From 700d4796ef59f5faf240d307839bd419e2b6bdff Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Tue, 9 Feb 2021 19:36:26 -0800
Subject: bpf: Optimize program stats

Move bpf_prog_stats from prog->aux into prog to avoid one extra load
in critical path of program execution.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20210210033634.62081-2-alexei.starovoitov@gmail.com
---
 include/linux/bpf.h    |  8 --------
 include/linux/filter.h | 14 +++++++++++---
 2 files changed, 11 insertions(+), 11 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 079162bbd387..5a388955e6b6 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -14,7 +14,6 @@
 #include <linux/numa.h>
 #include <linux/mm_types.h>
 #include <linux/wait.h>
-#include <linux/u64_stats_sync.h>
 #include <linux/refcount.h>
 #include <linux/mutex.h>
 #include <linux/module.h>
@@ -507,12 +506,6 @@ enum bpf_cgroup_storage_type {
  */
 #define MAX_BPF_FUNC_ARGS 12
 
-struct bpf_prog_stats {
-	u64 cnt;
-	u64 nsecs;
-	struct u64_stats_sync syncp;
-} __aligned(2 * sizeof(u64));
-
 struct btf_func_model {
 	u8 ret_size;
 	u8 nr_args;
@@ -845,7 +838,6 @@ struct bpf_prog_aux {
 	u32 linfo_idx;
 	u32 num_exentries;
 	struct exception_table_entry *extable;
-	struct bpf_prog_stats __percpu *stats;
 	union {
 		struct work_struct work;
 		struct rcu_head	rcu;
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 5b3137d7b690..cecb03c9d251 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -22,6 +22,7 @@
 #include <linux/vmalloc.h>
 #include <linux/sockptr.h>
 #include <crypto/sha1.h>
+#include <linux/u64_stats_sync.h>
 
 #include <net/sch_generic.h>
 
@@ -539,6 +540,12 @@ struct bpf_binary_header {
 	u8 image[] __aligned(BPF_IMAGE_ALIGNMENT);
 };
 
+struct bpf_prog_stats {
+	u64 cnt;
+	u64 nsecs;
+	struct u64_stats_sync syncp;
+} __aligned(2 * sizeof(u64));
+
 struct bpf_prog {
 	u16			pages;		/* Number of allocated pages */
 	u16			jited:1,	/* Is our filter JIT'ed? */
@@ -557,10 +564,11 @@ struct bpf_prog {
 	u32			len;		/* Number of filter blocks */
 	u32			jited_len;	/* Size of jited insns in bytes */
 	u8			tag[BPF_TAG_SIZE];
-	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
-	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */
+	struct bpf_prog_stats __percpu *stats;
 	unsigned int		(*bpf_func)(const void *ctx,
 					    const struct bpf_insn *insn);
+	struct bpf_prog_aux	*aux;		/* Auxiliary fields */
+	struct sock_fprog_kern	*orig_prog;	/* Original BPF program */
 	/* Instructions for interpreter */
 	struct sock_filter	insns[0];
 	struct bpf_insn		insnsi[];
@@ -581,7 +589,7 @@ DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key);
 		struct bpf_prog_stats *__stats;				\
 		u64 __start = sched_clock();				\
 		__ret = dfunc(ctx, (prog)->insnsi, (prog)->bpf_func);	\
-		__stats = this_cpu_ptr(prog->aux->stats);		\
+		__stats = this_cpu_ptr(prog->stats);			\
 		u64_stats_update_begin(&__stats->syncp);		\
 		__stats->cnt++;						\
 		__stats->nsecs += sched_clock() - __start;		\
-- 
cgit v1.2.3