diff options
| author | Paul Eggert <eggert@cs.ucla.edu> | 2018-03-23 12:57:39 -0700 |
|---|---|---|
| committer | Paul Eggert <eggert@cs.ucla.edu> | 2018-03-23 12:59:18 -0700 |
| commit | 42e7e267e5487f60f4d72e1b5c5cba001ba4d704 (patch) | |
| tree | 8c44c5c43eb779d4478b45b66095e94000b78032 /test/lisp/emacs-lisp/bytecomp-tests.el | |
| parent | d0881374734310eb77526823c0acad1fe556f41b (diff) | |
| download | emacs-42e7e267e5487f60f4d72e1b5c5cba001ba4d704.tar.gz | |
Avoid Fortran-style floating-point optimization
When optimizing arithmetic operations, avoid optimizations that
are valid for mathematical numbers but invalid for floating-point.
For example, do not optimize (+ 1 v 0.5) to (+ v 1.5), as they may
not be the same due to rounding errors. In general,
floating-point numbers cannot be constant-folded, since that would
make .elc files platform-dependent.
* lisp/emacs-lisp/byte-opt.el (byte-optimize-associative-math):
Do not optimize floats.
(byte-optimize-nonassociative-math, byte-optimize-approx-equal)
(byte-optimize-delay-constants-math, byte-compile-butlast)
(byte-optimize-logmumble):
Remove; no longer used.
(byte-optimize-minus): Do not optimize (- 0 x) to (- x).
(byte-optimize-multiply): Do not optimize (* -1 x) to (- x).
(byte-optimize-divide): Do not optimize (/ x -1) to (- x).
(logand, logior, logxor): Optimize with byte-optimize-predicate
instead of with byte-optimize-logmumble.
* test/lisp/emacs-lisp/bytecomp-tests.el:
(byte-opt-testsuite-arith-data): Add a couple of test cases.
Diffstat (limited to 'test/lisp/emacs-lisp/bytecomp-tests.el')
| -rw-r--r-- | test/lisp/emacs-lisp/bytecomp-tests.el | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/test/lisp/emacs-lisp/bytecomp-tests.el b/test/lisp/emacs-lisp/bytecomp-tests.el index 6ae7cdb9f9c..7330c676140 100644 --- a/test/lisp/emacs-lisp/bytecomp-tests.el +++ b/test/lisp/emacs-lisp/bytecomp-tests.el @@ -38,8 +38,7 @@ (let ((a 3) (b 2) (c 1.0)) (/ a b c)) (let ((a (+ 1 (expt 2 -64))) (b (expt 2 -65))) (+ a -1 b)) (let ((a (+ 1 (expt 2 -64))) (b (expt 2 -65))) (- a 1 (- b))) - ;; This fails. Should it be a bug? - ;; (let ((a (expt 2 -1074)) (b 0.125)) (* a 8 b)) + (let ((a (expt 2 -1074)) (b 0.125)) (* a 8 b)) (let ((a 1.0)) (* a 0)) (let ((a 1.0)) (* a 2.0 0)) (let ((a 1.0)) (/ 0 a)) @@ -244,6 +243,9 @@ (let ((a 3) (b 2) (c 1.0)) (/ a b c 0)) (let ((a 3) (b 2) (c 1.0)) (/ a b c 1)) (let ((a 3) (b 2) (c 1.0)) (/ a b c -1)) + + (let ((a t)) (logand 0 a)) + ;; Test switch bytecode (let ((a 3)) (cond ((eq a 1) 'one) ((eq a 2) 'two) ((eq a 3) 'three) (t t))) (let ((a 'three)) (cond ((eq a 'one) 1) ((eq a 2) 'two) ((eq a 'three) 3) |