Puts and reset an existing node in a slot, if one exists. Returns true
if a node was there, false if not. This helps cleanup some of the code
that does a lookup just to clear an existing node.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There are lots of spots open-coding this functionality, add a generic
helper that does the node lookup in a speculation safe way.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
For files, there's nr_user_files/file_table/file_data, and buffers have
nr_user_bufs/user_bufs/buf_data. There's no reason why file_table and
file_data can't be the same thing, and ditto for the buffer side. That
gets rid of more io_ring_ctx state that's in two spots rather than just
being in one spot, as it should be. Put all the registered file data in
one locations, and ditto on the buffer front.
This also avoids having both io_rsrc_data->nodes being an allocated
array, and ->user_bufs[] or ->file_table.nodes. There's no reason to
have this information duplicated. Keep it in one spot, io_rsrc_data,
along with how many resources are available.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
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 <axboe@kernel.dk>
If we're updating an existing slot, we clear the slot bitmap only to
set it again right after. Just leave the bit set rather than toggle
it off and on, and move the unused slot setting into the branch of
not already having a file occupy this slot.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Remove all the open coded magic on slot->file_ptr by introducing two
helpers that return the file pointer and the flags instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20230620113235.920399-9-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
struct io_rsrc_node carries a number of resources represented by struct
io_rsrc_put. That was handy before for sync overhead ammortisation, but
all complexity is gone and nodes are simple and lightweight. Let's
allocate a separate node for each resource.
Nodes and io_rsrc_put and not much different in size, and former are
cached, so node allocation should work better. That also removes some
overhead for nested iteration in io_rsrc_node_ref_zero() /
__io_rsrc_put_work().
Another reason for the patch is that it greatly reduces complexity
by moving io_rsrc_node_switch[_start]() inside io_queue_rsrc_removal(),
so users don't have to care about it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c7d3a45b30cc14cd93700a710dd112edc703db98.1681822823.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When fixed files are unregistered, file_alloc_end and alloc_hint
are not cleared. This can later cause a NULL pointer dereference in
io_file_bitmap_get() if auto index selection is enabled via
IORING_FILE_INDEX_ALLOC:
[ 6.519129] BUG: kernel NULL pointer dereference, address: 0000000000000000
[...]
[ 6.541468] RIP: 0010:_find_next_zero_bit+0x1a/0x70
[...]
[ 6.560906] Call Trace:
[ 6.561322] <TASK>
[ 6.561672] io_file_bitmap_get+0x38/0x60
[ 6.562281] io_fixed_fd_install+0x63/0xb0
[ 6.562851] ? __pfx_io_socket+0x10/0x10
[ 6.563396] io_socket+0x93/0xf0
[ 6.563855] ? __pfx_io_socket+0x10/0x10
[ 6.564411] io_issue_sqe+0x5b/0x3d0
[ 6.564914] io_submit_sqes+0x1de/0x650
[ 6.565452] __do_sys_io_uring_enter+0x4fc/0xb20
[ 6.566083] ? __do_sys_io_uring_register+0x11e/0xd80
[ 6.566779] do_syscall_64+0x3c/0x90
[ 6.567247] entry_SYSCALL_64_after_hwframe+0x72/0xdc
[...]
To fix the issue, set file alloc range and alloc_hint to zero after
file tables are freed.
