…bvec()

Currently, io_buffer_register_bvec() takes in a request. In preparation
for supporting kernel-populated buffers in fuse io-uring (which will
need to register bvecs directly, not through a struct request), rename
this to io_buffer_register_request().

A subsequent patch will commandeer the "io_buffer_register_bvec()"
function name to support registering bvecs directly.

Rename io_buffer_unregister_bvec() to a more generic name,
io_buffer_unregister(), as both io_buffer_register_request() and
io_buffer_register_bvec() callers will use it for unregistration.

Signed-off-by: Joanne Koong <[email protected]>
Reviewed-by: Caleb Sander Mateos <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Split the main initialization logic in io_buffer_register_request() into
a helper function.

This is a preparatory patch for supporting kernel-populated buffers in
fuse io-uring, which will be reusing this logic.

Signed-off-by: Joanne Koong <[email protected]>
Reviewed-by: Caleb Sander Mateos <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Add io_buffer_register_bvec() for registering a bvec array.

This is a preparatory patch for fuse-over-io-uring zero-copy.

Signed-off-by: Joanne Koong <[email protected]>
Reviewed-by: Caleb Sander Mateos <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

Rename IO_IMU_DEST and IO_IMU_SOURCE to IO_BUF_DEST and IO_BUF_SOURCE
and export it so subsystems may use it.

This is needed by the io_buffer_register_bvec() path for callers who may
need the buffer to be both readable and writable.

Signed-off-by: Joanne Koong <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

* for-7.1/io_uring-fuse:
  io_uring/rsrc: rename and export IO_IMU_DEST / IO_IMU_SOURCE
  io_uring/rsrc: add io_buffer_register_bvec()
  io_uring/rsrc: split io_buffer_register_request() logic
  io_uring/rsrc: rename io_buffer_register_bvec()/io_buffer_unregister_bvec()

* block-7.1:
  zloop: remove irq-safe locking
  zloop: factor out zloop_mark_{full,empty} helpers
  zloop: set RQF_QUIET when completing requests on deleted devices
  zloop: improve the unaligned write pointer warning
  zloop: use vfs_truncate
  zloop: fix write pointer calculation in zloop_forget_cache

* io_uring-7.1:
  io_uring/poll: fix signed comparison in io_poll_get_ownership()

As with the idling code before it, the error exit path should check for
a NULL tctx->io_wq before calling io_wq_put_and_exit().

Fixes: 7880174 ("io_uring/tctx: clean up __io_uring_add_tctx_node() error handling")
Reported-by: Dan Carpenter <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>

* io_uring-7.1:
  io_uring/tctx: check for setup tctx->io_wq before teardown

syzbot reports that it's hitting the below condition for exiting an
io_wq context:

WARN_ON_ONCE(!test_bit(IO_WQ_BIT_EXIT, &wq->state))

in io_wq_put_and_exit(), which can be triggered with memory allocation
fault injection. Ensure that the io_wq is marked as exiting to silence
this warning trigger.

Reported-by: [email protected]
Fixes: 7880174 ("io_uring/tctx: clean up __io_uring_add_tctx_node() error handling")
Signed-off-by: Jens Axboe <[email protected]>

* io_uring-7.1:
  io_uring/tctx: mark io_wq as exiting before error path teardown

* block-7.1:
  block/blk-throttle: Add WQ_PERCPU to alloc_workqueue users
  block: Add WQ_PERCPU to alloc_workqueue users
  block: relax pgmap check in bio_add_page for compatible zone device pages
  block: add pgmap check to biovec_phys_mergeable
  floppy: fix reference leak on platform_device_register() failure
  ublk: use unchecked copy helpers for bio page data
  t10-pi: reduce ref tag code duplication