| Message ID | 20181016145901.GA8811@embeddedor.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
| Series | RDMA/ucma: Fix Spectre v1 vulnerability | expand |
On Tue, 2018-10-16 at 16:59 +0200, Gustavo A. R. Silva wrote: > hdr.cmd can be indirectly controlled by user-space, hence leading to > a potential exploitation of the Spectre variant 1 vulnerability. > > This issue was detected with the help of Smatch: > > drivers/infiniband/core/ucma.c:1686 ucma_write() warn: potential > spectre issue 'ucma_cmd_table' [r] (local cap) > > Fix this by sanitizing hdr.cmd before using it to index > ucm_cmd_table. > > Notice that given that speculation windows are large, the policy is > to kill the speculation on the first load and not worry if it can be > completed with a dependent load/store [1]. > > [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2 > > Cc: stable@vger.kernel.org > Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Thanks, applied to for-rc.
On 10/16/18 6:49 PM, Doug Ledford wrote: >> >> Cc: stable@vger.kernel.org >> Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> > > Thanks, applied to for-rc. > Thanks, Doug. -- Gustavo
diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index 21863dd..01d68ed 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -44,6 +44,8 @@ #include <linux/module.h> #include <linux/nsproxy.h> +#include <linux/nospec.h> + #include <rdma/rdma_user_cm.h> #include <rdma/ib_marshall.h> #include <rdma/rdma_cm.h> @@ -1676,6 +1678,7 @@ static ssize_t ucma_write(struct file *filp, const char __user *buf, if (hdr.cmd >= ARRAY_SIZE(ucma_cmd_table)) return -EINVAL; + hdr.cmd = array_index_nospec(hdr.cmd, ARRAY_SIZE(ucma_cmd_table)); if (hdr.in + sizeof(hdr) > len) return -EINVAL;
hdr.cmd can be indirectly controlled by user-space, hence leading to a potential exploitation of the Spectre variant 1 vulnerability. This issue was detected with the help of Smatch: drivers/infiniband/core/ucma.c:1686 ucma_write() warn: potential spectre issue 'ucma_cmd_table' [r] (local cap) Fix this by sanitizing hdr.cmd before using it to index ucm_cmd_table. Notice that given that speculation windows are large, the policy is to kill the speculation on the first load and not worry if it can be completed with a dependent load/store [1]. [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2 Cc: stable@vger.kernel.org Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> --- drivers/infiniband/core/ucma.c | 3 +++ 1 file changed, 3 insertions(+)