RDMA/rxe: Reject unknown opcodes before ICRC processing
Subsystems (3)
PATCH v2 1/1 RDMA/rxe: Reject unknown opcodes before ICRC processing
Even after applying commit 7244491dab34 ("RDMA/rxe: Validate pad and ICRC before payload_size() in rxe_rcv"), a single unauthenticated UDP packet can still trigger panic. That patch handled payload_size() underflow only for valid opcodes with short packets, not for packets carrying an unknown opcode. The unknown-opcode OOB read described below predates that commit and reaches back to the initial Soft RoCE driver.
The check added there reads
pkt->paylen < header_size(pkt) + bth_pad(pkt) + RXE_ICRC_SIZE
where header_size(pkt) expands to rxe_opcode[pkt->opcode].length. The rxe_opcode[] array has 256 entries but is only populated for defined IB opcodes; any other entry (for example opcode 0xff) is zero-initialized, so length == 0 and the check degenerates to
pkt->paylen < 0 + bth_pad(pkt) + RXE_ICRC_SIZE
which does not constrain pkt->paylen enough. rxe_icrc_hdr() then computes
rxe_opcode[pkt->opcode].length - RXE_BTH_BYTES
which underflows when length == 0 and passes a huge value to rxe_crc32(), causing an out-of-bounds read of the skb payload.
Reproduced on v7.0-rc7 with that fix applied, QEMU/KVM with CONFIG_RDMA_RXE=y and CONFIG_KASAN=y, after
rdma link add rxe0 type rxe netdev eth0
A single 48-byte UDP packet to port 4791 with BTH opcode=0xff and QPN=IB_MULTICAST_QPN triggers:
BUG: KASAN: slab-out-of-bounds in crc32_le+0x115/0x170
Read of size 1 at addr ...
The buggy address is located 0 bytes to the right of
allocated 704-byte region
Call Trace:
crc32_le+0x115/0x170
rxe_icrc_hdr.isra.0+0x226/0x300
rxe_icrc_check+0x13f/0x3a0
rxe_rcv+0x6e1/0x16e0
rxe_udp_encap_recv+0x20a/0x320
udp_queue_rcv_one_skb+0x7ed/0x12c0Subsequent packets with the same shape fault on unmapped memory and panic the kernel. The trigger requires only module load and "rdma link add"; no QP, no connection, and no authentication.
Fix this by rejecting packets whose opcode has no rxe_opcode[] entry, detected via the zero mask or zero length, before any length arithmetic runs.
---
v2: also check rxe_opcode[].length per Zhu Yanjun; "||" rather than
"&&" so the guard tracks the actual underflow in rxe_icrc_hdr().
v1 was sent privately to [email protected].
drivers/infiniband/sw/rxe/rxe_recv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
drivers/infiniband/sw/rxe/rxe_recv.c +11 -0
@@ -330,6 +330,17 @@ void rxe_rcv(struct sk_buff *skb) pkt->qp =NULL ; pkt->mask |= rxe_opcode[pkt->opcode].mask; +/* +* Unknown opcodes have a zero-initialized rxe_opcode[] entry, so +* both mask and length are 0. Reject them before any length math: +* rxe_icrc_hdr() would otherwise compute length - RXE_BTH_BYTES +* and pass the underflowed value to rxe_crc32(), producing an +* out-of-bounds read. +*/ +if (unlikely (!rxe_opcode[pkt->opcode].mask || + !rxe_opcode[pkt->opcode].length)) + goto drop; +if (unlikely (pkt->paylen <header_size (pkt) +bth_pad (pkt) +RXE_ICRC_SIZE )) goto drop;
Signature · 1 line
2.53.0
Quoted 59 lines · Michael Bommarito
Even after applying commit
7244491dab34("RDMA/rxe: Validate pad and ICRC beforepayload_size()inrxe_rcv"), a single unauthenticated UDP packet can still trigger panic. That patch handledpayload_size()underflow only for valid opcodes with short packets, not for packets carrying an unknown opcode. The unknown-opcode OOB read described below predates that commit and reaches back to the initial Soft RoCE driver.The check added there reads
pkt->paylen < header_size(pkt) + bth_pad(pkt) + RXE_ICRC_SIZEwhere
header_size(pkt)expands to rxe_opcode[pkt->opcode].length. The rxe_opcode[] array has 256 entries but is only populated for defined IB opcodes; any other entry (for example opcode0xff) is zero-initialized, so length == 0 and the check degenerates topkt->paylen < 0 + bth_pad(pkt) + RXE_ICRC_SIZEwhich does not constrain
pkt->paylenenough.rxe_icrc_hdr()then computesrxe_opcode[pkt->opcode].length - RXE_BTH_BYTESwhich underflows when length == 0 and passes a huge value to
rxe_crc32(), causing an out-of-bounds read of the skb payload.Reproduced on v7.0-rc7 with that fix applied, QEMU/KVM with
CONFIG_RDMA_RXE=yandCONFIG_KASAN=y, afterrdma link add rxe0 type rxe netdev eth0
A single 48-byte UDP packet to port 4791 with BTH opcode=0xff and QPN=IB_MULTICAST_QPN triggers:
BUG: KASAN: slab-out-of-bounds in crc32_le+0x115/0x170 Read of size 1 at addr ... The buggy address is located 0 bytes to the right of allocated 704-byte region Call Trace: crc32_le+0x115/0x170 rxe_icrc_hdr.isra.0+0x226/0x300 rxe_icrc_check+0x13f/0x3a0 rxe_rcv+0x6e1/0x16e0 rxe_udp_encap_recv+0x20a/0x320 udp_queue_rcv_one_skb+0x7ed/0x12c0Subsequent packets with the same shape fault on unmapped memory and panic the kernel. The trigger requires only module load and "rdma link add"; no QP, no connection, and no authentication.
Fix this by rejecting packets whose opcode has no rxe_opcode[] entry, detected via the zero mask or zero length, before any length arithmetic runs.
Fixes 8700e3e7c485 ("Soft RoCE driver")Assisted-by Claude:claude-opus-4-6Signed-off-by Michael Bommarito <[email protected]>Reviewed-by Zhu Yanjun <[email protected]>
Quoted 8 lines
---
v2: also check rxe_opcode[].length per Zhu Yanjun; "||" rather than"&&" so the guard tracks the actual underflow in rxe_icrc_hdr().v1 was sent privately to [email protected].
drivers/infiniband/sw/rxe/rxe_recv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
Applied to rc thanks
Jason