racket/LOG

9.4 changes:
- updated version to 9.4
    bintar README NOTICE makefiles/Mf-install.in scheme.1.in
    c/Makefile.i3nt c/Makefile.a6nt c/Makefile.ti3nt c/Makefile.ta6nt
    mats/bldnt.bat workarea c/scheme.rc s/7.ss s/cmacros.ss
    release_notes/release_notes.stex csug/csug.stex
- added missing include
    killme.c
- added new mat verifying that primitives raise exceptions for invalid
  arguments based on primdata.ss signatures.  fixed some of those
  signatures, which weren't otherwise used except for arity checking.
  fixed some issues turned up by the test with who reports errors
  and how.
    primdata.ss, 5_1.ss, 7.ss, compile.ss, cpnanopass.ss, fasl.ss,
    interpret.ss, io.ss, record.ss, syntax.ss,
    primvars.ms, 5_1.ms, 7.ms, 8.ms, record.ms, mats/Mf-base,
    root-experr*, patch*
- string comparisons (string=?, string-ci<?, etc.) now get out fast
  when handed eq arguments.
    5_4.ss
- changed representation of most-negative iptr, I32, and I64 to make
  Visual C compiler happy.  updated windows make files, which had
  fallen out of date.  added missing hsrc= files sort.h and thread.h
  so they show up in the Windows workarea c directory.
    cmacros.ss,
    fasl.c, number.c, c/Mf-base, c/Makefile.{t,}{i3,a6}nt
- The scheme.h definition of Sfixnum(x) now uses multiply rather than
  left shift to avoid counting on the behavior of << on negative numbers,
  which is undefined in C.
    mkheader.ss
- Fixed a couple of casts, one harmless and the other causing
  Sinteger64_value to return 0 for the most-negative I64 on 32-bit
  builds.
    number.c
- The configure-generated Makefile distclean target no longer removes
  zlib and nanopass, since they are logically part of the git clone.
  It does run make distclean in zlib.
    makefiles/Makefile.in
- converted s_test_schlib shifts of -1 to equivalent shifts of 1 to
  avoid undefined left-shift behavior on negative numbers.
    prim5.c
- added if(---) {} wrapper around call to WRITE in display macro to
  silence unused return-value warnings.
    prim5.c
- liberalized get-mode check for ../mats.  it's not our business whether
  people make their directories group and/or other writeable.
    6.ms
- make test now prints the actual relative path to summary in the
  "check summary" message, whether invoked from the top-level directory
  or from the workarea.
    Makefile.in, Makefile-workarea.in
- configure now just uses cat to copy Makefile-workarea.in to $w/workarea,
  since the file is presently the same regardless of the configuration.
    configure
- fixed time-utc->date test in mat time&date-printing to work regardless of
  what locale (and time zone) the host machine has set.
    date.ms
- fixed date->time-utc to honor the zone-offset field when converting a date
  object to a time-utc object.
    stats.c,
    date.ms
- fixed incorrect handling of library-extension when searching wpo files
    compile.ss,
    7.ms
- modified floatify_normalize to properly round denormalized results.
  obviated scale_float in the process.
    number.c,
    ieee.ms
- fixed 0eNNNN for large NNNN to produce 0.0 rather than infinity
    strnum.ss,
    5_3.ms
- the reader now raises an exception with condition type implementation
  restriction (among the other usual lexical condition types), and
  string->number now raises #f, for #e<m>@<a>, where <m> and <a> are
  nonzero integers, since Chez Scheme can't represent polar numbers other
  than 0@<n> and <n>@0 exactly.  <m>@<a> still produces an inexact result,
  i.e., we're still extending the set of inexact numeric constants beyond
  what R6RS dictates.  doing this required a rework of $str->num, which
  turned into a fairly extensive rewrite that fixed up a few other minor
  issues (like r6rs:string->number improperly allowing 1/2e10) and
  eliminated the need for consumers to call $str->num twice in cases
  where it can actually produce a number.  added some related new tests,
  including several found missing by profiling.  added a couple of
  checks to number->string the absence of which was causing argument
  errors to be reported by other routines.
    strnum.ss, exceptions.ss, read.ss
    5_3.ms, 6.ms, root-experr*, patch*
- added pdtml flag, which if set to t causes profile-dump-html to be
  called at the end of a mat run.
    mats/Mf-base
- compile-whole-program and compile-whole-library now copy the hash-bang
  line from the wpo file (if it has one) to the object file.
    compile.ss,
    7.ms
- stex is now a submodule.  csug/Makefile and release_notes/Makefile
  set and use the required Scheme and STEXLIB variables accordingly.
  they default the machine type to a6le, but this can be overridden
  and is by the generated top-level Makefile.  the generated top-level
  Makefile now has a new docs target that runs make in both csug and
  release_notes, and an updated distclean target that cleans the same.
  the annoying csug Makefile .fig.pdf rule redefinition is now gone.
  copyright.stex and csug.stex now list May 2016 as the revision month
  and date; this will have to be updated for future releases.
    configure, makefiles/Makefile.in,
    csug/Makefile, copyright.stex, csug.stex,
    release_notes/Makefile
- added custom install options.  workarea creates an empty config.h,
  and configure creates a config.h that sets the default scheme heap
  path and scheme-script name based on the actual configuration.
    configure, newrelease, workarea, checkin,
    c/Mf-base, scheme.c, main.c,
    Mf-install.in
- renamed the installed example directory from lib to examples.
