Bonjour,
The behavior of memcpy is well defined:
"Copying overlapping buffers isn't guaranteed to work; use memmove() to
copy buffers that overlap"
In our case for the SH architecture we have a version of memcpy that
takes advantage at some specific instruction and above a certain size
the copy will be done in reverse (side effect of the optimization).
For overlapping copies, the code should be using memmove() as describe
in POSIX.
Comments?
Please note that I am not subscribed to the list.
Thanks
======
# svn diff
Index: inflate.c
===================================================================
--- inflate.c (revision 248771)
+++ inflate.c (working copy)
@@ -595,7 +595,7 @@
#if !defined(NOMEMCPY) && !defined(DEBUG)
if (w - d >= e) /* (this test assumes unsigned
comparison) */
{
- memcpy(slide + w, slide + d, e);
+ memmove(slide + w, slide + d, e);
w += e;
d += e;
}
Index: deflate.c
===================================================================
--- deflate.c (revision 248771)
+++ deflate.c (working copy)
@@ -547,7 +547,7 @@
*/
Assert(window_size == (ulg)2*WSIZE, "no sliding with BIG_MEM");
- memcpy((char*)window, (char*)window+WSIZE, (unsigned)WSIZE);
+ memmove((char*)window, (char*)window+WSIZE, (unsigned)WSIZE);
match_start -= WSIZE;
strstart -= WSIZE; /* we now have strstart >= MAX_DIST: */
