Avoid overflow in arena_run_regind().

Fix a regression due to:
    Remove an arena_bin_run_size_calc() constraint.
    2a6f2af6e4
The removed constraint required that small run headers fit in one page,
which indirectly limited runs such that they would not cause overflow in
arena_run_regind().  Add an explicit constraint to
arena_bin_run_size_calc() based on the largest number of regions that
arena_run_regind() can handle (2^11 as currently configured).

jemalloc/include/jemalloc/internal/arena.h

@@ -58,6 +58,10 @@
 #define	RUN_MAX_OVRHD		0x0000003dU
 #define	RUN_MAX_OVRHD_RELAX	0x00001800U
 
+/* Maximum number of regions in one run. */
+#define	LG_RUN_MAXREGS		11
+#define	RUN_MAXREGS		(1U << LG_RUN_MAXREGS)
+
 /*
  * The minimum ratio of active:dirty pages per arena is computed as:
  *
@@ -556,8 +560,8 @@ arena_run_regind(arena_run_t *run, arena_bin_info_t *bin_info, const void *ptr)
 		 * divide by 0, and 1 and 2 are both powers of two, which are
 		 * handled above.
 		 */
-#define	SIZE_INV_SHIFT 21
-#define	SIZE_INV(s) (((1U << SIZE_INV_SHIFT) / (s)) + 1)
+#define	SIZE_INV_SHIFT	((sizeof(unsigned) << 3) - LG_RUN_MAXREGS)
+#define	SIZE_INV(s)	(((1U << SIZE_INV_SHIFT) / (s)) + 1)
 		static const unsigned size_invs[] = {
 		    SIZE_INV(3),
 		    SIZE_INV(4), SIZE_INV(5), SIZE_INV(6), SIZE_INV(7),

jemalloc/include/jemalloc/internal/atomic.h

@@ -70,7 +70,9 @@ atomic_sub_uint64(uint64_t *p, uint64_t x)
 	return (OSAtomicAdd64(-((int64_t)x), (int64_t *)p));
 }
 #else
-#  error "Missing implementation for 64-bit atomic operations"
+#  if (LG_SIZEOF_PTR == 3)
+#    error "Missing implementation for 64-bit atomic operations"
+#  endif
 #endif
 
 /* 32-bit operations. */

jemalloc/include/jemalloc/internal/bitmap.h

@@ -2,7 +2,7 @@
 #ifdef JEMALLOC_H_TYPES
 
 /* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */
-#define	LG_BITMAP_MAXBITS	18
+#define	LG_BITMAP_MAXBITS	LG_RUN_MAXREGS
 
 typedef struct bitmap_level_s bitmap_level_t;
 typedef struct bitmap_info_s bitmap_info_t;

jemalloc/include/jemalloc/internal/jemalloc_internal.h.in

@@ -224,9 +224,9 @@ extern void	(*JEMALLOC_P(malloc_message))(void *wcbopaque, const char *s);
 #include "jemalloc/internal/ctl.h"
 #include "jemalloc/internal/mutex.h"
 #include "jemalloc/internal/mb.h"
-#include "jemalloc/internal/bitmap.h"
 #include "jemalloc/internal/extent.h"
 #include "jemalloc/internal/arena.h"
+#include "jemalloc/internal/bitmap.h"
 #include "jemalloc/internal/base.h"
 #include "jemalloc/internal/chunk.h"
 #include "jemalloc/internal/huge.h"

jemalloc/src/arena.c

@@ -2427,6 +2427,7 @@ small_size2bin_init_hard(void)
  *   *) bin_info->run_size >= min_run_size
  *   *) bin_info->run_size <= arena_maxclass
  *   *) run header overhead <= RUN_MAX_OVRHD (or header overhead relaxed).
+ *   *) bin_info->nregs <= RUN_MAXREGS
  *
  * bin_info->nregs, bin_info->bitmap_offset, and bin_info->reg0_offset are also
  * calculated here, since these settings are all interdependent.
@@ -2459,6 +2460,10 @@ bin_info_run_size_calc(arena_bin_info_t *bin_info, size_t min_run_size)
 	try_run_size = min_run_size;
 	try_nregs = ((try_run_size - sizeof(arena_run_t)) / bin_info->reg_size)
 	    + 1; /* Counter-act try_nregs-- in loop. */
+	if (try_nregs > RUN_MAXREGS) {
+		try_nregs = RUN_MAXREGS
+		    + 1; /* Counter-act try_nregs-- in loop. */
+	}
 	do {
 		try_nregs--;
 		try_hdr_size = sizeof(arena_run_t);
@@ -2500,6 +2505,10 @@ bin_info_run_size_calc(arena_bin_info_t *bin_info, size_t min_run_size)
 		try_nregs = ((try_run_size - sizeof(arena_run_t)) /
 		    bin_info->reg_size)
 		    + 1; /* Counter-act try_nregs-- in loop. */
+		if (try_nregs > RUN_MAXREGS) {
+			try_nregs = RUN_MAXREGS
+			    + 1; /* Counter-act try_nregs-- in loop. */
+		}
 		do {
 			try_nregs--;
 			try_hdr_size = sizeof(arena_run_t);
@@ -2526,7 +2535,8 @@ bin_info_run_size_calc(arena_bin_info_t *bin_info, size_t min_run_size)
 	} while (try_run_size <= arena_maxclass
 	    && try_run_size <= arena_maxclass
 	    && RUN_MAX_OVRHD * (bin_info->reg_size << 3) > RUN_MAX_OVRHD_RELAX
-	    && (try_reg0_offset << RUN_BFP) > RUN_MAX_OVRHD * try_run_size);
+	    && (try_reg0_offset << RUN_BFP) > RUN_MAX_OVRHD * try_run_size
+	    && try_nregs < RUN_MAXREGS);
 
 	assert(good_hdr_size <= good_reg0_offset);
 

jemalloc/test/bitmap.c

@@ -13,7 +13,11 @@
  */
 #include "../src/bitmap.c"
 
-#define	MAXBITS	4500
+#if (LG_BITMAP_MAXBITS > 12)
+#  define MAXBITS	4500
+#else
+#  define MAXBITS	(1U << LG_BITMAP_MAXBITS)
+#endif
 
 static void
 test_bitmap_size(void)