wrowe 01/06/27 15:07:25
Modified: mmap/win32 mmap.c
Log:
More crufty stuff, yes, it was my bug, and no, this is really horked since
we aren't distinguishing the size of the file backing the mmap from the
size of the mmap'ed view.
Revision Changes Path
1.6 +6 -3 apr/mmap/win32/mmap.c
Index: mmap.c
===================================================================
RCS file: /home/cvs/apr/mmap/win32/mmap.c,v
retrieving revision 1.5
retrieving revision 1.6
diff -u -r1.5 -r1.6
--- mmap.c 2001/06/06 16:05:21 1.5
+++ mmap.c 2001/06/27 22:07:24 1.6
@@ -126,11 +126,14 @@
(*new)->pstart = (offset / memblock) * memblock;
(*new)->psize = (apr_size_t)(offset % memblock) + size;
(*new)->poffset = offset - (*new)->pstart;
- offlo = (DWORD)(*new)->psize;
- offhi = (DWORD)((*new)->psize << 32);
+ /* XXX: psize below should be the MAXIMUM size of the mmap object,
+ * (e.g. file size) not the size of the mapped region!
+ * Since apr doesn't seem to acknowledge the discrepancy (the mmap
+ * size/view/off concepts are pretty horked) this will have to wait.
+ */
(*new)->mhandle = CreateFileMapping(file->filehand, NULL, fmaccess,
- offhi, offlo, NULL);
+ 0, (*new)->psize, NULL);
if (!(*new)->mhandle || (*new)->mhandle == INVALID_HANDLE_VALUE)
{
*new = NULL;
