

@1106

9 years 
mulligan 
changes necessary to get RTLabs>RTL to compile



@1105

9 years 
campbell 
Show that RTLabs graphs are closed on branch (i.e., all labels in …



@1104

9 years 
campbell 
A little more tidying.



@1103

9 years 
boender 
 reverted to old policy



@1102

9 years 
campbell 
Tidy up branch



@1101

9 years 
campbell 
Label preservation in Cminor initialisation and RTLabs translation
on …



@1100

9 years 
campbell 
Finally show that labels in generated Cminor programs are properly …



@1099

9 years 
ayache 
Bug fix in Deliverables/D2.2/8051: cast simplification.



@1098

9 years 
campbell 
Merge branch with trunk



@1097

9 years 
campbell 
Checkpoint labels work on branch again.



@1096

9 years 
campbell 
Checkpoint part way through adding proper C label checking to the branch.



@1095

9 years 
campbell 
Make add_exprs total (on branch).



@1094

9 years 
mulligan 
some changes from today to do with liveness analyses



@1093

9 years 
campbell 
Merge a bit more.



@1092

9 years 
campbell 
Some minor definitions for identifiers and lists.



@1091

9 years 
campbell 
Merge trunk into idlookupbranch



@1090

9 years 
mulligan 
small change to liveness analysis



@1089

9 years 
mulligan 
more changes from earlier in the week



@1088

9 years 
mulligan 
work on liveness analysis: an imperative nightmare



@1087

9 years 
campbell 
Experimental branch where lookups of local variables in Cminor code …



@1086

9 years 
campbell 
Branch for experimenting with identifier binding guarantees.



@1085

9 years 
mulligan 
removed stray files that are no longer needed



@1084

9 years 
mulligan 
more added on ertl pass: not sure how much should be axiomatised wrt …



@1083

9 years 
mulligan 
ertl > ltl statement generation nearly complete. required some …



@1082

9 years 
mulligan 
work from today on ertl > ltl pass



@1081

9 years 
mulligan 
completed rtlertl pass



@1080

9 years 
mulligan 
more added



@1079

9 years 
mulligan 
finished rtl to ertl pass modulo conversion of tailcall simplification code



@1078

9 years 
campbell 
Implement stack allocation for parameters whose address is taken.



@1077

9 years 
mulligan 
ack, dependent types are scary



@1076

9 years 
mulligan 
small changes



@1075

9 years 
mulligan 
nearly completed rtl > ertl pass removing all option types with dep. types



@1074

9 years 
boender 
 added lookup lemma



@1073

9 years 
mulligan 
more changes from today



@1072

9 years 
campbell 
Use not equals form of showing entry/exit labels.



@1071

9 years 
mulligan 
changes the specific form that the added proofs take to use None, not …



@1070

9 years 
campbell 
Show that entry and exit labels are in the RTLabs graph.



@1069

9 years 
campbell 
Change odd proof obligation problem back.



@1068

9 years 
mulligan 
rtlabs translation complete subject to axioms



@1067

9 years 
mulligan 
more smaller changes



@1066

9 years 
mulligan 
changes from today



@1065

9 years 
campbell 
Note a couple of deviations from the prototype.



@1064

9 years 
mulligan 
changes from today, nearly complete rtlabs translation pass



@1063

9 years 
mulligan 
changes from today



@1062

9 years 
mulligan 
separated jmeq and coercions from foldstuff.ma in order to fix the …



@1061

9 years 
mulligan 
more work, bug found, ridiculous map3 function with dep. types added



@1060

9 years 
mulligan 
work from this morning and yesterday



@1059

9 years 
mulligan 
work from today, bit of a mess at the moment



@1058

9 years 
campbell 
Evict CompCert? Maps interface in favour of BitVectorTries?.



@1057

9 years 
mulligan 
changes from today



@1056

9 years 
campbell 
Switch to delayed identifier error scheme.



@1055

9 years 
mulligan 
changes to how identifiers are generated



@1054

9 years 
boender 
 proven policy safe



@1053

9 years 
mulligan 
changes



@1052

9 years 
mulligan 
removed offsets after reading cerco mailing list



@1051

9 years 
mulligan 
removed superfluous addressing mode code from RTLabs/syntax.ma



@1050

9 years 
mulligan 
adding dependent types to map datastructure to remove all option …



@1049

9 years 
mulligan 
more stuff added



@1048

9 years 
mulligan 
added implementation of haskell associative maps to clean up the mess …



@1047

9 years 
mulligan 
more work from today



@1046

9 years 
mulligan 
syntax of rtlabs was wrong: cast not const. more added to rtlabs > …



@1045

9 years 
mulligan 
resolved conflict in rtlabs



@1044

9 years 
boender 
 more fold/forall stuff



@1043

9 years 
sacerdot 
Axiom commented out.



@1042

9 years 
sacerdot 
Dead code removed.
Slow code uncommented.



@1041

9 years 
sacerdot 
fetch_assembly is still working after bug fix



@1040

9 years 
sacerdot 
Bug fixed in assembly.



@1039

9 years 
sacerdot 
fetch_assembly_pseudo2 repaired from dependent type madness



@1038

9 years 
boender 
 some more BVT improvements



@1037

9 years 
sacerdot 
Main theorem: comments are working again.



@1036

9 years 
sacerdot 
…



@1035

9 years 
sacerdot 
Main theorem (broken because of dependent types) almost restored.



@1034

9 years 
boender 
various & sundry fold/forall lemmas



@1033

9 years 
sacerdot 
ispelled & submitted



@1032

9 years 
sacerdot 
Final version.



@1031

9 years 
sacerdot 
..



@1030

9 years 
sacerdot 
…



@1029

9 years 
sacerdot 
…



@1028

9 years 
sacerdot 
One more sentence restored and fitted in.



@1027

9 years 
sacerdot 
Bug fixed in figure.



@1026

9 years 
mulligan 
final version? under 16 pages



@1025

9 years 
mulligan 
removing stray single words to reduce page usage



@1024

9 years 
mulligan 
tidied explanation of proof



@1023

9 years 
mulligan 
changes to english in matita section, shrunk diagrams in introduction …



@1022

9 years 
sacerdot 
…



@1021

9 years 
mulligan 
tidied english in sect 3



@1020

9 years 
sacerdot 
More on Matita.



@1019

9 years 
sacerdot 
Finished rewriting of Section 3.



@1018

9 years 
mulligan 
tidying



@1017

9 years 
mulligan 
complete, just under 16 pages



@1016

9 years 
sacerdot 
Many fixes to the code snippets.



@1015

9 years 
sacerdot 
One intermediate version of main_thm0 close to be repaired.



@1014

9 years 
sacerdot 
The main theorem is completely broken (again).



@1013

9 years 
mulligan 
more added



@1012

9 years 
mulligan 
just a few things left to change



@1011

9 years 
mulligan 
…



@1010

9 years 
mulligan 
more added, finished up to end of subsect 3.2



@1009

9 years 
mulligan 
added line number counts, etc.



@1008

9 years 
mulligan 
…



@1007

9 years 
mulligan 
added explanation of sdcc


