ddidderr/zstd-rs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #1756 from mgrice/dev

Browse Source 
Improvements in zstd decode performance
This commit is contained in:
Yann Collet
2019-09-18 11:35:50 -07:00
committed by GitHub
parent 5329de1f1e 5d89771529
commit bfff5b30a4
3 changed files with 38 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/common/zstd_internal.h
+31 -15
View File
@@ -209,7 +209,7 @@ typedef enum {
/*! ZSTD_wildcopy() :
* custom version of memcpy(), can overwrite up to WILDCOPY_OVERLENGTH bytes (if length==0) */
MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype)
void ZSTD_wildcopy(void* dst, const void* src, BYTE* oend_g, ptrdiff_t length, ZSTD_overlap_e ovtype)
{
ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
const BYTE* ip = (const BYTE*)src;
@@ -217,25 +217,33 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e
BYTE* const oend = op + length;
assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8));
if (length < VECLEN || (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN)) {
do
COPY8(op, ip)
while (op < oend);
}
else {
if ((length & 8) == 0)
COPY8(op, ip);
do {
COPY16(op, ip);
if (oend < oend_g-16) {
/* common case */
do {
COPY16(op, ip);
}
while (op < oend);
}
else {
do {
COPY8(op, ip);
}
while (op < oend);
}
while (op < oend);
}
}
/*! ZSTD_wildcopy_16min() :
* same semantics as ZSTD_wilcopy() except guaranteed to be able to copy 16 bytes at the start */
* same semantics as ZSTD_wildcopy() except guaranteed to be able to copy 16 bytes at the start */
MEM_STATIC FORCE_INLINE_ATTR DONT_VECTORIZE
void ZSTD_wildcopy_16min(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e ovtype)
void ZSTD_wildcopy_16min(void* dst, const void* src, BYTE* oend_g, ptrdiff_t length, ZSTD_overlap_e ovtype)
{
ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src;
const BYTE* ip = (const BYTE*)src;
@@ -246,17 +254,25 @@ void ZSTD_wildcopy_16min(void* dst, const void* src, ptrdiff_t length, ZSTD_over
assert(diff >= 8 || (ovtype == ZSTD_no_overlap && diff < -8));
if (ovtype == ZSTD_overlap_src_before_dst && diff < VECLEN) {
do
COPY8(op, ip)
do {
COPY8(op, ip);
}
while (op < oend);
}
else {
if ((length & 8) == 0)
COPY8(op, ip);
do {
COPY16(op, ip);
if (oend < oend_g-16) {
/* common case */
do {
COPY16(op, ip);
}
while (op < oend);
}
else {
do {
COPY8(op, ip);
}
while (op < oend);
}
while (op < oend);
}
}