Go to file
Dennis Zhou (Facebook) 40064aeca3 percpu: replace area map allocator with bitmap
The percpu memory allocator is experiencing scalability issues when
allocating and freeing large numbers of counters as in BPF.
Additionally, there is a corner case where iteration is triggered over
all chunks if the contig_hint is the right size, but wrong alignment.

This patch replaces the area map allocator with a basic bitmap allocator
implementation. Each subsequent patch will introduce new features and
replace full scanning functions with faster non-scanning options when
possible.

Implementation:
This patchset removes the area map allocator in favor of a bitmap
allocator backed by metadata blocks. The primary goal is to provide
consistency in performance and memory footprint with a focus on small
allocations (< 64 bytes). The bitmap removes the heavy memmove from the
freeing critical path and provides a consistent memory footprint. The
metadata blocks provide a bound on the amount of scanning required by
maintaining a set of hints.

In an effort to make freeing fast, the metadata is updated on the free
path if the new free area makes a page free, a block free, or spans
across blocks. This causes the chunk's contig hint to potentially be
smaller than what it could allocate by up to the smaller of a page or a
block. If the chunk's contig hint is contained within a block, a check
occurs and the hint is kept accurate. Metadata is always kept accurate
on allocation, so there will not be a situation where a chunk has a
later contig hint than available.

Evaluation:
I have primarily done testing against a simple workload of allocation of
1 million objects (2^20) of varying size. Deallocation was done by in
order, alternating, and in reverse. These numbers were collected after
rebasing ontop of a80099a152. I present the worst-case numbers here:

  Area Map Allocator:

        Object Size | Alloc Time (ms) | Free Time (ms)
        ----------------------------------------------
              4B    |        310      |     4770
             16B    |        557      |     1325
             64B    |        436      |      273
            256B    |        776      |      131
           1024B    |       3280      |      122

  Bitmap Allocator:

        Object Size | Alloc Time (ms) | Free Time (ms)
        ----------------------------------------------
              4B    |        490      |       70
             16B    |        515      |       75
             64B    |        610      |       80
            256B    |        950      |      100
           1024B    |       3520      |      200

This data demonstrates the inability for the area map allocator to
handle less than ideal situations. In the best case of reverse
deallocation, the area map allocator was able to perform within range
of the bitmap allocator. In the worst case situation, freeing took
nearly 5 seconds for 1 million 4-byte objects. The bitmap allocator
dramatically improves the consistency of the free path. The small
allocations performed nearly identical regardless of the freeing
pattern.

While it does add to the allocation latency, the allocation scenario
here is optimal for the area map allocator. The area map allocator runs
into trouble when it is allocating in chunks where the latter half is
full. It is difficult to replicate this, so I present a variant where
the pages are second half filled. Freeing was done sequentially. Below
are the numbers for this scenario:

  Area Map Allocator:

        Object Size | Alloc Time (ms) | Free Time (ms)
        ----------------------------------------------
              4B    |       4118      |     4892
             16B    |       1651      |     1163
             64B    |        598      |      285
            256B    |        771      |      158
           1024B    |       3034      |      160

  Bitmap Allocator:

        Object Size | Alloc Time (ms) | Free Time (ms)
        ----------------------------------------------
              4B    |        481      |       67
             16B    |        506      |       69
             64B    |        636      |       75
            256B    |        892      |       90
           1024B    |       3262      |      147

The data shows a parabolic curve of performance for the area map
allocator. This is due to the memmove operation being the dominant cost
with the lower object sizes as more objects are packed in a chunk and at
higher object sizes, the traversal of the chunk slots is the dominating
cost. The bitmap allocator suffers this problem as well. The above data
shows the inability to scale for the allocation path with the area map
allocator and that the bitmap allocator demonstrates consistent
performance in general.

The second problem of additional scanning can result in the area map
allocator completing in 52 minutes when trying to allocate 1 million
4-byte objects with 8-byte alignment. The same workload takes
approximately 16 seconds to complete for the bitmap allocator.

V2:
Fixed a bug in pcpu_alloc_first_chunk end_offset was setting the bitmap
using bytes instead of bits.

