Quoting Alex Bennée (2017-09-21 06:06:25)
> The mmio path (see exec.c:prepare_mmio_access) already protects itself
> against recursive locking and it makes sense to do the same for
> io_readx/writex. Otherwise any helper running in the BQL context will
> assert when it attempts to write to device memory as in the case of
> the bug report.
>
> Signed-off-by: Alex Bennée <alex.ben...@linaro.org>
> CC: Richard Jones <rjo...@redhat.com>
> CC: Paolo Bonzini <bonz...@gnu.org>
FYI: this patch has been tagged for stable 2.10.1, but is not yet
upstream. Patch freeze for 2.10.1 is September 27th.
> ---
> accel/tcg/cputlb.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
> index e72415a882..bcbcc4db6c 100644
> --- a/accel/tcg/cputlb.c
> +++ b/accel/tcg/cputlb.c
> @@ -765,7 +765,7 @@ static uint64_t io_readx(CPUArchState *env, CPUIOTLBEntry
> *iotlbentry,
>
> cpu->mem_io_vaddr = addr;
>
> - if (mr->global_locking) {
> + if (mr->global_locking && !qemu_mutex_iothread_locked()) {
> qemu_mutex_lock_iothread();
> locked = true;
> }
> @@ -800,7 +800,7 @@ static void io_writex(CPUArchState *env, CPUIOTLBEntry
> *iotlbentry,
> cpu->mem_io_vaddr = addr;
> cpu->mem_io_pc = retaddr;
>
> - if (mr->global_locking) {
> + if (mr->global_locking && !qemu_mutex_iothread_locked()) {
> qemu_mutex_lock_iothread();
> locked = true;
> }
> --
> 2.14.1
>
>
