Ticket #3414: ghc-fix-build-dnl-windres.dpatch

File ghc-fix-build-dnl-windres.dpatch, 28.2 KB (added by ajd, 5 years ago)

Patch to GHC that seems to fix one of the issues

Line 
1Fri Jul 31 18:23:35 PDT 2009  Alexander Dunlap <alexander.dunlap@gmail.com>
2  * FIX BUILD: Remove harmful dnls and identifier typo in aclocal.m4
3
4New patches:
5
6[FIX BUILD: Remove harmful dnls and identifier typo in aclocal.m4
7Alexander Dunlap <alexander.dunlap@gmail.com>**20090801012335
8 Ignore-this: 825b21e13d4b3f04fab91bf07f4e693b
9] {
10hunk ./aclocal.m4 250
11 if test ! -f compiler/parser/Parser.hs || test ! -f compiler/main/ParsePkgConf.hs || test ! -f compiler/cmm/CmmParse.hs || test ! -f compiler/parser/ParserCore.hs
12 then
13     FP_COMPARE_VERSIONS([$fptools_cv_happy_version],[-lt],[1.15],
14-      [AC_MSG_ERROR([Happy version 1.15 or later is required to compile GHC.])])[]dnl
15+      [AC_MSG_ERROR([Happy version 1.15 or later is required to compile GHC.])])[]
16 fi
17 HappyVersion=$fptools_cv_happy_version;
18 AC_SUBST(HappyVersion)
19hunk ./aclocal.m4 285
20 if test ! -f compiler/cmm/CmmLex.hs || test ! -f compiler/parser/Lexer.hs
21 then
22     FP_COMPARE_VERSIONS([$fptools_cv_alex_version],[-lt],[2.1.0],
23-      [AC_MSG_ERROR([Alex version 2.1.0 or later is required to compile GHC.])])[]dnl
24+      [AC_MSG_ERROR([Alex version 2.1.0 or later is required to compile GHC.])])[]
25 fi
26 AlexVersion=$fptools_cv_alex_version;
27 AC_SUBST(AlexVersion)
28hunk ./aclocal.m4 325
29         changequote([, ])dnl
30     ])
31     FP_COMPARE_VERSIONS([$fptools_cv_windres_version],[-eq],[2.17.50],
32-      [AC_MSG_ERROR([Your windres version isn't compatible with GHC. 2.15.91 and 2.18.50 are known to work.])])[]dnl
33-    indResVersion=$fptools_cv_windres_version;
34+      [AC_MSG_ERROR([Your windres version isn't compatible with GHC. 2.15.91 and 2.18.50 are known to work.])])[]
35+    WindResVersion=$fptools_cv_windres_version;
36     AC_SUBST(WindResVersion)
37 fi
38 ])
39}
40
41Context:
42
43[Fix cleaning the integer package
44Ian Lynagh <igloo@earth.li>**20090730143129]
45[Fix warnings when building with the HEAD
46Ian Lynagh <igloo@earth.li>**20090730143031]
47[Remove some redundant fromIntegral's
48Ian Lynagh <igloo@earth.li>**20090730105532]
49[Fix space problems in ghci
50Ian Lynagh <igloo@earth.li>**20090730105351
51 We were making arrays with range (0, n-1) which is bad if n == 0 now
52 that we are using Word types.
53]
54[Make the types we use when creating GHCi bytecode better match reality
55Ian Lynagh <igloo@earth.li>**20090729130911
56 We were keeping things as Int, and then converting them to Word16 at
57 the last minute, when really they ought to have been Word16 all along.
58]
59[Add an (Outputable Word16) instance
60Ian Lynagh <igloo@earth.li>**20090728134436]
61[Fix whitespace in ByteCodeAsm.lhs
62Ian Lynagh <igloo@earth.li>**20090728123444]
63[Add CHECK(p), like ASSERT(p) but works even when !defined(DEBUG)
64Simon Marlow <marlowsd@gmail.com>**20090729075433
65 Ignore-this: a723f456e4b1eaeaa617a675da276aa2
66 For inexpensive assertions
67]
68[enable the x86-specific versions of atomic_inc()/atomic_dec()
69Simon Marlow <marlowsd@gmail.com>**20090729075307
70 Ignore-this: a271b7ade5502ec4d8444aac19f4b4c3
71]
72[fix warning
73Simon Marlow <marlowsd@gmail.com>**20090728103855
74 Ignore-this: ee626aa5de0a2aed8f44ae1131cc341d
75]
76[fix warning
77Simon Marlow <marlowsd@gmail.com>**20090728102707
78 Ignore-this: ad4f07a163921d17d242a9452b4bb578
79]
80[fix a warning
81Simon Marlow <marlowsd@gmail.com>**20090728101731
82 Ignore-this: 73ead0a0004723757c0a51b56681c991
83]
84[Be a bit more sensible about choosing external OccNames
85Simon Marlow <marlowsd@gmail.com>**20090728100434
86 Ignore-this: 4adcd661e76440deb2b4ab498ebd2d1e
87 
88 Instead of chr_$wchr, we now just get $wchr.  In general, when an
89 OccName is system-generated, we leave it out of the final external
90 name, preferring to use the name of the exported parent instead (which
91 is necessarily a user-written name).
