generic-library/isa-l
1
0
Fork 0
mirror of https://github.com/intel/isa-l.git synced 2025-02-25 07:41:06 +01:00
Code Issues Projects Releases Wiki Activity
227 Commits 8 Branches 19 Tags
Commit Graph

151 Commits

Author SHA1 Message Date
Roy Oursler
47e914f98f igzip: Fix Windows prologue for avx 512 gen_map and set_long 
Change-Id: I8e326dc7fb67f30101d03dc364ffba25242e1f67
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2018-03-01 13:27:48 -07:00
Roy Oursler
aedf4f8cff igzip: create optimized set_long_icf_fg for avx2 
Change-Id: I027e73eaa908ca69a5c1af5a52b464f1963fd2da
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2018-03-01 13:27:48 -07:00
Roy Oursler
9acc3ed2ac igzip: Create AVX2 optimized version of level 3 
Change-Id: Icfdb67445ee5afff85441cfee23beb66bfe15d5e
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2018-03-01 13:27:38 -07:00
Jean-Yves VET
52bb322912 build: fix compilation with CPU not supporting SSE 
This patch makes the project compile and run (tests and
performance tests as well) with CPUs which are not
supporting SSE instructions.

Signed-off-by: Jean-Yves Vet <jyvet@ddn.com>
 2018-02-23 06:13:10 -05:00
Roy Oursler
148898c6ba igizp: Print error when test_compress_file fails in igzip_rand_test 
Change-Id: I5b03b006708778cb61edb3d456d22d08f1169309
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2018-01-04 09:13:59 -07:00
Greg Tucker
22ec5c7469 igzip: Fix igzip fuzz test for lev2-3 buf size 
Also remove unnecessary deps in igzip makefile.

Change-Id: I1ff79461df6d60bfc52c99b574c39098f1fe238a
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-12-19 11:01:46 -07:00
Greg Tucker
6b1c9a95c8 igzip: Fix syntax for win yasm for when avx512 is supported 
Currently can't test because yasm doesn't support avx512 but should
get the syntax correct for when it does.

Change-Id: I672b47b83b96861d8b9bfb0af02e726a1949aca0
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-12-19 10:33:53 -07:00
Roy Oursler
48119c5c87 igzip: Increase long_code_lookup struct size to fix buffer overflow 
Change-Id: I6546dcb7ffcd5895292d06fdc748c3cf279a4542
 2017-12-19 10:24:23 -07:00
Roy Oursler
e3fad7c45a igzip: Fix out buffer overflow in write_type0_header 
Fix a possible 1 byte overflow by creating a combined write_bits and flush.

Change-Id: I2d2455e9e32a820522ff1d89d016db72a82baed9
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
5d413c8b12 igzip: Use full 32 bits histogram elements in igizp_icf_body/finish 
This fix prevents possible histogram overflow in compression.

Change-Id: Ie5f25d1bace7f443f432678fcfbd9050ac65113f
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
8965584ae3 igzip: Implement large hash table method as level 2 
Move current level 2 compression to be level 3 and create a level 2
compression which is the same algorithm as level 1 but with a larger
hashtable.

