mirror_zfs/include
Matthew Ahrens 8805abb8fc single-chunk scatter ABDs can be treated as linear
Scatter ABD's are allocated from a number of pages.  In contrast to
linear ABD's, these pages are disjoint in the kernel's virtual address
space, so they can't be accessed as a contiguous buffer.  Therefore
routines that need a linear buffer (e.g. abd_borrow_buf() and friends)
must allocate a separate linear buffer (with zio_buf_alloc()), and copy
the contents of the pages to/from the linear buffer.  This can have a
measurable performance overhead on some workloads.

https://github.com/zfsonlinux/zfs/commit/87c25d567fb7969b44c7d8af63990e
("abd_alloc should use scatter for >1K allocations") increased the use
of scatter ABD's, specifically switching 1.5K through 4K (inclusive)
buffers from linear to scatter.  For workloads that access blocks whose
compressed sizes are in this range, that commit introduced an additional
copy into the read code path.  For example, the
sequential_reads_arc_cached tests in the test suite were reduced by
around 5% (this is doing reads of 8K-logical blocks, compressed to 3K,
which are cached in the ARC).

This commit treats single-chunk scattered buffers as linear buffers,
because they are contiguous in the kernel's virtual address space.

All single-page (4K) ABD's can be represented this way.  Some multi-page
ABD's can also be represented this way, if we were able to allocate a
single "chunk" (higher-order "page" which represents a power-of-2 series
of physically-contiguous pages).  This is often the case for 2-page (8K)
ABD's.

Representing a single-entry scatter ABD as a linear ABD has the
performance advantage of avoiding the copy (and allocation) in
abd_borrow_buf_copy / abd_return_buf_copy.  A performance increase of
around 5% has been observed for ARC-cached reads (of small blocks which
can take advantage of this), fixing the regression introduced by
87c25d567.

Note that this optimization is only possible because all physical memory
is always mapped into the kernel's address space.  This is not the case
for HIGHMEM pages, so the optimization can not be made on 32-bit
systems.

Reviewed-by: Chunwei Chen <tuxoko@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #8580
2020-01-22 13:48:56 -08:00
..
linux Minor style cleanup 2019-09-25 11:27:49 -07:00
spl Retire unused spl_{mutex,rwlock}_{init_fini} 2019-09-25 11:27:49 -07:00
sys single-chunk scatter ABDs can be treated as linear 2020-01-22 13:48:56 -08:00
.gitignore Create /proc/sys/kernel/spl/gitrev with git hash 2018-10-08 21:57:02 -07:00
libnvpair.h Add JSON output support to channel programs 2018-03-19 12:40:58 -07:00
libuutil_common.h Support custom build directories and move includes 2010-09-08 12:38:56 -07:00
libuutil_impl.h Support custom build directories and move includes 2010-09-08 12:38:56 -07:00
libuutil.h Correct cppcheck errors 2017-09-19 12:17:29 -07:00
libzfs_core.h Add TRIM support 2019-03-29 09:13:20 -07:00
libzfs_impl.h OpenZFS 8115 - parallel zfs mount 2018-11-15 11:33:58 -08:00
libzfs.h Race condition between spa async threads and export 2019-09-25 11:27:50 -07:00
libzutil.h Fix libudev dependency in libzutil 2018-11-06 17:47:52 -08:00
Makefile.am Add libzutil for libzfs or libzpool consumers 2018-11-05 11:22:33 -08:00
thread_pool.h Add libtpool (thread pools) 2017-08-09 15:31:08 -07:00
zfeature_common.h Add bookmark v2 on-disk feature 2019-03-13 10:58:39 -07:00
zfs_comutil.h OpenZFS 9337 - zfs get all is slow due to uncached metadata 2018-07-12 10:49:27 -07:00
zfs_deleg.h OpenZFS 7614, 9064 - zfs device evacuation/removal 2018-04-14 12:16:17 -07:00
zfs_fletcher.h DLPX-44812 integrate EP-220 large memory scalability 2016-11-29 14:34:27 -08:00
zfs_namecheck.h Restrict filesystem creation if name referred either '.' or '..' 2019-09-25 11:27:47 -07:00
zfs_prop.h Add zfs module feature and property info to sysfs 2018-09-02 12:09:53 -07:00