linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Yunjeong Mun <yunjeong.mun@sk.com>
To: Gregory Price <gourry@gourry.net>
Cc: kernel_team@skhynix.com, Joshua Hahn <joshua.hahnjy@gmail.com>,
	harry.yoo@oracle.com, ying.huang@linux.alibaba.com,
	gregkh@linuxfoundation.org, rakie.kim@sk.com,
	akpm@linux-foundation.org, rafael@kernel.org, lenb@kernel.org,
	dan.j.williams@intel.com, Jonathan.Cameron@huawei.com,
	dave.jiang@intel.com, horen.chuang@linux.dev, hannes@cmpxchg.org,
	linux-kernel@vger.kernel.org, linux-acpi@vger.kernel.org,
	linux-mm@kvack.org, kernel-team@meta.com, yunjeong.mun@sk.com,
	Honggyu Kim <honggyu.kim@sk.com>
Subject: Re: [PATCH 2/2 v6] mm/mempolicy: Don't create weight sysfs for memoryless nodes
Date: Tue, 11 Mar 2025 11:07:27 +0900	[thread overview]
Message-ID: <20250311020806.404-1-yunjeong.mun@sk.com> (raw)
In-Reply-To: <Z871k085e-Th-jTz@gourry-fedora-PF4VCD3F>

Hi Gregory,

On Mon, 10 Mar 2025 10:22:11 -0400 Gregory Price <gourry@gourry.net> wrote:
> On Mon, Mar 10, 2025 at 09:26:48PM +0900, Honggyu Kim wrote:
> > Yeah, the proximity domain detects the node correctly as follows in dmesg.
> > 
> >  [  0.009915] ACPI: SRAT: Node 0 PXM 0 [mem 0x00000000-0x7fffffff]
> >  [  0.009917] ACPI: SRAT: Node 0 PXM 0 [mem 0x100000000-0x207fffffff]
> >  [  0.009919] ACPI: SRAT: Node 1 PXM 1 [mem 0x60f80000000-0x64f7fffffff]
> >  [  0.009924] ACPI: SRAT: Node 2 PXM 6 [mem 0x2080000000-0x807fffffff] hotplug
> >  [  0.009925] ACPI: SRAT: Node 3 PXM 7 [mem 0x64f80000000-0x6cf7fffffff] hotplug
> > 
> > It is printed even before CXL detection.
> > 
> 
> I wrote a some documentation on some example configurations last friday
> 
> https://lore.kernel.org/linux-mm/Z226PG9t-Ih7fJDL@gourry-fedora-PF4VCD3F/T/#m2780e47df7f0962a79182502afc99843bb046205
> 

I've checked our CFMWS configurations using the information from this link and 
wrote them below.

> This isn't exhaustive, but as I said with Rakie, I think your
> configuration is probably ok - if slightly confusing.
> 

I wonder if the below data is sufficient for your review in Rakies's
email [1]. Please let me know.

> What I'm going to guess is happening is you have 1 CFMWS per device that
> do not have matching SRAT entries, and then the CFMWS covering these:
> 
> >  [  0.009924] ACPI: SRAT: Node 2 PXM 6 [mem 0x2080000000-0x807fffffff] hotplug
> >  [  0.009925] ACPI: SRAT: Node 3 PXM 7 [mem 0x64f80000000-0x6cf7fffffff] hotplug
> 
> have interleave set up
> 

In my understanding, both SRAT and CFMWS have the above device and interleave setup.

Below are the SRAT entries (with some unnecessary lines removed) :

[A128h 41256 001h]               Subtable Type : 01 [Memory Affinity]  
[A12Ah 41258 004h]            Proximity Domain : 00000006  
[A130h 41264 008h]                Base Address : 0000002080000000
[A138h 41272 008h]              Address Length : 0000006000000000  
                                     Enabled : 1  
                               Hot Pluggable : 1  
...
  
