Ticket #2298: MoveFileEx.dpatch

New patches:

[Simplify implementation of __hscore_renameFile to be NT+ only
Duncan Coutts <duncan@haskell.org>**20080520172055
 Remove a bunch of hacky workarounds for Win9x systems.
 There is also a behaviour change for NT systems: previously the
 code also tried a workaround even on NT systems if calling
 MoveFileEx() failed. It would instead try to delete the target
 file and have another go at renaming. This seems most unwise as
 it looses any atomicity guarantee that MoveFileEx() provides.
] {
hunk ./cbits/dirUtils.c 79
- * the target if it exists (the MS CRT implementation of rename() returns
- * an error
+ * the target if it exists. The MS CRT implementation of rename() returns
+ * an error if the target file exists so we have to use MoveFileEx().
hunk ./cbits/dirUtils.c 87
-    static int forNT = -1;
-    
-    /* ToDo: propagate error codes back */
-    if (MoveFileA(src, dest)) {
+    if (MoveFileExA(src, dest, MOVEFILE_REPLACE_EXISTING)) {
hunk ./cbits/dirUtils.c 90
-       ;
-    }
-    
-    /* Failed...it could be because the target already existed. */
-    if ( !GetFileAttributes(dest) ) {
-       /* No, it's not there - just fail. */
-       maperrno();
-       return (-1);
-    }
-
-    if (forNT == -1) {
-       OSVERSIONINFO ovi;
-       ovi.dwOSVersionInfoSize = sizeof(ovi);
-       if ( !GetVersionEx(&ovi) ) {
-           maperrno();
-           return (-1);
-       }
-       forNT = ((ovi.dwPlatformId & VER_PLATFORM_WIN32_NT) != 0);
-    }
-    
-    if (forNT) {
-       /* Easy, go for MoveFileEx() */
-       if ( MoveFileExA(src, dest, MOVEFILE_REPLACE_EXISTING) ) {
-           return 0;
-       } else {
-           maperrno();
-           return (-1);
-       }
-    }
-
-    /* No MoveFileEx() for Win9x, try deleting the target. */
-    /* Similarly, if the MoveFile*() ops didn't work out under NT */
-    if (DeleteFileA(dest)) {
-       if (MoveFileA(src,dest)) {
-           return 0;
-       } else {
-           maperrno();
-           return (-1);
-       }
-    } else {
}

Context:

[Avoid calling varargs functions using the FFI
Simon Marlow <marlowsd@gmail.com>**20080509145334
 Calling varargs functions is explicitly deprecated according to the
 FFI specification.  It used to work, just about, but it broke with the
 recent changes to the via-C backend to not use header files.
] 
[Add comments about why rotate has an INLINE
simonpj@microsoft.com**20080502074137] 
[Inline Data.Bits.rotate@Int, enables rotate to be constant folded
Don Stewart <dons@galois.com>**20080501230152
 
 All other Bits instances seem to inline well enough on their own
 to enable constant folding, e.g.
 
     sumU . mapU (`shift` 3) . replicateU 10000000 $ (7 :: Int)
 
 goes to:
 
     Main.$wfold =
       \ (ww_sOb :: Int#) (ww1_sOf :: Int#) ->
         case ww1_sOf of wild_XM {
           __DEFAULT -> Main.$wfold (+# ww_sOb 56) (+# wild_XM 1);
           10000000 -> ww_sOb
         }
 
 With this patch, rotate gets inlined and folded too,
 
     sumU . mapU (`rotate` 3) . replicateU 10000000 $ (7 :: Int)
 
 to:
 
   Main.$wfold =
     \ (ww_sO7 :: Int#) (ww1_sOb :: Int#) ->
       case ww1_sOb of wild_XM {
         __DEFAULT -> Main.$wfold (+# ww_sO7 56) (+# wild_XM 1);
         10000000 -> ww_sO7
 
 Whereas it was left as a call to $wrotate before.
     
