Changes

Summary

1. formatting (details)
2. Unified and fixed handling of parameter attributes. (details)
3. second attempt to interpose dl_iterate_phdr with disable/enable interrupt protection (details)

formatting

Unified and fixed handling of parameter attributes.

second attempt to interpose dl_iterate_phdr with disable/enable interrupt protection

Summary

1. Save parser change for git pull (details)
2. Basic change for enable enumerated array syntax (details)
3. Array can use enum instance with explicit initializer as designator. More general case will be covered in later commit (details)
4. The invariant flags on the test script were set up incorrectly. They should do as documented now and be on by default now. (details)
5. Removed extra copy in FixInitNew. (details)
6. fixed unhandled exception bug causing deleted memory to be accessed sometimes with full coroutines (details)
7. add reading fixed strings, seperate skip input manipulator, getline checks for delimiter (details)
8. add string output manipulators, third attempt at input manipulators for strings (details)
9. test reading fixed strings (details)
10. use arm64 to select 64-bit random numbers (details)
11. add get_seed for thread random numbers (details)
12. simplify 32/64-bit random-number constructors (details)
13. fix get_seed signature (details)
14. remove ucpp-style collections used before Mike's lists (details)
15. Revert "remove ucpp-style collections used before Mike's lists" (details)
16. remove unnecessary include bits/sequence.hfa (details)
17. add include bits/sequence.hfa to pthread.cfa after removing it from weakso_lock.hfa (details)
18. remove empty rule in cfa_abstract_parameter_declaration to remove conflicts (details)
19. Rename internal macro away from `TYPE`, which is a cs343 name collision. (details)
20. Port basic input-manipulator test from cstring to string and fix resulting bug. (details)
21. formatting (details)
22. Duplicate manipulator read-to-string code for string_res, and test. (details)
23. Deduplicate manipulator read-to-string by delegating from string to string_res. (details)
24. edited thesis to incorporate Gregors comments (details)
25. Bug fix (details)
26. expanded upon intro and conclusion (details)
27. Clean-up in MultiLevelExit. Primary purpose is to remove some uneeded fill code. (details)
28. changes over all usages of uC++ collections to use dlist instead (details)
29. Allow uninitialized enum value as designators (details)
30. Implement full set of relational operators for strings (details)
31. proofread chapters CFA_concurrency, intro, and conclusion (details)
32. merged in Peter's edits with a few small changes (details)
33. made some small formatting changes (details)
34. Fix cstring input length interpretation issue that had a buffer overflow case. (details)
35. Insert additional checks so that impossible, or just unimplemented, local control flow raises an error in CFA. (details)
36. add getenv, strlen, strcmp, strncmp to extras (details)
37. rename clock routines processor and program to processor_cpu and program_cpu to deal with name clash with processor (details)
38. formatting (details)
39. add malloc_stats_clear routine to zero the heap master and all active thread heaps (details)
40. Incorporated changes in response to Trevor's comments. (details)
41. Added references when incorporating Trevor's comments (details)
42. Added cfa equivalent of cofor/cobegin/coend (details)
43. Added another check to the invariants for SizeofExpr/AlignofExpr. (details)
44. committed makefile change for cofor include (details)
45. Update max/min so n-ary version doesn't match against the 2 argument case. Also added a test. (details)
46. Some mistakes in isPolyType found while working on the box pass. There was a new-ast verison using the old type map and, perhaps relatedly, the new type map version recursed into the no type map form. Also some clean-up in related functions. (details)
47. add uC++toCFA document (details)
48. clean up command-line handling and I/O (details)
49. formatting, replace constructor calls with set_allocation calls (details)
50. added cofor macro (details)
51. updated cofor test to use macro (details)
52. refactor cofor.hfa into cofor.cfa, adjust Makefile.am to handle cofor.cfa (details)
53. formatting (details)
54. formatting (details)
55. remove superfluous inner_loop routine (details)
56. Fix designator value in enumerated array and implemented enumerated array with inlined enume declaration (details)
57. Revert "Fix designator value in enumerated array and implemented enumerated array with inlined enume declaration" (details)
58. Fix designation value error (details)
59. Fix typo{ (details)
60. formatting, remove unnecessary use of c_str(), add better debug printing for cc1 (details)