[A150h 41296 001h]               Subtable Type : 01 [Memory Affinity]  
[A152h 41298 004h]            Proximity Domain : 00000007  
[A158h 41304 008h]                Base Address : 0000064F80000000
[A160h 41312 008h]              Address Length : 0000008000000000  
                                     Enabled : 1  
                               Hot Pluggable : 1  

and below are the CFMWS configurations (with some unnecessary lines removed):

[0A4h 0164 001h]               Subtable Type : 01 [CXL Fixed Memory Window Structure]
[0ACh 0172 008h]         Window base address : 0000002080000000 <- Memory region
[0B4h 0180 008h]                 Window size : 0000032780000000
[0BCh 0188 001h]          Interleave Members : 01           <-- 2-way interleave
[0BDh 0189 001h]       Interleave Arithmetic : 01
[0C8h 0200 004h]                First Target : 00000043     <-- host bridge id  
[0CCh 0204 004h]                 Next Target : 00000053     <-- host bridge id 

...

[170h 0368 001h]               Subtable Type : 01 [CXL Fixed Memory Window Structure]
[178h 0376 008h]         Window base address : 0000064F80000000
[180h 0384 008h]                 Window size : 0000033000000000
[188h 0392 001h]          Interleave Members : 01          <-- 2-way interleave
[189h 0393 001h]       Interleave Arithmetic : 01
[194h 0404 004h]                First Target : 00000143    <-- host bridge id
[198h 0408 004h]                 Next Target : 00000153    <-- host bridge id

[1] https://lore.kernel.org/linux-mm/Z87zpg3TLRReikgu@gourry-fedora-PF4VCD3F/

Best regards,
Yunjeong


  reply	other threads:[~2025-03-11  2:08 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20250228001631.1102-1-yunjeong.mun@sk.com>
2025-02-26 21:35 ` [PATCH 1/2 v6] mm/mempolicy: Weighted Interleave Auto-tuning Joshua Hahn
2025-02-26 21:35   ` [PATCH 2/2 v6] mm/mempolicy: Don't create weight sysfs for memoryless nodes Joshua Hahn
2025-02-27  2:32     ` Honggyu Kim
2025-02-27  3:20       ` Honggyu Kim
2025-03-03 21:56         ` Joshua Hahn
2025-03-04 12:53           ` Honggyu Kim
2025-03-03 16:19       ` Gregory Price
2025-03-04 13:03         ` Honggyu Kim
2025-03-04 16:16           ` Gregory Price
2025-03-04 16:29       ` Gregory Price
2025-03-06 12:39         ` Honggyu Kim
2025-03-06 17:32           ` Gregory Price
2025-03-07 11:46             ` Honggyu Kim
2025-03-07 17:51               ` Gregory Price
2025-03-10 12:26                 ` Honggyu Kim
2025-03-10 14:22                   ` Gregory Price
2025-03-11  2:07                     ` Yunjeong Mun [this message]
2025-03-11  2:42                       ` Gregory Price
2025-03-11  4:02                         ` Yunjeong Mun
2025-03-11  4:42                           ` Gregory Price
2025-03-11  9:51                             ` Yunjeong Mun
2025-03-11 15:52                               ` Gregory Price
2025-03-18  8:02                             ` Yunjeong Mun
2025-03-18 11:02                               ` Honggyu Kim
2025-03-18 15:13                                 ` Gregory Price
2025-03-19  9:56                                   ` Yunjeong Mun
2025-03-19 14:54                                     ` Gregory Price
2025-02-28  0:16   ` [PATCH 1/2 v6] mm/mempolicy: Weighted Interleave Auto-tuning yunjeong.mun
2025-02-28  6:39   ` Yunjeong Mun
2025-02-28 16:24     ` Joshua Hahn
2025-03-04 21:56     ` Joshua Hahn
2025-03-04 22:22       ` Joshua Hahn
2025-03-05  9:49         ` Yunjeong Mun
2025-03-05 16:28           ` Joshua Hahn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250311020806.404-1-yunjeong.mun@sk.com \
    --to=yunjeong.mun@sk.com \
    --cc=Jonathan.Cameron@huawei.com \
    --cc=akpm@linux-foundation.org \
    --cc=dan.j.williams@intel.com \
    --cc=dave.jiang@intel.com \
    --cc=gourry@gourry.net \
    --cc=gregkh@linuxfoundation.org \
    --cc=hannes@cmpxchg.org \
    --cc=harry.yoo@oracle.com \
    --cc=honggyu.kim@sk.com \
    --cc=horen.chuang@linux.dev \
    --cc=joshua.hahnjy@gmail.com \
    --cc=kernel-team@meta.com \
    --cc=kernel_team@skhynix.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=rafael@kernel.org \
    --cc=rakie.kim@sk.com \
    --cc=ying.huang@linux.alibaba.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox