fs: fat: avoid multiplication overflow

The product of two 32 bit integers is a 32 bit integer. Hence
clustcount * bytesperclust may overflow on > 4 GiB devices.

Change the type of clustcount.

Fixes: cb8af8af5ba0 ("fs: fat: support write with non-zero offset")
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>

diff --git a/fs/fat/fat_write.c b/fs/fat/fat_write.c
index a6294419b8d66ad914668da57a39b353c85fde9f..8b5d669b00503d0306cf62eba45f23544e4f4d85 100644 (file)
--- a/fs/fat/fat_write.c
+++ b/fs/fat/fat_write.c
@@ -690,8 +690,8 @@ get_set_cluster(fsdata *mydata, __u32 clustnum, loff_t pos, __u8 *buffer,
        static u8 *tmpbuf_cluster;
        unsigned int bytesperclust = mydata->clust_size * mydata->sect_size;
        __u32 startsect;
-       loff_t wsize;
-       int clustcount, i, ret;
+       loff_t clustcount, wsize;
+       int i, ret;
 
        *gotsize = 0;
        if (!size)