Ticket #1933: timer-create.patch

File timer-create.patch, 24.0 KB (added by simonmar, 7 years ago)
2New patches:
4[check that timer_settime() really delivers a signal
5Simon Marlow <[email protected]>**20071129094759] {
6hunk ./aclocal.m4 1035
7-  [AC_CACHE_CHECK([for a working timer_create(CLOCK_REALTIME)],
8+  [AC_CACHE_CHECK([for a working timer_create()],
9hunk ./aclocal.m4 1038
10+#ifdef HAVE_STDLIB_H
11+#include <stdlib.h>
13hunk ./aclocal.m4 1050
14+static volatile int tock = 0;
15+static void handler(int i)
17+   tock = 1;
20hunk ./aclocal.m4 1061
21+    struct itimerspec it;
22+    struct sigaction action;
23+    unsigned int i;
25hunk ./aclocal.m4 1070
26+    if (timer_create(CLOCK_REALTIME, &ev, &timer) != 0) {
27+       exit(2);
28+    }
30+    action.sa_handler = handler;
31+    action.sa_flags = 0;
32+    sigemptyset(&action.sa_mask);
33+    if (sigaction(SIGVTALRM, &action, NULL) == -1) {
34+        exit(3);
35+    }
37+    it.it_value.tv_sec = 0;
38+    it.it_value.tv_nsec = 1;
39+    it.it_interval = it.it_value;
40+    if (timer_settime(timer, 0, &it, NULL) != 0) {
41+        exit(4);
42+    }
44+    usleep(100);
45+    if (tock) exit(0);
46+    exit(5);
47hunk ./aclocal.m4 1092
48-    exit(1)
49+    exit(6);
50hunk ./aclocal.m4 1102
51-                   [Define to 1 if we can use timer_create(CLOCK_PROCESS_CPUTIME_ID,...)]);;
52+                   [Define to 1 if we can use timer_create(),...)]);;
57[FIX Trac #1935: generate superclass constraints for derived classes
58[email protected]**20071128150541
60 This bug only reports a problem with phantom types, but actually
61 there was quite a long-standing and significant omission in the
62 constraint generation for derived classes.  See
63 Note [Superclasses of derived instance] in TcDeriv.
65 The test deriving-1935 tests both cases.
69[Print a bit more info in VarBinds (no need to merge)
70[email protected]**20071128150354]
71[Check for duplicate bindings in CoreLint
72[email protected]**20071128150214]
73[Rebuild utils with the stage1 compiler when making a bindist; fixes trac #1860
74Ian Lynagh <[email protected]>**20071127203959
75 This is a bit unpleasant, as "make binary-dist" really shouldn't actually
76 build anything, but it works.
78[Remove the --print-docdir flag
79Ian Lynagh <[email protected]>**20071127195605
80 It wasn't doing the right thing for bindists. Let's rethink...
82[add comment
83Simon Marlow <[email protected]>**20071128111417]
84[FIX #1916: don't try to convert float constants to int in CMM optimizer
85Bertram Felgenhauer <[email protected]>**20071122095513]
86[give a more useful message when the static flags have not been initialised (#1938)
87Simon Marlow <[email protected]>**20071127135435]
88[canonicalise the path to HsColour
89Simon Marlow <[email protected]>**20071126141614]
90[Consistently put www. on the front of haskell.org in URLs
91Ian Lynagh <[email protected]>**20071126215256]
92[Fix some more URLs
93Ian Lynagh <[email protected]>**20071126214147]
94[Tweak some URLs
95Ian Lynagh <[email protected]>**20071126194148]
96[Fix some links
97Ian Lynagh <[email protected]>**20071126184406]
98[FIX #1925: the interpreter was not maintaining tag bits correctly
99Simon Marlow <[email protected]>**20071127122614
100 See comment for details
102[add missing instruction: ALLOC_AP_NOUPD
103Simon Marlow <[email protected]>**20071127122604]
104[Copy gmp stamps into bindists, so we don't try and rebuild gmp
105Ian Lynagh <[email protected]>**20071125211919]
106[On Windows, Delete the CriticalSection's we Initialize
107Ian Lynagh <[email protected]>**20071125125845]
108[On Windows, add a start menu link to the flag reference
109Ian Lynagh <[email protected]>**20071125124429]
110[Remove html/ from the paths we put in the start menu on Windows
111Ian Lynagh <[email protected]>**20071125124150]
112[MERGED: Make ":" in GHCi repeat the last command
113Ian Lynagh <[email protected]>**20071125122020
114 Ian Lynagh <[email protected]>**20071124231857
115  It used to be a synonym for ":r" in 6.6.1, but this wasn't documented or
116  known about by the developers. In 6.8.1 it was accidentally broken.