61. add using namespace std to cfa.cc (details)
62. update help message from main.cc (details)
63. tweak comments for CFA comand-line options (details)
64. update tiobe index, update new command-line documentation (details)
65. fix lock/unlock for heap statistic-mode (details)
66. added enable/disable ehm with no params for coroutines (details)
67. added hotpotato non-local ehm test in both enable/disable and poll flavours (details)
68. a bit of cleanup of the hotpotato tests (details)
69. final hotpotato cleanup, removed two unneeded routines from explicit polling flavour (details)
70. make hotpotato test 32/64-bit specific due to different prng (details)
71. create 32-bit test results for hotpotato (details)
72. create ARM test results for hotpotato (details)
73. add keyword _Exception to uC++ lstlisting (details)
74. update citation entry (details)
75. add command setgcolumn to set lstlisting comment-column alignment, change size of lstlisting ^ character (details)
76. general updates to the Cforall user manual (details)
77. add keywords alignas, alignof, static_assert, and thread_local (details)
78. Translated the box pass to the new AST. This includes direct as possible translations of the existing passes are two fix in passes which correct AST problems the direct translation causes. Outside the box pass there have already been many changes, ad there is another in Instantiate Generics, which disconnects designators instead of leaving them connected to the original polymorphic type, which breaks the invarants once the fields are removed in the Eraser sub-pass. There was also a change that was made and un-made in one commit. If translate from the new-AST to the old-AST part way you must, where possible, sort the TypeEnvKey values by string comparison. However, it now passes over that so it would be just extra complexity and run time, so I removed it. I stand at the exit from a great woods, just shy of a year from when I entered it. It has been a difficult and tiring journey. The path has been long and at times comically winding; but most often it was invisible, hidden under an impenetrable canopy and I spend days looking for it. All for a short jog forward before getting lost again. In front of me is another woods. It looks smaller, but I can't see the other side. Anyways, time to keep walking. (details)
79. add corun/cofor statement, update old cofor (details)
80. Translated the Link-Once pass to the new ast. (details)
81. increase timeout in test.py from 180 to 240 seconds to deal with long locking tests on the ARM (details)
82. First clean-up pass on box pass. Some of it is stuff that could have been done in the initial commit but there was just so much of it. The rest is mostly simple restructuring that makes the translation less direct. (details)
83. Updated some documentation about the Instantiate Generic pass. (details)
84. Implement "scanset" semantics for manipulator `skip`, in place of exact-match. (details)
85. Support match-failure reporting for string input. (details)
86. Fixes to Box pass's DeclAdapter. The two largest are: Only local information in scopeTypeVars was used so it was replaced with localTypeVars, that greatly simplified the pass. The added parameters are also divided into groups more consistently. (details)
87. Fix read-to-variable-length-string cases when internal buffer fills. (details)
88. Fix compiler bug where duplicate type declarations caused crash. (details)
89. add arm64 to Jenkins nightly build (details)
90. add ARM architecture to nightly build (details)
91. add ARM architecture to nightly build (details)
92. add ARM architecture to regression tests (details)
93. fix typing mistake: arm4 to arm64 (details)
94. formatting (details)
95. formatting (details)
96. formatting, add PRNG copy to checkpoint PRNG state, remove autogen PRNG copy constructor and assignment to prevent accidental PRNG copy (details)
97. change host from arm64 to aarch64 for ARM nightly build (details)
98. Took the new EraseWith pass out of the box pass. It might be able to go even earlier or folded into an existing pass. C code generation now will not generate WithStmt nodes. (details)
99. formatting, SKULLDUGGERY to remove spurious free-nonheap-object warning (details)
100. allow precision < 3 for manipulator eng (details)
101. Implemented corun statement (details)
102. Updated cofor test to use new corun statement (details)
103. added error message for using corun statements without including the appropriate header (details)
104. Changed cofor files to support the corun statement (details)
105. Fixed up some whitespace. Pretty minor stuff mostly. (details)
106. Clean-up of some names and comments. Removed some TODO comments which are too old to do anything with. (details)
107. Added fence after channel handoff to prevent reordering on the arm that resulted in stale values (details)
108. temporary fix to linkonce problem by changing -7 to 7 so initialization does not invoke an operator (details)
109. change valide (French) to valid (English) (details)
110. Layout function builds its entire parameter list before creating the declaration now. (details)
111. Initial commit (details)
112. updates to the uC++toCFA cheat sheet (details)
113. Boxing no longer passes layout information about polymorphic types in the signature of the function. This simplifies the sub-passes in the box pass that added the arguments to the calls, the parameters to the declarations, and used the information internally. The effects in functions are less well quantified, because there are cases were the information has to be generated instead of passed in, but it is also never passed in (or generated) when not needed at all. (details)
114. Update the introduction (details)
115. Rename and move the file to proposal (details)
116. Remove some cruft from the box pass. Some pieces left over from earlier versions that are no longer needed and some rejected notes. (details)
117. temporally remove gcc-9 from ARM nightly build because it is failing (details)
118. TypeVarMap is now a subtype instead of an alias to remove the redundent constructor argument. Various bits of box pass clean-up. (details)
119. Small refactoring of some helper functions to avoid repeating a loop. (details)
120. add additional test for engineering and unit manipulator output (details)
121. Change string-read semantics so that reading nothing leaves the original value unmodified. (details)
122. Work on the box pass, fixing up the FunctionDecl update. (details)
123. added missing ARM fence on the signallee side of channel handoff (details)
124. first attempt at new C string input (details)
125. added fence to failing test to hopefully fix stale data issue on the arm, if this passes this fence will be put in waituntil code gen (details)
126. remove ISTYPE_VOID and ISTYPE_VOID_IMPL, and ends for input (details)
127. change genericUnion test fromt printf to sout (details)
128. ifdef'd the arm fences that were added to channels so that they only appear on the arm (details)
129. added WO fences for mcs spin lock to fix failing arm test (details)
130. added missing semicolons (details)
131. temporary change to as command to test ARM build (details)
132. add gcc-9 back into ARM full build (details)
133. Make all new declarations have a properly defined LinkageSpec. Also some general clean-up. (details)
134. test as.cc file on ARM (details)
135. comment out debug prints (details)
136. add __i386 with __x86_64 (details)
137. formatting (details)
138. Direct translation of code generation. (details)
139. Translation of the fix main pass to the new ast data structures. (details)
140. Updated some stats/counters to trigger off the new AST. Others will just have to be updated/re-added later because the support code to run them is missing and I don't know what stats we might need in the future. (details)
141. Code dumps (under the -P flag) no longer convert to the old ast to print. (details)
142. I don't actually know if this fixes anything, but the old version was definitely not doing anything and this is should have the same behaviour as the old one, and run just a bit faster. (details)
143. Parser now uses constants from the new ast types. (details)
144. Simple rework to BasicTypes-gen.cc to avoid refering to the old ast. No larger rework at this time. (details)
145. Added a missing include (not sure how that slipped through) and did some other replacements so the new ast is the source of constants. (details)
146. Hang on, eval doesn't have any side effects. (details)
147. Took out some old code from the new section of the Mangler. (details)
148. I forgot to move genName over to the new CodeGenerator module. It isn't really a code gen thing, more a modified lookup. (details)
149. Changed some new uses of UniqueId to the ast version. Changed where some debugging variables are stored. (details)
150. rename files gcc-builtins.cf builtins.cf extras.cf bootloader.cf and sync-builtins.cf with suffix .cfa (details)
151. Added test for the checks on the constructor/destructor attributes. (details)
152. removed print from static and dynamic tests that could be reordered due to a race (details)
153. Refactored the OperatorTable module. Most is just an update in style and some documentation. I did add a function that should be a useful utility for the demangler. (details)
154. added cofor implementation (details)
155. Added supporting library routines for cofor impl (details)
156. switched test to use new cofor syntax (details)
