Commit Graph

11 Commits

Author SHA1 Message Date
Attila Fülöp 78289b8458
zcommon: Refactor FPU state handling in fletcher4
Currently calls to kfpu_begin() and kfpu_end() are split between
the init() and fini() functions of the particular SIMD
implementation. This was done in #14247 as an optimization measure
for the ABD adapter. Unfortunately the split complicates FPU
handling on platforms that use a local FPU state buffer, like
Windows and macOS.

To ease porting, we introduce a boolean struct member in
fletcher_4_ops_t, indicating use of the FPU, and move the FPU state
handling from the SIMD implementations to the call sites.

Reviewed-by: Tino Reichardt <milky-zfs@mcmilk.de>
Reviewed-by: Richard Yao <richard.yao@alumni.stonybrook.edu>
Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Signed-off-by: Attila Fülöp <attila@fueloep.org>
Closes #14600
2023-03-14 09:45:28 -07:00
Richard Yao 59493b63c1
Micro-optimize fletcher4 calculations
When processing abds, we execute 1 `kfpu_begin()`/`kfpu_end()` pair on
every page in the abd. This is wasteful and slows down checksum
performance versus what the benchmark claimed. We correct this by moving
those calls to the init and fini functions.

Also, we always check the buffer length against 0 before calling the
non-scalar checksum functions. This means that we do not need to execute
the loop condition for the first loop iteration. That allows us to
micro-optimize the checksum calculations by switching to do-while loops.

Note that we do not apply that micro-optimization to the scalar
implementation because there is no check in
`fletcher_4_incremental_native()`/`fletcher_4_incremental_byteswap()`
against 0 sized buffers being passed.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Richard Yao <richard.yao@alumni.stonybrook.edu>
Closes #14247
2022-12-05 11:00:34 -08:00
наб d465fc5844 Forbid b{copy,zero,cmp}(). Don't include <strings.h> for <string.h>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #12996
2022-03-15 15:13:48 -07:00
наб 861166b027 Remove bcopy(), bzero(), bcmp()
bcopy() has a confusing argument order and is actually a move, not a
copy; they're all deprecated since POSIX.1-2001 and removed in -2008,
and we shim them out to mem*() on Linux anyway

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes #12996
2022-03-15 15:13:42 -07:00
Damian Szuberski 63652e1546
Add `--enable-asan` and `--enable-ubsan` switches
`configure` now accepts `--enable-asan` and `--enable-ubsan` switches
which results in passing `-fsanitize=address`
and `-fsanitize=undefined`, respectively, to the compiler. Those
flags are enabled in GitHub workflows for ZTS and zloop. Errors
reported by both instrumentations are corrected, except for:

- Memory leak reporting is (temporarily) suppressed. The cost of
  fixing them is relatively high compared to the gains.

- Checksum computing functions in `module/zcommon/zfs_fletcher*`
  have UBSan errors suppressed. It is completely impractical
  to enforce 64-byte payload alignment there due to performance
  impact.

- There's no ASan heap poisoning in `module/zstd/lib/zstd.c`. A custom
  memory allocator is used there rendering that measure
  unfeasible.

- Memory leaks detection has to be suppressed for `cmd/zvol_id`.
  `zvol_id` is run by udev with the help of `ptrace(2)`. Tracing is
  incompatible with memory leaks detection.

Reviewed-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: szubersk <szuberskidamian@gmail.com>
Closes #12928
2022-02-03 14:35:38 -08:00
Matthew Macy 006e9a4088 OpenZFS restructuring - move platform specific headers
Move platform specific Linux headers under include/os/linux/.
Update the build system accordingly to detect the platform.
This lays some of the initial groundwork to supporting building
for other platforms.

As part of this change it was necessary to create both a user
and kernel space sys/simd.h header which can be included in
either context.  No functional change, the source has been
refactored and the relevant #include's updated.

