linux-kernel - Re: Is 2nd Generation Intel(R) Xeon(R) Processors (Formerly Cascade Lake) affected by MDS

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:   Fri, 24 May 2019 17:14:38 +0200
From:   Jinpu Wang <jinpu.wang@...ud.ionos.com>
To:     Tony Luck <tony.luck@...il.com>
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        "Luck, Tony" <tony.luck@...el.com>,
        Arjan van de Ven <arjan@...ux.intel.com>,
        linux-kernel@...r.kernel.org,
        Elmar Gerdes <elmar.gerdes@...ud.ionos.com>
Subject: Re: Is 2nd Generation Intel(R) Xeon(R) Processors (Formerly Cascade
 Lake) affected by MDS

On Fri, May 24, 2019 at 4:45 PM Tony Luck <tony.luck@...il.com> wrote:
>
> Stepping 5 is preproduction of cascade lake. MDS mitigation is in production parts (stepping >= 6)
>
> Sent from my iPhone
>
> > On May 24, 2019, at 07:17, Greg Kroah-Hartman <gregkh@...uxfoundation.org> wrote:
> >
> >> On Fri, May 24, 2019 at 03:19:34PM +0200, Jinpu Wang wrote:
> >> Resend with plain text, and remove confidential unnecessary signature.
> >> sorry for spam.
> >>
> >> Hi Thomas, hi Greg, hi Tony, hi Arjan, hi other expert on the list
> >>
> >> I noticed on our Cascade lake with 4.14.120,  the kernel is reporting
> >> vulnerable:
> >> jwang@...01a-912:~$ head /sys/devices/system/cpu/vulnerabilities/mds
> >> Vulnerable: Clear CPU buffers attempted, no microcode; SMT vulnerable
> >>
> >> But according to INTEL,  they have built the mitigation in hardware
> >> for Cascade Lake:
> >> https://www.intel.com/content/www/us/en/architecture-and-technology/engineering-new-protections-into-hardware.html
> >>
> >> We are using latest microcode from debian:
> >> https://metadata.ftp-master.debian.org/changelogs//non-free/i/intel-microcode/intel-microcode_3.20190514.1~deb9u1_changelog
> >> lscpu:
> >> jwang@...01a-912:~$ lscpu
> >> Architecture:          x86_64
> >> CPU op-mode(s):        32-bit, 64-bit
> >> Byte Order:            Little Endian
> >> CPU(s):                96
> >> On-line CPU(s) list:   0-95
> >> Thread(s) per core:    2
> >> Core(s) per socket:    24
> >> Socket(s):             2
> >> NUMA node(s):          2
> >> Vendor ID:             GenuineIntel
> >> CPU family:            6
> >> Model:                 85
> >> Model name:            Intel(R) Xeon(R) Platinum 8268 CPU @ 2.90GHz
> >> Stepping:              5
> >> CPU MHz:               3228.226
> >> CPU max MHz:           3900.0000
> >> CPU min MHz:           1000.0000
> >> BogoMIPS:              5800.00
> >> Virtualization:        VT-x
> >> L1d cache:             32K
> >> L1i cache:             32K
> >> L2 cache:              1024K
> >> L3 cache:              33792K
> >> NUMA node0 CPU(s):     0-23,48-71
> >> NUMA node1 CPU(s):     24-47,72-95
> >> Flags:                 fpu vme de pse tsc msr pae mce cx8 apic sep
> >> mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht
> >> tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs
> >> bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq
> >> dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm
> >> pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes
> >> xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3
> >> cdp_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow
> >> vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2
> >> erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap
> >> clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec
> >> xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local
> >> dtherm ida arat pln pts pku ospke flush_l1d arch_capabilities
> >>
> >> Clearly we want to buy Cascade Lake if the hardware mitigations are in
> >> place, but the information on the internet is quite limited and
> >> misleading.
> >>
> >> Could anyone of you could clarify it for us?
> >> Does it mean 4.14.120 is missing patches for detecting Cascade Lake
> >> mitigation, Or maybe only small set of cascade lake has mitigation?
> >> Or the community/intel are not sure about it yet?
> >
> > Did you try 4.19 or 5.1?  Try those and see if anything changes.
> >
> > If not, odds are you need a microcode update from Intel, please contact
> > them, nothing we can do about that.
> >
> > If 4.14 works different from 4.19 or 5.1, please let us know.
> >
> > thanks,
> >
> > greg k-h

Thanks, Greg, thanks Tony,

I tried also 4.19.46 kernel, it show the same.

As Tony mentioned, it's due to the stepping difference, our test
machine is a preproduction one.
I would expect the production Cascade Lake supports md-clear feature,
so VM migration works from old generation.
Could some one give me an answer?

Thanks a lot.
-- 
Jack Wang
Linux Kernel Developer @ 1 & 1 Cloud IONOS GmbH