linux-kernel - Re: [PATCH v3 4/4] x86/cpu/topology: Check for X86_FEATURE_XTOPOLOGY instead of passing has

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: <ce208192-e1c0-4bcc-8c27-31b2af683872@amd.com>
Date: Mon, 25 Aug 2025 09:35:39 +0530
From: K Prateek Nayak <kprateek.nayak@....com>
To: Thomas Gleixner <tglx@...utronix.de>, Ingo Molnar <mingo@...hat.com>,
	Borislav Petkov <bp@...en8.de>, Dave Hansen <dave.hansen@...ux.intel.com>,
	Sean Christopherson <seanjc@...gle.com>, Paolo Bonzini <pbonzini@...hat.com>,
	<x86@...nel.org>
CC: Naveen rao <naveen.rao@....com>, Sairaj Kodilkar <sarunkod@....com>, "H.
 Peter Anvin" <hpa@...or.com>, "Peter Zijlstra (Intel)"
	<peterz@...radead.org>, "Xin Li (Intel)" <xin@...or.com>, Pawan Gupta
	<pawan.kumar.gupta@...ux.intel.com>, <linux-kernel@...r.kernel.org>,
	<kvm@...r.kernel.org>, Mario Limonciello <mario.limonciello@....com>,
	"Gautham R. Shenoy" <gautham.shenoy@....com>, Babu Moger
	<babu.moger@....com>, Suravee Suthikulpanit <suravee.suthikulpanit@....com>
Subject: Re: [PATCH v3 4/4] x86/cpu/topology: Check for X86_FEATURE_XTOPOLOGY
 instead of passing has_topoext

Hello Thomas,

On 8/24/2025 8:25 PM, Thomas Gleixner wrote:
> On Sun, Aug 24 2025 at 15:38, Thomas Gleixner wrote:
>>> -	if (!has_topoext) {
>>> +	if (!cpu_feature_enabled(X86_FEATURE_XTOPOLOGY)) {
>>>  		/*
>>>  		 * Prefer initial_apicid parsed from XTOPOLOGY leaf
>>>  		 * 0x8000026 or 0xb if available. Otherwise prefer the
>>
>> That's patently wrong.
>>
>> The leaves might be "available", but are not guaranteed to be valid. So
>> FEATURE_XTOPOLOGY gives you the wrong answer.
>>
>> The has_topoext logic is there for a reason.
> 
> Hrm. I have to correct myself. It's set by the 0xb... parsing when that
> finds a valid leaf. My memory tricked me on that.
> 
> So yes, it can be used for that, but that's a cleanup. The simple fix
> should be applied first as that's trivial to backport.

Thank you for taking a look at the series. I'll move the fixes ahead and
refresh the current Patch 2 with your suggested diff in the next
version.

-- 
Thanks and Regards,
Prateek