gzip/performance-neoncrc32-and-prfm.patch

From 00f4c93e6c6cae92714a96cdde3c07bdfd59c9dc Mon Sep 17 00:00:00 2001
From: root <root@localhost.localdomain>
Date: Tue, 28 Dec 2021 10:24:26 +0800
Subject: [PATCH] performance-neoncrc32-and-prfm-2

---
 deflate.c | 28 +++++++++++++++++++++++++++-
 util.c    | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 75 insertions(+), 1 deletion(-)

diff --git a/deflate.c b/deflate.c
index 55bdc0e..3add924 100644
--- a/deflate.c
+++ b/deflate.c
@@ -378,6 +378,9 @@ longest_match(IPos cur_match)
     register int len;                           /* length of current match */
     int best_len = prev_length;                 /* best match length so far */
     IPos limit = strstart > (IPos)MAX_DIST ? strstart - (IPos)MAX_DIST : NIL;
+    #ifdef __aarch64__
+    IPos next_match;
+    #endif
     /* Stop when cur_match becomes <= limit. To simplify the code,
      * we prevent matches with the string of window index 0.
      */
@@ -412,6 +415,11 @@ longest_match(IPos cur_match)
         Assert(cur_match < strstart, "no future");
         match = window + cur_match;
 
+        #ifdef __aarch64__
+        next_match = prev[cur_match & WMASK];
+        __asm__("PRFM PLDL1STRM, [%0]"::"r"(&(prev[next_match & WMASK])));
+        #endif
+
         /* Skip to next match if the match length cannot increase
          * or if the match length is less than 2:
          */
@@ -488,8 +496,14 @@ longest_match(IPos cur_match)
             scan_end   = scan[best_len];
 #endif
         }
-    } while ((cur_match = prev[cur_match & WMASK]) > limit
+    } 
+    #ifdef __aarch64__
+    while ((cur_match = next_match) > limit
+             && --chain_length != 0);
+    #else
+    while ((cur_match = prev[cur_match & WMASK]) > limit
              && --chain_length != 0);
+    #endif
 
     return best_len;
 }
@@ -777,7 +791,19 @@ deflate (int pack_level)
             lookahead -= prev_length-1;
             prev_length -= 2;
             RSYNC_ROLL(strstart, prev_length+1);
+            while (prev_length >= 4) {
+                prev_length -= 4;
+                strstart++;
+                INSERT_STRING(strstart, hash_head);
+                strstart++;
+                INSERT_STRING(strstart, hash_head);
+                strstart++;
+                INSERT_STRING(strstart, hash_head);
+                strstart++;
+                INSERT_STRING(strstart, hash_head);
+            }
             do {
+                if (prev_length == 0) break;
                 strstart++;
                 INSERT_STRING(strstart, hash_head);
                 /* strstart never exceeds WSIZE-MAX_MATCH, so there are
diff --git a/util.c b/util.c
index e4240a7..397fb3e 100644
--- a/util.c
+++ b/util.c
@@ -31,6 +31,9 @@
 #include "gzip.h"
 #include <dirname.h>
 #include <xalloc.h>
+#ifdef __aarch64__
+#include <arm_acle.h>
+#endif
 
 #ifndef CHAR_BIT
 #  define CHAR_BIT 8
@@ -41,6 +44,7 @@ static int write_buffer (int, voidp, unsigned int);
 /* ========================================================================
  * Table of CRC-32's of all single-byte values (made by makecrc.c)
  */
+#ifndef __aarch64__
 static const ulg crc_32_tab[] = {
   0x00000000L, 0x77073096L, 0xee0e612cL, 0x990951baL, 0x076dc419L,
   0x706af48fL, 0xe963a535L, 0x9e6495a3L, 0x0edb8832L, 0x79dcb8a4L,
@@ -95,6 +99,7 @@ static const ulg crc_32_tab[] = {
   0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
   0x2d02ef8dL
 };
+#endif
 
 /* Shift register contents.  */
 static ulg crc = 0xffffffffL;
@@ -132,6 +137,48 @@ ulg updcrc(s, n)
     const uch *s;           /* pointer to bytes to pump through */
     unsigned n;             /* number of bytes in s[] */
 {
+    #ifdef __aarch64__
+    register ulg c;
+    register const uint8_t  *buf1;
+    register const uint16_t *buf2;
+    register const uint32_t *buf4;
+    register const uint64_t *buf8;
+    int64_t length = (int64_t)n;
+    buf8 = (const  uint64_t *)(const void *)s;
+
+    if (s == NULL) {
+        c = 0xffffffffL;
+    } else {
+       c = crc;
+	
+    	while(length >= sizeof(uint64_t)) {
+    		c = __crc32d(c, *buf8++);
+    		length -= sizeof(uint64_t);
+    	}
+
+    	buf4 = (const uint32_t *)(const void *)buf8;
+    	if (length >= sizeof(uint32_t)) {
+    		c = __crc32w(c, *buf4++);
+    		length -= sizeof(uint32_t);
+    	}
+
+    	buf2 = (const uint16_t *)(const void *)buf4;
+    	if(length >= sizeof(uint16_t)) {
+    		c = __crc32h(c, *buf2++);
+    		length -= sizeof(uint16_t);
+    	}
+
+    	buf1 = (const uint8_t *)(const void *)buf2;
+    	if (length >= sizeof(uint8_t)) {
+    		c = __crc32b(c, *buf1);
+    		length -= sizeof(uint8_t);
+    	}
+    }
+
+	crc = c;
+
+    return (c ^ 0xffffffffL);
+#else
     register ulg c;         /* temporary variable */
 
     if (s == NULL) {
@@ -144,6 +191,7 @@ ulg updcrc(s, n)
     }
     crc = c;
     return c ^ 0xffffffffL;       /* (instead of ~c for 64-bit machines) */
+#endif
 }
 
 /* Return a current CRC value.  */
-- 
2.27.0