linux-kernel - Re: [PATCH v3 00/16] TDX: Enable Dynamic PAMT

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: <c5115ecbf5b6e3b135f91a5ccc2bf1aa8cdcf0b1.camel@intel.com>
Date: Tue, 30 Sep 2025 18:29:33 +0000
From: "Edgecombe, Rick P" <rick.p.edgecombe@...el.com>
To: "Hansen, Dave" <dave.hansen@...el.com>, "kas@...nel.org" <kas@...nel.org>,
	"Zhao, Yan Y" <yan.y.zhao@...el.com>
CC: "Gao, Chao" <chao.gao@...el.com>, "seanjc@...gle.com" <seanjc@...gle.com>,
	"Huang, Kai" <kai.huang@...el.com>, "dave.hansen@...ux.intel.com"
	<dave.hansen@...ux.intel.com>, "Annapurve, Vishal" <vannapurve@...gle.com>,
	"bp@...en8.de" <bp@...en8.de>, "mingo@...hat.com" <mingo@...hat.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"tglx@...utronix.de" <tglx@...utronix.de>, "kvm@...r.kernel.org"
	<kvm@...r.kernel.org>, "linux-coco@...ts.linux.dev"
	<linux-coco@...ts.linux.dev>, "Yamahata, Isaku" <isaku.yamahata@...el.com>,
	"pbonzini@...hat.com" <pbonzini@...hat.com>, "x86@...nel.org"
	<x86@...nel.org>
Subject: Re: [PATCH v3 00/16] TDX: Enable Dynamic PAMT

On Mon, 2025-09-29 at 09:58 -0700, Rick Edgecombe wrote:
> It might serve a purpose of proving that scalability is possible. I was thinking
> if we had line of sight to improving it we could go with the "simple" solution
> and wait until there is a problem. Is it reasonable?

We discussed this offline and the conclusion was to just keep a global
reader/writer lock as the backpocket solution. So DPAMT seamcalls take the
global reader, and if they get a BUSY code, they take the global lock as write
and retry. Similar to what KVM does on a per-VM scope, but globally and reduced
by the refcount.

But until there is a measured issue, just keep the simpler global exclusive lock
with the refcount to start. So unless anyone new chimes in, we can call this
closed.