117  This patch brings it back, but as "repeat the last command", similar to
118  pressing enter in gdb. This is almost as good for people who want it to
119  reload, and means that it can also be used to repeat commands like :step.
121[MERGED: Put library docs in a $pkg, rather than $pkgid, directory; fixes trac #1864
122Ian Lynagh <[email protected]>**20071124212305
123 Ian Lynagh <[email protected]>**20071124171220
125[Don't make a library documentation prologue
126Ian Lynagh <[email protected]>**20071124211943
127 It's far too large now, and no-one complained when 6.8.1 didn't have one.
129[Don't put package version numbers in links in index.html
130Ian Lynagh <[email protected]>**20071124211629]
131[Define install-strip in Makefile
132Ian Lynagh <[email protected]>**20071124205037]
133[Define install-strip in distrib/Makefile
134Ian Lynagh <[email protected]>**20071124204803]
135[Install gmp from bindists; fixes trac #1848
136Ian Lynagh <[email protected]>**20071124185240]
137[(native gen) fix code generated for GDTOI on x86_32
138Bertram Felgenhauer <[email protected]>**20071121063942
139 See trac #1910.
141[Copy the INSTALL hack from mk/config.mk.in into distrib/Makefile-bin-vars.in
142Ian Lynagh <[email protected]>**20071124163028
143 configure will set INSTALL to ./install-sh if it can't find it in the path,
144 so we need to replace the . with the path to our root.
146[Make install-sh executable /before/ we try to find it
147Ian Lynagh <[email protected]>**20071124162450]
148[Document --info in the +RTS -? help
149Ian Lynagh <[email protected]>**20071123204352]
150[MERGED: If we have hscolour then make source code links in teh haddock docs
151Ian Lynagh <[email protected]>**20071123233113
152 Fri Nov 23 13:15:59 PST 2007  Ian Lynagh <[email protected]>
154[Check tag bits on the fun pointer of a PAP
155Simon Marlow <[email protected]>**20071126160420]
156[Tidy and trim the type environment in mkBootModDetails
157[email protected]**20071123153519
159 Should fix Trac #1833
161 We were failing to trim the type envt in mkBootModDetails, so several
162 functions all called (*), for example, were getting into the interface.
163 Result chaos.  It only actually bites when we do the retyping-loop thing,
164 which is why it's gone so long without a fix.
168[refactor: HscNothing and boot modules do not need desugaring
169Simon Marlow <[email protected]>**20071123135237]
170[Properly ppr InstEqs in wanteds of implication constraints
171Manuel M T Chakravarty <[email protected]>**20071122093002]
172[FIX #1910: fix code generated for GDTOI on x86_32
173Bertram Felgenhauer <[email protected]>*-20071121102627]
174[Add built-in Double operations to vectorisation
175Roman Leshchinskiy <[email protected]>**20071122002517]
176[Teach vectorisation about Double
177Roman Leshchinskiy <[email protected]>**20071121054932]
178[Vectorise polyexprs with notes
179Roman Leshchinskiy <[email protected]>**20071121053102]
180[Make rebindable do-notation behave as advertised
181[email protected]**20071121174914
183 Adopt Trac #1537.  The patch ended up a bit bigger than I expected,
184 so I suggest we do not merge this into the 6.8 branch.  But there
185 is no funadamental reason why not.
187 With this patch, rebindable do-notation really does type as if you
188 had written the original (>>) and (>>=) operations in desguared form.
190 I ended up refactoring some of the (rather complicated) error-context
191 stuff in TcUnify, by pushing an InstOrigin into tcSubExp and its
192 various calls. That means we could get rid of tcFunResTy, and the
193 SubCtxt type.  This should improve error messages slightly
194 in complicated situations, because we have an Origin to hand
195 to instCall (in the (isSigmaTy actual_ty) case of tc_sub1).
