kernel/linux.git/lib/lz4, branch v5.10.36

lz4: fix kernel decompression speed

2020-08-15T02:56:56+00:00

This patch replaces all memcpy() calls with LZ4_memcpy() which calls __builtin_memcpy() so the compiler can inline it. LZ4 relies heavily on memcpy() with a constant size being inlined. In x86 and i386 pre-boot environments memcpy() cannot be inlined because memcpy() doesn't get defined as __builtin_memcpy(). An equivalent patch has been applied upstream so that the next import won't lose this change [1]. I've measured the kernel decompression speed using QEMU before and after this patch for the x86_64 and i386 architectures. The speed-up is about 10x as shown below. Code Arch Kernel Size Time Speed v5.8 x86_64 11504832 B 148 ms 79 MB/s patch x86_64 11503872 B 13 ms 885 MB/s v5.8 i386 9621216 B 91 ms 106 MB/s patch i386 9620224 B 10 ms 962 MB/s I also measured the time to decompress the initramfs on x86_64, i386, and arm. All three show the same decompression speed before and after, as expected. [1] https://github.com/lz4/lz4/pull/890 Signed-off-by: Nick Terrell Signed-off-by: Andrew Morton Cc: Yann Collet Cc: Gao Xiang Cc: Sven Schmidt <4sschmid@informatik.uni-hamburg.de> Cc: Greg Kroah-Hartman Cc: Ingo Molnar Cc: Arvind Sankar Link: http://lkml.kernel.org/r/20200803194022.2966806-1-nickrterrell@gmail.com Signed-off-by: Linus Torvalds

lib/lz4/lz4_decompress.c: document deliberate use of `&'

2020-06-11T02:14:17+00:00

This operation was intentional, but tools such as smatch will warn that it might not have been. Signed-off-by: Andrew Morton Cc: Yann Collet Cc: Vasily Averin Cc: Gao Xiang Link: http://lkml.kernel.org/r/3bf931c6ea0cae3e23f3485801986859851b4f04.camel@perches.com Signed-off-by: Linus Torvalds

lz4: do not export static symbol

2019-09-20T16:06:26+00:00

Kbuild now complains (rightly) about it. Signed-off-by: Linus Torvalds

treewide: Add SPDX license identifier - Makefile/Kconfig

2019-05-21T08:50:46+00:00

Add SPDX license identifiers to all Make/Kconfig files which: - Have no license information of any form These files fall under the project license, GPL v2 only. The resulting SPDX license identifier is: GPL-2.0-only Signed-off-by: Thomas Gleixner Signed-off-by: Greg Kroah-Hartman

lib/lz4: update LZ4 decompressor module

2018-10-31T15:54:14+00:00

Update the LZ4 compression module based on LZ4 v1.8.3 in order for the erofs file system to use the newest LZ4_decompress_safe_partial() which can now decode exactly the nb of bytes requested [1] to take place of the open hacked code in the erofs file system itself. Currently, apart from the erofs file system, no other users use LZ4_decompress_safe_partial, so no worry about the interface. In addition, LZ4 v1.8.x boosts up decompression speed compared to the current code which is based on LZ4 v1.7.3, mainly due to shortcut optimization for the specific common LZ4-sequences [2]. lzbench testdata (tested in kirin710, 8 cores, 4 big cores at 2189Mhz, 2GB DDR RAM at 1622Mhz, with enwik8 testdata [3]): Compressor name Compress. Decompress. Compr. size Ratio Filename memcpy 5004 MB/s 4924 MB/s 100000000 100.00 enwik8 lz4hc 1.7.3 -9 12 MB/s 653 MB/s 42203253 42.20 enwik8 lz4hc 1.8.0 -9 12 MB/s 908 MB/s 42203096 42.20 enwik8 lz4hc 1.8.3 -9 11 MB/s 965 MB/s 42203094 42.20 enwik8 [1] https://github.com/lz4/lz4/issues/566 https://github.com/lz4/lz4/commit/08d347b5b217b011ff7487130b79480d8cfdaeb8 [2] v1.8.1 perf: slightly faster compression and decompression speed https://github.com/lz4/lz4/commit/a31b7058cb97e4393da55e78a77a1c6f0c9ae038 v1.8.2 perf: slightly faster HC compression and decompression speed https://github.com/lz4/lz4/commit/45f8603aae389d34c689d3ff7427b314071ccd2c https://github.com/lz4/lz4/commit/1a191b3f8d26b50a7c1d41590b529ec308d768cd [3] http://mattmahoney.net/dc/textdata.html http://mattmahoney.net/dc/enwik8.zip Link: http://lkml.kernel.org/r/1537181207-21932-1-git-send-email-gaoxiang25@huawei.com Signed-off-by: Gao Xiang Tested-by: Guo Xuenan Cc: Colin Ian King Cc: Yann Collet Cc: Greg Kroah-Hartman Cc: Fang Wei Cc: Chao Yu Cc: Miao Xie Cc: Sven Schmidt <4sschmid@informatik.uni-hamburg.de> Cc: Kyungsik Lee Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

