linux-kernel - Re: [PATCH] x86/smp: Factor out parts of native_smp_prepare

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:   Mon, 8 Nov 2021 12:20:26 -0500
From:   Boris Ostrovsky <boris.ostrovsky@...cle.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     linux-kernel@...r.kernel.org, xen-devel@...ts.xenproject.org,
        tglx@...utronix.de, bp@...en8.de, dave.hansen@...ux.intel.com,
        x86@...nel.org, hpa@...or.com, jgross@...e.com
Subject: Re: [PATCH] x86/smp: Factor out parts of native_smp_prepare_cpus()


On 11/8/21 10:11 AM, Peter Zijlstra wrote:
> On Tue, Nov 02, 2021 at 07:36:36PM -0400, Boris Ostrovsky wrote:
>> Commit 66558b730f25 ("sched: Add cluster scheduler level for x86")
>> introduced cpu_l2c_shared_map mask which is expected to be initialized
>> by smp_op.smp_prepare_cpus(). That commit only updated
>> native_smp_prepare_cpus() version but not xen_pv_smp_prepare_cpus().
>> As result Xen PV guests crash in set_cpu_sibling_map().
>>
>> While the new mask can be allocated in xen_pv_smp_prepare_cpus() one can
>> see that both versions of smp_prepare_cpus ops share a number of common
>> operations that can be factored out. So do that instead.
>>
>> Fixes: 66558b730f25 ("sched: Add cluster scheduler level for x86")
>> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@...cle.com>
> Thanks! I'll go stick that somewhere /urgent (I've had another report on
> that here:
>
>    https://lkml.kernel.org/r/20211105074139.GE174703@worktop.programming.kicks-ass.net
> )


Thank you. (I don't see this message btw)


>
> But looking at those functions; there seems to be more spurious
> differences. For example, the whole sched_topology thing.


I did look at that and thought this should be benign given that Xen PV is not really topology-aware. I didn't see anything that would be a cause for concern but perhaps you can point me to things I missed.


>
> Should we re-architect this whole smp_prepare_cpus() thing instead? Have
> a common function and a guest function? HyperV for instance seems to
> call native_smp_prepare_cpus() and then does something extra (as does
> xen_hvm).


Something like


void smp_prepare_cpus()

{

     // Code that this patch moved to smp_prepare_cpus_common();


    smp_ops.smp_prepare_cpus();  // Including baremetal

}


?


XenHVM and hyperV will need to call native smp_op too. Not sure this will be prettier than what it is now?



-boris


-boris