92 
93 Names should be no less deterministic, but should be shorter and more
94 readable.
95]
96[Remove old 'foreign import dotnet' code
97Simon Marlow <marlowsd@gmail.com>**20090727144524
98 Ignore-this: 821ebea2c3897415195318f107421472
99 It still lives in darcs, if anyone wants to revive it sometime.
100]
101[remove a couple of ToDos
102Simon Marlow <marlowsd@gmail.com>**20090727094719
103 Ignore-this: ef00fc481821dff4381ba9efcd792708
104]
105[buildinfo files need a $$(wildcard)
106Simon Marlow <marlowsd@gmail.com>**20090727091012
107 Ignore-this: b4c3201dfa81fef32ee254dd9c955b2d
108]
109[Slight tweak to avoid overflowing the command-line size in bindist
110Simon Marlow <marlowsd@gmail.com>**20090727090946
111 Ignore-this: 1e2ff207d03fed08576ac59f0b46c08c
112 Not a real fix: if this bites us again we'll have to rethink
113]
114[avoid (benign) error about overriding rules for binary-dist
115Simon Marlow <marlowsd@gmail.com>**20090727090903
116 Ignore-this: ffb7f7bf1290f2faf96ac177f76a1422
117]
118[Give a better error message for hidden packages when building Cabal package
119Ian Lynagh <igloo@earth.li>**20090726194915
120 Fixes #3168
121]
122[Add a -fbuilding-cabal-package flag
123Ian Lynagh <igloo@earth.li>**20090726181934
124 This means GHC knows whether it's building a Cabal package, or just
125 Haskell sources. For example, it may wish to give different error
126 messages when building a Cabal package.
127]
128[Add an extension to disable n+k patterns
129Ian Lynagh <igloo@earth.li>**20090725134703]
130[Fix a warning on Windows
131Ian Lynagh <igloo@earth.li>**20090724221244]
132[Remove GHC's haskell98 dependency
133Ian Lynagh <igloo@earth.li>**20090724210825]
134[add number of bytes to +RTS -DS leak reports
135Simon Marlow <marlowsd@gmail.com>**20090724150010
136 Ignore-this: 3a66585c8fd2b58ce96abab1e154fb6e
137]
138[free the gc_thread structures during shutdown
139Simon Marlow <marlowsd@gmail.com>**20090724145956
140 Ignore-this: 25efeb189cbfb549af4550d266604f0e
141]
142[Add atomic_inc()/atomic_dec(), and use them to replace gc_running_mutex
143Simon Marlow <marlowsd@gmail.com>**20090724142620
144 Ignore-this: d775eeaf85fd0f9064d87a0909134bc0
145 This also fixes a memory leak on Windows with -threaded, because we
146 were calling initMutex(&gc_running_mutex) for each GC, which allocates
147 memory.
148]
149[Rewrite the foreign import string parser using ReadP
150Simon Marlow <marlowsd@gmail.com>**20090723152138
151 Ignore-this: 1c7db770a29d48710b05e2a3d216b2a8
152 And kill the duplicate one in HsSyn.Convert
153]
154[point to the wiki
155Simon Marlow <marlowsd@gmail.com>**20090723132345
156 Ignore-this: c11300bac62ce2f56d7fc271aa26dbcd
157]
158[Remove note about avoiding use of #def in libraries
159Simon Marlow <marlowsd@gmail.com>**20090723111026
160 Ignore-this: 8a027ed37b2d10094f7a31548aee2535
161 It should be safe to use now that we aren't relying on C prototypes
162 for foreign functions in via-C code.
