linux-kernel - Re: [PATCH] Revert "PCI: qcom: Remove custom ASPM enablement code"

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

Message-ID: <20251027223517.GA1484052@bhelgaas>
Date: Mon, 27 Oct 2025 17:35:17 -0500
From: Bjorn Helgaas <helgaas@...nel.org>
To: Johan Hovold <johan@...nel.org>
Cc: Manivannan Sadhasivam <mani@...nel.org>, linux-pci@...r.kernel.org,
	Manivannan Sadhasivam <manivannan.sadhasivam@....qualcomm.com>,
	Krzysztof Wilczyński <kwilczynski@...nel.org>,
	Frank Li <Frank.li@....com>, Shawn Lin <shawn.lin@...k-chips.com>,
	Rob Herring <robh@...nel.org>,
	"David E . Box" <david.e.box@...ux.intel.com>,
	Kai-Heng Feng <kai.heng.feng@...onical.com>,
	"Rafael J . Wysocki" <rafael@...nel.org>,
	Heiner Kallweit <hkallweit1@...il.com>,
	Chia-Lin Kao <acelan.kao@...onical.com>,
	Gustavo Pimentel <Gustavo.Pimentel@...opsys.com>,
	Han Jingoo <jingoohan1@...il.com>,
	Bjorn Andersson <andersson@...nel.org>,
	Konrad Dybcio <konrad.dybcio@....qualcomm.com>,
	Bartosz Golaszewski <brgl@...ev.pl>, linux-arm-msm@...r.kernel.org,
	linux-kernel@...r.kernel.org, Bjorn Helgaas <bhelgaas@...gle.com>
Subject: Re: [PATCH] Revert "PCI: qcom: Remove custom ASPM enablement code"

On Mon, Oct 27, 2025 at 11:07:51AM +0100, Johan Hovold wrote:
> On Sun, Oct 26, 2025 at 02:37:54PM -0500, Bjorn Helgaas wrote:
> > On Sun, Oct 26, 2025 at 08:58:29PM +0530, Manivannan Sadhasivam wrote:
> 
> > As far as I know, it's L1SS that has catastrophic effects.  I haven't
> > seen anything for L0s or L1.
> 
> Enabling L0s unconditionally certainly blew up on some Qualcomm
> machines. See commit d1997c987814 ("PCI: qcom: Disable ASPM L0s for
> sc8280xp, sa8540p and sa8295p").

Ah, right, thanks for that reminder.

IIUC the qcom_pcie_clear_aspm_l0s() quirk that removes L0s from the
advertised Link Capabilities is still there and prevents df5192d9bb0e
("PCI/ASPM: Enable only L0s and L1 for devicetree platforms") from
being a problem on those platforms, right?