break; case BPF_FUNC_find_vma: - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_find_vma_callback_state); + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_find_vma_callback_state); break; case BPF_FUNC_snprintf: err = check_bpf_snprintf_call(env, regs); break; case BPF_FUNC_loop: update_loop_inline_state(env, meta.subprogno); - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_loop_callback_state); + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_loop_callback_state); break; case BPF_FUNC_dynptr_from_mem: if (regs[BPF_REG_1].type != PTR_TO_MAP_VALUE) { @@ -10341,8 +10386,8 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn break; } case BPF_FUNC_user_ringbuf_drain: - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_user_ringbuf_callback_state); + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_user_ringbuf_callback_state); break; } @@ -11230,7 +11275,7 @@ static bool is_bpf_graph_api_kfunc(u32 btf_id) btf_id == special_kfunc_list[KF_bpf_refcount_acquire_impl]; } -static bool is_callback_calling_kfunc(u32 btf_id) +static bool is_sync_callback_calling_kfunc(u32 btf_id) { return btf_id == special_kfunc_list[KF_bpf_rbtree_add_impl]; } @@ -11982,6 +12027,21 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, return -EACCES; } + /* Check the arguments */ + err = check_kfunc_args(env, &meta, insn_idx); + if (err < 0) + return err; + + if (meta.func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) { + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_rbtree_add_callback_state); + if (err) { + verbose(env, "kfunc %s#%d failed callback verification\n", + func_name, meta.func_id); + return err; + } + } + rcu_lock = is_kfunc_bpf_rcu_read_lock(&meta); rcu_unlock = is_kfunc_bpf_rcu_read_unlock(&meta); @@ -12017,10 +12077,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, return -EINVAL; } - /* Check the arguments */ - err = check_kfunc_args(env, &meta, insn_idx); - if (err < 0) - return err; /* In case of release function, we get register number of refcounted * PTR_TO_BTF_ID in bpf_kfunc_arg_meta, do the release now. */ @@ -12054,16 +12110,6 @@ static int check_kfunc_call(struct bpf_verifier_env *env, struct bpf_insn *insn, } } - if (meta.func_id == special_kfunc_list[KF_bpf_rbtree_add_impl]) { - err = __check_func_call(env, insn, insn_idx_p, meta.subprogno, - set_rbtree_add_callback_state); - if (err) { - verbose(env, "kfunc %s#%d failed callback verification\n", - func_name, meta.func_id); - return err; - } - } - if (meta.func_id == special_kfunc_list[KF_bpf_throw]) { if (!bpf_jit_supports_exceptions()) { verbose(env, "JIT does not support calling kfunc %s#%d\n", @@ -15427,6 +15473,15 @@ static bool is_force_checkpoint(struct bpf_verifier_env *env, int insn_idx) return env->insn_aux_data[insn_idx].force_checkpoint; } +static void mark_calls_callback(struct bpf_verifier_env *env, int idx) +{ + env->insn_aux_data[idx].calls_callback = true; +} + +static bool calls_callback(struct bpf_verifier_env *env, int insn_idx) +{ + return env->insn_aux_data[insn_idx].calls_callback; +} enum { DONE_EXPLORING = 0, @@ -15540,6 +15595,21 @@ static int visit_insn(int t, struct bpf_verifier_env *env) * async state will be pushed for further exploration. */ mark_prune_point(env, t); + /* For functions that invoke callbacks it is not known how many times + * callback would be called. Verifier models callback calling functions + * by repeatedly visiting callback bodies and returning to origin call + * instruction. + * In order to stop such iteration verifier needs to identify when a + * state identical some state from a previous iteration is reached. + * Check below forces creation of checkpoint before callback calling + * instruction to allow search for such identical states. + */ + if (is_sync_callback_calling_insn(insn)) { + mark_calls_callback(env, t); + mark_force_checkpoint(env, t); + mark_prune_point(env, t); + mark_jmp_point(env, t); + } if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL) { struct bpf_kfunc_call_arg_meta meta; @@ -17009,10 +17079,16 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx) } goto skip_inf_loop_check; } + if (calls_callback(env, insn_idx)) { + if (states_equal(env, &sl->state, cur, true)) + goto hit; + goto skip_inf_loop_check; + } /* attempt to detect infinite loop to avoid unnecessary doomed work */ if (states_maybe_looping(&sl->state, cur) && states_equal(env, &sl->state, cur, false) && - !iter_active_depths_differ(&sl->state, cur)) { + !iter_active_depths_differ(&sl->state, cur) && + sl->state.callback_unroll_depth == cur->callback_unroll_depth) { verbose_linfo(env, insn_idx, "; "); verbose(env, "infinite loop detected at insn %d\n", insn_idx); verbose(env, "cur state:"); -- cgit v1.2.3 From cafe2c21508a38cdb3ed22708842e957b2572c3e Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 21 Nov 2023 04:06:58 +0200 Subject: bpf: widening for callback iterators Callbacks are similar to open coded iterators, so add imprecise widening logic for callback body processing. This makes callback based loops behave identically to open coded iterators, e.g. allowing to verify programs like below: struct ctx { u32 i; }; int cb(u32 idx, struct ctx* ctx) { ++ctx->i; return 0; } ... struct ctx ctx = { .i = 0 }; bpf_loop(100, cb, &ctx, 0); ... Acked-by: Andrii Nakryiko Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20231121020701.26440-9-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index a60dfa56ebb3..2f03e6b11bb9 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9799,9 +9799,10 @@ static bool in_rbtree_lock_required_cb(struct bpf_verifier_env *env) static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) { - struct bpf_verifier_state *state = env->cur_state; + struct bpf_verifier_state *state = env->cur_state, *prev_st; struct bpf_func_state *caller, *callee; struct bpf_reg_state *r0; + bool in_callback_fn; int err; callee = state->frame[state->curframe]; @@ -9856,7 +9857,8 @@ static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) * there function call logic would reschedule callback visit. If iteration * converges is_state_visited() would prune that visit eventually. */ - if (callee->in_callback_fn) + in_callback_fn = callee->in_callback_fn; + if (in_callback_fn) *insn_idx = callee->callsite; else *insn_idx = callee->callsite + 1; @@ -9871,6 +9873,24 @@ static int prepare_func_exit(struct bpf_verifier_env *env, int *insn_idx) * bpf_throw, this will be done by copy_verifier_state for extra frames. */ free_func_state(callee); state->frame[state->curframe--] = NULL; + + /* for callbacks widen imprecise scalars to make programs like below verify: + * + * struct ctx { int i; } + * void cb(int idx, struct ctx *ctx) { ctx->i++; ... } + * ... + * struct ctx = { .i = 0; } + * bpf_loop(100, cb, &ctx, 0); + * + * This is similar to what is done in process_iter_next_call() for open + * coded iterators. + */ + prev_st = in_callback_fn ? find_prev_entry(env, state, *insn_idx) : NULL; + if (prev_st) { + err = widen_imprecise_scalars(env, prev_st, state); + if (err) + return err; + } return 0; } -- cgit v1.2.3 From bb124da69c47dd98d69361ec13244ece50bec63e Mon Sep 17 00:00:00 2001 From: Eduard Zingerman Date: Tue, 21 Nov 2023 04:07:00 +0200 Subject: bpf: keep track of max number of bpf_loop callback iterations In some cases verifier can't infer convergence of the bpf_loop() iteration. E.g. for the following program: static int cb(__u32 idx, struct num_context* ctx) { ctx->i++; return 0; } SEC("?raw_tp") int prog(void *_) { struct num_context ctx = { .i = 0 }; __u8 choice_arr[2] = { 0, 1 }; bpf_loop(2, cb, &ctx, 0); return choice_arr[ctx.i]; } Each 'cb' simulation would eventually return to 'prog' and reach 'return choice_arr[ctx.i]' statement. At which point ctx.i would be marked precise, thus forcing verifier to track multitude of separate states with {.i=0}, {.i=1}, ... at bpf_loop() callback entry. This commit allows "brute force" handling for such cases by limiting number of callback body simulations using 'umax' value of the first bpf_loop() parameter. For this, extend bpf_func_state with 'callback_depth' field. Increment this field when callback visiting state is pushed to states traversal stack. For frame #N it's 'callback_depth' field counts how many times callback with frame depth N+1 had been executed. Use bpf_func_state specifically to allow independent tracking of callback depths when multiple nested bpf_loop() calls are present. Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20231121020701.26440-11-eddyz87@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 2f03e6b11bb9..af2819d5c8ee 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -9505,6 +9505,8 @@ static int push_callback_call(struct bpf_verifier_env *env, struct bpf_insn *ins return err; callback_state->callback_unroll_depth++; + callback_state->frame[callback_state->curframe - 1]->callback_depth++; + caller->callback_depth = 0; return 0; } @@ -10309,8 +10311,21 @@ static int check_helper_call(struct bpf_verifier_env *env, struct bpf_insn *insn break; case BPF_FUNC_loop: update_loop_inline_state(env, meta.subprogno); - err = push_callback_call(env, insn, insn_idx, meta.subprogno, - set_loop_callback_state); + /* Verifier relies on R1 value to determine if bpf_loop() iteration + * is finished, thus mark it precise. + */ + err = mark_chain_precision(env, BPF_REG_1); + if (err) + return err; + if (cur_func(env)->callback_depth < regs[BPF_REG_1].umax_value) { + err = push_callback_call(env, insn, insn_idx, meta.subprogno, + set_loop_callback_state); + } else { + cur_func(env)->callback_depth = 0; + if (env->log.level & BPF_LOG_LEVEL2) + verbose(env, "frame%d bpf_loop iteration limit reached\n", + env->cur_state->curframe); + } break; case BPF_FUNC_dynptr_from_mem: if (regs[BPF_REG_1].type != PTR_TO_MAP_VALUE) { -- cgit v1.2.3 From 4a6c5607d4502ccd1b15b57d57f17d12b6f257a7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 21 Nov 2023 11:39:36 -1000 Subject: workqueue: Make sure that wq_unbound_cpumask is never empty During boot, depending on how the housekeeping and workqueue.unbound_cpus masks are set, wq_unbound_cpumask can end up empty. Since 8639ecebc9b1 ("workqueue: Implement non-strict affinity scope for unbound workqueues"), this may end up feeding -1 as a CPU number into scheduler leading to oopses. BUG: unable to handle page fault for address: ffffffff8305e9c0 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page ... Call Trace: select_idle_sibling+0x79/0xaf0 select_task_rq_fair+0x1cb/0x7b0 try_to_wake_up+0x29c/0x5c0 wake_up_process+0x19/0x20 kick_pool+0x5e/0xb0 __queue_work+0x119/0x430 queue_work_on+0x29/0x30 ... An empty wq_unbound_cpumask is a clear misconfiguration and already disallowed once system is booted up. Let's warn on and ignore unbound_cpumask restrictions which lead to no unbound cpus. While at it, also remove now unncessary empty check on wq_unbound_cpumask in wq_select_unbound_cpu(). Signed-off-by: Tejun Heo Reported-and-Tested-by: Yong He Link: http://lkml.kernel.org/r/20231120121623.119780-1-alexyonghe@tencent.com Fixes: 8639ecebc9b1 ("workqueue: Implement non-strict affinity scope for unbound workqueues") Cc: stable@vger.kernel.org # v6.6+ Reviewed-by: Waiman Long --- kernel/workqueue.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 6e578f576a6f..2989b57e154a 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1684,9 +1684,6 @@ static int wq_select_unbound_cpu(int cpu) pr_warn_once("workqueue: round-robin CPU selection forced, expect performance impact\n"); } - if (cpumask_empty(wq_unbound_cpumask)) - return cpu; - new_cpu = __this_cpu_read(wq_rr_cpu_last); new_cpu = cpumask_next_and(new_cpu, wq_unbound_cpumask, cpu_online_mask); if (unlikely(new_cpu >= nr_cpu_ids)) { @@ -6515,6 +6512,17 @@ static inline void wq_watchdog_init(void) { } #endif /* CONFIG_WQ_WATCHDOG */ +static void __init restrict_unbound_cpumask(const char *name, const struct cpumask *mask) +{ + if (!cpumask_intersects(wq_unbound_cpumask, mask)) { + pr_warn("workqueue: Restricting unbound_cpumask (%*pb) with %s (%*pb) leaves no CPU, ignoring\n", + cpumask_pr_args(wq_unbound_cpumask), name, cpumask_pr_args(mask)); + return; + } + + cpumask_and(wq_unbound_cpumask, wq_unbound_cpumask, mask); +} + /** * workqueue_init_early - early init for workqueue subsystem * @@ -6534,11 +6542,11 @@ void __init workqueue_init_early(void) BUILD_BUG_ON(__alignof__(struct pool_workqueue) < __alignof__(long long)); BUG_ON(!alloc_cpumask_var(&wq_unbound_cpumask, GFP_KERNEL)); - cpumask_copy(wq_unbound_cpumask, housekeeping_cpumask(HK_TYPE_WQ)); - cpumask_and(wq_unbound_cpumask, wq_unbound_cpumask, housekeeping_cpumask(HK_TYPE_DOMAIN)); - + cpumask_copy(wq_unbound_cpumask, cpu_possible_mask); + restrict_unbound_cpumask("HK_TYPE_WQ", housekeeping_cpumask(HK_TYPE_WQ)); + restrict_unbound_cpumask("HK_TYPE_DOMAIN", housekeeping_cpumask(HK_TYPE_DOMAIN)); if (!cpumask_empty(&wq_cmdline_cpumask)) - cpumask_and(wq_unbound_cpumask, wq_unbound_cpumask, &wq_cmdline_cpumask); + restrict_unbound_cpumask("workqueue.unbound_cpus", &wq_cmdline_cpumask); pwq_cache = KMEM_CACHE(pool_workqueue, SLAB_PANIC); -- cgit v1.2.3 From bca4104b00fec60be330cd32818dd5c70db3d469 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 21 Nov 2023 12:41:26 +0100 Subject: lockdep: Fix block chain corruption Kent reported an occasional KASAN splat in lockdep. Mark then noted: > I suspect the dodgy access is to chain_block_buckets[-1], which hits the last 4 > bytes of the redzone and gets (incorrectly/misleadingly) attributed to > nr_large_chain_blocks. That would mean @size == 0, at which point size_to_bucket() returns -1 and the above happens. alloc_chain_hlocks() has 'size - req', for the first with the precondition 'size >= rq', which allows the 0. This code is trying to split a block, del_chain_block() takes what we need, and add_chain_block() puts back the remainder, except in the above case the remainder is 0 sized and things go sideways. Fixes: 810507fe6fd5 ("locking/lockdep: Reuse freed chain_hlocks entries") Reported-by: Kent Overstreet Signed-off-by: Peter Zijlstra (Intel) Tested-by: Kent Overstreet Link: https://lkml.kernel.org/r/20231121114126.GH8262@noisy.programming.kicks-ass.net --- kernel/locking/lockdep.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index e85b5ad3e206..151bd3de5936 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -3497,7 +3497,8 @@ static int alloc_chain_hlocks(int req) size = chain_block_size(curr); if (likely(size >= req)) { del_chain_block(0, size, chain_block_next(curr)); - add_chain_block(curr + req, size - req); + if (size > req) + add_chain_block(curr + req, size - req); return curr; } } -- cgit v1.2.3 From 75a442581d05edaee168222ffbe00d4389785636 Mon Sep 17 00:00:00 2001 From: Hou Tao Date: Sat, 11 Nov 2023 12:38:21 +0800 Subject: bpf: Add missed allocation hint for bpf_mem_cache_alloc_flags() bpf_mem_cache_alloc_flags() may call __alloc() directly when there is no free object in free list, but it doesn't initialize the allocation hint for the returned pointer. It may lead to bad memory dereference when freeing the pointer, so fix it by initializing the allocation hint. Fixes: 822fb26bdb55 ("bpf: Add a hint to allocated objects.") Signed-off-by: Hou Tao Acked-by: Yonghong Song Link: https://lore.kernel.org/r/20231111043821.2258513-1-houtao@huaweicloud.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/memalloc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/bpf/memalloc.c b/kernel/bpf/memalloc.c index 63b909d277d4..6a51cfe4c2d6 100644 --- a/kernel/bpf/memalloc.c +++ b/kernel/bpf/memalloc.c @@ -978,6 +978,8 @@ void notrace *bpf_mem_cache_alloc_flags(struct bpf_mem_alloc *ma, gfp_t flags) memcg = get_memcg(c); old_memcg = set_active_memcg(memcg); ret = __alloc(c, NUMA_NO_NODE, GFP_KERNEL | __GFP_NOWARN | __GFP_ACCOUNT); + if (ret) + *(struct bpf_mem_cache **)ret = c; set_active_memcg(old_memcg); mem_cgroup_put(memcg); } -- cgit v1.2.3 From cff5f49d433fcd0063c8be7dd08fa5bf190c6c37 Mon Sep 17 00:00:00 2001 From: Tim Van Patten Date: Wed, 15 Nov 2023 09:20:43 -0700 Subject: cgroup_freezer: cgroup_freezing: Check if not frozen __thaw_task() was recently updated to warn if the task being thawed was part of a freezer cgroup that is still currently freezing: void __thaw_task(struct task_struct *p) { ... if (WARN_ON_ONCE(freezing(p))) goto unlock; This has exposed a bug in cgroup1 freezing where when CGROUP_FROZEN is asserted, the CGROUP_FREEZING bits are not also cleared at the same time. Meaning, when a cgroup is marked FROZEN it continues to be marked FREEZING as well. This causes the WARNING to trigger, because cgroup_freezing() thinks the cgroup is still freezing. There are two ways to fix this: 1. Whenever FROZEN is set, clear FREEZING for the cgroup and all children cgroups. 