163]
164[refactorings
165Simon Marlow <marlowsd@gmail.com>**20090723091230
166 Ignore-this: 836feb0e819127603dd56623af6e48dc
167]
168[Fix Trac #3391: make generic to/from bindings only for newly-declared types
169simonpj@microsoft.com**20090723155803
170 Ignore-this: bb56c2ec054397d421dce13d5eb6c73f
171 
172 Before this patch we were bogusly making to/from bindings for all data types
173 in the TcGblEnv.  But that is wrong when we have multiple "chunks" of
174 bindings in Template Haskell.  We should start from the declarations
175 themselves.  Easy.
176 
177]
178[Print explicit braces and semicolons in do-notation
179simonpj@microsoft.com**20090723152411
180 Ignore-this: a97ddf19774d27d15a01d63787708b20
181 
182 By printing explicit braces we make it more likely that pretty-printed
183 code will be acceptable if fed back into GHC.
184 
185 See http://www.haskell.org/pipermail/glasgow-haskell-users/2009-July/017554.html
186 
187]
188[Documentation for stand-alone deriving (Trac #3012)
189simonpj@microsoft.com**20090723132558
190 Ignore-this: 54445c5984594eb7f82151b2ac118695
191]
192[Windows only: set the encoding on stdin to utf8
193Simon Marlow <marlowsd@gmail.com>**20090723121913
194 Ignore-this: d65115d9711b5fb68e77786565ef6de
195 Otherwise it defaults to latin1.
196]
197[Fix Trac #3012: allow more free-wheeling in standalone deriving
198simonpj@microsoft.com**20090723130145
199 Ignore-this: 357580b9388ccbe1da3c1da3ba90e456
200 
201 In standalone deriving, we now do *not* check side conditions.
202 We simply generate the code and typecheck it.  If there's a type
203 error, it's the programmer's problem.
204 
205 This means that you can do 'deriving instance Show (T a)', where
206 T is a GADT, for example, provided of course that the boilerplate
207 code does in fact typecheck.
208 
209 I put some work into getting a decent error message.  In particular
210 if there's a type error in a method, GHC will show the entire code
211 for that method (since, after all, the user did not write it).
212 Most of the changes are to achieve that goal.
213 
214 Still to come: changes in the documentation.
215 
216]
217[Use the ErrMsg record type
218simonpj@microsoft.com**20090723130108
219 Ignore-this: 7fb6dd78d3185da0c33901b8aac8d108
220]
221[Stop generating redundant parens in 'deriving' code
222simonpj@microsoft.com**20090723125903
223 Ignore-this: 6fc82df9648a82bcf7bf6fdfa9b4dad3
224 
225 This makes the code printed by -ddump-deriv look prettier
226 
227]
228[Wibble to printing tuple sections
229simonpj@microsoft.com**20090723125756
230 Ignore-this: af2a1b9784f6447fea0e11d454cf082f
231]
232[Fix Trac #3193: improve line number reporting for equality constraints
233simonpj@microsoft.com**20090723065504
234 Ignore-this: b45a68071bcaca48cad7855dccb9c9eb
235 
236 When reporting an error from a failed equality constraint, we were
237 setting the *context* but not the *line number* in TcTyFuns.eqInstMisMatch
238 As a result, the line number didn't match the context at all.  It's
239 trivial to fix.
240 
241 I'm 99% certain this fixes #3193, but it's too complicated to
242 reproduce, so I have not actually tested it.
243 
244]
245[Add fmapM_maybe :: Monad m => (a -> m b) -> Maybe a -> m (Maybe b)
246simonpj@microsoft.com**20090723064932
247 Ignore-this: db5f6319b52a5e6b5f85d76985f2a7c9
248 
249 This function isn't used at the moment, but Max added it, and it
250 looks useful.
251 
252]
253[Add tuple sections as a new feature
254simonpj@microsoft.com**20090723063859
255 Ignore-this: d42a26fc1efff112b852b5c1135c1746
256 
257 This patch adds tuple sections, so that
258 
259        (x,,z)  means   \y -> (x,y,z)
260 
261 Thanks for Max Bolinbroke for doing the hard work.
262 
263 In the end, instead of using two constructors in HsSyn, I used
264 just one (still called ExplicitTuple) whose arguments can be
265        Present (LHsExpr id)
266 or     Missing PostTcType
267 
268 While I was at it, I did a bit of refactoring too.
269 
270]
271[NetBSD defines _REENTRANT in its header files, so compiling ghc gives
272Simon Marlow <marlowsd@gmail.com>**20090723075030
273 Ignore-this: 4722c4ff0541c6080de8f433e498684
274 redefinition warnings for all files that are including includes/Rts.h.
