From: Hekuang <hekuang@huawei.com>
To: David Ahern <dsahern@gmail.com>, <peterz@infradead.org>,
<mingo@redhat.com>, <acme@kernel.org>,
<alexander.shishkin@linux.intel.com>, <jolsa@redhat.com>,
<wangnan0@huawei.com>, <jpoimboe@redhat.com>,
<ak@linux.intel.com>, <eranian@google.com>, <namhyung@kernel.org>,
<adrian.hunter@intel.com>, <sukadev@linux.vnet.ibm.com>,
<masami.hiramatsu.pt@hitachi.com>, <tumanova@linux.vnet.ibm.com>,
<kan.liang@intel.com>, <penberg@kernel.org>
Cc: <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v3 3/7 UPDATE] perf tools: Add option for the path of buildid dsos under symfs
Date: Mon, 16 May 2016 09:30:22 +0800 [thread overview]
Message-ID: <573922AE.5030002@huawei.com> (raw)
In-Reply-To: <4572f136-796d-059f-6b81-613e8f63be24@gmail.com>
在 2016/5/14 22:43, David Ahern 写道:
> On 5/14/16 2:19 AM, He Kuang wrote:
>> In the cross-platform perf record/script scenario, we need vdsos in
>> buildid-cache dir and other libs in symfs dir at the same time. For
>> the reason that to have every single file opened by perf is relative
>> to symfs dirctory, perf skips the buildid dir if symfs is given.
>>
>> This patch references the buildid dir under symfs if '--symfs' is
>> used, and adds new option '--dso-prefix' to specify the subdir path in
>> symfs which contains the buildid dsos.
>
> In the previous version of this patch you just wanted to drop the
> symfs check. That means there is a path that perf searches for buildid
> files and reading it worked for you. Why is adding symsfs to that path
> not enough? ie., Why do you need to specify a different location under
> the symfs?
>
In previous patch, I use 'perf buildid-cache -a' to add vdso
binary into the HOST buildid dir.
Currently, $(BUILDID_ROOT)/.debug/.buildid has a directory
structure organized by the buildid value, like this:
.build-id
- 3a
| e5ba6d4e532ad529e43ccf1ce1ddf8a64a4fdd
- 84
dbd75729adba57cc42f5544b25de571c0c8731
perf searchs for buildid binaries by the buildid value.
And symfs is a normal file tree and perf searchs for binaryies
by the file name.
lib
- libc.so
- ld.so
Tt's inappropriate to add symfs to .build-id dir becuase they
have different dirctory structure and perf searchs for them by
differnt ways.
So in this patch I add .buildid to symfs and got dirctory tree
like this:
symfs
- $(DSO_PREFIX)
- .build-id
- 3a
- 84
- libc.so
- ld.so
Thanks
next prev parent reply other threads:[~2016-05-16 1:32 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-12 8:43 [PATCH v3 0/7] Add support for remote unwind He Kuang
2016-05-12 8:43 ` [PATCH v3 1/7] perf tools: Set vdso name to vdso[64,32] depending on platform He Kuang
2016-05-12 10:06 ` Adrian Hunter
2016-05-13 8:51 ` [PATCH v3 1/7 UPDATE] perf tools: Find vdso with the consider of cross-platform He Kuang
2016-05-16 13:32 ` Arnaldo Carvalho de Melo
2016-05-17 7:34 ` Adrian Hunter
2016-05-17 7:33 ` Adrian Hunter
2016-05-17 9:04 ` [PATCH v3 1/7 UPDATE2] " He Kuang
2016-05-17 9:17 ` Adrian Hunter
2016-05-17 12:46 ` Arnaldo Carvalho de Melo
2016-06-15 13:34 ` Arnaldo Carvalho de Melo
2016-06-16 0:22 ` Hekuang
