Raphael, I'd like your help upstreaming this VMD power-saving patch, please

["Kenneth R. Crudup" <kenny@panix.com>]

[-- Attachment #1: Type: text/plain, Size: 1485 bytes --]


Hello Raphael,

For almost two years now, I've been trying to get patches from Ubuntu that
enable ASPM for devices behind Intel's VMD, necessary to get full lower-power
states (including very-reduced power usage during s0ix sleep) on my Alderlake
(et al.) laptop, upstreamed into mainline.

One such thread: https://lore.kernel.org/linux-pm/218aa81f-9c6-5929-578d-8dc15f83dd48@panix.com/

Since the original set of patches on this, most of the work has been pushed
upstream, with only this last patch required to get fully into the "CPU%LPI"
and "SYS%LPI" (names according to "turbostat") states.

I'm surprised that with the number of VMD-enabled laptops out there (which I
had to keep on so I could dual-boot into Win11 (the disk geometry changes if
I disable it, rendering the Win11 partition useless)), that there haven't been
many reports of excessive power usage in Linux during sleep; perhaps because
many installations are running stock Ubuntu kernels (where I assume variants
of this patch remain) it isn't an issue, but I do believe having this upstreamed
is still valuable.

I don't have the resources you've got to test this fully for regressions, nor
the expertise getting a patch into the kernel, so I'd like to again bring this
up for discussion (hence the phone-book of a CC: here).

If there's anything I can do to help get this done, please let me know.

Thank you,

-Kenneth Crudup

-- 
Kenneth R. Crudup / Sr. SW Engineer, Scott County Consulting, Orange County CA

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: Type: text/x-diff; name=0001-PCI-ASPM-Fixup-ASPM-for-VMD-bridges.patch, Size: 2172 bytes --]

From ee3618a598a261bbc8a875557d42d6dbbbc4cdd0 Mon Sep 17 00:00:00 2001
From: "Kenneth R. Crudup" <kenny@panix.com>
Date: Fri, 13 Dec 2024 15:28:42 -0800
Subject: [PATCH] PCI/ASPM: Fixup ASPM for VMD bridges

Effectively a squashed commit of:
UBUNTU: SAUCE: PCI/ASPM: Enable ASPM for links under VMD domain
UBUNTU: SAUCE: PCI/ASPM: Enable LTR for endpoints behind VMD
UBUNTU: SAUCE: vmd: fixup bridge ASPM by driver name instead
---
 drivers/pci/pcie/aspm.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 29fcb0689a91..fdc1ce2755ff 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -788,6 +788,31 @@ static void aspm_l1ss_init(struct pcie_link_state *link)
 		aspm_calc_l12_info(link, parent_l1ss_cap, child_l1ss_cap);
 }
 
+/*
+ * BIOS may not be able to access config space of devices under VMD domain, so
+ * it relies on software to enable ASPM for links under VMD.
+ */
+static bool pci_fixup_vmd_bridge_enable_aspm(struct pci_dev *pdev)
+{
+       struct pci_bus *bus = pdev->bus;
+       struct device *dev;
+       struct pci_driver *pdrv;
+
+       if (!pci_is_root_bus(bus))
+               return false;
+
+       dev = bus->bridge->parent;
+       if (dev == NULL)
+               return false;
+
+       pdrv = pci_dev_driver(to_pci_dev(dev));
+       if (pdrv == NULL || strcmp("vmd", pdrv->name))
+               return false;
+
+       pci_info(pdev, "enable ASPM for pci bridge behind vmd");
+       return true;
+}
+
 static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 {
 	struct pci_dev *child = link->downstream, *parent = link->pdev;
@@ -866,7 +891,8 @@ static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 	}
 
 	/* Save default state */
-	link->aspm_default = link->aspm_enabled;
+	link->aspm_default = pci_fixup_vmd_bridge_enable_aspm(parent) ?
+		PCIE_LINK_STATE_ASPM_ALL : link->aspm_enabled;
 
 	/* Setup initial capable state. Will be updated later */
 	link->aspm_capable = link->aspm_support;
-- 
2.48.1