bcc: Update to master tip

In order to use latest clang, latest bcc sources are needed, this also
means that we need to use vendored libbpf as thats the newest one needed
and avoids dependencies on kernel bpf headers which maybe old and out of
sync

examples use error.h header which is not available on musl therefore
disable examples when building on musl

Signed-off-by: Khem Raj <raj.khem@gmail.com>

1 files changed, 0 insertions, 50 deletions

diff --git a/dynamic-layers/openembedded-layer/recipes-devtools/bcc/bcc/0001-tools-opensnoop-snoop-do_sys_openat2-for-kernel-v5.6.patch b/dynamic-layers/openembedded-layer/recipes-devtools/bcc/bcc/0001-tools-opensnoop-snoop-do_sys_openat2-for-kernel-v5.6.patch
deleted file mode 100644
index fb659cc..0000000
--- a/dynamic-layers/openembedded-layer/recipes-devtools/bcc/bcc/0001-tools-opensnoop-snoop-do_sys_openat2-for-kernel-v5.6.patch
+++ /dev/null
@@ -1,50 +0,0 @@
-From 8e12b10e7576a6d47e0dc2cdc36caeb9ba26fa12 Mon Sep 17 00:00:00 2001
-From: He Zhe <zhe.he@windriver.com>
-Date: Mon, 15 Jun 2020 07:05:24 +0000
-Subject: [PATCH] tools: opensnoop: snoop do_sys_openat2 for kernel v5.6 and
- later
-
-Since kernel v5.6, fddb5d430ad9 ("open: introduce openat2(2) syscall"),
-do_sys_openat2 instead of do_sys_open has been used as entry function for open.
-
-Upstream-Status: Inappropriate, upstream context has changed and needs more
-                 tweak.
-
-Signed-off-by: He Zhe <zhe.he@windriver.com>
----
- tools/opensnoop.py | 12 ++++++++++--
- 1 file changed, 10 insertions(+), 2 deletions(-)
-
-diff --git a/tools/opensnoop.py b/tools/opensnoop.py
-index 51d3dc05..522812d4 100755
---- a/tools/opensnoop.py
-+++ b/tools/opensnoop.py
-@@ -22,6 +22,8 @@ from bcc.utils import printb
- import argparse
- from datetime import datetime, timedelta
- import os
-+import platform
-+from pkg_resources import parse_version
- 
- # arguments
- examples = """examples:
-@@ -235,8 +237,14 @@ if debug or args.ebpf:
- # initialize BPF
- b = BPF(text=bpf_text)
- if not is_support_kfunc:
--    b.attach_kprobe(event="do_sys_open", fn_name="trace_entry")
--    b.attach_kretprobe(event="do_sys_open", fn_name="trace_return")
-+    # Since kernel v5.6, do_sys_openat2 instead of do_sys_open has been used as entry function for open
-+    if parse_version(platform.uname().release.split('-')[0]) > parse_version('5.6.0'):
-+        entry = "do_sys_openat2"
-+    else:
-+        entry = "do_sys_open"
-+
-+    b.attach_kprobe(event=entry, fn_name="trace_entry")
-+    b.attach_kretprobe(event=entry, fn_name="trace_return")
- 
- initial_ts = 0
- 
--- 
-2.17.1
-