RFC: guest INTEL GDS mitigation status on patched host

RFC: guest INTEL GDS mitigation status on patched host

[Jinpu Wang]

Hi folks on the list:

I'm testing the latest Downfall cpu vulnerability mitigation. what I
notice is when both host and guest are using patched kernel +
microcode eg kernel 5.15.125 +  intel-microcode 20230808 on affected
server eg Icelake server.

The mitigation status inside guest is:

Vulnerabilities:
  Gather data sampling:  Unknown: Dependent on hyp
                         ervisor status
-----------------------------------> this one.
  Itlb multihit:         Not affected
  L1tf:                  Not affected
  Mds:                   Not affected
  Meltdown:              Not affected
  Mmio stale data:       Vulnerable: Clear CPU buf
                         fers attempted, no microc
                         ode; SMT Host state unkno
                         wn
  Retbleed:              Not affected
  Spec rstack overflow:  Not affected
  Spec store bypass:     Mitigation; Speculative S
                         tore Bypass disabled via
                         prctl and seccomp
  Spectre v1:            Mitigation; usercopy/swap
                         gs barriers and __user po
                         inter sanitization
  Spectre v2:            Mitigation; Enhanced IBRS
                         , IBPB conditional, RSB f
                         illing, PBRSB-eIBRS SW se
                         quence
  Srbds:                 Not affected
  Tsx async abort:       Not affected

According to kernel commit below
commit 81ac7e5d741742d650b4ed6186c4826c1a0631a7
Author: Daniel Sneddon <daniel.sneddon@linux.intel.com>
Date:   Wed Jul 12 19:43:14 2023 -0700

    KVM: Add GDS_NO support to KVM

    Gather Data Sampling (GDS) is a transient execution attack using
    gather instructions from the AVX2 and AVX512 extensions. This attack
    allows malicious code to infer data that was previously stored in
    vector registers. Systems that are not vulnerable to GDS will set the
    GDS_NO bit of the IA32_ARCH_CAPABILITIES MSR. This is useful for VM
    guests that may think they are on vulnerable systems that are, in
    fact, not affected. Guests that are running on affected hosts where
    the mitigation is enabled are protected as if they were running
    on an unaffected system.

    On all hosts that are not affected or that are mitigated, set the
    GDS_NO bit.

    Signed-off-by: Daniel Sneddon <daniel.sneddon@linux.intel.com>
    Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
    Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>

KVM also has the support of GDS_NO, but seems qemu side doesn't pass
the info to guest, that's why it is unknown. IMO qemu should pass
GDS_NO if the host is already patched.

Is Intel or anyone already working on the qemu patch? I know it's not
a must, but good to do.

Thx!
Jinpu Wang @ IONOS Cloud

[PATCH] target/i386: Export GDS_NO bit to guests

[Pawan Gupta]

Gather Data Sampling (GDS) is a side-channel attack using Gather
instructions. Some Intel processors will set ARCH_CAP_GDS_NO bit in
MSR IA32_ARCH_CAPABILITIES to report that they are not vulnerable to
GDS.

Make this bit available to guests.

Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
---
This is just compile tested.


 target/i386/cpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 97ad229d8ba3..48709b77689f 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -1155,7 +1155,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
             NULL, "sbdr-ssdp-no", "fbsdp-no", "psdp-no",
             NULL, "fb-clear", NULL, NULL,
             NULL, NULL, NULL, NULL,
-            "pbrsb-no", NULL, NULL, NULL,
+            "pbrsb-no", NULL, "gds-no", NULL,
             NULL, NULL, NULL, NULL,
         },
         .msr = {
-- 
2.34.1

Re: [PATCH] target/i386: Export GDS_NO bit to guests

[Daniel Sneddon]

On 8/11/23 10:18, Pawan Gupta wrote:
> This is just compile tested.

I used this patch when testing the GDS Linux patches, so you can add a T-B from me.

Re: RFC: guest INTEL GDS mitigation status on patched host

[Jinpu Wang]

Hi Pawan, hi Daniel

Thanks for the patch.