2. Update cgroup_freezing() to also verify that FROZEN is not set. This patch implements option (2), since it's smaller and more straightforward. Signed-off-by: Tim Van Patten Tested-by: Mark Hasemeyer Fixes: f5d39b020809 ("freezer,sched: Rewrite core freezer logic") Cc: stable@vger.kernel.org # v6.1+ Signed-off-by: Tejun Heo --- kernel/cgroup/legacy_freezer.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c index 122dacb3a443..66d1708042a7 100644 --- a/kernel/cgroup/legacy_freezer.c +++ b/kernel/cgroup/legacy_freezer.c @@ -66,9 +66,15 @@ static struct freezer *parent_freezer(struct freezer *freezer) bool cgroup_freezing(struct task_struct *task) { bool ret; + unsigned int state; rcu_read_lock(); - ret = task_freezer(task)->state & CGROUP_FREEZING; + /* Check if the cgroup is still FREEZING, but not FROZEN. The extra + * !FROZEN check is required, because the FREEZING bit is not cleared + * when the state FROZEN is reached. + */ + state = task_freezer(task)->state; + ret = (state & CGROUP_FREEZING) && !(state & CGROUP_FROZEN); rcu_read_unlock(); return ret; -- cgit v1.2.3 From 23ab79e8e469e2605beec2e3ccb40d19c68dd2e0 Mon Sep 17 00:00:00 2001 From: Elliot Berman Date: Mon, 20 Nov 2023 09:36:31 -0800 Subject: freezer,sched: Do not restore saved_state of a thawed task It is possible for a task to be thawed multiple times when mixing the *legacy* cgroup freezer and system-wide freezer. To do this, freeze the cgroup, do system-wide freeze/thaw, then thaw the cgroup. When this happens, then a stale saved_state can be written to the task's state and cause task to hang indefinitely. Fix this by only trying to thaw tasks that are actually frozen. This change also has the marginal benefit avoiding unnecessary wake_up_state(p, TASK_FROZEN) if we know the task is already thawed. There is not possibility of time-of-compare/time-of-use race when we skip the wake_up_state because entering/exiting TASK_FROZEN is guarded by freezer_lock. Fixes: 8f0eed4a78a8 ("freezer,sched: Use saved_state to reduce some spurious wakeups") Signed-off-by: Elliot Berman Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Abhijeet Dharmapurikar Link: https://lore.kernel.org/r/20231120-freezer-state-multiple-thaws-v1-1-f2e1dd7ce5a2@quicinc.com --- kernel/freezer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/freezer.c b/kernel/freezer.c index c450fa8b8b5e..759006a9a910 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -201,7 +201,7 @@ void __thaw_task(struct task_struct *p) if (WARN_ON_ONCE(freezing(p))) goto unlock; - if (task_call_func(p, __restore_freezer_state, NULL)) + if (!frozen(p) || task_call_func(p, __restore_freezer_state, NULL)) goto unlock; wake_up_state(p, TASK_FROZEN); -- cgit v1.2.3 From 382c27f4ed28f803b1f1473ac2d8db0afc795a1b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 29 Nov 2023 15:24:52 +0100 Subject: perf: Fix perf_event_validate_size() Budimir noted that perf_event_validate_size() only checks the size of the newly added event, even though the sizes of all existing events can also change due to not all events having the same read_format. When we attach the new event, perf_group_attach(), we do re-compute the size for all events. Fixes: a723968c0ed3 ("perf: Fix u16 overflows") Reported-by: Budimir Markovic Signed-off-by: Peter Zijlstra (Intel) --- kernel/events/core.c | 61 ++++++++++++++++++++++++++++++++-------------------- 1 file changed, 38 insertions(+), 23 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index b704d83a28b2..c9d123e13b57 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1814,31 +1814,34 @@ static inline void perf_event__state_init(struct perf_event *event) PERF_EVENT_STATE_INACTIVE; } -static void __perf_event_read_size(struct perf_event *event, int nr_siblings) +static int __perf_event_read_size(u64 read_format, int nr_siblings) { int entry = sizeof(u64); /* value */ int size = 0; int nr = 1; - if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) size += sizeof(u64); - if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) size += sizeof(u64); - if (event->attr.read_format & PERF_FORMAT_ID) + if (read_format & PERF_FORMAT_ID) entry += sizeof(u64); - if (event->attr.read_format & PERF_FORMAT_LOST) + if (read_format & PERF_FORMAT_LOST) entry += sizeof(u64); - if (event->attr.read_format & PERF_FORMAT_GROUP) { + if (read_format & PERF_FORMAT_GROUP) { nr += nr_siblings; size += sizeof(u64); } - size += entry * nr; - event->read_size = size; + /* + * Since perf_event_validate_size() limits this to 16k and inhibits + * adding more siblings, this will never overflow. + */ + return size + nr * entry; } static void __perf_event_header_size(struct perf_event *event, u64 sample_type) @@ -1888,8 +1891,9 @@ static void __perf_event_header_size(struct perf_event *event, u64 sample_type) */ static void perf_event__header_size(struct perf_event *event) { - __perf_event_read_size(event, - event->group_leader->nr_siblings); + event->read_size = + __perf_event_read_size(event->attr.read_format, + event->group_leader->nr_siblings); __perf_event_header_size(event, event->attr.sample_type); } @@ -1920,24 +1924,35 @@ static void perf_event__id_header_size(struct perf_event *event) event->id_header_size = size; } +/* + * Check that adding an event to the group does not result in anybody + * overflowing the 64k event limit imposed by the output buffer. + * + * Specifically, check that the read_size for the event does not exceed 16k, + * read_size being the one term that grows with groups size. Since read_size + * depends on per-event read_format, also (re)check the existing events. + * + * This leaves 48k for the constant size fields and things like callchains, + * branch stacks and register sets. + */ static bool perf_event_validate_size(struct perf_event *event) { - /* - * The values computed here will be over-written when we actually - * attach the event. - */ - __perf_event_read_size(event, event->group_leader->nr_siblings + 1); - __perf_event_header_size(event, event->attr.sample_type & ~PERF_SAMPLE_READ); - perf_event__id_header_size(event); + struct perf_event *sibling, *group_leader = event->group_leader; - /* - * Sum the lot; should not exceed the 64k limit we have on records. - * Conservative limit to allow for callchains and other variable fields. - */ - if (event->read_size + event->header_size + - event->id_header_size + sizeof(struct perf_event_header) >= 16*1024) + if (__perf_event_read_size(event->attr.read_format, + group_leader->nr_siblings + 1) > 16*1024) return false; + if (__perf_event_read_size(group_leader->attr.read_format, + group_leader->nr_siblings + 1) > 16*1024) + return false; + + for_each_sibling_event(sibling, group_leader) { + if (__perf_event_read_size(sibling->attr.read_format, + group_leader->nr_siblings + 1) > 16*1024) + return false; + } + return true; } -- cgit v1.2.3 From d839a656d0f3caca9f96e9bf912fd394ac6a11bc Mon Sep 17 00:00:00 2001 From: JP Kobryn Date: Fri, 1 Dec 2023 14:53:55 +0900 Subject: kprobes: consistent rcu api usage for kretprobe holder It seems that the pointer-to-kretprobe "rp" within the kretprobe_holder is RCU-managed, based on the (non-rethook) implementation of get_kretprobe(). The thought behind this patch is to make use of the RCU API where possible when accessing this pointer so that the needed barriers are always in place and to self-document the code. The __rcu annotation to "rp" allows for sparse RCU checking. Plain writes done to the "rp" pointer are changed to make use of the RCU macro for assignment. For the single read, the implementation of get_kretprobe() is simplified by making use of an RCU macro which accomplishes the same, but note that the log warning text will be more generic. I did find that there is a difference in assembly generated between the usage of the RCU macros vs without. For example, on arm64, when using rcu_assign_pointer(), the corresponding store instruction is a store-release (STLR) which has an implicit barrier. When normal assignment is done, a regular store (STR) is found. In the macro case, this seems to be a result of rcu_assign_pointer() using smp_store_release() when the value to write is not NULL. Link: https://lore.kernel.org/all/20231122132058.3359-1-inwardvessel@gmail.com/ Fixes: d741bf41d7c7 ("kprobes: Remove kretprobe hash") Cc: stable@vger.kernel.org Signed-off-by: JP Kobryn Acked-by: Masami Hiramatsu (Google) Signed-off-by: Masami Hiramatsu (Google) --- kernel/kprobes.