 ed
|
cf78cfa120
|
Rename _shift enums to _pos (may change to offset, but in either case not as accurate to call them _shift, and _shift_smount is longer).
|
2026-08-19 13:17:08 -04:00 |
|
|
ed
|
b345ccd60e
|
don't bother wiht icache flush atom.
|
2026-08-18 16:10:44 -04:00 |
|
|
ed
|
bbda5efaea
|
build_noramlized_v3s4 mostly reviewed.
|
2026-08-18 16:10:29 -04:00 |
|
|
ed
|
290bb0e07a
|
WIP: Adjusting normalize atom proc to to the new tape runtime convention. Trying to reduce redundant scuff to scratch.
|
2026-08-18 13:22:40 -04:00 |
|
|
ed
|
86fe189b4e
|
Moving definitions to use dedicated scratch register.
|
2026-08-18 09:26:49 -04:00 |
|
|
ed
|
da007d342e
|
Reviewing. Successfuly reworked register allocation for tape runs.
|
2026-08-18 00:26:36 -04:00 |
|
|
ed
|
5a4bfb1224
|
Collapse of atom 6-9 into a single atom (finaly). Generalized cross product atom proc and atom component. Still working on normalize_v3s4.
|
2026-08-17 18:18:18 -04:00 |
|
|
ed
|
d4795cf9de
|
Extract out a cross roduct component.
|
2026-08-17 10:34:43 -04:00 |
|
|
ed
|
e79c364b40
|
reducing cross product atom procs to a single one in gte for once.
|
2026-08-17 01:05:58 -04:00 |
|
|
ed
|
18b1d5a04b
|
remove outdated comments.
|
2026-08-16 12:07:09 -04:00 |
|
|
ed
|
3faccfc283
|
more reviewing, thinking about atom bundles...
|
2026-08-16 01:22:53 -04:00 |
|
|
ed
|
3301826f5c
|
reviewed: resolve_look_at__populate_proc
|
2026-08-15 21:34:55 -04:00 |
|
|
ed
|
d9b9241e2c
|
resolve_look_at__cross_uz_ux_to_up_proc reviewed
|
2026-08-15 19:53:08 -04:00 |
|
|
ed
|
8a825a59c7
|
Add RegUse_ support to the lua metaprogram. Ideated further on type mapping atom comonents to their base component op (math distinctions annotated in the asm).
|
2026-08-15 15:51:06 -04:00 |
|
|
ed
|
ffc66052f8
|
Curating duffle, preparing to update metaprogram for latest atom asm ideation. Reviewing the resolve_look_at atoms further...
|
2026-08-15 11:21:28 -04:00 |
|
|
ed
|
67a84d34f3
|
oops: endregion
|
2026-08-14 13:41:45 -04:00 |
|
|
ed
|
baaff12f33
|
Ideating on "RegUse_" patterned structs for describe register allocatins to mips atom proc.
|
2026-08-14 12:38:00 -04:00 |
|
|
ed
|
b695056b9a
|
finished reviewing normalize_v3s4 for now
|
2026-08-14 03:45:34 -04:00 |
|
|
ed
|
a535d381ed
|
remove encoding masks from gp (unnecessary, hides errors)
|
2026-08-14 01:22:36 -04:00 |
|
|
ed
|
c447bfa877
|
fixes to the reg file allocator, exploring...
|
2026-08-14 00:43:19 -04:00 |
|
|
ed
|
d88e0d0487
|
remove mask from mips and gte instruction encoders. missing math changes.
|
2026-08-13 23:39:35 -04:00 |
|
|
ed
|
9a6eca6047
|
more review, made a register file allocator (drafted, kinda iffy, want todo comp-time as well).
|
2026-08-13 23:39:03 -04:00 |
|
|
ed
|
5c9c61720f
|
Redesign: Not making local var in MipsAtom_Proc_ or MipsAtomComp_Proc_ have sym tied to proc name. Adjusted parser as well base do that.
