[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <0309f5f9-9a95-485c-a442-e9fba603d676@os.amperecomputing.com>
Date: Thu, 28 Mar 2024 16:39:44 -0700
From: Daniel Ferguson <danielf@...amperecomputing.com>
To: shiju.jose@...wei.com, linux-cxl@...r.kernel.org,
linux-acpi@...r.kernel.org, linux-mm@...ck.org, dan.j.williams@...el.com,
dave@...olabs.net, jonathan.cameron@...wei.com, dave.jiang@...el.com,
alison.schofield@...el.com, vishal.l.verma@...el.com, ira.weiny@...el.com
Cc: linux-edac@...r.kernel.org, linux-kernel@...r.kernel.org,
david@...hat.com, Vilas.Sridharan@....com, leo.duran@....com,
Yazen.Ghannam@....com, rientjes@...gle.com, jiaqiyan@...gle.com,
tony.luck@...el.com, Jon.Grimm@....com, dave.hansen@...ux.intel.com,
rafael@...nel.org, lenb@...nel.org, naoya.horiguchi@....com,
james.morse@....com, jthoughton@...gle.com, somasundaram.a@....com,
erdemaktas@...gle.com, pgonda@...gle.com, duenwen@...gle.com,
mike.malvestuto@...el.com, gthelen@...gle.com,
wschwartz@...erecomputing.com, dferguson@...erecomputing.com,
tanxiaofei@...wei.com, prime.zeng@...ilicon.com,
kangkang.shen@...urewei.com, wanghuiqiang@...wei.com, linuxarm@...wei.com,
wbs@...amperecomputing.com
Subject: Re: [RFC PATCH v7 00/12] memory: scrub: introduce subsystem +
CXL/ACPI-RAS2 drivers
> RAS2 HW based memory patrol scrub needs RAS2 PCC interfaces
> and ACPI RAS2 driver for communication b/w kernel and firmware.
> ACPI RAS2 Driver adds platform device, for each memory feature,
> which binds to the RAS2 memory driver.
> Memory RAS2 driver registers with the memory scrub subsystem to
> expose the RAS2 scrub controls to the user.
Hi Shiju,
Thanks for this work. This has been very useful for us, as we've
been using it to test a RAS2 implementation here at Ampere
Computing.
In general, the pieces implementing RAS2 are useful and functional. We did,
however, encounter a few places where we had to make some bug fixes.
I'll reply to specific patches with changes I can recommend. Also, there are
a few implementation choices that I will have questions about.
Here is an outline containing the generalized comments I plan on making
within/near relevant patches:
1) Found a bug in ras2.c in how the pcc_desc_list pointer is incremented.
2) Executing a RAS2 command seems unnecessary for feature detection.
specifically, in the functions ras2_is_patrol_scrub_support() and
ras2_get_patrol_scrub_params()
3) Consider adding more error detection/reporting in ras2_check_pcc_chan()
4) Concerns with multiple scrub devices attempting to map the same PCC
channel.
-Daniel
Powered by blists - more mailing lists