On Tue, 2019-03-05 at 10:10 +0000, Radoslav Gerganov wrote:
> In f_hidg_write() the write_spinlock is acquired before calling
> usb_ep_queue() which causes a deadlock when dummy_hcd is being used.
> This is because dummy_queue() callbacks into f_hidg_req_complete()
> which tries to acquire the same spinlock. This is (part of) the
> backtrace when the deadlock occurs:
>
> 0xffffffffc06b1410 in f_hidg_req_complete
> 0xffffffffc06a590a in usb_gadget_giveback_request
> 0xffffffffc06cfff2 in dummy_queue
> 0xffffffffc06a4b96 in usb_ep_queue
> 0xffffffffc06b1eb6 in f_hidg_write
> 0xffffffff8127730b in __vfs_write
> 0xffffffff812774d1 in vfs_write
> 0xffffffff81277725 in SYSC_write
>
> Fix this by releasing the write_spinlock before calling
> usb_ep_queue()
This fixes a usb_f_hid deadlock I've also been seeing with a FIDO key
emulator. This is a serious bug in that it brings down my entire
system if I use the hid gadget in any way, so can we please get it
applied? It turns out to have been introduced by this commit:
commit 749494b6bdbbaf0899aa1c62a1ad74cd747bce47
Author: Krzysztof Opasiak <kopasiak90@xxxxxxxxx>
Date: Tue Jan 24 03:27:24 2017 +0100
usb: gadget: f_hid: fix: Move IN request allocation to set_alt()
So you can add
Reviewed-by: James Bottomley <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>
Tested-by: James Bottomley <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx # 4.11+
Fixes: 749494b6bdbb ("usb: gadget: f_hid: fix: Move IN request allocation to set_alt()")
James
