| /freebsd/sys/contrib/zstd/ |
| H A D | CHANGELOG | 8 perf: faster huffman decoder, using x64 assembly, by @terrelln
9 perf: slightly faster high speed modes (strategies fast & dfast), by @felixhandte
10 perf: improved binary size and faster compilation times, by @terrelln
12 perf: faster mid-level compression speed in presence of highly repetitive patterns, by @senhuang42
15 perf: faster compression speed on incompressible data, by @bindhvo
112 perf: faster dictionary compression at medium compression levels, by @felixhandte
207 cli: faster `-t` test mode
237 perf: Slightly faster compression speed when re-using a context by @cyan4973 (#1658)
239 perf: Faster compression speed in high compression mode for repetitive data by @terrelln (#1635)
328 perf: much faster dictionary builder, by @jenniferliu
[all …]
|
| /freebsd/contrib/lua/doc/ |
| H A D | luac.1 | 20 faster loading,
24 Precompiling does not imply faster execution
29 The main goal in precompiling is faster loading.
|
| /freebsd/sys/contrib/libsodium/ |
| H A D | ChangeLog | 3 - Signatures computations and verifications are now way faster on
25 - Argon2 and scrypt are slightly faster on Linux.
123 - SHA-2 is about 20% faster, which also gives a speed boost to
161 - AVX2 detection was fixed, resulting in faster Blake2b hashing on
195 - Some implementations are now slightly faster when using the Clang
209 - Faster (2x) scalarmult_base() when using the ref10 implementation.
223 - Faster `sodium_increment()` for common nonce sizes.
231 Intel platforms. `crypto_generichash()` is now faster than MD5 and SHA1
454 - poly1305-ref has been replaced by a faster implementation,
|
| /freebsd/contrib/bearssl/samples/ |
| H A D | custom_profile.c | 196 * "i32". The "i31" code is somewhat faster than the "i32" code. in example_client_profile()
197 * Usually, "i31" is faster than "i15", except on some specific in example_client_profile()
200 * the "i31" is not, and the "i15" code is faster anyway). in example_client_profile()
203 * case of RSA, the "i31" code is faster, except on the small in example_client_profile()
204 * ARM Cortex M, where the "i15" code is faster and safer. in example_client_profile()
222 * For a given curve, "m15" is faster than "i15" (but possibly in example_client_profile()
223 * with a larger code footprint) and "m31" is faster than "i31" in example_client_profile()
285 * aes_ct64 is about twice faster than aes_ct for in example_client_profile()
414 * "i31" code for RSA (it is somewhat faster than the "i32" in example_client_profile()
|
| /freebsd/contrib/llvm-project/compiler-rt/lib/builtins/i386/ |
| H A D | floatundisf.S | 12 // actually several cycles faster.
15 // faster for the common case if you know something a priori about the input
52 // branch-free, x87-free implementation - faster at the expense of code size
|
| /freebsd/contrib/llvm-project/llvm/include/llvm/CodeGen/ |
| H A D | SelectionDAGTargetInfo.h | 91 /// faster than a libcall. The first returned SDValue is the result of the
103 /// faster than a libcall. The first returned SDValue is the result of the
114 /// where that is faster than a libcall.
128 /// faster than a libcall.
|
| /freebsd/contrib/ntp/sntp/libevent/ |
| H A D | whatsnew-2.1.txt | 524 2.2. Faster activations and timeouts
569 The EVBUFFER_EOL_CRLF line-ending type is now much faster, thanks to
583 has approximately the resolution corresponding to epoll, and is much faster
596 code run a little faster on change-heavy loads.
608 significantly faster and more accurate than our old ftime()-based approach.
638 3.6. Faster cross-thread notification on kqueue
646 thread from kqueue. This should be a tiny bit faster than the
651 4.1. Faster tests
653 I've spent some time to try to make the unit tests run faster in
663 Faster unit tests are great, since they let programmers test their
|
| /freebsd/contrib/libevent/ |
| H A D | whatsnew-2.1.txt | 524 2.2. Faster activations and timeouts
569 The EVBUFFER_EOL_CRLF line-ending type is now much faster, thanks to
583 has approximately the resolution corresponding to epoll, and is much faster
596 code run a little faster on change-heavy loads.
608 significantly faster and more accurate than our old ftime()-based approach.
638 3.6. Faster cross-thread notification on kqueue
646 thread from kqueue. This should be a tiny bit faster than the
651 4.1. Faster tests
653 I've spent some time to try to make the unit tests run faster in
663 Faster unit tests are great, since they let programmers test their
|
| /freebsd/sys/contrib/openzfs/man/man7/ |
| H A D | zpool-features.7 | 390 It is much faster than copying the data (as the actual data is neither read nor
483 hot spares that allow faster resilvering while retaining the benefits of RAID-Z.
509 .Pq over 350% faster than SHA-256 ,
727 This feature allows clones to be deleted faster than the traditional method
763 features significantly faster compression and decompression as well as a
769 compression is approximately 50% faster on compressible data and 200% faster
772 It is also approximately 80% faster on decompression,
913 but these systems cannot for whatever reason utilize the faster
924 .Pq 80% faster tha
[all...] |
| /freebsd/sbin/hastd/ |
| H A D | lzf.h | 105 * for small blocks (and 14 is usually a bit faster).
