#1376 closed defect (fixed)
DIG: overlapping memory in memcpy() call
| Reported by: | SF/khalek | Owned by: | aquadran |
|---|---|---|---|
| Priority: | normal | Component: | Engine: SCUMM |
| Version: | Keywords: | ||
| Cc: | Game: | The Dig |
Description
eriktorbjorn mentioned on irc the following valgrind warning that occurs with the dig just after the intro has been escaped, I subsequently confirmed it and here it is...
==20575== Thread 2: ==20575== Source and destination overlap in memcpy(0x3CFE726C, 0x3CFE8C43, 7674)==20575== at 0x3C01F0DE: memcpy (mac_replace_strmem.c:113) ==20575== by 0x809FD5E: Scumm::IMuseDigital::bundleMusicHandler() (imuse_digi.cpp:1339) ==20575== by 0x809F5DA: Scumm::IMuseDigital::music_handler(void*) (imuse_digi.cpp:1189) ==20575== by 0x8199DB4: Timer::handler(int) (timer.cpp:99)
If these sections of memory are indeed supposed to be overlapping memmove() should be used, otherwise this needs to be fixed. Either way it needs attention.
Latest CVS, Linux et cetera
Ticket imported from: #871656. Ticket imported from: bugs/1376.
Change History (8)
comment:1 by , 20 years ago
comment:2 by , 20 years ago
| Owner: | set to |
|---|
comment:3 by , 20 years ago
| Resolution: | → wontfix |
|---|---|
| Status: | new → closed |
comment:5 by , 20 years ago
| Resolution: | wontfix |
|---|---|
| Status: | closed → new |
comment:6 by , 20 years ago
That's why I asked whether erik or khalek could please run Valgrind over the new code again, to verify that the problem is indeed gone.
comment:8 by , 20 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
Well, aquadran just moved all the IMuseDigital stuff around.
Can you check if the problem still is there with latest CVS?