kernel.git/kernel/bpf/btf.c, branch linux-6.4.y

bpf: Fix an error in verifying a field in a union

2023-09-13T07:48:00Z

[ Upstream commit 33937607efa050d9e237e0c4ac4ada02d961c466 ] We are utilizing BPF LSM to monitor BPF operations within our container environment. When we add support for raw_tracepoint, it hits below error. ; (const void *)attr->raw_tracepoint.name); 27: (79) r3 = *(u64 *)(r2 +0) access beyond the end of member map_type (mend:4) in struct (anon) with off 0 size 8 It can be reproduced with below BPF prog. SEC("lsm/bpf") int BPF_PROG(bpf_audit, int cmd, union bpf_attr *attr, unsigned int size) { switch (cmd) { case BPF_RAW_TRACEPOINT_OPEN: bpf_printk("raw_tracepoint is %s", attr->raw_tracepoint.name); break; default: break; } return 0; } The reason is that when accessing a field in a union, such as bpf_attr, if the field is located within a nested struct that is not the first member of the union, it can result in incorrect field verification. union bpf_attr { struct { __u32 map_type; <<<< Actually it will find that field. __u32 key_size; __u32 value_size; ... }; ... struct { __u64 name; <<<< We want to verify this field. __u32 prog_fd; } raw_tracepoint; }; Considering the potential deep nesting levels, finding a perfect solution to address this issue has proven challenging. Therefore, I propose a solution where we simply skip the verification process if the field in question is located within a union. Fixes: 7e3617a72df3 ("bpf: Add array support to btf_struct_access") Signed-off-by: Yafang Shao Link: https://lore.kernel.org/r/20230713025642.27477-4-laoar.shao@gmail.com Signed-off-by: Alexei Starovoitov Signed-off-by: Sasha Levin

bpf: Fix an error around PTR_UNTRUSTED

2023-09-13T07:48:00Z

[ Upstream commit 7ce4dc3e4a9d954c8a1fb483c7a527e9b060b860 ] Per discussion with Alexei, the PTR_UNTRUSTED flag should not been cleared when we start to walk a new struct, because the struct in question may be a struct nested in a union. We should also check and set this flag before we walk its each member, in case itself is a union. We will clear this flag if the field is BTF_TYPE_SAFE_RCU_OR_NULL. Fixes: 6fcd486b3a0a ("bpf: Refactor RCU enforcement in the verifier.") Signed-off-by: Yafang Shao Link: https://lore.kernel.org/r/20230713025642.27477-2-laoar.shao@gmail.com Signed-off-by: Alexei Starovoitov Signed-off-by: Sasha Levin

bpf: Silence a warning in btf_type_id_size()

2023-07-27T06:56:49Z

[ Upstream commit e6c2f594ed961273479505b42040782820190305 ] syzbot reported a warning in [1] with the following stacktrace: WARNING: CPU: 0 PID: 5005 at kernel/bpf/btf.c:1988 btf_type_id_size+0x2d9/0x9d0 kernel/bpf/btf.c:1988 ... RIP: 0010:btf_type_id_size+0x2d9/0x9d0 kernel/bpf/btf.c:1988 ... Call Trace: map_check_btf kernel/bpf/syscall.c:1024 [inline] map_create+0x1157/0x1860 kernel/bpf/syscall.c:1198 __sys_bpf+0x127f/0x5420 kernel/bpf/syscall.c:5040 __do_sys_bpf kernel/bpf/syscall.c:5162 [inline] __se_sys_bpf kernel/bpf/syscall.c:5160 [inline] __x64_sys_bpf+0x79/0xc0 kernel/bpf/syscall.c:5160 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x39/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd With the following btf [1] DECL_TAG 'a' type_id=4 component_idx=-1 [2] PTR '(anon)' type_id=0 [3] TYPE_TAG 'a' type_id=2 [4] VAR 'a' type_id=3, linkage=static and when the bpf_attr.btf_key_type_id = 1 (DECL_TAG), the following WARN_ON_ONCE in btf_type_id_size() is triggered: if (WARN_ON_ONCE(!btf_type_is_modifier(size_type) && !btf_type_is_var(size_type))) return NULL; Note that 'return NULL' is the correct behavior as we don't want a DECL_TAG type to be used as a btf_{key,value}_type_id even for the case like 'DECL_TAG -> STRUCT'. So there is no correctness issue here, we just want to silence warning. To silence the warning, I added DECL_TAG as one of kinds in btf_type_nosize() which will cause btf_type_id_size() returning NULL earlier without the warning. [1] https://lore.kernel.org/bpf/000000000000e0df8d05fc75ba86@google.com/ Reported-by: syzbot+958967f249155967d42a@syzkaller.appspotmail.com Signed-off-by: Yonghong Song Link: https://lore.kernel.org/r/20230530205029.264910-1-yhs@fb.com Signed-off-by: Martin KaFai Lau Signed-off-by: Sasha Levin

