[src/netbsd-7]: src/lib/libc/db/hash Pull up following revision(s) (requested...

[bouyer <bouyer%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/73fcf52f3c4f
branches:  netbsd-7
changeset: 799720:73fcf52f3c4f
user:      bouyer <bouyer%NetBSD.org@localhost>
date:      Sun Nov 22 14:15:14 2015 +0000

description:
Pull up following revision(s) (requested by christos in ticket #1046):
        lib/libc/db/hash/hash_page.c: revision 1.27
        lib/libc/db/hash/hash_page.c: revision 1.28
        lib/libc/db/hash/hash.h: revision 1.16
        lib/libc/db/hash/hash.c: revision 1.36
        lib/libc/db/hash/hash.c: revision 1.37
        lib/libc/db/hash/hash.c: revision 1.38
        lib/libc/db/hash/hash_bigkey.c: revision 1.25
Account for the -1 hack to fit 0x10000 in a short in hash_page.c
Introduce a HASH_BSIZE macro to return the blocksize; in the 64K case this
returns 0xffff to avoid overflow. This is used where sizes are stored.
If MAX_BSIZE == hashp->BSIZE (65536) then it does not fit in a short, and
we end up storing 0... This means that every entry needs a page. We store
MAX_BSIZE - 1 here, but it would be better to always store (avail - 1) here
so that we don't waste a byte and be consistent.
PR/50441: Manuel Bouyer: hash seq enumeration skips keys on big data.
XXX: pullup-7

diffstat:

 lib/libc/db/hash/hash.c        |  17 +++++++++--------
 lib/libc/db/hash/hash.h        |   7 ++++++-
 lib/libc/db/hash/hash_bigkey.c |  16 ++++++++--------
 lib/libc/db/hash/hash_page.c   |  24 ++++++++++++------------
 4 files changed, 35 insertions(+), 29 deletions(-)

diffs (242 lines):

diff -r bac643a67a3c -r 73fcf52f3c4f lib/libc/db/hash/hash.c
--- a/lib/libc/db/hash/hash.c   Sun Nov 22 14:02:31 2015 +0000
+++ b/lib/libc/db/hash/hash.c   Sun Nov 22 14:15:14 2015 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: hash.c,v 1.33.4.1 2015/08/06 21:50:36 snj Exp $        */
+/*     $NetBSD: hash.c,v 1.33.4.2 2015/11/22 14:15:14 bouyer Exp $     */
 
 /*-
  * Copyright (c) 1990, 1993, 1994
@@ -37,7 +37,7 @@
 #endif
 
 #include <sys/cdefs.h>
-__RCSID("$NetBSD: hash.c,v 1.33.4.1 2015/08/06 21:50:36 snj Exp $");
+__RCSID("$NetBSD: hash.c,v 1.33.4.2 2015/11/22 14:15:14 bouyer Exp $");
 
 #include "namespace.h"
 #include <sys/param.h>
@@ -585,7 +585,7 @@
        hash_accesses++;
 #endif
 
-       off = hashp->BSIZE;
+       off = HASH_BSIZE(hashp);
        size = key->size;
        kp = (char *)key->data;
        rbufp = __get_buf(hashp, __call_hash(hashp, kp, (int)size), NULL, 0);
@@ -617,7 +617,7 @@
                        bp = (uint16_t *)(void *)rbufp->page;
                        n = *bp++;
                        ndx = 1;
-                       off = hashp->BSIZE;
+                       off = HASH_BSIZE(hashp);
                } else if (bp[1] < REAL_KEY) {
                        if ((ndx =
                            __find_bigpair(hashp, rbufp, ndx, kp, (int)size)) > 0)
@@ -640,7 +640,7 @@
                                bp = (uint16_t *)(void *)rbufp->page;
                                n = *bp++;
                                ndx = 1;
-                               off = hashp->BSIZE;
+                               off = HASH_BSIZE(hashp);
                        } else {
                                save_bufp->flags &= ~BUF_PIN;
                                return (ERROR);
@@ -770,7 +770,7 @@
                                hashp->cndx = 1;
                        }
                } else {
-                       bp = (uint16_t *)(void *)hashp->cpage->page;
+                       bp = (uint16_t *)(void *)bufp->page;
                        if (flag == R_NEXT || flag == 0) {
                                if (hashp->cndx > bp[0]) {
                                        hashp->cpage = NULL;
@@ -802,15 +802,16 @@
        if (bp[ndx + 1] < REAL_KEY) {
                if (__big_keydata(hashp, bufp, key, data, 1))
                        return (ERROR);
+               hashp->cndx = 1;
        } else {
                if (hashp->cpage == NULL)
                        return (ERROR);
                key->data = (uint8_t *)hashp->cpage->page + bp[ndx];
-               key->size = (ndx > 1 ? bp[ndx - 1] : hashp->BSIZE) - bp[ndx];
+               key->size = (ndx > 1 ? bp[ndx - 1] : HASH_BSIZE(hashp)) - bp[ndx];
                data->data = (uint8_t *)hashp->cpage->page + bp[ndx + 1];
                data->size = bp[ndx] - bp[ndx + 1];
+               hashp->cndx += 2;
        }
-       hashp->cndx += 2;
        return (SUCCESS);
 }
 
