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: <bac9837c-6dea-4ca1-9bda-9a4985403ed2@siemens.com>
Date: Thu, 15 Aug 2024 16:19:03 +0200
From: Jan Kiszka <jan.kiszka@...mens.com>
To: Nishanth Menon <nm@...com>, Judith Mendez <jm@...com>
Cc: devicetree@...r.kernel.org, Vignesh Raghavendra <vigneshr@...com>,
 linux-kernel@...r.kernel.org, Rob Herring <robh@...nel.org>,
 Conor Dooley <conor+dt@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>
Subject: Re: [PATCH v2 0/6] Add and fix ESM nodes

On 15.08.24 14:29, Nishanth Menon wrote:
> On 08:59-20240814, Judith Mendez wrote:
>> Hi Jan,
>>
>> On 8/13/24 11:04 PM, Jan Kiszka wrote:
>>> On 14.08.24 01:03, Judith Mendez wrote:
>>>> The following patch adds ESM nodes and fixes ESM source
>>>> interrupts for Sitara K3 platforms. Currently watchdog cannot
>>>> reset the CPU because of misconfiguration or missing ESM node
>>>> in DT.
>>>>
>>>> ESM node was added for am62ax and am65x. For am62px ESM source
>>>> interrupts are fixed. Comments were also added for clarity on what
>>>> source interrupts are routed to ESM based on device TRM.
>>>>
>>>> ESM nodes like MCU ESM for am65x are added for device completion,
>>>> currently, some ESM0 events are not routed to MCU ESM, so watchdog
>>>> cannot reset the CPU using the current implementation.
>>>
>>> Yes, that's why there is https://github.com/siemens/k3-rti-wdt and
>>> probably similar bits in other R5 firmware. I was always told that is
>>> the only way to reset the /system/ (CPU alone would not help). That
>>> information is still correct?
>>
>> If you look at 9.4.14 MCU_ESM0 Interrupt Map, ESM0_ESM_INT_CFG_LVL_0,
>> ESM0_ESM_INT_HI_LVL_0, and ESM0_ESM_INT_LOW_LVL_0 are not routed to
>> MCU_ESM0. So the current implementation to route events from ESM0 to
>> MCU_ESM0 to reset the CPU will not work for AM65x, this is the
>> implementation on other K3 Sitara platforms and how watchdog can reset
>> the cpu.
>>
>> I did find MAIN_ESM_ERROR_INT which should be SOC_SAFETY_ERRORn, look
>> at Figure 12-3690. Perhaps the ESMs could be configured to use
>> SOC_SAFETY_ERRORn instead, not sure.
>>
>> The above should apply to both SR1 and SR2 devices according to the TRM.
> 
> Thanks for clarifying - you should add that in the commit message.
> 

So the short answer to my question is "yes".

Jan

-- 
Siemens AG, Technology
Linux Expert Center


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