I tried similar patch on Icelake server:
Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Byte Order:                         Little Endian
Address sizes:                      46 bits physical, 57 bits virtual
CPU(s):                             64
On-line CPU(s) list:                0-63
Thread(s) per core:                 2
Core(s) per socket:                 16
Socket(s):                          2
NUMA node(s):                       2
Vendor ID:                          GenuineIntel
CPU family:                         6
Model:                              106
Model name:                         Intel(R) Xeon(R) Gold 6346 CPU @ 3.1
                                    0GHz
Stepping:                           6
CPU MHz:                            3100.000
CPU max MHz:                        3600,0000
CPU min MHz:                        800,0000
BogoMIPS:                           6200.00
Virtualization:                     VT-x
L1d cache:                          1,5 MiB
L1i cache:                          1 MiB
L2 cache:                           40 MiB
L3 cache:                           72 MiB
NUMA node0 CPU(s):                  0,2,4,6,8,10,12,14,16,18,20,22,24,26
                                    ,28,30,32,34,36,38,40,42,44,46,48,50
                                    ,52,54,56,58,60,62
NUMA node1 CPU(s):                  1,3,5,7,9,11,13,15,17,19,21,23,25,27
                                    ,29,31,33,35,37,39,41,43,45,47,49,51
                                    ,53,55,57,59,61,63
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Mitigation; Clear CPU buffers; SMT v
                                    ulnerable
Vulnerability Retbleed:             Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass
                                     disabled via prctl and seccomp
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers
                                     and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Enhanced IBRS, IBPB cond
                                    itional, RSB filling, PBRSB-eIBRS SW
                                     sequence
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected

 target/i386/cpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/target/i386/cpu.c b/target/i386/cpu.c
index 97ad229d8ba3..48709b77689f 100644
--- a/target/i386/cpu.c
+++ b/target/i386/cpu.c
@@ -1155,7 +1155,7 @@ FeatureWordInfo feature_word_info[FEATURE_WORDS] = {
             NULL, "sbdr-ssdp-no", "fbsdp-no", "psdp-no",
             NULL, "fb-clear", NULL, NULL,
             NULL, NULL, NULL, NULL,
-            "pbrsb-no", NULL, NULL, NULL,
+            "pbrsb-no", NULL, "gds-no", NULL,
             NULL, NULL, NULL, NULL,
         },
         .msr = {
-- 
2.34.1
For the change Pawan provided, I tested on Icelake server, it works as expected.
Somehow I'm not cc for the patch, but please consider it tested

Reported-by: Jack Wang <jinpu.wang@ionos.com>
Tested-by: Jack Wang <jinpu.wang@ionos.com>

Thx!
Jinpu Wang


while if I patches QEMU below:



On Fri, Aug 11, 2023 at 3:12 PM Jinpu Wang <jinpu.wang@ionos.com> wrote:
>
> Hi folks on the list:
>
> I'm testing the latest Downfall cpu vulnerability mitigation. what I
> notice is when both host and guest are using patched kernel +
> microcode eg kernel 5.15.125 +  intel-microcode 20230808 on affected
> server eg Icelake server.
>
> The mitigation status inside guest is:
>
> Vulnerabilities:
>   Gather data sampling:  Unknown: Dependent on hyp
>                          ervisor status
> -----------------------------------> this one.
>   Itlb multihit:         Not affected
>   L1tf:                  Not affected
>   Mds:                   Not affected
>   Meltdown:              Not affected
>   Mmio stale data:       Vulnerable: Clear CPU buf
>                          fers attempted, no microc
>                          ode; SMT Host state unkno
>                          wn
>   Retbleed:              Not affected
>   Spec rstack overflow:  Not affected
>   Spec store bypass:     Mitigation; Speculative S
>                          tore Bypass disabled via
>                          prctl and seccomp
>   Spectre v1:            Mitigation; usercopy/swap
>                          gs barriers and __user po
>                          inter sanitization
>   Spectre v2:            Mitigation; Enhanced IBRS
>                          , IBPB conditional, RSB f
>                          illing, PBRSB-eIBRS SW se
>                          quence
>   Srbds:                 Not affected
>   Tsx async abort:       Not affected
>
> According to kernel commit below
> commit 81ac7e5d741742d650b4ed6186c4826c1a0631a7
> Author: Daniel Sneddon <daniel.sneddon@linux.intel.com>
> Date:   Wed Jul 12 19:43:14 2023 -0700
>
>     KVM: Add GDS_NO support to KVM
>
>     Gather Data Sampling (GDS) is a transient execution attack using
>     gather instructions from the AVX2 and AVX512 extensions. This attack
>     allows malicious code to infer data that was previously stored in
>     vector registers. Systems that are not vulnerable to GDS will set the
>     GDS_NO bit of the IA32_ARCH_CAPABILITIES MSR. This is useful for VM
>     guests that may think they are on vulnerable systems that are, in
>     fact, not affected. Guests that are running on affected hosts where
>     the mitigation is enabled are protected as if they were running
>     on an unaffected system.
>
>     On all hosts that are not affected or that are mitigated, set the
>     GDS_NO bit.
>
>     Signed-off-by: Daniel Sneddon <daniel.sneddon@linux.intel.com>
>     Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
>     Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
>
> KVM also has the support of GDS_NO, but seems qemu side doesn't pass
> the info to guest, that's why it is unknown. IMO qemu should pass
> GDS_NO if the host is already patched.
>
> Is Intel or anyone already working on the qemu patch? I know it's not
> a must, but good to do.
>
> Thx!
> Jinpu Wang @ IONOS Cloud