Memory Management - Revision history

Tim Bird: /* Additional Resources/Mailing Lists */ add McKenney article on caches

2012-07-10T19:59:06Z

‎Additional Resources/Mailing Lists: add McKenney article on caches

2012-05-02T17:37:21Z

2011-06-22T19:45:20Z

Added syntax highlighting to code.

2011-01-12T19:15:58Z

‎Reserving (and accessing) the top of memory on startup: removed source highlighting, which was causing a problem with the sample code display

Wmat: Swapped

 tags for #source SyntaxHighlighting.

2010-03-26T19:53:04Z

Swapped <pre> tags for #source SyntaxHighlighting.

2009-12-02T21:13:50Z

2009-10-01T22:24:14Z

‎Enhanced Out-Of-Memory handling: missed one

2009-10-01T22:23:40Z

‎Enhanced Out-Of-Memory handling: add subsections

2009-10-01T22:07:41Z

‎Enhanced Out-Of-Memory (OOM) handling

2009-10-01T22:06:13Z

‎Enhanced Out-Of-Memory (OOM) handling: add link to ELC presentation for OOM cgroup stuff

← Older revision		Revision as of 19:59, 10 July 2012
Line 119:		Line 119:

	*[http://lwn.net/Articles/250967/ Everything about memory that a programmer should know]		*[http://lwn.net/Articles/250967/ Everything about memory that a programmer should know]
		+
		+	=== Articles on caches ===
		+	Excellent paper (2010) by Paul McKenney on how CPU caches operate: http://irl.cs.ucla.edu/~yingdi/paperreading/whymb.2010.06.07c.pdf

	[[Category:Linux]]		[[Category:Linux]]

@@ Line 5: / Line 5: @@
 Most of these areas have wider reaching implications, but are relatively simpler in the embedded case, largely thanks to not having to contend with swap and things of that nature. Simpler memory management as well as vendors not afraid of deviation from mainline for product programs makes for an excellent playground for experimenting with new things in the memory management and virtual memory space.
-=== Memory Measurement ===
+=== Memory Measurement and Analysis ===
 Analyzing the amount of system memory in use and available is trickier than it sounds.
@@ Line 11: / Line 11: @@
 * See [[Accurate Memory Measurement]] for some different techniques for dealing with inadequacies in current memory measurement systems.
 === Huge/large/superpages ===

@@ Line 38: / Line 38: @@
 but you use the cmdline mem=60M to reserve 4MB at the end. Then your
 reserved memory will be at 0x83c00000, so all you need to do is
-<pre>
+<syntaxhighlight lang="c">
 int fd;
 char *reserved_memory;
@@ Line 44: / Line 44: @@
 fd = open("/dev/mem",O_RDWR);
 reserved_memory = (char *) mmap(0,4*1024*1024,PROT_READ|PROT_WRITE,MAP_SHARED,fd,0x83c00000);
-</pre>
+</syntaxhighlight>
 ----

@@ Line 38: / Line 38: @@
 but you use the cmdline mem=60M to reserve 4MB at the end. Then your
 reserved memory will be at 0x83c00000, so all you need to do is
-{{#source: c |
+<pre>
 int fd;
 char *reserved_memory;
 fd = open("/dev/mem",O_RDWR);
-reserved_memory = (char *) mmap(0,4*1024*1024,PROT_READ| PROT_WRITE,MAP_SHARED,fd,0x83c00000);
+reserved_memory = (char *) mmap(0,4*1024*1024,PROT_READ|PROT_WRITE,MAP_SHARED,fd,0x83c00000);
-}}
+</pre>
 ----

@@ Line 38: / Line 38: @@
 but you use the cmdline mem=60M to reserve 4MB at the end. Then your
 reserved memory will be at 0x83c00000, so all you need to do is
+{{#source: c |
-<pre>
 int fd;
 char *reserved_memory;
@@ Line 45: / Line 44: @@
 fd = open("/dev/mem",O_RDWR);
 reserved_memory = (char *) mmap(0,4*1024*1024,PROT_READ| PROT_WRITE,MAP_SHARED,fd,0x83c00000);
-</pre>
+}}
 ----

@@ Line 100: / Line 100: @@
 condition. Applications visible to user or otherwise non-background
 killable are then supposed to free their caches and/or disable features
-that could take a lot of additional memory.  If the caches are from heap
+that could take a lot of additional memory.  If the caches are from [[Heap_memory|heap]]
-instead of memory mapped, it's less likely to help because of heap
+instead of memory mapped, it's less likely to help because of [[Heap_memory|heap]]
 fragmentation and it requiring more work/time though.
 </pre>

@@ Line 64: / Line 64: @@
 ** http://lkml.org/lkml/2009/7/7/410 - thread about second submission to LKML in July 2009
 *** http://lkml.org/lkml/2009/7/16/468 - Balbir Singh wants to take time to evaluate the change....
 * mem_notify patches
 ** This set of patches provided a mechanism to notify user-space when memory is getting low, allowing for application-based handling of the condition.  These patches were submitted in January 2008.

@@ Line 57: / Line 57: @@
 Here are some technologies that I know about (these need to be researched and documented better):
 * Memory usage limit notification (By Embedded Alley, sponsored by CE Linux Forum)
 ** This patch updates the Memory Controller cgroup to add a configurable memory usage limit notification.  The feature was presented at the April 2009 Embedded Linux Conference.
@@ Line 67: / Line 68: @@
 ** This patch cannot be applied to versions beyond 2.6.28 because the memory management reclaiming sequence have changed.
 ** See http://lwn.net/Articles/267013/
 * Google per-cgroup OOM handler
 ** Google posted a Request For Comments (RFC) for OOM handling implemented in a per-cgroup fashion.  See http://article.gmane.org/gmane.linux.kernel.mm/28376
 * Nokia OOM enhancements
 ** Maemo application enhancements referenced at: http://lwn.net/Articles/267013/ (search for "killable" in the comments)
@@ Line 98: / Line 103: @@
 fragmentation and it requiring more work/time though.
 </pre>
 * Paul Mundt submitted a patch to CELF for the 2.6.12 kernel which provided low-memory notifications to user space.  See [[Accurate_Memory_Measurement#Nokia_out-of-memory_notifier_module]] for more information.
 ** This module was based on the Linux Security Module system, which has been removed from recent kernels.

@@ Line 48: / Line 48: @@
 ----
-=== Enhanced Out-Of-Memory (OOM) handling ===
+=== Enhanced Out-Of-Memory handling ===
-Several technologies have been developed and suggested for improving the handling out-of-memory conditions with Linux systems.
+Several technologies have been developed and suggested for improving the handling Out-Of-Memory (OOM) conditions with Linux systems.
 See http://linux-mm.org/OOM_Killer for information about the OOM killer in the Linux kernel.