tjh.dev / kernel
Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux
1
fork

Configure Feed

Select the types of activity you want to include in your feed.

nommu: make the initial mmap allocation excess behaviour Kconfig configurable

NOMMU mmap() has an option controlled by a sysctl variable that determines
whether the allocations made by do_mmap_private() should have the excess
space trimmed off and returned to the allocator. Make the initial setting
of this variable a Kconfig configuration option.

The reason there can be excess space is that the allocator only allocates
in power-of-2 size chunks, but mmap()'s can be made in sizes that aren't a
power of 2.

There are two alternatives:

(1) Keep the excess as dead space. The dead space then remains unused for the
lifetime of the mapping. Mappings of shared objects such as libc, ld.so
or busybox's text segment may retain their dead space forever.

(2) Return the excess to the allocator. This means that the dead space is
limited to less than a page per mapping, but it means that for a transient
process, there's more chance of fragmentation as the excess space may be
reused fairly quickly.

During the boot process, a lot of transient processes are created, and
this can cause a lot of fragmentation as the pagecache and various slabs
grow greatly during this time.

By turning off the trimming of excess space during boot and disabling
batching of frees, Coldfire can manage to boot.

A better way of doing things might be to have /sbin/init turn this option
off. By that point libc, ld.so and init - which are all long-duration
processes - have all been loaded and trimmed.

Reported-by: Lanttor Guo <lanttor.guo@freescale.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: Lanttor Guo <lanttor.guo@freescale.com>
Cc: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

authored by

David Howells and committed by
Linus Torvalds fc4d5c29 3a6be87f

+29 -1
+28
mm/Kconfig
··· 225 225 226 226 config MMU_NOTIFIER 227 227 bool 228 + 229 + config NOMMU_INITIAL_TRIM_EXCESS 230 + int "Turn on mmap() excess space trimming before booting" 231 + depends on !MMU 232 + default 1 233 + help 234 + The NOMMU mmap() frequently needs to allocate large contiguous chunks 235 + of memory on which to store mappings, but it can only ask the system 236 + allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently 237 + more than it requires. To deal with this, mmap() is able to trim off 238 + the excess and return it to the allocator. 239 + 240 + If trimming is enabled, the excess is trimmed off and returned to the 241 + system allocator, which can cause extra fragmentation, particularly 242 + if there are a lot of transient processes. 243 + 244 + If trimming is disabled, the excess is kept, but not used, which for 245 + long-term mappings means that the space is wasted. 246 + 247 + Trimming can be dynamically controlled through a sysctl option 248 + (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of 249 + excess pages there must be before trimming should occur, or zero if 250 + no trimming is to occur. 251 + 252 + This option specifies the initial value of this option. The default 253 + of 1 says that all excess pages should be trimmed. 254 + 255 + See Documentation/nommu-mmap.txt for more information.
+1 -1
mm/nommu.c
··· 66 66 int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */ 67 67 int sysctl_overcommit_ratio = 50; /* default is 50% */ 68 68 int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT; 69 - int sysctl_nr_trim_pages = 1; /* page trimming behaviour */ 69 + int sysctl_nr_trim_pages = CONFIG_NOMMU_INITIAL_TRIM_EXCESS; 70 70 int heap_stack_gap = 0; 71 71 72 72 atomic_long_t mmap_pages_allocated;