Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <E5BAF7C1-74B1-4C71-9B01-53609D3E5971@gmail.com>
Date: Mon, 9 Jan 2017 09:47:35 +0100
From: Julien Ramseier <j.ramseier@...il.com>
To: musl@...ts.openwall.com
Subject: [PATCH] malloc: always fail with ENOMEM

malloc may set errno to something else than ENOMEM indirectly
through mmap, though ENOMEM is the only error allowed by POSIX.

There are cases where mmap will return EPERM instead of ENOMEM,
as highlighted by libc-test[1]. This can happen when mmap tries to
map pages near `mmap_min_addr` [2][3], as a security measure.

[1] http://www.openwall.com/lists/musl/2016/03/30/9
[2] https://ghc.haskell.org/trac/ghc/ticket/7500
[3] https://github.com/torvalds/linux/blob/master/security/min_addr.c

diff --git a/src/malloc/expand_heap.c b/src/malloc/expand_heap.c
index d8c0be7..4051b1b 100644
--- a/src/malloc/expand_heap.c
+++ b/src/malloc/expand_heap.c
@@ -65,7 +65,10 @@ void *__expand_heap(size_t *pn)
 	if (n < min) n = min;
 	void *area = __mmap(0, n, PROT_READ|PROT_WRITE,
 		MAP_PRIVATE|MAP_ANONYMOUS, -1, 0);
-	if (area == MAP_FAILED) return 0;
+	if (area == MAP_FAILED) {
+		errno = ENOMEM;
+		return 0;
+	}
 	*pn = n;
 	mmap_step++;
 	return area;
diff --git a/src/malloc/malloc.c b/src/malloc/malloc.c
index c38c46f..593c4dd 100644
--- a/src/malloc/malloc.c
+++ b/src/malloc/malloc.c
@@ -328,7 +328,10 @@ void *malloc(size_t n)
 		size_t len = n + OVERHEAD + PAGE_SIZE - 1 & -PAGE_SIZE;
 		char *base = __mmap(0, len, PROT_READ|PROT_WRITE,
 			MAP_PRIVATE|MAP_ANONYMOUS, -1, 0);
-		if (base == (void *)-1) return 0;
+		if (base == MAP_FAILED) {
+			errno = ENOMEM;
+			return 0;
+		}
 		c = (void *)(base + SIZE_ALIGN - OVERHEAD);
 		c->csize = len - (SIZE_ALIGN - OVERHEAD);
 		c->psize = SIZE_ALIGN - OVERHEAD;

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.