275 
276 Contributed by: Krister Walfridsson <krister.walfridsson@gmail.com>
277]
278[includes/TSO.h: kill trailing whitespace
279Samuel Bronson <naesten@gmail.com>**20090722170354
280 Ignore-this: 36d4afd1a21188d604ed6b432942dcdc
281]
282[Say what StgTSOBlockInfo is for, where to read about it.
283Samuel Bronson <naesten@gmail.com>**20090722163011
284 Ignore-this: 6c09e11f23594251cdc2db1bc642edc9
285]
286[Make the Integer library used directly configurable in GHC and base
287Ian Lynagh <igloo@earth.li>**20090722151048
288 Rather than indirecting through an integer package
289]
290[Fix cleaning with the new integer changes
291Ian Lynagh <igloo@earth.li>**20090722142545]
292[Add integer-simple as a build option
293Ian Lynagh <igloo@earth.li>**20090722013137]
294[Tweak whitespace
295Ian Lynagh <igloo@earth.li>**20090719221303]
296[thenIO, bindIO, returnIO moved to GHC.Base
297Simon Marlow <marlowsd@gmail.com>**20090722102219
298 Ignore-this: 5d6c5988e0abab2c5169540aa8ddedb9
299]
300[remove unused $(HscIfaceFileVersion)
301Simon Marlow <marlowsd@gmail.com>**20090720145053
302 Ignore-this: d725cbbde3c68673e2342b370460d87f
303]
304[Choose external names more predictably
305Simon Marlow <marlowsd@gmail.com>**20090720144751
306 Ignore-this: 17513bc93af646108f21bbed1c8f4a3a
307 
308 Now, for a compiler-generated binding "x", if "x" is referred to by
309 the exported "f", then it will be named "f_x" rather than something
310 like "x23".  This means that hopefully
311 
312   - compilation will more often product the same results given the
313     same input (the choice of names is not dependent on the
314     non-deterministic order of bindings within the compiler).
315 
316   - less recompilation will be necessary after making changes
317 
318   - navigating Core might be a bit easier.
319 
320 unfortunately, compilation with -O still does not consistently produce
321 the same ABI.  The simplifier sometimes does different things,
322 apparently.
323 
324 Names will be longer, but I can't see a way around that.
325]
326[Use stable ordering in the dependencies
327Simon Marlow <marlowsd@gmail.com>**20090717123449
328 Ignore-this: e20bac233cf6f834e69c027ff60b5b50
329 Fixes another cause of wobbly interface files and unnecessary recompilation.
330]
331[fall back on libffi for 'foreign import "wrapper"' if necessary
332Simon Marlow <marlowsd@gmail.com>**20090716134549
333 Ignore-this: e1073e1ad77e720326865a6d3c4f3790
334]
335[Take account of GADTs when reporting patterm-match overlap
336simonpj@microsoft.com**20090722050933
337 Ignore-this: 7dcbdcb91021e83e6e6208a2e68c50c9
338 
339 When matching against a GADT, some of the constructors may be impossible.
340 For example
341        data T a where
342           T1 :: T Int
343           T2 :: T Bool
344           T3 :: T a
345 
346         f :: T Int -> Int
347         f T1 = 3
348         f T3 = 4
349 
350 Here, does not have any missing cases, despite omittting T2, because
351 T2 :: T Bool.
352 
353 This patch teaches the overlap checker about GADTs, which happily
354 turned out to be rather easy even though the overlap checker needs
355 a serious rewrite.
356 
357]
358[Fix Trac #3382: desugaring of NPats
359simonpj@microsoft.com**20090720061226
360 Ignore-this: 4dccdaf2b7d6428141dcf174cb455a20
361 
362 Max spotted that the short-cut rules for desugaring NPats (where
363 we compare against a literal) were wrong now that we have overloaded
364 strings.
365 
366]
367[Add a -fwarn-dodgy-exports flag; fixes #1911
368Ian Lynagh <igloo@earth.li>**20090719200124
369 This is used to control warnings that were previously unconditional.
370]
371[Build terminfo if we /aren't/ on Windows, not if we /are/
372Ian Lynagh <igloo@earth.li>**20090719111709]
373[Change how PACKAGES is constructed, so that everything gets cleaned properly
374Ian Lynagh <igloo@earth.li>**20090718210058
375 If Windows wasn't defined properly then the Win32 package wasn't being
376 cleaned, as it wasn't added to PACKAGES. Now we always add everything to
377 PACKAGES when CLEANING=YES.
378]
379[temporarily turn off unused import warnings for the time library
380Ian Lynagh <igloo@earth.li>**20090718183445]
381[Follow the split directory rename in the GHC build system rules
382Ian Lynagh <igloo@earth.li>**20090718155618]
383[Add osuf to the name we use for the split dir
384Ian Lynagh <igloo@earth.li>**20090718145522
385 This avoids a collision between the directories we use when compiling
386 multiple ways, which in turn leads to a race condition in parallel
387 builds.
