both firmwares force 4096 sector size

[schmorp]

Both of the firmwares (index.html && index2.html) force 4096 byte sector size, regardless of any switches selected. This makes the majority of devices (which are 512e nowadays) unusable with these.

It would be helpful if this were at least documented (if not fixed), as otherwise everybody flashing this will wonder where their partitions are, as linux no longer detects the partition table on those disks.

[Zibri]

Nope.. it does not. The switches in index2 all work.
Perhaps you didn't flash correctly or you flashed the wrong version.

[gitercn]

@Zibri
I ran into the same issue. I found that firmware built via https://zibri.github.io/JMS579/index2.html (version 5.1.5) yields a LogicalSectorSize of 4096 in Windows (checked via Get-Disk | Format-List Number, FriendlyName, LogicalSectorSize, PhysicalSectorSize in PowerShell), regardless of whether “Enable 4K block size” is enabled or not.

In contrast, firmware built via https://zibri.github.io/JMS579/index.html (version 4.1.4) shows LogicalSectorSize = 512 when “Enable 4K block emulation” is turned off.

That struck me as odd, so I inspected the JavaScript. In script2.js we have:

(!ken.checked << 7) | (uasp.checked << 6) | (multisec.checked << 5) | (z[0xc4e7].charCodeAt(0) & 0x1F)

While in script.js:

(usb_storage.checked << 7) | (uasp.checked << 6) | (kemu.checked << 5) | (z[0xc4e7].charCodeAt(0) & 0x1F)

I suspect that in index2, the MultiSec checkbox might be controlling the 4K block size behavior. When I disabled that in index2, the 5.1.5 firmware it produced now shows LogicalSectorSize = 512.

That said, I’m not yet sure whether the script is wrong, or whether my method of checking for 4K block size is incorrect. Also, what exactly do the bit-shifts by 7 and 5 represent in that script? (I couldn’t find documentation explaining them.)

By the way, here is the stock firmware of a WD Elements 12 TB (which reports LogicalSectorSize = 512). But its sleep time is too short, plus I want to flash a more general-purpose firmware so I can reuse it with other drives. That’s how I traced this down.

jms579-original-wd-elements.zip