lib/lz4: make arrays static const, reduces object code size

2017-10-04T00:54:25+00:00

Don't populate the read-only arrays dec32table and dec64table on the stack, instead make them both static const. Makes the object code smaller by over 10K bytes: Before: text data bss dec hex filename 31500 0 0 31500 7b0c lib/lz4/lz4_decompress.o After: text data bss dec hex filename 20237 176 0 20413 4fbd lib/lz4/lz4_decompress.o (gcc version 7.2.0 x86_64) Link: http://lkml.kernel.org/r/20170921221939.20820-1-colin.king@canonical.com Signed-off-by: Colin Ian King Cc: Christophe JAILLET Cc: Sven Schmidt <4sschmid@informatik.uni-hamburg.de> Cc: Arnd Bergmann Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

lib/lz4: remove back-compat wrappers

2017-02-25T01:46:57+00:00

Remove the functions introduced as wrappers for providing backwards compatibility to the prior LZ4 version. They're not needed anymore since there's no callers left. Link: http://lkml.kernel.org/r/1486321748-19085-6-git-send-email-4sschmid@informatik.uni-hamburg.de Signed-off-by: Sven Schmidt <4sschmid@informatik.uni-hamburg.de> Cc: Bongkyu Kim Cc: Rui Salvaterra Cc: Sergey Senozhatsky Cc: Greg Kroah-Hartman Cc: Herbert Xu Cc: David S. Miller Cc: Anton Vorontsov Cc: Colin Cross Cc: Kees Cook Cc: Tony Luck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

