One of our customer has reported this issue. This is easily reproducible in our environment.

What were you trying to do that didn't work?

Running below command led to segfault:

~~~
unused devices: <none>

1. mdadm --assemble --uuid=35fae301:a3a87ac9:8e931def:26314e58
   Segmentation fault (core dumped) <<<

[ 372.605303] mdadm[22362]: segfault at 0 ip 000055b86114f62e sp 00007ffc3b993fd0 error 4 in mdadm[55b86114c000+6e000] likely on CPU 0 (core 0, socket 0)
[ 372.605390] Code: fb 83 f8 03 0f 84 d5 fe ff ff 83 3c 24 01 0f 85 26 fc ff ff 83 bc 24 54 02 00 00 00 0f 85 18 fc ff ff 48 8b 9c 24 d8 00 00 00 <48> 8b 3b e8 aa d9 ff ff 83 7c 24 2c 00 0f 8e 0e 03 00 00 48 8b 33
~~~

What is the impact of this issue to you?

Please provide the package NVR for which the bug is seen:

1. rpm -q mdadm
   mdadm-4.4-2.el9.x86_64

1. uname -r
   5.14.0-611.20.1.el9_7.x86_64

How reproducible is this bug?:

It is 100% reproducible.

Steps to reproduce

1] Create a Raid6 mdadm array with mdadm package version mdadm-4.4-2.el9.x86_64 over rhel 9.7
2] Run below command and gives segfault

1. mdadm --assemble --uuid=35fae301:a3a87ac9:8e931def:26314e58 (Replace uuid with the array uuid)
    
   If we roll back to older version of mdadm then it works fine:

~~
Downgraded:
mdadm-4.3-4.el9_5.x86_64

Complete!

1. mdadm --assemble --uuid=35fae301:a3a87ac9:8e931def:26314e58
   mdadm: an md device must be given in this mode
   ~~~

Expected results

It should give an error message that command is incorrect.

Actual results

It reports segfault.