Allow keeping hash table around on pool freeze, adjust callers

- Pool id -> string always works with a frozen pool, but in some cases
  we'll need to go the other way, allow caller to specify whether
  string -> id lookups should be possible on frozen pool.
- On glibc, realloc() to smaller size doesn't move the data but on
  other platforms (including valgrind) it can and does move, which
  would require a full rehash. For now, just leave all the data
  alone unless we're also freeing the hash, the memory savings
  isn't much for a global pool (which is where this matters)

5 files changed, 34 insertions, 23 deletions

diff --git a/lib/depends.c b/lib/depends.c
index 79d86d8db..b9d87d94e 100644
--- a/lib/depends.c
+++ b/lib/depends.c
@@ -330,7 +330,7 @@ rpmal rpmtsCreateAl(rpmts ts, rpmElementTypes types)
 	rpmstrPool tspool = rpmtsPool(ts);
 
 	/* Required for now to lock string pointers in memory */
-	rpmstrPoolFreeze(tspool);
+	rpmstrPoolFreeze(tspool, 1);
 
 	al = rpmalCreate(tspool, (rpmtsNElements(ts) / 4) + 1, rpmtsFlags(ts),
 				rpmtsColor(ts), rpmtsPrefColor(ts));
diff --git a/lib/rpmds.c b/lib/rpmds.c
index dcf2c39ab..3b7a836f8 100644
--- a/lib/rpmds.c
+++ b/lib/rpmds.c
@@ -212,7 +212,7 @@ rpmds rpmdsNewPool(rpmstrPool pool, Header h, rpmTagVal tagN, int flags)
 
 	/* freeze the pool to save memory, but only if private pool */
 	if (ds->pool != pool)
-	    rpmstrPoolFreeze(ds->pool);
+	    rpmstrPoolFreeze(ds->pool, 0);
     }
 
 exit:
@@ -310,7 +310,7 @@ static rpmds singleDS(rpmTagVal tagN, const char * N, const char * EVR,
 
     /* freeze the pool on success to save memory */
     if (ds != NULL)
-	rpmstrPoolFreeze(ds->pool);
+	rpmstrPoolFreeze(ds->pool, 0);
 
     /* free or unlink: ds now owns the reference */
     rpmstrPoolFree(pool);
@@ -989,7 +989,7 @@ int rpmdsRpmlib(rpmds * dsp, const void * tblp)
 	rpmdsFree(ds);
     }
     if (*dsp)
-	rpmstrPoolFreeze((*dsp)->pool);
+	rpmstrPoolFreeze((*dsp)->pool, 0);
     return rc;
 }
 
diff --git a/lib/rpmfi.c b/lib/rpmfi.c
index 6f61b70bb..826b0d2b1 100644
--- a/lib/rpmfi.c
+++ b/lib/rpmfi.c
@@ -1253,7 +1253,7 @@ rpmfi rpmfiNewPool(rpmstrPool pool, Header h, rpmTagVal tagN, rpmfiFlags flags)
 
 	    /* freeze the pool to save memory, but only if private pool */
 	    if (fi->pool != pool)
-		rpmstrPoolFreeze(fi->pool);
+		rpmstrPoolFreeze(fi->pool, 0);
 
 	    fi->h = (fi->fiflags & RPMFI_KEEPHEADER) ? headerLink(h) : NULL;
 	} else {
diff --git a/rpmio/rpmstrpool.c b/rpmio/rpmstrpool.c
index 38b676c03..5792471d8 100644
--- a/rpmio/rpmstrpool.c
+++ b/rpmio/rpmstrpool.c
@@ -25,6 +25,7 @@ struct rpmstrPool_s {
     size_t data_size;		/* string data area size */
     size_t data_alloced;	/* string data area allocation size */
     strHash hash;		/* string -> sid hash table */
+    int frozen;			/* are new id additions allowed? */
     int nrefs;			/* refcount */
 };
 
@@ -58,28 +59,38 @@ rpmstrPool rpmstrPoolLink(rpmstrPool pool)
     return pool;
 }
 
-void rpmstrPoolFreeze(rpmstrPool pool)
+void rpmstrPoolFreeze(rpmstrPool pool, int keephash)
 {
-    if (pool && pool->hash) {
-	pool->hash = strHashFree(pool->hash);
-	pool->data_alloced = pool->data_size;
-	pool->data = xrealloc(pool->data, pool->data_alloced);
-	pool->offs_alloced = pool->offs_size + 1;
-	pool->offs = xrealloc(pool->offs,
-			      pool->offs_alloced * sizeof(*pool->offs));
+    if (pool && !pool->frozen) {
+	/*
+	 * realloc() might require rehashing even when downsizing,
+	 * dont bother unless we're also discarding the hash.
+	 */
+	if (!keephash) {
+	    pool->hash = strHashFree(pool->hash);
+	    pool->data_alloced = pool->data_size;
+	    pool->data = xrealloc(pool->data, pool->data_alloced);
+	    pool->offs_alloced = pool->offs_size + 1;
+	    pool->offs = xrealloc(pool->offs,
+				  pool->offs_alloced * sizeof(*pool->offs));
+	}
+	pool->frozen = 1;
     }
 }
 
 void rpmstrPoolUnfreeze(rpmstrPool pool)
 {
-    if (pool && pool->hash == NULL) {
-	int sizehint = (pool->offs_size / 2) - 1;
-	if (sizehint < STRHASH_INITSIZE)
-	    sizehint = STRHASH_INITSIZE;
-	pool->hash = strHashCreate(sizehint, rstrhash, strcmp, NULL, NULL);
-	for (int i = 1; i < pool->offs_size; i++) {
-	    strHashAddEntry(pool->hash, rpmstrPoolStr(pool, i), i);
+    if (pool) {
+	if (pool->hash == NULL) {
+	    int sizehint = (pool->offs_size / 2) - 1;
+	    if (sizehint < STRHASH_INITSIZE)
+		sizehint = STRHASH_INITSIZE;
+	    pool->hash = strHashCreate(sizehint, rstrhash, strcmp, NULL, NULL);
+	    for (int i = 1; i < pool->offs_size; i++) {
+		strHashAddEntry(pool->hash, rpmstrPoolStr(pool, i), i);
+	    }
 	}
+	pool->frozen = 0;
     }
 }
 
@@ -130,7 +141,7 @@ rpmsid rpmstrPoolIdn(rpmstrPool pool, const char *s, size_t slen, int create)
 	rpmsid *sids;
 	if (strHashGetHEntry(pool->hash, s, hash, &sids, NULL, NULL)) {
 	    sid = sids[0];
-	} else if (create) {
+	} else if (create && !pool->frozen) {
 	    sid = rpmstrPoolPut(pool, s, slen, hash);
 	}
     }
diff --git a/rpmio/rpmstrpool.h b/rpmio/rpmstrpool.h
index 958a47033..67e162056 100644
--- a/rpmio/rpmstrpool.h
+++ b/rpmio/rpmstrpool.h
@@ -18,8 +18,8 @@ rpmstrPool rpmstrPoolFree(rpmstrPool sidpool);
 /* reference a string pool */
 rpmstrPool rpmstrPoolLink(rpmstrPool sidpool);
 
-/* freeze pool to free memory */
-void rpmstrPoolFreeze(rpmstrPool sidpool);
+/* freeze pool to free memory (keephash required for string -> id lookups) */
+void rpmstrPoolFreeze(rpmstrPool sidpool, int keephash);
 
 /* unfreeze pool (ie recreate hash table) */
 void rpmstrPoolUnfreeze(rpmstrPool sidpool);