Easy Flasher Custom Operation: ad-hoc read/write test results

Easy Flasher custom operation dump thread for success/fails - updated as stuff get tried


read of 0x0 length 0x11000 of a flashed firmware compared to original full dump extract


I attempted to transplant the BL from another BK7231n dump so chose the whole 2mb file but only selected 0x0 - 0x11000 to flash, but it burnt the whole file. I guess chosen file needs to be exactly the size to be flashed.

after setting a custom write operation going back to standard "do firmware write (no backup!)" will perform a small write to 0x00 instead of the full QIO as expected

---

Added after 9 [minutes]:

to confirm custom write offset selection I erased the whole flash of a BK7231N using BKFIL then flashed 4kb file TLV_calib_moes.bin to 0x1E3000


then took a whole flash backup to check location and content. looks good

and with 0x0 up to 0x1E3000 removed from the full dump:



Added after 28 [minutes]:

---

not directly related but I've proposed the erase all info text is updated to say it doesn't include BL
https://github.com/openshwprojects/BK7231GUIFlashTool/pull/52

It's not immediately obvious to users that BK7252 mode in EF is currently for 2mb BK7252U and will skip bootloader before they begin backup.



to get 4mb, still with no BL, a custom read operation can be performed




ideally there'd be some wrap-around (?) check thing to dump full 2 or 4mb, rename of BK7252 to BK7252U and then the addition of BK7252N, which includes ROM like BK7238 and BK7231N.

minor note. With no firmware selected writing wrong firmware can occur. only tested with BK7252