kernel/linux.git/io_uring/notif.c, branch v6.18.21

Merge tag 'for-6.18/io_uring-20250929' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux

2025-10-02T16:56:23+00:00

Pull io_uring updates from Jens Axboe: - Store ring provided buffers locally for the users, rather than stuff them into struct io_kiocb. These types of buffers must always be fully consumed or recycled in the current context, and leaving them in struct io_kiocb is hence not a good ideas as that struct has a vastly different life time. Basically just an architecture cleanup that can help prevent issues with ring provided buffers in the future. - Support for mixed CQE sizes in the same ring. Before this change, a CQ ring either used the default 16b CQEs, or it was setup with 32b CQE using IORING_SETUP_CQE32. For use cases where a few 32b CQEs were needed, this caused everything else to use big CQEs. This is wasteful both in terms of memory usage, but also memory bandwidth for the posted CQEs. With IORING_SETUP_CQE_MIXED, applications may use request types that post both normal 16b and big 32b CQEs on the same ring. - Add helpers for async data management, to make it harder for opcode handlers to mess it up. - Add support for multishot for uring_cmd, which ublk can use. This helps improve efficiency, by providing a persistent request type that can trigger multiple CQEs. - Add initial support for ring feature querying. We had basic support for probe operations, but the API isn't great. Rather than expand that, add support for QUERY which is easily expandable and can cover a lot more cases than the existing probe support. This will help applications get a better idea of what operations are supported on a given host. - zcrx improvements from Pavel: - Improve refill entry alignment for better caching - Various cleanups, especially around deduplicating normal memory vs dmabuf setup. - Generalisation of the niov size (Patch 12). It's still hard coded to PAGE_SIZE on init, but will let the user to specify the rx buffer length on setup. - Syscall / synchronous bufer return. It'll be used as a slow fallback path for returning buffers when the refill queue is full. Useful for tolerating slight queue size misconfiguration or with inconsistent load. - Accounting more memory to cgroups. - Additional independent cleanups that will also be useful for mutli-area support. - Various fixes and cleanups * tag 'for-6.18/io_uring-20250929' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux: (68 commits) io_uring/cmd: drop unused res2 param from io_uring_cmd_done() io_uring: fix nvme's 32b cqes on mixed cq io_uring/query: cap number of queries io_uring/query: prevent infinite loops io_uring/zcrx: account niov arrays to cgroup io_uring/zcrx: allow synchronous buffer return io_uring/zcrx: introduce io_parse_rqe() io_uring/zcrx: don't adjust free cache space io_uring/zcrx: use guards for the refill lock io_uring/zcrx: reduce netmem scope in refill io_uring/zcrx: protect netdev with pp_lock io_uring/zcrx: rename dma lock io_uring/zcrx: make niov size variable io_uring/zcrx: set sgt for umem area io_uring/zcrx: remove dmabuf_offset io_uring/zcrx: deduplicate area mapping io_uring/zcrx: pass ifq to io_zcrx_alloc_fallback() io_uring/zcrx: check all niovs filled with dma addresses io_uring/zcrx: move area reg checks into io_import_area io_uring/zcrx: don't pass slot to io_zcrx_create_area ...

io_uring: fix incorrect io_kiocb reference in io_link_skb

2025-09-19T12:00:57+00:00

In io_link_skb function, there is a bug where prev_notif is incorrectly assigned using 'nd' instead of 'prev_nd'. This causes the context validation check to compare the current notification with itself instead of comparing it with the previous notification. Fix by using the correct prev_nd parameter when obtaining prev_notif. Signed-off-by: Yang Xiuwei Reviewed-by: Pavel Begunkov Fixes: 6fe4220912d19 ("io_uring/notif: implement notification stacking") Signed-off-by: Jens Axboe

io_uring/zctx: check chained notif contexts

2025-08-24T17:41:11+00:00

Send zc only links ubuf_info for requests coming from the same context. There are some ambiguous syz reports, so let's check the assumption on notification completion. Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/fd527d8638203fe0f1c5ff06ff2e1d8fd68f831b.1755179962.git.asml.silence@gmail.com Signed-off-by: Jens Axboe

io_uring: remove io_preinit_req()

2025-05-06T16:11:23+00:00

Apart from setting ->ctx, io_preinit_req() zeroes a bunch of fields of a request, from which only ->file_node is mandatory. Remove the function and zero the entire request on first allocation. With that, we also need to initialise ->ctx every time, which might be a good thing for performance as now we're likely overwriting the entire cache line, and so it can write combined and avoid RMW. Suggested-by: Jens Axboe Signed-off-by: Pavel Begunkov Link: https://lore.kernel.org/r/ba5485dc913f1e275862ce88f5169d4ac4a33836.1746533807.git.asml.silence@gmail.com Signed-off-by: Jens Axboe

io_uring: introduce type alias for io_tw_state

2025-02-17T12:34:50+00:00

In preparation for changing how io_tw_state is passed, introduce a type alias io_tw_token_t for struct io_tw_state *. This allows for changing the representation in one place, without having to update the many functions that just forward their struct io_tw_state * argument. Also add a comment to struct io_tw_state to explain its purpose. Signed-off-by: Caleb Sander Mateos Link: https://lore.kernel.org/r/20250217022511.1150145-1-csander@purestorage.com Signed-off-by: Jens Axboe

