Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

Daniel Borkmann says:

====================
pull-request: bpf-next 2024-03-25

We've added 38 non-merge commits during the last 13 day(s) which contain
a total of 50 files changed, 867 insertions(+), 274 deletions(-).

The main changes are:

1) Add the ability to specify and retrieve BPF cookie also for raw
   tracepoint programs in order to ease migration from classic to raw
   tracepoints, from Andrii Nakryiko.

2) Allow the use of bpf_get_{ns_,}current_pid_tgid() helper for all
   program types and add additional BPF selftests, from Yonghong Song.

3) Several improvements to bpftool and its build, for example, enabling
   libbpf logs when loading pid_iter in debug mode, from Quentin Monnet.

4) Check the return code of all BPF-related set_memory_*() functions during
   load and bail out in case they fail, from Christophe Leroy.

5) Avoid a goto in regs_refine_cond_op() such that the verifier can
   be better integrated into Agni tool which doesn't support backedges
   yet, from Harishankar Vishwanathan.

6) Add a small BPF trie perf improvement by always inlining
   longest_prefix_match, from Jesper Dangaard Brouer.

7) Small BPF selftest refactor in bpf_tcp_ca.c to utilize start_server()
   helper instead of open-coding it, from Geliang Tang.

8) Improve test_tc_tunnel.sh BPF selftest to prevent client connect
   before the server bind, from Alessandro Carminati.

9) Fix BPF selftest benchmark for older glibc and use syscall(SYS_gettid)
   instead of gettid(), from Alan Maguire.

10) Implement a backward-compatible method for struct_ops types with
    additional fields which are not present in older kernels,
    from Kui-Feng Lee.

11) Add a small helper to check if an instruction is addr_space_cast
    from as(0) to as(1) and utilize it in x86-64 JIT, from Puranjay Mohan.

12) Small cleanup to remove unnecessary error check in
    bpf_struct_ops_map_update_elem, from Martin KaFai Lau.

13) Improvements to libbpf fd validity checks for BPF map/programs,
    from Mykyta Yatsenko.

* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (38 commits)
  selftests/bpf: Fix flaky test btf_map_in_map/lookup_update
  bpf: implement insn_is_cast_user() helper for JITs
  bpf: Avoid get_kernel_nofault() to fetch kprobe entry IP
  selftests/bpf: Use start_server in bpf_tcp_ca
  bpf: Sync uapi bpf.h to tools directory
  libbpf: Add new sec_def "sk_skb/verdict"
  selftests/bpf: Mark uprobe trigger functions with nocf_check attribute
  selftests/bpf: Use syscall(SYS_gettid) instead of gettid() wrapper in bench
  bpf-next: Avoid goto in regs_refine_cond_op()
  bpftool: Clean up HOST_CFLAGS, HOST_LDFLAGS for bootstrap bpftool
  selftests/bpf: scale benchmark counting by using per-CPU counters
  bpftool: Remove unnecessary source files from bootstrap version
  bpftool: Enable libbpf logs when loading pid_iter in debug mode
  selftests/bpf: add raw_tp/tp_btf BPF cookie subtests
  libbpf: add support for BPF cookie for raw_tp/tp_btf programs
  bpf: support BPF cookie in raw tracepoint (raw_tp, tp_btf) programs
  bpf: pass whole link instead of prog when triggering raw tracepoint
  bpf: flatten bpf_probe_register call chain
  selftests/bpf: Prevent client connect before server bind in test_tc_tunnel.sh
  selftests/bpf: Add a sk_msg prog bpf_get_ns_current_pid_tgid() test
  ...
====================

Link: https://lore.kernel.org/r/20240325233940.7154-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

5 files changed, 48 insertions, 26 deletions

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 4f20f62f9d63..62762390c93d 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1116,8 +1116,7 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image, void *i
 				void *func_addr);
 void *arch_alloc_bpf_trampoline(unsigned int size);
 void arch_free_bpf_trampoline(void *image, unsigned int size);
-void arch_protect_bpf_trampoline(void *image, unsigned int size);
-void arch_unprotect_bpf_trampoline(void *image, unsigned int size);
+int __must_check arch_protect_bpf_trampoline(void *image, unsigned int size);
 int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
 			     struct bpf_tramp_links *tlinks, void *func_addr);
 
@@ -1608,6 +1607,12 @@ struct bpf_tracing_link {
 	struct bpf_prog *tgt_prog;
 };
 
+struct bpf_raw_tp_link {
+	struct bpf_link link;
+	struct bpf_raw_event_map *btp;
+	u64 cookie;
+};
+
 struct bpf_link_primer {
 	struct bpf_link *link;
 	struct file *file;
diff --git a/include/linux/filter.h b/include/linux/filter.h
index c99bc3df2d28..44934b968b57 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -228,6 +228,16 @@ static inline bool insn_is_zext(const struct bpf_insn *insn)
 	return insn->code == (BPF_ALU | BPF_MOV | BPF_X) && insn->imm == 1;
 }
 
+/* addr_space_cast from as(0) to as(1) is for converting bpf arena pointers
+ * to pointers in user vma.
+ */
+static inline bool insn_is_cast_user(const struct bpf_insn *insn)
+{
+	return insn->code == (BPF_ALU64 | BPF_MOV | BPF_X) &&
+			      insn->off == BPF_ADDR_SPACE_CAST &&
+			      insn->imm == 1U << 16;
+}
+
 /* BPF_LD_IMM64 macro encodes single 'load 64-bit immediate' insn */
 #define BPF_LD_IMM64(DST, IMM)					\
 	BPF_LD_IMM64_RAW(DST, 0, IMM)
@@ -887,20 +897,22 @@ bpf_ctx_narrow_access_offset(u32 off, u32 size, u32 size_default)
 
