linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wang Nan <wangnan0@huawei.com>
To: <namhyung@kernel.org>, <alexei.starovoitov@gmail.com>,
	<masami.hiramatsu.pt@hitachi.com>, <acme@kernel.org>,
	<paulus@samba.org>, <a.p.zijlstra@chello.nl>, <mingo@redhat.com>,
	<jolsa@kernel.org>, <dsahern@gmail.com>, <daniel@iogearbox.net>,
	<brendan.d.gregg@gmail.com>
Cc: <lizefan@huawei.com>, <hekuang@huawei.com>, <xiakaixu@huawei.com>,
	<linux-kernel@vger.kernel.org>, <pi3orama@163.com>
Subject: [RFC PATCH v5 17/30] bpf tools: Relocate eBPF programs
Date: Mon, 1 Jun 2015 07:38:03 +0000	[thread overview]
Message-ID: <1433144296-74992-18-git-send-email-wangnan0@huawei.com> (raw)
In-Reply-To: <1433144296-74992-1-git-send-email-wangnan0@huawei.com>

If an eBPF program accesses a map, LLVM generates a load instruction
which loads an absolute address into a register, like this:

  ld_64   r1, <MCOperand Expr:(mymap)>
  ...
  call    2

That ld_64 instruction will be recorded in relocation section.
To enable the usage of that map, relocation must be done by replacing
the immediate value by real map file descriptor so it can be found by
eBPF map functions.

This patch to the relocation work based on information collected by
patch 'bpf tools: Collect relocation sections from SHT_REL sections'.
For each instruction which needs relocation, it inject corresponding
file descriptor to imm field. As a part of protocol, src_reg is set to
BPF_PSEUDO_MAP_FD to notify kernel this is a map loading instruction.

Signed-off-by: Wang Nan <wangnan0@huawei.com>
---
 tools/lib/bpf/libbpf.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 51 insertions(+)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 6e4fc62..c581baa 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -596,6 +596,55 @@ bpf_object__create_maps(struct bpf_object *obj)
 	return 0;
 }
 
+static int
+bpf_program__relocate(struct bpf_program *prog, int *map_fds)
+{
+	int i;
+
+	if (!prog || !prog->reloc_desc)
+		return 0;
+
+	for (i = 0; i < prog->nr_reloc; i++) {
+		int insn_idx, map_idx;
+		struct bpf_insn *insns = prog->insns;
+
+		insn_idx = prog->reloc_desc[i].insn_idx;
+		map_idx = prog->reloc_desc[i].map_idx;
+
+		if (insn_idx >= (int)prog->insns_cnt) {
+			pr_warning("relocation out of range: '%s'\n",
+				   prog->section_name);
+			return -ERANGE;
+		}
+		insns[insn_idx].src_reg = BPF_PSEUDO_MAP_FD;
+		insns[insn_idx].imm = map_fds[map_idx];
+	}
+
+	zfree(&prog->reloc_desc);
+	prog->nr_reloc = 0;
+	return 0;
+}
+
+
+static int
+bpf_object__relocate(struct bpf_object *obj)
+{
+	struct bpf_program *prog;
+	size_t i;
+	int err;
+
+	for (i = 0; i < obj->nr_programs; i++) {
+		prog = &obj->programs[i];
+
+		if ((err = bpf_program__relocate(prog, obj->map_fds))) {
+			pr_warning("failed to relocate '%s'\n",
+				   prog->section_name);
+			return err;
+		}
+	}
+	return 0;
+}
+
 static int bpf_object__collect_reloc(struct bpf_object *obj)
 {
 	int i, err;
@@ -709,6 +758,8 @@ int bpf_object__load(struct bpf_object *obj)
 	obj->loaded = true;
 	if (bpf_object__create_maps(obj))
 		goto out;
+	if (bpf_object__relocate(obj))
+		goto out;
 