io_uring: move struct io_kiocb from task_struct to io_uring_task

2024-11-06T20:55:38+00:00

Rather than store the task_struct itself in struct io_kiocb, store the io_uring specific task_struct. The life times are the same in terms of io_uring, and this avoids doing some dereferences through the task_struct. For the hot path of putting local task references, we can deref req->tctx instead, which we'll need anyway in that function regardless of whether it's local or remote references. This is mostly straight forward, except the original task PF_EXITING check needs a bit of tweaking. task_work is _always_ run from the originating task, except in the fallback case, where it's run from a kernel thread. Replace the potentially racy (in case of fallback work) checks for req->task->flags with current->flags. It's either the still the original task, in which case PF_EXITING will be sane, or it has PF_KTHREAD set, in which case it's fallback work. Both cases should prevent moving forward with the given request. Signed-off-by: Jens Axboe

io_uring/rsrc: split io_kiocb node type assignments

2024-11-06T20:55:36+00:00

Currently the io_rsrc_node assignment in io_kiocb is an array of two pointers, as two nodes may be assigned to a request - one file node, and one buffer node. However, the buffer node can co-exist with the provided buffers, as currently it's not supported to use both provided and registered buffers at the same time. This crucially brings struct io_kiocb down to 4 cache lines again, as before it spilled into the 5th cacheline. Signed-off-by: Jens Axboe

io_uring/rsrc: get rid of the empty node and dummy_ubuf

2024-11-02T21:45:30+00:00

The empty node was used as a placeholder for a sparse entry, but it didn't really solve any issues. The caller still has to check for whether it's the empty node or not, it may as well just check for a NULL return instead. The dummy_ubuf was used for a sparse buffer entry, but NULL will serve the same purpose there of ensuring an -EFAULT on attempted import. Just use NULL for a sparse node, regardless of whether or not it's a file or buffer resource. Signed-off-by: Jens Axboe

io_uring/rsrc: add an empty io_rsrc_node for sparse buffer entries

2024-11-02T21:44:30+00:00

Rather than allocate an io_rsrc_node for an empty/sparse buffer entry, add a const entry that can be used for that. This just needs checking for writing the tag, and the put check needs to check for that sparse node rather than NULL for validity. This avoids allocating rsrc nodes for sparse buffer entries. Signed-off-by: Jens Axboe

io_uring/rsrc: get rid of per-ring io_rsrc_node list

2024-11-02T21:44:18+00:00

Work in progress, but get rid of the per-ring serialization of resource nodes, like registered buffers and files. Main issue here is that one node can otherwise hold up a bunch of other nodes from getting freed, which is especially a problem for file resource nodes and networked workloads where some descriptors may not see activity in a long time. As an example, instantiate an io_uring ring fd and create a sparse registered file table. Even 2 will do. Then create a socket and register it as fixed file 0, F0. The number of open files in the app is now 5, with 0/1/2 being the usual stdin/out/err, 3 being the ring fd, and 4 being the socket. Register this socket (eg "the listener") in slot 0 of the registered file table. Now add an operation on the socket that uses slot 0. Finally, loop N times, where each loop creates a new socket, registers said socket as a file, then unregisters the socket, and finally closes the socket. This is roughly similar to what a basic accept loop would look like. At the end of this loop, it's not unreasonable to expect that there would still be 5 open files. Each socket created and registered in the loop is also unregistered and closed. But since the listener socket registered first still has references to its resource node due to still being active, each subsequent socket unregistration is stuck behind it for reclaim. Hence 5 + N files are still open at that point, where N is awaiting the final put held up by the listener socket. Rewrite the io_rsrc_node handling to NOT rely on serialization. Struct io_kiocb now gets explicit resource nodes assigned, with each holding a reference to the parent node. A parent node is either of type FILE or BUFFER, which are the two types of nodes that exist. A request can have two nodes assigned, if it's using both registered files and buffers. Since request issue and task_work completion is both under the ring private lock, no atomics are needed to handle these references. It's a simple unlocked inc/dec. As before, the registered buffer or file table each hold a reference as well to the registered nodes. Final put of the node will remove the node and free the underlying resource, eg unmap the buffer or put the file. Outside of removing the stall in resource reclaim described above, it has the following advantages: 1) It's a lot simpler than the previous scheme, and easier to follow. No need to specific quiesce handling anymore. 2) There are no resource node allocations in the fast path, all of that happens at resource registration time. 3) The structs related to resource handling can all get simplified quite a bit, like io_rsrc_node and io_rsrc_data. io_rsrc_put can go away completely. 4) Handling of resource tags is much simpler, and doesn't require persistent storage as it can simply get assigned up front at registration time. Just copy them in one-by-one at registration time and assign to the resource node. The only real downside is that a request is now explicitly limited to pinning 2 resources, one file and one buffer, where before just assigning a resource node to a request would pin all of them. The upside is that it's easier to follow now, as an individual resource is explicitly referenced and assigned to the request. With this in place, the above mentioned example will be using exactly 5 files at the end of the loop, not N. Signed-off-by: Jens Axboe