RDMA/rxe: Prevent access to wr->next ptr afrer wr is posted to send queue

[ Upstream commit 5f0b2a6093 ]

rxe_post_send_kernel() iterates over linked list of wr's, until the
wr->next ptr is NULL.  However if we've got an interrupt after last wr is
posted, control may be returned to the code after send completion callback
is executed and wr memory is freed.

As a result, wr->next pointer may contain incorrect value leading to
panic. Store the wr->next on the stack before posting it.

Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20200716190340.23453-1-m.malygin@yadro.com
Signed-off-by: Mikhail Malygin <m.malygin@yadro.com>
Signed-off-by: Sergey Kojushev <s.kojushev@yadro.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
This commit is contained in:
Mikhail Malygin 2020-07-16 22:03:41 +03:00 committed by Greg Kroah-Hartman
parent 4cf66d70b5
commit 7ecfbee3b9
1 changed files with 4 additions and 1 deletions

View File

@ -679,6 +679,7 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, const struct ib_send_wr *wr,
unsigned int mask;
unsigned int length = 0;
int i;
struct ib_send_wr *next;
while (wr) {
mask = wr_opcode_mask(wr->opcode, qp);
@ -695,6 +696,8 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, const struct ib_send_wr *wr,
break;
}
next = wr->next;
length = 0;
for (i = 0; i < wr->num_sge; i++)
length += wr->sg_list[i].length;
@ -705,7 +708,7 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, const struct ib_send_wr *wr,
*bad_wr = wr;
break;
}
wr = wr->next;
wr = next;
}
rxe_run_task(&qp->req.task, 1);