From: "Wangnan (F)" <wangnan0@huawei.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: <ast@plumgrid.com>, <linux-kernel@vger.kernel.org>,
<lizefan@huawei.com>, <hekuang@huawei.com>, <xiakaixu@huawei.com>,
<pi3orama@163.com>
Subject: Re: [PATCH 06/39] bpf tools: Create eBPF maps defined in an object file
Date: Tue, 14 Jul 2015 11:59:25 +0800 [thread overview]
Message-ID: <55A4891D.9070600@huawei.com> (raw)
In-Reply-To: <20150713195453.GF2885@kernel.org>
On 2015/7/14 3:54, Arnaldo Carvalho de Melo wrote:
> Em Thu, Jul 09, 2015 at 12:35:09PM +0000, Wang Nan escreveu:
>
> <SNIP>
>
>> +static int
>> +bpf_object__create_maps(struct bpf_object *obj)
>> +{
>> + unsigned int i;
>> + size_t nr_maps;
>> + int *pfd;
>> +
>> + nr_maps = obj->maps_buf_sz / sizeof(struct bpf_map_def);
>> + if (!obj->maps_buf || !nr_maps) {
>> + pr_debug("don't need create maps for %s\n",
>> + obj->path);
>> + return 0;
>> + }
>> +
>> + obj->map_fds = malloc(sizeof(int) * nr_maps);
>> + if (!obj->map_fds) {
>> + pr_warning("realloc perf_bpf_map_fds failed\n");
>> + return -ENOMEM;
>> + }
>> + obj->nr_map_fds = nr_maps;
>> +
>> + /* fill all fd with -1 */
>> + memset(obj->map_fds, 0xff, sizeof(int) * nr_maps);
> Huh? Each entry is an int... If you want to set it all to -1, why don't
> you do it as:
>
> memset(obj->map_fds, -1, sizeof(int) * nr_maps);
>
> That way we can even remove that comment :) Doing it, please let me know
> if this has some magic twist that makes it be right as-is, if so, please
> provide a better comment.
Yes, it should be better.
Thank you,
> - Arnaldo
>
>> +
>> + pfd = obj->map_fds;
>> + for (i = 0; i < nr_maps; i++) {
>> + struct bpf_map_def def;
>> +
>> + def = *(struct bpf_map_def *)(obj->maps_buf +
>> + i * sizeof(struct bpf_map_def));
>> +
>> + *pfd = bpf_create_map(def.type,
>> + def.key_size,
>> + def.value_size,
>> + def.max_entries);
>> + if (*pfd < 0) {
>> + size_t j;
>> + int err = *pfd;
>> +
>> + pr_warning("failed to create map: %s\n",
>> + strerror(errno));
>> + for (j = 0; j < i; j++)
>> + zclose(obj->map_fds[j]);
>> + obj->nr_map_fds = 0;
>> + zfree(&obj->map_fds);
>> + return err;
>> + }
>> + pr_debug("create map: fd=%d\n", *pfd);
>> + pfd++;
>> + }
>> +
>> + zfree(&obj->maps_buf);
>> + obj->maps_buf_sz = 0;
>> + return 0;
>> +}
>> +
>> static int bpf_object__collect_reloc(struct bpf_object *obj)
>> {
>> int i, err;
>> @@ -671,6 +736,42 @@ struct bpf_object *bpf_object__open_buffer(void *obj_buf,
>> return __bpf_object__open("[buffer]", obj_buf, obj_buf_sz);
>> }
>>
>> +int bpf_object__unload(struct bpf_object *obj)
>> +{
>> + size_t i;
>> +
>> + if (!obj)
>> + return -EINVAL;
>> +
>> + for (i = 0; i < obj->nr_map_fds; i++)
>> + zclose(obj->map_fds[i]);
>> + zfree(&obj->map_fds);
>> + obj->nr_map_fds = 0;
>> +
>> + return 0;
>> +}
>> +
>> +int bpf_object__load(struct bpf_object *obj)
>> +{
>> + if (!obj)
>> + return -EINVAL;
>> +
>> + if (obj->loaded) {
>> + pr_warning("object should not be loaded twice\n");
>> + return -EINVAL;
>> + }
>> +
>> + obj->loaded = true;
>> + if (bpf_object__create_maps(obj))
>> + goto out;
>> +
>> + return 0;
>> +out:
>> + bpf_object__unload(obj);
>> + pr_warning("failed to load object '%s'\n", obj->path);
>> + return -EINVAL;
>> +}
>> +
>> void bpf_object__close(struct bpf_object *obj)
>> {
>> size_t i;
>> @@ -679,6 +780,7 @@ void bpf_object__close(struct bpf_object *obj)
>> return;
>>
>> bpf_object__elf_finish(obj);
>> + bpf_object__unload(obj);
>>
>> zfree(&obj->maps_buf);
>>
>> diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
>> index 6e75acd..3e69600 100644
>> --- a/tools/lib/bpf/libbpf.h
>> +++ b/tools/lib/bpf/libbpf.h
>> @@ -30,6 +30,10 @@ struct bpf_object *bpf_object__open_buffer(void *obj_buf,
>> size_t obj_buf_sz);
>> void bpf_object__close(struct bpf_object *object);
>>
>> +/* Load/unload object into/from kernel */
>> +int bpf_object__load(struct bpf_object *obj);
>> +int bpf_object__unload(struct bpf_object *obj);
>> +
>> /*
>> * We don't need __attribute__((packed)) now since it is
>> * unnecessary for 'bpf_map_def' because they are all aligned.
