Closed wwarthen closed 3 years ago
The issue with the Mark IV is not related to FFS. There is some other problem when the slack space in HBIOS gets small. I will need to try and track that down.
So, the only real issue (kind of) seems to be the issue with CPU speed being high.
-Wayne
Confirmed the Mark IV issue was actually an HBIOS heap overflow error that was not well handled.
The 36 MHz issue seems to be a very slight issue with the write sequence in the FFS that was causing the flash write errors on high speed CPUs.
Fixes for both of these were just checked in.
Phil, would you take a look to see if the FFS change makes sense to you?
Thanks,
Wayne
will do. also testing an optional verify after write option at the moment.
On Wed, 23 Dec 2020, 08:46 Wayne Warthen, notifications@github.com wrote:
Confirmed the Mark IV issue was actually an HBIOS heap overflow error that was not well handled.
The 36 MHz issue seems to be a very slight issue with the write sequence in the FFS that was causing the flash write errors on high speed CPUs.
Fixes for both of these were just checked in.
Phil, would you take a look to see if the FFS change makes sense to you?
Thanks,
Wayne
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/wwarthen/RomWBW/issues/166#issuecomment-749854497, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJM7HV2GUJAOJ5RNSNZP4RDSWE4VTANCNFSM4VGAE7ZA .
These issues appear to be resolved, so I am closing this issue thread.
@b1ackmai1er
Going to open an issue to track testing of the Flash File System.
I have tested the latest Flash File System code on a few systems. Generally checking out OK. I have encountered a couple failures.
First, on a Z180 based system running at 36 MHz, the ROM was corrupted (log below). No issues at normal 18 MHz. While the ROM is reading fine at 36 MHz, it seems to fail to write at 36 MHz.
Second failure was on my Mark IV system (running at 18 MHz). It fails to complete the boot into ZSDOS (or CP/M 2.2). However, CP/M 3 seems to work fine.