public inbox for libc-stable@sourceware.org
 help / color / mirror / Atom feed
* [2.26 COMMITTED] malloc: Always call memcpy in _int_realloc [BZ #24027]
@ 2019-01-01  0:00 Florian Weimer
  0 siblings, 0 replies; only message in thread
From: Florian Weimer @ 2019-01-01  0:00 UTC (permalink / raw)
  To: libc-stable

From: Florian Weimer <fw@deneb.enyo.de>

This commit removes the custom memcpy implementation from _int_realloc
for small chunk sizes.  The ncopies variable has the wrong type, and
an integer wraparound could cause the existing code to copy too few
elements (leaving the new memory region mostly uninitialized).
Therefore, removing this code fixes bug 24027.

(cherry picked from commit b50dd3bc8cbb1efe85399b03d7e6c0310c2ead84)

2018-12-31  Florian Weimer  <fw@deneb.enyo.de>

	[BZ #24027]
	* malloc/malloc.c (_int_realloc): Always call memcpy for the
	copying operation.  (ncopies had the wrong type, resulting in an
	integer wraparound and too few elements being copied.)

diff --git a/NEWS b/NEWS
index f6d26425ff..ef857b6e6c 100644
--- a/NEWS
+++ b/NEWS
@@ -163,6 +163,7 @@ The following bugs are resolved with this release:
   [23579] libc: Errors misreported in preadv2
   [23709] Fix CPU string flags for Haswell-type CPUs
   [23927] Linux if_nametoindex() does not close descriptor (CVE-2018-19591)
+  [24027] malloc: Integer overflow in realloc
 \f
 Version 2.26
 
diff --git a/malloc/malloc.c b/malloc/malloc.c
index 862337466e..49e8ed69c2 100644
--- a/malloc/malloc.c
+++ b/malloc/malloc.c
@@ -4528,11 +4528,6 @@ _int_realloc(mstate av, mchunkptr oldp, INTERNAL_SIZE_T oldsize,
   mchunkptr        bck;             /* misc temp for linking */
   mchunkptr        fwd;             /* misc temp for linking */
 
-  unsigned long    copysize;        /* bytes to copy */
-  unsigned int     ncopies;         /* INTERNAL_SIZE_T words to copy */
-  INTERNAL_SIZE_T* s;               /* copy source */
-  INTERNAL_SIZE_T* d;               /* copy destination */
-
   /* oldmem size */
   if (__builtin_expect (chunksize_nomask (oldp) <= 2 * SIZE_SZ, 0)
       || __builtin_expect (oldsize >= av->system_mem, 0))
@@ -4600,43 +4595,7 @@ _int_realloc(mstate av, mchunkptr oldp, INTERNAL_SIZE_T oldsize,
             }
           else
             {
-              /*
-                 Unroll copy of <= 36 bytes (72 if 8byte sizes)
-                 We know that contents have an odd number of
-                 INTERNAL_SIZE_T-sized words; minimally 3.
-               */
-
-              copysize = oldsize - SIZE_SZ;
-              s = (INTERNAL_SIZE_T *) (chunk2mem (oldp));
-              d = (INTERNAL_SIZE_T *) (newmem);
-              ncopies = copysize / sizeof (INTERNAL_SIZE_T);
-              assert (ncopies >= 3);
-
-              if (ncopies > 9)
-                memcpy (d, s, copysize);
-
-              else
-                {
-                  *(d + 0) = *(s + 0);
-                  *(d + 1) = *(s + 1);
-                  *(d + 2) = *(s + 2);
-                  if (ncopies > 4)
-                    {
-                      *(d + 3) = *(s + 3);
-                      *(d + 4) = *(s + 4);
-                      if (ncopies > 6)
-                        {
-                          *(d + 5) = *(s + 5);
-                          *(d + 6) = *(s + 6);
-                          if (ncopies > 8)
-                            {
-                              *(d + 7) = *(s + 7);
-                              *(d + 8) = *(s + 8);
-                            }
-                        }
-                    }
-                }
-
+	      memcpy (newmem, chunk2mem (oldp), oldsize - SIZE_SZ);
               _int_free (av, oldp, 1);
               check_inuse_chunk (av, newp);
               return chunk2mem (newp);

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2019-01-02 13:01 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-01-01  0:00 [2.26 COMMITTED] malloc: Always call memcpy in _int_realloc [BZ #24027] Florian Weimer

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).