Reviewed-by: Jorgen Lundman <lundman@lundman.net>
Reviewed-by: Igor Kozhukhov <igor@dilos.org>
Signed-off-by: Matthew Macy <mmacy@FreeBSD.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #9198
2019-09-05 09:34:54 -07:00
Brian Behlendorf e5db313494
Linux 5.0 compat: SIMD compatibility
Restore the SIMD optimization for 4.19.38 LTS, 4.14.120 LTS,
and 5.0 and newer kernels.  This is accomplished by leveraging
the fact that by definition dedicated kernel threads never need
to concern themselves with saving and restoring the user FPU state.
Therefore, they may use the FPU as long as we can guarantee user
tasks always restore their FPU state before context switching back
to user space.

For the 5.0 and 5.1 kernels disabling preemption and local
interrupts is sufficient to allow the FPU to be used.  All non-kernel
threads will restore the preserved user FPU state.

For 5.2 and latter kernels the user FPU state restoration will be
skipped if the kernel determines the registers have not changed.
Therefore, for these kernels we need to perform the additional
step of saving and restoring the FPU registers.  Invalidating the
per-cpu global tracking the FPU state would force a restore but
that functionality is private to the core x86 FPU implementation
and unavailable.

In practice, restricting SIMD to kernel threads is not a major
restriction for ZFS.  The vast majority of SIMD operations are
already performed by the IO pipeline.  The remaining cases are
relatively infrequent and can be handled by the generic code
without significant impact.  The two most noteworthy cases are:

  1) Decrypting the wrapping key for an encrypted dataset,
     i.e. `zfs load-key`.  All other encryption and decryption
     operations will use the SIMD optimized implementations.

  2) Generating the payload checksums for a `zfs send` stream.

In order to avoid making any changes to the higher layers of ZFS
all of the `*_get_ops()` functions were updated to take in to
consideration the calling context.  This allows for the fastest
implementation to be used as appropriate (see kfpu_allowed()).

The only other notable instance of SIMD operations being used
outside a kernel thread was at module load time.  This code
was moved in to a taskq in order to accommodate the new kernel
thread restriction.

Finally, a few other modifications were made in order to further
harden this code and facilitate testing.  They include updating
each implementations operations structure to be declared as a
constant.  And allowing "cycle" to be set when selecting the
preferred ops in the kernel as well as user space.

Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #8754 
Closes #8793 
Closes #8965
2019-07-12 09:31:20 -07:00
Brian Behlendorf 93ce2b4ca5 Update build system and packaging
Minimal changes required to integrate the SPL sources in to the
ZFS repository build infrastructure and packaging.

Build system and packaging:
  * Renamed SPL_* autoconf m4 macros to ZFS_*.
  * Removed redundant SPL_* autoconf m4 macros.
  * Updated the RPM spec files to remove SPL package dependency.
  * The zfs package obsoletes the spl package, and the zfs-kmod
    package obsoletes the spl-kmod package.
  * The zfs-kmod-devel* packages were updated to add compatibility
    symlinks under /usr/src/spl-x.y.z until all dependent packages
    can be updated.  They will be removed in a future release.
  * Updated copy-builtin script for in-kernel builds.
  * Updated DKMS package to include the spl.ko.
  * Updated stale AUTHORS file to include all contributors.
  * Updated stale COPYRIGHT and included the SPL as an exception.
  * Renamed README.markdown to README.md
  * Renamed OPENSOLARIS.LICENSE to LICENSE.
  * Renamed DISCLAIMER to NOTICE.

Required code changes:
  * Removed redundant HAVE_SPL macro.
  * Removed _BOOT from nvpairs since it doesn't apply for Linux.
  * Initial header cleanup (removal of empty headers, refactoring).
  * Remove SPL repository clone/build from zimport.sh.
  * Use of DEFINE_RATELIMIT_STATE and DEFINE_SPINLOCK removed due
    to build issues when forcing C99 compilation.
  * Replaced legacy ACCESS_ONCE with READ_ONCE.
  * Include needed headers for `current` and `EXPORT_SYMBOL`.

Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Pavel Zakharov <pavel.zakharov@delphix.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
TEST_ZIMPORT_SKIP="yes"
Closes #7556
2018-05-29 16:00:33 -07:00
Gvozden Neskovic 5bf703b8f3 Fletcher4: save/reload implementation context
Init, compute, and fini methods are changed to work on internal context object.
This is necessary because ABI does not guarantee that SIMD registers will be preserved
on function calls. This is technically the case in Linux kernel in between
`kfpu_begin()/kfpu_end()`, but it breaks user-space tests and some kernels that
don't require disabling preemption for using SIMD (osx).

Use scalar compute methods in-place for small buffers, and when the buffer size
does not meet SIMD size alignment.

Signed-off-by: Gvozden Neskovic <neskovic@gmail.com>
2016-10-05 16:41:46 +02:00
Gvozden Neskovic fc897b24b2 Rework of fletcher_4 module
- Benchmark memory block is increased to 128kiB to reflect real block sizes more
accurately. Measurements include all three stages needed for checksum generation,
i.e. `init()/compute()/fini()`. The inner loop is repeated multiple times to offset
overhead of time function.

- Fastest implementation selects native and byteswap methods independently in
benchmark. To support this new function pointers `init_byteswap()/fini_byteswap()`
are introduced.

- Implementation mutex lock is replaced by atomic variable.

- To save time, benchmark is not executed in userspace. Instead, highest supported
implementation is used for fastest. Default userspace selector is still 'cycle'.

- `fletcher_4_native/byteswap()` methods use incremental methods to finish
calculation if data size is not multiple of vector stride (currently 64B).

- Added `fletcher_4_native_varsize()` special purpose method for use when buffer size
is not known in advance. The method does not enforce 4B alignment on buffer size, and
will ignore last (size % 4) bytes of the data buffer.

- Benchmark `kstat` is changed to match the one of vdev_raidz. It now shows
throughput for all supported implementations (in B/s), native and byteswap,
as well as the code [fastest] is running.

Example of `fletcher_4_bench` running on `Intel(R) Xeon(R) CPU E5-2660 v3 @ 2.60GHz`:
implementation   native         byteswap
scalar           4768120823     3426105750
sse2             7947841777     4318964249
ssse3            7951922722     6112191941
avx2             13269714358    11043200912
fastest          avx2           avx2

Example of `fletcher_4_bench` running on `Intel(R) Xeon Phi(TM) CPU 7210 @ 1.30GHz`:
implementation   native         byteswap
scalar           1291115967     1031555336
sse2             2539571138     1280970926
ssse3            2537778746     1080016762
avx2             4950749767     1078493449
avx512f          9581379998     4010029046
fastest          avx512f        avx512f

Signed-off-by: Gvozden Neskovic <neskovic@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #4952
2016-08-16 14:11:55 -07:00
Jinshan Xiong 1eeb4562a7 Implementation of AVX2 optimized Fletcher-4
New functionality:
- Preserves existing scalar implementation.
- Adds AVX2 optimized Fletcher-4 computation.
- Fastest routines selected on module load (benchmark).
- Test case for Fletcher-4 added to ztest.

New zcommon module parameters:
-  zfs_fletcher_4_impl (str): selects the implementation to use.
    "fastest" - use the fastest version available
    "cycle"   - cycle trough all available impl for ztest
    "scalar"  - use the original version
    "avx2"    - new AVX2 implementation if available

Performance comparison (Intel i7 CPU, 1MB data buffers):
- Scalar:  4216 MB/s
- AVX2:   14499 MB/s

See contents of `/sys/module/zcommon/parameters/zfs_fletcher_4_impl`
to get list of supported values. If an implementation is not supported
on the system, it will not be shown. Currently selected option is
enclosed in `[]`.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #4330
2016-06-02 14:30:51 -07:00