On Mon, Sep 24, 2018 at 08:39:11PM +0200, Greg KH wrote:
On Mon, Sep 24, 2018 at 11:15:00AM -0700, Greg Hackmann wrote:
__close_fd() is reachable via the close() syscall with a userspace-controlled fd. Ensure that it can't be used to speculatively access past the end of current->fdt.
Reported-by: Omer Tripp trippo@google.com Cc: stable@vger.kernel.org Signed-off-by: Greg Hackmann ghackmann@google.com
fs/file.c | 2 ++ 1 file changed, 2 insertions(+)
diff --git a/fs/file.c b/fs/file.c index 7ffd6e9d103d..a80cf82be96b 100644 --- a/fs/file.c +++ b/fs/file.c @@ -18,6 +18,7 @@ #include <linux/bitops.h> #include <linux/spinlock.h> #include <linux/rcupdate.h> +#include <linux/nospec.h> unsigned int sysctl_nr_open __read_mostly = 1024*1024; unsigned int sysctl_nr_open_min = BITS_PER_LONG; @@ -626,6 +627,7 @@ int __close_fd(struct files_struct *files, unsigned fd) fdt = files_fdtable(files); if (fd >= fdt->max_fds) goto out_unlock;
- fd = array_index_nospec(fd, fdt->max_fds); file = fdt->fd[fd];
Don't you need 2 "halfs" of a gadget in order to make it work? This is one half, where is the second half?
Or am I reading this code wrong here somehow?
We don't want to play whack-a-mole with only 1/2 spectre gadgets, otherwise the 700+ patches that Red Hat added to their kernel would have been merged already.
Which reminds me, did the Red Hat tooling catch this one as well? If not, someone need to go fix it :)
Ping?