>> --
>> 1.8.3.4
next prev parent reply other threads:[~2015-07-14 4:03 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-09 12:35 [GIT PULL 00/39] perf tools: filtering events using eBPF programs Wang Nan
2015-07-09 12:35 ` [PATCH 01/39] bpf tools: Collect symbol table from SHT_SYMTAB section Wang Nan
2015-07-09 12:35 ` [PATCH 02/39] bpf tools: Collect eBPF programs from their own sections Wang Nan
2015-07-09 15:58 ` Arnaldo Carvalho de Melo
2015-07-10 3:07 ` Wangnan (F)
2015-07-13 19:51 ` Arnaldo Carvalho de Melo
2015-07-14 3:58 ` Wangnan (F)
2015-07-09 12:35 ` [PATCH 03/39] bpf tools: Collect relocation sections from SHT_REL sections Wang Nan
2015-07-09 12:35 ` [PATCH 04/39] bpf tools: Record map accessing instructions for each program Wang Nan
2015-07-09 12:35 ` [PATCH 05/39] bpf tools: Add bpf.c/h for common bpf operations Wang Nan
2015-07-09 12:35 ` [PATCH 06/39] bpf tools: Create eBPF maps defined in an object file Wang Nan
2015-07-13 19:54 ` Arnaldo Carvalho de Melo
2015-07-14 3:59 ` Wangnan (F) [this message]
2015-07-09 12:35 ` [PATCH 07/39] bpf tools: Relocate eBPF programs Wang Nan
2015-07-09 12:35 ` [PATCH 08/39] bpf tools: Introduce bpf_load_program() to bpf.c Wang Nan
2015-07-09 12:35 ` [PATCH 09/39] bpf tools: Load eBPF programs in object files into kernel Wang Nan
2015-07-09 12:35 ` [PATCH 10/39] bpf tools: Introduce accessors for struct bpf_program Wang Nan
2015-07-09 12:35 ` [PATCH 11/39] bpf tools: Link all bpf objects onto a list Wang Nan
2015-07-09 12:35 ` [PATCH 12/39] perf tools: Introduce llvm config options Wang Nan
2015-07-21 11:13 ` [PATCH 12/39 update] " Wang Nan
2015-08-08 8:16 ` [tip:perf/core] " tip-bot for Wang Nan
2015-07-09 12:35 ` [PATCH 13/39] perf tools: Call clang to compile C source to object code Wang Nan
2015-07-09 12:35 ` [PATCH 14/39] perf tools: Auto detecting kernel build directory Wang Nan
2015-07-13 21:46 ` Arnaldo Carvalho de Melo
2015-07-14 6:56 ` Wangnan (F)
2015-07-14 7:16 ` Wangnan (F)
2015-07-09 12:35 ` [PATCH 15/39] perf tools: Auto detecting kernel include options Wang Nan
2015-08-08 8:17 ` [tip:perf/core] " tip-bot for Wang Nan
2015-07-09 12:35 ` [PATCH 16/39] perf tests: Add LLVM test for eBPF on-the-fly compiling Wang Nan
2015-07-15 10:54 ` [PATCH updated " Wang Nan
2015-07-22 4:46 ` [PATCH 16/39 RESEND] " Wang Nan
2015-07-09 12:35 ` [PATCH 17/39] perf tools: Make perf depend on libbpf Wang Nan
2015-07-09 12:35 ` [PATCH 18/39] perf record: Enable passing bpf object file to --event Wang Nan
2015-07-09 12:35 ` [PATCH 19/39] perf record: Compile scriptlets if pass '.c' " Wang Nan
2015-07-09 12:35 ` [PATCH 20/39] perf tools: Parse probe points of eBPF programs during preparation Wang Nan
2015-07-09 12:35 ` [PATCH 21/39] perf probe: Attach trace_probe_event with perf_probe_event Wang Nan
2015-07-09 12:35 ` [PATCH 22/39] perf record: Probe at kprobe points Wang Nan
2015-07-09 12:35 ` [PATCH 23/39] perf record: Load all eBPF object into kernel Wang Nan
