My computer has a roughly 512 GB NVMe SSD, but Linux live environments detect the drive as 0 B, so the installer reports that there is not enough storage. I have tried enabling AHCI, disabling RST, confirming that the drive is not configured as RAID, and removing BitLocker encryption. The SSD is an Intel model, and I suspect Intel VMD or a firmware configuration may be preventing Linux from accessing it. Windows reports that the system uses UEFI, but the firmware interface appears to boot in legacy mode, and selecting UEFI firmware settings sends me back to the legacy interface. Could I shrink the Windows partition, leave around 12 GB for a Windows-based installer, and install Linux into the remaining space without using a USB drive?
2 Answers
Linux can be installed without a USB, including through a Windows-based installer or network boot, but those methods are more complicated and can damage the existing boot setup if something goes wrong. Creating a Windows installer or recovery drive and making a complete backup first is strongly recommended. Also, simply reserving 12 GB for Windows will not make Linux see the other 500 GB—the installer must be able to access the SSD at the controller level.
If commands such as `fdisk -l`, `lsblk`, and `blkid` detect the NVMe device but report its size as 0 B, the problem is probably not the filesystem or BitLocker. It points more toward the firmware storage mode or a controller such as Intel VMD/RST not being supported or correctly exposed to the Linux kernel. Check the firmware for VMD, RST, RAID, and AHCI settings, and make sure the system is consistently configured for UEFI rather than mixing legacy and UEFI boot modes. Back up Windows before changing storage settings, because switching modes can make Windows unbootable.
The fact that the drive appears but has a zero-byte capacity is an important clue. A Linux installer running inside Windows would not necessarily solve the underlying controller issue; Linux still needs a working way to access the NVMe device.