From 06fe9b1df1086b42718d632aa57e8f7cd1a66a21 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 29 May 2024 09:38:38 -0600 Subject: io_uring: don't attempt to mmap larger than what the user asks for MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If IORING_FEAT_SINGLE_MMAP is ignored, as can happen if an application uses an ancient liburing or does setup manually, then 3 mmap's are required to map the ring into userspace. The kernel will still have collapsed the mappings, however userspace may ask for mapping them individually. If so, then we should not use the full number of ring pages, as it may exceed the partial mapping. Doing so will yield an -EFAULT from vm_insert_pages(), as we pass in more pages than what the application asked for. Cap the number of pages to match what the application asked for, for the particular mapping operation. Reported-by: Lucas Mülling Link: https://github.com/axboe/liburing/issues/1157 Fixes: 3ab1db3c6039 ("io_uring: get rid of remap_pfn_range() for mapping rings/sqes") Signed-off-by: Jens Axboe --- io_uring/memmap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/io_uring/memmap.c b/io_uring/memmap.c index 4785d6af5fee..a0f32a255fd1 100644 --- a/io_uring/memmap.c +++ b/io_uring/memmap.c @@ -244,6 +244,7 @@ __cold int io_uring_mmap(struct file *file, struct vm_area_struct *vma) struct io_ring_ctx *ctx = file->private_data; size_t sz = vma->vm_end - vma->vm_start; long offset = vma->vm_pgoff << PAGE_SHIFT; + unsigned int npages; void *ptr; ptr = io_uring_validate_mmap_request(file, vma->vm_pgoff, sz); @@ -253,8 +254,8 @@ __cold int io_uring_mmap(struct file *file, struct vm_area_struct *vma) switch (offset & IORING_OFF_MMAP_MASK) { case IORING_OFF_SQ_RING: case IORING_OFF_CQ_RING: - return io_uring_mmap_pages(ctx, vma, ctx->ring_pages, - ctx->n_ring_pages); + npages = min(ctx->n_ring_pages, (sz + PAGE_SIZE - 1) >> PAGE_SHIFT); + return io_uring_mmap_pages(ctx, vma, ctx->ring_pages, npages); case IORING_OFF_SQES: return io_uring_mmap_pages(ctx, vma, ctx->sqe_pages, ctx->n_sqe_pages); -- cgit From e112311615a24e1618a591c73506571dc304eb8d Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 30 May 2024 07:23:39 -0700 Subject: io_uring/rw: Free iovec before cleaning async data kmemleak shows that there is a memory leak in io_uring read operation, where a buffer is allocated at iovec import, but never de-allocated. The memory is allocated at io_async_rw->free_iovec, but, then io_async_rw is kfreed, taking the allocated memory with it. I saw this happening when the read operation fails with -11 (EAGAIN). This is the kmemleak splat. unreferenced object 0xffff8881da591c00 (size 256): ... backtrace (crc 7a15bdee): [<00000000256f2de4>] __kmalloc+0x2d6/0x410 [<000000007a9f5fc7>] iovec_from_user.part.0+0xc6/0x160 [<00000000cecdf83a>] __import_iovec+0x50/0x220 [<00000000d1d586a2>] __io_import_iovec+0x13d/0x220 [<0000000054ee9bd2>] io_prep_rw+0x186/0x340 [<00000000a9c0372d>] io_prep_rwv+0x31/0x120 [<000000001d1170b9>] io_prep_readv+0xe/0x30 [<0000000070b8eb67>] io_submit_sqes+0x1bd/0x780 [<00000000812496d4>] __do_sys_io_uring_enter+0x3ed/0x5b0 [<0000000081499602>] do_syscall_64+0x5d/0x170 [<00000000de1c5a4d>] entry_SYSCALL_64_after_hwframe+0x76/0x7e This occurs because the async data cleanup functions are not set for read/write operations. As a result, the potentially allocated iovec in the rw async data is not freed before the async data is released, leading to a memory leak. With this following patch, kmemleak does not show the leaked memory anymore, and all liburing tests pass. Fixes: a9165b83c193 ("io_uring/rw: always setup io_async_rw for read/write requests") Signed-off-by: Breno Leitao Link: https://lore.kernel.org/r/20240530142340.1248216-1-leitao@debian.org Signed-off-by: Jens Axboe --- io_uring/opdef.