197 Thanks to Pepe for the first draft of the patch.
200[Add DEBUG-only flag -dsuppress-uniques to suppress printing of uniques
201[email protected]**20071116152446
203 This is intended only for debugging use: it makes it easier to
204 compare two variants without the variations between uniques mattering.
206 (Of course, you can't actually feed the output to the C compiler
207 or assembler and expect anything sensible to happen!)
210[Add -dcore-lint when validating libraries
211[email protected]**20071105164733]
212[Fix Trac #1913: check data const for derived types are in scope
213[email protected]**20071121151428
215 When deriving an instance, the data constructors should all be in scope.
216 This patch checks the condition.
220[FIX #1910: fix code generated for GDTOI on x86_32
221Bertram Felgenhauer <[email protected]>**20071121102627]
222[Fix Trac #1909: type of map in docs
223[email protected]**20071120160152]
224[Move file locking into the RTS, fixing #629, #1109
225Simon Marlow <[email protected]>**20071120140859
226 File locking (of the Haskell 98 variety) was previously done using a
227 static table with linear search, which had two problems: the array had
228 a fixed size and was sometimes too small (#1109), and performance of
229 lockFile/unlockFile was suboptimal due to the linear search.
230 Also the algorithm failed to count readers as required by Haskell 98
231 (#629).
233 Now it's done using a hash table (provided by the RTS).  Furthermore I
234 avoided the extra fstat() for every open file by passing the dev_t and
235 ino_t into lockFile.  This and the improvements to the locking
236 algorithm result in a healthy 20% or so performance increase for
237 opening/closing files (see openFile008 test).
239[FIX Trac #1825: standalone deriving Typeable
240[email protected]**20071120125732
242 Standalone deriving of typeable now requires you to say
243        instance Typeable1 Maybe
244 which is exactly the shape of instance decl that is generated
245 by a 'deriving( Typeable )' clause on the data type decl.
247 This is a bit horrid, but it's the only consistent way, at least
248 for now.  If you say something else, the error messages are helpful.
250 MERGE to 6.8 branch
253[FIX #1715: egregious bug in ifaceDeclSubBndrs
254[email protected]**20071120111723
256 ifaceDeclSubBndrs didn't have an IfaceSyn case; but with type
257 families an IfaceSyn can introduce subordinate binders.  Result:
258 chaos.
260 The fix is easy though.  Merge to 6.8 branch.
264[Always do 'setup makefile' before building each library
265Simon Marlow <[email protected]>**20071120103329
266 This forces preprocessing to happen, which is necessary if any of the
267 .hsc files have been modified.  Without this change, a 'setup
268 makefile' would be required by hand after a .hsc file changed.
269 Fortunately 'setup makefile' isn't much extra work, and I've made it
270 not overwrite GNUmakefile if it hasn't changed, which avoids
271 recalculating the dependencies each time.
273[FIX #1847 (improve :browse! docs, fix unqual)
274[email protected]**20071108013147
276 - add example to docs, explain how to interpret
277   output of `:browse! Data.Maybe`
278 - print unqualified names according to current
279   context, not the context of the target module
282[Track changes to package ndp
283Roman Leshchinskiy <[email protected]>**20071120033716]
284[Temporary hack for passing PArrays from unvectorised to vectorised code
285Roman Leshchinskiy <[email protected]>**20071120024545]
286[Bind NDP stuff to [:.:] arrays
287Roman Leshchinskiy <[email protected]>**20071119020302]
288[Don't treat enumerations specially during vectorisation for the moment
289Roman Leshchinskiy <[email protected]>**20071119013729]
290[Fix bugs in vectorisation of case expressions
291Roman Leshchinskiy <[email protected]>**20071119013714]
292[More built-in NDP combinators
293Roman Leshchinskiy <[email protected]>**20071119012205]
294[New vectorisation built-ins
295Roman Leshchinskiy <[email protected]>**20071118051940]
296[Fix bug in conversion unvect/vect
297Roman Leshchinskiy <[email protected]>**20071118051926]
298[Extend built-in vectorisation environments
299Roman Leshchinskiy <[email protected]>**20071118045219]
300[Fix bug in generation of environments for vectorisation
301Roman Leshchinskiy <[email protected]>**20071118045203]
302[Add builtin var->var mapping to vectorisation
303Roman Leshchinskiy <[email protected]>**20071118042605]
304[Extend vectorisation built-in mappings with datacons