157. Updated the demangler to also demangle operator names. You can also pass a filename to demangle now. There is also a test, but I don't really know what good tests are for the demangler. (details)
158. removed fences for ARM that are likely not needed, insertion of fences into waituntil codegen TBD (details)
159. forgot to include in last push. Added a get_proc_count that doesn't need a cluster argument (details)
160. The demangler now uses the compiler's genType. The only difference I can find is that the demangler genType wrote functions ': RET_TYPE' instead of the C style. If this or a different change turns out to be imported it will have to be replicated post-translation. (details)
161. Translated the demangling code from the old ast to the new ast. (details)
162. Made UniqueName use other tools we have, removed the unused default and did some general clean-up. (details)
163. Remove BaseSyntaxNode and clean-up. (details)
164. Removed an extra forward declaration. (details)
165. Took the special main linkage code out of FunctionDecl and put it into a new pass. There is also a lot of related clean-up done. (details)
166. input manipulators for getline, excl, and incl reset argument to null string for failed read, temporarily make reading a C-string have the same semantics (details)
167. update bibliography entries (details)
168. formatting (details)
169. some updates (details)
170. latex formatting changes (details)
171. Removed forward declarations missed in the BaseSyntaxNode removal. Removed code and modified names to support two versions of the ast. (details)
172. remove old enum.md file (details)
173. Remove the warning directive from SpecCost, you can't actually use the standard maybe_accept here, it does not allow the user to insert code between visiting elements in a container. (details)
174. Fixed non-local ehm memory leak that occurs when a coroutine is deleted while suspended in a catchresume handler (details)
175. added test case for non-local ehm memory leak (details)
176. Some clean-up in src/Makefile.am (no change in behaviour). (details)
177. It turns out enumerate does work if you use this very particular form of the for loop added in C++17. Ironically in this simple vector case it is about as complex. Swapped the order of fields in the enuemate value to make the binding read key-then-value. (details)
178. Major round of clean-up in the GenPoly directory. (details)
179. added architecture specific expect files for non-local exception test since it uses prng (details)
180. added arm64 expect file for non-local exception test (details)
181. added x86 expect file for non-local exception test cardgame (details)
182. Round of clean-up in Tuples directory. (Skipping TupleExpansion, which will be fused together later.) (details)
183. Reformating and clean-up in SymTab directory. (details)
184. fixed bug where uninit mem was sometimes printed in test (details)
185. regen expected to match test change (details)
186. Most of ResolvExpr was written before the new style standard. Some files updated, focus on headers. (details)
187. fixed error where order of argument eval in compiler could cause different code to be generated (details)
188. formatting (details)
189. raise exception missing_data when read fails to find value, initial code for quoted manipulator (details)
190. proofread enumeration proposal (details)
191. formatting (details)
192. slight restructuring of automake rules for generating the demangler (details)
193. disable building the demangler (executable and library) (details)
194. fixed bug where cofor runner could derefence a null pointer (details)
195. Did some investigation of WithStmt. It may not be possible to convert it to a Stmt without changing how SymbolTable handles the with clauses. (details)
196. Added a test showing how the conflict handling in the with clause currently works. (details)
197. Removed Decl::fromId as it was unused. There are a few places that use uniqueId directly. (details)
198. Round of significant clean-up and reindentation of InitTweak directory. (details)
199. Round of clean-up in the Concurrency directory headers. (details)
200. Box pass Eraser now removes the polymorphic arguments now that the functions are no longer actually polymorphic. (details)
201. Slight improvement to forall clause erasing. (details)
202. formatting (details)
203. formatting (details)
204. formatting (details)
205. formatting (details)
206. Update enum.tex (details)
207. Removed unused case in CurrentObject::findNext. Did a bit of clean-up in the area. (details)
208. Completed the second demangler/automake task: The demangler (when built) in played in the driver directory beside cfa-cc. (details)