bpf, btf: Warn but return no error for NULL btf from __register_btf_kfunc_id_set()

2023-07-19T14:36:47Z

[ Upstream commit 3de4d22cc9ac7c9f38e10edcf54f9a8891a9c2aa ] __register_btf_kfunc_id_set() assumes .BTF to be part of the module's .ko file if CONFIG_DEBUG_INFO_BTF is enabled. If that's not the case, the function prints an error message and return an error. As a result, such modules cannot be loaded. However, the section could be stripped out during a build process. It would be better to let the modules loaded, because their basic functionalities have no problem [0], though the BTF functionalities will not be supported. Make the function to lower the level of the message from error to warn, and return no error. [0] https://lore.kernel.org/bpf/20220219082037.ow2kbq5brktf4f2u@apollo.legion Fixes: c446fdacb10d ("bpf: fix register_btf_kfunc_id_set for !CONFIG_DEBUG_INFO_BTF") Reported-by: Alexander Egorenkov Suggested-by: Kumar Kartikeya Dwivedi Signed-off-by: SeongJae Park Signed-off-by: Daniel Borkmann Acked-by: Jiri Olsa Link: https://lore.kernel.org/bpf/87y228q66f.fsf@oc8242746057.ibm.com Link: https://lore.kernel.org/bpf/20220219082037.ow2kbq5brktf4f2u@apollo.legion Link: https://lore.kernel.org/bpf/20230701171447.56464-1-sj@kernel.org Signed-off-by: Sasha Levin

bpf/btf: Accept function names that contain dots

2023-06-21T08:32:22Z

When building a kernel with LLVM=1, LLVM_IAS=0 and CONFIG_KASAN=y, LLVM leaves DWARF tags for the "asan.module_ctor" & co symbols. In turn, pahole creates BTF_KIND_FUNC entries for these and this makes the BTF metadata validation fail because they contain a dot. In a dramatic turn of event, this BTF verification failure can cause the netfilter_bpf initialization to fail, causing netfilter_core to free the netfilter_helper hashmap and netfilter_ftp to trigger a use-after-free. The risk of u-a-f in netfilter will be addressed separately but the existence of "asan.module_ctor" debug info under some build conditions sounds like a good enough reason to accept functions that contain dots in BTF. Although using only LLVM=1 is the recommended way to compile clang-based kernels, users can certainly do LLVM=1, LLVM_IAS=0 as well and we still try to support that combination according to Nick. To clarify: - > v5.10 kernel, LLVM=1 (LLVM_IAS=0 is not the default) is recommended, but user can still have LLVM=1, LLVM_IAS=0 to trigger the issue - <= 5.10 kernel, LLVM=1 (LLVM_IAS=0 is the default) is recommended in which case GNU as will be used Fixes: 1dc92851849c ("bpf: kernel side support for BTF Var and DataSec") Signed-off-by: Florent Revest Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Cc: Yonghong Song Cc: Nick Desaulniers Link: https://lore.kernel.org/bpf/20230615145607.3469985-1-revest@chromium.org