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 075a632e6c7c..d5a0ee40bf66 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -2252,7 +2252,7 @@ int register_kretprobe(struct kretprobe *rp) rp->rph = NULL; return -ENOMEM; } - rp->rph->rp = rp; + rcu_assign_pointer(rp->rph->rp, rp); rp->nmissed = 0; /* Establish function entry probe point */ ret = register_kprobe(&rp->kp); @@ -2300,7 +2300,7 @@ void unregister_kretprobes(struct kretprobe **rps, int num) #ifdef CONFIG_KRETPROBE_ON_RETHOOK rethook_free(rps[i]->rh); #else - rps[i]->rph->rp = NULL; + rcu_assign_pointer(rps[i]->rph->rp, NULL); #endif } mutex_unlock(&kprobe_mutex); -- cgit v1.2.3 From a1461f1fd6cfdc4b8917c9d4a91e92605d1f28dc Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Fri, 1 Dec 2023 14:53:56 +0900 Subject: rethook: Use __rcu pointer for rethook::handler Since the rethook::handler is an RCU-maganged pointer so that it will notice readers the rethook is stopped (unregistered) or not, it should be an __rcu pointer and use appropriate functions to be accessed. This will use appropriate memory barrier when accessing it. OTOH, rethook::data is never changed, so we don't need to check it in get_kretprobe(). NOTE: To avoid sparse warning, rethook::handler is defined by a raw function pointer type with __rcu instead of rethook_handler_t. Link: https://lore.kernel.org/all/170126066201.398836.837498688669005979.stgit@devnote2/ Fixes: 54ecbe6f1ed5 ("rethook: Add a generic return hook") Cc: stable@vger.kernel.org Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202311241808.rv9ceuAh-lkp@intel.com/ Tested-by: JP Kobryn Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/rethook.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/rethook.c b/kernel/trace/rethook.c index 6fd7d4ecbbc6..fa03094e9e69 100644 --- a/kernel/trace/rethook.c +++ b/kernel/trace/rethook.c @@ -48,7 +48,7 @@ static void rethook_free_rcu(struct rcu_head *head) */ void rethook_stop(struct rethook *rh) { - WRITE_ONCE(rh->handler, NULL); + rcu_assign_pointer(rh->handler, NULL); } /** @@ -63,7 +63,7 @@ void rethook_stop(struct rethook *rh) */ void rethook_free(struct rethook *rh) { - WRITE_ONCE(rh->handler, NULL); + rethook_stop(rh); call_rcu(&rh->rcu, rethook_free_rcu); } @@ -82,6 +82,12 @@ static int rethook_fini_pool(struct objpool_head *head, void *context) return 0; } +static inline rethook_handler_t rethook_get_handler(struct rethook *rh) +{ + return (rethook_handler_t)rcu_dereference_check(rh->handler, + rcu_read_lock_any_held()); +} + /** * rethook_alloc() - Allocate struct rethook. * @data: a data to pass the @handler when hooking the return. @@ -107,7 +113,7 @@ struct rethook *rethook_alloc(void *data, rethook_handler_t handler, return ERR_PTR(-ENOMEM); rh->data = data; - rh->handler = handler; + rcu_assign_pointer(rh->handler, handler); /* initialize the objpool for rethook nodes */ if (objpool_init(&rh->pool, num, size, GFP_KERNEL, rh, @@ -135,9 +141,10 @@ static void free_rethook_node_rcu(struct rcu_head *head) */ void rethook_recycle(struct rethook_node *node) { - lockdep_assert_preemption_disabled(); + rethook_handler_t handler; - if (likely(READ_ONCE(node->rethook->handler))) + handler = rethook_get_handler(node->rethook); + if (likely(handler)) objpool_push(node, &node->rethook->pool); else call_rcu(&node->rcu, free_rethook_node_rcu); @@ -153,9 +160,7 @@ NOKPROBE_SYMBOL(rethook_recycle); */ struct rethook_node *rethook_try_get(struct rethook *rh) { - rethook_handler_t handler = READ_ONCE(rh->handler); - - lockdep_assert_preemption_disabled(); + rethook_handler_t handler = rethook_get_handler(rh); /* Check whether @rh is going to be freed. */ if (unlikely(!handler)) @@ -300,7 +305,7 @@ unsigned long rethook_trampoline_handler(struct pt_regs *regs, rhn = container_of(first, struct rethook_node, llist); if (WARN_ON_ONCE(rhn->frame != frame)) break; - handler = READ_ONCE(rhn->rethook->handler); + handler = rethook_get_handler(rhn->rethook); if (handler) handler(rhn, rhn->rethook->data, correct_ret_addr, regs); -- cgit v1.2.3 From dfce9cb3140592b886838e06f3e0c25fea2a9cae Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 30 Nov 2023 18:46:40 -0800 Subject: bpf: Fix a verifier bug due to incorrect branch offset comparison with cpu=v4 Bpf cpu=v4 support is introduced in [1] and Commit 4cd58e9af8b9 ("bpf: Support new 32bit offset jmp instruction") added support for new 32bit offset jmp instruction. Unfortunately, in function bpf_adj_delta_to_off(), for new branch insn with 32bit offset, the offset (plus/minor a small delta) compares to 16-bit offset bound [S16_MIN, S16_MAX], which caused the following verification failure: $ ./test_progs-cpuv4 -t verif_scale_pyperf180 ... insn 10 cannot be patched due to 16-bit range ... libbpf: failed to load object 'pyperf180.bpf.o' scale_test:FAIL:expect_success unexpected error: -12 (errno 12) #405 verif_scale_pyperf180:FAIL Note that due to recent llvm18 development, the patch [2] (already applied in bpf-next) needs to be applied to bpf tree for testing purpose. The fix is rather simple. For 32bit offset branch insn, the adjusted offset compares to [S32_MIN, S32_MAX] and then verification succeeded. [1] https://lore.kernel.org/all/20230728011143.3710005-1-yonghong.song@linux.dev [2] https://lore.kernel.org/bpf/20231110193644.3130906-1-yonghong.song@linux.dev Fixes: 4cd58e9af8b9 ("bpf: Support new 32bit offset jmp instruction") Signed-off-by: Yonghong Song Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20231201024640.3417057-1-yonghong.song@linux.dev --- kernel/bpf/core.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index cd3afe57ece3..fe254ae035fe 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -371,14 +371,18 @@ static int bpf_adj_delta_to_imm(struct bpf_insn *insn, u32 pos, s32 end_old, static int bpf_adj_delta_to_off(struct bpf_insn *insn, u32 pos, s32 end_old, s32 end_new, s32 curr, const bool probe_pass) { - const s32 off_min = S16_MIN, off_max = S16_MAX; + s64 off_min, off_max, off; s32 delta = end_new - end_old; - s32 off; - if (insn->code == (BPF_JMP32 | BPF_JA)) + if (insn->code == (BPF_JMP32 | BPF_JA)) { off = insn->imm; - else + off_min = S32_MIN; + off_max = S32_MAX; + } else { off = insn->off; + off_min = S16_MIN; + off_max = S16_MAX; + } if (curr < pos && curr + off + 1 >= end_old) off += delta; -- cgit v1.2.3 From 7be76461f302ec05cbd62b90b2a05c64299ca01f Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 5 Dec 2023 16:52:09 -0500 Subject: tracing: Always update snapshot buffer size It use to be that only the top level instance had a snapshot buffer (for latency tracers like wakeup and irqsoff). The update of the ring buffer size would check if the instance was the top level and if so, it would also update the snapshot buffer as it needs to be the same as the main buffer. Now that lower level instances also has a snapshot buffer, they too need to update their snapshot buffer sizes when the main buffer is changed, otherwise the following can be triggered: # cd /sys/kernel/tracing # echo 1500 > buffer_size_kb # mkdir instances/foo # echo irqsoff > instances/foo/current_tracer # echo 1000 > instances/foo/buffer_size_kb Produces: WARNING: CPU: 2 PID: 856 at kernel/trace/trace.c:1938 update_max_tr_single.part.0+0x27d/0x320 Which is: ret = ring_buffer_swap_cpu(tr->max_buffer.buffer, tr->array_buffer.buffer, cpu); if (ret == -EBUSY) { [..] } WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY); <== here That's because ring_buffer_swap_cpu() has: int ret = -EINVAL; [..] /* At least make sure the two buffers are somewhat the same */ if (cpu_buffer_a->nr_pages != cpu_buffer_b->nr_pages) goto out; [..] out: return ret; } Instead, update all instances' snapshot buffer sizes when their main buffer size is updated. Link: https://lkml.kernel.org/r/20231205220010.454662151@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 6d9b3fa5e7f6 ("tracing: Move tracing_max_latency into trace_array") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 9aebf904ff97..231c173ec04f 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6392,8 +6392,7 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, return ret; #ifdef CONFIG_TRACER_MAX_TRACE - if (!