diff -r bac643a67a3c -r 73fcf52f3c4f lib/libc/db/hash/hash.h
--- a/lib/libc/db/hash/hash.h   Sun Nov 22 14:02:31 2015 +0000
+++ b/lib/libc/db/hash/hash.h   Sun Nov 22 14:15:14 2015 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: hash.h,v 1.15 2008/08/26 21:18:38 joerg Exp $  */
+/*     $NetBSD: hash.h,v 1.15.40.1 2015/11/22 14:15:14 bouyer Exp $    */
 
 /*-
  * Copyright (c) 1990, 1993, 1994
@@ -123,6 +123,11 @@
  * Constants
  */
 #define        MAX_BSIZE               65536           /* 2^16 */
+/*
+ * Make it fit in uint16_t; a better way would be to store size - 1, but
+ * then we'd need to bump the version.
+ */
+#define HASH_BSIZE(hp) ((hp)->BSIZE == MAX_BSIZE ? MAX_BSIZE - 1 : (hp)->BSIZE)
 #define MIN_BUFFERS            6
 #define MINHDRSIZE             512
 #define DEF_BUFSIZE            65536           /* 64 K */
diff -r bac643a67a3c -r 73fcf52f3c4f lib/libc/db/hash/hash_bigkey.c
--- a/lib/libc/db/hash/hash_bigkey.c    Sun Nov 22 14:02:31 2015 +0000
+++ b/lib/libc/db/hash/hash_bigkey.c    Sun Nov 22 14:15:14 2015 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: hash_bigkey.c,v 1.24 2012/03/13 21:13:32 christos Exp $        */
+/*     $NetBSD: hash_bigkey.c,v 1.24.10.1 2015/11/22 14:15:14 bouyer Exp $     */
 
 /*-
  * Copyright (c) 1990, 1993, 1994
@@ -37,7 +37,7 @@
 #endif
 
 #include <sys/cdefs.h>
-__RCSID("$NetBSD: hash_bigkey.c,v 1.24 2012/03/13 21:13:32 christos Exp $");
+__RCSID("$NetBSD: hash_bigkey.c,v 1.24.10.1 2015/11/22 14:15:14 bouyer Exp $");
 
 /*
  * PACKAGE: hash
@@ -274,10 +274,10 @@
                bufp->ovfl = NULL;
        n -= 2;
        bp[0] = n;
-       temp = hashp->BSIZE - PAGE_META(n);
+       temp = HASH_BSIZE(hashp) - PAGE_META(n);
        _DBFIT(temp, uint16_t);
        FREESPACE(bp) = (uint16_t)temp;
-       OFFSET(bp) = hashp->BSIZE;
+       OFFSET(bp) = HASH_BSIZE(hashp);
 
        bufp->flags |= BUF_MOD;
        if (rbufp)
@@ -309,9 +309,9 @@
        ksize = size;
        kkey = key;
 
