https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=240466
Adriaan de Groot <adr...@freebsd.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |adr...@freebsd.org
--- Comment #5 from Adriaan de Groot <adr...@freebsd.org> ---
Here's a backtrace (after I rebuilt it with debugging symbols):
* thread #1, name = 'digikam', stop reason = signal SIGSEGV: invalid address
(fault address: 0x0)
frame #0: 0x0000000800f821f1
libdigikamcore.so.6.4.0`::WXMPMeta_RegisterNamespace_1(namespaceURI="http://ns.adobe.com/lightroom/1.0/";,
suggestedPrefix="lr", registeredPrefix=0x00007fffffffd6a0,
prefixSize=0x0000000000000001, wResult=0x0000000000000000) at
WXMPMeta.cpp:224:5
221 XMP_StringLen * prefixSize,
222 WXMP_Result * wResult )
223 {
-> 224 XMP_ENTER_WRAPPER ( "WXMPMeta_RegisterNamespace_1" )
225
226 if ( (namespaceURI == 0) || (*namespaceURI == 0) ) XMP_Throw (
"Empty namespace URI", kXMPErr_BadSchema );
227 if ( (suggestedPrefix == 0) || (*suggestedPrefix == 0) )
XMP_Throw ( "Empty suggested prefix", kXMPErr_BadSchema );
(lldb) bt
* thread #1, name = 'digikam', stop reason = signal SIGSEGV: invalid address
(fault address: 0x0)
* frame #0: 0x0000000800f821f1
libdigikamcore.so.6.4.0`::WXMPMeta_RegisterNamespace_1(namespaceURI="http://ns.adobe.com/lightroom/1.0/";,
suggestedPrefix="lr", registeredPrefix=0x00007fffffffd6a0,
prefixSize=0x0000000000000001, wResult=0x0000000000000000) at
WXMPMeta.cpp:224:5
frame #1: 0x0000000805825b44
libexiv2.so.27`Exiv2::XmpParser::initialize(void (*)(void*, bool), void*) + 132
frame #2: 0x0000000800d30a0a
libdigikamcore.so.6.4.0`Digikam::MetaEngine::initializeExiv2() at
metaengine.cpp:84:10
frame #3: 0x0000000000208d22 digikam`main(argc=1, argv=0x00007fffffffe458)
at main.cpp:138:5
frame #4: 0x000000000020810f digikam`_start(ap=<unavailable>,
cleanup=<unavailable>) at crt1.c:76:7
What's happening is that Exiv2::XmpParser::initialize() is calling
WXMPMeta_RegisterNamespace_1 which is *supposed* to be a function inside Exiv2.
Instead, it ends up in Digikam's reimplementation of that function (instead of
Exiv2's version). And then it goes boom.
--
You are receiving this mail because:
You are the assignee for the bug.