106 * For a low-memory/faster configuration, use HLOG == 13;
116 * (very roughly) 15% faster. This is the preferred mode of operation.
145 * You may choose to pre-set the hash table (might be faster on some
|
| /freebsd/crypto/openssl/crypto/bn/ |
| H A D | bn_ppc.c | 31 * but above subroutine is faster on contemporary processors. in bn_mul_mont()
33 * FPU code path would be faster, POWER6 perhaps, but there was in bn_mul_mont()
|
| /freebsd/contrib/llvm-project/llvm/lib/Target/X86/ |
| H A D | X86FixupInstTuning.cpp | 10 // with faster ones. We do this here, as opposed to during normal ISel, as
131 // `vshufpd` is always as fast or faster than `vpermilpd` and takes in processInstruction()
146 // `vshufps` is always as fast or faster than `vpermilps` and takes in processInstruction()
160 // `vpshufd` is always as fast or faster than `vpermilps` and takes 1 less in processInstruction()
180 // 1) If no bypass delay and `vunpck{l|h}qdq` faster than `vunpck{l|h}pd` in processInstruction()
182 // 2) If `vshufpd` faster than `vunpck{l|h}pd` in processInstruction()
198 // target we won't find it faster. in processInstruction()
|
| /freebsd/crypto/openssl/doc/man3/ |
| H A D | BF_encrypt.pod | 47 of the faster block ciphers. It is quite a bit faster than DES, and much
48 faster than IDEA or RC2.
|
| /freebsd/crypto/openssl/crypto/ec/ |
| H A D | ec_cvt.c | 40 * generally faster, sometimes "respectfully" faster, sometimes in EC_GROUP_new_curve_GFp()
|
| /freebsd/lib/libc/i386/string/ |
| H A D | swab.S | 37 * On the i486, this code is negligibly faster than the code generated
39 * should be considerably faster than the gcc code on an i386.
|
| /freebsd/contrib/llvm-project/llvm/include/llvm/Transforms/Utils/ |
| H A D | BypassSlowDivision.h | 10 // execute short instructions significantly faster than longer instructions.
65 /// profitably bypassed and carried out with a shorter, faster divide.
|
| /freebsd/contrib/mandoc/ |
| H A D | catman.8 | 80 overhead and uses the much faster
84 it may be about one order of magnitude faster than other
|
| /freebsd/secure/lib/libcrypto/man/man3/ |
| H A D | BF_encrypt.3 | 184 of the faster block ciphers. It is quite a bit faster than \s-1DES,\s0 and much
185 faster than \s-1IDEA\s0 or \s-1RC2.\s0
|
| /freebsd/contrib/bzip2/ |
| H A D | README | 68 effort to upgrade to use version 1.0, since it is both faster and more
147 * Approx 10% faster compression, 30% faster decompression
|
| /freebsd/sys/libkern/ |
| H A D | murmur3_32.c | 38 * faster if the caller guarantees that the input is correctly aligned for
39 * 32-bit reads, and slightly faster yet if the caller guarantees that the
|
| /freebsd/contrib/unbound/compat/ |
| H A D | getentropy_solaris.c | 256 int save_errno = errno, e, pgs = getpagesize(), faster = 0, repeat; in getentropy_fallback() local
272 faster = 1; in getentropy_fallback()
275 faster = 0; in getentropy_fallback()
299 if (!faster) { in getentropy_fallback()
|
| /freebsd/share/doc/smm/05.fastfs/ |
| H A D | 4.t | 137 Both reads and writes are faster in the new system than in the old system.
160 faster at writing files than reading them.
163 requests much faster than they can be serviced,
|
| /freebsd/sys/contrib/zstd/programs/ |
| H A D | zstd.1 | 105 …=#\fR is not present, it defaults to \fB1\fR\. The higher the value, the faster the compression sp…
135 …c\-friendly\. There is a negligible impact to compression ratio, and the faster compression levels…
156 …ter optimize compression parameters, resulting in better and potentially faster compression, espec…
271 …e in the range [6, 8], occasionally up to 16, but the algorithm will run faster with d <= \fI8\fR\…
367 There are 9 strategies numbered from 1 to 9, from faster to stronger: 1=ZSTD_fast, 2=ZSTD_dfast, 3=…
384 Bigger hash tables cause less collisions which usually makes compression faster, but requires more …
|
| /freebsd/contrib/ntp/libntp/ |
| H A D | ntp_random.c | 102 * The new version is somewhat faster than the original. As the
107 * 128 bytes, this new version runs about 19 percent faster and for a 16
108 * byte buffer it is about 5 percent faster.
278 /* seed the likely faster (and poorer) rand() as well */ in ntp_srandom()
316 * Array versions of the above information to make code run faster --
|
| /freebsd/contrib/bc/ |
| H A D | README.md | 225 math is *much* faster, so any non-trivial script should run faster in my `bc`.)
333 * This `bc` is faster. (See [Performance](#performance).)
338 `bc` is faster than GNU `bc` and can be *much* faster. Full benchmarks can be
|