libsparse: fix 32 bit overflow when calculating last chunk
last_block * s->block_size can overflow when writing large filesystems,
cast to 64 bits before multiplying.
Change-Id: I3e54097852ce7d0fd271eab53d65e666284898e4
diff --git a/libsparse/sparse.c b/libsparse/sparse.c
index 189b4c0..741e8c6 100644
--- a/libsparse/sparse.c
+++ b/libsparse/sparse.c
@@ -139,7 +139,7 @@
DIV_ROUND_UP(backed_block_len(bb), s->block_size);
}
- pad = s->len - last_block * s->block_size;
+ pad = s->len - (int64_t)last_block * s->block_size;
assert(pad >= 0);
if (pad > 0) {
write_skip_chunk(out, pad);