    Mf-install.in,
    scheme.1.in
- added force option to gzip during man page install to prevent gzip from
  asking for permission to overwrite existing man page files.
    Mf-install.in
- removed ~/lib/csv%v/%m from the default scheme heap path on unix-like
  systems.  documented inclusion of %x\..\..\boot\%m in the Windows
  default scheme heap path.
    main.c,
    use.stex
- added new configuration options: --installbin, --installlib,
  --installschemename, --installpetitename, and --installscriptname.
    configure
- updated the example library link to the nanopass framework.
    CONTRIBUTING.md
- now cleaning up petite.1 and scheme.1 left behind by make install
    Makefile-workarea.in, checkin
- now removing workarea after cleaning csug and release_notes so
  Mf-stex (included from csug/Makefile and release_notes/Makefile)
  doesn't complain trying to determine the machine type.
    Makefile.in
- added installsh support for --ifdiff so the csug make file can use it
  for the install target.
    installsh,
    csug/Makefile
- added instructions for building (cross-compiling) a boot file for
  a supported machine type for which a boot file is not built by default.
    BUILDING
- corrected CHEZSCHEMELIBS and CHEZSCHEMEEXTS index entries to be
  CHEZSCHEMELIBDIRS and CHEZSCHEMELIBEXTS.
    use.stex
- updated to curl stex version 1.2.1
    configure
- updated the fix to S_mktime to work on windows.  the struct tm
  structure on windows does not have the tm_gmtoff field used in the
  mac and linux version of the code.
    stats.c
- updated the Windows makefiles for building and testing to remove links for
  files that no longer exist, which was getting in the way of creating links
  for files that do exist.  Also updated the build batch file for Windows to
  compile the nanopass framework separately before building the rest of the
  scheme compiler and libraries.
    s/Mf-{a6,i3,ta6,ti3}nt, s/bldnt.bat,
    mats/Mf-{a6,i3,ta6,ti3}nt
- updated the read me to include a link to the Chez Scheme project page.
    README.md
- fixed embarrassing typo in read me.
    README.md
- profiler's html output refresh: mark the files as HTML5 rather
  than HTML4; use target attributes rather than onclick events to
  open links in specific windows; add a missing table row element;
  replace the deprecated name attribute with an id attribute (and
  replace the anchors with spans); and replace the deprecated valign
  attribute with a style attribute.
    pdhtml.ss

9.4.1 changes:
- updated version to 9.4.1
    bintar BUILDING NOTICE makefiles/Mf-install.in scheme.1.in
    c/Makefile.i3nt c/Makefile.a6nt c/Makefile.ti3nt c/Makefile.ta6nt
    mats/bldnt.bat workarea c/scheme.rc s/7.ss s/cmacros.ss
    release_notes/release_notes.stex csug/csug.stex
- updated newrelease to produce the correct log-entry format and
  fixed the existing 9.4.1 log entry
    newrelease, LOG
- replaced a couple of tabs
    number.c
- updated the descriptions of statistics and related functions to
  reflect the Version 9.3.1 change from sstats structures to sstats
  records, with sstats times represented as time objects and sstats
  bytes and counts represented as exact integers; also updated the
  sstats-difference description to reflect that it no longer coerces
  negative differences to zero.  added a corresponding release note.
    system.stex,
    release_notes.stex
- added a cast to eliminate a warning
    c/number.c
- fixed bug in Windows version of directory-separator-predicate when
  path-* procedures are passed a path that is not a string.
    s/6.ss
- fixed bug in cp0 on Windows with $foreign-wchar?.
    s/cp0.ss
- Cygwin is now used on Windows, updated mats, eliminated unused killme
    BUILDING c/*nt c/Mf-base c/scheme.exe.manifest configure
    examples/Makefile mats/6.ms mats/Mf-* mats/foreign.ms mats/ftype.ms
    mats/patch-* mats/windows.ms s/Mf-*nt s/Mf-base workarea
    release_notes.stex
- fixed spelling of non-existent
    s/syntax.ss
- now forcing zlib configuration before compiling files that depend on
  the zlib header files, since zlib's configure script can modify the
  header files.  removed ZlibInclude variable, which no longer serves
  a purpose.
    c/Mf-*, c/Makefile.*nt
- removed unnecessary datestamp.c target
    c/Mf.*nt
- fixed unnessesary blocking in expeditor on Windows. 
    c/expeditor.c
- eliminated a couple of thread-safety issues and limitations on the
  sizes of pathnames produced by expansion of tilde (home-directory)
  prefixes by replacing S_pathname, S_pathname_impl, and S_homedir
  with S_malloc_pathname, which always mallocs space for the result.
  one thread-safety issue involved the use of static strings for expanded
  pathnames and affected various file-system operations.  the other
  affected the file open routines and involved use of the incoming
  pathname while deactivated.  the incoming pathname is sometimes if not
  always a pointer into a Scheme bytevector, which can be overwritten if a
  collection occurs while the thread is deactivated.  the size limitation
  corresponded to the use of the static strings, which were limited to
  PATH_MAX bytes.  (PATH_MAX typically isn't actually the maximum path
  length in contemporary operating systems.)  eliminated similar issues
  for wide pathnames under Windows by adding S_malloc_wide_pathname.
  consumers of the old routines have been modified to use the new
  routines and to free the result strings.  the various file operations
  now consistently treat a pathname with an unresolvable home directory
  as a pathname that happens to start with a tilde.  eliminated unused
  foreign-symbol binding of "(cs)pathname" to S_pathname.
    io.c, externs.h, new_io.c, prim5.c, scheme.c, prim.c
- various places where a call to close or gzclose was retried when
  the close operation was interrupted no longer do so, since this can
  cause problems when another thread has reallocated the same file
  descriptor.
    new_io.c
- now using vcvarsall type x86_amd64 rather than amd64 when the
  former appears to supported and the latter does not, as is the
  case with VS Express 2015.
    c/Mf-a6nt, c/Mf-ta6nt
- commented out one of the thread mats that consistently causes
  indefinite delays under Windows and OpenBSD due to starvation.
    thread.ms
- increased wait time for a couple of subprocess responses
    6.ms
- added call to collector to close files opened during iconv mats
  specifically for when mats are run under Windows with no iconv dll.
    io.ms
- fixed typo: VC/bin/vcvars64.bat => VC/bin/amd64/vcvars64.bat
    c/Mf-a6nt, c/Mf-ta6nt
- scheme_mutex_t now uses volatile keyword for owner and count fields
  because these fields can be accessed from multiple threads
  concurrently. Updated $yield and $thread-check in mats/thread.ms to
  be more tolerant of timing variability.
    c/types.h, mats/thread.ms, release_notes.stex
- fixed three instances of unchecked mallocs reported by laqrix in
  github issue #77.
    io.c, schlib.c, thread.c
- continue the profiler's html output refresh: refine the styling
  (and palette) and update CSUG to match. update the CSUG screenshots
  to reflect the refined look.
    s/pdhtml.ss
    csug/system.stex 
    csug/canned/profilehtml-orig.png
    csug/canned/profilehtml.png
    csug/canned/fatfibhtml-orig.png
    csug/canned/fatfibhtml.png
- add unicode support to the expression editor.  entry and display now work
  except that combining characters are not treated correctly for
  line-wrapping.  this addresses github issue #32 and part of issue #81.
    c/expeditor.c, s/expeditor.ss
- moved s_ee_write_char function within the WIN32 check to allow the unicode
  change to compile on windows.  unicode is not yet supported in the windows
  version of the repl.
    c/expeditor.c
- reworked the S_create_thread_object to print an error and exit when
  allocating the thread context fails from Sactivate_thread.  before
  this change, the error was raised on the main thread, which resulted
  in strange behavior at best.  also added who argument to
  S_create_thread_object to allow it to report either Sactivate_thread
  or fork-thread led to the error.
    externs.h, schsig.c, scheme.c, thread.c
- fixed a bug in cpvalid resulting in it leaving behind a cpvalid-defer
  form for later passes to choke on.  also fixed cp0 to print the correct
  name for cpvalid when it does this.
    cpvalid.ss, cp0.ss,
    misc.ms
- updated the prototype for s_ee_write_char to match the definition
    expeditor.c
- fixed a side-effect preservation bug with non-trivial test-context
  not-like patterns.
    cp0.ss,
    cp0.ms, 4.ms
- instead of default-exception handler, new-cafe establishes a handler
  that calls the current value of base-exception-handler so the handler
  can be overridden, as we do in our own make files.
    cafe.ss,
    7.ms
- fixed a bug in case and exclusive-cond syntax-error calls causing an
  exception in syntax-error instead of the intended error message.
    syntax.ss
- added tests for the case and exclusive-cond syntax-error calls
    4.ms, root-experr-compile-0-f-f-f
- added print-extended-identifiers parameter.  when #t, symbols like
  1+ and +++ print without escapes.
    priminfo.ss, print.ss,
    6.ms
- added descriptions of print-extended-identifiers to the user's guide
  and release notes.  updated the release notes to account for a couple
  of other log entries.
    release_notes.stex,
    intro.stex, io.stex
- updated the sockets example to work with the current version of Chez.
  Change the foreign procedure definitions to use the more portable int
  rather than integer-32.  Switch to a custom port
  [make-custom-binary-input/output-port] rather than a generic port
  [make-input/output-port], which resulted in deleting quite a bit of
  code.  Fix various compiler warnings in the C code, and along the way,
  fix a signedness bug in c_write that could have resulted in not writing
  the full buffer (but reporting that it did) in the case of errors from
  write.
    examples/csocket.c, examples/socket.ss
- use high-precision clock time on Windows 8 and up
    c/stats.c
- fixed profiling code that keyed profiling locations off of only the
  bfp to instead key off of both the bfp and efp.
    pdhtml.ss
- added Windows installer using the WiX Toolset
    BUILDING, install/* (new)
- fix typo in ordinal format for 12
    format.ss,
    format.ms
- renamed install directory to wininstall to avoid conflict with
  top-level Makefile
    BUILDING, install/* (removed), wininstall/* (new)
- updated zlib to version 1.2.11
    configure
- added procedure-arity-mask to report the allowed argument counts of
  a compiled function.  On a procedure from interpret or from one of
  the trace procedures or syntactic forms, procedure-arity-mask
  may report counts that are not actually allowed by the source
  procedure.
    cmacros.ss, compile.ss, cpnanopass.ss, mkheader.ss, primdata.ss,
    prims.ss, strip.ss,
    fasl.c, gc.c, globals.h, prim.c, prim5.c, scheme.c, schsig.c,
    misc.ms, root-experr*,
    objects.stex
- for non-win32 systems, now setting al register to a count of the
  floating-point register arguments as required for varargs functions
  by the System V ABI.
    x86_64.ss,
    foreign.ms
- added a missing quote mark in new printf mat Windows case
    foreign.ms
- added travis-ci automation script and a 'partialx' testing target to
  allow us to perform more than a single run of testing without running
  afoul of travis-ci's 50-minute build timeout.  'partialx' tests six
  of the twelve configurations tested by 'allx'.
     .travis.yml (new),
     mats/Mf-base
- paired the 'partialx' down to just four test configurations, with one
  interpreter run, to try to get the threaded builds into line with
  travis-ci's timeout.
     mats/Mf-base
- eliminated some direct assumptions that a vector's type/length field
  is a fixnum and added meta-asserts to verify that it is in a couple of
  others, to facilitate future changes to vector typing.  vectors are
  now treated essentially like fxvectors, strings, and bytevectors.
    cmacros.ss, cpnanopass.ss, prims.ss, mkheader.ss,
    alloc.c, gc.c, scheme.c
- fixed a few comments to refer to scheme.c rather than main.c
    externs.h, globals.h, thread.c
- for 64-bit Windows systems, now copying foreign-procedure
  double-precision floating-point register arguments to integer
  registers as required for varargs functions. Windows does not
  support single-precision floating-point arguments as varargs.
    foreign.ms, np-languages.ss, x86_64.ss
- added an optional timeout argument to condition-wait
    externs.h, stats.c, thread.c, thread.h, csug/threads.stex,
    primvars.ms, thread.ms, release_notes.stex,
    date.ss, primdata.ss, prims.ss
- added immutable strings, vectors, fxvector, bytevectors, and boxes
    5_4.ss, 5_6.ss, bytevector.ss, cmacros.ss, cpnanopass.ss,
    fasl.ss, library.ss, mkheader.ss, primdata.ss, prims.ss,
    externs.h, types.h, alloc.c, fasl.c, gc.c, scheme.c,
    5_5.ms, 5_6.ms, bytevector.ms, misc.ms, root-experr*
    objects.stex
- various tweaks to the immutable object support; also taught cp0
  to simplify ($fxu< (most-positive-fixnum) e) => (fx< e 0) so we
  don't have any incentive in special casing length checks where
  the maximum length happens to be (most-positive-fixnum).
    5_4.ss, 5_6.ss, bytevector.ss, cmacros.ss, cp0.ss, cpnanopass.ss,
    mkheader.ss, primdata.ss, prims.ss,
    fasl.c, gc.c, types.h
    root-experr*, patch*
- generated bytevector=? procedure now gets out quickly on eq
  arguments.  cp0 optimizes away a number of additional equality
  operations at optimize-level 3 (including bytevector=?) when
  applied to the same variable references, as it already did for
  eq?, eqv?, and equal?, at all optimize levels.
    cpnanopass.ss, cp0.ss, primdata.ss,
    cp0.ms
- updated bullyx patches
    patch*
- updated release notes and tweaked user's guide.
    release-notes.stex, objects.stex
- fixed typo: fxvector-immutable-flag used in place of
  bytevector-immutable-flag in computation of type-immutable-bytevector
    cmacros.ss
- reallocated typed-object types, using previously unused tag #b010
  for strings and giving bytevectors both #b001 and #b101 (the
  latter for immutable bytevectors) so that the maximum bytevector
  length on 32-bit machines is once again the most-positive fixnum.
  treating bytevectors rather than strings or fxvectors (or even
  vectors) special in this regard is appropriate since the maximum
  number of bytes in a bytevector is maximum-length x 1 rather than
  maximum-length x 4 for strings, fxvectors, and vectors on 32-bit
  machines.  with this change on 32-bit machines, a vector can
  occupy up to 1/2 of virtual memory, strings and fxvectors 1/4,
  and bytevectors 1/8.
    cmacros.ss
- added record-type-equal-procedure, record-type-hash-procedure,
  record-equal-procedure, and record-hash-procedure to enable
  per-type customization of the behavior of equal? and equal-hash
  for a record value
     5_1.ss, newhash.ss, primdata.ss,
     record.ms, root-experr*,
     objects.stex
- adding dropped changes
    record.ss,
    root-experr*
- added entry for record-type-equal-procedure and friends
    release_notes.stex
- changed copyright year to 2017
    7.ss, scheme.1.in, comments of many files
- expanded the CSUG description of the handling of command-line
  arguments not recognized as standard options and added a description
  of the same to the COMMAND-LINE OPTIONS section of the man page.
  did a bit of minor cleanup elsewhere in the man page.
    use.stex, scheme.1.in
- destroy_thread now processes guardian entries
    thread.c, 4.ms, release_notes.stex
- mutexes and conditions are now freed when no longer used via
  $close-resurrected-mutexes&conditions, $keep-live primitive added
    externs.h, prim5.c, thread.c, 4.ms, thread.ms, release_notes.stex,
    7.ss, cpnanopass.ss, np-languages.ss, primdata.ss, prims.ss
- fix reduction for map and for-each with optimization level 3
  to drop the expression, check that procedure has the correct
  arity and is discardable or unsafe.