Change-Id: I398a6017c2ff218dded24e229fb51c23ea69f314
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
7a12bcb2a8 igzip: Separate concept of level and compression method 
Change-Id: I82a5fbeb93adc77057893c643e044e311e4f393c
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
fe68f02dac igzip: Move hash_table and symbol histogram for icf compression 
Change-Id: I50df9c8915ff3e1af450aeb8e4c0cc3baf9624ae
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
2573f3cd87 igzip: Create defines for lvl1 hash table and histogram 
Change-Id: I4cdf7af4e482b8105aef024085323d0b1cd622ef
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
06cd70481c igzip: Separate defines for LVL1 and LVL0 hash tables 
Change-Id: I19bdec8d2d0c74083bc1695763c9630516995885
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
9203f96c2a igzip: Change inputs to deflate_hash functions 
Change-Id: I9ffff9aee82d5d1ece51853b4709f13ccd80c8ad
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
fe90c14bc3 igzip: Fix issues for large file inputs to igzip_rand_test 
Change-Id: Ic77834d771b2eebdfb7eeed782ee0d69e7905fa5
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
b4ebd1c6e7 igzip: Fix bug in igzip_rand_test where invalid flag is set 
Change-Id: I0d9418bcb48bd40e2fb8bb9105f38cffe77ba33d
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
4ae2d1be29 igzip: Implement optimized level 2 compression 
Change-Id: I8cf5bcd56f290d17205ac36dc2828c8acfc66947
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
3c62216aa1 igzip: Remove some uneccessary random data generation from igzip_rand_test 
Change-Id: I7c10c95549b9d825360d8f033f8dc3d8546d4d4c
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
4de751f35a igzip: Improve VERBOSE printing in igzip_rand_test 
Change-Id: Ib04361d0980920e81f901da3df234f9299669a98
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
864634a62b igzip: Modify rand_test to not call rand() within VERBOSE blocks 
Change-Id: Ice907a9eae72662a741d3d02a3926683e65c173d
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
a7d474a55c igzip: Fix rand_test_bug where invalid flush case can have a valid flush 
Change-Id: I6459432cb14b0c91f07b87a54c224f71ee03a05a
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
49156c3568 igzip: Implement block buffer 
Change-Id: I9d7942740557e4ffaf8e223e190f4bd4e4f47b1e
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
6049ce3ca7 igzip: Modify state to record total_in on deflate call 
Change-Id: I13e5878a227732545aee5a762bf5a9a75ce73f02
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Roy Oursler
59b9990a39 igzip: Remove file_start from zstate 
Change-Id: Ia4cb6dc86da54cc771f25a6d958bea730caa4801
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-12-15 14:27:14 -07:00
Greg Tucker
4f59eeda90 test: Add llvm fuzz testing 
Moved the afl fuzz test and added llvm fuzz tests including inflate
and round trip compress and inflate.  Currently only works with clang,
std makefile and libFuzzer installed.  Need to add checking and
support later when libfuzzer is more tightly integrated into the
compiler.

Change-Id: I2db9ad2335d6c5ed846886703b58225f67bcc935
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-11-17 17:28:27 -07:00
Roy Oursler
f7b6e73146 igzip: Improve igzip_rand_test compress_multi_pass by modifying in buffer 
Modify igzip_rand_test to test that data after next_in can be modified without
producing incorrect compressed data.

Change-Id: Ic6b62f269c6783407fbec7222ae73f24d3735717
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-10-20 15:29:39 -07:00
Greg Tucker
48d74d6e51 igzip: Fix perf test to only open dictionary if specified 
Change-Id: Idf5048589af5b61da5ccccf25c575dfc05ea15ec
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-10-11 15:57:41 -07:00
Roy Oursler
e79c57c7e3 igzip: Fix issue with isal_create_hufftables_subset 
isal_create_hufftables_subset failed to generate length symbols, but should
generate those symbols as a histogram does not guarantee finding all lengths
found in compression

Change-Id: I880210fe1b1078de8617cab0ecb93c9810b9c9de
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-09-21 17:24:56 -07:00
Greg Tucker
f78601adb9 igzip: Pick faster crc32 gzip function for sse4.2 machines 
For machines with sse4.2 but not avx we can choose a faster version of
crc32_gzip. For highly compressible files using gzip format (rfc 1952) this can
be a significant advantage.

Change-Id: Ib699a68999290eed4b83f9bbf54dc6c304a5c445
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-09-20 10:50:16 -07:00
Peng Xiao
9099918dcb fixing huff_codes.c's compiling problem on CentOS 6.5 as gcc 4.4.7 only support C89 
Signed-off-by: Peng Xiao <xiao.peng61@zte.com.cn>
 2017-08-25 11:07:23 +00:00
Roy Oursler
aff6555226 igzip: Optimized deflate_hash 
Optimize deflate hash by unrolling crc calculations.