388]
389[Temporarily turn off unused-do-bind warnings for the time package
390Ian Lynagh <igloo@earth.li>**20090718134536]
391[Make ghc-cabal handle "Custom" Setup.hs files that have a configure script
392Ian Lynagh <igloo@earth.li>**20090718131555]
393[Add the time library, and support for libraries in tarballs
394Ian Lynagh <igloo@earth.li>**20090718121649]
395[Always serialise Int as 64bit values; fixes trac #3041
396Ian Lynagh <igloo@earth.li>**20090717224203
397 This means that, provided the values are small enough, files
398 serialized are portable between architectures. In particular,
399 .haddock files are portable.
400]
401[Remove some code that has always been commented out
402Ian Lynagh <igloo@earth.li>**20090717224100]
403[Fix Trac #3346: tcSimplify for LHS of RULES with type equalities
404simonpj@microsoft.com**20090717155722
405 Ignore-this: dfdd0f9a62d78d63276a4d558831099c
406]
407[Allow mixed case in the LINE pragma; patch from squadette; fixes #1817
408Ian Lynagh <igloo@earth.li>**20090717133522]
409[Comment only
410simonpj@microsoft.com**20090717120154
411 Ignore-this: f96b11e602fe4b311c1e466af9aa1908
412]
413[Add missing case for eq_note.
414t-peterj@microsoft.com**20090624134407]
415[Rename parameters to make debugging code compile.
416t-peterj@microsoft.com**20090626105440]
417[Comment fix: use the same variable names in the conclusion as in the premise.
418t-peterj@microsoft.com**20090618092235]
419[Typo fixes, from Alexey Mahotkin
420Ian Lynagh <igloo@earth.li>**20090717010817]
421[Use names like '$fOrdInt' for dfuns (and TF instances), rather than '$f21'
422Simon Marlow <marlowsd@gmail.com>**20090716125643
423 Ignore-this: d0b4632cf8ed9e05b67a19aa19ab3e19
424 
425 2 reasons for this:
426   - compilation is more predictable.  Adding or removing an instance
427     is less likely to force unnecessary recompilation due to
428     renumbering other dfun names.
429   - it makes it easier to read Core / C-- / asm
430 
431 The names aren't completely deterministic.  To do that, we'd have to
432 include package and module names, which would make the symbol names
433 long and reduce readability.  So the compromise is that if there's a
434 clash, we disambiguate by adding an integer suffix.  This is fairly
435 unlikely in practice unless you're using overlapping instances.
436 
437 Type family instances are handled in the same way, with the same
438 disambiguation strategy.
439]
440[Use a stable ordering for the export list in the interface
441Simon Marlow <marlowsd@gmail.com>**20090716122601
442 Ignore-this: 847dd7adc8b52e56f28d2478c78c925
443 The export list was ordered according to the whim of FastStrings,
444 which meant that interface fingerprints could change for no good
445 reason, causing spurious recompilation.
446]
447[Don't put all of $CFLAGS into $SRC_CC_OPTS
448Ian Lynagh <igloo@earth.li>**20090716131309
449 Instead, we just put the flags we need in there (e.g. -m64 on OS X 64).
450 This fixes a problem found by Simon M, where we were compiling
451 everything with -g, leading to a bloated RTS.
452]
453[Move showOpt into DynFlags
454Ian Lynagh <igloo@earth.li>**20090716005314]
455[Make the --info values printable with "ghc --print-foo"; trac #3122
456Ian Lynagh <igloo@earth.li>**20090716001718
457 Also, libdir is now part of the --info output, so this subsumes the old
458 --print-libdir flag.
459 The mode parsing was getting rather adhoc, so I've tidied it up a bit
460 in the process.