209. Removed the old-ast-compatable FunctionDecl constructor. However, enough cases pass nothing polymorphic along some of the uses of the constructor now go to a new monomorphic function constructor. (details)
210. Updated .gitignore for the new location of demangler. (details)
211. Combined the new and old TupleExpansion files. (details)
212. Removed 'New' suffixes, they are no longer needed for disambiguation. (details)
213. Renamed ResolvMode to ResolveMode. This is less consistent with the namespace, but is more consistent with almost everything else. (details)
214. Rename the ScrubTyVars moduel to ScrubTypeVars. This makes it more consistent with the data types and function names now used. (details)
215. Rename the linkReferenceTypes pass and the file that contains it. BaseInstType used to be called ReferenceToType, so the reason for the pass being called that is no longer true. (details)
216. Removed some code that supported ast conversion, it supported ast translation, and that is done. (details)
217. Implicit and layout parameters are now marked as const. This is primarily a safety feature to catch bad writes. (details)
218. Just some random clean-up in Box pass while I was musing. (details)
219. Some general clean-up, removing bits and pieces from the old version. (details)
220. Moved include from Decl header. Removed some old ast code from the Parser. (details)
221. reduce runtime testing duration of PRNG (details)
222. update PRNG output for x86 (details)
223. change incorrect architecture name __arm_64__ to __aarch64__ (details)
224. add PRNG expected output for arm64 (details)
225. Clean-up of the chain mutator. Seems like it is underused. (details)
226. (Not really readable right now. Will clean up today and probably tomorrow) 1. Remove implicit conversion; 2. Companion trait; (details)
227. Try terminate now does not call the catch function, now they have the same caller. This involved updating some platform dependent code which should be correct in all cases. (details)
228. Accept Peter's proofreading and adjustment of examples to current syntax (details)
229. Recent thesis writing (details)
230. change containers/list.hfa to collections/list.hfa in thesis test files (details)
231. comment out assert( s1_mid == fc ) bug (details)
232. comment out missing lstinputlisting files, fix missing @'s for lstinline code (details)
233. Missing files from last thesis push (details)
234. comment back in lstinputlisting files after missing files pushed (details)
235. Minimal change to get exceptions running the old way while a fix is found. Or even more details on the problem. (details)
236. add papers directory and start populating it (details)
237. first attempt at simplifying SemanticError and its usage (details)
238. formatting (details)
239. update expected output from tests after changes in semantic error messages (details)
240. Change the unification scheme (details)
241. A harder revert of exception changes, hopefully it will get the tests working. (details)
242. Started the storage section (inling and weak reference) (details)
243. first attempt at simplifying SemanticWarning, inline SemanticError routine (details)
244. second attempt at simplifying SemanticError messages (details)
245. small proofreading changes (details)
246. Small fix in code generation. The first case in a switch is formatted properly. (details)
247. Update enumeration data structure (details)
248. Another attempt at fixing execptions. It is very close to the last attempt but the offsets have been updated and checked. (details)
249. Clean-up in the exception translate pass. This changes some patterns from pre-translation and sets some things up for later reworks. (details)
250. You can how use local control flow out of 'catch' clauses. Added a test to show that it works. (details)
251. Forgot to remove an unused function from the exception code. (details)
252. Reorganization of Linkage::Spec. is_mangled represented two properties with is_gcc_builtin separating them in one case. The second property is_overloadable which replaces is_gcc_builtin. (details)
253. Small bug fix in code generation. Seems to only come up while generating intermediate code. (details)
254. Remove extra compound statement and clean-up in InitTweak. (details)
255. Hoist nested enum in Struct (details)
256. formatting, change cmp to strcmp, add strlen and strcat (details)
257. formatting (details)
258. continue to work on quoted formatting and generalization (details)
259. first attempt to interpose dl_iterate_phdr with disable/enable interrupt protection (details)
260. update string testing and expected output (details)
261. update manipulator testing and expected output (details)
262. update command-line processing (details)
263. Enable array RAII and provide uninit(-), a uNoCtor equivalent. (details)

