linux-kernel - Re: [RFC 0/4] trace_uprobe: Support SDT markers having semaphore

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]

Message-Id: <0fdb07bd-fe66-5087-9740-1f87f14e4b60@linux.vnet.ibm.com>
Date:   Thu, 1 Mar 2018 11:02:07 +0530
From:   Ravi Bangoria <ravi.bangoria@...ux.vnet.ibm.com>
To:     Masami Hiramatsu <mhiramat@...nel.org>
Cc:     peterz@...radead.org, mingo@...hat.com, acme@...nel.org,
        alexander.shishkin@...ux.intel.com, jolsa@...hat.com,
        namhyung@...nel.org, linux-kernel@...r.kernel.org,
        rostedt@...dmis.org, ananth@...ux.vnet.ibm.com,
        naveen.n.rao@...ux.vnet.ibm.com, srikar@...ux.vnet.ibm.com,
        oleg@...hat.com, Ravi Bangoria <ravi.bangoria@...ux.vnet.ibm.com>
Subject: Re: [RFC 0/4] trace_uprobe: Support SDT markers having semaphore



On 02/28/2018 07:55 PM, Masami Hiramatsu wrote:
> Hi Ravi,
>
> Thank you for your great work!

Thanks Masami.

> On Wed, 28 Feb 2018 13:23:41 +0530
> Ravi Bangoria <ravi.bangoria@...ux.vnet.ibm.com> wrote:
>
>> Userspace Statically Defined Tracepoints[1] are dtrace style markers
>> inside userspace applications. These markers are added by developer at
>> important places in the code. Each marker source expands to a single
>> nop instruction in the compiled code but there may be additional
>> overhead for computing the marker arguments which expands to couple of
>> instructions. If this computaion is quite more, execution of it can be
>> ommited by runtime if() condition when no one is tracing on the marker:
...
>>
>> Semaphore offset is 0x10036. I don't have educated 'perf probe'
>> about semaphore. So instead of using 'perf probe' command, I'm
>> manually adding entry in the <tracefs>/uprobe_events file.
> Ok, it is easy to pass semaphore address via perf probe :)

Yes, it should be fairly easy to parse semaphore at buildid-cache time.
Will add a patch for that.

>
>> Special char * denotes semaphore offset.
>>
>>
>>   # echo "p:sdt_tick/loop2 /tmp/tick:0x6e4 *0x10036" > uprobe_events
> IMHO, this syntax is no good, separate with space is only for arguments.
> Since the semaphore is per-probe-point based, that should be specified with probe point.
> (there are no 2 or more semaphores on 1 event, are there?)
> So something like
>
> # echo "p:sdt_tick/loop2 /tmp/tick:0x6e4(0x10036)" > uprobe_events

This is great suggestion. Will change it.

Please review patch 3 and 4 which contains actual implementation.

Thanks for the review,
Ravi