461]
462[whitespace only
463Simon Marlow <marlowsd@gmail.com>**20090716104217
464 Ignore-this: 38cff291d9ef15c30e3ed685ffc3c9f9
465]
466[refactor: use packageConfigId in place of mkPackageId . package
467Simon Marlow <marlowsd@gmail.com>**20090716104145
468 Ignore-this: f3d73e7bd1b307a67d26585c49f3d89f
469]
470[Fix a flag name in the docs
471Ian Lynagh <igloo@earth.li>**20090714165943]
472[Add the -fno-shared-implib flag
473Ian Lynagh <igloo@earth.li>**20090714165631
474 Patch from
475     Max Bolingbroke <batterseapower@hotmail.com>
476 Rerecorded to avoid conflicts.
477]
478[Derived Foldable instances should use Data.Foldable.foldr
479m.niloc@gmail.com**20090711130647
480 Ignore-this: e3eb841e9535a842a98bb1ae0532c6e8
481]
482[remove Solaris-specific hacks, now unnecessary
483Simon Marlow <marlowsd@gmail.com>**20090713083524
484 Ignore-this: 500077008e463532e0677ee82f5284bb
485]
486[Simplify timestamp restoration
487Matthias Kilian <kili@outback.escape.de>**20090711100244
488 Ignore-this: 7eaede224befa6b5368c91b92366211
489]
490[FIX #3272
491Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090714054559
492 Ignore-this: 225fe4d82d4eed02e9b1377687661bac
493]
494[Fix warnings
495Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713092032
496 Ignore-this: 3631b87164fc54d82e3a02875dc08f7d
497]
498[Separate length from data in DPH arrays
499Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713044212
500 Ignore-this: aa2cc3b5ae43bd2c493ce4b330c883cd
501]
502[Stop using -fno-warn-unused-do-bind when compiling the libraries
503Ian Lynagh <igloo@earth.li>**20090709160422
504 They're now fixed to not generate those warnings
505]
506[Remove maybePrefixMatch, using stripPrefix instead
507Ian Lynagh <igloo@earth.li>**20090709160412
508 We already require GHC 6.8 to build, and that included stripPrefix
509 in Data.List.
510]
511[TFs: FIX #2203 (second half)
512Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710064834
513 Ignore-this: 46a46feaa73f74feb08524b9e7547414
514]
515[TFs: Fix should_compile/Simple8
516Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710042728
517 Ignore-this: 471ab67e3df1c5245921be5286a45f93
518]
519[workaround new Cygwin bash CRLF behaviour
520Simon Marlow <marlowsd@gmail.com>**20090709132850
521 Ignore-this: 5cfa2cc9d776ebe315c0f6ad7ab56d98
522]
523[Use /usr/bin/test if it exists, and fix test syntax.
524Simon Marlow <marlowsd@gmail.com>**20090709124616
525 Ignore-this: 83a75ba7c3ce2a1d02bddb7bfe414bfe
526 Should fix Solaris build failures
527]
528[Allow mixed case pragmas; #1817. Patch from squadette
529Ian Lynagh <igloo@earth.li>**20090709153737
530 This patch allow you to use "Language CPP", or even "LaNgUaGe CPP",
531 if you wish, as the manual claims you can.
532]
533[don't create inplace/bin/ghc-<version>
534Simon Marlow <marlowsd@gmail.com>**20090706092031
535 Ignore-this: 2584d7bf56e77b27ca5b7b557c152c5e
536]
537[Fix ignored-monadic-result warnings
538Ian Lynagh <igloo@earth.li>**20090707181857]
539[Fix an unused import warning
540Ian Lynagh <igloo@earth.li>**20090707144706]
541[Fix unused import warnings
542Ian Lynagh <igloo@earth.li>**20090707143216]
543[Fix unused import warnings
544Ian Lynagh <igloo@earth.li>**20090707133537]
545[When exporting F(..), all the children of F are also exported
546Ian Lynagh <igloo@earth.li>**20090707133427
547 This fixes the unused imports warning when
548     Foo (F(x,y,z))
549 is imported and
550     Foo (F(..))
551 is exported.
552]
553[Remove unused imports
554Ian Lynagh <igloo@earth.li>**20090707121548]
555[Major patch to fix reporting of unused imports
556simonpj@microsoft.com**20090706112503
557 Ignore-this: 3b5ecdd880474493d73bdbdc0fa0b782
558 
559 This patch, joint work between and Ian and Simon, fixes Trac #1074
560 by reporting unused import declarations much more accuratly than
561 before.  The specification is described at
562 
563 http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/UnusedImports
564 
565 The implementation is both easier to understand than before, and shorter
566 too. 
