d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Use vm_page_replace_checked() instead of vm_page_rename() for implementing

Browse Source 
optimized copy-on-write faults.  This has two advantages: (1) one less radix
tree operation is performed and (2) vm_page_replace_checked() cannot fail,
making the code simpler.

Submitted by:	Ryan Libby
Reviewed by:	kib
Sponsored by:	EMC / Isilon Storage Division
Differential Revision:	https://reviews.freebsd.org/D4478
This commit is contained in:
alc 2016-05-27 06:05:12 +00:00
parent b295eb2eaf
commit 2cbd677fae
1 changed files with 10 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
sys/vm/vm_fault.c
View File

@ -799,26 +799,15 @@ RetryFault:;
* We don't chase down the shadow chain
*/
fs.object == fs.first_object->backing_object) {
/*
* get rid of the unnecessary page
*/
vm_page_lock(fs.first_m);
vm_page_remove(fs.first_m);
vm_page_unlock(fs.first_m);
/*
* grab the page and put it into the
* process'es object. The page is
* automatically made dirty.
*/
if (vm_page_rename(fs.m, fs.first_object,
fs.first_pindex)) {
VM_OBJECT_WUNLOCK(fs.first_object);
unlock_and_deallocate(&fs);
goto RetryFault;
}
vm_page_lock(fs.m);
vm_page_remove(fs.m);
vm_page_unlock(fs.m);
vm_page_lock(fs.first_m);
vm_page_replace_checked(fs.m, fs.first_object,
fs.first_pindex, fs.first_m);
vm_page_free(fs.first_m);
vm_page_unlock(fs.first_m);
vm_page_dirty(fs.m);
#if VM_NRESERVLEVEL > 0
/*
* Rename the reservation.
@ -827,6 +816,10 @@ RetryFault:;
fs.object, OFF_TO_IDX(
fs.first_object->backing_object_offset));
#endif
/*
* Removing the page from the backing object
* unbusied it.
*/
vm_page_xbusy(fs.m);
fs.first_m = fs.m;
fs.m = NULL;