- Dec 23, 2013
-
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
translation outside of the test.
-
Raymond Toy authored
-
Raymond Toy authored
o Derive the value of asin(2), lest we forget again why it is the way it is. o Use the formulas from the CLHS in the docstrings.
-
Raymond Toy authored
-
- Dec 22, 2013
-
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
- Dec 21, 2013
-
-
Raymond Toy authored
-
Raymond Toy authored
Clean up comments for asin too.
-
Raymond Toy authored
-
Raymond Toy authored
code/irrat-dd.lisp o Make REDUCE-ARG return an extra result. Since the reduction returns 3 double's, return a double-double result and the third double result for extra accuracy o Update dd-%tan and dd-%%tan to take the extra arg. o Add new constant dd-pi/4-lo. o Increase accuracy of tan by using the relationship tan(pi/4-y)=(1-tan(y))/(1+tan(y)). tests/trig.lisp: o Update the allowed error threshold for two tests to reflect the increased accuracy.
-
Raymond Toy authored
-
Raymond Toy authored
separately.
-
Raymond Toy authored
-
Raymond Toy authored
-0w0).
-
- Dec 20, 2013
-
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
options just like for DISASSEM:DISASSEMBLE. * DISASSEMBLE-SEGMENTS actually implements the base, case and radix options. * DISASSEMBLE doesn't implement the options but calls DISASSEMBLE-SEGMENTS with the appropriate args. * DISASSEMBLE-FUNCTION, DISASSEMBLE-MEMORY, and DISASSEMBLE-CODE-COMPONENT accepts the new keyword args. * Update the docstrings to indicate the new keyword args.
-
- Dec 18, 2013
-
-
Raymond Toy authored
comments.
-
Raymond Toy authored
x86-arch.c and ppc-arch.c
-
Raymond Toy authored
-
Raymond Toy authored
-
Raymond Toy authored
-
- Dec 17, 2013
-
-
Raymond Toy authored
-
- Dec 15, 2013
-
-
Raymond Toy authored
-
Raymond Toy authored
* Add CC_REM_PIO2 variable when compiling e_rem_pio2.c and k_rem_pio2.c. * On linux and freebsd, set CC_REM_PIO2 to -ffloat-store so that proper rounding happens. Not needed on Darwin because Darwin always uses sse2 and not x87.
-
Raymond Toy authored
* code/irrat.lisp * Fix typo in kernel-tan. * tests/trig.lisp * Add test case.
-
Raymond Toy authored
* Remove unneeded package specifier for %ieee754-rem-pi/2 * Add some comments for %tan.
-
Raymond Toy authored
-
Raymond Toy authored
used for all platforms. code/irrat.lisp:: * Implement %SINCOS compiler/float-tran.lisp:: * Update deftransforms for CIS. %SINCOS can be used on any platform. tests/trig.lisp: * Add tests to verify %sincos returns exactly the same values as for sin and cos.
-
Raymond Toy authored
code/irrat.lisp:: * Add Lisp implementation for sin, cos, and tan, based on code from fdlibm. Requires the C reduction routines. Only working so far on systems that already include the reduction routies. tests/trig.lisp:: * Tests for the new sin, cos, and tan functions. Tests pass on x86/darwin.
-
Raymond Toy authored
src/contrib/rt:: * Add RT code, including asdf. src/code/module.lisp:: * Add RT as a module
-
- Dec 13, 2013
-
-
Raymond Toy authored
These routines did argument reduction, but since we use __kernel_rem_pio2 to do accurate argument reduction, the argument reduction in these routines is a waste of time. This greatly simplifies the routines to just the polynomial (or rational) approximations.
-
Raymond Toy authored
* Allow the second arg to be a descriptor for the basic arithmetic operations on single floats and double floats. Previously, the boxed number would be loaded to a temporary reg for the operation. This saves the load and the extra register. * Allow a descriptor for the second arg in the comparison vops, the float conversion vops and the float truncation vops. * Allow descriptor for sqrt vop.
-