linux-kernel - x86: CR4 update when IRQs are enabled

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]

Date:   Wed, 8 Nov 2017 13:52:17 -0800
From:   Nadav Amit <nadav.amit@...il.com>
To:     LKML <linux-kernel@...r.kernel.org>,
        Andy Lutomirski <luto@...nel.org>
Subject: x86: CR4 update when IRQs are enabled

I am puzzled by the comment in tlb_state.cr4 , which says:

        /*
         * Access to this CR4 shadow and to H/W CR4 is protected by
         * disabling interrupts when modifying either one.
         */

This does not seem to be true and adding a warning to CR4 writes when
!irqs_disabled() immediately fires in identify_cpu() and
__mcheck_cpu_init_generic(). While these two are called during boot, I think
there other CR4 changes with enabled IRQs, for example, PR_SET_TSC.

So my question(s): Is the comment correct? Is the current behavior correct?

Thanks,
Nadav