android_kernel_msm-6.1_noth.../include
Ilya Dryomov 75dbb685f4 libceph: fix potential use-after-free on linger ping and resends
request_reinit() is not only ugly as the comment rightfully suggests,
but also unsafe.  Even though it is called with osdc->lock held for
write in all cases, resetting the OSD request refcount can still race
with handle_reply() and result in use-after-free.  Taking linger ping
as an example:

    handle_timeout thread                     handle_reply thread

                                              down_read(&osdc->lock)
                                              req = lookup_request(...)
                                              ...
                                              finish_request(req)  # unregisters
                                              up_read(&osdc->lock)
                                              __complete_request(req)
                                                linger_ping_cb(req)

      # req->r_kref == 2 because handle_reply still holds its ref

    down_write(&osdc->lock)
    send_linger_ping(lreq)
      req = lreq->ping_req  # same req
      # cancel_linger_request is NOT
      # called - handle_reply already
      # unregistered
      request_reinit(req)
        WARN_ON(req->r_kref != 1)  # fires
        request_init(req)
          kref_init(req->r_kref)

                   # req->r_kref == 1 after kref_init

                                              ceph_osdc_put_request(req)
                                                kref_put(req->r_kref)

            # req->r_kref == 0 after kref_put, req is freed

        <further req initialization/use> !!!

This happens because send_linger_ping() always (re)uses the same OSD
request for watch ping requests, relying on cancel_linger_request() to
unregister it from the OSD client and rip its messages out from the
messenger.  send_linger() does the same for watch/notify registration
and watch reconnect requests.  Unfortunately cancel_request() doesn't
guarantee that after it returns the OSD client would be completely done
with the OSD request -- a ref could still be held and the callback (if
specified) could still be invoked too.

The original motivation for request_reinit() was inability to deal with
allocation failures in send_linger() and send_linger_ping().  Switching
to using osdc->req_mempool (currently only used by CephFS) respects that
and allows us to get rid of request_reinit().

Cc: stable@vger.kernel.org
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Xiubo Li <xiubli@redhat.com>
Acked-by: Jeff Layton <jlayton@kernel.org>
2022-05-18 21:21:05 +02:00
..
acpi ACPI: bus: Eliminate acpi_bus_get_device() 2022-04-05 19:49:26 +02:00
asm-generic bug: Have __warn() prototype defined unconditionally 2022-04-26 10:59:57 +02:00
clocksource
crypto Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 2022-03-21 16:02:36 -07:00
drm
dt-bindings dt-bindings: clk: mpfs: add defines for two new clocks 2022-04-22 18:40:15 -07:00
keys
kunit
kvm
linux libceph: fix potential use-after-free on linger ping and resends 2022-05-18 21:21:05 +02:00
math-emu
media
memory memory: renesas-rpc-if: Fix HF/OSPI data transfer in Manual Mode 2022-04-21 17:00:24 +02:00
misc
net Bluetooth: Fix the creation of hdev->name 2022-05-11 17:18:42 -07:00
pcmcia
ras mm/memory-failure.c: fix race with changing page compound again 2022-03-22 15:57:07 -07:00
rdma
scsi scsi: iscsi: Fix NOP handling during conn recovery 2022-04-11 22:09:35 -04:00
soc net: mscc: ocelot: mark traps with a bool instead of keeping them in a list 2022-05-05 19:15:14 -07:00
sound ASoC: Fixes for v5.18 2022-04-19 17:26:01 +02:00
target
trace sched/tracing: Append prev_state to tp args instead 2022-05-12 00:37:11 +02:00
uapi Networking fixes for 5.18-rc7, including fixes from wireless, 2022-05-12 11:51:45 -07:00
vdso
video
xen xen/grant-table: remove readonly parameter from functions 2022-03-15 20:34:40 -05:00