305Roman Leshchinskiy <[email protected]>**20071118034351]
306[Change representation of parallel arrays of enumerations
307Roman Leshchinskiy <[email protected]>**20071118033355]
308[Add vectorisation-related builtin
309Roman Leshchinskiy <[email protected]>**20071118031513]
310[Teach vectorisation about Bool
311Roman Leshchinskiy <[email protected]>**20071117042714]
312[Incomplete support for boxing during vectorisation
313Roman Leshchinskiy <[email protected]>**20071117040739]
314[Make sure some TyCons always vectorise to themselves
315Roman Leshchinskiy <[email protected]>**20071117040537]
316[Simple conversion vectorised -> unvectorised
317Roman Leshchinskiy <[email protected]>**20071117023029]
318[Fix bug in case vectorisation
319Roman Leshchinskiy <[email protected]>**20071117015014]
320[Vectorisation of algebraic case expressions
321Roman Leshchinskiy <[email protected]>**20071116074814]
322[More vectorisation-related built-ins
323Roman Leshchinskiy <[email protected]>**20071116061831]
324[Vectorisation utilities
325Roman Leshchinskiy <[email protected]>**20071116051037]
326[Add vectorisation built-ins
327Roman Leshchinskiy <[email protected]>**20071116050959]
328[Fix vectorisation of binders in case expressions
329Roman Leshchinskiy <[email protected]>**20071116021833]
330[Two small typos in the flags summary (merge to 6.8 branch)
331[email protected]**20071119134639]
332[Improve the situation for Trac #959: civilised warning instead of a trace msg
333[email protected]**20071119122938
335 This doesn't fix the root cause of the bug, but it makes the report
336 more civilised, and points to further info.
339[FIX Trac #1806: test for correct arity for datacon in infix pattern patch
340[email protected]**20071119114301
342 Happily the fix is easy; pls merge
345[Accept x86_64-*-freebsd* as well as amd64-*-freebsd* in configure.ac
346Ian Lynagh <[email protected]>**20071117154502
347 Patch from Brian P. O'Hanlon
349[Attempt at fixing #1873, #1360
350Simon Marlow <[email protected]>**20071116152148
352 I think I figured out a reasonable way to manage the GHCi context,
353 comments welcome.
355 Rule 1: external package modules in the context are persistent.  That
356 is, when you say 'import Data.Maybe' it survives over :load, :add,
357 :reload and :cd.
359 Rule 2: :load and :add remove all home-package modules from the
360 context and add the rightmost target, as a *-module if possible.  This
361 is as before, and makes sense for :load because we're starting a new
362 program; the old home-package modules don't make sense any more.  For
363 :add, it usually does what you want, because the new target will
364 become the context.
366 Rule 3: any modules from the context that fail to load during a
367 :reload are remembered, and re-added to the context at the next
368 successful :reload.
370 Claus' suggestion about adding the "remembered" modules to the prompt
371 prefixed with a ! is implemented but commented out.  I couldn't
372 decide whether it was useful or confusing.
374 One difference that people might notice is that after a :reload where
375 there were errors, GHCi would previously dump you in the most recent
376 module that it loaded.  Now it dumps you in whatever subset of the
377 current context still makes sense, and in the common case that will
378 probably be {Prelude}.
380[Wibble to fix Trac #1901 (shorten messsage slightly)
381[email protected]**20071116150341]
382[Improve links from flag reference to the relevant section; and improve doc of RankN flags
383[email protected]**20071116145816]
384[FIX Trac #1901: check no existential context in H98 mode
385[email protected]**20071116145609]
386[Improve documentation of data type declarations (Trac #1901)
387[email protected]**20071116081841]
388[Change the command-line semantics for query commands
389Simon Marlow <[email protected]>**20071116132046
391 From the help text:
393   Commands that query the package database (list, latest, describe,
394   field) operate on the list of databases specified by the flags
395   --user, --global, and --package-conf.  If none of these flags are
396   given, the default is --global --user.
398 This makes it possible to query just a single database (e.g. the
399 global one without the user one), which needed tricks to accomplish
400 before.
402[use "ghc-pkg latest --global" instead of "ghc-pkg list --simple-output"
403Simon Marlow <[email protected]>**20071116122018
404 The former now does the right thing: it uses the global database only,
405 and picks the most recent package with the given name.
