linux-kernel - Re: [PATCH v4 00/30] coresight: Support for ACPI bindings

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: <20190528080853.GB5753@leoy-ThinkPad-X240s>
Date:   Tue, 28 May 2019 16:08:53 +0800
From:   Leo Yan <leo.yan@...aro.org>
To:     Suzuki K Poulose <suzuki.poulose@....com>
Cc:     linux-arm-kernel@...ts.infradead.org, coresight@...ts.linaro.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 00/30] coresight: Support for ACPI bindings

Hi Suzuki, Mathieu,

On Tue, May 28, 2019 at 01:19:24PM +0800, Leo Yan wrote:

[...]

> After I applied this patch set, I found all device names under
> '/sys/bus/event_source/devices/cs_etm/sinks/' have been changed as
> below on my DB410c board:
> # ls /sys/bus/event_source/devices/cs_etm/sinks/
> tmc_etf0  tmc_etr0  tpiu0
> 
> This leads to below command failure when open PMU device:
> # perf record -e cs_etm/@...000.etr/ --per-thread uname
> failed to set sink "826000.etr" on event cs_etm/@...000.etr/ with 2 (No such file or directory)
> 
> I must use below command so that perf can match string with the
> device name under '/sys/bus/event_source/devices/cs_etm/sinks/':
> # perf record -e cs_etm/@..._etr0/ --per-thread uname
> 
> Seems to me, this is an unexpected change and when I worked on the
> patch set v2, IIRC that version still can use '826000.etr' to open PMU
> device.
> 
> Please help confirm for this.  Thanks!

Finally, this is narrowed down to the patch 09/30 'coresight: Use
coresight device names for sinks in PMU attribute', so this is
delibrately to change to use new name format for perf command;
if so, maybe also update the documentation to reflect this change?

Thanks,
Leo Yan