 #define bpf_classic_proglen(fprog) (fprog->len * sizeof(fprog->filter[0]))
 
-static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
+static inline int __must_check bpf_prog_lock_ro(struct bpf_prog *fp)
 {
 #ifndef CONFIG_BPF_JIT_ALWAYS_ON
 	if (!fp->jited) {
 		set_vm_flush_reset_perms(fp);
-		set_memory_ro((unsigned long)fp, fp->pages);
+		return set_memory_ro((unsigned long)fp, fp->pages);
 	}
 #endif
+	return 0;
 }
 
-static inline void bpf_jit_binary_lock_ro(struct bpf_binary_header *hdr)
+static inline int __must_check
+bpf_jit_binary_lock_ro(struct bpf_binary_header *hdr)
 {
 	set_vm_flush_reset_perms(hdr);
-	set_memory_rox((unsigned long)hdr, hdr->size >> PAGE_SHIFT);
+	return set_memory_rox((unsigned long)hdr, hdr->size >> PAGE_SHIFT);
 }
 
 int sk_filter_trim_cap(struct sock *sk, struct sk_buff *skb, unsigned int cap);
diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
index 6f9bdfb09d1d..9df3e2973626 100644
--- a/include/linux/trace_events.h
+++ b/include/linux/trace_events.h
@@ -765,8 +765,11 @@ unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx);
 int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog, u64 bpf_cookie);
 void perf_event_detach_bpf_prog(struct perf_event *event);
 int perf_event_query_prog_array(struct perf_event *event, void __user *info);
-int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog *prog);
-int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_prog *prog);
+
+struct bpf_raw_tp_link;
+int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link);
+int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link);
+
 struct bpf_raw_event_map *bpf_get_raw_tracepoint(const char *name);
 void bpf_put_raw_tracepoint(struct bpf_raw_event_map *btp);
 int bpf_get_perf_event_info(const struct perf_event *event, u32 *prog_id,
@@ -794,11 +797,12 @@ perf_event_query_prog_array(struct perf_event *event, void __user *info)
 {
 	return -EOPNOTSUPP;
 }
-static inline int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog *p)
+struct bpf_raw_tp_link;
+static inline int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link)
 {
 	return -EOPNOTSUPP;
 }
-static inline int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_prog *p)
+static inline int bpf_probe_unregister(struct bpf_raw_event_map *btp, struct bpf_raw_tp_link *link)
 {
 	return -EOPNOTSUPP;
 }
@@ -909,31 +913,31 @@ void *perf_trace_buf_alloc(int size, struct pt_regs **regs, int *rctxp);
 int perf_event_set_bpf_prog(struct perf_event *event, struct bpf_prog *prog, u64 bpf_cookie);
 void perf_event_free_bpf_prog(struct perf_event *event);
 
-void bpf_trace_run1(struct bpf_prog *prog, u64 arg1);
-void bpf_trace_run2(struct bpf_prog *prog, u64 arg1, u64 arg2);
-void bpf_trace_run3(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run1(struct bpf_raw_tp_link *link, u64 arg1);
+void bpf_trace_run2(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2);
+void bpf_trace_run3(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3);
-void bpf_trace_run4(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run4(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4);
-void bpf_trace_run5(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run5(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4, u64 arg5);
-void bpf_trace_run6(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run6(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4, u64 arg5, u64 arg6);
-void bpf_trace_run7(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run7(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7);
-void bpf_trace_run8(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run8(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7,
 		    u64 arg8);
-void bpf_trace_run9(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run9(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		    u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7,
 		    u64 arg8, u64 arg9);
-void bpf_trace_run10(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run10(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		     u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7,
 		     u64 arg8, u64 arg9, u64 arg10);
-void bpf_trace_run11(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run11(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		     u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7,
 		     u64 arg8, u64 arg9, u64 arg10, u64 arg11);
-void bpf_trace_run12(struct bpf_prog *prog, u64 arg1, u64 arg2,
+void bpf_trace_run12(struct bpf_raw_tp_link *link, u64 arg1, u64 arg2,
 		     u64 arg3, u64 arg4, u64 arg5, u64 arg6, u64 arg7,
 		     u64 arg8, u64 arg9, u64 arg10, u64 arg11, u64 arg12);
 void perf_trace_run_bpf_submit(void *raw_data, int size, int rctx,
diff --git a/include/trace/bpf_probe.h b/include/trace/bpf_probe.h
index e609cd7da47e..a2ea11cc912e 100644
--- a/include/trace/bpf_probe.h
+++ b/include/trace/bpf_probe.h
@@ -46,8 +46,7 @@
 static notrace void							\
 __bpf_trace_##call(void *__data, proto)					\
 {									\
-	struct bpf_prog *prog = __data;					\
-	CONCATENATE(bpf_trace_run, COUNT_ARGS(args))(prog, CAST_TO_U64(args));	\
+	CONCATENATE(bpf_trace_run, COUNT_ARGS(args))(__data, CAST_TO_U64(args));	\
 }
 
 #undef DECLARE_EVENT_CLASS
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 3c42b9f1bada..9585f5345353 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -1662,8 +1662,10 @@ union bpf_attr {
 	} query;
 
 	struct { /* anonymous struct used by BPF_RAW_TRACEPOINT_OPEN command */
-		__u64 name;
-		__u32 prog_fd;
+		__u64		name;
+		__u32		prog_fd;
+		__u32		:32;
+		__aligned_u64	cookie;
 	} raw_tracepoint;
 
 	struct { /* anonymous struct for BPF_BTF_LOAD */