407[Disallow installing packages whose names  differ in case only.
408Simon Marlow <[email protected]>**20071116121153
409 --force overrides.  Requested by Duncan Coutts, with a view to
410 treating package names as case-insensitive in the future.
412[FIX BUILD (with GHC 6.2.x): update .hi-boot file
413Simon Marlow <[email protected]>**20071116101227]
414[FIX #1828: installing to a patch with spaces in
415Simon Marlow <[email protected]>**20071115155747
416 We have to pass the path to gcc when calling windres, which itself
417 might have spaces in.  Furthermore, we have to pass the path to gcc's
418 tools to gcc.  This means getting the quoting right, and after much
419 experimentation and reading of the windres sources I found something
420 that works: passing --use-temp-files to windres makes it use its own
421 implementation of quoting instead of popen(), and this does what we
422 want.  Sigh.
424[on Windows, install to a directory with spaces (test for #1828)
425Simon Marlow <[email protected]>**20071115155327]
426[FIX #1679: crash on returning from a foreign call
427Simon Marlow <[email protected]>**20071115131635
428 We forgot to save a pointer to the BCO over the foreign call.  Doing
429 enough allocation and GC during the call could provoke a crash.
431[Avoid the use of unversioned package dependencies
432Simon Marlow <[email protected]>**20071115103249
433 Fortunately "ghc-pkg list $pkg --simple-output" is a good way to add
434 the version number.
436[FIX #1596 (remove deprecated --define-name)
437Simon Marlow <[email protected]>**20071114165323
438 Also remove the old command-line syntax for ghc-pkg, which was not
439 documented.  Do not merge.
441[FIX #1837: remove deprecated support for unversioned dependencies (do not merge)
442Simon Marlow <[email protected]>**20071114161044
446Pepe Iborra <[email protected]>**20071114233356]
447[Make pprNameLoc more robust in absence of loc information
448Pepe Iborra <[email protected]>**20071114233343]
449[Try to manage the size of the text rendered for ':show bindings'
450Pepe Iborra <[email protected]>**20071114231601]
451[Make the Term ppr depth aware
452Pepe Iborra <mnislaih@gmail.com>**20071114183417]
453[Use paragraph fill sep where possible
454Pepe Iborra <mnislaih@gmail.com>**20071114181233]
455[Make SpecConstr work again
458 In a typo I'd written env instead of env', and as a result RULES are
459 practically guaranteed not to work in a recursive group.  This pretty
460 much kills SpecConstr in its tracks!
462 Well done Kenny Lu for spotting this.  The fix is easy.
464 Merge into 6.8 please.
469[Documentation only - fix typo in flags reference
470Tim Chevalier <chevalier@alum.wellesley.edu>**20071115055748]
471[Avoid making Either String an instance of Monad in the Haddock parser
472David Waern <david.waern@gmail.com>**20071114204050]
473[FIX 1463 (implement 'ghc-pkg find-module')
476 - the ticket asks for a module2package lookup in ghc-pkg
477   (this would be useful to have in cabal, as well)
479 - we can now ask which packages expose a module we need,
480   eg, when preparing a cabal file or when getting errors
481   after package reorganisations:
483   $ ./ghc-pkg-inplace find-module Var
484   c:/fptools/ghc/driver/package.conf.inplace:
485       (ghc-6.9.20071106)
487   $ ./ghc-pkg-inplace find-module Data.Sequence
488   c:/fptools/ghc/driver/package.conf.inplace:
489       containers-0.1
491 - implemented as a minor variation on listPackages
493 (as usual, it would be useful if one could combine
494 multiple queries into one)
497[remove --define-name from the --help usage message (#1596)
498Simon Marlow <simonmar@microsoft.com>**20071114153417
501[FIX #1837: emit deprecated message for unversioned dependencies
502Simon Marlow <simonmar@microsoft.com>**20071114153010]
503[Fix #782, #1483, #1649: Unicode GHCi input
504Simon Marlow <simonmar@microsoft.com>**20071114151411
505 GHCi input is now treated universally as UTF-8, except for the Windows
506 console where we do the correct conversion from the current code
507 page (see System.Win32.stringToUnicode).
509 That leaves non-UTF-8 locales on Unix as unsupported, but (a) we only
510 accept source files in UTF-8 anyway, and (b) UTF-8 is quite ubiquitous
511 as the default locale.
