Release Notes: A new float type EXT:DOUBLE-DOUBLE-FLOAT is supported. A DOUBLE-DOUBLE-FLOAT uses two DOUBLE-FLOATs to represent a number with >= 106 bits of precision (about 33 digits). Hash tables now support weak value, weak key- and-value, and weak key-or-value tables. LONG-LONG and UNSIGNED-LONG-LONG are recognized types in the C-CALL package for signed and unsigned 64-bit integers. The generational garbage collector has been ported to Darwin/ PPC. Numerous bugs and ANSI-compliance problems have been fixed.
Release Notes: Cross-reference information can now be saved to a FASL file. Annotation support has been added to the pretty printer. Logical path names are supported when using LOAD-OBJECT-FILE. CMUCL recognizes multi-arch (fat) shared libs on Darwin/PPC. It's now possible to load libSystem.dylib. ASIN and ACOS return NaN when given NaN args instead of signaling an error. Many bugfixes and ANSI compliance fixes have been implemented.
Release Notes: This release features stack overflow checking, package locks, weak hash tables, modular arithmetic, helpful local function names, compiler checking of format strings and arguments, callbacks from foreign code to LISP, fwrappers-style encapsulation, enhanced disassemble, heap overflow checking, and enhanced source location recording that works for macros. C functions are visible in backtraces (instead of just “Foreign function call land”). This release also includes numerous bugfixes and ANSI compliance changes, as well as significant performance enhancements to the CLOS implementation.
Release Notes: This release adds a Karatsuba multiplier to give faster multiplication for big bignums, better support for loading foreign libraries, a cross-referencing facility for the compiler, support for generalized function names, access to the hardware cycle counters on Pentium and UltraSPARC processors, largefile support on Linux, significant improvements to the CLOS implementation, and very numerous ANSI compliance fixes and bugfixes. The new platforms now supported are OpenBSD/x86, Linux/Alpha, and IRIX/MIPS. Experimental support for native 64-bit integers is available on UltraSPARC.
Release Notes: OpenBSD/x86 is supported. There are significant performance enhancements, support for hierarchical packages, improvements to disassembly listings on x86 and SPARC, use of branch prediction on UltraSPARC, and stack clearing for improved garbage collection characteristics on SPARC. The size of the dynamic space can be controlled from the command line. This release also includes numerous ANSI compliance fixes (affecting eval-when, proclaim, defmacro, and logical pathnames), and many bugfixes. Code generated with this release is not binary compatible with the previous release.
Release Notes: This release includes many bugfixes, improvements to ANSI CL conformance, compiler enhancements, support for larger heaps, improved logical pathname support, equalp hashtables, faster sequence functions, (complex single-float) and (complex double-float) support on all platforms, better SPARC v9 support, and support for FreeBSD 4.X and Solaris 2.7 and 8. This release is not binary compatible with release 18b: FASL files will need to be recompiled.