Hi,
problem seems to come from my last commit on this file (big patch to
correct lot of bugs and strange behavior of the module).
I see the root cause : self->dev is NULL. The question is more how to
get a work around to get the image size.
Will try to correct this as soon as possible.
thanks for reporting
Aldric
Le 17/03/2013 12:16, Richard Levitte a écrit :
Hi,
I've built the latest revision I can get (1.1+1329~g1728e4d), and I'm
getting a sigsegv. The automatically generated backtrace shows me
that it happens in src/iop/clipping.c:190. A manual session later, I
find myself with this:
Program received signal SIGSEGV, Segmentation fault.
0x8b99fe40 in legacy_params (self=0x9c257ff0, old_params=0x9c2576d8,
old_version=4, new_params=0x9c258520, new_version=5)
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/iop/clipping.c:190
190 float whratio = ((float)(self->dev->image_storage.width - 2 *
interpolation->width) * (fabsf(n->cw) - n->cx)) /
(gdb) print *n
$1 = {angle = -0.699996948, cx = 0.0587422848, cy = 0.0702050403,
cw = 0.992118239, ch = 0.691418171, k_h = 0, k_v = 0, kxa = 0.200000003,
kya = 0.200000003, kxb = 0.800000012, kyb = 0.200000003, kxc = 0.800000012,
kyc = 0.800000012, kxd = 0.200000003, kyd = 0.800000012, k_type = 0,
k_sym = 0, k_apply = 0, crop_auto = 1, ratio_n = 0, ratio_d = 0}
(gdb) list
185 // if no clipping yet, use default aspect ratio
186 if (fabsf(n->cw) == 1.0 && n->cx == 0.0 && fabsf(n->ch) == 1.0 && n->cy ==
0.0) n->ratio_d=-1, n->ratio_n=-1;
187 else
188 {
189 const struct dt_interpolation* interpolation =
dt_interpolation_new(DT_INTERPOLATION_USERPREF);
190 float whratio = ((float)(self->dev->image_storage.width - 2 *
interpolation->width) * (fabsf(n->cw) - n->cx)) /
191 ((float)(self->dev->image_storage.height - 2 * interpolation->width) *
(fabsf(n->ch) - n->cy));
192 float ri = self->dev->image_storage.width /
(float)self->dev->image_storage.height;
193
194 float prec = 0.0003f;
(gdb) print *interpolation
value has been optimized out
(gdb) print interpolation->width
value has been optimized out
(gdb) print self->dev->image_storage
Cannot access memory at address 0x5c
(gdb) print self->dev
$2 = (struct dt_develop_t *) 0x0
(gdb) print self
$3 = (dt_iop_module_t *) 0x9c257ff0
(gdb) bt
#0 0x8b99fe40 in legacy_params (self=0x9c257ff0, old_params=0x9c2576d8,
old_version=4, new_params=0x9c258520, new_version=5)
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/iop/clipping.c:190
#1 0xf7e58295 in init_presets (module_so=0x9c255648)
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/develop/imageop.c:1307
#2 dt_iop_load_modules_so ()
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/develop/imageop.c:1438
#3 0xf7df3d73 in dt_init (argc=argc@entry=1, argv=argv@entry=0xffffd014,
init_gui=init_gui@entry=1)
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/common/darktable.c:669
#4 0x08048963 in main (argc=1, argv=0xffffd014)
at
/home/levitte/gitwrk/github.com/darktable-1.1+1329~g1728e4d/src/main.c:24
(gdb)
Cheers,
Richard
P.S. Attaching the automagically genetared backtrace.
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
darktable-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-devel
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
darktable-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-devel