  Also add a simplification for for-each with empty lists
  with optimization level 2.
    cp0.ss, 4.ms, primdata.ss
- fix invalid memory reference when enum-set-indexer procedure is not
  passed a symbol
    enum.ss, enum.ms, root-experr*, release_notes.stex
- fix overflow detection for fxsll, fxarithmetic-shift-left, and
  fxarithmetic-shift
    library.ss, fx.ms, release_notes.stex
- added ephemeron pairs and changed weak hashtables to use
  ephemeron pairs for key--value mapping to avoid the key-in-value
  problem
    prims.ss, primdata.ss, newhash.ss, fasl.ss, mkheader.ss
    cmacro.ss, prim5.c, fasl.c, gc.c, gcwrapper.c, types.h,
    4.ms, hash.ms, objects.stex, smgmt.stex, csug.bib
- check_dirty_ephemeron now puts ephemerons whose keys haven't yet
  been seen on the pending list rather than the trigger lists.
    gc.c
- removed scan of space_ephemeron from check_heap because check_heap
  as written can't handle the two link fields properly.
    gcwrapper.c
- in the ephemerons mat that checks interaction between mutation and
  collection, added generation arguments to the first two collect
  calls so they always collect into the intended generation.
    4.ms
- updated allx and bullyx patches
    patch*
- fix strip-fasl-file for immutable strings and vectors,
  fix an $oops call, and fix a vector-index increment in hashing
    strip.ss, 7.ss, newhash.ss, misc.ms
- fix signature of fxbit-set?
    primdata.ss
- more optimizations for map and for-each with explicit list
  extend the reductions for map and for-each when the arguments are
  explicit lists like (list 1 2 3 ...) or '(1 2 3 ...). 
    cp0.ss,
    4.ms
- reverted to the preceding version of cp0 due to failure to preserve
  the expected evaluation order in one of the mats; removed the
  corresponding equivalent-expansion tests.
    cp0.ss,
    4.ms
- restored the map and for-each optimizations with a fix for the
  evaluation-order bug.
    cp0.ss,
    4.ms
- added date-dst? to access the previously-hidden DST information in
  date records, and added date-zone-name to provide a time zone name.
    date.ss, primdata.ss,
    stats.c,
    date.ms, root-experr*, patch-compile*,
    system.stex
- fixed a bug in flonum-extractor, which on 64-bit machines was using an
  8-byte read instead of a 4-byte read to pick up the 4 highest-order
  bytes of a little-endian flonum, potentially reading past the end of
  mapped memory for flonums produced by taking the imaginary part of an
  inexact complexnum (which, unlike other flonums, are not aligned on
  16-byte boundaries).  The 8-byte load would also have failed to produce
  correct results on 64-bit big-endian machines (of which we presently
  have none) because the offsets passed to flonum-extractor assume the
  bits are in the lowest-order 4 bytes of the extracted field.
    cp0.ss,
    misc.ms,
    release_notes.stex
- support Windows build on Bash/WSL
    BUILDING, configure, workarea, c/vs.bat (new), mats/vs.bat (new),
    c/Mf-*nt, mats/Mf-*, s/Mf-base
- fix c/version.h for FreeBDS (machine types i3fb, ti3fb, a6fb, ta6fb)
- fix reference to libc.so to be libc.so.7 for FreeBSD (machine types
  i3fb, ti3fb, a6fb, ta6fb)
    foreign.ms
- added CC option to configure for selecting the compiler
    configure,
    c/Mf-*
- Suppress warnings from implicit fallthrough in case labels.
