SSD partitions not available anymore after partition resizing from Windows

[elgrande]

Hello,

I performed a partition resizing from within Samsung Magician on my Windows 11 partition.

After that my EFI partition /dev/nvd0p1 mounted under /boot/efi was not available anymore under /dev:

elgrande@elfreebsdo:~ % ls /dev/nvd*
/dev/nvd0
elgrande@elfreebsdo:~

The partition layout is as follows now:

elgrande@elfreebsdo:~ % gpart show
=>        34  3907029101  diskid/DISK-S69ENF0RXXXXXXF  GPT  (1.8T)
          34        2014                               - free -  (1.0M)
        2048     3072000                            1  efi  (1.5G)
     3074048       32768                            2  ms-reserved  (16M)
     3106816  2843703296                            3  ms-basic-data  (1.3T)
  2846810112   390701056                               - free -  (186G)
  3237511168     1298432                            4  ms-recovery  (634M)
  3238809600    39071744                               - free -  (19G)
  3277881344   268435456                            5  freebsd-zfs  (128G)
  3546316800   360712335                               - free -  (172G)

elgrande@elfreebsdo:~ %

I could solve this by mounting via /dev/diskid/DISK-S69ENF0RXXXXXXF, but I would like to understand the issue.

Any ideas?
To me that free space at the beginning looks strange, but Windows does not even recognize this free space.

 

[dave01]

The free space at the beginning is "normal" for Windows. My dual boot system is similar, but without the EFI slice.

I can't help further, but a first guess would be that the partition manager you used to shrink the Windows partitions did something bad to your partition table, possibly because it didn't recognise the non-Windows partitions.

What are the multi-GB gaps in your slices for? Is that the results of the shrinking?

 

[elgrande]

Well everything works now, so I will probably not mess around anymore. Only the /dev/nvd0p* devices are missing, but I can work around that.
The multi-GB gaps are due to historical reasons, I played a lot around with different OS'es and so.
I am just curious what is wrong now on the FreeBSD side with the /dev/nvd0p* devices.