linux-acpi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Duran, Leo" <leo.duran@amd.com>
To: "linux-acpi@vger.kernel.org" <linux-acpi@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>, Len Brown <lenb@kernel.org>
Subject: Parsing PXM from ACPI (DSDT)
Date: Wed, 14 Aug 2019 21:29:49 +0000	[thread overview]
Message-ID: <CY4PR12MB1815251BBCB062AAA8EEB39CF9AD0@CY4PR12MB1815.namprd12.prod.outlook.com> (raw)
In-Reply-To: <CY4PR12MB1815E6208E334CE4EA8F2670F9AD0@CY4PR12MB1815.namprd12.prod.outlook.com>

Hello,
Is not clear or evident to me if the kernel parses _PXM values below (or under) the root-complex.

For example, in my experience:
This ASL sample for PXM at the root-complex level produces the expected NUMA assignment from “lstopo”:
Scope (\_SB) {
  // ...
  Device (PCI0) { // Root PCI Bus (Host-Bridge)
    Name (_HID, EISAID("PNP0A08"))
    Name (_CID, EISAID("PNP0A03"))
    Name (_BBN, 0)
    Method (_CRS,0) {
      // Return current resources for host bridge 0
    }
    Name (_PRT, Package() {
     // Package with PCI IRQ routing table information
    })
    Method (_PXM, 0, NotSerialized) {
      Return (0)
    }
  }
  // ...
}

However,
This ASL sample for PXM at the P2P root-bridge level does not produce the expected NUMA assignment from “lstopo”:
 (Of course, the assumption is that multiple NUMA nodes actually exist on the system)

 Scope (\_SB) {
  // ...
  Device (PCI0) { // Root PCI Bus (Host-Bridge)
    Name (_HID, EISAID ("PNP0A08"))
    Name (_CID, EISAID ("PNP0A03"))
    Name (_BBN, 0)
    Method (_CRS,0) {
      // Return current resources for host bridge 0
    }
    Name (_PRT, Package() {
      // Package with PCI IRQ routing table information
    })

    Device (P2P0) { // First PCI-to-PCI bridge (Port0)
      Name (_ADR, 0x00070001) // Device#7h, Func#1 on bus PCI0
      Name (_PRT, Package() {
        // Package with PCI IRQ routing table information
      })
      Method (_PXM, 0, NotSerialized) {
        Return (0)
      }
    }

    Device (P2P1) { // Second PCI-to-PCI bridge (Port1)
      Name (_ADR ,0x00080001) // Device#8h, Func#1 on bus PCI0
      Name (_PRT, Package() {
        // Package with PCI IRQ routing table information
      }) 
      Method (_PXM, 0, NotSerialized) {
         Return (1)
      }
    }
  }
  // ...
}

Thanks,
Leo.


       reply	other threads:[~2019-08-14 21:30 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CY4PR12MB1815927E36A8A9A0E5B18734F9D20@CY4PR12MB1815.namprd12.prod.outlook.com>
     [not found] ` <CY4PR12MB18156AA493BA8F3C28F0377FF9AD0@CY4PR12MB1815.namprd12.prod.outlook.com>
     [not found]   ` <CY4PR12MB1815E6208E334CE4EA8F2670F9AD0@CY4PR12MB1815.namprd12.prod.outlook.com>
2019-08-14 21:29     ` Duran, Leo [this message]
2019-08-14 21:44       ` Parsing PXM from ACPI (DSDT) Schmauss, Erik

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=CY4PR12MB1815251BBCB062AAA8EEB39CF9AD0@CY4PR12MB1815.namprd12.prod.outlook.com \
    --to=leo.duran@amd.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rjw@rjwysocki.net \
    /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;
as well as URLs for NNTP newsgroup(s).