    Mf-{a6,arm32,i3,ppc,ta6,ti3,tpp32}le
- added bytevector-compress and bytevector-uncompress
    bytevector.ss, primdata.ss, new-io.c, prim5.c, externs.h,
    objects.stex, release_notes.stex,
    bytevector.ms, root-experr*
- fixed typo in S_abnormal_exit
    schsig.c
- don't remove the pariah form in the cp0 pass
    cp0.ss,
    misc.ms
- revert use of ephemerons in weak hashtables, add ephemeron
  hashtables
    newhash.ss, hashtable-types.ss, library.ss, primdata.ss,
    fasl.ss, fasl.c, gc.c, globals.h,
    hash.ms, objects.stex, release_notes.stex
- fixed pariah mat
    misc.ms
- minor wordsmithing and fix for an overfull hbox
    objects.stex, system.stex
- fix (define-values () ....) to expand to a definition
    syntax.ss, 3.ms
- added optional line and column components to a source object, a
  locate-source-object-source function that uses the new components,
  a current-locate-source-object-source parameter to control looking up
  line and column information, a current-make-source-object parameter to
  control loation recording, an optional use-cache argument to
  locate-source, and a 'source-object message for code and continuation
  inspectors
    read.ss, syntax.ss, 7.ss, compile.ss, cpnanopass.ss, exceptions.ss,
    inspect.ss, primdata.ss, prims.ss, print.ss, cmacros.ss, types.ss,
    mat.ss, 8.ms, root-experr*,
    syntax.stex, debug.stex, system.stex, release_notes.stex
- fixed broken mats on Windows caused by Bash/WSL changes
    7.ms, ftype.ms
- added "ez-grammar" example program
    examples/ez-grammar.ss, examples/ez-grammar-test.ss,
    examples/Makefile, examples.ms
- updated ez-grammar-test to write temp files to current directory and delete them when finished
    examples/ez-grammar-test.ss
- added support for Microsoft Visual Studio 2017 on Windows
    BUILDING, c/Mf-a6nt, c/Mf-ta6nt, c/vs.bat,
    mats/Mf-a6nt, mats/Mf-ta6nt, mats/ftype.ms
- added support for building Windows installs with Bash/WSL
    wininstall/Makefile, candle.bat, light.bat
- added support for building with Visual Studio 2017's BuildTools
    c/vs.bat
- check for git before using to get submodules
    configure
- fixed windows installer failure when vcredist is not preinstalled by
  using the vcredist merge module, split the 32 and 64 bit MSIs and
  added a wix bundle to combine the MSIs into a single exe installer,
  added a batch script for locating Visual Studio's vcredist merge
  modules, updated installer paths and names.
    wininstall/*
- fixed np-normalize-context pass to process trivs list in mvset forms
  in tail and predicate context and added regression tests.  Thanks to
  @marcomaggi for reporting the bug and @yjqww6 for providing a
  simplified test and finding the initial solution.
    cpnanopass.ss,
    3.ms
- removed a useless check in foreign-alloc
    record.ss
- fix cp0 reduction of fx[+-*]/carry and their signatures
    cp0.ss
    primdata.ss
    fx.ms
- renamed s_gettime => S_gettime to remain consistent with the
  convention that the only undocumented externs are prefixed with
  S_.
    externs.h, stats.c, thread.c
- added version number to scheme.1.in trailer; updated date.
    scheme.1.in, newrelease
- removed version update of no-longer-existing bldnt.bat.  "fixed"
  sed patterns to replace \? with * for the benefit of the deficient
  mac sed.
    newrelease

9.5 changes:
- updated version to 9.5
    bintar BUILDING NOTICE makefiles/Mf-install.in scheme.1.in
    c/Makefile.i3nt c/Makefile.a6nt c/Makefile.ti3nt c/Makefile.ta6nt
    workarea c/scheme.rc s/7.ss s/cmacros.ss
    release_notes/release_notes.stex csug/csug.stex
- updated release notes and fixed user's guide overfull hbox.
    release-notes.stex, syntax.stex
- updated install target to do something more sensible
    release_notes/Makefile

9.5.1 changes:
- updated version to 9.5.1
    bintar BUILDING NOTICE makefiles/Mf-install.in scheme.1.in
    c/Makefile.i3nt c/Makefile.a6nt c/Makefile.ti3nt c/Makefile.ta6nt
    workarea c/scheme.rc s/7.ss s/cmacros.ss
    release_notes/release_notes.stex csug/csug.stex csug/use.stex
    examples/ez-grammar-test.ss examples/socket.ss
    wininstall/Makefile wininstall/*nt.wxs
- Added setting of CHEZSCHEMELIBDIRS to s and mats make files so that
  an existing setting will not interfere with the build process, and
  added a note to BUILDING that CHEZSCHEMELIBDIRS should be unset in
  Version 9.5 and before.
    s/Mf-base, mats/Mf-base, BUILDING
- the $case macro used by r6rs:case and case now unconditionally trims
  redundant keys and expands into exclusive-cond rather than cond.
  it catches references to => before expanding into exclusive-cond
  to avoid supporting => as an undocumented and useless extension
  of the case syntax.  the r6rs:case and case macros now require
  multiple clauses rather than leaving the enforcement to exclusive-cond,
  and the exclusive-cond macro now requires multiple clauses rather
  than leaving the enforcement to cond.
    syntax.ss,
    4.ms, root-experr*, patch*
- ifdef'd out include of xlocale.h for glibc, since the glibc
  locale.h includes xlocale.h or, in glibc 2.26, its replacement.
    expeditor.c
- Updated CSUG to replace \INSERTREVISIONMONTHSPACEYEAR with the current
  month and year at the time of generation.
    csug.stex, copyright.stex
- Updated configuration to set machine types in the CSUG and release notes 
  make files, and updated distclean target to remove these files.
    configure, makefiles/Makefile-csug.in (renamed from csug/Makefile),
    makefiles/Makefile-release_notes.in
      (renamed from release_notes/Makefile),
    makefiles/Makefile
- added pass-time tracking for pre-cpnanopass passes to compile.
