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: <CAJ9a7VgBx+4Vgr_2VOY76wJMyjoyO48YD3Pp7Qx0C7hUY8xzKg@mail.gmail.com>
Date:   Mon, 27 Feb 2017 12:37:45 +0000
From:   Mike Leach <mike.leach@...aro.org>
To:     Leo Yan <leo.yan@...aro.org>
Cc:     Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Mathieu Poirier <mathieu.poirier@...aro.org>,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v1 1/2] coresight: bindings for debug module

Hi Leo

On 23 February 2017 at 01:57, Leo Yan <leo.yan@...aro.org> wrote:
> According to ARMv8 architecture reference manual (ARM DDI 0487A.k)
> Chapter 'Part H: External debug', the CPU can integrate debug module
> and it can support self-hosted debug and external debug. Especially
> for supporting self-hosted debug, this means the program can access
> the debug module from mmio region; and usually the mmio region is
> integrated with coresight.
>
> So add document for binding debug component, includes binding to two
> clocks, one is apb clock for bus and another is debug clock for debug
> module self; and also need specify the CPU node which the debug module
> is dedicated to specific CPU.
>
> Signed-off-by: Leo Yan <leo.yan@...aro.org>
> ---
>  .../devicetree/bindings/arm/coresight-debug.txt    | 39 ++++++++++++++++++++++
>  1 file changed, 39 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/arm/coresight-debug.txt
>
> diff --git a/Documentation/devicetree/bindings/arm/coresight-debug.txt b/Documentation/devicetree/bindings/arm/coresight-debug.txt
> new file mode 100644
> index 0000000..6e03e9b
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/arm/coresight-debug.txt
> @@ -0,0 +1,39 @@
> +* CoreSight Debug Component:
> +
> +CoreSight debug component are compliant with the ARMv8 architecture reference
> +manual (ARM DDI 0487A.k) Chapter 'Part H: External debug'. The external debug
> +module is mainly used for two modes: self-hosted debug and external debug, and
> +it can be accessed from mmio region from Coresight and eventually the debug
> +module connects with CPU for debugging. And the debug module provides
> +sample-based profiling extension, which can be used to sample CPU program
> +counter, secure state and exception level, etc; usually every CPU has one
> +dedicated debug module to be connected.
> +
> +Required properties:
> +
> +- compatible : should be
> +            * "arm,coresight-debug", "arm,primecell"; supplemented with
> +              "arm,primecell" as driver is using the AMBA bus interface.
> +
> +- reg : physical base address and length of the register set.
> +
> +- clocks : the clocks associated to this component.
> +
> +- clock-names : the name of the clocks referenced by the code. Since we are
> +                using the AMBA framework, the name of the clock providing
> +               the interconnect should be "apb_pclk", and the debug module
> +               has an additional clock "dbg_clk", which is used to provide
> +               clock for debug module itself. Both clocks are mandatory.
> +

Perhaps I am misunderstanding the nature of the .dts files, but I'm
puzzled about the dbg_clk. I cannot see anything in the architecture
or normal A53 implementation that mentions this.
To access external debug from the core/external debug agent then we do
need the APB clock, but the interface between the debug logic and the
processor core is clocked by the internal CPU clocks.


> +- cpu : the cpu phandle the debug module is affined to. When omitted
> +       the source is considered to belong to CPU0.
> +
> +Example:
> +
> +       debug@...90000 {
> +               compatible = "arm,coresight-debug","arm,primecell";
> +               reg = <0 0xf6590000 0 0x1000>;
> +               clocks = <&sys_ctrl HI6220_CS_ATB>, <&acpu_ctrl HI6220_ACPU_DBG_CLK0>;
> +               clock-names = "apb_pclk", "dbg_clk";
> +               cpu = <&cpu0>;
> +       };
> --
> 2.7.4
>

When I was looking at clocks for trace on the HiKey board I noted:
HI6220_CS_DAPB -- which I assumed was the debug APB clock.
HI6220_CS_ATB - which I assumed was the ATB (trace bus) clock.


Regards

Mike

-- 
Mike Leach
Principal Engineer, ARM Ltd.
Blackburn Design Centre. UK

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