|
2026-08-13 21:42:06 -04:00 |
|
|
ed
|
b8e31123e4
|
editing/reading.
|
2026-08-13 21:22:29 -04:00 |
|
|
ed
|
ea3e30a11e
|
oops
|
2026-08-13 20:51:45 -04:00 |
|
|
ed
|
37f4712237
|
gutting nosiy comments. Looking into some atom components..
|
2026-08-13 19:55:17 -04:00 |
|
|
ed
|
b699b47b28
|
intiial review on: resolve_look_at__input_and_sub_proc
|
2026-08-13 19:42:43 -04:00 |
|
|
ed
|
640dab7e61
|
wip: starting to review and update lua metaprogram with more modeling of gte.
|
2026-08-13 18:51:09 -04:00 |
|
|
ed
|
4688566767
|
FINALLY?
|
2026-08-13 17:42:55 -04:00 |
|
|
ed
|
5ebaa6e083
|
still failing
|
2026-08-13 13:18:27 -04:00 |
|
|
ed
|
7f0bdefbcb
|
checkpoint nothing
|
2026-08-13 02:13:47 -04:00 |
|
|
ed
|
d5f28b83ea
|
minor
|
2026-08-12 22:41:52 -04:00 |
|
|
ed
|
3ea3e8d105
|
sssiiighhhh
|
2026-08-12 20:36:11 -04:00 |
|
|
ed
|
6b60cef2e8
|
sigh
|
2026-08-12 20:30:26 -04:00 |
|
|
ed
|
77f19321cd
|
pain
|
2026-08-12 20:24:13 -04:00 |
|
|
ed
|
9501bbbcc2
|
WIP
|
2026-08-12 20:17:53 -04:00 |
|
|
ed
|
9b6b5535f5
|
wip
|
2026-08-12 20:09:57 -04:00 |
|
|
ed
|
7807047dc0
|
Atoms 2-3 work for resolve look at. Don't need OA_ macro so going to stop using.
|
2026-08-11 21:35:59 -04:00 |
|
|
ed
|
7daeec0ee3
|
checkpoint: atom 0-1 works for resolve look at.
|
2026-08-11 14:05:23 -04:00 |
|
|
ed
|
3f3b691ac0
|
Making a proper distinction between atom arenas and atom builders.
|
2026-08-11 11:25:54 -04:00 |
|
|
ed
|
a2d79d65eb
|
amazing bug
|
2026-08-11 01:25:40 -04:00 |
|
|
ed
|
bebcc6a585
|
wip: going to incremnetally test this.
|
2026-08-11 01:25:09 -04:00 |
|
|
ed
|
ece21ed368
|
mark current crashing path.
|
2026-08-10 23:29:38 -04:00 |
|
|
ed
|
144c605ad8
|
some more review. not working still.
|
2026-08-10 23:04:43 -04:00 |
|
|
ed
|
4afd1af0fd
|
started to review this...
|
2026-08-10 19:53:34 -04:00 |
|
|
ed
|
004a7eff19
|
WIP: not fully reviewed. Adds auto-register allocation + mips atom procs + wip resolve look at atoms + atom bundle...
|
2026-08-10 14:13:02 -04:00 |
|
|
ed
|
e42c75a26a
|
WIP: preparing for major changes to atoms to fullfill needs of resolve_look_at and atom ported normalize_v3s4.
|
2026-08-09 18:49:59 -04:00 |
|
|
ed
|
69f2c0d036
|
Prepping for: resolve_look_at impl.
|
2026-08-08 23:13:18 -04:00 |
|
|
ed
|
b045856dd6
|
converted pad input for cam to mips atom
|
2026-08-08 18:23:28 -04:00 |
|
|
ed
|
68b87f1c8b
|
Completed C-side of: Camera Transformation chapter. Now todo atom tape translation...
|
2026-08-08 16:42:07 -04:00 |
|