[git] CMU Common Lisp annotated tag snapshot-2014-08 created. snapshot-2014-08
Raymond Toy
rtoy at common-lisp.net
Fri Aug 8 04:31:37 UTC 2014
This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "CMU Common Lisp".
The annotated tag, snapshot-2014-08 has been created
at 3d3259cd2f0d437388b69ecf50feedc73246096d (tag)
tagging 98319835e05b93ea62e17f6bc3da4144f96ac2c5 (commit)
replaces snapshot-2014-06
tagged by Raymond Toy
on Thu Aug 7 21:30:43 2014 -0700
- Log -----------------------------------------------------------------
Snapshot 2014-08
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.13 (Darwin)
iEYEABECAAYFAlPkUnoACgkQJ5IjUmgZO7JzgwCgt4mPwW0Qx0sI3qtsXPOttVrq
klAAnAxpd/KCElQ6ggJf82ZSVlhRBOJm
=4mEx
-----END PGP SIGNATURE-----
Raymond Toy (101):
Change max gen to GC to 3; add inteface to set it.
Fix declaration of min-av-mem-age in generation-stats.
Import fdlibm trig functions; as is, from netlib.
Update to use unions, using the style of e_rem_pio2.c.
Add declarations for the fdlibm trig functions.
Compile fdlibm trig functions on linux and darwin.
Darwin doesn't need fdlibm because the system library is good.
Add implmeentation of sincos in C.
Comment out all of the trig functions in lisp in favor of calling out
Oops. Didn't comment out quite enough.
Add sincos for x86 linux.
For consistency with sincos, we need to include the fdlibm trig
ppc doesn't have sincos in libm so use our own version.
Oops. Need k_sin and k_cos for sincos.
Add rest of fdblim trig functios so that sincos is consistent with sin
Put back the deftransforms for sin/cos/tan so we can call out directly
Add fdlibm trig for sparc.
Add fdlibm trig routines for NetBSD.
Fix stupid typo that causes sin to converted to asin.
All platforms have sincos now, so remove the deftransform for cis that
Add declaration to ignore unused variable.
Rename sin/cos/tan/sincos so we don't collide with libm.
Add include guard.
Some cleanup of the trig code.
Comment code, add license header, and conform to cmucl C style.
Ignore emacs ~ files and fasls.
Finally remove the Lisp implementation of the trig functions that are
Add expm1 to match log1p. All platforms seem to have expm1 as well.
Remove the mathcalls section. It's not used and some of it (log1p) is
Update to 3.1.3.
Fix ticket ##104: Source location for define-condition
Add README to document the naming convention of boot files.
Add info for cross-compile scripts.
CLEAR-OUTPUT should call STREAM-CLEAR-OUTPUT for Gray streams.
Implement CLEAR-OUTPUT for FD-STREAM's.
Add tests for CLEAR-OUTPUT.
Import log1p from fdlibm, as is.
Import expm1 from fdlibm, as is.
Modify to use unions and change name to include fdlibm prefix.
Compile s_log1p and s_expm1.
Use fdlibm versions of log1p and expm1.
Import pow (ieee754_pow) from fdlibm, as is.
Update to use unions.
Compile e_pow as a part of lisp.
Use __ieee754_pow instead of pow.
Compile the new fdlibm routines.
Compile the new fdlibm routines.
Compile the new fdlibm routines.
Add fdlibm routines e_exp and e_log, as is.
Update to use unions.
Compile fdlibm routines e_exp.c and e_log.c
Use the fdlibm routines for exp and log.
Import inverse trig functions from fdlibm, as is.
Use unions to access the high and low parts of a double.
Compile the asin, acos, and atan routines.
Use the fdlibm asin, acos, and atan routines.
Initialize t to get rid of a compiler warning from clang.
Initialize k to get rid of a compiler warning from clang.
Remove the sccsid variable.
Add some braces to silence the warning from clang about dangling else
Add some braces to silence the warning from clang about dangling else
Import hyperbolic functions from fdlibm, as is.
Use unions to access the high and low parts of a double and update
Add parens around && expressions to silence a clang warning.
Compile the fdlibm hyperbolic functions.
Use fdlibm hyperbolic functions instead of libm.
Import inverse hyperbolic functions from fdlibm, as is.
Fix up the inverse hyperbolics.
Compile the fdlibm inverse hyperbolics.
Use the fdlibm inverse hyperbolic functions.
Import atan2 function from fdlibm, as is.
Fix up atan2 by using unions.
Use fdlibm_atan instead of atan.
Compile the fdlibm atan2.
Use fdlibm atan2.
Use correct flags to generate sse2 instructions in the C code and
Clean up config files.
Remove the list of fdlibm files from here because they're in
Forgot to remove e_rem_pio2 and friends.
-march=pentium4 imples -msse2, so drop the latter.
Update netbsd config. Untested.
Forgot to remove e_rem_pio2 and friends from a few more Config files
Remove CC_REM_PIO2 and add new CPPFLAGS to force compiling with sse2
add -mtune=generic to get optimization for the most common x86
Update from commit logs.
Initialize c to get rid of a compiler warning on Linux with gcc 4.4.1.
Add parens around operands of | to silence a compiler warning with gcc
Initilize hi and lo to get rid of a compiler warning from gcc 4.4.1 on
Fix aliasing issue noted by gcc 4.4.1 on Linux.
Fix aliasing issue noted by gcc 4.4.1 on Linux.
Oops. Fix typo by removing the extraneous comma that was left in.
Fix bug in DOUBLE-FLOAT-BITS vop.
Remove -mtune=generic from Config.x86_common.
Update LDFLAGS for NetBSD so motifd will build and run correctly on
Oops. x86-assem needs -m32.
Document that the argument/result timespec starts at 0, not 1.
Simplify SSE2-FLOATING-POINT-MODES vop by removing an unneeded
Addresses ticket #84 for Linux.
Fix typo and add -m32 to LDFLAGS.
Remove commented-out LDFLAGS.
Regenerated and updated.
-----------------------------------------------------------------------
hooks/post-receive
--
CMU Common Lisp
More information about the cmucl-cvs
mailing list