mem: mmap the backing store with MAP_NORESERVE

This patch ensures we can run simulations with very large simulated
memories (at least 64 TB based on some quick runs on a Linux
workstation). In essence this allows us to efficiently deal with
sparse address maps without having to implement a redirection layer in
the backing store.

This opens up for run-time errors if we eventually exhausts the hosts
memory and swap space, but this should hopefully never happen.

2 files changed, 8 insertions, 1 deletions

diff --git a/src/sim/System.py b/src/sim/System.py
index 630cd2a84..e24a1e6b2 100644
--- a/src/sim/System.py
+++ b/src/sim/System.py
@@ -59,6 +59,13 @@ class System(MemObject):
                                           "All memories in the system")
     mem_mode = Param.MemoryMode('atomic', "The mode the memory system is in")
 
+    # When reserving memory on the host, we have the option of
+    # reserving swap space or not (by passing MAP_NORESERVE to
+    # mmap). By enabling this flag, we accomodate cases where a large
+    # (but sparse) memory is simulated.
+    mmap_using_noreserve = Param.Bool(False, "mmap the backing store " \
+                                          "without reserving swap")
+
     # The memory ranges are to be populated when creating the system
     # such that these can be passed from the I/O subsystem through an
     # I/O bridge or cache
diff --git a/src/sim/system.cc b/src/sim/system.cc
index c311d65b9..9bd487b03 100644
--- a/src/sim/system.cc
+++ b/src/sim/system.cc
@@ -88,7 +88,7 @@ System::System(Params *p)
       loadAddrMask(p->load_addr_mask),
       loadAddrOffset(p->load_offset),
       nextPID(0),
-      physmem(name() + ".physmem", p->memories),
+      physmem(name() + ".physmem", p->memories, p->mmap_using_noreserve),
       memoryMode(p->mem_mode),
       _cacheLineSize(p->cache_line_size),
       workItemsBegin(0),