src-test2/sys/vm, branch release/5.3.0_cvs FreeBSD source tree https://cgit-dev.freebsd.org/src-test2/atom?h=release%2F5.3.0_cvs 2004-11-04T19:12:42Z This commit was manufactured by cvs2svn to create tag 2004-11-04T19:12:42Z cvs2svn cvs2svn@FreeBSD.org 2004-11-04T19:12:42Z urn:sha1:3f86d8a2ea3f3265afaa1fd263b0004c5c000e69 'RELENG_5_3_0_RELEASE'. This commit was manufactured to restore the state of the 5.3-RELEASE image. MFC uma_dbg.c r1.16, uma_core.c r1.106-1.107: critical stability/leak fixes 2004-10-16T01:41:34Z Brian Feldman green@FreeBSD.org 2004-10-16T01:41:34Z urn:sha1:185be6a9728be8a279919a120ae331d3b82cbaaa (for MBUF clusters) implemented by UMA. Approved by: re MFC src/sys/vm/swap_pager.c,v 1.257 2004-09-27T04:53:18Z David Schultz das@FreeBSD.org 2004-09-27T04:53:18Z urn:sha1:4ecfa4c49bc86b4b3fe7cc7dba480ceb7b339b53 | Don't look for swap blocks in objects that aren't swap-backed. | I expect that this will fix the following panic, reported by Jun: | swap_pager_isswapped: failed to locate all swap meta blocks Approved by: re (scottl) MFC: Turn PREEMPTION into a kernel option. It is NOT enabled at this time. 2004-09-10T00:04:18Z Scott Long scottl@FreeBSD.org 2004-09-10T00:04:18Z urn:sha1:69ba7f59694787612ef7782b368ccf60a6b8ab75 Approved by: re MFC of: 2004-09-09T09:45:27Z Julian Elischer julian@FreeBSD.org 2004-09-09T09:45:27Z urn:sha1:22d3d6d45be74782cce332ab5983380a55035683 Give setrunqueue() and sched_add() more of a clue as to where they are coming from and what is expected from them. ( 2004-09-01 02:11:28 UTC ) Approved by: re (scottl) MFC: 1.27; Prevent vm_page_zero_idle_wakeup() from attempting to 2004-09-03T19:58:00Z Ian Dowse iedowse@FreeBSD.org 2004-09-03T19:58:00Z urn:sha1:53ec2be3647f272013904eef733de6905d8cad24 wake up the page zeroing thread before it has been created. Approved by: re - Introduce and use a new tunable "debug.mpsafevm". At present, setting 2004-08-16T06:16:12Z Alan Cox alc@FreeBSD.org 2004-08-16T06:16:12Z urn:sha1:c1fbc251cd43f57703976219d84bfb31a5716cf8 "debug.mpsafevm" results in (almost) Giant-free execution of zero-fill page faults. (Giant is held only briefly, just long enough to determine if there is a vnode backing the faulting address.) Also, condition the acquisition and release of Giant around calls to pmap_remove() on "debug.mpsafevm". The effect on performance is significant. On my dual Opteron, I see a 3.6% reduction in "buildworld" time. - Use atomic operations to update several counters in vm_fault(). Rather than bringing back all of the changes to make VM map deletion 2004-08-16T03:11:09Z Brian Feldman green@FreeBSD.org 2004-08-16T03:11:09Z urn:sha1:7c938963a49475acf5c7033d8dbacf5bac16347d wait for system wires to disappear, do so (much more trivially) by instead only checking for system wires of user maps and not kernel maps. Alternative by: tor Reviewed by: alc Remove spl calls. 2004-08-14T18:57:41Z Alan Cox alc@FreeBSD.org 2004-08-14T18:57:41Z urn:sha1:6eaee3fee48befdee02d5119e6722e2c286f0098 Replace the linear search in vm_map_findspace() with an O(log n) 2004-08-13T08:06:34Z Alan Cox alc@FreeBSD.org 2004-08-13T08:06:34Z urn:sha1:0164e0578108997ceb567576fc14cc0a8658a9fe algorithm built into the map entry splay tree. This replaces the first_free hint in struct vm_map with two fields in vm_map_entry: adj_free, the amount of free space following a map entry, and max_free, the maximum amount of free space in the entry's subtree. These fields make it possible to find a first-fit free region of a given size in one pass down the tree, so O(log n) amortized using splay trees. This significantly reduces the overhead in vm_map_findspace() for applications that mmap() many hundreds or thousands of regions, and has a negligible slowdown (0.1%) on buildworld. See, for example, the discussion of a micro-benchmark titled "Some mmap observations compared to Linux 2.6/OpenBSD" on -hackers in late October 2003. OpenBSD adopted this approach in March 2002, and NetBSD added it in November 2003, both with Red-Black trees. Submitted by: Mark W. Krentel