properly label PCFS mount type

[Mythra]

now that we've got an open-source way of booting a cat-dev using PCFS without the official SDK (https://codeberg.org/rem-verse/sprig, this also contains the source code i used to generate the image below at: https://codeberg.org/rem-verse/sprig/src/commit/760514b5e762de0edbf466a0d3f3ce06ee7a5513/console/apps/hfiod), we can confirm that this second mount source is indeed PCFS. Testing this code on a real cat-dev, allows me to list files on the attached host system.

---

I chose to label this "PCFS" even though based on the mount name says "HFIO" (I assume for "host file i/o"), while these deviates from the mount name, PCFS is much more clear as PCFS is not the only way a cat-dev can do i/o with an attached 'host' system (none of the others are supported by this mount type), not to mention PCFS has certain limitations (e.g. can only do 'simple' filesystem permissions, supports the max length of PCFS, etc.).

[GaryOderNichts]

I'm not sure about your reasoning with the name in this case.
Isn't PCFS the filesystem driver? Wouldn't that be the same as naming the SD mountpoint FS_MOUNT_SOURCE_FAT?

[Mythra]

Yes, and no. "PCFS" is a filesystem driver. However, it is also the general name for all the different filesystem drivers (e.g. the web page on a cat dev calls settings for various filesystem drivers driven by a host PC "PCFS settings", which then lists the various host filesystem drivers including another PCFS), and it's the name of the boot mode of the device (e.g. when the boot the device it will report as "NAND", "PCFS", or "DUAL" in its UI).

You're right that in this case calling it PCFS because of its limitations is akin to calling it the "FAT" driver, but also it is kind of the generic term.

I can rename it to HFIO, and leave a comment explaining, but that may also have its own confusions (not only "which limitations am I under", but also where's the thing called PCFS").