lib: update LZ4 compressor module

2017-02-25T01:46:57+00:00

Patch series "Update LZ4 compressor module", v7. This patchset updates the LZ4 compression module to a version based on LZ4 v1.7.3 allowing to use the fast compression algorithm aka LZ4 fast which provides an "acceleration" parameter as a tradeoff between high compression ratio and high compression speed. We want to use LZ4 fast in order to support compression in lustre and (mostly, based on that) investigate data reduction techniques in behalf of storage systems. Also, it will be useful for other users of LZ4 compression, as with LZ4 fast it is possible to enable applications to use fast and/or high compression depending on the usecase. For instance, ZRAM is offering a LZ4 backend and could benefit from an updated LZ4 in the kernel. LZ4 homepage: http://www.lz4.org/ LZ4 source repository: https://github.com/lz4/lz4 Source version: 1.7.3 Benchmark (taken from [1], Core i5-4300U @1.9GHz): ----------------|--------------|----------------|---------- Compressor | Compression | Decompression | Ratio ----------------|--------------|----------------|---------- memcpy | 4200 MB/s | 4200 MB/s | 1.000 LZ4 fast 50 | 1080 MB/s | 2650 MB/s | 1.375 LZ4 fast 17 | 680 MB/s | 2220 MB/s | 1.607 LZ4 fast 5 | 475 MB/s | 1920 MB/s | 1.886 LZ4 default | 385 MB/s | 1850 MB/s | 2.101 [1] http://fastcompression.blogspot.de/2015/04/sampling-or-faster-lz4.html [PATCH 1/5] lib: Update LZ4 compressor module [PATCH 2/5] lib/decompress_unlz4: Change module to work with new LZ4 module version [PATCH 3/5] crypto: Change LZ4 modules to work with new LZ4 module version [PATCH 4/5] fs/pstore: fs/squashfs: Change usage of LZ4 to work with new LZ4 version [PATCH 5/5] lib/lz4: Remove back-compat wrappers This patch (of 5): Update the LZ4 kernel module to LZ4 v1.7.3 by Yann Collet. The kernel module is inspired by the previous work by Chanho Min. The updated LZ4 module will not break existing code since the patchset contains appropriate changes. API changes: New method LZ4_compress_fast which differs from the variant available in kernel by the new acceleration parameter, allowing to trade compression ratio for more compression speed and vice versa. LZ4_decompress_fast is the respective decompression method, featuring a very fast decoder (multiple GB/s per core), able to reach RAM speed in multi-core systems. The decompressor allows to decompress data compressed with LZ4 fast as well as the LZ4 HC (high compression) algorithm. Also the useful functions LZ4_decompress_safe_partial and LZ4_compress_destsize were added. The latter reverses the logic by trying to compress as much data as possible from source to dest while the former aims to decompress partial blocks of data. A bunch of streaming functions were also added which allow compressig/decompressing data in multiple steps (so called "streaming mode"). The methods lz4_compress and lz4_decompress_unknownoutputsize are now known as LZ4_compress_default respectivley LZ4_decompress_safe. The old methods will be removed since there's no callers left in the code. [arnd@arndb.de: fix KERNEL_LZ4 support] Link: http://lkml.kernel.org/r/20170208211946.2839649-1-arnd@arndb.de [akpm@linux-foundation.org: simplify] [akpm@linux-foundation.org: fix the simplification] [4sschmid@informatik.uni-hamburg.de: fix performance regressions] Link: http://lkml.kernel.org/r/1486898178-17125-2-git-send-email-4sschmid@informatik.uni-hamburg.de [4sschmid@informatik.uni-hamburg.de: v8] Link: http://lkml.kernel.org/r/1487182598-15351-2-git-send-email-4sschmid@informatik.uni-hamburg.de Link: http://lkml.kernel.org/r/1486321748-19085-2-git-send-email-4sschmid@informatik.uni-hamburg.de Signed-off-by: Sven Schmidt <4sschmid@informatik.uni-hamburg.de> Signed-off-by: Arnd Bergmann Cc: Bongkyu Kim Cc: Rui Salvaterra Cc: Sergey Senozhatsky Cc: Greg Kroah-Hartman Cc: Herbert Xu Cc: David S. Miller Cc: Anton Vorontsov Cc: Colin Cross Cc: Kees Cook Cc: Tony Luck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

lib: lz4: cleanup unaligned access efficiency detection

2016-04-13T16:22:49+00:00

These identifiers are bogus. The interested architectures should define HAVE_EFFICIENT_UNALIGNED_ACCESS whenever relevant to do so. If this isn't true for some arch, it should be fixed in the arch definition. Signed-off-by: Rui Salvaterra Reviewed-by: Sergey Senozhatsky Signed-off-by: Greg Kroah-Hartman

lib: lz4: fixed zram with lz4 on big endian machines

2016-04-13T16:22:49+00:00

Based on Sergey's test patch [1], this fixes zram with lz4 compression on big endian cpus. Note that the 64-bit preprocessor test is not a cleanup, it's part of the fix, since those identifiers are bogus (for example, __ppc64__ isn't defined anywhere else in the kernel, which means we'd fall into the 32-bit definitions on ppc64). Tested on ppc64 with no regression on x86_64. [1] http://marc.info/?l=linux-kernel&m=145994470805853&w=4 Cc: stable@vger.kernel.org Suggested-by: Sergey Senozhatsky Signed-off-by: Rui Salvaterra Reviewed-by: Sergey Senozhatsky Signed-off-by: Greg Kroah-Hartman