BPF List
 help / color / mirror / Atom feed
From: Yonghong Song <yhs@fb.com>
To: Hou Tao <houtao1@huawei.com>, Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>,
	Andrii Nakryiko <andrii@kernel.org>,
	Martin KaFai Lau <kafai@fb.com>, Song Liu <songliubraving@fb.com>,
	<bpf@vger.kernel.org>, <netdev@vger.kernel.org>
Subject: Re: [PATCH 1/3] bpf: support writable context for bare tracepoint
Date: Fri, 17 Sep 2021 07:48:53 -0700	[thread overview]
Message-ID: <3001e875-9a74-8e22-3a7c-be3d280cd866@fb.com> (raw)
In-Reply-To: <b76d4051-abff-5e75-c812-41c6f283327f@huawei.com>



On 9/17/21 6:45 AM, Hou Tao wrote:
> Hi,
> 
> On 9/17/2021 7:16 AM, Yonghong Song wrote:
>>
>>
>> On 9/16/21 6:55 AM, Hou Tao wrote:
>>> Commit 9df1c28bb752 ("bpf: add writable context for raw tracepoints")
>>> supports writable context for tracepoint, but it misses the support
>>> for bare tracepoint which has no associated trace event.
>>>
>>> Bare tracepoint is defined by DECLARE_TRACE(), so adding a corresponding
>>> DECLARE_TRACE_WRITABLE() macro to generate a definition in __bpf_raw_tp_map
>>> section for bare tracepoint in a similar way to DEFINE_TRACE_WRITABLE().
>>>
>>> Signed-off-by: Hou Tao <houtao1@huawei.com>
>>> ---
>>>    include/trace/bpf_probe.h | 19 +++++++++++++++----
>>>    1 file changed, 15 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/include/trace/bpf_probe.h b/include/trace/bpf_probe.h
>>> index a23be89119aa..d08ee1060d82 100644
>>> --- a/include/trace/bpf_probe.h
>>> +++ b/include/trace/bpf_probe.h
>>> @@ -93,8 +93,7 @@ __section("__bpf_raw_tp_map") = {                    \
>>>      #define FIRST(x, ...) x
>>>    -#undef DEFINE_EVENT_WRITABLE
>>> -#define DEFINE_EVENT_WRITABLE(template, call, proto, args, size)    \
>>> +#define __CHECK_WRITABLE_BUF_SIZE(call, proto, args, size)        \
>>>    static inline void bpf_test_buffer_##call(void)                \
>>>    {                                    \
>>>        /* BUILD_BUG_ON() is ignored if the code is completely eliminated, but \
>>> @@ -103,8 +102,12 @@ static inline void
>>> bpf_test_buffer_##call(void)                \
>>>         */                                \
>>>        FIRST(proto);                            \
>>>        (void)BUILD_BUG_ON_ZERO(size != sizeof(*FIRST(args)));        \
>>> -}                                    \
>>> -__DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args), size)
>>> +}
>>> +
>>> +#undef DEFINE_EVENT_WRITABLE
>>> +#define DEFINE_EVENT_WRITABLE(template, call, proto, args, size) \
>>> +    __CHECK_WRITABLE_BUF_SIZE(call, PARAMS(proto), PARAMS(args), size) \
>>> +    __DEFINE_EVENT(template, call, PARAMS(proto), PARAMS(args), size)
>>>      #undef DEFINE_EVENT
>>>    #define DEFINE_EVENT(template, call, proto, args)            \
>>> @@ -119,10 +122,18 @@ __DEFINE_EVENT(template, call, PARAMS(proto),
>>> PARAMS(args), size)
>>>        __BPF_DECLARE_TRACE(call, PARAMS(proto), PARAMS(args))        \
>>>        __DEFINE_EVENT(call, call, PARAMS(proto), PARAMS(args), 0)
>>>    +#undef DECLARE_TRACE_WRITABLE
>>> +#define DECLARE_TRACE_WRITABLE(call, proto, args, size) \
>>> +    __CHECK_WRITABLE_BUF_SIZE(call, PARAMS(proto), PARAMS(args), size) \
>>> +    __BPF_DECLARE_TRACE(call, PARAMS(proto), PARAMS(args)) \
>>> +    __DEFINE_EVENT(call, call, PARAMS(proto), PARAMS(args), size)
>>> +
>>>    #include TRACE_INCLUDE(TRACE_INCLUDE_FILE)
>>>      #undef DEFINE_EVENT_WRITABLE
>>> +#undef DECLARE_TRACE_WRITABLE
>>>    #undef __DEFINE_EVENT
>>> +#undef __CHECK_WRITABLE_BUF_SIZE
>>
>> Put "#undef __CHECK_WRITABLE_BUF_SIZE" right after "#undef
>> DECLARE_TRACE_WRITABLE" since they are related to each other
>> and also they are in correct reverse order w.r.t. __DEFINE_EVENT?
> If considering __CHECK_WRITABLE_BUF_SIZE is used in both DECLARE_TRACE_WRITABLE and
> DEFINE_EVENT_WRITABLE and the order of definitions, is the following order better ?
> 
> #undef DECLARE_TRACE_WRITABLE
> #undef DEFINE_EVENT_WRITABLE
> #undef __CHECK_WRITABLE_BUF_SIZE

This should be okay.

> 
>>
>>>    #undef FIRST
>>>      #endif /* CONFIG_BPF_EVENTS */
>>>
>> .
> 

  reply	other threads:[~2021-09-17 14:49 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-16 13:55 [PATCH 0/3] add support for writable bare tracepoint Hou Tao
2021-09-16 13:55 ` [PATCH 1/3] bpf: support writable context for " Hou Tao
2021-09-16 23:16   ` Yonghong Song
2021-09-17 13:45     ` Hou Tao
2021-09-17 14:48       ` Yonghong Song [this message]
2021-09-16 13:55 ` [PATCH 2/3] libbpf: support detecting and attaching of writable tracepoint program Hou Tao
2021-09-16 23:35   ` Yonghong Song
2021-09-16 13:55 ` [PATCH 3/3] bpf/selftests: add test for writable bare tracepoint Hou Tao
2021-09-16 23:46   ` Yonghong Song
2021-09-17 14:03     ` Hou Tao
2021-09-16 22:58 ` [PATCH 0/3] add support " Yonghong Song

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=3001e875-9a74-8e22-3a7c-be3d280cd866@fb.com \
    --to=yhs@fb.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=houtao1@huawei.com \
    --cc=kafai@fb.com \
    --cc=netdev@vger.kernel.org \
    --cc=songliubraving@fb.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