RE: [PATCH v9 2/2] ras: mem: Add memory ACPI RAS2 driver

[Shiju Jose <shiju.jose@huawei.com>]

>-----Original Message-----
>From: Shiju Jose <shiju.jose@huawei.com>
>Sent: 01 August 2025 18:29
>To: Daniel Ferguson <danielf@os.amperecomputing.com>; rafael@kernel.org;
>linux-edac@vger.kernel.org; linux-acpi@vger.kernel.org; linux-
>doc@vger.kernel.org; bp@alien8.de; tony.luck@intel.com; lenb@kernel.org;
>leo.duran@amd.com; Yazen.Ghannam@amd.com; mchehab@kernel.org
>Cc: Jonathan Cameron <jonathan.cameron@huawei.com>; linux-
>mm@kvack.org; Linuxarm <linuxarm@huawei.com>; rientjes@google.com;
>jiaqiyan@google.com; Jon.Grimm@amd.com; dave.hansen@linux.intel.com;
>naoya.horiguchi@nec.com; james.morse@arm.com; jthoughton@google.com;
>somasundaram.a@hpe.com; erdemaktas@google.com; pgonda@google.com;
>duenwen@google.com; gthelen@google.com;
>wschwartz@amperecomputing.com; dferguson@amperecomputing.com;
>wbs@os.amperecomputing.com; nifan.cxl@gmail.com; tanxiaofei
><tanxiaofei@huawei.com>; Zengtao (B) <prime.zeng@hisilicon.com>; Roberto
>Sassu <roberto.sassu@huawei.com>; kangkang.shen@futurewei.com;
>wanghuiqiang <wanghuiqiang@huawei.com>;
>vanshikonda@os.amperecomputing.com
>Subject: RE: [PATCH v9 2/2] ras: mem: Add memory ACPI RAS2 driver
>
>>-----Original Message-----
>>From: Daniel Ferguson <danielf@os.amperecomputing.com>
[...]
>>3) This is a special case of #2. During driver load, ras2_probe issues
>>a GET_PATROL_PARAMETERS using a base and size equal to 0. The base and
>>size are zero because requested_address_range is allocated with
>>kzalloc. In this case, we may not be getting the initial values from the correct
>range.
>In the probe stage, while issue GET_PATROL_PARAMETERS, driver has no
>information about the  'requested_address_range' supposed to set other than
>set to 0 because
>RAS2 ('RAS2 Platform Communication Channel Shared Memory Region' table?)
>does not define fields for the platform to advertise the supported full address
>range for a scrubber.
Please ignore this as it was solved.

>>
>>Proposed Solution:
>>What we propose, is to instead of zeroing out the base and size after
>>an error, use the full range of the current NUMA node. We believe that
>>a superset of a currently active scrub range can properly report all
>>the relevant and correct information.
>>To be compliant with the specification, FW should set "Flags" field if
>>there is any on-demand scrub in progress on any memory range in the
>>NUMA node. Again, this solution assumes that the driver does not allow
>>more than one scrubber to run within a single NUMA node. All three
>>problems can be solved in the same way.
>>
>>What do you think?
>Hi Daniel,
>
>Please check the v10 of the RAS2 series sent with changes for your
>requirements.
>and request please test in your system and share the feedback.
>
>https://lore.kernel.org/all/20250801172040.2175-1-shiju.jose@huawei.com/
>>
>>Regards,
>>~Daniel
>
>Thanks,
>Shiju