about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAgeFilesLines
* load debug info into code diskKartik K. Agaram2021-05-143-4/+12
|
* insert a compile phase to emit some debug infoKartik K. Agaram2021-05-148-2317/+2918
|
* .Kartik K. Agaram2021-05-142-53/+4
|
* .Kartik K. Agaram2021-05-142-0/+1
| | | | | Fix a stack bug in survey_baremetal. I'm not sure how my tests weren't crashing, but I won't bother digging further.
* .Kartik K. Agaram2021-05-142-5/+4
|
* .Kartik K. Agaram2021-05-141-3/+1
|
* .Kartik Agaram2021-05-091-26/+32
| | | | | Keep the html version of mandelbrot-fixed at the correct monochrome still image for now. (commit 4bfc80ce)
* mandelbrot: add some garish colorKartik K. Agaram2021-05-092-19/+12
|
* similarly zoom into fixed and fp mandelbrotKartik K. Agaram2021-05-092-27/+70
|
* mandelbrot-fixed: introduce some viewport parametersKartik K. Agaram2021-05-091-17/+42
|
* bugfix in mandelbrot-fixedKartik K. Agaram2021-05-091-13/+19
| | | | | | | | Thanks to comparing regular ints with fixed-point ints, I was computing iterations for many screenfuls of pixels to the right of the current one. (Y-axis behaved similarly, but there it was just doing other work instead of busy-looping forever. If you don't have anything else to do, it doesn't matter what you do.)
* .Kartik K. Agaram2021-05-092-31/+31
| | | | | Yet another step in the slow divergence of survey_baremetal from its survey_elf roots.
* .Kartik Agaram2021-05-091-7/+43
|
* .Kartik Agaram2021-05-092-6/+6
|
* .Kartik Agaram2021-05-094-4/+4
|
* .Kartik Agaram2021-05-094-352/+372
|
* .Kartik Agaram2021-05-092-34/+42
|
* mandelbrot: streamline expositionKartik Agaram2021-05-094-283/+254
|
* .Kartik Agaram2021-05-092-195/+197
|
* .Kartik Agaram2021-05-082-0/+499
|
* mandelbrot set in fixed-pointKartik K. Agaram2021-05-082-15/+239
|
* .Kartik K. Agaram2021-05-081-1/+1
|
* support checking overflow flag everywhereKartik K. Agaram2021-05-088-9/+517
|
* .Kartik K. Agaram2021-05-081-1/+1
|
* initial, monochrome mandelbrot setKartik K. Agaram2021-05-081-0/+149
|
* A 'bowboard', a chessboard of rainbow circlesKartik K. Agaram2021-05-081-23/+31
| | | | Compare with (chessboard screen 256)
* a full-circle rainbowKartik K. Agaram2021-05-082-3/+30
|
* new shell macro: doKartik K. Agaram2021-05-071-2/+3
|
* always check for null in 'index' instructionsKartik K. Agaram2021-05-076-95/+132
|
* all apps working again after null get checkKartik K. Agaram2021-05-072-2/+34
|
* .Kartik K. Agaram2021-05-072-39/+39
|
* always check for null in 'get' instructionsKartik K. Agaram2021-05-075-14/+51
|
* a little bit more information when lookup failsKartik K. Agaram2021-05-071-1/+31
| | | | | | Basically this should never, ever happen until I start reclaiming heap memory. I believe the only reason it happens is unprotected writes to address 0 or thereabouts.
* an often-overflowing stream when running brcircleKartik K. Agaram2021-05-072-1/+3
| | | | We really need to systematically check our trace streams.
* .Kartik K. Agaram2021-05-072-6/+8
|
* some boot-time heartbeat messagesKartik K. Agaram2021-05-072-1/+20
| | | | This will help us with some common debug scenarios.
* .Kartik K. Agaram2021-05-071-19/+19
|
* clean up all definitionsKartik K. Agaram2021-05-071-21/+20
|
* clean up chessboardKartik K. Agaram2021-05-072-11/+15
| | | | | We still benefit from some helpers here because of the unrolling and multiple calls to helpers.
* clean up read_lineKartik K. Agaram2021-05-071-9/+6
|
* clean up Bresenham line-drawingKartik K. Agaram2021-05-072-26/+54
|
* .Kartik K. Agaram2021-05-071-6/+6
|
* no, we need hline1 for fill_rectKartik K. Agaram2021-05-071-8/+10
|
* clean up hline and vlineKartik K. Agaram2021-05-071-10/+10
|
* starting to implement first macrosKartik K. Agaram2021-05-074-0/+20
| | | | | | | | | | | | | | | Another commit, another bugfix. Some snippets from my currently exploding todo list: - always investigate lookup errors immediately. Beyond the root cause, they should never happen at the moment, while we aren't reclaiming memory. we should always return a more precise error message. Usually involving null pointer checks. - on abort, print out stack trace - emit mapping of labels to addresses during survey - store a mapping of symbols somewhere in the code image - stop allocating 1KB per token; expand space for tokens as needed
* give up on nested backquotes for nowKartik K. Agaram2021-05-072-7/+82
|
* macros almost doneKartik K. Agaram2021-05-061-23/+69
| | | | Just one case left: macros within unquote.
* some more casesKartik K. Agaram2021-05-061-11/+60
|
* first passing test for macroexpandKartik K. Agaram2021-05-066-41/+249
| | | | | | | | In the process I spent a long time tracking down a stray TODO in 108write.subx that I thought would abort but didn't since the switch to baremetal. Then after I reintroduced that assertion I had to go track down a bunch of buffer sizes. Stream sizes continue to be a huge mess.
* .Kartik K. Agaram2021-05-061-1/+1
|
generated by cgit-pink 1.4.1-2-gfad0 (git 2.36.2.497.gbbea4dcf42) at 2024-11-15 21:25:09 +0000