io_uring/net: don't truncate the bundle segment length to int
Subsystems (2)
PATCH 1/1 io_uring/net: don't truncate the bundle segment length to int
io_bundle_nbufs() works out how many buffers a short bundle transfer consumed by walking the mapped iovec and subtracting each segment from the transferred length:
do {
int this_len = min_t(int, iov[nbufs].iov_len, ret);
nbufs++;
ret -= this_len;
} while (ret);iov_len is a size_t while this_len is an int, so a segment longer than INT_MAX comes out negative. ret then grows instead of shrinking and the loop keeps walking, reading past the end of the array it was handed.
The recv bundle path can hand it such a segment. io_recv_buf_select() folds sr->mshot_total_len into arg.max_len, and that field is unsigned and taken straight from sqe->optlen:
if (sr->flags & IORING_RECV_MSHOT_LIM) arg.max_len = min_not_zero(arg.max_len, sr->mshot_total_len);
With sqe->len left at 0 nothing else lowers it, so arg.max_len can be 0xffffffff. io_ring_buffers_peek() clamps each buffer against that rather than against INT_MAX and records an iov_len above INT_MAX; a short receive over that bundle then enters the loop above.
An unprivileged multishot IORING_OP_RECV with IOSQE_BUFFER_SELECT and IORING_RECVSEND_BUNDLE, a provided buffer ring registered with IOU_PBUF_RING_INC holding two buffers, and two bytes of data is enough to reach it. The two-entry iovec is a 32-byte kmalloc:
BUG: KASAN: slab-out-of-bounds in io_bundle_nbufs+0x114/0x1a8 Read of size 8 at addr ffff0000d39a1b68 by task io_repro/165 Call trace: io_bundle_nbufs+0x114/0x1a8 io_recv_finish+0x138/0xa38 io_recv+0x38c/0x1008 io_issue_sqe+0xc4/0x155c io_submit_sqes+0x6d0/0x1fac __arm64_sys_io_uring_enter+0x30c/0x1180 Allocated by task 165: __kmalloc_noprof+0x1b8/0x444 io_ring_buffers_peek+0x57c/0xbb4 io_buffers_peek+0x168/0x2a0 io_recv+0x598/0x1008
The buggy address is located 8 bytes to the right of allocated 32-byte region [ffff0000d39a1b40, ffff0000d39a1b60)
Count in size_t. ret is positive by the time the loop runs, since the function returns early for ret <= 0, so widening loses nothing; this_len stays bounded by ret, and the mapped length is never less than ret, so the walk ends inside the array.
Fixing it here rather than by capping arg.max_len keeps the accounting correct for every caller and does not disturb buffer selection: an arg.max_len of 0 is a distinct "not set yet" state that io_ring_buffers_peek() tests before it applies its own INT_MAX default, and pre-filling it changes which bundles get expanded.
Discovered by XBOW, triaged by Baul Lee <[email protected]>
---
The sr->len path into this loop is already rejected in io_{send,recv}msg_prep(); this one comes in through sqe->optlen instead, which is unsigned and so never trips a negative-length check.
io_uring/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
io_uring/net.c +1 -1
@@ -500,7 +500,7 @@ static int io_bundle_nbufs(struct io_async_msghdr *kmsg, int ret)/* short transfer, count segments */ nbufs =0 ; do { - int this_len = min_t(int, iov[nbufs].iov_len, ret); + size_t this_len = min_t(size_t, iov[nbufs].iov_len, ret); nbufs++; ret -= this_len;
Signature · 1 line
2.53.0
Quoted 62 lines · Baul Lee
io_bundle_nbufs()works out how many buffers a short bundle transfer consumed by walking the mapped iovec and subtracting each segment from the transferred length:do { int this_len = min_t(int, iov[nbufs].iov_len, ret); nbufs++; ret -= this_len; } while (ret);
iov_lenis asize_twhilethis_lenis an int, so a segment longer thanINT_MAXcomes out negative. ret then grows instead of shrinking and the loop keeps walking, reading past the end of the array it was handed.The recv bundle path can hand it such a segment.
io_recv_buf_select()foldssr->mshot_total_leninto arg.max_len, and that field is unsigned and taken straight fromsqe->optlen:if (sr->flags & IORING_RECV_MSHOT_LIM) arg.max_len = min_not_zero(arg.max_len, sr->mshot_total_len);With
sqe->lenleft at 0 nothing else lowers it, so arg.max_len can be0xffffffff.io_ring_buffers_peek()clamps each buffer against that rather than againstINT_MAXand records aniov_lenaboveINT_MAX; a short receive over that bundle then enters the loop above.An unprivileged multishot
IORING_OP_RECVwithIOSQE_BUFFER_SELECTandIORING_RECVSEND_BUNDLE, a provided buffer ring registered withIOU_PBUF_RING_INCholding two buffers, and two bytes of data is enough to reach it. The two-entry iovec is a 32-byte kmalloc:BUG: KASAN: slab-out-of-bounds in io_bundle_nbufs+0x114/0x1a8 Read of size 8 at addr ffff0000d39a1b68 by task io_repro/165 Call trace: io_bundle_nbufs+0x114/0x1a8 io_recv_finish+0x138/0xa38 io_recv+0x38c/0x1008 io_issue_sqe+0xc4/0x155c io_submit_sqes+0x6d0/0x1fac __arm64_sys_io_uring_enter+0x30c/0x1180 Allocated by task 165: __kmalloc_noprof+0x1b8/0x444 io_ring_buffers_peek+0x57c/0xbb4 io_buffers_peek+0x168/0x2a0 io_recv+0x598/0x1008The buggy address is located 8 bytes to the right of allocated 32-byte region [
ffff0000d39a1b40,ffff0000d39a1b60)Count in
size_t. ret is positive by the time the loop runs, since the function returns early for ret <= 0, so widening loses nothing;this_lenstays bounded by ret, and the mapped length is never less than ret, so the walk ends inside the array.Fixing it here rather than by capping arg.max_len keeps the accounting correct for every caller and does not disturb buffer selection: an arg.max_len of 0 is a distinct "not set yet" state that
io_ring_buffers_peek()tests before it applies its ownINT_MAXdefault, and pre-filling it changes which bundles get expanded.
Quoted 5 lines
Discovered by XBOW, triaged by Baul Lee <[email protected]>
Fixes a05d1f625c7a ("io_uring/net: support bundles for send")Signed-off-by Baul Lee <[email protected]>
Signature · 1 line
Gabriel Krisman Bertazi
Jens Axboe
maintainerQuoted 3 lines · Baul Lee
io_bundle_nbufs()works out how many buffers a short bundle transfer consumed by walking the mapped iovec and subtracting each segment from the transferred length:
Should already be fixed by:
Signature · 1 line
Jens Axboe