summary refs log tree commit diff
path: root/lisc
AgeCommit message (Collapse)Author
2016-02-25add some bitset functionsQuentin Carbonneaux
2016-02-25remove FP notesQuentin Carbonneaux
2016-02-24create new doc directoryQuentin Carbonneaux
2016-02-24fix tight assertion in MEM()Quentin Carbonneaux
2016-02-24tentative big args supportQuentin Carbonneaux
2016-02-24use default use/def counts in newtmp()Quentin Carbonneaux
This is paliative, before I check that all use counts in isel.c are correct.
2016-02-24do not print null offsetsQuentin Carbonneaux
2016-02-24fix swapped operands on emitcopy() callQuentin Carbonneaux
2016-02-24prepare for big structs passing codeQuentin Carbonneaux
2016-02-24support memory class arguemntsQuentin Carbonneaux
2016-02-24oh oh, subs are not folded!Quentin Carbonneaux
2016-02-24simply use memset to 0 stuffQuentin Carbonneaux
2016-02-24oops, phi nodes rewrite for fast locals was trashedQuentin Carbonneaux
The phi fixing mechanism can use emit(), so we need to set curi before performing the rewrite. Otherwise, we are writing at random places in the instruction buffer (not so bad because it is bounds checked), but then we loose the instructions written (bad)!
2016-02-23add fun example, does not compile yetQuentin Carbonneaux
2016-02-23cosmetics in isel.cQuentin Carbonneaux
2016-02-23print new jump instructionsQuentin Carbonneaux
2016-02-23fix uninitialized variable in selpar()Quentin Carbonneaux
2016-02-23patch return, might not workQuentin Carbonneaux
2016-02-22fix buggy name changes in iselQuentin Carbonneaux
2016-02-22simplify emit tableQuentin Carbonneaux
2016-02-22cosmeticsQuentin Carbonneaux
2016-02-22use isstore() in parserQuentin Carbonneaux
2016-02-21do not spill dead phisQuentin Carbonneaux
Regalloc will be able to handle these spurious phis, however, some other spurious dead instructions can be emitted. It would be better to get rid of them upfront; maybe by modifying isel, or by inserting a proper dce pass. An example of that undesirable behavior is exposed below. @l1 %foo =w ... @l2 %dead =w phi @l1 %foo, ...
2016-02-21complete fp support for small structsQuentin Carbonneaux
There is an oddity/bug though, we use OStorel to store possibly fp registers. Gas does not complain, but this is wrong. The fix is probably to have a simple OStore, like in the OLoad case.
2016-02-21genius or idiot?Quentin Carbonneaux
2016-02-19add lexing sugar for backward compatibilityQuentin Carbonneaux
2016-02-18stop using OXxx1 and use new OLoadQuentin Carbonneaux
2016-02-18use classes in arg classificationQuentin Carbonneaux
2016-02-18complete argcls (pretty ugly...)Quentin Carbonneaux
2016-02-18start completing the fp abiQuentin Carbonneaux
2016-02-16fix dumb bug found by Andrew ChambersQuentin Carbonneaux
2016-02-15better variable name in selcall()Quentin Carbonneaux
2016-02-15always print RACall with 3 digitsQuentin Carbonneaux
2016-02-15use arguments in mandelbrot testQuentin Carbonneaux
2016-02-15more fp calling conventionsQuentin Carbonneaux
2016-02-15be future proof in usage message :)Quentin Carbonneaux
2016-02-15fix comments in emitQuentin Carbonneaux
2016-02-15uniformize notations in mandelbrot testQuentin Carbonneaux
2016-02-15scale up the mandelbrot testQuentin Carbonneaux
2016-02-15quickly patch emit (fp load and store)Quentin Carbonneaux
2016-02-15initialize class correctly in pmgen()Quentin Carbonneaux
2016-02-15patch isel for store{s,d}Quentin Carbonneaux
2016-02-15completely hide xmm15Quentin Carbonneaux
2016-02-15more stressful test for fpQuentin Carbonneaux
2016-02-15partial fix argcls() in iselQuentin Carbonneaux
2016-02-15collect and emit fp constantsQuentin Carbonneaux
2016-02-12test int syntax for one fp constantQuentin Carbonneaux
2016-02-12cut code, use scanf a little moreQuentin Carbonneaux
2016-02-12new syntax for float literalsQuentin Carbonneaux
2016-02-12scan ints using scanfQuentin Carbonneaux
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-30 09:04:02 +0000