- Dec 02, 2008
-
-
rtoy authored
valid on Darwin because the sigcontext structure is no longer defined in Lisp, so we can't check this way anymore. Hence, remove the code to clean things up a bit.
-
- Nov 12, 2008
-
-
rtoy authored
sse2-packed-2008-11-12).
-
- Oct 24, 2008
-
-
rtoy authored
o Forgot to implement FLOAT-PRECISION for double-doubles. general-info/release-19f.txt: o Document it.
-
- Oct 22, 2008
-
-
rtoy authored
o Instead of a continuable error, use a restart to allow the user to continue the calculation with the option of updating the limit to the new power. Should we also increase the initial limit to something larger? Should we just eliminate this altogether? Or maybe allow the *intexp-maximum-exponent* to be nil to mean no limit at all?
-
rtoy authored
-
- Oct 21, 2008
-
-
rtoy authored
and *truncate-y*.
-
rtoy authored
*truncate-x* and *truncate-y*. Accessing these in tight loops hurts performance. So pass truncate-x and truncate-y as parameters to the routines that need them. Reorder the truncate support routines so we can block-compile all of the routines for truncate. With these changes, we see the following changes according to cl-bench. The first column and reference is 2008-10 snapshot (darwin), and the next column is the ratio of this new code to the reference. So we see improvements from 10% to 50%. BIGNUM/ELEM-100-1000 [ 0.23] 0.84 BIGNUM/ELEM-1000-100 [ 0.50] 0.89 BIGNUM/ELEM-10000-1 [ 0.51] 0.90 BIGNUM/PARI-100-10 [ 0.05] 0.63 BIGNUM/PARI-200-5 [ 0.22] 0.51 PI-DECIMAL/SMALL [ 2.44] 0.70 PI-DECIMAL/BIG [ 2.24] 0.73 PI-ATAN [ 1.07] 1.00 PI-RATIOS [ 2.56] 0.88
-
- Oct 08, 2008
-
-
rtoy authored
Take the rule of float precision contagion (CLHS 12.1.4.4) to also mean that the result should be as accurate as the most accurate argument. Effectively, all args are coerced to the highest precision first before computing expt. There's a simple test program to check that every case is covered with the expected precision. (I think). (defun test-expt () (dolist (base '(2 2f0 2d0 2w0) t) (dolist (power '(1/2 .5f0 .5d0 .5w0)) (flet ((test-it (b p a eps expected) (let* ((res (expt b p)) (absdiff (abs (- res a)))) (unless (or (typep (realpart res) (type-of expected)) (<= absdiff (* 10 eps))) (format t "FAILED: ~A^~A = ~A (~A)~%" b p res a))))) ;; Compute base^power. (let* ((expected-type (let ((prod (* base power))) (if (rationalp prod) 1f0 prod))) (eps (etypecase expected-type ((or rational single-float) single-float-epsilon) (double-float double-float-epsilon) (double-double-float 1w-31))) (answer (sqrt (float base expected-type)))) (test-it base power answer eps expected-type) (test-it base (complex power) answer eps expected-type) (test-it (complex base) power answer eps expected-type) (test-it (complex base) (complex power) eps answer expected-type))))))
-
- Oct 03, 2008
- Sep 24, 2008
-
-
cshapiro authored
an available devices on our own. This resolves an issue with Linux systems that do not populate /dev with BSD pseudo-terminal devices.
-
- Sep 23, 2008
-
-
rtoy authored
are. This makes the interpreter produces a message when the compiler would produce a message.
-
- Sep 16, 2008
-
-
rtoy authored
UNIX-SYSINFO.
-
- Sep 07, 2008
- Sep 02, 2008
-
-
rtoy authored
o Update link new initializer.
-
- Aug 31, 2008
-
-
rtoy authored
random number state. This should give better random numbers since we initialize the state with 627 "truly" random values. Previously, only one word was used, which meant that the state vector really only had 32 bits of randomness.
-
- Aug 16, 2008
-
-
rtoy authored
returning the fixed string on x86/darwin.
-
- Aug 15, 2008
-
-
rtoy authored
followed by lower case letters. This is needed to match what format ~:C produces. (Issue caught by ansi-tests.) Add a comment here so we don't forget.
