X-Git-Url: http://repo.macrolet.net/gitweb/?a=blobdiff_plain;f=NEWS;h=382cc2e18ecfb3823715d5af65241166280be1fc;hb=74a1797f60e26c7adbc491840f89bbaab08e504d;hp=b3c2c6dfa589ba9ddbd0b14a3e28486f357edf51;hpb=2e002dae2f9a3c64f147ca651751ed833806ad5e;p=sbcl.git diff --git a/NEWS b/NEWS index b3c2c6d..382cc2e 100644 --- a/NEWS +++ b/NEWS @@ -1,5 +1,80 @@ ;;;; -*- coding: utf-8; -*- +changes in sbcl-1.0.19 relative to 1.0.18: + * bug fix: compiler no longer makes erronous assumptions in the + presense of non-foldable SATISFIES types. + * fixed some bugs revealed by Paul Dietz' test suite: + ** interval arithmetic during type derivation used inexact integer + to single-float coercions. + ** arithmetic operations involving large integers and single + floats give the same results in compiled and interpreted code. + ** deriving the result type of COERCE no longer signals an error + if the derived type of the second argument is a MEMBER type + containing invalid type specifiers. + ** ADJOIN with constant NIL as second argument works correctly. + ** FUNCTION-KEYWORDS was called SB-PCL::FUNCTION-KEYWORD-PARAMETERS. + +changes in sbcl-1.0.18 relative to 1.0.17: + * minor incompatible change: SB-SPROF:WITH-PROFILING now by default + profiles only the current thread. + * minor incompatible change: changes to SYMBOL-VALUE of constants + defined with DEFCONSTANT now signal an error. + * enhancement: SB-SPROF now has support for wallclock profiling, + and is also able to profile specific threads. REPORT output + has also additional sorting options. + * enhancement: better pretty-printing of DEFPACKAGE forms. (Thanks + to Michael Weber) + * optimization: structure allocation has been improved + ** constructors created by non-toplevel DEFSTRUCTs are ~40% faster. + ** out of line constructors are ~10% faster. + ** inline constructors are ~15% faster. + ** inline constructors are capable of dynamic extent allocation + (generally on x86 and x86-64, in some cases on other platforms + as well.) + * optimization: simple uses of HANDLER-CASE and HANDLER-BIND no + longer cons. + * optimization: file compiler is now able to coalesce non-circular + lists, non-base strings, and bit-vectors. Additionally, constants + are never referenced through SYMBOL-VALUE at runtime. + * optimization: code defining methods on PRINT-OBJECT (and other + generic functions in the COMMON-LISP package) now loads faster. + * bug fix: EAI_NODATA is deprecated since RFC 3493. Stop using it + in sb-bsd-sockets. + * bug fix: if COMPILE-FILE aborts due to an unwind, the partial + fasl is now deleted. (reported by Attila Lendvai) + * bug fix: READ-LINE always returned NIL for the last line in files. + (reported by Yoshinori Tahara) + * bug fix: more accurate disassembly annotations of foreign function + calls. (thanks to Andy Hefner) + * bug fix: trimming non-simple strings and non-string string + designators when the there is nothing to trim works properly. + (thanks to James Knight) + * new feature: SB-POSIX bindings for mlockall, munlockall, and setsid. + (thanks to Travis Cross) + * fixed some bugs revealed by Paul Dietz' test suite: + ** NIL is a valid function name (regression at 1.0.13.38) + ** FILL on lists was missing its return value (regression at 1.0.12.27) + ** STRING-TRIM, STRING-LEFT-TRIM, and STRING-RIGHT-TRIM did not respect + fill pointers properly (regression at 1.0.12.23) + ** STRING-TRIM, STRING-LEFT-TRIM, and STRING-RIGHT-TRIM did not respect + displacement indices properly (regression at 1.0.12.23) + changes in sbcl-1.0.17 relative to 1.0.16: + * temporary regression: user code can no longer allocate closure + variable storage on stack, due to bug 419 without explicitly + requesting it. Please consult sbcl-devel for advice if you need to + use this feature in the meanwhile. + * new feature: runtime argument --control-stack-size can be used to + adjust thread default control stack size. + * enhancement: improved TIME output + ** all times are reported using the measured accuracy (milliseconds + for real and GC times, microseconds for everything else.) + ** processor cycle counts on x86 and x86-64. + ** interpreted forms are counted for both evaluator modes. + ** number of lambdas converted by the compiler is reported. + ** CPU percentage report (computed from real and total run time.) + ** more comprehensive run time reporting, using a condenced format + ** interperted form, lambda, and page fault counts are omitted + when zero. * optimization: ADJOIN and PUSHNEW are upto ~70% faster in normal SPEED policies. * optimization: APPEND is upto ~10% faster in normal SPEED policies. @@ -7,6 +82,12 @@ changes in sbcl-1.0.17 relative to 1.0.16: in normal SPEED policies. * optimization: NCONC no longer needs to heap cons its &REST list in normal SPEED policies. + * bug fix: SB-FLUID build feature no longer breaks the build. (thanks + to Sidney Markowitz) + * bug fix: UNION and NUNION work with :TEST-NOT once more, + regression since 1.0.9.1. (thanks to Eric Marsden) + * bug fix: result of MAKE-ARRAY can be stack allocated - regression + since 1.0.15.36. (thanks to Paul Khuong) * bug fix: LAST when always returned the whole list when given a bignum as the second argument. * bug fix: dynamic extent allocation of nested lists and vectors @@ -17,6 +98,10 @@ changes in sbcl-1.0.17 relative to 1.0.16: platform word lengths. * bug fix: more correct assembler syntax for GNU binutils 2.18.50.0.4 support. (thanks to Marijn Schouten) + * bug fix: fix ECASE warnings from CMUCL-as-xc-host. (reported by + Andreas Franke) + * bug fix: the fopcompiler can handle LOCALLY forms (with no + declarations) successfully. (reported by Attila Lendvai) changes in sbcl-1.0.16 relative to 1.0.15: * minor incompatible change: revert the changes to sb-posix's error