xref: /netbsd-src/sys/uvm/uvm_pglist.c (revision 3b01aba77a7a698587faaae455bbfe740923c1f5)
1 /*	$NetBSD: uvm_pglist.c,v 1.17 2001/06/27 21:18:34 thorpej Exp $	*/
2 
3 /*-
4  * Copyright (c) 1997 The NetBSD Foundation, Inc.
5  * All rights reserved.
6  *
7  * This code is derived from software contributed to The NetBSD Foundation
8  * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
9  * NASA Ames Research Center.
10  *
11  * Redistribution and use in source and binary forms, with or without
12  * modification, are permitted provided that the following conditions
13  * are met:
14  * 1. Redistributions of source code must retain the above copyright
15  *    notice, this list of conditions and the following disclaimer.
16  * 2. Redistributions in binary form must reproduce the above copyright
17  *    notice, this list of conditions and the following disclaimer in the
18  *    documentation and/or other materials provided with the distribution.
19  * 3. All advertising materials mentioning features or use of this software
20  *    must display the following acknowledgement:
21  *      This product includes software developed by the NetBSD
22  *      Foundation, Inc. and its contributors.
23  * 4. Neither the name of The NetBSD Foundation nor the names of its
24  *    contributors may be used to endorse or promote products derived
25  *    from this software without specific prior written permission.
26  *
27  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
28  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
29  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
30  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
31  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
32  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
33  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
34  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
35  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
37  * POSSIBILITY OF SUCH DAMAGE.
38  */
39 
40 /*
41  * uvm_pglist.c: pglist functions
42  */
43 
44 #include <sys/param.h>
45 #include <sys/systm.h>
46 #include <sys/malloc.h>
47 #include <sys/proc.h>
48 
49 #include <uvm/uvm.h>
50 
51 #ifdef VM_PAGE_ALLOC_MEMORY_STATS
52 #define	STAT_INCR(v)	(v)++
53 #define	STAT_DECR(v)	do { \
54 		if ((v) == 0) \
55 			printf("%s:%d -- Already 0!\n", __FILE__, __LINE__); \
56 		else \
57 			(v)--; \
58 	} while (0)
59 u_long	uvm_pglistalloc_npages;
60 #else
61 #define	STAT_INCR(v)
62 #define	STAT_DECR(v)
63 #endif
64 
65 /*
66  * uvm_pglistalloc: allocate a list of pages
67  *
68  * => allocated pages are placed at the tail of rlist.  rlist is
69  *    assumed to be properly initialized by caller.
70  * => returns 0 on success or errno on failure
71  * => XXX: implementation allocates only a single segment, also
72  *	might be able to better advantage of vm_physeg[].
73  * => doesn't take into account clean non-busy pages on inactive list
74  *	that could be used(?)
75  * => params:
76  *	size		the size of the allocation, rounded to page size.
77  *	low		the low address of the allowed allocation range.
78  *	high		the high address of the allowed allocation range.
79  *	alignment	memory must be aligned to this power-of-two boundary.
80  *	boundary	no segment in the allocation may cross this
81  *			power-of-two boundary (relative to zero).
82  */
83 
84 int
85 uvm_pglistalloc(size, low, high, alignment, boundary, rlist, nsegs, waitok)
86 	psize_t size;
87 	paddr_t low, high, alignment, boundary;
88 	struct pglist *rlist;
89 	int nsegs, waitok;
90 {
91 	paddr_t try, idxpa, lastidxpa;
92 	int psi;
93 	struct vm_page *pgs;
94 	int s, tryidx, idx, pgflidx, end, error, free_list, color;
95 	struct vm_page *m;
96 	u_long pagemask;
97 #ifdef DEBUG
98 	struct vm_page *tp;
99 #endif
100 
101 	KASSERT((alignment & (alignment - 1)) == 0);
102 	KASSERT((boundary & (boundary - 1)) == 0);
103 
104 	/*
105 	 * Our allocations are always page granularity, so our alignment
106 	 * must be, too.
107 	 */
108 	if (alignment < PAGE_SIZE)
109 		alignment = PAGE_SIZE;
110 
111 	size = round_page(size);
112 	try = roundup(low, alignment);
113 
114 	if (boundary != 0 && boundary < size)
115 		return (EINVAL);
116 
117 	pagemask = ~(boundary - 1);
118 
119 	/* Default to "lose". */
120 	error = ENOMEM;
121 
122 	/*
123 	 * Block all memory allocation and lock the free list.
124 	 */
125 	s = uvm_lock_fpageq();
126 
127 	/* Are there even any free pages? */
128 	if (uvmexp.free <= (uvmexp.reserve_pagedaemon + uvmexp.reserve_kernel))
129 		goto out;
130 
131 	for (;; try += alignment) {
132 		if (try + size > high) {
133 
134 			/*
135 			 * We've run past the allowable range.
