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.

Yang, can you please check that this works?


From bd6289bc3ffc89aecad3bd8798d76626c8c16d39 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Fri, 5 May 2023 10:13:09 -0300
Subject: [PATCH 1/1] perf kwork_trace.bpf: Stop using vmlinux.h, grab copies
 of used structs

And mark them with __attribute__((preserve_access_index)) so that
libbpf's CO-RE code can fixup offsets if they differ with the kernel
data structure.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
 tools/perf/util/bpf_skel/kwork_trace.bpf.c | 70 +++++++++++++++++++++-
 1 file changed, 69 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/bpf_skel/kwork_trace.bpf.c b/tools/perf/util/bpf_skel/kwork_trace.bpf.c
index 063c124e099938ed..e38fe54c7667fa74 100644
--- a/tools/perf/util/bpf_skel/kwork_trace.bpf.c
+++ b/tools/perf/util/bpf_skel/kwork_trace.bpf.c
@@ -1,13 +1,81 @@
 // SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
 // Copyright (c) 2022, Huawei
 
-#include "vmlinux.h"
+#include <linux/bpf.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
 #define KWORK_COUNT 100
 #define MAX_KWORKNAME 128
 
+
+// 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.
+
+enum {
+	HI_SOFTIRQ = 0,
+	TIMER_SOFTIRQ,
+	NET_TX_SOFTIRQ,
+	NET_RX_SOFTIRQ,
+	BLOCK_SOFTIRQ,
+	IRQ_POLL_SOFTIRQ,
+	TASKLET_SOFTIRQ,
+	SCHED_SOFTIRQ,
+	HRTIMER_SOFTIRQ,
+	RCU_SOFTIRQ,    /* Preferable RCU should always be the last softirq */
+
+	NR_SOFTIRQS
+};
+
+struct trace_entry {
+	short unsigned int type;
+	unsigned char	   flags;
+	unsigned char	   preempt_count;
+	int		   pid;
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_irq_handler_entry {
+	struct trace_entry ent;
+	int		   irq;
+	__u32		   __data_loc_name;
+	char		   __data[];
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_irq_handler_exit {
+	struct trace_entry ent;
+	int		   irq;
+	int		   ret;
+	char		   __data[];
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_softirq {
+	struct trace_entry ent;
+	unsigned int	   vec;
+	char		   __data[];
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_workqueue_execute_start {
+	struct trace_entry ent;
+	void		   *work;
+	void		   *function;
+	char		   __data[];
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_workqueue_execute_end {
+	struct trace_entry ent;
+	void		   *work;
+	void		   *function;
+	char		  __data[];
+} __attribute__((preserve_access_index));
+
+struct trace_event_raw_workqueue_activate_work {
+	struct trace_entry ent;
+	void		   *work;
+	char		   __data[];
+} __attribute__((preserve_access_index));
+
 /*
  * This should be in sync with "util/kwork.h"
  */
-- 
2.39.2