linux-kernel - Re: x86/amd late microcode thread loading slows down boot

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: <6449aa6c-43b1-40eb-ab54-48803481d11b@citrix.com>
Date: Wed, 6 Nov 2024 18:50:12 +0000
From: Andrew Cooper <andrew.cooper3@...rix.com>
To: thomas.de_schampheleire@...ia.com
Cc: bp@...en8.de, linux-kernel@...r.kernel.org, x86@...nel.org
Subject: Re: x86/amd late microcode thread loading slows down boot

> Hi Borislav, all,
>
> I am encountering varying delays in the boot process

I recognise those symptoms.

https://xenbits.xen.org/gitweb/?p=xen.git;a=commitdiff;h=f19a199281a23725beb73bef61eb8964d8e225ce

We found it in Xen in 2019, and were instructed by AMD to insert a TLB
flush immediately after patchloading.

Looking at Linux, there's no such workaround.

It turns out that a side effect of patchloading on some CPUs leaves the
mapping of the blob in the TLB (at whatever granularity) as fully UC. 
When this happens to be a 2M/1G superpage, it causes whole lot of perf
problems in unrelated areas.

I suggest Linux take the same approach.

~Andrew