bpf: minimal support for programs hooked into netfilter framework

2023-04-21T18:34:14Z

This adds minimal support for BPF_PROG_TYPE_NETFILTER bpf programs that will be invoked via the NF_HOOK() points in the ip stack. Invocation incurs an indirect call. This is not a necessity: Its possible to add 'DEFINE_BPF_DISPATCHER(nf_progs)' and handle the program invocation with the same method already done for xdp progs. This isn't done here to keep the size of this chunk down. Verifier restricts verdicts to either DROP or ACCEPT. Signed-off-by: Florian Westphal Link: https://lore.kernel.org/r/20230421170300.24115-3-fw@strlen.de Signed-off-by: Alexei Starovoitov

bpf: Fix race between btf_put and btf_idr walk.

2023-04-21T14:27:56Z

Florian and Eduard reported hard dead lock: [ 58.433327] _raw_spin_lock_irqsave+0x40/0x50 [ 58.433334] btf_put+0x43/0x90 [ 58.433338] bpf_find_btf_id+0x157/0x240 [ 58.433353] btf_parse_fields+0x921/0x11c0 This happens since btf->refcount can be 1 at the time of btf_put() and btf_put() will call btf_free_id() which will try to grab btf_idr_lock and will dead lock. Avoid the issue by doing btf_put() without locking. Fixes: 3d78417b60fb ("bpf: Add bpf_btf_find_by_name_kind() helper.") Fixes: 1e89106da253 ("bpf: Add bpf_core_add_cands() and wire it into bpf_core_apply_relo_insn().") Reported-by: Florian Westphal Reported-by: Eduard Zingerman Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann Tested-by: Eduard Zingerman Link: https://lore.kernel.org/bpf/20230421014901.70908-1-alexei.starovoitov@gmail.com

bpf: support access variable length array of integer type

2023-04-20T04:29:39Z

After this commit: bpf: Support variable length array in tracing programs (9c5f8a1008a1) Trace programs can access variable length array, but for structure type. This patch adds support for integer type. Example: Hook load_balance struct sched_domain { ... unsigned long span[]; } The access: sd->span[0]. Co-developed-by: Chengming Zhou Signed-off-by: Chengming Zhou Signed-off-by: Feng Zhou Link: https://lore.kernel.org/r/20230420032735.27760-2-zhoufeng.zf@bytedance.com Signed-off-by: Alexei Starovoitov