    compile.ss
- added inline handler for fxdiv-and-mod
    cp0.ss, primdata.ss
- changed order in which return-point operations are done (adjust
  sfp first, then store return values, then restore local saves) to
  avoid storing return values to homes beyond the end of the stack
  in cases where adjusting sfp might result in a call to dooverflood.
    cpnanopass.ss, np-languages.ss
- removed unused {make-,}asm-return-registers bindings
    cpnanopass.ss
- corrected the max-fv value field of the lambda produced by the
  hand-coded bytevector=? handler.
    cpnanopass.ss
- reduced live-pointer and inspector free-variable mask computation
  overhead
    cpnanopass.ss
- moved regvec cset copies to driver so they aren't copied each
  time a uvar is assigned to a register.  removed checks for
  missing register csets, since registers always have csets.
    cpnanopass.ss
- added closure-rep else clause in record-inspector-information!.
    cpnanopass.ss
- augmented tree representation with a constant representation
  for full trees to reduce the overhead of manipulating trees or
  subtress with all bits set.
    cpnanopass.ss
- tree-for-each now takes start and end offsets; this cuts the
  cost of traversing and applying the action when the range of
  applicable offsets is other than 0..tree-size.
    cpnanopass.ss
- introduced the notion of poison variables to reduce the cost of
  register/frame allocation for procedures with large sets of local
  variables.  When the number of local variables exceeds a given
  limit (currently hardwired to 1000), each variable with a large
  live range is considered poison.  A reasonable set of variables
  with large live ranges (the set of poison variables) is computed
  by successive approximation to avoid excessive overhead.  Poison
  variables directly conflict with all spillables, and all non-poison
  spillables indirectly conflict with all poison spillables through
  a shared poison-cset.  Thus poison variables cannot live in the
  same location as any other variable, i.e., they poison the location.
  Conflicts between frame locations and poison variables are handled
  normally, which allows poison variables to be assigned to
  move-related frame homes.  Poison variables are spilled prior to
  register allocation, so conflicts between registers and poison
  variables are not represented.  move relations between poison
  variables and frame variables are recorded as usual, but other
  move relations involving poison variables are not recorded.
    cpnanopass.ss, np-languages.ss
- changed the way a uvar's degree is decremented by remove-victim!.
  instead of checking for a conflict between each pair of victim
  and keeper and decrementing when the conflict is found, remove-victim!
  now decrements the degree of each var in each victim's conflict
  set.  while this might decrement other victims' degrees unnecessarily,
  it can be much less expensive when large numbers of variables are
  involved, since the number of conflicts between two non-poison
  variables should be small due to the selection process for
  (non-)poison variables and the fact that the unspillables introduced
  by instruction selection should also have few conflicts.  That
  is, it reduces the worst-case complexity of decrementing degrees
  from O(n^2) to O(n).
    cpnanopass.ss
- took advice in compute-degree! comment to increment the uvars in
  each registers csets rather than looping over the registers for
  each uvar asking whether the register conflicts with the uvar.
    cpnanopass.ss
- assign-new-frame! now zeros out save-weight for local saves, since
  once they are explicitly saved and restored, they are no longer
  call-live and thus have no save cost.
    cpnanopass.ss
- desensitized the let-values source-caching timing test slightly
    8.ms
- updated allx, bullyx patches
    patch*
- attempt to stabilize timing tests let-values source-caching
  test and ephemeron gc test while resensitizing the former
    8.ms, 4.ms
- various formatting and comment corrections
    workarea,
    s/Mf-base, bytevector.ss, cpnanopass.ss, date.ss,
    5_6.ms, examples.ms
- updated newrelease to handle mats/Mf-*nt
    newrelease mats/Mf-a6nt mats/Mf-i3nt mats/Mf-ta6nt mats/Mf-ti3nt
- fixed gather-filedata's sort of profile entries.  for any two
  entries x and y in the list produced by the sort call, if x's
  bfp = y's bfp, x should come before y if x's efp < y's efp.
  The idea is that enclosing entries should always come later
  in the list.  this affects only languages where two expressions
  can start at the same character position.
    pdhtml.ss
- expanded capability of ez-grammar with support for simpl
  parsing of binary operators w/precedence and associativity
  and automatically generated markdown grammar descriptions.
  ez-grammar-test.ss now also doubles as a test of pdhtml for
  algebraic languages.
    mats/examples.ms,
    examples/ez-grammar.ss, examples/ez-grammar-test.ss,
    examples/Makefile
- maybe-compile-{file,program,library} and automatic import
  compilation now treat a malformed object file as if it were
  not present and needs to be regenerated.  A malformed object
  file (particularly a truncated one) might occur if the compiling
  processes is killed or aborts before it has a chance to delete
  a partial object file.
    syntax.ss,
    7.ms
- fix signature of bytevector-[u/s]16-native-set!
    primdata.ss
- fix enumerate signature 
    primdata.ss
- added support for Visual Studio 2017.15.5
    wininstall/locate-vcredist.bat
- fixed substring-fill! and vector-fill! to return void, reflecting the
  documented return value of unspecified value.  Also changes substring-fill!
  to use define-who instead of repeating 'substring-fill! in all the error
  messages.
