lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Wed, 19 Aug 2015 17:30:12 +0900
From:	Namhyung Kim <namhyung@...nel.org>
To:	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
Cc:	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	linux-kernel@...r.kernel.org,
	Adrian Hunter <adrian.hunter@...el.com>,
	Ingo Molnar <mingo@...hat.com>,
	Paul Mackerras <paulus@...ba.org>,
	Jiri Olsa <jolsa@...nel.org>, Borislav Petkov <bp@...e.de>,
	Hemant Kumar <hemant@...ux.vnet.ibm.com>
Subject: Re: [RFC PATCH perf/core v3 00/17] perf-probe --cache and SDT support

Hi Masami,

On Sat, Aug 15, 2015 at 08:42:53PM +0900, Masami Hiramatsu wrote:
> Hi,
> 
> Here is the 3rd version of the patchset for probe-cache and 
> initial SDT support which are going to be perf-cache finally.
> 
> The previous version is here; https://lkml.org/lkml/2015/7/15/221
> 
> This version improves SDT support for perf-list and perf-record.
> It is still just a basic support (no wildcard support, nor
> @FILE/@...LDID support). So this is an ongoing work report :)
> 
> As we discussed on the previous series, now all SDT events have
> "sdt_" prefix on the provider name. We also should support "@FILE"
> or "@BUILDID" suffixes for same-name SDTs.
> 
> perf-list shows only SDTs on existing binaries (no old/deleted
> files) and if the same-name events are shown, it adds 
> "@FILE@...LDID" suffixes after the events to clarify on which
> binary the SDT exists.
> 
>   ----
>   # perf buildid-cache -a /usr/bin/gcc
>   # perf list sdt
> 
>   List of pre-defined events (to be used in -e):
> 
>     sdt_libgcc:unwind                                  [SDT event]
>     sdt_libstdcxx:catch                                [SDT event]
>     sdt_libstdcxx:rethrow                              [SDT event]
>     sdt_libstdcxx:throw                                [SDT event]
>   # perf buildid-cache -a /usr/lib/libstdc++.so.6
>   # perf list sdt
> 
>   List of pre-defined events (to be used in -e):
> 
>     sdt_libgcc:unwind                                  [SDT event]
>     sdt_libstdcxx:catch@...r/bin/gcc@...5d261236bbb    [SDT event]
>     sdt_libstdcxx:catch@...r/lib/libstdc++.so.6.0.19@...d6ac0e6236c [SDT event]
>     sdt_libstdcxx:rethrow@...r/bin/gcc@...5d261236bbb  [SDT event]
>     sdt_libstdcxx:rethrow@...r/lib/libstdc++.so.6.0.19@...d6ac0e6236c [SDT event]
>     sdt_libstdcxx:throw@...r/bin/gcc@...5d261236bbb    [SDT event]
>     sdt_libstdcxx:throw@...r/lib/libstdc++.so.6.0.19@...d6ac0e6236c [SDT event]
>   ----
> 
> In this version, I used "@" separater for both FILE and BUILDID,
> but I think it is also possible to use "%" for BUILDID.
> (BTW, I've cut down the BUILDID to the first 12 chars, and added "0x") 
> 
> TODOs:
>  - (perf record) Allow glob matching for SDT event to specify events.
>  - (perf record) Support @FILE/@...LDID suffix to record specific SDTs.
>  - (perf record) Try to unregister SDT events after record.
>  - (perf probe) Allow glob matching for pre-cached events.
>  - (perf probe) Support @FILE/@...LDID suffix for pre-cached events.
>                (also removes -x option when using pre-cached events)
>  - (ftrace) Support multiple SDTs on single event.
> 
> Since multiple same SDTs are defined in a single binary (e.g. libc:setjump
> has 3 different entries on libc-2.17.so), we need the last feature on
> ftrace, so that a single uprobe event can occur several different
> probe points.

Do you want to have multiple events of same name in the kernel?  Or
to make perf handle those same names with different kernel events?

Thanks,
Namhyung
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