linux-kernel - Re: [External] Re: [PATCH v6 06/11] x86/smpboot: Support parallel startup of secondary CPUs

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]

Date:   Fri, 3 Feb 2023 06:41:50 -0800
From:   Arjan van de Ven <arjan@...ux.intel.com>
To:     David Woodhouse <dwmw2@...radead.org>,
        Usama Arif <usama.arif@...edance.com>, tglx@...utronix.de
Cc:     mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com,
        hpa@...or.com, x86@...nel.org, pbonzini@...hat.com,
        paulmck@...nel.org, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org, rcu@...r.kernel.org, mimoja@...oja.de,
        hewenliang4@...wei.com, thomas.lendacky@....com, seanjc@...gle.com,
        pmenzel@...gen.mpg.de, fam.zheng@...edance.com,
        punit.agrawal@...edance.com, simon.evans@...edance.com,
        liangma@...ngbit.com
Subject: Re: [External] Re: [PATCH v6 06/11] x86/smpboot: Support parallel
 startup of secondary CPUs

On 2/3/2023 12:14 AM, David Woodhouse wrote:
> On Thu, 2023-02-02 at 22:50 +0000, Usama Arif wrote:
>>
>> Ah, sorry about that, should have caught it while reviewing before
>> posting. To think of it, it might be better to squash this and next AMD
>> disabling patch, if anyone testing on AMD ever happens to check out at
>> this specific patch, their machine might not boot :).
>>
> 
> Nah, we don't actually *enable* parallel boot until the next patch
> anyway. At this point we're only laying the groundwork. I think it's
> best separate.
> 
>> Will repost only after getting the OK for the rest of the patches
>> along with addressing any other comments that come up.
> 
> Ack. And hopefully the missing SoB from both Thomas and Arjan for their
> parts. Part of the reasoning for the lack of SoB from Arjan might have
> been that we haven't finished doing the thinking about how this works
> in all cases, with non-constant TSC and other stuff. Let's make sure we
> give that a long hard stare.

most of it is that I need to sit down for some real time and have a think through it ;)

for all the common cases / modern HW it will be ok... but all the rest???

maybe we need a strict(er) safety valve on where this is used and then over time
widen the usage?

(after all, a regression means revert, and maybe it's better to only revert "add CPU type X" than
revert the whole series)