Added a comment to pcpu_cnt_pop_pages to explain bitmap_weight.

Signed-off-by: Dennis Zhou <dennisszhou@gmail.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2017-07-26 17:41:05 -04:00
Documentation This series converts a number of top-level documents to the RST format 2017-07-15 12:58:58 -07:00
arch Merge branch 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2017-07-15 12:00:42 -07:00
block Merge branch 'for-linus' of git://git.kernel.dk/linux-block 2017-07-11 15:36:52 -07:00
certs modsign: add markers to endif-statements in certs/Makefile 2017-07-14 11:01:37 +10:00
crypto Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 2017-07-14 22:49:50 -07:00
drivers Add wait_for_random_bytes() and get_random_*_wait() functions so that 2017-07-15 12:44:02 -07:00
firmware firmware/Makefile: force recompilation if makefile changes 2017-05-08 17:15:10 -07:00
fs Add wait_for_random_bytes() and get_random_*_wait() functions so that 2017-07-15 12:44:02 -07:00
include percpu: replace area map allocator with bitmap 2017-07-26 17:41:05 -04:00
init percpu: replace area map allocator with bitmap 2017-07-26 17:41:05 -04:00
ipc ipc/util.h: update documentation for ipc_getref() and ipc_putref() 2017-07-12 16:26:02 -07:00
kernel Merge branch 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2017-07-15 12:00:42 -07:00
lib Add wait_for_random_bytes() and get_random_*_wait() functions so that 2017-07-15 12:44:02 -07:00
mm percpu: replace area map allocator with bitmap 2017-07-26 17:41:05 -04:00
net Add wait_for_random_bytes() and get_random_*_wait() functions so that 2017-07-15 12:44:02 -07:00
samples Merge branch 'akpm' (patches from Andrew) 2017-07-13 12:38:49 -07:00
scripts Kbuild updates for v4.13 (2nd) 2017-07-13 13:37:57 -07:00
security KEYS: DH: validate __spare field 2017-07-14 11:01:38 +10:00
sound sound fixes for 4.13-rc1 2017-07-14 12:44:00 -07:00
tools Merge branch 'akpm' (patches from Andrew) 2017-07-14 21:57:25 -07:00
usr ramfs: clarify help text that compression applies to ramfs as well as legacy ramdisk. 2017-07-06 16:24:30 -07:00
virt Second batch of KVM updates for v4.13 2017-07-15 10:18:16 -07:00
.cocciconfig scripts: add Linux .cocciconfig for coccinelle 2016-07-22 12:13:39 +02:00
.get_maintainer.ignore Add hch to .get_maintainer.ignore 2015-08-21 14:30:10 -07:00
.gitattributes .gitattributes: set git diff driver for C source code files 2016-10-07 18:46:30 -07:00
.gitignore kbuild: Add support to generate LLVM assembly files 2017-04-25 08:13:52 +09:00
.mailmap power supply and reset changes for the v4.12 series (part 2) 2017-05-12 12:02:21 -07:00
COPYING
CREDITS avr32: remove support for AVR32 architecture 2017-05-01 09:27:15 +02:00
Kbuild kbuild: Consolidate header generation from ASM offset information 2017-04-13 05:43:37 +09:00
Kconfig
MAINTAINERS Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus 2017-07-15 10:59:54 -07:00
Makefile Linux v4.13-rc1 2017-07-15 15:22:10 -07:00
README README: add a new README file, pointing to the Documentation/ 2016-10-24 08:12:35 -02:00

README

Linux kernel
============

This file was moved to Documentation/admin-guide/README.rst

Please notice that there are several guides for kernel developers and users.
These guides can be rendered in a number of formats, like HTML and PDF.

In order to build the documentation, use ``make htmldocs`` or
``make pdfdocs``.

There are various text files in the Documentation/ subdirectory,
several of them using the Restructured Text markup notation.
See Documentation/00-INDEX for a list of what is contained in each file.

Please read the Documentation/process/changes.rst file, as it contains the
requirements for building and running the kernel, and information about
the problems which may result by upgrading your kernel.