-
- Aug 12, 2008
-
-
rtoy authored
o Remove definition of vm::ash-left-mod32. (It's already in numbers.lisp.) code/numbers.lisp: o Declaim vm::ash-left-mod32 as inline to work around x86 issue where vm::ash-left-mod32 never gets translated to a vop.
-
- Jul 31, 2008
-
-
rtoy authored
-
- Jul 30, 2008
-
-
rtoy authored
code/internet.lisp: o Add BIND-INET-SOCKET o Allow CONNECT-TO-INET-SOCKET to allow binding the new socket to a local address. code/exports.lisp: o Update export list for BIND-INET-SOCKET. docs/cmu-user/internet.tex: o Add and update docs for BIND-INET-SOCKET and CONNECT-TO-INET-SOCKET. o Add docs for ACCEPT-NETWORK-STREAM and OPEN-NETWORK-STREAM. o Fix docs for CREATE-UNIX-LISTENER---the function doesn't actually have REUSE-ADDRESS keyword arg. general-info/release-19f.txt: o Update with new network info.
-
- Jul 21, 2008
-
-
rtoy authored
-
- Jul 18, 2008
-
-
rtoy authored
-
- Jul 15, 2008
- Jul 14, 2008
-
-
rtoy authored
code/loop.lisp: o In LOOP-TRANSLATE, create block for loop named NIL by checking for the end of the list, not for NIL in the list. o In LOOP-DO-NAMED, don't append NIL to *LOOP-NAMES* so we can create blocks named NIL. (Why was this being done before?) general-info/release-19f.txt: o Update
-
rtoy authored
space so everything gets spaced correctly. Don't print the space in FORMAT-PRINT-CARDINAL-AUX.
-
- Jun 27, 2008
-
-
rtoy authored
I just removed the leading space from each name, and modified FORMAT-PRINT-CARDINAL-AUX to print the space instead.
-
- Jun 26, 2008
-
-
rtoy authored
o Recursively call RANDOM for the integer case to let the deftransform handle the case. compiler/float-tran.lisp: o Remove comment about modifying RANDOM to match the deftransform. We're automatically in sync now.
-
- Jun 25, 2008
-
-
rtoy authored
-
- Jun 24, 2008
-
-
rtoy authored
Fix bug in deftransform where the case for x86 and num-high > random-fixnum-max was never used. We now always use the bignum multiply code for all platforms and use that for all cases for which the deftransform applies. Simplify the constant-continuation case too so we don't have to the check. I think the bignum multiply version does not have the bias problem that rem has. code/rand-mt19937.lisp: Change the RANDOM function to compute the same value as the deftransform would. general-info/release-19f.txt: Update Make a note that both the function and the deftransform should be synchronized to match. This change also changes the sequence of numbers produced.
-
- Jun 23, 2008
-
-
rtoy authored
ACCEPT-NETWORK-STREAM. From Paul Foley.
-
rtoy authored
o Not that it really matters, since they're not printed and case-insensitive on read, but the control character names should really be upper-case o Also cleans up some anachronistic mention of character fonts and bits.
-
rtoy authored
From Paul Foley.
-
- Jun 20, 2008
-
-
rtoy authored
I've moved some slots out of external-format so they can be shared between external-formats that are identical in all but some variables. Also fixed a bug in octets-to-string that made it return one character short, and used char-code-limit instead of #x100 to determine when octets-to-char returns a "?", so now it'll work without change on 8 or 16 bit lisps.
-
- Jun 19, 2008
-
-
rtoy authored
more like the actual call (var filespec &rest open-args) instead of (var &rest open-args). Fix up the macro documentation too since that didn't really match the arguments very well either.
-
rtoy authored
Basically, don't use delete-duplicates but do it ourselves since the list is already sorted. From Lynn Quam, cmucl-imp, 2008/06/05, slightly modified. code/filesys.lisp: o New DIRECTORY function general-info/release-19f.txt: o Update
-
rtoy authored
"library:ext-formats/". This makes it easier to add new directories where external formats can be found. The previous use made it difficult because the formats had to be in the subdirectory ext-formats. save.lisp: o Create and initialize new search-list. extfmts.lisp: o Use the new search-list instead of "library:ext-formats/".
-