From: Mario Nunes Date: Mon, 31 Aug 2015 01:35:28 +0000 (-0300) Subject: Cosmetic changes X-Git-Url: https://git.ndcode.org/public/gitweb.cgi?a=commitdiff_plain;h=0bdac0b3afb97dfea16b94b6b4c0b1710041aa26;p=FUZIX.git Cosmetic changes --- diff --git a/Kernel/platform-msx1/README.md b/Kernel/platform-msx1/README.md index 0f5233fe..5ccee0b0 100644 --- a/Kernel/platform-msx1/README.md +++ b/Kernel/platform-msx1/README.md @@ -1,4 +1,5 @@ First wild guess at an MSX1 target +--- And it is exactly that a guess 8) @@ -13,20 +14,23 @@ may be able to bank the OS kernel in a cartridge. Q: should we work to the following map - +``` 0x0000-0x7FFF User space / switchable with kernel (in theory if kernel code is cartridge could we even use RAM + megaram?) 0x8000-0xBFFF Kernel (end of code, data, highcode, common) 0xC000-0xFFFF +``` or do we bite the bullet about relocatable binaries (which we need to do for the sinclair at least) and do +``` 0x0000-0x00FF Vectors 0x0100-???? 0x4000-0xBFFF User space/switchable with kernel rest 0xC000-0xFFFF +``` which would allow the use of more typs of megaram without LDIR tricks. @@ -44,4 +48,4 @@ Other mappers that might be found include - MSX2 mapper (in which case we should probably make the msx2 platform detect the early type VDP and use that ?) - - Zemina (requires LDIR copying stuff) \ No newline at end of file + - Zemina (requires LDIR copying stuff)