Why does Windows 10 use a separate 100MB partition for UEFI since with UEFI it can read in the filesystem in C:\Windows\System32\winload.efi?

Question

What I've read from UEFI so far is that it has features that BIOS didn't have: it can look in the filesystem and find the boot loader there.

Then, since the UEFI boot system can read directly

C:\Windows\System32\winload.efi

from the main partition for C:\, why is there the need for a separate 100 MB partition when it could just read winload.efi directly in the main partition?

Can't the GPT link that it should use C:\Windows\System32\winload.efi and avoid the 100 MB partition?

(illustrative image, not taken from my system, thus the different partition sizes)

Answer 1

1. UEFI doesn't support NTFS. The spec calls for FAT family support. Vendors could add NTFS support, but:

   • It's not really necessary, because FAT32 is completely sufficient and much less complex
   • Unless all vendors agreed to add NTFS support, it wouldn't be a universally usable configuration anyway.

2. EFI System Partition isn't tied to any particular OS. If you're multibooting, you can store multiple bootloaders on a single partition.

Answer 2

In addition to gronostaj's answer there's another feature that require a separate partition: full disk encryption, aka BitLocker in Microsoft world.

From Microsoft documentation:

BitLocker Drive Encryption Partitioning Requirements

BitLocker must use a system partition that is separate from the Windows partition. The system partition:

• Must be configured as the active partition.
• Must not be encrypted or used to store user files.
• Must have at least 100 megabytes (MB) of space.
• Must have at least 50 MB of free space.
• May be shared with a recovery partition.

The unencrypted partition contains the info needed to read the encrypted partition, and the UEFI can't access this encrypted partition.