514[Fix build
515David Waern <david.waern@gmail.com>**20071114125842
516 I had forgot to update HaddockLex.hi-boot-6, so the build with 6.2.2
517 failed. This fixes that.
519[FIX Trac 1662: actually check for existentials in proc patterns
522 I'd fixed the bug for code that should be OK, but had forgotten to
523 make the test for code that should be rejected!
525 Test is arrowfail004
528[FIX Trac 1888; duplicate INLINE pragmas
531 There are actually three things here
532 - INLINE pragmas weren't being pretty-printed properly
533 - They were being classified into too-narrow boxes by eqHsSig
534 - They were being printed in to much detail by hsSigDoc
536 All easy.  Test is rnfail048.
539[Run the -frule-check pass more often (when asked)
541[GHCi debugger: added a new flag, -fno-print-binding-contents
542Pepe Iborra <mnislaih@gmail.com>**20071113174539
544 The contents of bindings show at breakpoints and by :show bindings
545 is rendered using the same printer that :print uses.
546 But sometimes the output it gives spans over too many lines and the
547 user may want to be able to disable it.
549[Fix Trac 1865: GHCi debugger crashes with :print
550Pepe Iborra <mnislaih@gmail.com>**20071113170113]
551[Replaced two uses of head b explicit pattern matching
552Pepe Iborra <mnislaih@gmail.com>**20071013113136]
553[Print binding contents in :show bindings
554Pepe Iborra <mnislaih@gmail.com>**20071006123952]
555[ Leftovers from the 1st GHCi debugger prototype
556Pepe Iborra <mnislaih@gmail.com>**20071004204718]
557[Following an indirection doesn't count as a RTTI step
558Pepe Iborra <mnislaih@gmail.com>**20070928091941]
559[FIX #1653 (partially): add -X flags to completion for :set
560Simon Marlow <simonmar@microsoft.com>**20071113153257]
561[Merge from Haddock: Add <<url>> for images
562David Waern <david.waern@gmail.com>**20071112220537
563 A merge of this patch:
565   Mon Aug  7 16:22:14 CEST 2006  Simon Marlow <simonmar@microsoft.com>
566     * Add <<url>> for images
567     Submitted by: Lennart Augustsson
569 Please merge to the 6.8.2 branch.
571[Improve documentation of INLINE, esp its interactions with other transformations
573[Comment re Trac #1220
575[Merge from Haddock: Modify lexing of /../
576David Waern <david.waern@gmail.com>**20071112023856
578   Tue Aug 28 11:19:54 CEST 2007  Simon Marlow <simonmar@microsoft.com>
579     * Modify lexing of /../
580     This makes /../ more like '..', so that a single / on a line doesn't
581     trigger a parse error.  This should reduce the causes of accidental
582     parse errors in Haddock comments; apparently stray / characters are
583     a common source of failures.
585 Please merge this to the 6.8.2 branch.
587[Merge from Haddock: allow blank lines inside code blocks
588David Waern <david.waern@gmail.com>**20071112013439
590   Tue Jan  9 14:14:34 CET 2007  Simon Marlow <simonmar@microsoft.com>
591     * allow blank lines inside a @...@ code block
593 Please merge this to the 6.8.2 branch
595[Merge of a patch from the old Haddock branch:
596David Waern <david.waern@gmail.com>**20071112013143
598   Fri Jan  5 12:13:41 CET 2007  Simon Marlow <simonmar@microsoft.com>
599     * Fix up a case of extra vertical space after a code block
601 Please merge this to the 6.8.2 branch
603[Remove ex-extralibs from libraries/Makefile
604Ian Lynagh <igloo@earth.li>**20071111213618]
605[Remove the X11 and HGL libraries from extralibs
606Ian Lynagh <igloo@earth.li>**20071111213447
607 Don Stewart, X11 maintainer, requested we remove X11, and HGL depends on it
608 on Linux (and we don't try to build HGL on Windows).
610[arrows is no longer an extralib
611Ian Lynagh <igloo@earth.li>**20071027123656]
612[Turn -fprint-bind-result off by default
613Ian Lynagh <igloo@earth.li>**20071111001126]
614[TAG 2007-11-11
615Ian Lynagh <igloo@earth.li>**20071111161540]
616Patch bundle hash: