linux/Documentation/vm/page_owner.rst
Zhenliang Wei f7df2b1cf0 tools/vm/page_owner_sort.c: count and sort by mem
When viewing page owner information, we may be more concerned about the
total memory rather than the times of stack appears.  Therefore, the
following adjustments are made:

1. Added the statistics on the total number of pages.

2. Added the optional parameter "-m" to configure the program to sort by
   memory (total pages).

The general output of page_owner is as follows:

	Page allocated via order XXX, ...
	PFN XXX ...
	 // Detailed stack

	Page allocated via order XXX, ...
	PFN XXX ...
	 // Detailed stack

The original page_owner_sort ignores PFN rows, puts the remaining rows
in buf, counts the times of buf, and finally sorts them according to the
times.  General output:

	XXX times:
	Page allocated via order XXX, ...
	 // Detailed stack

Now, we use regexp to extract the page order value from the buf, and
count the total pages for the buf.  General output:

	XXX times, XXX pages:
	Page allocated via order XXX, ...
	 // Detailed stack

By default, it is still sorted by the times of buf; If you want to sort
by the pages nums of buf, use the new -m parameter.

Link: https://lkml.kernel.org/r/1631678242-41033-1-git-send-email-weizhenliang@huawei.com
Signed-off-by: Zhenliang Wei <weizhenliang@huawei.com>
Cc: Tang Bin <tangbin@cmss.chinamobile.com>
Cc: Zhang Shengju <zhangshengju@cmss.chinamobile.com>
Cc: Zhenliang Wei <weizhenliang@huawei.com>
Cc: Xiaoming Ni <nixiaoming@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2021-11-06 13:30:40 -07:00

111 lines
4.4 KiB
ReStructuredText

.. _page_owner:
==================================================
page owner: Tracking about who allocated each page
==================================================
Introduction
============
page owner is for the tracking about who allocated each page.
It can be used to debug memory leak or to find a memory hogger.
When allocation happens, information about allocation such as call stack
and order of pages is stored into certain storage for each page.
When we need to know about status of all pages, we can get and analyze
this information.
Although we already have tracepoint for tracing page allocation/free,
using it for analyzing who allocate each page is rather complex. We need
to enlarge the trace buffer for preventing overlapping until userspace
program launched. And, launched program continually dump out the trace
buffer for later analysis and it would change system behaviour with more
possibility rather than just keeping it in memory, so bad for debugging.
page owner can also be used for various purposes. For example, accurate
fragmentation statistics can be obtained through gfp flag information of
each page. It is already implemented and activated if page owner is
enabled. Other usages are more than welcome.
page owner is disabled in default. So, if you'd like to use it, you need
to add "page_owner=on" into your boot cmdline. If the kernel is built
with page owner and page owner is disabled in runtime due to no enabling
boot option, runtime overhead is marginal. If disabled in runtime, it
doesn't require memory to store owner information, so there is no runtime
memory overhead. And, page owner inserts just two unlikely branches into
the page allocator hotpath and if not enabled, then allocation is done
like as the kernel without page owner. These two unlikely branches should
not affect to allocation performance, especially if the static keys jump
label patching functionality is available. Following is the kernel's code
size change due to this facility.
- Without page owner::
text data bss dec hex filename
48392 2333 644 51369 c8a9 mm/page_alloc.o
- With page owner::
text data bss dec hex filename
48800 2445 644 51889 cab1 mm/page_alloc.o
6662 108 29 6799 1a8f mm/page_owner.o
1025 8 8 1041 411 mm/page_ext.o
Although, roughly, 8 KB code is added in total, page_alloc.o increase by
520 bytes and less than half of it is in hotpath. Building the kernel with
page owner and turning it on if needed would be great option to debug
kernel memory problem.
There is one notice that is caused by implementation detail. page owner
stores information into the memory from struct page extension. This memory
is initialized some time later than that page allocator starts in sparse
memory system, so, until initialization, many pages can be allocated and
they would have no owner information. To fix it up, these early allocated
pages are investigated and marked as allocated in initialization phase.
Although it doesn't mean that they have the right owner information,
at least, we can tell whether the page is allocated or not,
more accurately. On 2GB memory x86-64 VM box, 13343 early allocated pages
are catched and marked, although they are mostly allocated from struct
page extension feature. Anyway, after that, no page is left in
un-tracking state.
Usage
=====
1) Build user-space helper::
cd tools/vm
make page_owner_sort
2) Enable page owner: add "page_owner=on" to boot cmdline.
3) Do the job what you want to debug
4) Analyze information from page owner::
cat /sys/kernel/debug/page_owner > page_owner_full.txt
./page_owner_sort page_owner_full.txt sorted_page_owner.txt
The general output of ``page_owner_full.txt`` is as follows:
Page allocated via order XXX, ...
PFN XXX ...
// Detailed stack
Page allocated via order XXX, ...
PFN XXX ...
// Detailed stack
The ``page_owner_sort`` tool ignores ``PFN`` rows, puts the remaining rows
in buf, uses regexp to extract the page order value, counts the times
and pages of buf, and finally sorts them according to the times.
See the result about who allocated each page
in the ``sorted_page_owner.txt``. General output:
XXX times, XXX pages:
Page allocated via order XXX, ...
// Detailed stack
By default, ``page_owner_sort`` is sorted according to the times of buf.
If you want to sort by the pages nums of buf, use the ``-m`` parameter.