2016-06-16 13:00 ` Adrian Hunter
2016-06-16 16:33 ` Arnaldo Carvalho de Melo
2016-05-17 9:05 ` [PATCH v3 1/7 UPDATE] " Hekuang
2016-05-13 8:53 ` [PATCH v3 1/7] perf tools: Set vdso name to vdso[64,32] depending on platform Hekuang
2016-05-12 8:43 ` [PATCH v3 2/7] perf tools: Store vdso buildid unconditionally He Kuang
2016-05-12 13:09 ` Arnaldo Carvalho de Melo
2016-05-20 6:43 ` [tip:perf/urgent] perf symbols: " tip-bot for He Kuang
2016-05-12 8:43 ` [PATCH v3 3/7] perf tools: Remove the logical that skip buildid cache if symfs is given He Kuang
2016-05-12 13:09 ` Arnaldo Carvalho de Melo
2016-05-12 20:23 ` David Ahern
2016-05-12 20:33 ` Arnaldo Carvalho de Melo
2016-05-13 7:19 ` Hekuang
2016-05-13 14:27 ` David Ahern
2016-05-13 18:01 ` Arnaldo Carvalho de Melo
2016-05-14 8:19 ` [PATCH v3 3/7 UPDATE] perf tools: Add option for the path of buildid dsos under symfs He Kuang
2016-05-14 14:43 ` David Ahern
2016-05-16 1:30 ` Hekuang [this message]
2016-05-16 2:50 ` David Ahern
2016-05-16 6:40 ` Hekuang
2016-05-18 1:47 ` Hekuang
2016-05-18 1:51 ` David Ahern
2016-05-18 2:48 ` Hekuang
2016-05-18 3:04 ` David Ahern
2016-05-12 8:43 ` [PATCH v3 4/7] perf tools: Promote proper messages for cross-platform unwind He Kuang
2016-05-12 8:43 ` [PATCH v3 5/7] perf callchain: Add support " He Kuang
2016-05-12 8:43 ` [PATCH v3 6/7] perf callchain: Support x86 target platform He Kuang
2016-05-12 8:43 ` [PATCH v3 7/7] perf callchain: Support aarch64 cross-platform He Kuang
-- strict thread matches above, loose matches on Subject: below --
2016-06-22 6:57 [PATCH 0/5] Fixes on perf unwind He Kuang
2016-06-22 6:57 ` [PATCH 1/5] perf unwind: Change macro names of perf register He Kuang
2016-06-26 10:56 ` [tip:perf/core] " tip-bot for He Kuang
2016-06-22 6:57 ` [PATCH 2/5] perf unwind: Fix wrongly used regs for x86_32 unwind He Kuang
2016-06-26 10:56 ` [tip:perf/core] " tip-bot for He Kuang
2016-06-22 6:57 ` [PATCH 3/5] perf unwind: Fix wrongly used regs for aarch64 unwind He Kuang
2016-06-26 10:57 ` [tip:perf/core] " tip-bot for He Kuang
2016-06-22 6:57 ` [PATCH 4/5] perf tools: Let python use correct gcc for build_ext He Kuang
2016-06-26 10:54 ` [tip:perf/core] " tip-bot for He Kuang
2016-06-22 6:57 ` [PATCH 5/5] perf tools: Fix NULL pointer deference when vdso not found He Kuang
2016-06-23 2:02 ` Wangnan (F)
2016-06-23 3:03 ` Hekuang
2016-06-23 12:07 ` Arnaldo Carvalho de Melo
2016-06-23 13:28 ` Arnaldo Carvalho de Melo
2016-06-26 10:55 ` [tip:perf/core] perf tools: Find right DSO taking into account if binary is 32 or 64-bit tip-bot for He Kuang
2016-06-23 13:30 ` [PATCH 0/5] Fixes on perf unwind 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=573922AE.5030002@huawei.com \
--to=hekuang@huawei.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=dsahern@gmail.com \
--cc=eranian@google.com \
--cc=jolsa@redhat.com \
--cc=jpoimboe@redhat.com \
--cc=kan.liang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=penberg@kernel.org \
--cc=peterz@infradead.org \
--cc=sukadev@linux.vnet.ibm.com \
--cc=tumanova@linux.vnet.ibm.com \
--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).