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/io_uring/opdef.c b/io_uring/opdef.c index 2de5cca9504e..2e3b7b16effb 100644 --- a/io_uring/opdef.c +++ b/io_uring/opdef.c @@ -516,10 +516,12 @@ const struct io_cold_def io_cold_defs[] = { }, [IORING_OP_READ_FIXED] = { .name = "READ_FIXED", + .cleanup = io_readv_writev_cleanup, .fail = io_rw_fail, }, [IORING_OP_WRITE_FIXED] = { .name = "WRITE_FIXED", + .cleanup = io_readv_writev_cleanup, .fail = io_rw_fail, }, [IORING_OP_POLL_ADD] = { @@ -582,10 +584,12 @@ const struct io_cold_def io_cold_defs[] = { }, [IORING_OP_READ] = { .name = "READ", + .cleanup = io_readv_writev_cleanup, .fail = io_rw_fail, }, [IORING_OP_WRITE] = { .name = "WRITE", + .cleanup = io_readv_writev_cleanup, .fail = io_rw_fail, }, [IORING_OP_FADVISE] = { @@ -692,6 +696,7 @@ const struct io_cold_def io_cold_defs[] = { }, [IORING_OP_READ_MULTISHOT] = { .name = "READ_MULTISHOT", + .cleanup = io_readv_writev_cleanup, }, [IORING_OP_WAITID] = { .name = "WAITID", -- cgit From 18414a4a2eabb0281d12d374c92874327e0e3fe3 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 30 May 2024 13:35:50 -0600 Subject: io_uring/net: assign kmsg inq/flags before buffer selection syzbot reports that recv is using an uninitialized value: ===================================================== BUG: KMSAN: uninit-value in io_req_cqe_overflow io_uring/io_uring.c:810 [inline] BUG: KMSAN: uninit-value in io_req_complete_post io_uring/io_uring.c:937 [inline] BUG: KMSAN: uninit-value in io_issue_sqe+0x1f1b/0x22c0 io_uring/io_uring.c:1763 io_req_cqe_overflow io_uring/io_uring.c:810 [inline] io_req_complete_post io_uring/io_uring.c:937 [inline] io_issue_sqe+0x1f1b/0x22c0 io_uring/io_uring.c:1763 io_wq_submit_work+0xa17/0xeb0 io_uring/io_uring.c:1860 io_worker_handle_work+0xc04/0x2000 io_uring/io-wq.c:597 io_wq_worker+0x447/0x1410 io_uring/io-wq.c:651 ret_from_fork+0x6d/0x90 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Uninit was stored to memory at: io_req_set_res io_uring/io_uring.h:215 [inline] io_recv_finish+0xf10/0x1560 io_uring/net.c:861 io_recv+0x12ec/0x1ea0 io_uring/net.c:1175 io_issue_sqe+0x429/0x22c0 io_uring/io_uring.c:1751 io_wq_submit_work+0xa17/0xeb0 io_uring/io_uring.c:1860 io_worker_handle_work+0xc04/0x2000 io_uring/io-wq.c:597 io_wq_worker+0x447/0x1410 io_uring/io-wq.c:651 ret_from_fork+0x6d/0x90 arch/x86/kernel/process.c:147 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 Uninit was created at: slab_post_alloc_hook mm/slub.c:3877 [inline] slab_alloc_node mm/slub.c:3918 [inline] __do_kmalloc_node mm/slub.c:4038 [inline] __kmalloc+0x6e4/0x1060 mm/slub.c:4052 kmalloc include/linux/slab.h:632 [inline] io_alloc_async_data+0xc0/0x220 io_uring/io_uring.c:1662 io_msg_alloc_async io_uring/net.c:166 [inline] io_recvmsg_prep_setup io_uring/net.c:725 [inline] io_recvmsg_prep+0xbe8/0x1a20 io_uring/net.c:806 io_init_req io_uring/io_uring.c:2135 [inline] io_submit_sqe io_uring/io_uring.c:2182 [inline] io_submit_sqes+0x1135/0x2f10 io_uring/io_uring.c:2335 __do_sys_io_uring_enter io_uring/io_uring.c:3246 [inline] __se_sys_io_uring_enter+0x40f/0x3c80 io_uring/io_uring.c:3183 __x64_sys_io_uring_enter+0x11f/0x1a0 io_uring/io_uring.c:3183 x64_sys_call+0x2c0/0x3b50 arch/x86/include/generated/asm/syscalls_64.h:427 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcf/0x1e0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f which appears to be io_recv_finish() reading kmsg->msg.msg_inq to decide if it needs to set IORING_CQE_F_SOCK_NONEMPTY or not. If the recv is entered with buffer selection, but no buffer is available, then we jump error path which calls io_recv_finish() without having assigned kmsg->msg_inq. This might cause an errant setting of the NONEMPTY flag for a request get gets errored with -ENOBUFS. Reported-by: syzbot+b1647099e82b3b349fbf@syzkaller.appspotmail.com Fixes: 4a3223f7bfda ("io_uring/net: switch io_recv() to using io_async_msghdr") Signed-off-by: Jens Axboe --- io_uring/net.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/io_uring/net.c b/io_uring/net.c index 0a48596429d9..7c98c4d50946 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -1127,6 +1127,9 @@ int io_recv(struct io_kiocb *req, unsigned int issue_flags) flags |= MSG_DONTWAIT; retry_multishot: + kmsg->msg.msg_inq = -1; + kmsg->msg.msg_flags = 0; + if (io_do_buffer_select(req)) { ret = io_recv_buf_select(req, kmsg, &len, issue_flags); if (unlikely(ret)) @@ -1134,9 +1137,6 @@ retry_multishot: sr->buf = NULL; } - kmsg->msg.msg_inq = -1; - kmsg->msg.msg_flags = 0; - if (flags & MSG_WAITALL) min_ret = iov_iter_count(&kmsg->msg.msg_iter); -- cgit