- Timestamp:
- Sep 2, 2009 12:43:32 PM (16 years ago)
- svn:sync-xref-src-repo-rev:
- 51841
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/src/VBox/VMM/PGMPhys.cpp
r22710 r22715 386 386 PGMPoolFlushPage(pVM, *pGCPhys); 387 387 Assert(!PGM_PAGE_HAS_ACTIVE_HANDLERS(pPage)); 388 /** @todo r=bird: return VERR_PGM_PHYS_PAGE_RESERVED here if it still has 389 * active handlers, see the PGMR3PhysGCPhys2CCPtrExternal docs. */ 388 390 } 389 391 } … … 2947 2949 Req.idChunkUnmap = pgmR3PhysChunkFindUnmapCandidate(pVM); 2948 2950 /** @todo This is wrong. Any thread in the VM process should be able to do this, 2949 * there are depenenecies on this. What currently saves the day is that 2951 * there are depenenecies on this. What currently saves the day is that 2950 2952 * we don't unmap anything and that all non-zero memory will therefore 2951 2953 * be present when non-EMTs tries to access it. */
Note:
See TracChangeset
for help on using the changeset viewer.