(tr->flags & TRACE_ARRAY_FL_GLOBAL) || - !tr->current_trace->use_max_tr) + if (!tr->current_trace->use_max_tr) goto out; ret = ring_buffer_resize(tr->max_buffer.buffer, size, cpu); -- cgit v1.2.3 From d78ab792705c7be1b91243b2544d1a79406a2ad7 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 5 Dec 2023 16:52:10 -0500 Subject: tracing: Stop current tracer when resizing buffer When the ring buffer is being resized, it can cause side effects to the running tracer. For instance, there's a race with irqsoff tracer that swaps individual per cpu buffers between the main buffer and the snapshot buffer. The resize operation modifies the main buffer and then the snapshot buffer. If a swap happens in between those two operations it will break the tracer. Simply stop the running tracer before resizing the buffers and enable it again when finished. Link: https://lkml.kernel.org/r/20231205220010.748996423@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 3928a8a2d9808 ("ftrace: make work with new ring buffer") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 231c173ec04f..e978868b1a22 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6387,9 +6387,12 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, if (!tr->array_buffer.buffer) return 0; + /* Do not allow tracing while resizng ring buffer */ + tracing_stop_tr(tr); + ret = ring_buffer_resize(tr->array_buffer.buffer, size, cpu); if (ret < 0) - return ret; + goto out_start; #ifdef CONFIG_TRACER_MAX_TRACE if (!tr->current_trace->use_max_tr) @@ -6417,7 +6420,7 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, WARN_ON(1); tracing_disabled = 1; } - return ret; + goto out_start; } update_buffer_entries(&tr->max_buffer, cpu); @@ -6426,7 +6429,8 @@ static int __tracing_resize_ring_buffer(struct trace_array *tr, #endif /* CONFIG_TRACER_MAX_TRACE */ update_buffer_entries(&tr->array_buffer, cpu); - + out_start: + tracing_start_tr(tr); return ret; } -- cgit v1.2.3 From b538bf7d0ec11ca49f536dfda742a5f6db90a798 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Tue, 5 Dec 2023 16:52:11 -0500 Subject: tracing: Disable snapshot buffer when stopping instance tracers It use to be that only the top level instance had a snapshot buffer (for latency tracers like wakeup and irqsoff). When stopping a tracer in an instance would not disable the snapshot buffer. This could have some unintended consequences if the irqsoff tracer is enabled. Consolidate the tracing_start/stop() with tracing_start/stop_tr() so that all instances behave the same. The tracing_start/stop() functions will just call their respective tracing_start/stop_tr() with the global_array passed in. Link: https://lkml.kernel.org/r/20231205220011.041220035@goodmis.org Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Cc: Andrew Morton Fixes: 6d9b3fa5e7f6 ("tracing: Move tracing_max_latency into trace_array") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 110 ++++++++++++++++----------------------------------- 1 file changed, 34 insertions(+), 76 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index e978868b1a22..2492c6c76850 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2360,13 +2360,7 @@ int is_tracing_stopped(void) return global_trace.stop_count; } -/** - * tracing_start - quick start of the tracer - * - * If tracing is enabled but was stopped by tracing_stop, - * this will start the tracer back up. - */ -void tracing_start(void) +static void tracing_start_tr(struct trace_array *tr) { struct trace_buffer *buffer; unsigned long flags; @@ -2374,119 +2368,83 @@ void tracing_start(void) if (tracing_disabled) return; - raw_spin_lock_irqsave(&global_trace.start_lock, flags); - if (--global_trace.stop_count) { - if (global_trace.stop_count < 0) { + raw_spin_lock_irqsave(&tr->start_lock, flags); + if (--tr->stop_count) { + if (WARN_ON_ONCE(tr->stop_count < 0)) { /* Someone screwed up their debugging */ - WARN_ON_ONCE(1); - global_trace.stop_count = 0; + tr->stop_count = 0; } goto out; } /* Prevent the buffers from switching */ - arch_spin_lock(&global_trace.max_lock); + arch_spin_lock(&tr->max_lock); - buffer = global_trace.array_buffer.buffer; + buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); #ifdef CONFIG_TRACER_MAX_TRACE - buffer = global_trace.max_buffer.buffer; + buffer = tr->max_buffer.buffer; if (buffer) ring_buffer_record_enable(buffer); #endif - arch_spin_unlock(&global_trace.max_lock); - - out: - raw_spin_unlock_irqrestore(&global_trace.start_lock, flags); -} - -static void tracing_start_tr(struct trace_array *tr) -{ - struct trace_buffer *buffer; - unsigned long flags; - - if (tracing_disabled) - return; - - /* If global, we need to also start the max tracer */ - if (tr->flags & TRACE_ARRAY_FL_GLOBAL) - return tracing_start(); - - raw_spin_lock_irqsave(&tr->start_lock, flags); - - if (--tr->stop_count) { - if (tr->stop_count < 0) { - /* Someone screwed up their debugging */ - WARN_ON_ONCE(1); - tr->stop_count = 0; - } - goto out; - } - - buffer = tr->array_buffer.buffer; - if (buffer) - ring_buffer_record_enable(buffer); + arch_spin_unlock(&tr->max_lock); out: raw_spin_unlock_irqrestore(&tr->start_lock, flags); } /** - * tracing_stop - quick stop of the tracer + * tracing_start - quick start of the tracer * - * Light weight way to stop tracing. Use in conjunction with - * tracing_start. + * If tracing is enabled but was stopped by tracing_stop, + * this will start the tracer back up. */ -void tracing_stop(void) +void tracing_start(void) + +{ + return tracing_start_tr(&global_trace); +} + +static void tracing_stop_tr(struct trace_array *tr) { struct trace_buffer *buffer; unsigned long flags; - raw_spin_lock_irqsave(&global_trace.start_lock, flags); - if (global_trace.stop_count++) + raw_spin_lock_irqsave(&tr->start_lock, flags); + if (tr->stop_count++) goto out; /* Prevent the buffers from switching */ - arch_spin_lock(&global_trace.max_lock); + arch_spin_lock(&tr->max_lock); - buffer = global_trace.array_buffer.buffer; + buffer = tr->array_buffer.buffer; if (buffer) ring_buffer_record_disable(buffer); #ifdef CONFIG_TRACER_MAX_TRACE - buffer = global_trace.max_buffer.buffer; + buffer = tr->max_buffer.buffer; if (buffer) ring_buffer_record_disable(buffer); #endif - arch_spin_unlock(&global_trace.max_lock); + arch_spin_unlock(&tr->max_lock); out: - raw_spin_unlock_irqrestore(&global_trace.start_lock, flags); + raw_spin_unlock_irqrestore(&tr->start_lock, flags); } -static void tracing_stop_tr(struct trace_array *tr) +/** + * tracing_stop - quick stop of the tracer + * + * Light weight way to stop tracing. Use in conjunction with + * tracing_start. + */ +void tracing_stop(void) { - struct trace_buffer *buffer; - unsigned long flags; - - /* If global, we need to also stop the max tracer */ - if (tr->flags & TRACE_ARRAY_FL_GLOBAL) - return tracing_stop(); - - raw_spin_lock_irqsave(&tr->start_lock, flags); - if (tr->stop_count++) - goto out; - - buffer = tr->array_buffer.buffer; - if (buffer) - ring_buffer_record_disable(buffer); - - out: - raw_spin_unlock_irqrestore(&tr->start_lock, flags); + return