Maybe it is time to start a table of FLASH Chips that do or not do work with BT81x.
Detect: REG_FLASH_STATUS reports FLASH_STATUS_BASIC and REG_FLASH_SIZE is correct
Manufacturer | Chip | Size(MBit) | Detect | Erase | Read/Write | CMD_FLASHFAST |
Adesto | AT25SF641-SUB-T | 128 | Yes | Yes | Yes | 0xE004 - device/blob mismatch |
Adesto | AT25QF128A-SHB-T | 128 | Yes | Yes | Yes | 0xE004 - device/blob mismatch |
Cypress | S25FL064LABMFI010 | 64 | Yes | Yes | Yes | Yes |
Cypress | S25FL128LAGMFV010 | 128 | Yes | Yes | Yes | Yes |
GigaDevice | GD25Q64CSIGR | 64 | Yes | Yes | Yes | Yes |
GigaDevice | GD25Q127CSIGR | 128 | Yes | Yes | Yes | Yes |
ISSI | IS25LP064A-JBLE | 64 | Yes | Yes | Yes | Yes |
ISSI | IS25LP128F-JBLE | 128 | Yes | Yes | Yes | Yes |
Macronix | MX25L6433FM2I-08G | 64 | Yes | Yes | Yes | Yes |
Macronix | MX25L12833FM2I-10G | 128 | Yes | Yes | Yes | Yes |
Macronix | MX25L12872FM2I-10G | 128 | Yes | Yes | Yes | Yes |
Macronix | MX25L25645FM2I-08G | 256 | Yes | Yes | Yes | Yes |
Macronix | MX25L25673FM2I-08G | 256 | Yes | Yes | Yes | Yes |
Microchip | SST26VF064BA-104I/SM | 64 | Yes | Yes | Write fails | - |
Micron | MT25QL128ABA1ESE-0SIT | 128 | Yes | Yes | Yes | Yes |
Micron | MT25QL01GBBB | 1024 | Yes | Yes | Yes | Yes |
Winbond | W25Q32JVSSIQ | 32 | Yes | Yes | Yes | Yes |
Winbond | W25Q64JVSSIQ | 64 | Yes | Yes | Yes | Yes |
Winbond | W25Q128JVSIQ | 128 | Yes | Yes | Yes | Yes |
Winbond | W25Q256JV | 256 | Yes | Yes | Yes | Yes |
Please report any other FLASH Chip that you found to be working or not and I will add it to the table.
Okay, for science! :)
I just replaced the W25Q32JVSSIQ-ND on an EVE3-43G with a socket for 208 mil SOIC chips.
The first chip I have in there is the W25Q128JVSIQ-ND and it is doing just fine.
If you have other chips to test, I can try to get them and test them as well.
Push.
Although I am all alone with this so far I added four more chips to the list that I just tested.
This time the first 256 MBit chips, Macronix seem to be the only company that manufacures these
in a SOIC 208 mil package.
Oh, I also sorted the list and added the manufacturer as it was grown a bit now.
Most of the chips I have tested in the socket now, so I can retest these.
The test now is to connect a USB/SPI interface to my EVE3-43G and use progflash.exe from the command line:
progflash module MPSSE write flash.bin
progflash module MPSSE detect except_writing
progflash module MPSSE erase
Currently I am using the progflash.exe that was part of the EVE Asset Builder 1.6.0 RC1 archive, it identifies itself as "BT81X Flash Programming Utility v1.1".
Can confirm that AT25SF641F (64MBit) fails to enter FAST mode with error 0xE004 - device/blob mismatch
MX25L6433F (64MBit) works correctly.
AT25SF appears to have non standard commands and registers for enabling quad mode and reading status registers.
Furthermore, I'm finding that the AT25 will randomly fail to be detected on startup, and continue to do so until some unknown sequence of events resets it.
When in this failed-to-detect state, my attempts to read flash registers using cmd_flash_spitx/_spirx return 0xFF
I can't make much progress with debugging this as it starts working when I need it not to, and I don't know how to make it fail again.
Manufacturer | Chip | Size(MBit) | Detect | Erase | Read/Write | CMD_FLASHFAST |
Winbond | W25Q256JV | 256 | Yes | Yes | Yes | Yes |
Hi Rudolph, you can add MT25QL01GBBB to the list (specifically tested MT25QL01GBBB1EW9-0SIT):
Manufacturer Chip Size(MBit) Detect Erase Read/Write CMD_FLASHFAST
Micron MT25QL01GBBB 1024 Yes Yes Yes Yes
I added the MT25QL01GBBB, thank you, but I wonder if it actually works properly since I saw in the datasheet that it uses two 512MBit dies in the package.
But now I wonder what Flash exactly is Riverdi using on their EVE4 modules.
My RVT70HSBNWC00-B is populated with a chip from Micron, the markings are "MT" "IGA17", "RW213" and "FBNP".
My RVT50HSBNWC00-B is populated with a chip from Micron, the markings are "MT" "IIA17", "RW213" and "JBRQ".
My RVT101HSBNWC00-B is populated with a chip from Micron, the markings are "MT" "0DA17", "RW256" and "J2F1".
All do have BGA cases.
A quick search for Micron 512MBit flash chips did not tell me what devices exactly these are.