linux-kernel - Re: [PATCH v7 0/2] PCI: Configure Root Port MPS during host probing

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <f5322597-c6dd-494a-863b-2caf24363f2d@pardini.net>
Date: Wed, 31 Dec 2025 03:58:33 +0100
From: Ricardo Pardini <ricardo@...dini.net>
To: Hans Zhang <18255117159@....com>, lpieralisi@...nel.org,
 kwilczynski@...nel.org, bhelgaas@...gle.com, helgaas@...nel.org,
 heiko@...ech.de, mani@...nel.org, yue.wang@...ogic.com
Cc: pali@...nel.org, neil.armstrong@...aro.org, robh@...nel.org,
 jingoohan1@...il.com, khilman@...libre.com, jbrunet@...libre.com,
 martin.blumenstingl@...glemail.com, cassel@...nel.org,
 linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-arm-kernel@...ts.infradead.org, linux-amlogic@...ts.infradead.org,
 linux-rockchip@...ts.infradead.org
Subject: Re: [PATCH v7 0/2] PCI: Configure Root Port MPS during host probing

On 27/11/2025 18:09, Hans Zhang wrote:
> Current PCIe initialization exhibits a key optimization gap: Root Ports
> may operate with non-optimal Maximum Payload Size (MPS) settings. While
> downstream device configuration is handled during bus enumeration, Root
> Port MPS values inherited from firmware or hardware defaults often fail
> to utilize the full capabilities supported by controller hardware. This
> results in suboptimal data transfer efficiency throughout the PCIe
> hierarchy.
> 
> This patch series addresses this by:
> 
> 1. Core PCI enhancement (Patch 1):
> - Proactively configures Root Port MPS during host controller probing
> - Sets initial MPS to hardware maximum (128 << dev->pcie_mpss)
> - Conditional on PCIe bus tuning being enabled (PCIE_BUS_TUNE_OFF unset)
>    and not in PCIE_BUS_PEER2PEER mode (which requires default 128 bytes)
> - Maintains backward compatibility via PCIE_BUS_TUNE_OFF check
> - Preserves standard MPS negotiation during downstream enumeration
> 
> 2. Driver cleanup (Patch 2):
> - Removes redundant MPS configuration from Meson PCIe controller driver
> - Functionality is now centralized in PCI core
> - Simplifies driver maintenance long-term
> 
> ---
> Changes in v7:
> - Exclude PCIE_BUS_PEER2PEER mode from Root Port MPS configuration
> - Remove redundant check for upstream bridge (Root Ports don't have one)
> - Improve commit message and code comments as per Bjorn.
Hi Hans,

I've tested on an Odroid-HC4 with a SATA SSD (via an ASM1061) by 
applying your v7 on v6.19-rc3 + Bjorn's 
20251103221930.1831376-1-helgaas@...nel.org ("PCI: meson: Remove 
meson_pcie_link_up() timeout, message, speed check" which is required to 
get the meson PCIe to work at all since 6.18). With that setup I get:

# hdparm --direct -t /dev/sda
  Timing O_DIRECT disk reads: 832 MB in  3.00 seconds = 277.33 MB/sec

I've an identical machine, with a similar disk (even slightly faster, on 
paper), running plain 6.12.y and there I get:

# hdparm --direct -t /dev/sda
  Timing O_DIRECT disk reads: 764 MB in  3.00 seconds = 254.26 MB/sec

I repeated those a few times, not very scientific, I know; but anyway:

Tested-by: Ricardo Pardini <ricardo@...dini.net> # on Odroid-HC4

I've also feedback from another user running with this series with 
success on a different meson PCIe machine, will ask them to TB as well; 
they had reported a significant drop in performance since v6.18 without 
this.

Thanks,
Ricardo