View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000319 | ascend | packages | public | 2007-01-21 02:13 | 2010-03-23 18:30 |
| Reporter | john | ||||
| Assigned To | ben | ||||
| Priority | normal | Severity | minor | Reproducibility | always |
| Status | assigned | Resolution | open | ||
| Product Version | 0.9.5.x | ||||
| Target Version | 1.0 | Fixed in Version | |||
| Summary | 0000319: sensitivity_all fails | ||||
| Description | > I added a test case for the 'do_sensitivity_all' code, but it seems to > > cause a crash. > > > > Program received signal SIGSEGV, Segmentation fault. > > [Switching to Thread -1209825088 (LWP 2280)] > > 0xb6f36c04 in mtx_org_to_col (mtx=0x8268cd8, org=136572504) at > > base/generic/solver/mtx_perms.c:483 > > 483 return( mtx->perm.col.org_to_cur[org] ); > > (gdb) where > > #0 0xb6f36c04 in mtx_org_to_col (mtx=0x8268cd8, org=136572504) at > > base/generic/solver/mtx_perms.c:483 Pretty clearly org is a bogus column number from somewhere or uninitialized. Running the same code with valgrind might differentiate it or stepping through Compute_dy_dx_smart and seeing where some loop gets lost and find this goofy index may tell the story. ben | ||||
| Tags | No tags attached. | ||||