567 
568 Also fixed are #1148, #2267
569 
570 Also fixed is -ddump-minimal imports, which now works properly, fixing
571 Trac #1792.
572 
573 
574]
575[Trim unused imports detected by new unused-import code
576simonpj@microsoft.com**20090706112201
577 Ignore-this: c6ca46d3a750c1cd1d58ea2c0de9f14f
578]
579[Avoid unnecessary recompilation after ./configure (helps #3228)
580Simon Marlow <marlowsd@gmail.com>**20090707085040
581 Ignore-this: f8b3e7a2a96bc23cd29505ab9c8dbd7d
582 We cache the old versions of files generated by configure, so that if
583 configure touches the file without changing it, we can detect that and
584 restore the timestamp.
585]
586[check for tabs in compiler/ghc.cabal.in (#3344)
587Simon Marlow <marlowsd@gmail.com>**20090707081845
588 Ignore-this: 6073db47eafd52e13e76c58ef738afcf
589]
590[remove tabs
591Simon Marlow <marlowsd@gmail.com>**20090707081823
592 Ignore-this: 3d65831fc019f76cefac03291904842a
593]
594[fix cleaning of libraries (now 'make clean' in libraries/* works again)
595Simon Marlow <marlowsd@gmail.com>**20090703114638
596 Ignore-this: b3af731d50ff5bfbd453f94aa40cb92c
597]
598[FIX #2677
599Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090707055442
600 Ignore-this: e224dd09d0d1c9ec4f3b46c7accb8d57
601]
602[Update driver/Makefile for the new build system
603Ian Lynagh <igloo@earth.li>**20090705204041]
604[Fix generational GC bug (#3348)
605Simon Marlow <marlowsd@gmail.com>**20090706112227
606 Ignore-this: 5938338efa0ad1550968c664a5a76f31
607]
608[Windows fixes to build system: use the 'find' and 'sort' found by configure
609simonpj@microsoft.com**20090706103413
610 Ignore-this: a96197917f388a637118bafefb427495
611 
612 The build system should use 'find' and 'sort' that are discovered by
613 configure, not the ones in your path.  On Windows the ones in your path
614 might well be the non-Unixy Windows versions.
615 
616 This patch fixes the ones I tripped over. There may be more.
617 
618]
619[Follow Cabal changes
620Ian Lynagh <igloo@earth.li>**20090705180414]
621[Update TODO list
622Ian Lynagh <igloo@earth.li>**20090705165009]
623[Make -fext-core a dynamic flag (it was a static flag)
624Ian Lynagh <igloo@earth.li>**20090705132420]
625[Update a few points about shared libs in other sections
626Duncan Coutts <duncan@well-typed.com>**20090704212212
627 And add links to the new shared libs section.
628]
629[Document -dynload flag. Also add it and -shared to the flags reference.
630Duncan Coutts <duncan@well-typed.com>**20090704212119]
631[Add new section on using shared libs
632Duncan Coutts <duncan@well-typed.com>**20090704212003]
633[Document foreign import prim in the user guide
634Duncan Coutts <duncan@well-typed.com>**20090704180547
635 Basically just stat that it exists and refer to the ghc dev wiki
636 for the details, because we don't really want people using it.
637]
638[For now, use -fno-warn-unused-do-bind when building the libraries
639Ian Lynagh <igloo@earth.li>**20090704210654]
640[Make changes to -fwarn-unused-do-bind and -fwarn-wrong-do-bind suggested by SPJ
641Max Bolingbroke <batterseapower@hotmail.com>**20090702150943
642 Ignore-this: 595368298d2e11623c0bd280ff89d8de
643]
644[Support for -fwarn-unused-do-bind and -fwarn-wrong-do-bind, as per #3263
645Max Bolingbroke <batterseapower@hotmail.com>**20090701200344
646 Ignore-this: 511117ffc10d4b656e530b751559b8b8
647]
648[Improved infrastructure for fast-rebuilding of parts of the tree
649Simon Marlow <marlowsd@gmail.com>**20090703074527
650 Ignore-this: ab348d0988d8bbc28c2b4babbd6bbfb8
651 
652 e.g.
653 
654   cd compiler
655   make FAST=YES stage1/build/HscTypes.o
656 
657 builds just the specified .o file, without rebuilding dependencies,
658 and omitting some of the makefile phases.  FAST=YES works anywhere, to
659 omit depenencies and phases.  'make fast' is shorthand for 'make
660 all FAST=YES'.
