SCM Repository
Annotation of /branches/vis15/src/compiler/WISHLIST.md
Parent Directory
| 
Revision Log
Revision 4384 - (view) (download)
| 1 : | jhr | 4381 | ## Current status as of August 8, 2016 |
| 2 : | jhr | 4103 | |
| 3 : | ### Major bugs | ||
| 4 : | |||
| 5 : | * https://github.com/Diderot-Language/examples dvr/dvr.diderot (Feb 14 email from glk subject | ||
| 6 : | "memory errors, possibly cause of crashing"): memory errors cause crashing. Error messages | ||
| 7 : | suggest incorrect use of free(). | ||
| 8 : | jhr | 4223 | - fixed; the crashing bugs are gone. There is a small amount of memory leaking, but it |
| 9 : | appears to be leaking from teem (hest). | ||
| 10 : | jhr | 4103 | * bug054.diderot: return from inside a conditional in a function causes uncaught exception |
| 11 : | jhr | 4175 | - fixed |
| 12 : | jhr | 4103 | * bug047.diderot: don't know what's wrong; this is a program that used to work in the vis12 branch. |
| 13 : | - fixed | ||
| 14 : | * bug052.diderot: inside() doesn't work on 1D fields | ||
| 15 : | jhr | 4229 | - fixed |
| 16 : | jhr | 4103 | * bug053.diderot: differentiation doesn't work on 1D fields |
| 17 : | cchiw | 4227 | - fixed |
| 18 : | jhr | 4103 | * bug043.diderot: generated broken C code (was: no output ints?) |
| 19 : | - fixed | ||
| 20 : | * bug045.diderot: missing determinant on tensor[2,2] and tensor[3,3] | ||
| 21 : | - fixed | ||
| 22 : | * bug050.diderot: tensor[3,3] output not working | ||
| 23 : | - fixed | ||
| 24 : | * bug055.diderot: conditional field assignment causes an uncaught exception | ||
| 25 : | - fixed (typechecker rejects program) | ||
| 26 : | |||
| 27 : | ### Minor bugs | ||
| 28 : | |||
| 29 : | * bug029.diderot missing code generation for printing matrices and higher-order tensors. | ||
| 30 : | - fixed | ||
| 31 : | * bug051.diderot: 10x increase in compile time changing from bspln3 to c4hexic (can use --disable-mid-check) | ||
| 32 : | - vis15 compiles the c4hexic version in 1.16 seconds without checking and in 29s with --check-all | ||
| 33 : | So it is faster than vis12, but still slow when internal checking is enabled. | ||
| 34 : | jhr | 4229 | * bug028.diderot: missing forms of matrix multiplication; Charisee's work will address this issue |
| 35 : | jhr | 4103 | - fixed |
| 36 : | jhr | 4381 | * bug048.diderot: can't use one int and one real in a conditional expression |
| 37 : | jhr | 4103 | - fixed |
| 38 : | jhr | 4381 | * bug049.diderot: bogus types parsed as strand names create uncaught exceptions |
| 39 : | jhr | 4103 | - fixed, but the error message could be improved |
| 40 : | * bug007.diderot: no codegen support for tensor slices; Charisee's work will address this issue | ||
| 41 : | - fixed | ||
| 42 : | * bug038.diderot: (re-opening an older bug) run-time should check on existence of field orientation meta-data before use | ||
| 43 : | - fixed | ||
| 44 : | * bug044.diderot: either implement or disallow "output bool" | ||
| 45 : | - fixed | ||
| 46 : | * bug040.diderot: "The avoidance of redundant Inside tests is not working." | ||
| 47 : | - still open | ||
| 48 : | |||
| 49 : | ### Wish list (from high to low priority) | ||
| 50 : | |||
| 51 : | jhr | 4384 | * add `#version` directive. `#version 1.0` will be old syntax (some merge of charisee and lamont's |
| 52 : | branches); `#version 2.0` will be the new syntax. No directive will be interpreted as the | ||
| 53 : | same as `#version 1.0`. Include a `--warn-deprecated` compiler flag to generate | ||
| 54 : | warnings about deprecated syntax. | ||
| 55 : | |||
| 56 : | jhr | 4120 | * add clerp, with the following semantics: |
| 57 : | jhr | 4162 | ```` |
| 58 : | clerp(A,B,U,x,V) == clamp(A,B,lerp(A,B,U,x,V)) | ||
| 59 : | clerp(A,B,x) == clamp(A,B,lerp(A,B,x)) | ||
| 60 : | ```` | ||
| 61 : | jhr | 4120 | - still open |
| 62 : | jhr | 4103 | * allow a "continue" statement within the update method to go to the beginning of |
| 63 : | the next update iteration | ||
| 64 : | - fixed | ||
| 65 : | * print(): either implement for --target=pthread, or have compiler warn that it won't work | ||
| 66 : | - still open; pthread support is not yet complete in vis15 | ||
| 67 : | * print(): should be possible during strand initialization. Otherwise have to create an iteration | ||
| 68 : | counter initialized to zero, and print from inside update() only on first iteration | ||
| 69 : | - fixed | ||
| 70 : | * remove restriction on .diderot filename extension for Diderot program (or at least allow .ddro) | ||
| 71 : | jhr | 4124 | - fixed |
| 72 : | jhr | 4103 | * syntax errors from the compiler should just say "syntax error". The "try" suggestions are |
| 73 : | almost always confusing, sometimes laughable (e.g. printf() not a valid function; should be | ||
| 74 : | print() but compiler suggests to try inserting "%=". nope!). For now, all suggestions should | ||
| 75 : | be removed, because on average they create more confusion. More carefully crafted suggestions | ||
| 76 : | can be added on a case by case basis. | ||
| 77 : | jhr | 4381 | - fixed; we now just print "syntax error". A better fix will require improvements to ml-antlr. |
| 78 : | jhr | 4120 | * with "if (true) { foo } else { bar }", prune the dead code (in simplify?) |
| 79 : | jhr | 4124 | - fixed |
| 80 : | jhr | 4103 | * using "kern#N" syntax to assert C^N continuity |
| 81 : | - fixed | ||
| 82 : | * odd-support kernels | ||
| 83 : | - still open | ||
| 84 : | jhr | 4124 |
| root@smlnj-gforge.cs.uchicago.edu | ViewVC Help |
| Powered by ViewVC 1.0.0 |