tracing_stop_tr(&global_trace); } static int trace_save_cmdline(struct task_struct *tsk) -- cgit v1.2.3 From 7fed14f7ac9cf5e38c693836fe4a874720141845 Mon Sep 17 00:00:00 2001 From: Petr Pavlu Date: Tue, 5 Dec 2023 17:17:34 +0100 Subject: tracing: Fix incomplete locking when disabling buffered events The following warning appears when using buffered events: [ 203.556451] WARNING: CPU: 53 PID: 10220 at kernel/trace/ring_buffer.c:3912 ring_buffer_discard_commit+0x2eb/0x420 [...] [ 203.670690] CPU: 53 PID: 10220 Comm: stress-ng-sysin Tainted: G E 6.7.0-rc2-default #4 56e6d0fcf5581e6e51eaaecbdaec2a2338c80f3a [ 203.670704] Hardware name: Intel Corp. GROVEPORT/GROVEPORT, BIOS GVPRCRB1.86B.0016.D04.1705030402 05/03/2017 [ 203.670709] RIP: 0010:ring_buffer_discard_commit+0x2eb/0x420 [ 203.735721] Code: 4c 8b 4a 50 48 8b 42 48 49 39 c1 0f 84 b3 00 00 00 49 83 e8 01 75 b1 48 8b 42 10 f0 ff 40 08 0f 0b e9 fc fe ff ff f0 ff 47 08 <0f> 0b e9 77 fd ff ff 48 8b 42 10 f0 ff 40 08 0f 0b e9 f5 fe ff ff [ 203.735734] RSP: 0018:ffffb4ae4f7b7d80 EFLAGS: 00010202 [ 203.735745] RAX: 0000000000000000 RBX: ffffb4ae4f7b7de0 RCX: ffff8ac10662c000 [ 203.735754] RDX: ffff8ac0c750be00 RSI: ffff8ac10662c000 RDI: ffff8ac0c004d400 [ 203.781832] RBP: ffff8ac0c039cea0 R08: 0000000000000000 R09: 0000000000000000 [ 203.781839] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 [ 203.781842] R13: ffff8ac10662c000 R14: ffff8ac0c004d400 R15: ffff8ac10662c008 [ 203.781846] FS: 00007f4cd8a67740(0000) GS:ffff8ad798880000(0000) knlGS:0000000000000000 [ 203.781851] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 203.781855] CR2: 0000559766a74028 CR3: 00000001804c4000 CR4: 00000000001506f0 [ 203.781862] Call Trace: [ 203.781870] [ 203.851949] trace_event_buffer_commit+0x1ea/0x250 [ 203.851967] trace_event_raw_event_sys_enter+0x83/0xe0 [ 203.851983] syscall_trace_enter.isra.0+0x182/0x1a0 [ 203.851990] do_syscall_64+0x3a/0xe0 [ 203.852075] entry_SYSCALL_64_after_hwframe+0x6e/0x76 [ 203.852090] RIP: 0033:0x7f4cd870fa77 [ 203.982920] Code: 00 b8 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 66 90 b8 89 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d e9 43 0e 00 f7 d8 64 89 01 48 [ 203.982932] RSP: 002b:00007fff99717dd8 EFLAGS: 00000246 ORIG_RAX: 0000000000000089 [ 203.982942] RAX: ffffffffffffffda RBX: 0000558ea1d7b6f0 RCX: 00007f4cd870fa77 [ 203.982948] RDX: 0000000000000000 RSI: 00007fff99717de0 RDI: 0000558ea1d7b6f0 [ 203.982957] RBP: 00007fff99717de0 R08: 00007fff997180e0 R09: 00007fff997180e0 [ 203.982962] R10: 00007fff997180e0 R11: 0000000000000246 R12: 00007fff99717f40 [ 204.049239] R13: 00007fff99718590 R14: 0000558e9f2127a8 R15: 00007fff997180b0 [ 204.049256] For instance, it can be triggered by running these two commands in parallel: $ while true; do echo hist:key=id.syscall:val=hitcount > \ /sys/kernel/debug/tracing/events/raw_syscalls/sys_enter/trigger; done $ stress-ng --sysinfo $(nproc) The warning indicates that the current ring_buffer_per_cpu is not in the committing state. It happens because the active ring_buffer_event doesn't actually come from the ring_buffer_per_cpu but is allocated from trace_buffered_event. The bug is in function trace_buffered_event_disable() where the following normally happens: * The code invokes disable_trace_buffered_event() via smp_call_function_many() and follows it by synchronize_rcu(). This increments the per-CPU variable trace_buffered_event_cnt on each target CPU and grants trace_buffered_event_disable() the exclusive access to the per-CPU variable trace_buffered_event. * Maintenance is performed on trace_buffered_event, all per-CPU event buffers get freed. * The code invokes enable_trace_buffered_event() via smp_call_function_many(). This decrements trace_buffered_event_cnt and releases the access to trace_buffered_event. A problem is that smp_call_function_many() runs a given function on all target CPUs except on the current one. The following can then occur: * Task X executing trace_buffered_event_disable() runs on CPU 0. * The control reaches synchronize_rcu() and the task gets rescheduled on another CPU 1. * The RCU synchronization finishes. At this point, trace_buffered_event_disable() has the exclusive access to all trace_buffered_event variables except trace_buffered_event[CPU0] because trace_buffered_event_cnt[CPU0] is never incremented and if the buffer is currently unused, remains set to 0. * A different task Y is scheduled on CPU 0 and hits a trace event. The code in trace_event_buffer_lock_reserve() sees that trace_buffered_event_cnt[CPU0] is set to 0 and decides the use the buffer provided by trace_buffered_event[CPU0]. * Task X continues its execution in trace_buffered_event_disable(). The code incorrectly frees the event buffer pointed by trace_buffered_event[CPU0] and resets the variable to NULL. * Task Y writes event data to the now freed buffer and later detects the created inconsistency. The issue is observable since commit dea499781a11 ("tracing: Fix warning in trace_buffered_event_disable()") which moved the call of trace_buffered_event_disable() in __ftrace_event_enable_disable() earlier, prior to invoking call->class->reg(.. TRACE_REG_UNREGISTER ..). The underlying problem in trace_buffered_event_disable() is however present since the original implementation in commit 0fc1b09ff1ff ("tracing: Use temp buffer when filtering events"). Fix the problem by replacing the two smp_call_function_many() calls with on_each_cpu_mask() which invokes a given callback on all CPUs. Link: https://lore.kernel.org/all/20231127151248.7232-2-petr.pavlu@suse.com/ Link: https://lkml.kernel.org/r/20231205161736.19663-2-petr.pavlu@suse.com Cc: stable@vger.kernel.org Fixes: 0fc1b09ff1ff ("tracing: Use temp buffer when filtering events") Fixes: dea499781a11 ("tracing: Fix warning in trace_buffered_event_disable()") Signed-off-by: Petr Pavlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 2492c6c76850..6aeffa4a6994 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2781,11 +2781,9 @@ void trace_buffered_event_disable(void) if (--trace_buffered_event_ref) return; - preempt_disable(); /* For each CPU, set the buffer as used. */ - smp_call_function_many(tracing_buffer_mask, - disable_trace_buffered_event, NULL, 1); - preempt_enable(); + on_each_cpu_mask(tracing_buffer_mask, disable_trace_buffered_event, + NULL, true); /* Wait for all current users to finish */ synchronize_rcu(); @@ -2800,11 +2798,9 @@ void trace_buffered_event_disable(void) */ smp_wmb(); - preempt_disable(); /* Do the work on each cpu */ - smp_call_function_many(tracing_buffer_mask, - enable_trace_buffered_event, NULL, 1); - preempt_enable(); + on_each_cpu_mask(tracing_buffer_mask, enable_trace_buffered_event, NULL, + true); } static struct trace_buffer *temp_buffer; -- cgit v1.2.3 From 34209fe83ef8404353f91ab4ea4035dbc9922d04 Mon Sep 17 00:00:00 2001 From: Petr Pavlu Date: Tue, 5 Dec 2023 17:17:35 +0100 Subject: tracing: Fix a warning when allocating buffered events fails Function trace_buffered_event_disable() produces an unexpected warning when the previous call to trace_buffered_event_enable() fails to allocate pages for buffered events. The situation can occur as follows: * The counter trace_buffered_event_ref is at 0. * The soft mode gets enabled for some event and trace_buffered_event_enable() is called. The function increments trace_buffered_event_ref to 1 and starts allocating event pages. * The allocation fails for some page and trace_buffered_event_disable() is called for cleanup. * Function trace_buffered_event_disable() decrements trace_buffered_event_ref back to 0, recognizes that it was the last use of buffered events and frees all allocated pages. * The control goes back to trace_buffered_event_enable() which returns. The caller of trace_buffered_event_enable() has no information that the function actually failed. * Some time later, the soft mode is disabled for the same event. Function trace_buffered_event_disable() is called. It warns on "WARN_ON_ONCE(!trace_buffered_event_ref)" and returns. Buffered events are just