Re: BPF skels in perf .Re: [GIT PULL] perf tools changes for v6.4

[Arnaldo Carvalho de Melo <acme@kernel.org>]

Em Thu, May 04, 2023 at 07:01:51PM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Thu, May 04, 2023 at 06:48:50PM -0300, Arnaldo Carvalho de Melo escreveu:
> > Em Thu, May 04, 2023 at 04:07:29PM -0300, Arnaldo Carvalho de Melo escreveu:
> > > Em Thu, May 04, 2023 at 11:50:07AM -0700, Andrii Nakryiko escreveu:
> > > > On Thu, May 4, 2023 at 10:52 AM Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> > > > > Andrii, can you add some more information about the usage of vmlinux.h
> > > > > instead of using kernel headers?
> > >  
> > > > I'll just say that vmlinux.h is not a hard requirement to build BPF
> > > > programs, it's more a convenience allowing easy access to definitions
> > > > of both UAPI and kernel-internal structures for tracing needs and
> > > > marking them relocatable using BPF CO-RE machinery. Lots of real-world
> > > > applications just check-in pregenerated vmlinux.h to avoid build-time
> > > > dependency on up-to-date host kernel and such.
> > >  
> > > > If vmlinux.h generation and usage is causing issues, though, given
> > > > that perf's BPF programs don't seem to be using many different kernel
> > > > types, it might be a better option to just use UAPI headers for public
> > > > kernel type definitions, and just define CO-RE-relocatable minimal
> > > > definitions locally in perf's BPF code for the other types necessary.
> > > > E.g., if perf needs only pid and tgid from task_struct, this would
> > > > suffice:
> > >  
> > > > struct task_struct {
> > > >     int pid;
> > > >     int tgid;
> > > > } __attribute__((preserve_access_index));
> > > 
> > > Yeah, that seems like a way better approach, no vmlinux involved, libbpf
> > > CO-RE notices that task_struct changed from this two integers version
> > > (of course) and does the relocation to where it is in the running kernel
> > > by using /sys/kernel/btf/vmlinux.
> > 
> > Doing it for one of the skels, build tested, runtime untested, but not
> > using any vmlinux, BTF to help, not that bad, more verbose, but at least
> > we state what are the fields we actually use, have those attribute
> > documenting that those offsets will be recorded for future use, etc.
> > 

Namhyung, can you please check that this one for the recent sample works?

From c6972dae6c962d7be5ba006ab90c9955268debc5 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Fri, 5 May 2023 09:55:18 -0300
Subject: [PATCH 1/2] perf sample_filter.bpf: Stop using vmlinux.h generated by
 bpftool, use CO-RE

Including linux/bpf.h and linux/perf_events.h we get the UAPI structs
and then define a subset  'struct perf_sample_data' with the fields we
use in this tool while using __attribute__((preserve_access_index)) so
that at libbpf load time it can fixup the offsets according to the
'struct perf_data_sample' obtained from the running kernel BTF
(/sys/kernel/btf/vmlinux).

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/bpf_skel/sample_filter.bpf.c | 37 +++++++++++++++++++-
 1 file changed, 36 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/bpf_skel/sample_filter.bpf.c b/tools/perf/util/bpf_skel/sample_filter.bpf.c
index cffe493af1ed5f31..045532c2366d74ef 100644
--- a/tools/perf/util/bpf_skel/sample_filter.bpf.c
+++ b/tools/perf/util/bpf_skel/sample_filter.bpf.c
@@ -1,12 +1,47 @@
 // SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
 // Copyright (c) 2023 Google
-#include "vmlinux.h"
+#include <linux/bpf.h>
+#include <linux/perf_event.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 #include <bpf/bpf_core_read.h>
 
 #include "sample-filter.h"
 
+// non-UAPI kernel data structures, just the fields used in this tool,
+// preserving the access index so that libbpf can fixup offsets with the ones
+// used in the kernel when loading the BPF bytecode, if they differ from what
+// is used here.
+
+struct perf_sample_data {
+	__u64			 addr;
+	__u64			 period;
+	union perf_sample_weight weight;
+	__u64			 txn;
+	union perf_mem_data_src  data_src;
+	__u64			 ip;
+	struct {
+		__u32		 pid;
+		__u32		 tid;
+	} tid_entry;
+	__u64			 time;
+	__u64			 id;
+	struct {
+		__u32		 cpu;
+	} cpu_entry;
+	__u64			 phys_addr;
+	__u64			 data_page_size;
+	__u64			 code_page_size;
+} __attribute__((__aligned__(64))) __attribute__((preserve_access_index));
+
+struct bpf_perf_event_data_kern {
+	struct perf_sample_data *  data;
+	struct perf_event *        event;
+
+	/* size: 24, cachelines: 1, members: 3 */
+	/* last cacheline: 24 bytes */
+} __attribute__((preserve_access_index));
+
 /* BPF map that will be filled by user space */
 struct filters {
 	__uint(type, BPF_MAP_TYPE_ARRAY);
-- 
2.39.2