X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f Sender: rich AT phekda DOT freeserve DOT co DOT uk Message-ID: <3F76CD05.D44CA7DC@phekda.freeserve.co.uk> Date: Sun, 28 Sep 2003 12:59:01 +0100 From: Richard Dawe X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.23 i586) X-Accept-Language: de,fr MIME-Version: 1.0 To: djgpp-workers AT delorie DOT com Subject: Re: Integrating K. B. Williams's maths functions References: <3F5B82C8 DOT 2EE2A73D AT phekda DOT freeserve DOT co DOT uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Reply-To: djgpp-workers AT delorie DOT com Hello. I've put a new work-in-progress patch here: http://www.phekda.freeserve.co.uk/richdawe/kbw-maths-20030928.diff.gz It's around a megabyte uncompressed. I didn't think there was much point posting it to djgpp-workers. I have no idea if it works. It compiles, at least. You can find some notes I made here: http://www.phekda.freeserve.co.uk/richdawe/kbw-maths-notes.txt New notes, since the last release: * Ignore standard.results for now - we don't have a standard set of C99 FP functions yet! * Code for: cosl, erfcl, y0l, y1l, ynl? * Documentation for: jy0l, jy1l, jynl, nonesuchl, sinl, tanl? * Integrated docs, except those for functions that have not been integrated. - Changed portability info to "!ansi-c89, ansi-c99" from "ansi, posix". * modfl not integrated. - DJGPP already has modfl. * mdnorm prototype in *.h inconsistent with the one in qsqrta.c. - Is this a problem? Causes "make all" to fail in tests/cygnus/tgen/qfloat. * Warnings in tests/cygnus/tgen/qfloat: bash-2.04$ make gcc -Wall -march=pentium -O2 -c qexp2.c gcc -Wall -march=pentium -O2 -c qflt.c qflt.c: In function `e113toq': qflt.c:1771: warning: passing arg 1 of `qclear' from incompatible pointer type qflt.c:1799: warning: assignment from incompatible pointer type qflt.c:1820: warning: passing arg 1 of `shift' from incompatible pointer type qflt.c:1823: warning: passing arg 1 of `normlz' from incompatible pointer type qflt.c:1824: warning: passing arg 1 of `qclear' from incompatible pointer type qflt.c:1828: warning: passing arg 1 of `qmov' from incompatible pointer type ar -rs qfloatx.a qexp2.o qflt.o * Need to integrate changes to tests/cygnus/tgen/makefile. * Rename DblMath.[ch], FltMath.[ch], LdblTest.h to have lower-case names, to be consistent with other files in DJGPP CVS. * keepthis.h? Build broken, until I have that file. KB: I've integrated your documentation, new qfloat code and testzips that you sent me. As far as I know, I'm up-to-date with everything you've sent me. (Except , which I'm fixing up. And a couple of makefiles, which I'm merging manually.) Bye, Rich =] -- Richard Dawe [ http://www.phekda.freeserve.co.uk/richdawe/ ]