Custom Query (1030 matches)
Results (274 - 276 of 1030)
| Ticket | Resolution | Summary | Owner | Reporter |
|---|---|---|---|---|
| #9 | fixed | problem compiling CLASP | ||
| Description |
Hi Gary, I found another problem related to the compilation of CLASP. OpenMCL 1.1-pre-070408: darwin-ppc32, darwin-ppc64, linux-x86-64
CLASP> (defun invert-matrix (matrix &optional into-matrix)
"If matrix is singular returns nil, else returns its inverse.
If into-matrix is supplied, inverse is returned in it,
otherwise a new array is created."
(let* ((dims (array-dimensions matrix))
(rows (car dims))
(cols (cadr dims)))
(unless (= rows cols)
(error () "Matrix ~s not square" matrix))
(if (null into-matrix)
(setq into-matrix (make-array dims))
(let ((dims-into (array-dimensions into-matrix)))
(unless (and (= rows (car dims-into)) (= cols (cadr dims-into)))
(error () "into-matrix ~s does not have same dimensions as matrix ~s" into-matrix
matrix))))
(let ((xcols (+ rows cols))
(vv (make-array rows)))
(dotimes (r rows)
(let ((v (make-array xcols :initial-element #o0)))
(setf (aref v (+ r rows)) #o1)
(dotimes (c cols)
(setf (aref v c) (aref matrix r c)))
(setf (aref vv r) v)))
(dotimes (r rows)
(let ((imax #o-1)
(vmax #o-1))
(do ((i r (1+ i)))
((= i rows))
(let ((v (abs (aref (aref vv i) r))))
(cond
((> v vmax) (setq vmax v) (setq imax i)))))
(cond ((zerop vmax) (return-from invert-matrix ())))
(cond ((not (= r imax))
(let ((exch-temp (aref vv r)))
(setf (aref vv r) (aref vv imax))
(setf (aref vv imax) exch-temp)))))
(let ((pivot-row (aref vv r)))
(do ((d (aref pivot-row r))
(i (1+ r) (1+ i)))
((= i xcols))
(setf (aref pivot-row i)
(/ (aref pivot-row i) d)))
(do ((i (1+ r) (1+ i)))
((= i rows))
(let ((row (aref vv i)))
(do ((d (aref row r))
(j (1+ r) (1+ j)))
((= j xcols))
(setf (aref row j) (- (aref row j) (* d (aref pivot-row j)))))))))
(do ((r (1- rows) (1- r)))
((= r #o0))
(let ((pivot-row (aref vv r)))
(do ((i (1- r) (1- i)))
((< i #o0))
(do ((d (aref row r))
(j rows (1+ j)))
((= j xcols))
(setf (aref row j) (- (aref row j) (* d (aref pivot-row j)))))))))
(dotimes (r rows)
(let ((v (aref vv r)))
(dotimes (c cols)
(setf (aref into-matrix r c) (aref v (+ c cols)))))))
(values into-matrix))) =>
value (VARIABLE . "ccl:l1;l1-typesys.lisp.newest") is not of the expected type (SATISFIES CCL::CTYPE-P). Backtrace: 0: (CCL::TYPE-SPECIFIER) 1: (CCL::ACODE-FORM-TYPE '(#<VAR #x30004147ECAD>) #<VAR #x30004147EC2D>) 2: (CCL::NX1-NOTE-VAR-BINDING '(25674 (24614 #<VAR #x30004147F7DD>) (24614 #<VAR #x30004144330D>)) '(#<VAR #x30004147ECAD>)) 3: (CCL::NX1-NOTE-VAR-BINDINGS '(#<VAR #x30004147EC2D> #<VAR #x30004147ECAD>) 'NIL) 4: (CCL::NX1-LET '(LET ((D #) (I #)) (TAGBODY (GO #1=#:G38) #:G37 (TAGBODY #) (PSETQ I #) #1# (UNLESS # #))) #<LEXICAL-ENVIRONMENT #x30004147F48D>) 5: (CCL::NX1-COMBINATION #<LEXICAL-ENVIRONMENT #x30004147F48D> '(LET ((D #) (I #)) (TAGBODY (GO #1=#:G38) #:G37 (TAGBODY #) (PSETQ I #) #1# (UNLESS # #)))) 6: (CCL::NX1-FORM '(LET ((D #) (I #)) (TAGBODY (GO #1=#:G38) #:G37 (TAGBODY #) (PSETQ I #) #1# (UNLESS # #)))) 7: (CCL::NX1-BLOCK '(BLOCK NIL (LET (# #) (TAGBODY # #:G37 # # #:G38 #))) #<LEXICAL-ENVIRONMENT #x30004147F96D>) 8: (CCL::NX1-COMBINATION #<LEXICAL-ENVIRONMENT #x30004147F96D> '(DO ((D #) (I # #)) ((= I XCOLS)) (SETF (AREF PIVOT-ROW I) (/ # D)))) 9: (CCL::NX1-FORM '(DO ((D #) (I # #)) ((= I XCOLS)) (SETF (AREF PIVOT-ROW I) (/ # D)))) 10: (CCL::NX1-FORMLIST '((DO (# #) (#) (SETF # #)) (DO (#) (#) (LET # #)))) 11: (CCL::NX1-PROGN-BODY 'NIL) 12: (CCL::NX1-ENV-BODY '((DO (# #) (#) (SETF # #)) (DO (#) (#) (LET # #))) #<LEXICAL-ENVIRONMENT #x3000414430FD>) 13: (CCL::NX1-LET* '(LET* ((PIVOT-ROW #)) (DO (# #) (#) (SETF # #)) (DO (#) (#) (LET # #))) #<LEXICAL-ENVIRONMENT #x3000414430FD>) This function can be found in clasp/dev/matrices.lisp. This is old code (pre-1990) that compiles fine under ACL, MCL, LispWorks, and the OpenMCL 2006-10-24 release. I've attached a tarball of the CLASP asdf system (unix-newlines). Thanks, jjm |
|||
| #10 | fixed | problem compiling ppc64-target: typo in defcallback xcmain | ||
| Description |
Hi Gary, I found a typo in ccl/level-1/ppc-transport.lisp in the cvs snapshot I checked out on 16 April 2007 at 12:00 pm. in #+ppc64-target defcallback xcmain, :single-doubleword should replace :single-doublewod jjm |
|||
| #11 | fixed | hand-rolled :darwin64-thread-state64 struct isn't mis-aligned, but should be. | ||
| Description |
Apple has decided that renaming some structures dealing with thread contexts and the fields in those structures would help to ensure compatibility with some C standard or other; the names of these fields therefore depend on what header files one looks at (and what -preserve-broken-legacy-behavior options are in effect.) Lisp code occasionally needs to poke around in these structures, and therefore defines its own set of record types with its own field names "manually". The manual definition of :darwin-ppc-thread-state64 in ccl/level-1/ppc-trap-support.lisp has been wrong; the structure is defined with #pragma pack(4) in effect, so some 64-bit fields that should be misaligned are actually (and incorrectly) naturally aligned. One consequence of this is that backtrace doesn't work on darwinppc64 in some contexts, e.g., after a keyboard interrupt. There may be other consequences, as well. |
|||
