linux-kernel - Re: [PATCH v5 20/34] x86/fred: add a machine check entry stub for FRED

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:   Tue, 21 Mar 2023 09:59:53 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Li, Xin3" <xin3.li@...el.com>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>,
        "kvm@...r.kernel.org" <kvm@...r.kernel.org>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "bp@...en8.de" <bp@...en8.de>,
        "dave.hansen@...ux.intel.com" <dave.hansen@...ux.intel.com>,
        "hpa@...or.com" <hpa@...or.com>,
        "andrew.cooper3@...rix.com" <andrew.cooper3@...rix.com>,
        "Christopherson,, Sean" <seanjc@...gle.com>,
        "pbonzini@...hat.com" <pbonzini@...hat.com>,
        "Shankar, Ravi V" <ravi.v.shankar@...el.com>
Subject: Re: [PATCH v5 20/34] x86/fred: add a machine check entry stub for
 FRED

On Tue, Mar 21, 2023 at 12:04:47AM +0000, Li, Xin3 wrote:
> > > Unlike IDT, no need to save/restore dr7 in FRED machine check handler.
> > 
> > Given how fragile MCE is, the question should be, do we ever want hw
> > breakpoints to happen while it is running?
> 
> HW breakpoints still work if they are properly configured.
> 
> > If the hw-breakpoint handler trips on the same memory fail that got us into the
> > mce the first time, we're dead.
> 
> Right.
> 
> Unless the MCIP bit is turned off any subsequent #MC goes to shutdown
> ("machine is screwed").
> 
> It's the kernel debugger's responsibility to decide how to proceed in such
> cases. But if the kernel debugger itself is in a screwed memory region, we
> are soooooo dead.

Yeah, so I would much prefer, for robustness sake, to start out with not
allowing #DB in MCE -- much like today.