[malloc] Avoid false positive warnings from valgrind

Calling discard_cache() is likely to result in a call to free_memblock(), which will call valgrind_make_blocks_noaccess() before returning. This causes valgrind to report an invalid read on the next iteration through the loop in alloc_memblock(). Fix by explicitly calling valgrind_make_blocks_defined() after discard_cache() returns. Also call valgrind_make_blocks_noaccess() before calling discard_cache(), to guard against free list corruption while executing cache discarders. Signed-off-by: Michael Brown <mcb30@ipxe.org>
ipxe · Sep 4, 2017 · 42eedb0 · 42eedb0
1 parent 75acb3c
commit 42eedb0
Showing 1 changed file with 8 additions and 1 deletion.
diff --git a/src/core/malloc.c b/src/core/malloc.c
@@ -284,6 +284,7 @@ void * alloc_memblock ( size_t size, size_t align, size_t offset ) {
 	size_t post_size;
 	struct memory_block *pre;
 	struct memory_block *post;
+	unsigned int discarded;
 	void *ptr;
 
 	/* Sanity checks */
@@ -371,7 +372,13 @@ void * alloc_memblock ( size_t size, size_t align, size_t offset ) {
 		}
 
 		/* Try discarding some cached data to free up memory */
-		if ( ! discard_cache() ) {
+		DBGC ( &heap, "Attempting discard for %#zx (aligned %#zx+%zx), "
+		       "used %zdkB\n", size, align, offset, ( usedmem >> 10 ) );
+		valgrind_make_blocks_noaccess();
+		discarded = discard_cache();
+		valgrind_make_blocks_defined();
+		check_blocks();
+		if ( ! discarded ) {
 			/* Nothing available to discard */
 			DBGC ( &heap, "Failed to allocate %#zx (aligned "
 			       "%#zx)\n", size, align );