linux-kernel - Re: [PATCH v6] platform/chrome: cros_ec: Add trace event to trace EC commands

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] [day] [month] [year] [list]

Message-ID: <a966c393-70ed-48e9-988e-a869c670e25c@collabora.com>
Date:   Tue, 23 Apr 2019 12:02:20 +0200
From:   Enric Balletbo i Serra <enric.balletbo@...labora.com>
To:     Raul E Rangel <rrangel@...omium.org>,
        linux-trace-devel@...r.kernel.org
Cc:     zwisler@...omium.org, djkurtz@...omium.org,
        Ross Zwisler <zwisler@...gle.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Guenter Roeck <groeck@...omium.org>,
        linux-kernel@...r.kernel.org, Benson Leung <bleung@...omium.org>
Subject: Re: [PATCH v6] platform/chrome: cros_ec: Add trace event to trace EC
 commands



On 16/4/19 23:03, Raul E Rangel wrote:
> This is useful to see which EC commands are being executed and when.
> 
> To enable:
> 
>     echo 1 > /sys/kernel/debug/tracing/events/cros_ec/enable
> 
> Example:
> 
>     /* cros_ec_cmd: version: 0, command: EC_CMD_GET_VERSION */
>     /* cros_ec_cmd: version: 0, command: EC_CMD_GET_PROTOCOL_INFO */
>     /* cros_ec_cmd: version: 1, command: EC_CMD_GET_CMD_VERSIONS */
>     /* cros_ec_cmd: version: 1, command: EC_CMD_USB_PD_CONTROL */
> 
> Signed-off-by: Raul E Rangel <rrangel@...omium.org>
> Reviewed-by: Ross Zwisler <zwisler@...gle.com>
> Reviewed-by: Steven Rostedt (VMware) <rostedt@...dmis.org>
> ---
> I did not add cros_ec_commands.h as an include to cros_ec_trace.c.
> It's not strictly required because the macro doesn't get evaluated until
> after cros_ec_trace.h has been included. That transitively brings in
> cros_ec_ommands.h. I did try it, but since cros_ec_commands.h doesn't
> include linux/types.h it fails to find uint32_t. I could have included
> both headers, but since they aren't required I just skipped it.
> 
> My reasoning for not putting EC_CMDS in the cros_ec_trace.h was that I
> didn't want to clutter the file. I think it's visually easier to find
> all the trace events without having to parse through all the
> implementation details.
> 
> Changes in v6:
> - Added script to generate the EC_CMDS trace symbols. They were
>   originally sorted by value, now they are sorted by declaration order.
>   There is one duplicate: EC_CMD_PORT80_LAST_BOOT and
>   EC_CMD_PORT80_READ. They both resolve to the same value. Not sure if
>   this is a bug in the cros_ec_commands.h file or not. Anyway it doesn't
>   hurt anything.
> - Changed the example in the commit
> 
> Changes in v5:
> - Rebased on upstream/master. It turns out upstream/master is missing a
>   lot of changes to cros_ec_commands.h.
> - Replace SYM with SYMBOL in macro name.
> 
> Changes in v4:
> - Use the full command name so go to definition continues to work in my
> editor.
> 
> Changes in v3:
> - Use a macro to avoid duplicating the ec command names.
> 
> Changes in v2:
> - Changed comment style to match other cros_ec files.
> - Fixed commit tag.
> 
>  drivers/platform/chrome/Makefile        |   5 +-
>  drivers/platform/chrome/cros_ec_proto.c |   4 +
>  drivers/platform/chrome/cros_ec_trace.c | 124 ++++++++++++++++++++++++
>  drivers/platform/chrome/cros_ec_trace.h |  51 ++++++++++
>  4 files changed, 183 insertions(+), 1 deletion(-)
>  create mode 100644 drivers/platform/chrome/cros_ec_trace.c
>  create mode 100644 drivers/platform/chrome/cros_ec_trace.h
> 

Tweaked a bit the commit message and applied for 5.2

Thanks,
 Enric