 
] 
[Moved def. of emptyP 
keller@cse.unsw.edu.au**20080501024223] 
[emptyP def to gHC.PArr
keller@cse.unsw.edu.au**20080501012547] 
[Reexport (>>>) and (<<<) from Control.Arrow. Preserves API compatibility
Don Stewart <dons@galois.com>**20080430231055] 
[Add realToFrac RULE comments from patch message into the source code
simonpj@microsoft.com**20080430214355] 
[Add RULES for realToFrac from Int.
Don Stewart <dons@galois.com>**20080430174126
 
     {-# RULES
     "realToFrac/Int->Double"    realToFrac   = int2Double
     "realToFrac/Int->Float"     realToFrac   = int2Float
       #-}
 
 Note that this only matters for realToFrac. If you've been using
 fromIntegral to promote Int to Doubles, things should be fine as they are.
 
 The following program, using stream fusion to eliminate arrays:
 
     import Data.Array.Vector
 
     n = 40000000
 
     main = do
           let c = replicateU n (2::Double)
               a = mapU realToFrac (enumFromToU 0 (n-1) ) :: UArr Double
           print (sumU (zipWithU (*) c a))
 
 Yields this loop body without the RULE:
 
     case $wtoRational sc_sY4 of ww_aM7 { (# ww1_aM9, ww2_aMa #) ->
     case $wfromRat ww1_aM9 ww2_aMa of tpl_X1P { D# ipv_sW3 ->
     Main.$s$wfold
       (+# sc_sY4 1)
       (+# wild_X1i 1)
       (+## sc2_sY6 (*## 2.0 ipv_sW3))
 
 And with the rule:
 
    Main.$s$wfold
       (+# sc_sXT 1)
       (+# wild_X1h 1)
       (+## sc2_sXV (*## 2.0 (int2Double# sc_sXT)))
 
 The running time of the program goes from 120 seconds to 0.198 seconds
 with the native backend, and 0.143 seconds with the C backend.
 
 
 And just so I don't forget, here's the difference in resulting
 assembly (x86_64), between the native code generator, and the
 C backend. 
 
 -fasm
 
     Main_zdszdwfold_info:
       movq %rdi,%rax
       cmpq $40000000,%rax
       jne .LcZK
       jmp *(%rbp)
     .LcZK:
       cmpq $39999999,%rsi
       jg .LcZN
       cvtsi2sdq %rsi,%xmm0
       mulsd .LnZP(%rip),%xmm0
       movsd %xmm5,%xmm7
       addsd %xmm0,%xmm7
       incq %rax
       incq %rsi
       movq %rax,%rdi
       movsd %xmm7,%xmm5
       jmp Main_zdszdwfold_info
 
 With the C backend we get the even better assembly, (-fvia-C -optc-O3)
 
     Main_zdszdwfold_info:
         cmpq    $40000000, %rdi
         je  .L9
     .L5:
         cmpq    $39999999, %rsi
         jg  .L9
         cvtsi2sdq   %rsi, %xmm0
         leaq    1(%rdi), %rdi
         addq    $1, %rsi
         addsd   %xmm0, %xmm0
         addsd   %xmm0, %xmm5
         jmp Main_zdszdwfold_info
     .L9:
         jmp *(%rbp)
 
 So might make a useful test once the native codegen project starts up.
 