    5_4.ss, 5_6.ss
- fix mat of substring-fill!
  after the recent change, the result of substring-fill! is void
    5_5.ms
- fix a few signatures
    primdata.ss
- fix comment about Sscheme_program
    main.c
- fix even? and odd? to error on exceptional flonums
    5_3.ss, 5_3.ms, fl.ms, root-experr*, patch*
- fix bug in date->time-utc caused by incorrect use of difftime in Windows
    stats.c, date.ms, release_notes.stex
- Check that first argument of map is a procedure in cp02 expansion
  to raise the same error that the non expanded version
    cp0.ss
- avoid building the result list in a map that is called for effect
    cp0.ss
- added tests to ensure the optimize-level version 2 of map and for-each raise
  a non-procedure exception when the first argument is not a procedure, even
  when the rest of the program is compiled at optimize level 3.
    4.ms, root-experr-compile-0-f-f-f, patch-compile-0-t-f-f,
    patch-compile-0-f-t-f, patch-interpret-0-f-t-f, patch-interpret-0-f-f-f,
    patch-interpret-3-f-t-f, patch-interpret-3-f-f-f
- fix bounds checking with an immediate index on immutable vectors,
  fxvectors, strings, and bytevectors
    cpnanopass.ss, 5_5.ms, 5_6.ms, bytevector.ms
- fix a few signatures
    primdata.ss
- more staid and consistent Mf-cross main target
    Mf-cross
- cpletrec now replaces the incoming prelexes with new ones so
  that it doesn't have to alter the flags on the incoming ones, since
  the same expander output is passed through the compiler twice while
  compiling a file with macro definitions or libraries.  we were
  getting away without this just by luck.
    cpletrec.ss
- pure? and ivory? now return #t for a primref only if the prim is
  declared to be a proc, since some non-proc prims are mutable, e.g.,
  $active-threads and $collect-request-pending.
    cp0.ss
- $error-handling-mode? and $eol-style? are now properly declared to
  be procs rather than system state variables.
    primdata.ss
- the new pass $check-prelex-flags verifies that prelex referenced,
  multiply-referenced, and assigned flags are set when they
  should be.  (it doesn't, however, complain if a flag is set
  when it need not be.)  when the new system parameter
  $enable-check-prelex-flags is set, $check-prelex-flags is
  called after each major pass that produces Lsrc forms to verify
  that the flags are set correctly in the output of the pass.
  this parameter is unset by default but set when running the
  mats.
    cprep.ss, back.ss, compile.ss, primdata.ss,
    mats/Mf-base
- removed the unnecessary set of prelex referenced flag from the
  build-ref routines when we've just established that it is set.
    syntax.ss, compile.ss
- equivalent-expansion? now prints differences to the current output
  port to aid in debugging.
    mat.ss
- the nanopass that patches calls to library globals into calls to
  their local counterparts during whole-program optimization now
  creates new prelexes and sets the prelex referenced, multiply
  referenced, and assigned flags on the new prelexes rather than
  destructively setting flags on the incoming prelexes.  The
  only known problems this fixes are (1) the multiply referenced
  flag was not previously being set for cross-library calls when
  it should have been, resulting in overly aggressive inlining
  of library exports during whole-program optimization, and (2)
  the referenced flag could sometimes be set for library exports
  that aren't actually used in the final program, which could
  prevent some unreachable code from being eliminated.
    compile.ss
- added support for specifying default record-equal and
  record-hash procedures.
    primdata.ss, cmacros.ss, cpnanopass.ss, prims.ss, newhash.ss,
    gc.c,
    record.ms
- added missing call to relocate for subset-mode tc field, which
  wasn't burning us because the only valid non-false value, the
  symbol system, is in the static generation after the initial heap
  compaction.
    gc.c
- added a lambda-commonization pass that runs after the other
  source optimizations, particularly inlining, and a new parameter
  that controls how hard it works.  the value of commonization-level
  ranges from 0 through 9, with 0 disabling commonization and 9
  maximizing it.  The default value is 0 (disabled).  At present,
  for non-zero level n, the commonizer attempts to commonize
  lambda expressions consisting of 2^(10-n) or more nodes.
  commonization of one or more lambda expressions requires that
  they have identical structure down to the leaf nodes for quote
  expressions, references to unassigned variables, and primitives.
  So that various downstream optimizations aren't disabled, there
  are some additional restrictions, the most important of which
  being that call-position expressions must be identical.  The
  commonizer works by abstracting the code into a helper that
  takes the values of the differing leaf nodes as arguments.
  the name of the helper is formed by concatenating the names of
  the original procedures, separated by '&', and this is the name
  that will show up in a stack trace.  The source location will
  be that of one of the original procedures.  Profiling inhibits
  commonization, because commonization requires profile source
  locations to be identical.
    cpcommonize.ss (new), compile.ss, interpret.ss, cprep.ss,
    primdata.ss, s/Mf-base,
    mats/Mf-base
- cpletrec now always produces a letrec rather than a let for
  single immutable lambda bindings, even when not recursive, for
  consistent expand/optimize output whether the commonizer is
  run or not.
    cpletrec.ss,
    record.ms
- trans-make-ftype-pointer no longer generates a call to
  $verify-ftype-address if the address expression is a call to
  ftype-pointer-address.
    ftype.ss