<oriansj>stikonas: yeah, I'll move the files from mescc-tools to stage0-posix <stikonas>I was thinking of bootstrap-seeds/POSIX/PV64/hex0_RV64.S <oriansj>stikonas: I thought you were referencing the M0 example in mescc-tools <stikonas>isn't M0 in mesc-tools actual source file for cc_x86 <oriansj>well C version of hex2 really has grown to deal with a great deal of architecture specific crap <oriansj>and 208 lines of dealing with just the arguments <oriansj>stikonas[m]: I have a very rough draft of word based support in hex2 up in the latest commit. <oriansj>The logic is entirely in hex2_word.c <oriansj>I've only confirmed the essentials but the jump and branch immdiate block bits might be wrong. Could you please do a sanity check. <siraben>oriansj: lol I love that forth.c file <siraben>##forth would have a field day with that <siraben>but truly and honestly, Forth is not a bootstrapping language <siraben>the language prevents you from building up abstractions well <oriansj>The big question is are U format instruction relative displacement or absolute addresses <oriansj>stikonas: good, then as it is now, should function for your needs <stikonas>although, before I start writing hex1, I might do kaem too, we'll see <oriansj>I'll start on matching work for M1 as otherwise all immediates will be of the jump and load relative type. ***jackhill is now known as KM4MBG