 
] 
[Just (-0/1) is now printed as Just (-0.0), not Just -0.0; trac #2036
Ian Lynagh <igloo@earth.li>**20080427133230] 
[record libraries@haskell.org as maintainer
Ross Paterson <ross@soi.city.ac.uk>**20080425095200] 
[Added emptyP def
keller@cse.unsw.edu.au**20080424021403] 
[don't set O_NONBLOCK on FDs passed to fdToHandle
Simon Marlow <simonmarhaskell@gmail.com>**20080422204719] 
[Improve docs for unsafeCoerce
Malcolm.Wallace@cs.york.ac.uk**20080422130547
 Make it clear that compilers differ.  Point to GHC docs in GHC.Base, and
 add a short description of nhc98's representation-safe conversions.
] 
[Spelling only
simonpj@microsoft.com**20080422082033] 
[In docs for unsafeCoerce, point to unsafeCoerce#
simonpj@microsoft.com**20080421152251] 
[Turn off the gcd/lcm optimisations for Integer for now
Ian Lynagh <igloo@earth.li>**20080418190238
 This makes it easier to experiment with other implementations
] 
[Ordering has now moved to ghc-prim
Ian Lynagh <igloo@earth.li>**20080412100741] 
[fix types for __hscore_st_dev() and __hscore_st_ino()
Simon Marlow <simonmarhaskell@gmail.com>**20080319174101] 
[give an absolute path to 'harch'
Malcolm.Wallace@cs.york.ac.uk**20080327142912] 
[Move Word64/Int64/Word32/Int32 primitives into ghc-prim
Ian Lynagh <igloo@earth.li>**20080325202634] 
[Update .darcs-boring
Ian Lynagh <igloo@earth.li>**20080324205808
 GHC/Prim.hs, GHC/PrimopWrappers.hs are no longer generated in this package
] 
[Remove GHC.PrimopWrappers from base's exposed modules list
Ian Lynagh <igloo@earth.li>**20080324205734] 
[base now uses build-type: Configure
Ian Lynagh <igloo@earth.li>**20080323191146] 
[Move Integer out into its own package
Ian Lynagh <igloo@earth.li>**20080323181342
 We now depend on the new integer package.
 We also depend on a new ghc-prim package, which has GHC.Prim,
 GHC.PrimopWrappers, and new modules GHC.Bool and GHC.Generics,
 containing Bool and Unit/Inl/Inr respectively.
] 
[List extensions used rather than using the -fglasgow-exts hammer
Ian Lynagh <igloo@earth.li>**20080322133622] 
[Remove a gratuitous pattern type sig
Ian Lynagh <igloo@earth.li>**20080321011356] 
[An even better definition for (^) (trac #1687)
Ian Lynagh <igloo@earth.li>**20080320003957] 
[Replace (^) with a faster variant (from trac #1687)
Ian Lynagh <igloo@earth.li>**20080318003017] 
[Add partitionEithers, lefts, and rights.
Ian Lynagh <igloo@earth.li>**20080314193037
 Patch from Russell O'Connor, trac proposal #974.
] 
[System.Console.GetOpt mistakenly rejects options as ambiguous.
Malcolm.Wallace@cs.york.ac.uk**20080312111047
 From "Eelis van der Weegen" <haskell-libs@contacts.eelis.net>.
 Testcase:
 
 > import System.Console.GetOpt
 >
 > type Color = String
 >
 > optsDesc :: [OptDescr Color]
 > optsDesc = [Option "" ["color", "colour"]
 >                    (ReqArg id "color") "Foreground color"]
 >
 > main = do
 >   let args = ["--col=blue"]
 >   case getOpt RequireOrder optsDesc args of
 >     (_, _, err:_) -> putStrLn err
 >     _ -> return ()
 
 Output:
   option `--col' is ambiguous; could be one of:
       --color=color, --colour=color  Foreground color
       --color=color, --colour=color  Foreground color
 
 This error is silly, because the two alternatives listed are the same
 option.  The problem is caused by incorrect use of a generator in a list
 comprehension.
] 
[untabify
Don Stewart <dons@galois.com>**20080310005455] 
[untabify
Don Stewart <dons@galois.com>**20080308014256] 
[untabify
Don Stewart <dons@galois.com>**20080308014129] 
[untabify
Don Stewart <dons@galois.com>**20080308014040] 
[untabify
Don Stewart <dons@galois.com>**20080308013556] 
[untabify
Don Stewart <dons@galois.com>**20080308012457] 
[untabify
Don Stewart <dons@galois.com>**20080308012059] 
[untabify
Don Stewart <dons@galois.com>**20080307192727] 
[untabify
Don Stewart <dons@galois.com>**20080305033712] 
[untabify
Don Stewart <dons@galois.com>**20080305015827] 
[untabify
Don Stewart <dons@galois.com>**20080305012530] 
[untabify
Don Stewart <dons@galois.com>**20080305010343] 
[untabify
Don Stewart <dons@galois.com>**20080305010255] 
[untabify
Don Stewart <dons@galois.com>**20080305005041] 
[untabify
Don Stewart <dons@galois.com>**20080305005025] 
[untabify
Don Stewart <dons@galois.com>**20080304235330] 
[untabify
Don Stewart <dons@galois.com>**20080304225120] 
[untabify
Don Stewart <dons@galois.com>**20080304174827] 
[untabify
Don Stewart <dons@galois.com>**20080303195109] 
[untabify
Don Stewart <dons@galois.com>**20080303195002] 
[untabify
Don Stewart <dons@galois.com>**20080303194454] 
[untabify
Don Stewart <dons@galois.com>**20080228234443] 
[untabify
Don Stewart <dons@galois.com>**20080228185409] 
[untabify
Don Stewart <dons@galois.com>**20080228185356] 
[untabify
Don Stewart <dons@galois.com>**20080228185331] 
[export MVar, TVar, and STM non-abstractly
Simon Marlow <simonmar@microsoft.com>**20080228113035
 As requested by Sterling Clover on ghc-users
] 
[Added Down class and improved groupWith fusion
Max Bolingbroke <batterseapower@hotmail.com>**20080213212246] 
[untabify
Don Stewart <dons@galois.com>**20080227062836] 
[untabify
Don Stewart <dons@galois.com>**20080226070630] 
[mention explicitly that hIsEOF may block
Simon Marlow <simonmar@microsoft.com>**20080220141209] 
[untabify
Don Stewart <dons@galois.com>**20080219233644] 
[untabify
Don Stewart <dons@galois.com>**20080219233047] 
[untabify
Don Stewart <dons@galois.com>**20080219232910] 
[untabify
Don Stewart <dons@galois.com>**20080219225437] 
[untabify
Don Stewart <dons@galois.com>**20080219061513] 
[Add exitSuccess :: IO a. For symmetry with exitFailure
Don Stewart <dons@galois.com>**20080213222644] 
[untabify
Don Stewart <dons@galois.com>**20080218075732] 
[untabify
Don Stewart <dons@galois.com>**20080218065411] 
[untabify
Don Stewart <dons@galois.com>**20080215005543] 
[FIX dynamic001, dynamic002: further fixes to tuple printing
Simon Marlow <simonmar@microsoft.com>**20080211101908] 
[untabify
Don Stewart <dons@galois.com>**20080213221950] 
[untabify only
Don Stewart <dons@galois.com>**20080213221856] 
[whitespace only
Don Stewart <dons@galois.com>**20080207191939] 
[Whitespace only
Don Stewart <dons@galois.com>**20080207183954] 
[FIX dynamic001 dynamic002: isTupleTyCon had rotted
Simon Marlow <simonmar@microsoft.com>**20080205103904
 In the patch "Tuple tycons have parens around their names", the names
 of the tuple tycons were changed to include parens, but isTupleTyCon
 was not updated to match, which made tuple types show as "(,) a b"
 rather than "(a,b)"
] 
[deforestation rules for enumFromThenTo; based on a patch from Robin Houston
Ian Lynagh <igloo@earth.li>**20080203152755] 
[Generalise type of forever :: (Monad m) => m a -> m b
Don Stewart <dons@galois.com>**20080129191940] 
[FIX #1936: hGetBufNonBlocking was blocking on stdin/stdout/stderr
Simon Marlow <simonmar@microsoft.com>**20080124092203] 
[The default uncaught exception handler was adding an extra \n
Simon Marlow <simonmar@microsoft.com>**20080124091216] 
[add comment about lack of _chsize_s()
Simon Marlow <simonmar@microsoft.com>**20080123131248] 
[Windows: large file support for hFileSize and hSeek (#1771)
Simon Marlow <simonmar@microsoft.com>**20080123102904
 
 
] 
[Export topHandler, topHandlerFastExit from GHC.TopHandler
Ian Lynagh <igloo@earth.li>**20080120182429
 We now use one of these in ghc when running with ghc -e
] 
[haddock attributes for haddock-2.0
Ross Paterson <ross@soi.city.ac.uk>**20080120022308] 
[Data.List.sort: force elements from start to end.
Bertram Felgenhauer <int-e@gmx.de>**20071121101458
 this prevents a stack overflow on  sort (take 10^6 [1..])
] 
[Fix comment on GHC.Ptr.minusPtr
simonpj@microsoft.com**20080109114736] 
[Remove redundant imports of GHC.Err
simonpj@microsoft.com**20080104091314
 
 GHC.Base SOURCE-imports GHC.Err, and re-exports 'error'.  So 
 other modules need only import GHC.Base.
 
 This doesn't change the fact that these other modules are all compiled
 before GHC.Err, so they are all part of the module loop that starts with
 GHC.Base and finishes with GHC.Err.  But it does reduce the occurrence
 of those SOURCE imports.
 
] 
[Tuple tycons have parens around their names
simonpj@microsoft**20071220171812
 
 The name of the pair TyCon, in the Typeable instance,
 should be "(,)" not ",".
 
 Don't merge to 6.8; it's a minor API change. 
 
] 
[Add groupWith, sortWith, the, to support generalised list comprehensions
simonpj@microsoft.com**20071220111929
 
   This the base-library patch to support the main compiler patch
      Implement generalised list comprehensions
   
   It just adds three functions to GHC.Exts.
 
] 
[Add GHC.Prim to exposedModules in the Haddock 0.x hook
David Waern <david.waern@gmail.com>*-20071209173931
 
 Please merge to the stable branch
] 
[Add GHC.Prim to exposedModules in the Haddock 0.x hook
David Waern <david.waern@gmail.com>**20071209173931
 
 Please merge to the stable branch
] 
[Simplify the GHC.Prim hack in base.cabal/Setup.hs
Ian Lynagh <igloo@earth.li>**20071202215758] 
[Implement 'openTempFile' for nhc98.
Malcolm.Wallace@cs.york.ac.uk**20071207133335] 
[docs: describe the changes to forkIO, and document forkOnIO
Simon Marlow <simonmar@microsoft.com>**20071205091423] 
[doc only: use realToFrac instead of fromRational.toRational
Simon Marlow <simonmar@microsoft.com>**20071205091334] 
[Add singletonP to GHC.PArr
Roman Leshchinskiy <rl@cse.unsw.edu.au>**20071205220859] 
[FIX #1621: bug in Windows code for getCPUTime
Simon Marlow <simonmar@microsoft.com>**20071205120118
 We were reading the components of FILETIME as CLong, when they should
 be unsigned.  Word32 seems to be the correct type here.
] 
[protect console handler against concurrent access (#1922)
Simon Marlow <simonmar@microsoft.com>**20071204153940] 
[protect against concurrent access to the signal handlers (#1922)
Simon Marlow <simonmar@microsoft.com>**20071204110817] 
[restore fdToHandle' to avoid breaking clients (#1109)
Simon Marlow <simonmar@microsoft.com>**20071130135122
 
] 
[note about how to convert CTime (aka EpochTime) to UTCTime
Simon Marlow <simonmar@microsoft.com>**20071130101648] 
[Fix some URLs
Ian Lynagh <igloo@earth.li>**20071126214213] 
[Fix some links in haddock docs
Ian Lynagh <igloo@earth.li>**20071126184428] 
[Don't try to make haddock links to the mtl package as we don't depend on it
Ian Lynagh <igloo@earth.li>**20071126170631] 
[Escape some special characters in haddock docs
Ian Lynagh <igloo@earth.li>**20071126163443] 
[FIX BUILD: maybeUpdateFile: ignore failures when removing the target
Simon Marlow <simonmar@microsoft.com>**20071123092219] 
[FIX #1753
Simon Marlow <simonmar@microsoft.com>**20071122094207
 hClose should close the Handle and unlock the file even if calling
 close() fails for some reason.
] 
[remove lockFile.h from install-includes
Simon Marlow <simonmar@microsoft.com>**20071121102248] 
[oops, we forgot to export traceShow
Simon Marlow <simonmar@microsoft.com>**20071121094300] 
[Fix compilation with GHC 6.2.x
Simon Marlow <simonmar@microsoft.com>**20071121084341] 
[Move file locking into the RTS, fixing #629, #1109
Simon Marlow <simonmar@microsoft.com>**20071120121053
 File locking (of the Haskell 98 variety) was previously done using a
 static table with linear search, which had two problems: the array had
 a fixed size and was sometimes too small (#1109), and performance of
 lockFile/unlockFile was suboptimal due to the linear search.
 Also the algorithm failed to count readers as required by Haskell 98
 (#629).
 