2015-07-09 12:35 ` [PATCH 24/39] perf tools: Add bpf_fd field to evsel and config it Wang Nan
2015-07-09 12:35 ` [PATCH 25/39] perf tools: Attach eBPF program to perf event Wang Nan
2015-07-09 12:35 ` [PATCH 26/39] perf tools: Suppress probing messages when probing by BPF loading Wang Nan
2015-07-09 12:35 ` [PATCH 27/39] perf record: Add clang options for compiling BPF scripts Wang Nan
2015-07-09 12:35 ` [PATCH 28/39] bpf tools: Load a program with different instances using preprocessor Wang Nan
2015-07-13 5:44 ` Wangnan (F)
2015-07-09 12:35 ` [PATCH 29/39] perf tools: Fix probe-event.h include Wang Nan
2015-07-09 12:35 ` [PATCH 30/39] perf probe: Reset args and nargs for probe_trace_event when failure Wang Nan
2015-07-09 12:35 ` [PATCH 31/39] perf tools: Move linux/filter.h to tools/include Wang Nan
2015-07-09 12:35 ` [PATCH 32/39] perf tools: Add BPF_PROLOGUE config options for further patches Wang Nan
2015-07-09 12:35 ` [PATCH 33/39] perf tools: Introduce arch_get_reg_info() for x86 Wang Nan
2015-07-09 12:35 ` [PATCH 34/39] perf tools: Add prologue for BPF programs for fetching arguments Wang Nan
2015-07-09 12:35 ` [PATCH 35/39] perf tools: Generate prologue for BPF programs Wang Nan
2015-07-09 12:35 ` [PATCH 36/39] perf tools: Use same BPF program if arguments are identical Wang Nan
2015-07-09 12:35 ` [PATCH 37/39] perf record: Support custom vmlinux path Wang Nan
2015-07-09 12:35 ` [PATCH 38/39] perf probe: Init symbol as kprobe if any event is kprobe Wang Nan
2015-07-09 12:35 ` [PATCH 39/39] perf tools: Support attach BPF program on uprobe events Wang Nan
2015-07-14 15:36 ` perf test LLVM was: Re: [GIT PULL 00/39] perf tools: filtering events using eBPF programs Arnaldo Carvalho de Melo
2015-07-15 10:49 ` Wangnan (F)
2015-07-15 11:20 ` Arnaldo Carvalho de Melo
2015-07-17 2:34 ` Wangnan (F)
[not found] ` <CA+JHD93=ZMFP0gB1NqTXGBjTfSyYC-53Uj7r11gQJnKtQRKcDg@mail.gmail.com>
2015-07-21 11:09 ` Wangnan (F)
2015-07-21 11:41 ` Arnaldo Carvalho de Melo
2015-07-22 4:40 ` Wangnan (F)
2015-07-31 15:35 ` perf eBPF patch ordering. was: " Arnaldo Carvalho de Melo
2015-07-31 20:31 ` Arnaldo Carvalho de Melo
2015-08-03 2:37 ` Wangnan (F)
2015-08-03 15:07 ` Arnaldo Carvalho de Melo
2015-08-03 15:19 ` Arnaldo Carvalho de Melo
2015-08-03 15:53 ` pi3orama
2015-08-03 16:11 ` Arnaldo Carvalho de Melo
2015-08-03 19:49 ` Arnaldo Carvalho de Melo
2015-08-04 5:28 ` Wangnan (F)
2015-08-04 10:39 ` Wangnan (F)
2015-08-04 15:55 ` Arnaldo Carvalho de Melo
2015-08-04 16:11 ` Arnaldo Carvalho de Melo
2015-08-04 16:13 ` pi3orama
2015-08-06 1:44 ` Wangnan (F)
2015-08-06 14:46 ` Arnaldo Carvalho de Melo
2015-08-06 15:58 ` 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=55A4891D.9070600@huawei.com \
--to=wangnan0@huawei.com \
--cc=acme@kernel.org \
--cc=ast@plumgrid.com \
--cc=hekuang@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizefan@huawei.com \
--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).