Cc: stable@vger.kernel.org
Fixes: 4278a0deb1 ("io_uring: defer alloc_hint update to io_file_bitmap_set()")
Signed-off-by: Savino Dicanosa <sd7.dev@pm.me>
[axboe: add explicit bitmap == NULL check as well]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
There is an interesting reference bug when -ENOMEM occurs in calling of
io_install_fixed_file(). KASan report like below:
[ 14.057131] ==================================================================
[ 14.059161] BUG: KASAN: use-after-free in unix_get_socket+0x10/0x90
[ 14.060975] Read of size 8 at addr ffff88800b09cf20 by task kworker/u8:2/45
[ 14.062684]
[ 14.062768] CPU: 2 PID: 45 Comm: kworker/u8:2 Not tainted 6.1.0-rc4 #1
[ 14.063099] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
[ 14.063666] Workqueue: events_unbound io_ring_exit_work
[ 14.063936] Call Trace:
[ 14.064065] <TASK>
[ 14.064175] dump_stack_lvl+0x34/0x48
[ 14.064360] print_report+0x172/0x475
[ 14.064547] ? _raw_spin_lock_irq+0x83/0xe0
[ 14.064758] ? __virt_addr_valid+0xef/0x170
[ 14.064975] ? unix_get_socket+0x10/0x90
[ 14.065167] kasan_report+0xad/0x130
[ 14.065353] ? unix_get_socket+0x10/0x90
[ 14.065553] unix_get_socket+0x10/0x90
[ 14.065744] __io_sqe_files_unregister+0x87/0x1e0
[ 14.065989] ? io_rsrc_refs_drop+0x1c/0xd0
[ 14.066199] io_ring_exit_work+0x388/0x6a5
[ 14.066410] ? io_uring_try_cancel_requests+0x5bf/0x5bf
[ 14.066674] ? try_to_wake_up+0xdb/0x910
[ 14.066873] ? virt_to_head_page+0xbe/0xbe
[ 14.067080] ? __schedule+0x574/0xd20
[ 14.067273] ? read_word_at_a_time+0xe/0x20
[ 14.067492] ? strscpy+0xb5/0x190
[ 14.067665] process_one_work+0x423/0x710
[ 14.067879] worker_thread+0x2a2/0x6f0
[ 14.068073] ? process_one_work+0x710/0x710
[ 14.068284] kthread+0x163/0x1a0
[ 14.068454] ? kthread_complete_and_exit+0x20/0x20
[ 14.068697] ret_from_fork+0x22/0x30
[ 14.068886] </TASK>
[ 14.069000]
[ 14.069088] Allocated by task 289:
[ 14.069269] kasan_save_stack+0x1e/0x40
[ 14.069463] kasan_set_track+0x21/0x30
[ 14.069652] __kasan_slab_alloc+0x58/0x70
[ 14.069899] kmem_cache_alloc+0xc5/0x200
[ 14.070100] __alloc_file+0x20/0x160
[ 14.070283] alloc_empty_file+0x3b/0xc0
[ 14.070479] path_openat+0xc3/0x1770
[ 14.070689] do_filp_open+0x150/0x270
[ 14.070888] do_sys_openat2+0x113/0x270
[ 14.071081] __x64_sys_openat+0xc8/0x140
[ 14.071283] do_syscall_64+0x3b/0x90
[ 14.071466] entry_SYSCALL_64_after_hwframe+0x63/0xcd
[ 14.071791]
[ 14.071874] Freed by task 0:
[ 14.072027] kasan_save_stack+0x1e/0x40
[ 14.072224] kasan_set_track+0x21/0x30
[ 14.072415] kasan_save_free_info+0x2a/0x50
[ 14.072627] __kasan_slab_free+0x106/0x190
[ 14.072858] kmem_cache_free+0x98/0x340
[ 14.073075] rcu_core+0x427/0xe50
[ 14.073249] __do_softirq+0x110/0x3cd
[ 14.073440]
[ 14.073523] Last potentially related work creation:
[ 14.073801] kasan_save_stack+0x1e/0x40
[ 14.074017] __kasan_record_aux_stack+0x97/0xb0
[ 14.074264] call_rcu+0x41/0x550
[ 14.074436] task_work_run+0xf4/0x170
[ 14.074619] exit_to_user_mode_prepare+0x113/0x120
[ 14.074858] syscall_exit_to_user_mode+0x1d/0x40
[ 14.075092] do_syscall_64+0x48/0x90
[ 14.075272] entry_SYSCALL_64_after_hwframe+0x63/0xcd
[ 14.075529]
[ 14.075612] Second to last potentially related work creation:
[ 14.075900] kasan_save_stack+0x1e/0x40
[ 14.076098] __kasan_record_aux_stack+0x97/0xb0
[ 14.076325] task_work_add+0x72/0x1b0
[ 14.076512] fput+0x65/0xc0
[ 14.076657] filp_close+0x8e/0xa0
[ 14.076825] __x64_sys_close+0x15/0x50
[ 14.077019] do_syscall_64+0x3b/0x90
[ 14.077199] entry_SYSCALL_64_after_hwframe+0x63/0xcd
[ 14.077448]
[ 14.077530] The buggy address belongs to the object at ffff88800b09cf00
[ 14.077530] which belongs to the cache filp of size 232
[ 14.078105] The buggy address is located 32 bytes inside of
[ 14.078105] 232-byte region [ffff88800b09cf00, ffff88800b09cfe8)
[ 14.078685]
[ 14.078771] The buggy address belongs to the physical page:
[ 14.079046] page:000000001bd520e7 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88800b09de00 pfn:0xb09c
[ 14.079575] head:000000001bd520e7 order:1 compound_mapcount:0 compound_pincount:0
[ 14.079946] flags: 0x100000000010200(slab|head|node=0|zone=1)
[ 14.080244] raw: 0100000000010200 0000000000000000 dead000000000001 ffff88800493cc80
[ 14.080629] raw: ffff88800b09de00 0000000080190018 00000001ffffffff 0000000000000000
[ 14.081016] page dumped because: kasan: bad access detected
[ 14.081293]
[ 14.081376] Memory state around the buggy address:
[ 14.081618] ffff88800b09ce00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 14.081974] ffff88800b09ce80: 00 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc
[ 14.082336] >ffff88800b09cf00: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[ 14.082690] ^
[ 14.082909] ffff88800b09cf80: fb fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc
[ 14.083266] ffff88800b09d000: fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb
[ 14.083622] ==================================================================
The actual tracing of this bug is shown below:
commit 8c71fe7502 ("io_uring: ensure fput() called correspondingly
when direct install fails") adds an additional fput() in
io_fixed_fd_install() when io_file_bitmap_get() returns error values. In
that case, the routine will never make it to io_install_fixed_file() due
to an early return.
static int io_fixed_fd_install(...)
{
if (alloc_slot) {
...
ret = io_file_bitmap_get(ctx);
if (unlikely(ret < 0)) {
io_ring_submit_unlock(ctx, issue_flags);
fput(file);
return ret;
}
...
}
...
ret = io_install_fixed_file(req, file, issue_flags, file_slot);
...
}
In the above scenario, the reference is okay as io_fixed_fd_install()
ensures the fput() is called when something bad happens, either via
bitmap or via inner io_install_fixed_file().
However, the commit 61c1b44a21 ("io_uring: fix deadlock on iowq file
slot alloc") breaks the balance because it places fput() into the common
path for both io_file_bitmap_get() and io_install_fixed_file(). Since
io_install_fixed_file() handles the fput() itself, the reference
underflow come across then.
There are some extra commits make the current code into
io_fixed_fd_install() -> __io_fixed_fd_install() ->
io_install_fixed_file()
However, the fact that there is an extra fput() is called if
io_install_fixed_file() calls fput(). Traversing through the code, I
find that the existing two callers to __io_fixed_fd_install():
io_fixed_fd_install() and io_msg_send_fd() have fput() when handling
error return, this patch simply removes the fput() in
io_install_fixed_file() to fix the bug.
Fixes: 61c1b44a21 ("io_uring: fix deadlock on iowq file slot alloc")
Signed-off-by: Lin Ma <linma@zju.edu.cn>
Link: https://lore.kernel.org/r/be4ba4b.5d44.184a0a406a4.Coremail.linma@zju.edu.cn
Signed-off-by: Jens Axboe <axboe@kernel.dk>
From recently io_uring provides an option to allocate a file index for
operation registering fixed files. However, it's utterly unusable with
mixed approaches when for a part of files the userspace knows better
where to place it, as it may race and users don't have any sane way to
pick a slot and hoping it will not be taken.
Let the userspace to register a range of fixed file slots in which the
auto-allocation happens. The use case is splittting the fixed table in
two parts, where on of them is used for auto-allocation and another for
slot-specified operations.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/66ab0394e436f38437cf7c44676e1920d09687ad.1656154403.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Put it with the filetable code, which is where it belongs. While doing
so, have the helpers take a ctx rather than an io_kiocb. It doesn't make
sense to use a request, as it's not an operation on the request itself.
It applies to the ring itself.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