 Now it's done using a hash table (provided by the RTS).  Furthermore I
 avoided the extra fstat() for every open file by passing the dev_t and
 ino_t into lockFile.  This and the improvements to the locking
 algorithm result in a healthy 20% or so performance increase for
 opening/closing files (see openFile008 test).
] 
[Only overwrite GHC/Prim.hs and GHC/Primopwrappers.hs if they change
Simon Marlow <simonmar@microsoft.com>**20071120102042
 This avoids make doing unnecessary work after 'setup makefile'.
] 
[fix comment
Simon Marlow <simonmar@microsoft.com>**20071116091227] 
[Fix ` characters in elem's haddock docs
Ian Lynagh <igloo@earth.li>**20071110173052] 
[Filter out GHC.Prim also for the Haddock step
David Waern <david.waern@gmail.com>**20071109000806
 Please merge to the GHC 6.8.2 branch
] 
[Add module of special magic GHC desugaring helper functions
Simon Marlow <simonmar@microsoft.com>**20071102160054
 Currently containing only one such helper: (>>>) for arrow desugaring
] 
[add Control.Category to the nhc98 build
Malcolm.Wallace@cs.york.ac.uk**20071030120459] 
[fix nhc98 build: need a qualified Prelude import
Malcolm.Wallace@cs.york.ac.uk**20071030120410] 
[Fix performance regression: re-instate -funbox-strict-fields
Simon Marlow <simonmar@microsoft.com>**20071029150730
 Yikes!  While investigating the increase in code size with GHC 6.8
 relative to 6.6, I noticed that in the transition to Cabal for the
 libraries we lost -funbox-strict-fields, which is more or less
 depended on by the IO library for performance.  I'm astonished that we
 didn't notice this earlier!
 
 To reduce the chances of this happening again, I put
 -funbox-strict-fields in the OPTIONS_GHC pragma of the modules that
 need it.  {-# UNPACK #-} pragmas would be better, though.
] 
[FIX BUILD: Haddock 1.x fails to parse (Prelude..)
Simon Marlow <simonmar@microsoft.com>**20071029131921] 
[new Control.Category, ghc ticket #1773
Ashley Yakeley <ashley@semantic.org>**20071029022526] 
[new Control.Compositor module
Ashley Yakeley <ashley@semantic.org>**20071013074851
 
 The Compositor class is a superclass of Arrow.
] 
[Fix doc building with Haddock 0.9
Simon Marlow <simonmar@microsoft.com>**20071024090947
 I was using a recent build here, which is more tolerant.
] 
[FIX #1258: document that openTempFile is secure(ish)
Simon Marlow <simonmar@microsoft.com>**20071023130928
 Also change the mode from 0666 to 0600, which seems like a more
 sensible value and matches what C's mkstemp() does.
] 
[Clean up .cabal file a bit
Duncan Coutts <duncan@haskell.org>**20071022132708
 specify build-type and cabal-version >= 1.2
 put extra-tmp-files in the right place
 use os(windows) rather than os(mingw32)
] 
[base in 6.8 and head branch should be version 3.0
Don Stewart <dons@galois.com>**20071007150408] 
[FIX #1652: openTempFile should accept an empty string for the directory
Simon Marlow <simonmar@microsoft.com>**20071018122345] 
[clean up duplicate code
Simon Marlow <simonmar@microsoft.com>**20071017141311] 
[expose the value of +RTS -N as GHC.Conc.numCapabilities (see #1733)
Simon Marlow <simonmar@microsoft.com>**20071009132042] 
[typo
Simon Marlow <simonmar@microsoft.com>**20070917130703] 
[put extra-tmp-files field in the right place
Simon Marlow <simonmar@microsoft.com>**20070914140812] 
[Add more entries to boring file
Ian Lynagh <igloo@earth.li>**20070913210500] 
[Add a boring file
Ian Lynagh <igloo@earth.li>**20070913204641] 
[TAG 2007-09-13
Ian Lynagh <igloo@earth.li>**20070913215720] 
Patch bundle hash:
3a528267f25f2781d8a633e99cc6731e37dae8d2