an optimization and can handle failures. Make trace_buffered_event_enable() exit on the first failure and left any cleanup later to when trace_buffered_event_disable() is called. Link: https://lore.kernel.org/all/20231127151248.7232-2-petr.pavlu@suse.com/ Link: https://lkml.kernel.org/r/20231205161736.19663-3-petr.pavlu@suse.com Fixes: 0fc1b09ff1ff ("tracing: Use temp buffer when filtering events") Signed-off-by: Petr Pavlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 6aeffa4a6994..ef72354f61ce 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2728,8 +2728,11 @@ void trace_buffered_event_enable(void) for_each_tracing_cpu(cpu) { page = alloc_pages_node(cpu_to_node(cpu), GFP_KERNEL | __GFP_NORETRY, 0); - if (!page) - goto failed; + /* This is just an optimization and can handle failures */ + if (!page) { + pr_err("Failed to allocate event buffer\n"); + break; + } event = page_address(page); memset(event, 0, sizeof(*event)); @@ -2743,10 +2746,6 @@ void trace_buffered_event_enable(void) WARN_ON_ONCE(1); preempt_enable(); } - - return; - failed: - trace_buffered_event_disable(); } static void enable_trace_buffered_event(void *data) -- cgit v1.2.3 From c0591b1cccf708a47bc465c62436d669a4213323 Mon Sep 17 00:00:00 2001 From: Petr Pavlu Date: Tue, 5 Dec 2023 17:17:36 +0100 Subject: tracing: Fix a possible race when disabling buffered events Function trace_buffered_event_disable() is responsible for freeing pages backing buffered events and this process can run concurrently with trace_event_buffer_lock_reserve(). The following race is currently possible: * Function trace_buffered_event_disable() is called on CPU 0. It increments trace_buffered_event_cnt on each CPU and waits via synchronize_rcu() for each user of trace_buffered_event to complete. * After synchronize_rcu() is finished, function trace_buffered_event_disable() has the exclusive access to trace_buffered_event. All counters trace_buffered_event_cnt are at 1 and all pointers trace_buffered_event are still valid. * At this point, on a different CPU 1, the execution reaches trace_event_buffer_lock_reserve(). The function calls preempt_disable_notrace() and only now enters an RCU read-side critical section. The function proceeds and reads a still valid pointer from trace_buffered_event[CPU1] into the local variable "entry". However, it doesn't yet read trace_buffered_event_cnt[CPU1] which happens later. * Function trace_buffered_event_disable() continues. It frees trace_buffered_event[CPU1] and decrements trace_buffered_event_cnt[CPU1] back to 0. * Function trace_event_buffer_lock_reserve() continues. It reads and increments trace_buffered_event_cnt[CPU1] from 0 to 1. This makes it believe that it can use the "entry" that it already obtained but the pointer is now invalid and any access results in a use-after-free. Fix the problem by making a second synchronize_rcu() call after all trace_buffered_event values are set to NULL. This waits on all potential users in trace_event_buffer_lock_reserve() that still read a previous pointer from trace_buffered_event. Link: https://lore.kernel.org/all/20231127151248.7232-2-petr.pavlu@suse.com/ Link: https://lkml.kernel.org/r/20231205161736.19663-4-petr.pavlu@suse.com Cc: stable@vger.kernel.org Fixes: 0fc1b09ff1ff ("tracing: Use temp buffer when filtering events") Signed-off-by: Petr Pavlu Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index ef72354f61ce..fbcd3bafb93e 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2791,13 +2791,17 @@ void trace_buffered_event_disable(void) free_page((unsigned long)per_cpu(trace_buffered_event, cpu)); per_cpu(trace_buffered_event, cpu) = NULL; } + /* - * Make sure trace_buffered_event is NULL before clearing - * trace_buffered_event_cnt. + * Wait for all CPUs that potentially started checking if they can use + * their event buffer only after the previous synchronize_rcu() call and + * they still read a valid pointer from trace_buffered_event. It must be + * ensured they don't see cleared trace_buffered_event_cnt else they + * could wrongly decide to use the pointed-to buffer which is now freed. */ - smp_wmb(); + synchronize_rcu(); - /* Do the work on each cpu */ + /* For each CPU, relinquish the buffer */ on_each_cpu_mask(tracing_buffer_mask, enable_trace_buffered_event, NULL, true); } -- cgit v1.2.3 From b2dd797543cfa6580eac8408dd67fa02164d9e56 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 6 Dec 2023 10:02:44 -0500 Subject: ring-buffer: Force absolute timestamp on discard of event There's a race where if an event is discarded from the ring buffer and an interrupt were to happen at that time and insert an event, the time stamp is still used from the discarded event as an offset. This can screw up the timings. If the event is going to be discarded, set the "before_stamp" to zero. When a new event comes in, it compares the "before_stamp" with the "write_stamp" and if they are not equal, it will insert an absolute timestamp. This will prevent the timings from getting out of sync due to the discarded event. Link: https://lore.kernel.org/linux-trace-kernel/20231206100244.5130f9b3@gandalf.local.home Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Fixes: 6f6be606e763f ("ring-buffer: Force before_stamp and write_stamp to be different on discard") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ring_buffer.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 43cc47d7faaf..a6da2d765c78 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -3030,22 +3030,19 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, local_read(&bpage->write) & ~RB_WRITE_MASK; unsigned long event_length = rb_event_length(event); + /* + * For the before_stamp to be different than the write_stamp + * to make sure that the next event adds an absolute + * value and does not rely on the saved write stamp, which + * is now going to be bogus. + */ + rb_time_set(&cpu_buffer->before_stamp, 0); + /* Something came in, can't discard */ if (!rb_time_cmpxchg(&cpu_buffer->write_stamp, write_stamp, write_stamp - delta)) return false; - /* - * It's possible that the event time delta is zero - * (has the same time stamp as the previous event) - * in which case write_stamp and before_stamp could - * be the same. In such a case, force before_stamp - * to be different than write_stamp. It doesn't - * matter what it is, as long as its different. - */ - if (!delta) - rb_time_set(&cpu_buffer->before_stamp, 0); - /* * If an event were to come in now, it would see that the * write_stamp and the before_stamp are different, and assume -- cgit v1.2.3 From f458a1453424e03462b5bb539673c9a3cddda480 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (Google)" Date: Wed, 6 Dec 2023 10:00:50 -0500 Subject: ring-buffer: Test last update in 32bit version of __rb_time_read() Since 64 bit cmpxchg() is very expensive on 32bit architectures, the timestamp used by the ring buffer does some interesting tricks to be able to still have an atomic 64 bit number. It originally just used 60 bits and broke it up into two 32 bit words where the extra 2 bits were used for synchronization. But this was not enough for all use cases, and all 64 bits were required. The 32bit version of the ring buffer timestamp was then broken up into 3 32bit words using the same counter trick. But one update was not done. The check to see if the read operation was done without interruption only checked the first two words and not last one (like it had before this update). Fix it by making sure all three updates happen without interruption by comparing the initial counter with the last updated counter. Link: https://lore.kernel.org/linux-trace-kernel/20231206100050.3100b7bb@gandalf.local.home Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mark Rutland Cc: Mathieu Desnoyers Fixes: f03f2abce4f39 ("ring-buffer: Have 32 bit time stamps use all 64 bits") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ring_buffer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index a6da2d765c78..8d2a4f00eca9 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -644,8 +644,8 @@ static inline bool __rb_time_read(rb_time_t *t, u64 *ret, unsigned long *cnt) *cnt = rb_time_cnt(top); - /* If top and bottom counts don't match, this interrupted a write */ - if (*cnt != rb_time_cnt(bottom)) + /* If top and msb counts don't match, this interrupted a write */ + if (*cnt != rb_time_cnt(msb)) return false; /* The shift to msb will lose its cnt bits */ -- cgit v1.2.3 From 4b7de801606e504e69689df71475d27e35336fb3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Wed, 6 Dec 2023 09:30:40 +0100 Subject: bpf: Fix prog_array_map_poke_run map poke update Lee pointed out issue found by syscaller [0] hitting BUG in prog array map poke update in prog_array_map_poke_run function due to error value returned from bpf_arch_text_poke function. There's race window where bpf_arch_text_poke can fail due to missing bpf program kallsym symbols, which is accounted for with check for -EINVAL in that BUG_ON call. The problem is that in such case we won't update the tail call jump and cause imbalance for the next tail call update check which will fail with -EBUSY in bpf_arch_text_poke. I'm hitting following race during the program load: CPU 0 CPU 1 bpf_prog_load bpf_check do_misc_fixups prog_array_map_poke_track map_update_elem bpf_fd_array_map_update_elem prog_array_map_poke_run bpf_arch_text_poke returns -EINVAL bpf_prog_kallsyms_add After bpf_arch_text_poke (CPU 1) fails to update the tail call jump, the next poke update fails on expected jump instruction check in bpf_arch_text_poke with -EBUSY and triggers the BUG_ON in prog_array_map_poke_run. Similar race exists on the program unload. Fixing this by moving the update to bpf_arch_poke_desc_update function which makes sure we call __bpf_arch_text_poke that skips the bpf address check. Each architecture has slightly different approach wrt looking up bpf address in bpf_arch_text_poke, so instead of splitting the function or adding new 'checkip' argument in previous version, it seems best to move the whole map_poke_run update as arch specific code. [0] https://syzkaller.appspot.com/bug?extid=97a4fe20470e9bc30810 Fixes: ebf7d1f508a7 ("bpf, x64: rework pro/epilogue and tailcall handling in JIT") Reported-by: syzbot+97a4fe20470e9bc30810@syzkaller.appspotmail.com Signed-off-by: Jiri Olsa Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song Cc: Lee Jones Cc: Maciej Fijalkowski Link: https://lore.kernel.org/bpf/20231206083041.1306660-2-jolsa@kernel.org --- kernel/bpf/arraymap.c | 58 +++++++++------------------------------------------ 1 file changed, 10 insertions(+), 48 deletions(-) (limited to 'kernel') diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 2058e89b5ddd..c85ff9162a5c 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -1012,11 +1012,16 @@ static void prog_array_map_poke_untrack(struct bpf_map *map, mutex_unlock(&aux->poke_mutex); } +void __weak bpf_arch_poke_desc_update(struct bpf_jit_poke_descriptor *poke, + struct bpf_prog *new, struct bpf_prog *old) +{ + WARN_ON_ONCE(1); +} + static void prog_array_map_poke_run(struct bpf_map *map, u32 key, struct bpf_prog *old, struct bpf_prog *new) { - u8 *old_addr, *new_addr, *old_bypass_addr; struct prog_poke_elem *elem; struct bpf_array_aux *aux; @@ -1025,7 +1030,7 @@ static void prog_array_map_poke_run(struct bpf_map *map, u32 key, list_for_each_entry(elem, &aux->poke_progs, list) { struct bpf_jit_poke_descriptor *poke; - int i, ret; + int i; for (i = 0; i < elem->aux->size_poke_tab; i++) { poke = &elem->aux->poke_tab[i]; @@ -1044,21 +1049,10 @@ static void prog_array_map_poke_run(struct bpf_map *map, u32 key, * activated, so tail call updates can arrive from here * while JIT is still finishing its final fixup for * non-activated poke entries. - * 3) On program teardown, the program's kallsym entry gets - * removed out of RCU callback, but we can only untrack - * from sleepable context, therefore bpf_arch_text_poke() - * might not see that this is in BPF text section and - * bails out with -EINVAL. As these are unreachable since - * RCU grace period already passed, we simply skip them. - * 4) Also programs reaching refcount of zero while patching + * 3) Also programs reaching refcount of zero while patching * is in progress is okay since we're protected under * poke_mutex and untrack the programs before the JIT - * buffer is freed. When we're still in the middle of - * patching and suddenly kallsyms entry of the program - * gets evicted, we just skip the rest which is fine due - * to point 3). - * 5) Any other error happening below from bpf_arch_text_poke() - * is a unexpected bug. + * buffer is freed. */ if (!READ_ONCE(poke->tailcall_target_stable)) continue; @@ -1068,39 +1062,7 @@ static void prog_array_map_poke_run(struct bpf_map *map, u32 key, poke->tail_call.key != key) continue; - old_bypass_addr = old ? NULL : poke->bypass_addr; - old_addr = old ? (u8 *)old->bpf_func + poke->adj_off : NULL; - new_addr = new ? (u8 *)new->bpf_func + poke->adj_off : NULL; - - if (new) { - ret = bpf_arch_text_poke(poke->tailcall_target, - BPF_MOD_JUMP, - old_addr, new_addr); - BUG_ON(ret < 0 && ret != -EINVAL); - if (!old) { - ret = bpf_arch_text_poke(poke->tailcall_bypass, - BPF_MOD_JUMP, - poke->bypass_addr, - NULL); - BUG_ON(ret < 0 && ret != -EINVAL); - } - } else { - ret = bpf_arch_text_poke(poke->tailcall_bypass, - BPF_MOD_JUMP, - old_bypass_addr, - poke->bypass_addr); - BUG_ON(ret < 0 && ret != -EINVAL); - /* let other CPUs finish the execution of program - * so that it will not possible to expose them - * to invalid nop, stack unwind, nop state - */ - if (!ret) - synchronize_rcu(); - ret = bpf_arch_text_poke(poke->tailcall_target, - BPF_MOD_JUMP, - old_addr, NULL); - BUG_ON(ret < 0 && ret != -EINVAL); - } + bpf_arch_poke_desc_update(poke, new, old); } } } -- cgit v1.2.3 From dccf78d39f1069a5ddf4328bf0c97aa5f2f4296e Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Tue, 28 Nov 2023 13:44:57 +0800 Subject: kernel/Kconfig.kexec: drop select of KEXEC for CRASH_DUMP Ignat Korchagin complained that a potential config regression was introduced by commit 89cde455915f ("kexec: consolidate kexec and crash options into kernel/Kconfig.kexec"). Before the commit, CONFIG_CRASH_DUMP has no dependency on CONFIG_KEXEC. After the commit, CRASH_DUMP selects KEXEC. That enforces system to have CONFIG_KEXEC=y as long as CONFIG_CRASH_DUMP=Y which people may not want. In Ignat's case, he sets CONFIG_CRASH_DUMP=y, CONFIG_KEXEC_FILE=y and CONFIG_KEXEC=n because kexec_load interface could have security issue if kernel/initrd has no chance to be signed and verified. CRASH_DUMP has select of KEXEC because Eric, author of above commit, met a LKP report of build failure when posting patch of earlier version. Please see below link to get detail of the LKP report: https://lore.kernel.org/all/3e8eecd1-a277-2cfb-690e-5de2eb7b988e@oracle.com/T/#u In fact, that LKP report is triggered because arm's is wrapped in CONFIG_KEXEC ifdeffery scope. That is wrong. CONFIG_KEXEC controls the enabling/disabling of kexec_load interface, but not kexec feature. Removing the wrongly added CONFIG_KEXEC ifdeffery scope in of arm allows us to drop the select KEXEC for CRASH_DUMP. Meanwhile, change arch/arm/kernel/Makefile to let machine_kexec.o relocate_kernel.o depend on KEXEC_CORE. Link: https://lkml.kernel.org/r/20231128054457.659452-1-bhe@redhat.com Fixes: 89cde455915f ("kexec: consolidate kexec and crash options into kernel/Kconfig.kexec") Signed-off-by: Baoquan He Reported-by: Ignat Korchagin Tested-by: Ignat Korchagin [compile-time only] Tested-by: Alexander Gordeev Reviewed-by: Eric DeVolder Tested-by: Eric DeVolder Signed-off-by: Andrew Morton --- kernel/Kconfig.kexec | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel') diff --git a/kernel/Kconfig.kexec b/kernel/Kconfig.kexec index 7aff28ded2f4..1cc3b1c595d7 100644 --- a/kernel/Kconfig.kexec +++ b/kernel/Kconfig.kexec @@ -97,7 +97,6 @@ config CRASH_DUMP depends on ARCH_SUPPORTS_KEXEC select CRASH_CORE select KEXEC_CORE - select KEXEC help Generate crash dump after being started by kexec. This should be normally only set in special crash dump kernels -- cgit v1.2.3