Significantly reduce the cost, i.e., run time, of calls to madvise(...,

MADV_DONTNEED) and madvise(..., MADV_FREE).  Specifically, introduce a new
pmap function, pmap_advise(), that operates on a range of virtual addresses
within the specified pmap, allowing for a more efficient implementation of
MADV_DONTNEED and MADV_FREE.  Previously, the implementation of
MADV_DONTNEED and MADV_FREE relied on per-page pmap operations, such as
pmap_clear_reference().  Intuitively, the problem with this implementation
is that the pmap-level locks are acquired and released and the page table
traversed repeatedly, once for each resident page in the range
that was specified to madvise(2).  A more subtle flaw with the previous
implementation is that pmap_clear_reference() would clear the reference bit
on all mappings to the specified page, not just the mapping in the range
specified to madvise(2).

Since our malloc(3) makes heavy use of madvise(2), this change can have a
measureable impact.  For example, the system time for completing a parallel
"buildworld" on a 6-core amd64 machine was reduced by about 1.5% to 2.0%.

Note: This change only contains pmap_advise() implementations for a subset
of our supported architectures.  I will commit implementations for the
remaining architectures after further testing.  For now, a stub function is
sufficient because of the advisory nature of pmap_advise().

Discussed with: jeff, jhb, kib
Tested by:      pho (i386), marcel (ia64)
Sponsored by:   EMC / Isilon Storage Division

2 files changed, 28 insertions, 0 deletions

diff --git a/sys/powerpc/powerpc/mmu_if.m b/sys/powerpc/powerpc/mmu_if.m
index 0382bd8..f9f37cb 100644
--- a/sys/powerpc/powerpc/mmu_if.m
+++ b/sys/powerpc/powerpc/mmu_if.m
@@ -133,6 +133,25 @@ CODE {
 
 
 /**
+ * @brief Apply the given advice to the specified range of addresses within
+ * the given pmap.  Depending on the advice, clear the referenced and/or
+ * modified flags in each mapping and set the mapped page's dirty field.
+ *
+ * @param _pmap		physical map
+ * @param _start	virtual range start
+ * @param _end		virtual range end
+ * @param _advice	advice to apply
+ */
+METHOD void advise {
+	mmu_t		_mmu;
+	pmap_t		_pmap;
+	vm_offset_t	_start;
+	vm_offset_t	_end;
+	int		_advice;
+};
+
+
+/**
  * @brief Change the wiring attribute for the page in the given physical
  * map and virtual address.
  *
diff --git a/sys/powerpc/powerpc/pmap_dispatch.c b/sys/powerpc/powerpc/pmap_dispatch.c
index 7fd98f4..24e6076 100644
--- a/sys/powerpc/powerpc/pmap_dispatch.c
+++ b/sys/powerpc/powerpc/pmap_dispatch.c
@@ -91,6 +91,15 @@ RB_GENERATE(pvo_tree, pvo_entry, pvo_plink, pvo_vaddr_compare);
 	
 
 void
+pmap_advise(pmap_t pmap, vm_offset_t start, vm_offset_t end, int advice)
+{
+
+	CTR5(KTR_PMAP, "%s(%p, %#x, %#x, %d)", __func__, pmap, start, end,
+	    advice);
+	MMU_ADVISE(mmu_obj, pmap, start, end, advice);
+}
+
+void
 pmap_change_wiring(pmap_t pmap, vm_offset_t va, boolean_t wired)
 {