Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Fix assembly | 2017-09-12 | |
| | |||
* | WIP on reg alloc for asm output | 2017-09-11 | |
| | |||
* | Fix (hopefully) overlap in reg | 2017-09-05 | |
| | |||
* | Fix schedule to be in correct order | 2017-09-05 | |
| | |||
* | WIP | 2017-09-05 | |
| | |||
* | Fix schedule | 2017-09-05 | |
| | |||
* | WIP on reg alloc | 2017-09-04 | |
| | |||
* | chmod +x | 2017-09-04 | |
| | |||
* | Add display files (temporary, kind-of) | 2017-09-04 | |
| | |||
* | WIP Update compile with registers | 2017-09-04 | |
| | |||
* | More WIP on register allocation | 2017-08-17 | |
| | | | | | | | | | | The current allocation is terrible, probably because we are currently requiring that all instructions output to registers. My current guess at a decent thing to do is to make a pass, after register allocation, and eliminate all registers that simply get stored to memory, replacing the relevant instructions with the memory-using versions. Then we can re-register allocate, ignoring values that go straight to memory. | ||
* | Handle most of register allocation | 2017-08-14 | |
| | |||
* | Handle equality in parsing | 2017-08-14 | |
| | |||
* | Fixup header and footer | 2017-08-14 | |
| | |||
* | Update scheduler to know about implicit mulx arg | 2017-08-14 | |
| | | | | | Now it prefers putting together mulx with the same implicit arg (approximated as the same variable with the lower number). | ||
* | Use a more realistic processor model | 2017-08-13 | |
| | |||
* | Revert "Revert "Subset compiler differently"" | 2017-08-13 | |
| | | | | This reverts commit c8c82f2edf416b667ba487dfa2cff0795d37cbe6. | ||
* | Revert "Subset compiler differently" | 2017-08-13 | |
| | | | | This reverts commit 7038444e6515ec51a3b6d1cdfe972664d3f16c81. | ||
* | Subset compiler differently | 2017-08-13 | |
| | |||
* | Add heuristic search | 2017-08-13 | |
| | |||
* | Update the graph maker | 2017-08-13 | |
| | |||
* | Update exhaustive search compiler | 2017-08-13 | |
| | |||
* | Add memoize.py to zinc compiler folder | 2017-08-11 | |
| | |||
* | wip on register allocation in python | 2017-08-09 | |
| | |||
* | Fix the sense of dependencies in zinc generation | 2017-08-09 | |
| | | | | | This brings the maximum window size back down to 53, so we use the old method of making dependencies, rather than the new one. | ||
* | Get all scheduling done in one frame | 2017-08-09 | |
| | |||
* | Fix wip | 2017-08-09 | |
| | |||
* | Larger instruction window for zinc | 2017-08-09 | |
| | |||
* | Add a faster version of the zinc compiler | 2017-08-06 | |
| | |||
* | Add initial stab at C-compilation by optimization | 2017-08-06 | |