|
From: | Jim Langston |
Subject: | Re: audio check? |
Date: | Sun, 09 Dec 2007 21:04:51 -0500 |
User-agent: | Thunderbird 1.5.0.8 (X11/20061204) |
This is what the debugger says created the segfault: address@hidden (address@hidden) program terminated by signal SEGV (Segmentation Fault) 0xffffffffffffffff: <bad address 0xffffffffffffffff> Current function is MatrixType::MatrixType 222 found [j] = false; (dbx) list 222 found [j] = false; 223 224 for (octave_idx_type j = i; j < nm; j++) 225 { 226 if ((a.cidx(j+1) > a.cidx(j) + 1) || 227 ((a.cidx(j+1) == a.cidx(j) + 1) && found [a.ridx(j)])) 228 { 229 tmp_typ = MatrixType::Full; 230 break; 231 } Jim //////////////////////////////// Jim Langston wrote: Hi Moritz, Just pulled down 2.9.18 and recompiled, no problem with the compile, one thing to do for the compile: - change in configure: BLAS_LIBS="-xlic_lib=sunperf -lsunmath" to: BLAS_LIBS="-lsunperf -lsunmath" also, I am now getting a segfault when running check which I did not get with 2.9.13 Integrated test scripts: src/DLD-FUNCTIONS/bsxfun.cc ............................ PASS 55/55 src/DLD-FUNCTIONS/cellfun.cc ........................... PASS 31/31 src/DLD-FUNCTIONS/conv2.cc ............................. PASS 1/1 src/DLD-FUNCTIONS/dispatch.cc .......................... PASS 13/13 src/DLD-FUNCTIONS/fft.cc ............................... PASS 9/9 src/DLD-FUNCTIONS/luinc.cc ............................. src/DLD-FUNCTIONS/matrix_type.cc .......................panic: Segmentation Fault -- stopping myself... gmake[2]: *** [check] Segmentation Fault (core dumped) gmake[2]: Leaving directory `/export/home/langston/Octave/octave-2.9.18/test' gmake[1]: *** [check] Error 2 gmake[1]: Leaving directory `/export/home/langston/Octave/octave-2.9.18' gmake: *** [check] Error 2 Jim ///////////////////////////////// Moritz Borgmann wrote:Hi John & Jim, you guys were discussing this issue here back in September: http://www.nabble.com/audio-check--tf4379058.html#a12867897 The issue is still present in current octave cvs. Did you find some solution for it back then? Jim - this may be of interest for you: current cvs is now bugfixed such that it (almost) compiles with Sun Studio. Currently hunting a few last issue, see octave bug mailing list <address@hidden> for details. Thanks, Moritz -- ///////////////////////////////////////////// Jim Langston Sun Microsystems, Inc. (877) 854-5583 (AccessLine) AIM: jl9594 address@hidden |
[Prev in Thread] | Current Thread | [Next in Thread] |