Save parser change for git pull

Basic change for enable enumerated array syntax

Array can use enum instance with explicit initializer as designator. More general case will be covered in later commit

The invariant flags on the test script were set up incorrectly. They should do as documented now and be on by default now.

Removed extra copy in FixInitNew.

fixed unhandled exception bug causing deleted memory to be accessed sometimes with full coroutines

add reading fixed strings, seperate skip input manipulator, getline checks for delimiter

add string output manipulators, third attempt at input manipulators for strings

test reading fixed strings

use arm64 to select 64-bit random numbers

add get_seed for thread random numbers

simplify 32/64-bit random-number constructors

fix get_seed signature

remove ucpp-style collections used before Mike's lists

Revert "remove ucpp-style collections used before Mike's lists"

This reverts commit b388d1baad1af006ac9ea384e38dd2abb5f0ba6e.

remove unnecessary include bits/sequence.hfa

add include bits/sequence.hfa to pthread.cfa after removing it from weakso_lock.hfa

remove empty rule in cfa_abstract_parameter_declaration to remove conflicts

Rename internal macro away from `TYPE`, which is a cs343 name collision.

Port basic input-manipulator test from cstring to string and fix resulting bug.

Bug was `ignore(...s...)` overwriting `s` with garbage.

formatting

Duplicate manipulator read-to-string code for string_res, and test.

Deduplicate manipulator read-to-string by delegating from string to string_res.

edited thesis to incorporate Gregors comments

Bug fix

expanded upon intro and conclusion

Clean-up in MultiLevelExit. Primary purpose is to remove some uneeded fill code.

changes over all usages of uC++ collections to use dlist instead

Allow uninitialized enum value as designators

Implement full set of relational operators for strings

proofread chapters CFA_concurrency, intro, and conclusion

merged in Peter's edits with a few small changes

made some small formatting changes

Fix cstring input length interpretation issue that had a buffer overflow case.

The cases added to the manipulatorsInput test are runnable against an old libcfa build.  In this setup, the test fails with an illustration of the bug.

The testing in this commit drives the following inputs through a length-8 buffer.
- 123456
- 123456789
The obviously-missing cases, like 1234567, will be added later.
They will accompany fixes for further bugs not solved yet.

Insert additional checks so that impossible, or just unimplemented, local control flow raises an error in CFA.

add getenv, strlen, strcmp, strncmp to extras

rename clock routines processor and program to processor_cpu and program_cpu to deal with name clash with processor

formatting

add malloc_stats_clear routine to zero the heap master and all active thread heaps

Incorporated changes in response to Trevor's comments.

Added references when incorporating Trevor's comments

Added cfa equivalent of cofor/cobegin/coend

Added another check to the invariants for SizeofExpr/AlignofExpr.

committed makefile change for cofor include

Update max/min so n-ary version doesn't match against the 2 argument case. Also added a test.

Some mistakes in isPolyType found while working on the box pass. There was a new-ast verison using the old type map and, perhaps relatedly, the new type map version recursed into the no type map form. Also some clean-up in related functions.

add uC++toCFA document

clean up command-line handling and I/O

formatting, replace constructor calls with set_allocation calls

added cofor macro

updated cofor test to use macro

refactor cofor.hfa into cofor.cfa, adjust Makefile.am to handle cofor.cfa

formatting

formatting

remove superfluous inner_loop routine

Fix designator value in enumerated array and implemented enumerated array with inlined enume declaration

Revert "Fix designator value in enumerated array and implemented enumerated array with inlined enume declaration"

This reverts commit c1e66d966aadf6846330871033458d4a398bd576.

Fix designation value error

Fix typo{

formatting, remove unnecessary use of c_str(), add better debug printing for cc1

add using namespace std to cfa.cc

update help message from main.cc

tweak comments for CFA comand-line options

update tiobe index, update new command-line documentation

fix lock/unlock for heap statistic-mode

added enable/disable ehm with no params for coroutines

added hotpotato non-local ehm test in both enable/disable and poll flavours

a bit of cleanup of the hotpotato tests

final hotpotato cleanup, removed two unneeded routines from explicit polling flavour

make hotpotato test 32/64-bit specific due to different prng

create 32-bit test results for hotpotato

create ARM test results for hotpotato

add keyword _Exception to uC++ lstlisting

update citation entry

add command setgcolumn to set lstlisting comment-column alignment, change size of lstlisting ^ character

general updates to the Cforall user manual

add keywords alignas, alignof, static_assert, and thread_local

Translated the box pass to the new AST. This includes direct as possible translations of the existing passes are two fix in passes which correct AST problems the direct translation causes. Outside the box pass there have already been many changes, ad there is another in Instantiate Generics, which disconnects designators instead of leaving them connected to the original polymorphic type, which breaks the invarants once the fields are removed in the Eraser sub-pass. There was also a change that was made and un-made in one commit. If translate from the new-AST to the old-AST part way you must, where possible, sort the TypeEnvKey values by string comparison. However, it now passes over that so it would be just extra complexity and run time, so I removed it. I stand at the exit from a great woods, just shy of a year from when I entered it. It has been a difficult and tiring journey. The path has been long and at times comically winding; but most often it was invisible, hidden under an impenetrable canopy and I spend days looking for it. All for a short jog forward before getting lost again. In front of me is another woods. It looks smaller, but I can't see the other side. Anyways, time to keep walking.