bpf: Migrate bpf_rbtree_remove to possibly fail

2023-04-16T00:36:50Z

This patch modifies bpf_rbtree_remove to account for possible failure due to the input rb_node already not being in any collection. The function can now return NULL, and does when the aforementioned scenario occurs. As before, on successful removal an owning reference to the removed node is returned. Adding KF_RET_NULL to bpf_rbtree_remove's kfunc flags - now KF_RET_NULL | KF_ACQUIRE - provides the desired verifier semantics: * retval must be checked for NULL before use * if NULL, retval's ref_obj_id is released * retval is a "maybe acquired" owning ref, not a non-owning ref, so it will live past end of critical section (bpf_spin_unlock), and thus can be checked for NULL after the end of the CS BPF programs must add checks ============================ This does change bpf_rbtree_remove's verifier behavior. BPF program writers will need to add NULL checks to their programs, but the resulting UX looks natural: bpf_spin_lock(&glock); n = bpf_rbtree_first(&ghead); if (!n) { /* ... */} res = bpf_rbtree_remove(&ghead, &n->node); bpf_spin_unlock(&glock); if (!res) /* Newly-added check after this patch */ return 1; n = container_of(res, /* ... */); /* Do something else with n */ bpf_obj_drop(n); return 0; The "if (!res)" check above is the only addition necessary for the above program to pass verification after this patch. bpf_rbtree_remove no longer clobbers non-owning refs ==================================================== An issue arises when bpf_rbtree_remove fails, though. Consider this example: struct node_data { long key; struct bpf_list_node l; struct bpf_rb_node r; struct bpf_refcount ref; }; long failed_sum; void bpf_prog() { struct node_data *n = bpf_obj_new(/* ... */); struct bpf_rb_node *res; n->key = 10; bpf_spin_lock(&glock); bpf_list_push_back(&some_list, &n->l); /* n is now a non-owning ref */ res = bpf_rbtree_remove(&some_tree, &n->r, /* ... */); if (!res) failed_sum += n->key; /* not possible */ bpf_spin_unlock(&glock); /* if (res) { do something useful and drop } ... */ } The bpf_rbtree_remove in this example will always fail. Similarly to bpf_spin_unlock, bpf_rbtree_remove is a non-owning reference invalidation point. The verifier clobbers all non-owning refs after a bpf_rbtree_remove call, so the "failed_sum += n->key" line will fail verification, and in fact there's no good way to get information about the node which failed to add after the invalidation. This patch removes non-owning reference invalidation from bpf_rbtree_remove to allow the above usecase to pass verification. The logic for why this is now possible is as follows: Before this series, bpf_rbtree_add couldn't fail and thus assumed that its input, a non-owning reference, was in the tree. But it's easy to construct an example where two non-owning references pointing to the same underlying memory are acquired and passed to rbtree_remove one after another (see rbtree_api_release_aliasing in selftests/bpf/progs/rbtree_fail.c). So it was necessary to clobber non-owning refs to prevent this case and, more generally, to enforce "non-owning ref is definitely in some collection" invariant. This series removes that invariant and the failure / runtime checking added in this patch provide a clean way to deal with the aliasing issue - just fail to remove. Because the aliasing issue prevented by clobbering non-owning refs is no longer an issue, this patch removes the invalidate_non_owning_refs call from verifier handling of bpf_rbtree_remove. Note that bpf_spin_unlock - the other caller of invalidate_non_owning_refs - clobbers non-owning refs for a different reason, so its clobbering behavior remains unchanged. No BPF program changes are necessary for programs to remain valid as a result of this clobbering change. A valid program before this patch passed verification with its non-owning refs having shorter (or equal) lifetimes due to more aggressive clobbering. Also, update existing tests to check bpf_rbtree_remove retval for NULL where necessary, and move rbtree_api_release_aliasing from progs/rbtree_fail.c to progs/rbtree.c since it's now expected to pass verification. Signed-off-by: Dave Marchevsky Link: https://lore.kernel.org/r/20230415201811.343116-8-davemarchevsky@fb.com Signed-off-by: Alexei Starovoitov

bpf: Introduce opaque bpf_refcount struct and add btf_record plumbing

2023-04-16T00:36:49Z

A 'struct bpf_refcount' is added to the set of opaque uapi/bpf.h types meant for use in BPF programs. Similarly to other opaque types like bpf_spin_lock and bpf_rbtree_node, the verifier needs to know where in user-defined struct types a bpf_refcount can be located, so necessary btf_record plumbing is added to enable this. bpf_refcount is sized to hold a refcount_t. Similarly to bpf_spin_lock, the offset of a bpf_refcount is cached in btf_record as refcount_off in addition to being in the field array. Caching refcount_off makes sense for this field because further patches in the series will modify functions that take local kptrs (e.g. bpf_obj_drop) to change their behavior if the type they're operating on is refcounted. So enabling fast "is this type refcounted?" checks is desirable. No such verifier behavior changes are introduced in this patch, just logic to recognize 'struct bpf_refcount' in btf_record. Signed-off-by: Dave Marchevsky Link: https://lore.kernel.org/r/20230415201811.343116-3-davemarchevsky@fb.com Signed-off-by: Alexei Starovoitov