661]
662[Fix Trac #3342: missed zonking in TcHsSyn
663simonpj@microsoft.com**20090702124331
664 Ignore-this: 9b97b2142dfc665b503f59df7c55dd17
665 
666 The type in a ViewPat wasn't being zonked.  Easily fixed.
667 
668]
669[Type synonym families may be nullary
670Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702084826
671 Ignore-this: bcfe6ed62c901206daf5a5088890bbea
672]
673[New syntax for GADT-style record declarations, and associated refactoring
674simonpj@microsoft.com**20090702094657
675 Ignore-this: bd9817230d3773b3b01fae3d7f04c57d
676 
677 The main purpose of this patch is to fix Trac #3306, by fleshing out the
678 syntax for GADT-style record declraations so that you have a context in
679 the type.  The new form is
680    data T a where
681      MkT :: forall a. Eq a => { x,y :: !a } -> T a
682 See discussion on the Trac ticket.
683 
684 The old form is still allowed, but give a deprecation warning.
685 
686 When we remove the old form we'll also get rid of the one reduce/reduce
687 error in the grammar. Hurrah!
688 
689 While I was at it, I failed as usual to resist the temptation to do lots of
690 refactoring.  The parsing of data/type declarations is now much simpler and
691 more uniform.  Less code, less chance of errors, and more functionality.
692 Took longer than I planned, though.
693 
694 ConDecl has record syntax, but it was not being used consistently, so I
695 pushed that through the compiler.
696 
697]
698[White space only
699simonpj@microsoft.com**20090702094627
700 Ignore-this: 19f654cbf371c8dcc6517fd4934855b4
701]
702[Comments only
703simonpj@microsoft.com**20090702094531
704 Ignore-this: 384fc2729c7c50a1680775a1f9ff89e4
705]
706[Look through Notes when matching
707simonpj@microsoft.com**20090702094444
708 Ignore-this: 7daea81e905ec6061d3e0fd588d7e61b
709]
710[FIX #3197
711Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702070905
712 Ignore-this: ebf829f0ae025e82bccdfa4345828ffe
713]
714[Fix #2197 (properly this time)
715Simon Marlow <marlowsd@gmail.com>**20090701122354
716 Ignore-this: 39b6e4b0bcdd8c2f4660f976b7db768d
717 
718 $ ./inplace/bin/ghc-stage2 --interactive
719 GHCi, version 6.11.20090701: http://www.haskell.org/ghc/  :? for help
720 ghc-stage2: GHCi cannot be used when compiled with -prof
721 [1]    32473 exit 1     ./inplace/bin/ghc-stage2 --interactive
722]
723[make GhcProfiled work, and add a "prof" flavour to build.mk
724Simon Marlow <marlowsd@gmail.com>**20090701114211
725 Ignore-this: 386d347e4ad8b6c2bd40a2ba7da31ba6
726 
727 Building a profiled GHC is as simple as adding
728 
729 GhcLibWays += p
730 GhcProfiled = YES
731 
732 to your build.mk and saying 'make'.  Then you have a profiled
733 inplace/bin/ghc-stage2.
734]
735[remove unnecessary $(RM)s
736Simon Marlow <marlowsd@gmail.com>**20090701110609
737 Ignore-this: f326ec8931d0d484a66b67ce1270cc6e
738]
739['make html' in a library builds the Haddock docs
740Simon Marlow <marlowsd@gmail.com>**20090630111137
741 Ignore-this: 781bf10e2d4bca23b7f70c6f0465d120
742]
743[fix GC bug introduced with the C finalizer support
744Simon Marlow <marlowsd@gmail.com>**20090630080834
745 Ignore-this: 3567e3adb5ae4a5dcbce81733487f348
746]
747[Add a configure test for whether or not __mingw_vfprintf exists
748Ian Lynagh <igloo@earth.li>**20090627150501]
749[Fix #3319, and do various tidyups at the same time
750Simon Marlow <marlowsd@gmail.com>**20090626095421
751 Ignore-this: ea54175f6bd49e101d7b33392764f643
752  - converting a THSyn FFI declaration to HsDecl was broken; fixed
753  - pretty-printing of FFI declarations was variously bogus; fixed
754  - there was an unused "library" field in CImport; removed
755]
756[rename cache variable to keep recent autoconfs happy
757Ross Paterson <ross@soi.city.ac.uk>**20090626131410
758 Ignore-this: 187091bbe78f2b14402162acfb98180f
759]
760[TAG 2009-06-25
761Ian Lynagh <igloo@earth.li>**20090625155528]
762Patch bundle hash:
7638fefa6ab08626da662d19cefe2469e73f3b2796f