add corun/cofor statement, update old cofor

Translated the Link-Once pass to the new ast.

increase timeout in test.py from 180 to 240 seconds to deal with long locking tests on the ARM

First clean-up pass on box pass. Some of it is stuff that could have been done in the initial commit but there was just so much of it. The rest is mostly simple restructuring that makes the translation less direct.

Updated some documentation about the Instantiate Generic pass.

Implement "scanset" semantics for manipulator `skip`, in place of exact-match.

That is, `sin | skip("abc")` now consumes all input before the "+" on input "cccccb+".

Exact-match semantics is available with just `sin | "abc"`.

(Prior exact-match semantics was discovered by surprise, while debugging a test program for an upcoming commit.)

Support match-failure reporting for string input.

On failed match, assign the empty-string value to the target.
Prior handling did not modify the target,
which in context like `char s[10]; sin | exclude("\n", s);`,
left `s` uninitialized, and invalid for `strlen`, when the input's next character is a newline.
In this case, `scanf` offers the return value, the count of output arguments written,
to say, "I did not write to your buffer," a feature lost in the pipe abstraction.

Add tests to exercise such cases.

(Prior "leave uninitialized" treatment was discovered by surprise, while debugging a test program for an upcoming commit.)

Fixes to Box pass's DeclAdapter. The two largest are: Only local information in scopeTypeVars was used so it was replaced with localTypeVars, that greatly simplified the pass. The added parameters are also divided into groups more consistently.

Fix read-to-variable-length-string cases when internal buffer fills.

Also fix read-to-cstring ability to give no-exception cases when an entire buffer fills.

The added test cases run, and fail, when run against prior libcfa.
Doing so illustrates a CFA-string-level bug.
Doing so illustrates a C-string-level changed semantics.

At the CFA-string level, the bug was, when reading strings of just the right length,
what should be two reads ("abc" then "def") gets mashed into one ("abcdef").
These cases are clearly bugs because a test program that just echoes chuncks of delimeted input would do so inaccurately.
They're just hard to drive because the relevant chunk lengths are implementation-dependent, and sometimes big.

At the C-string level, the semantic change concerns when to throw the cstring_length exception.
By this change, make the original semantics,
"An exception means the maximum number of characters was read," into
"An exception means that if the buffer were larger, then more characters would have been read."

The added test cases cover the respective stop conditions for manipulator state "%s", include, exclude, getline, and getline/delimiter.

Fix compiler bug where duplicate type declarations caused crash.

And add missing test.  The test runs and fails with the previous cfa compiler.

add arm64 to Jenkins nightly build

add ARM architecture to nightly build

add ARM architecture to nightly build

add ARM architecture to regression tests

fix typing mistake: arm4 to arm64

formatting

formatting

formatting, add PRNG copy to checkpoint PRNG state, remove autogen PRNG copy constructor and assignment to prevent accidental PRNG copy

change host from arm64 to aarch64 for ARM nightly build

Took the new EraseWith pass out of the box pass. It might be able to go even earlier or folded into an existing pass. C code generation now will not generate WithStmt nodes.

formatting, SKULLDUGGERY to remove spurious free-nonheap-object warning

allow precision < 3 for manipulator eng

Implemented corun statement

Updated cofor test to use new corun statement

added error message for using corun statements without including the appropriate header

Changed cofor files to support the corun statement

Fixed up some whitespace. Pretty minor stuff mostly.

Clean-up of some names and comments. Removed some TODO comments which are too old to do anything with.

Added fence after channel handoff to prevent reordering on the arm that resulted in stale values

temporary fix to linkonce problem by changing -7 to 7 so initialization does not invoke an operator

change valide (French) to valid (English)

Layout function builds its entire parameter list before creating the declaration now.

Initial commit

updates to the uC++toCFA cheat sheet

Boxing no longer passes layout information about polymorphic types in the signature of the function. This simplifies the sub-passes in the box pass that added the arguments to the calls, the parameters to the declarations, and used the information internally. The effects in functions are less well quantified, because there are cases were the information has to be generated instead of passed in, but it is also never passed in (or generated) when not needed at all.

Update the introduction

Rename and move the file to proposal

Remove some cruft from the box pass. Some pieces left over from earlier versions that are no longer needed and some rejected notes.

temporally remove gcc-9 from ARM nightly build because it is failing

TypeVarMap is now a subtype instead of an alias to remove the redundent constructor argument. Various bits of box pass clean-up.

Small refactoring of some helper functions to avoid repeating a loop.

add additional test for engineering and unit manipulator output

Change string-read semantics so that reading nothing leaves the original value unmodified.

Logically revert #8cffa4f21, and apply the change onto work done since.

At cstring layer, libcfa change is a simplification, and test suite change is merely a different expectation.

At var-len-string layer, libcfa change is a patch to enable detecting the nothing-read case,
and test sute change (shows the different expectation but also) puts a similar patch into its "stop trying if nothing read" loop exits.

Work on the box pass, fixing up the FunctionDecl update.

added missing ARM fence on the signallee side of channel handoff

first attempt at new C string input

added fence to failing test to hopefully fix stale data issue on the arm, if this passes this fence will be put in waituntil code gen

remove ISTYPE_VOID and ISTYPE_VOID_IMPL, and ends for input

change genericUnion test fromt printf to sout

ifdef'd the arm fences that were added to channels so that they only appear on the arm

added WO fences for mcs spin lock to fix failing arm test

added missing semicolons

temporary change to as command to test ARM build

add gcc-9 back into ARM full build

Make all new declarations have a properly defined LinkageSpec. Also some general clean-up.

test as.cc file on ARM

comment out debug prints

add __i386 with __x86_64

formatting

Direct translation of code generation.

Translation of the fix main pass to the new ast data structures.

Updated some stats/counters to trigger off the new AST. Others will just have to be updated/re-added later because the support code to run them is missing and I don't know what stats we might need in the future.

Code dumps (under the -P flag) no longer convert to the old ast to print.

I don't actually know if this fixes anything, but the old version was definitely not doing anything and this is should have the same behaviour as the old one, and run just a bit faster.

Parser now uses constants from the new ast types.

Simple rework to BasicTypes-gen.cc to avoid refering to the old ast. No larger rework at this time.

Added a missing include (not sure how that slipped through) and did some other replacements so the new ast is the source of constants.

Hang on, eval doesn't have any side effects.

Took out some old code from the new section of the Mangler.

I forgot to move genName over to the new CodeGenerator module. It isn't really a code gen thing, more a modified lookup.

Changed some new uses of UniqueId to the ast version. Changed where some debugging variables are stored.

rename files gcc-builtins.cf builtins.cf extras.cf bootloader.cf and sync-builtins.cf with suffix .cfa

Added test for the checks on the constructor/destructor attributes.

removed print from static and dynamic tests that could be reordered due to a race

Refactored the OperatorTable module. Most is just an update in style and some documentation. I did add a function that should be a useful utility for the demangler.

added cofor implementation

Added supporting library routines for cofor impl

switched test to use new cofor syntax

Updated the demangler to also demangle operator names. You can also pass a filename to demangle now. There is also a test, but I don't really know what good tests are for the demangler.

removed fences for ARM that are likely not needed, insertion of fences into waituntil codegen TBD

forgot to include in last push. Added a get_proc_count that doesn't need a cluster argument

The demangler now uses the compiler's genType. The only difference I can find is that the demangler genType wrote functions ': RET_TYPE' instead of the C style. If this or a different change turns out to be imported it will have to be replicated post-translation.

Translated the demangling code from the old ast to the new ast.

Made UniqueName use other tools we have, removed the unused default and did some general clean-up.

Remove BaseSyntaxNode and clean-up.

Removed an extra forward declaration.

Took the special main linkage code out of FunctionDecl and put it into a new pass. There is also a lot of related clean-up done.

input manipulators for getline, excl, and incl reset argument to null string for failed read, temporarily make reading a C-string have the same semantics