linux-kernel - [PATCH v2 0/6] x86: Add stright-line-speculation mitigations

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

Message-ID: <20211204134338.760603010@infradead.org>
Date:   Sat, 04 Dec 2021 14:43:38 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     x86@...nel.org
Cc:     linux-kernel@...r.kernel.org, peterz@...radead.org,
        keescook@...omium.org, hjl.tools@...il.com,
        andrew.cooper3@...rix.com, mark.rutland@....com, will@...nel.org,
        ndesaulniers@...gle.com
Subject: [PATCH v2 0/6] x86: Add stright-line-speculation mitigations

Respin and split of the SLS patch.

Since the old patch hit a snag, I've split up the patch as per popular demand.

New this time are patches 1 and 5 and some minor edits to the last patch, the
rest is mostly identical.

Patch 1 removes a 32bit (abuse) of the RET macro; which got exposed by the
build robot due to a change in the last patch (defining RET even for
!CONFIG_SLS). It would've been possible to revert to the old method of relying
on AS being case insensitive and RET being a valid instruction etc.. but I
figured that code was convoluted enough to warrant cleaning up anyway.

Patch 5 is what made Boris' machine go BUG when he tried to use the SLS patch.

Combined it seems to boot and build a kernel for me, so it must be perfect this
time (fingers crossed).