 	return 0;
 out:
-- 
1.8.3.4


  parent reply	other threads:[~2015-06-01  9:56 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-01  7:37 [RFC PATCH v5 00/30] perf tools: filtering events using eBPF programs Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 01/30] perf tools: Move linux/kernel.h to tools/include Wang Nan
2015-06-04 14:11   ` [tip:perf/core] " tip-bot for Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 02/30] perf tools: Move linux/{list.h,poison.h} " Wang Nan
2015-06-04 14:11   ` [tip:perf/core] tools: Move tools/perf/util/include/linux/{list.h ,poison.h} " tip-bot for Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 03/30] tools build: Add feature check for eBPF API Wang Nan
2015-06-02 22:11   ` Arnaldo Carvalho de Melo
2015-06-03  1:15     ` Wangnan (F)
2015-06-03 13:47       ` Arnaldo Carvalho de Melo
2015-06-01  7:37 ` [RFC PATCH v5 04/30] bpf tools: Introduce 'bpf' library to tools Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 05/30] bpf tools: Allow caller to set printing function Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 06/30] bpf tools: Open eBPF object file and do basic validation Wang Nan
2015-06-03 21:33   ` Arnaldo Carvalho de Melo
2015-06-03 21:35     ` Arnaldo Carvalho de Melo
2015-06-04  2:18       ` Wangnan (F)
2015-06-01  7:37 ` [RFC PATCH v5 07/30] bpf tools: Check endianess and make libbpf fail early Wang Nan
2015-06-03 21:52   ` Arnaldo Carvalho de Melo
2015-06-01  7:37 ` [RFC PATCH v5 08/30] bpf tools: Iterate over ELF sections to collect information Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 09/30] bpf tools: Collect version and license from ELF sections Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 10/30] bpf tools: Collect map definitions from 'maps' section Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 11/30] bpf tools: Collect symbol table from SHT_SYMTAB section Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 12/30] bpf tools: Collect eBPF programs from their own sections Wang Nan
2015-06-01  7:37 ` [RFC PATCH v5 13/30] bpf tools: Collect relocation sections from SHT_REL sections Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 14/30] bpf tools: Record map accessing instructions for each program Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 15/30] bpf tools: Add bpf.c/h for common bpf operations Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 16/30] bpf tools: Create eBPF maps defined in an object file Wang Nan
2015-06-01  7:38 ` Wang Nan [this message]
2015-06-01  7:38 ` [RFC PATCH v5 18/30] bpf tools: Introduce bpf_load_program() to bpf.c Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 19/30] bpf tools: Load eBPF programs in object files into kernel Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 20/30] bpf tools: Introduce accessors for struct bpf_program Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 21/30] bpf tools: Introduce accessors for struct bpf_object Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 22/30] bpf tools: Link all bpf objects onto a list Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 23/30] perf tools: Make perf depend on libbpf Wang Nan
2015-06-03  2:45   ` [RFC PATCH v6 " Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 24/30] perf record: Enable passing bpf object file to --event Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 25/30] perf tools: Parse probe points of eBPF programs during preparation Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 26/30] perf probe: Attach trace_probe_event with perf_probe_event Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 27/30] perf record: Probe at kprobe points Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 28/30] perf record: Load all eBPF object into kernel Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 29/30] perf tools: Add bpf_fd field to evsel and config it Wang Nan
2015-06-01  7:38 ` [RFC PATCH v5 30/30] perf tools: Attach eBPF program to perf event Wang Nan
2015-06-02  0:48 ` [RFC PATCH v5 00/30] perf tools: filtering events using eBPF programs Alexei Starovoitov
2015-06-02  1:15   ` Arnaldo Carvalho de Melo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1433144296-74992-18-git-send-email-wangnan0@huawei.com \
    --to=wangnan0@huawei.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@kernel.org \
    --cc=alexei.starovoitov@gmail.com \
    --cc=brendan.d.gregg@gmail.com \
    --cc=daniel@iogearbox.net \
    --cc=dsahern@gmail.com \
    --cc=hekuang@huawei.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lizefan@huawei.com \
    --cc=masami.hiramatsu.pt@hitachi.com \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=paulus@samba.org \
    --cc=pi3orama@163.com \
    --cc=xiakaixu@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).