Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.compilers > #2658
| From | Fernando <pronesto@gmail.com> |
|---|---|
| Newsgroups | comp.compilers |
| Subject | Re: 8086 register allocation |
| Date | 2021-05-10 04:46 -0700 |
| Organization | Compilers Central |
| Message-ID | <21-05-006@comp.compilers> (permalink) |
| References | <21-05-005@comp.compilers> |
Hi Alex, I read part of your code. Very nice! The core algorithm (disregarding the x86 specific part) seems like 'local register allocation' using Belady's heuristic for spilling. I have a class about it here: https://youtu.be/XmNXeNCGSIA There has been some academic work about models for register allocation for x86. Two papers follow below: * Register Allocation by Puzzle Solving, https://llvm.org/pubs/2008-06-PLDI-PuzzleSolving.pdf * A generalized algorithm for graph-coloring register allocation, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.10.3076&rep=rep1&type=pdf Regards, Fernando > For the fun of it I've implemented a greedy bottom-up local register > allocator for the intel 8086 CPU, which is known for its non-uniform > use of registers in ALU instructions and memory operands > (https://github.com/alexfru/regal86).
Back to comp.compilers | Previous | Next — Previous in thread | Next in thread | Find similar
8086 register allocation "Alexei A. Frounze" <alexfrunews@gmail.com> - 2021-05-09 14:28 -0700
Re: 8086 register allocation Fernando <pronesto@gmail.com> - 2021-05-10 04:46 -0700
Re: 8086 register allocation gah4 <gah4@u.washington.edu> - 2021-05-10 14:49 -0700
Re: 8086 register allocation Hans-Peter Diettrich <DrDiettrich1@netscape.net> - 2021-05-11 03:35 +0200
Re: 8086 register allocation gah4 <gah4@u.washington.edu> - 2021-05-10 21:19 -0700
Re: 8086 register allocation Hans-Peter Diettrich <DrDiettrich1@netscape.net> - 2021-05-11 09:35 +0200
Re: 8086 register allocation Thomas Koenig <tkoenig@netcologne.de> - 2021-05-11 06:44 +0000
csiph-web