-       for (bytes = hashp->BSIZE - bp[ndx];
+       for (bytes = HASH_BSIZE(hashp) - bp[ndx];
            bytes <= size && bp[ndx + 1] == PARTIAL_KEY;
-           bytes = hashp->BSIZE - bp[ndx]) {
+           bytes = HASH_BSIZE(hashp) - bp[ndx]) {
                if (memcmp(p + bp[ndx], kkey, (size_t)bytes))
                        return (-2);
                kkey += bytes;
@@ -479,7 +479,7 @@
 
        p = bufp->page;
        bp = (uint16_t *)(void *)p;
-       mylen = hashp->BSIZE - bp[1];
+       mylen = HASH_BSIZE(hashp) - bp[1];
        save_addr = bufp->addr;
 
        if (bp[2] == FULL_KEY_DATA) {           /* End of Data */
@@ -546,7 +546,7 @@
 
        p = bufp->page;
        bp = (uint16_t *)(void *)p;
-       mylen = hashp->BSIZE - bp[1];
+       mylen = HASH_BSIZE(hashp) - bp[1];
 
        save_addr = bufp->addr;
        totlen = len + mylen;
diff -r bac643a67a3c -r 73fcf52f3c4f lib/libc/db/hash/hash_page.c
--- a/lib/libc/db/hash/hash_page.c      Sun Nov 22 14:02:31 2015 +0000
+++ b/lib/libc/db/hash/hash_page.c      Sun Nov 22 14:15:14 2015 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: hash_page.c,v 1.26 2013/12/01 00:22:48 christos Exp $  */
+/*     $NetBSD: hash_page.c,v 1.26.4.1 2015/11/22 14:15:14 bouyer Exp $        */
 
 /*-
  * Copyright (c) 1990, 1993, 1994
@@ -37,7 +37,7 @@
 #endif
 
 #include <sys/cdefs.h>
-__RCSID("$NetBSD: hash_page.c,v 1.26 2013/12/01 00:22:48 christos Exp $");
+__RCSID("$NetBSD: hash_page.c,v 1.26.4.1 2015/11/22 14:15:14 bouyer Exp $");
 
 /*
  * PACKAGE:  hashing
@@ -83,9 +83,9 @@
 #define        PAGE_INIT(P) { \
        ((uint16_t *)(void *)(P))[0] = 0; \
        temp = 3 * sizeof(uint16_t); \
-       _DIAGASSERT((size_t)hashp->BSIZE >= temp); \
-       ((uint16_t *)(void *)(P))[1] = (uint16_t)(hashp->BSIZE - temp); \
-       ((uint16_t *)(void *)(P))[2] = hashp->BSIZE; \
+       _DIAGASSERT((size_t)HASH_BSIZE(hashp) >= temp); \
+       ((uint16_t *)(void *)(P))[1] = (uint16_t)(HASH_BSIZE(hashp) - temp); \
+       ((uint16_t *)(void *)(P))[2] = HASH_BSIZE(hashp); \
 }
 
 /*
@@ -145,7 +145,7 @@
        if (ndx != 1)
                newoff = bp[ndx - 1];
        else
-               newoff = hashp->BSIZE;
+               newoff = HASH_BSIZE(hashp);
        pairlen = newoff - bp[ndx + 1];
 
        if (ndx != (n - 1)) {
@@ -194,8 +194,8 @@
        char *op;
        size_t temp;
 
-       copyto = (uint16_t)hashp->BSIZE;
-       off = (uint16_t)hashp->BSIZE;
+       copyto = HASH_BSIZE(hashp);
+       off = HASH_BSIZE(hashp);
        old_bufp = __get_buf(hashp, obucket, NULL, 0);
        if (old_bufp == NULL)
                return (-1);
@@ -346,7 +346,7 @@
 
                        ino = (uint16_t *)(void *)bufp->page;
                        n = 1;
-                       scopyto = hashp->BSIZE;
+                       scopyto = HASH_BSIZE(hashp);
                        moved = 0;
 
                        if (last_bfp)
@@ -354,7 +354,7 @@
                        last_bfp = bufp;
                }
                /* Move regular sized pairs of there are any */
-               off = hashp->BSIZE;
+               off = HASH_BSIZE(hashp);
                for (n = 1; (n < ino[0]) && (ino[n + 1] >= REAL_KEY); n += 2) {
                        cino = (char *)(void *)ino;
                        key.data = (uint8_t *)cino + ino[n];
@@ -541,7 +541,7 @@
        size_t temp;
 
        fd = hashp->fp;
-       size = hashp->BSIZE;
+       size = HASH_BSIZE(hashp);
 
        if ((fd == -1) || !is_disk) {
                PAGE_INIT(p);
@@ -594,7 +594,7 @@
        int fd, page, size;
        ssize_t wsize;
 
-       size = hashp->BSIZE;
+       size = HASH_BSIZE(hashp);
        if ((hashp->fp == -1) && (hashp->fp = __dbtemp("_hash", NULL)) == -1)
                return (-1);
        fd = hashp->fp;