Author: alc
Date: Mon Jun 10 01:48:21 2013
New Revision: 251591
URL: http://svnweb.freebsd.org/changeset/base/251591

Log:
  Revise the interface between vm_object_madvise() and vm_page_dontneed() so
  that pointless calls to pmap_is_modified() can be easily avoided when
  performing madvise(..., MADV_FREE).
  
  Sponsored by: EMC / Isilon Storage Division

Modified:
  head/sys/vm/vm_object.c
  head/sys/vm/vm_page.c
  head/sys/vm/vm_page.h

Modified: head/sys/vm/vm_object.c
==============================================================================
--- head/sys/vm/vm_object.c     Sun Jun  9 23:51:26 2013        (r251590)
+++ head/sys/vm/vm_object.c     Mon Jun 10 01:48:21 2013        (r251591)
@@ -1175,28 +1175,8 @@ shadowlookup:
                }
                if (advise == MADV_WILLNEED) {
                        vm_page_activate(m);
-               } else if (advise == MADV_DONTNEED) {
-                       vm_page_dontneed(m);
-               } else if (advise == MADV_FREE) {
-                       /*
-                        * Mark the page clean.  This will allow the page
-                        * to be freed up by the system.  However, such pages
-                        * are often reused quickly by malloc()/free()
-                        * so we do not do anything that would cause
-                        * a page fault if we can help it.
-                        *
-                        * Specifically, we do not try to actually free
-                        * the page now nor do we try to put it in the
-                        * cache (which would cause a page fault on reuse).
-                        *
-                        * But we do make the page is freeable as we
-                        * can without actually taking the step of unmapping
-                        * it.
-                        */
-                       pmap_clear_modify(m);
-                       m->dirty = 0;
-                       m->act_count = 0;
-                       vm_page_dontneed(m);
+               } else {
+                       vm_page_advise(m, advise);
                }
                vm_page_unlock(m);
                if (advise == MADV_FREE && tobject->type == OBJT_SWAP)

Modified: head/sys/vm/vm_page.c
==============================================================================
--- head/sys/vm/vm_page.c       Sun Jun  9 23:51:26 2013        (r251590)
+++ head/sys/vm/vm_page.c       Mon Jun 10 01:48:21 2013        (r251591)
@@ -93,6 +93,7 @@ __FBSDID("$FreeBSD$");
 #include <sys/kernel.h>
 #include <sys/limits.h>
 #include <sys/malloc.h>
+#include <sys/mman.h>
 #include <sys/msgbuf.h>
 #include <sys/mutex.h>
 #include <sys/proc.h>
@@ -2238,15 +2239,15 @@ vm_page_cache(vm_page_t m)
 }
 
 /*
- * vm_page_dontneed
+ * vm_page_advise
  *
  *     Cache, deactivate, or do nothing as appropriate.  This routine
- *     is typically used by madvise() MADV_DONTNEED.
+ *     is used by madvise().
  *
  *     Generally speaking we want to move the page into the cache so
  *     it gets reused quickly.  However, this can result in a silly syndrome
  *     due to the page recycling too quickly.  Small objects will not be
- *     fully cached.  On the otherhand, if we move the page to the inactive
+ *     fully cached.  On the other hand, if we move the page to the inactive
  *     queue we wind up with a problem whereby very large objects 
  *     unnecessarily blow away our inactive and cache queues.
  *
@@ -2261,13 +2262,31 @@ vm_page_cache(vm_page_t m)
  *     The object and page must be locked.
  */
 void
-vm_page_dontneed(vm_page_t m)
+vm_page_advise(vm_page_t m, int advice)
 {
-       int dnw;
-       int head;
+       int dnw, head;
 
-       vm_page_lock_assert(m, MA_OWNED);
+       vm_page_assert_locked(m);
        VM_OBJECT_ASSERT_WLOCKED(m->object);
+       if (advice == MADV_FREE) {
+               /*
+                * Mark the page clean.  This will allow the page to be freed
+                * up by the system.  However, such pages are often reused
+                * quickly by malloc() so we do not do anything that would
+                * cause a page fault if we can help it.
+                *
+                * Specifically, we do not try to actually free the page now
+                * nor do we try to put it in the cache (which would cause a
+                * page fault on reuse).
+                *
+                * But we do make the page is freeable as we can without
+                * actually taking the step of unmapping it.
+                */
+               pmap_clear_modify(m);
+               m->dirty = 0;
+               m->act_count = 0;
+       } else if (advice != MADV_DONTNEED)
+               return;
        dnw = PCPU_GET(dnweight);
        PCPU_INC(dnweight);
 
@@ -2294,7 +2313,7 @@ vm_page_dontneed(vm_page_t m)
        pmap_clear_reference(m);
        vm_page_aflag_clear(m, PGA_REFERENCED);
 
-       if (m->dirty == 0 && pmap_is_modified(m))
+       if (advice != MADV_FREE && m->dirty == 0 && pmap_is_modified(m))
                vm_page_dirty(m);
 
        if (m->dirty || (dnw & 0x0070) == 0) {

Modified: head/sys/vm/vm_page.h
==============================================================================
--- head/sys/vm/vm_page.h       Sun Jun  9 23:51:26 2013        (r251590)
+++ head/sys/vm/vm_page.h       Mon Jun 10 01:48:21 2013        (r251591)
@@ -370,6 +370,7 @@ void vm_page_free_zero(vm_page_t m);
 void vm_page_wakeup(vm_page_t m);
 
 void vm_page_activate (vm_page_t);
+void vm_page_advise(vm_page_t m, int advice);
 vm_page_t vm_page_alloc (vm_object_t, vm_pindex_t, int);
 vm_page_t vm_page_alloc_contig(vm_object_t object, vm_pindex_t pindex, int req,
     u_long npages, vm_paddr_t low, vm_paddr_t high, u_long alignment,
@@ -381,7 +382,6 @@ void vm_page_cache_free(vm_object_t, vm_
 void vm_page_cache_transfer(vm_object_t, vm_pindex_t, vm_object_t);
 int vm_page_try_to_cache (vm_page_t);
 int vm_page_try_to_free (vm_page_t);
-void vm_page_dontneed(vm_page_t);
 void vm_page_deactivate (vm_page_t);
 void vm_page_dequeue(vm_page_t m);
 void vm_page_dequeue_locked(vm_page_t m);
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "[email protected]"

Reply via email to