Change-Id: Ief882910619a2cc3b052416d30499f6226e47419
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-08-18 14:42:37 -07:00
Roy Oursler
cf936f0d84 igzip: Remove igzip_sync_flush_perf 
Remove igzip_sync_flush perf because the perf test does not give any new
information, and the data used does is not a representative work load.

Change-Id: I7b68f8b7c6da0944ace5a2a9e31db378135689ff
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-08-18 14:39:14 -07:00
Roy Oursler
80bfbb33df igzip: Remove DECLARE_ALIGNED and optimize structure layout 
Change-Id: I95bc3b8e2e30aff0d596c743158337400c4eb486
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-08-18 14:35:44 -07:00
Greg Tucker
a7fad4b9d2 igzip: Change a few conflicting functions to static 
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-08-18 14:34:14 -07:00
Roy Oursler
48d7def5d9 igzip: Fix inflate total_out behavior to align with expected behavior 
Modify total_out to be total data decompressed into user supplied buffers,
rather than to user supplied and internal buffers.

Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-08-18 14:28:10 -07:00
Xiaodong Liu
3ab8239097 multibinary: move WRT_OPT macro to common header 
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
 2017-08-18 14:24:57 -07:00
Roy Oursler
34c341db35 igzip: Add reset functions for both deflate and inflate. 
Change-Id: I8677a4365ac5c2343751660176f3b2eb4746ddfe
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-26 04:14:44 -04:00
Xiaodong Liu
39ce870235 igzip: slver typo fix 
Change-Id: I13e6d150d0c661ee6dda9c25162c9ade5136d367
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
 2017-06-26 04:12:19 -04:00
Roy Oursler
ed15402f5b igzip: Add stateful dictionary support 
Change-Id: I75dbac947787bc0041674468c88d0aa41b8b082f
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-26 04:11:48 -04:00
Roy Oursler
1a7c640ef9 igzip: Fix 0 length file and looping errors in igzip_inflate_test 
Change-Id: I328f241ba07d8a0ae4fbc4c7de2ea8913912a188
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-26 04:11:03 -04:00
Greg Tucker
a8966b6709 igzip: Add unit tests for adler and crc32_gzip 
Also renamed test helper function, fixed clang warnings and adler usage.

Change-Id: I4ad22d046809483456608be1f4fdc4adbf0e09e4
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-06-26 04:03:35 -04:00
Greg Tucker
e1f5284ff8 igzip: Add sse optimized adler32 checksum 
Change-Id: Id07727b8a8da4b41aa983b487ca881552d5190ee
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-06-26 04:01:29 -04:00
Greg Tucker
3025e83b91 igzip: Add avx2 optimized adler32 checksum 
Change-Id: I019a38cf98836e3e6c7215a6914b85abb9399e33
Signed-off-by: Greg Tucker <greg.b.tucker@intel.com>
 2017-06-26 03:56:49 -04:00
Roy Oursler
f4a5b303e2 igzip: Remove BITBUF8 and BITBUF_ELSE compile options 
Change-Id: Iad3b2e6f9a32473b6e59910494c75d82558fc28e
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-21 06:03:51 -04:00
Roy Oursler
edacadc8fb igzip: Modify encode_df_04 to behave more like encode_df_06 algorithm 
Change-Id: I39c5d0d8182efb0fe8aa6bea97d9361df4ee8ddf
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-21 06:02:13 -04:00
Roy Oursler
5a55e3096c igzip: Avx512 version for encdode_df 
Change-Id: I1625a3d7e016805791cfd09e31909562f432fd71
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-21 05:56:05 -04:00
Roy Oursler
b3c09b9b7c igzip: Improve random data generation in igzip-rand_test 
Change-Id: I4835a9e376b4fa24080d765255703a959389487d
Signed-off-by: Roy Oursler <roy.j.oursler@intel.com>
 2017-06-21 05:47:04 -04:00
Roy Oursler
64143a741e igzip: Implement type 0 blocks for level 1 compress 
Change-Id: If55ab161623d29fa6fb08df3bc813e654918e592
 2017-06-21 05:46:38 -04:00