136 			 */
137 
138 			goto out;
139 		}
140 
141 		/*
142 		 * Make sure this is a managed physical page.
143 		 */
144 
145 		if ((psi = vm_physseg_find(atop(try), &idx)) == -1)
146 			continue; /* managed? */
147 		if (vm_physseg_find(atop(try + size), NULL) != psi)
148 			continue; /* end must be in this segment */
149 
150 		tryidx = idx;
151 		end = idx + (size / PAGE_SIZE);
152 		pgs = vm_physmem[psi].pgs;
153 
154 		/*
155 		 * Found a suitable starting page.  See of the range is free.
156 		 */
157 
158 		for (; idx < end; idx++) {
159 			if (VM_PAGE_IS_FREE(&pgs[idx]) == 0) {
160 				break;
161 			}
162 			idxpa = VM_PAGE_TO_PHYS(&pgs[idx]);
163 			if (idx > tryidx) {
164 				lastidxpa = VM_PAGE_TO_PHYS(&pgs[idx - 1]);
165 				if ((lastidxpa + PAGE_SIZE) != idxpa) {
166 
167 					/*
168 					 * Region not contiguous.
169 					 */
170 
171 					break;
172 				}
173 				if (boundary != 0 &&
174 				    ((lastidxpa ^ idxpa) & pagemask) != 0) {
175 
176 					/*
177 					 * Region crosses boundary.
178 					 */
179 
180 					break;
181 				}
182 			}
183 		}
184 		if (idx == end) {
185 			break;
186 		}
187 	}
188 
189 #if PGFL_NQUEUES != 2
190 #error uvm_pglistalloc needs to be updated
191 #endif
192 
193 	/*
194 	 * we have a chunk of memory that conforms to the requested constraints.
195 	 */
196 	idx = tryidx;
197 	while (idx < end) {
198 		m = &pgs[idx];
199 		free_list = uvm_page_lookup_freelist(m);
200 		color = VM_PGCOLOR_BUCKET(m);
201 		pgflidx = (m->flags & PG_ZERO) ? PGFL_ZEROS : PGFL_UNKNOWN;
202 #ifdef DEBUG
203 		for (tp = TAILQ_FIRST(&uvm.page_free[
204 			free_list].pgfl_buckets[color].pgfl_queues[pgflidx]);
205 		     tp != NULL;
206 		     tp = TAILQ_NEXT(tp, pageq)) {
207 			if (tp == m)
208 				break;
209 		}
210 		if (tp == NULL)
211 			panic("uvm_pglistalloc: page not on freelist");
212 #endif
213 		TAILQ_REMOVE(&uvm.page_free[free_list].pgfl_buckets[
214 		    color].pgfl_queues[pgflidx], m, pageq);
215 		uvmexp.free--;
216 		if (m->flags & PG_ZERO)
217 			uvmexp.zeropages--;
218 		m->flags = PG_CLEAN;
219 		m->pqflags = 0;
220 		m->uobject = NULL;
221 		m->uanon = NULL;
222 		m->version++;
223 		TAILQ_INSERT_TAIL(rlist, m, pageq);
224 		idx++;
225 		STAT_INCR(uvm_pglistalloc_npages);
226 	}
227 	error = 0;
228 
229 out:
230 	/*
231 	 * check to see if we need to generate some free pages waking
232 	 * the pagedaemon.
233 	 */
234 
235 	UVM_KICK_PDAEMON();
236 
237 	uvm_unlock_fpageq(s);
238 
239 	return (error);
240 }
241 
242 /*
243  * uvm_pglistfree: free a list of pages
244  *
245  * => pages should already be unmapped
246  */
247 
248 void
249 uvm_pglistfree(list)
250 	struct pglist *list;
251 {
252 	struct vm_page *m;
253 	int s;
254 
255 	/*
256 	 * Block all memory allocation and lock the free list.
257 	 */
258 	s = uvm_lock_fpageq();
259 
260 	while ((m = TAILQ_FIRST(list)) != NULL) {
261 		KASSERT((m->pqflags & (PQ_ACTIVE|PQ_INACTIVE)) == 0);
262 		TAILQ_REMOVE(list, m, pageq);
263 		m->pqflags = PQ_FREE;
264 		TAILQ_INSERT_TAIL(&uvm.page_free[
265 		    uvm_page_lookup_freelist(m)].pgfl_buckets[
266 		    VM_PGCOLOR_BUCKET(m)].pgfl_queues[PGFL_UNKNOWN], m, pageq);
267 		uvmexp.free++;
268 		if (uvmexp.zeropages < UVM_PAGEZERO_TARGET)
269 			uvm.page_idle_zero = vm_page_zero_enable;
270 		STAT_DECR(uvm_pglistalloc_npages);
271 	}
272 
273 	uvm_unlock_fpageq(s);
274 }
275