On 4/15/2019 10:32 AM, Mia Magnusson wrote: > Den Sun, 14 Apr 2019 21:43:28 -0500 skrev Jim Brain <brain_at_jbrain.com>: >> Thanks Michał for the testing. >> >> I have updated the github repository with the v4 code here: > Cool, nice work! > >> * I have also added support for the 65C816S in native mode, in which >> the bank register is driven directly by the 'C816. Placing the CPU >> into native mode will also switch into this mode, which means care >> must be practiced to set up the '816 bank to match the current >> 6509 'Execution Bank' before switching into this mode. Location $0 >> and $1 will still be visible while in native mode in the current HDL. > Something that Michal and you might want conteplate: Would there be any > benefit for the new 8088 support software if it were possible to hide > the $0/$1 registers in '816 native mode? I apologize. I misread my code last night, and the behavior you note is indeed in place: always _at_(*) begin if(oe_bank & !flag_ext & !flag_full) // accessing bank register lower 4 bits and we're not in '816 mode data_6502_out = {4'b0000, data_bank[3:0]}; else if(oe_bank & !flag_ext & flag_full) // accessing bank register 8 bits and we're not in '816 mode data_6502_out = data_bank; else if(r_w & _rdy & phi2_6509) // read data from bus and ready is high and phi2 is high data_6502_out = data_6509; else // there is no option to read bank register in 816 mode. data_6502_out = 8'bz; end However, I did not hide the writes to the $0/$1 registers. I assume writing to those registers should also be prevented, but that means that one must remember what bank one was in before one went into '816 mode. -- Jim Brain brain_at_jbrain.com www.jbrain.comReceived on 2020-05-29 21:34:35
Archive generated by hypermail 2.3.0.