From: Alexei Starovoitov <ast@fb.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
"David S . Miller" <davem@davemloft.net>,
Ingo Molnar <mingo@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Arnaldo Carvalho de Melo <acme@infradead.org>,
Wang Nan <wangnan0@huawei.com>, Josef Bacik <jbacik@fb.com>,
Brendan Gregg <brendan.d.gregg@gmail.com>,
<netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
<kernel-team@fb.com>
Subject: Re: [PATCH net-next 2/8] perf, bpf: allow bpf programs attach to tracepoints
Date: Mon, 18 Apr 2016 14:43:07 -0700 [thread overview]
Message-ID: <571554EB.9010702@fb.com> (raw)
In-Reply-To: <20160418162905.220df2f4@gandalf.local.home>
On 4/18/16 1:29 PM, Steven Rostedt wrote:
> On Mon, 4 Apr 2016 21:52:48 -0700
> Alexei Starovoitov <ast@fb.com> wrote:
>
>> introduce BPF_PROG_TYPE_TRACEPOINT program type and allow it to be
>> attached to tracepoints.
>> The tracepoint will copy the arguments in the per-cpu buffer and pass
>> it to the bpf program as its first argument.
>> The layout of the fields can be discovered by doing
>> 'cat /sys/kernel/debug/tracing/events/sched/sched_switch/format'
>> prior to the compilation of the program with exception that first 8 bytes
>> are reserved and not accessible to the program. This area is used to store
>> the pointer to 'struct pt_regs' which some of the bpf helpers will use:
>> +---------+
>> | 8 bytes | hidden 'struct pt_regs *' (inaccessible to bpf program)
>> +---------+
>> | N bytes | static tracepoint fields defined in tracepoint/format (bpf readonly)
>> +---------+
>> | dynamic | __dynamic_array bytes of tracepoint (inaccessible to bpf yet)
>> +---------+
>>
>> Not that all of the fields are already dumped to user space via perf ring buffer
>> and some application access it directly without consulting tracepoint/format.
>> Same rule applies here: static tracepoint fields should only be accessed
>> in a format defined in tracepoint/format. The order of fields and
>> field sizes are not an ABI.
>>
>> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
>> ---
...
>> - entry = perf_trace_buf_prepare(__entry_size, \
>> - event_call->event.type, &__regs, &rctx); \
>> + event_type = prog ? TRACE_EVENT_TYPE_MAX : event_call->event.type; \
>
> Can you move this into perf_trace_entry_prepare?
that's the old version.
The last one are commits 1e1dcd93b46 and 98b5c2c65c295 in net-next.
>> + if (prog) { \
>> + *(struct pt_regs **)entry = __regs; \
>> + if (!trace_call_bpf(prog, entry) || hlist_empty(head)) { \
>> + perf_swevent_put_recursion_context(rctx); \
>> + return; \
>> + } \
>> + memset(&entry->ent, 0, sizeof(entry->ent)); \
>> + } \
>
> And perhaps this into perf_trace_buf_submit()?
>
> Tracepoints are a major cause of bloat, and the reasons for these
> prepare and submit functions is to move code out of the macros. Every
> tracepoint in the kernel (1000 and counting) will include this code.
> I've already had complaints that each tracepoint can add up to 5k to
> the core.
I was worried about this too, but single 'if' and two calls
(as in commit 98b5c2c65c295) is a better way, since it's faster, cleaner
and doesn't need to refactor the whole perf_trace_buf_submit() to pass
extra event_call argument to it.
perf_trace_buf_submit() is already ugly with 8 arguments!
Passing more args or creating a struct to pass args only going to
hurt performance without much reduction in .text size.
tinyfication folks will disable tracepoints anyway.
Note that the most common case is bpf returning 0 and not even
calling perf_trace_buf_submit() which is already slow due
to so many args passed on stack.
This stuff is called million times a second, so every instruction
counts.
next prev parent reply other threads:[~2016-04-18 21:44 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-05 4:52 [PATCH net-next 0/8] allow bpf attach to tracepoints Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 1/8] perf: optimize perf_fetch_caller_regs Alexei Starovoitov
2016-04-05 12:06 ` Peter Zijlstra
2016-04-05 17:41 ` Alexei Starovoitov
2016-04-08 22:12 ` Steven Rostedt
2016-04-05 4:52 ` [PATCH net-next 2/8] perf, bpf: allow bpf programs attach to tracepoints Alexei Starovoitov
2016-04-05 14:18 ` Peter Zijlstra
2016-04-05 18:09 ` Alexei Starovoitov
2016-04-05 18:16 ` Peter Zijlstra
2016-04-05 18:21 ` Alexei Starovoitov
2016-04-18 20:29 ` Steven Rostedt
2016-04-18 21:43 ` Alexei Starovoitov [this message]
2016-04-18 22:16 ` Steven Rostedt
2016-04-19 1:15 ` Alexei Starovoitov
2016-04-19 2:58 ` Steven Rostedt
2016-04-05 4:52 ` [PATCH net-next 3/8] bpf: register BPF_PROG_TYPE_TRACEPOINT program type Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 4/8] bpf: support bpf_get_stackid() and bpf_perf_event_output() in tracepoint programs Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 5/8] bpf: sanitize bpf tracepoint access Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 6/8] samples/bpf: add tracepoint support to bpf loader Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 7/8] samples/bpf: tracepoint example Alexei Starovoitov
2016-04-05 4:52 ` [PATCH net-next 8/8] samples/bpf: add tracepoint vs kprobe performance tests Alexei Starovoitov
2016-04-18 16:13 ` [PATCH net-next 0/8] allow bpf attach to tracepoints Steven Rostedt
2016-04-18 19:51 ` Alexei Starovoitov
2016-04-18 20:47 ` Steven Rostedt
2016-04-18 21:25 ` Alexei Starovoitov
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=571554EB.9010702@fb.com \
--to=ast@fb.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@infradead.org \
--cc=brendan.d.gregg@gmail.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=jbacik@fb.com \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=wangnan0@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).