[v4,bpf-next,00/14] Support BTF-powered BPF tracing programs for kernel modules

Message ID	20201202001616.3378929-1-andrii@kernel.org
Headers	show Return-Path: <netdev-owner@kernel.org> From: Andrii Nakryiko <andrii@kernel.org> To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>, <ast@fb.com>, <daniel@iogearbox.net> CC: <andrii@kernel.org>, <kernel-team@fb.com> Subject: [PATCH v4 bpf-next 00/14] Support BTF-powered BPF tracing programs for kernel modules Date: Tue, 1 Dec 2020 16:16:02 -0800 Message-ID: <20201202001616.3378929-1-andrii@kernel.org> Content-Type: text/plain Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 Precedence: bulk
Series	Support BTF-powered BPF tracing programs for kernel modules \| expand [v4,bpf-next,00/14] Support BTF-powered BPF tracing programs for kernel modules [v4,bpf-next,01/14] bpf: fix bpf_put_raw_tracepoint()'s use of __module_address() [v4,bpf-next,02/14] bpf: keep module's btf_data_size intact after load [v4,bpf-next,03/14] libbpf: add internal helper to load BTF data by FD [v4,bpf-next,04/14] libbpf: refactor CO-RE relocs to not assume a single BTF object [v4,bpf-next,05/14] libbpf: add kernel module BTF support for CO-RE relocations [v4,bpf-next,06/14] selftests/bpf: add bpf_testmod kernel module for testing [v4,bpf-next,07/14] selftests/bpf: add support for marking sub-tests as skipped [v4,bpf-next,08/14] selftests/bpf: add CO-RE relocs selftest relying on kernel module BTF [v4,bpf-next,09/14] bpf: remove hard-coded btf_vmlinux assumption from BPF verifier [v4,bpf-next,10/14] bpf: allow to specify kernel module BTFs when attaching BPF programs [v4,bpf-next,11/14] libbpf: factor out low-level BPF program loading helper [v4,bpf-next,12/14] libbpf: support attachment of BPF tracing programs to kernel modules [v4,bpf-next,13/14] selftests/bpf: add tp_btf CO-RE reloc test for modules [v4,bpf-next,14/14] selftests/bpf: add fentry/fexit/fmod_ret selftest for kernel module

Message ID

20201202001616.3378929-1-andrii@kernel.org

Headers

From: Andrii Nakryiko <andrii@kernel.org>
To: <bpf@vger.kernel.org>, <netdev@vger.kernel.org>, <ast@fb.com>,
	<daniel@iogearbox.net>
CC: <andrii@kernel.org>, <kernel-team@fb.com>
Subject: [PATCH v4 bpf-next 00/14] Support BTF-powered BPF tracing programs
	for kernel modules
Date: Tue, 1 Dec 2020 16:16:02 -0800
Message-ID: <20201202001616.3378929-1-andrii@kernel.org>
Content-Type: text/plain
Content-Transfer-Encoding: 8BIT
MIME-Version: 1.0
Precedence: bulk

Series

Support BTF-powered BPF tracing programs for kernel modules | expand

Message

Andrii Nakryiko Dec. 2, 2020, 12:16 a.m. UTC

This patch sets extends kernel and libbpf with support for attaching
BTF-powered raw tracepoint (tp_btf) and tracing (fentry/fexit/fmod_ret/lsm)
BPF programs to BPF hooks defined in kernel modules. As part of that, libbpf
now supports performing CO-RE relocations against types in kernel module BTFs,
in addition to existing vmlinux BTF support.

Kernel UAPI for BPF_PROG_LOAD is extended with extra parameter
(attach_btf_obj_id) which allows to specify kernel module BTF in which the BTF
type is identifed by attach_btf_id.

This patch set also sets up a convenient and fully-controlled custom kernel
module (called "bpf_testmod"), that is a predictable playground for all the
BPF selftests, that rely on module BTFs. Currently pahole doesn't generate
BTF_KIND_FUNC info for ftrace-able static functions in kernel modules, so
expose traced function in bpf_sidecar.ko. Once pahole is enhanced, we can go
back to static function.