Published release SBCL 2.4.2 (Steel Bank Common Lisp), a free implementation of the Common Lisp programming language. The project code is written in Common Lisp and C languages, and distributed by under BSD license.
In the new release:
- Self-compilation on x86-64 Linux systems now produces bit-identical cross-compiled fasls where the build host is cmucl, ccl, clisp, or sbcl itself.
- Optimizations:
- A large number of internal tables, particularly those related to Unicode support, have been converted to use collision-free hash functions (perfect), which improved speed and reduced size.
- FIND, POSITION, ASSOC, and RASSOC with constant-sequence arguments containing symbols as keys now compile for lookups using collision-free hashes.
- On x86-64 systems, the compiler eliminates the jump-to-jump transition.
- Bug fixes:
- The ability to inherit from both SEQUENCE and SB-MOP:FUNCALLABLE-STANDARD-OBJECT has been restored.
- COERCE will not convert form lambdas to functions if the given type name is a subclass of FUNCTION.
- LOG with double-float and rational arguments (in any order) does not lose precision through the intermediate single-float argument.
- LOG to base 2 integer power of two will most likely give a mathematically accurate answer.
- LOG on rational numbers close to one with a numerator or denominator close to a power of two will use log1p and thus will lose less precision.
- The external utf-8 format with Unix line endings updates its character size information when applying the fast path for ASCII character buffers.
- The contents of probably no longer valid dynamically existing cons in the PRINT-OBJECT method for THREAD objects were not printed.
- Always place external entry points for functions before any other local functions.
- Remove inactive optimization notes for quoted forms and single APPEND calls at high speed.
- Infinite loop in COPY-SEQ on zero-length arrays with NIL element type.
- Fixed compilation of non-top-level structure constructors.
- Fixes in SB-SIMD:
- Improved bounds checking in SB-SIMD.
- Fix SB-SIMD AVX f64.4-reverse.
- Fix SB-SIMD shuffling on AVX and SSE2.
- Fixed lifetime in sse+xmm0 VOPs.
Thanks for reading: