From: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Date: Fri, 9 Dec 2022 04:59:26 +0000 (+0900)
Subject: RDMA/rxe: Fix oops with zero length reads
X-Git-Url: http://git.maquefel.me/?a=commitdiff_plain;h=3282a549cf9b300e2d1b007925ed007ab24e4131;p=linux.git

RDMA/rxe: Fix oops with zero length reads

The commit 686d348476ee ("RDMA/rxe: Remove unnecessary mr testing") causes
a kernel crash. If responder get a zero-byte RDMA Read request,
qp->resp.mr is not set in check_rkey() (see IBA C9-88). The mr is NULL in
this case, and a NULL pointer dereference occurs as shown below.

 BUG: kernel NULL pointer dereference, address: 0000000000000010
 #PF: supervisor write access in kernel mode
 #PF: error_code(0x0002) - not-present page
 PGD 0 P4D 0
 Oops: 0002 [#1] PREEMPT SMP PTI
 CPU: 2 PID: 3622 Comm: python3 Kdump: loaded Not tainted 6.1.0-rc3+ #34
 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
 RIP: 0010:__rxe_put+0xc/0x60 [rdma_rxe]
 Code: cc cc cc 31 f6 e8 64 36 1b d3 41 b8 01 00 00 00 44 89 c0 c3 cc cc cc cc 41 89 c0 eb c1 90 0f 1f 44 00 00 41 54 b8 ff ff ff ff <f0> 0f c1 47 10 83 f8 01 74 11 45 31 e4 85 c0 7e 20 44 89 e0 41 5c
 RSP: 0018:ffffb27bc012ce78 EFLAGS: 00010246
 RAX: 00000000ffffffff RBX: ffff9790857b0580 RCX: 0000000000000000
 RDX: ffff979080fe145a RSI: 000055560e3e0000 RDI: 0000000000000000
 RBP: ffff97909c7dd800 R08: 0000000000000001 R09: e7ce43d97f7bed0f
 R10: ffff97908b29c300 R11: 0000000000000000 R12: 0000000000000000
 R13: 0000000000000000 R14: ffff97908b29c300 R15: 0000000000000000
 FS:  00007f276f7bd740(0000) GS:ffff9792b5c80000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 0000000000000010 CR3: 0000000114230002 CR4: 0000000000060ee0
 Call Trace:
  <IRQ>
  read_reply+0xda/0x310 [rdma_rxe]
  rxe_responder+0x82d/0xe50 [rdma_rxe]
  do_task+0x84/0x170 [rdma_rxe]
  tasklet_action_common.constprop.0+0xa7/0x120
  __do_softirq+0xcb/0x2ac
  do_softirq+0x63/0x90
  </IRQ>

Support a NULL mr during read_reply()

Fixes: 686d348476ee ("RDMA/rxe: Remove unnecessary mr testing")
Fixes: b5f9a01fae42 ("RDMA/rxe: Fix mr leak in RESPST_ERR_RNR")
Link: https://lore.kernel.org/r/20221209045926.531689-1-matsuda-daisuke@fujitsu.com
Link: https://lore.kernel.org/r/20221202145713.13152-1-lizhijian@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
---

diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c
index 12eb85e8d4154..0c8bec2819376 100644
--- a/drivers/infiniband/sw/rxe/rxe_resp.c
+++ b/drivers/infiniband/sw/rxe/rxe_resp.c
@@ -460,7 +460,7 @@ static enum resp_states check_rkey(struct rxe_qp *qp,
 		return RESPST_EXECUTE;
 	}
 
-	/* A zero-byte op is not required to set an addr or rkey. */
+	/* A zero-byte op is not required to set an addr or rkey. See C9-88 */
 	if ((pkt->mask & RXE_READ_OR_WRITE_MASK) &&
 	    (pkt->mask & RXE_RETH_MASK) &&
 	    reth_len(pkt) == 0) {
@@ -880,13 +880,15 @@ static enum resp_states read_reply(struct rxe_qp *qp,
 	skb = prepare_ack_packet(qp, &ack_pkt, opcode, payload,
 				 res->cur_psn, AETH_ACK_UNLIMITED);
 	if (!skb) {
-		rxe_put(mr);
+		if (mr)
+			rxe_put(mr);
 		return RESPST_ERR_RNR;
 	}
 
 	err = rxe_mr_copy(mr, res->read.va, payload_addr(&ack_pkt),
 			  payload, RXE_FROM_MR_OBJ);
-	rxe_put(mr);
+	if (mr)
+		rxe_put(mr);
 	if (err) {
 		kfree_skb(skb);
 		return RESPST_ERR_RKEY_VIOLATION;