upstream/opnsense-src
1
0
Fork 0
mirror of https://github.com/opnsense/src.git synced 2026-05-28 04:12:45 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

Enable both sbrk(2)- and mmap(2)-based memory acquisition methods by

Browse source 
default.  This has the disadvantage of rendering the datasize resource
limit irrelevant, but without this change, legitimate uses of more
memory than will fit in the data segment are thwarted by default.

Fix chunk_alloc_mmap() to work correctly if initial mapping is not
chunk-aligned and mapping extension fails.
This commit is contained in:
Jason Evans 2008-01-03 23:22:13 +00:00
parent 06019eff46
commit f38512f4af
2 changed files with 20 additions and 19 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
lib/libc/stdlib/malloc.3
View file

@ -32,7 +32,7 @@
.\" @(#)malloc.3 8.1 (Berkeley) 6/4/93
.\" $FreeBSD$
.\"
.Dd December 27, 2007
.Dd January 3, 2008
.Dt MALLOC 3
.Os
.Sh NAME
@ -245,7 +245,7 @@ This option is not available for some configurations (non-PIC).
Use
.Xr mmap 2
to acquire anonymously mapped memory.
This option is disabled by default.
This option is enabled by default.
If both the
.Dq D
and
@ -335,24 +335,24 @@ Traditionally, allocators have used
to obtain memory, which is suboptimal for several reasons, including race
conditions, increased fragmentation, and artificial limitations on maximum
usable memory.
This allocator uses
This allocator uses both
.Xr sbrk 2
by default in order to facilitate resource limits, but it can be configured at
run time to use
.Xr sbrk 2
and/or
.Xr mmap 2 .
and
.Xr mmap 2
by default, but it can be configured at run time to use only one or the other.
If resource limits are not a primary concern, the preferred configuration is
.Ev MALLOC_OPTIONS=dM
or
.Ev MALLOC_OPTIONS=DM .
When so configured, the
.Ar datasize
resource limit has little practical effect for typical applications.
The
resource limit has little practical effect for typical applications; use
.Ev MALLOC_OPTIONS=Dm
if that is a concern.
Regardless of allocator configuration, the
.Ar vmemoryuse
resource limit, however, can be used to bound the total virtual memory used by
a process, as described in
resource limit can be used to bound the total virtual memory used by a
process, as described in
.Xr limits 1 .
.Pp
This allocator uses multiple arenas in order to reduce lock contention for

15
lib/libc/stdlib/malloc.c
View file

@ -1,5 +1,5 @@
/*-
* Copyright (C) 2006,2007 Jason Evans <jasone@FreeBSD.org>.
* Copyright (C) 2006-2008 Jason Evans <jasone@FreeBSD.org>.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@ -131,7 +131,7 @@
* unnecessary, but we are burdened by history and the lack of resource limits
* for anonymous mapped memory.
*/
#define MALLOC_DSS
#define MALLOC_DSS
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
@ -821,7 +821,7 @@ static bool opt_junk = false;
#endif
#ifdef MALLOC_DSS
static bool opt_dss = true;
static bool opt_mmap = false;
static bool opt_mmap = true;
#endif
static bool opt_hint = false;
#ifdef MALLOC_LAZY_FREE
@ -1646,6 +1646,7 @@ chunk_alloc_mmap(size_t size)
return (NULL);
/* Clean up unneeded leading/trailing space. */
offset = CHUNK_ADDR2OFFSET(ret);
if (offset != 0) {
/* Leading space. */
pages_unmap(ret, chunksize - offset);
@ -1661,11 +1662,11 @@ chunk_alloc_mmap(size_t size)
pages_unmap((void *)((uintptr_t)ret + size),
chunksize);
}
} else {
/* Clean up unneeded leading space. */
pages_unmap(ret, chunksize - offset);
ret = (void *)((uintptr_t)ret + (chunksize - offset));
}
/* Clean up unneeded leading space. */
pages_unmap(ret, chunksize - offset);
ret = (void *)((uintptr_t)ret + (chunksize - offset));
}
return (ret);