forked from luck/tmp_suning_uos_patched
0ad31889c4
[ Upstream commit 01e31bea7e622f1890c274f4aaaaf8bccd296aa5 ]
Currently the vhost_zerocopy_callback() maybe be called to decrease
the refcount when sendmsg fails in tun. The error handling in vhost
handle_tx_zerocopy() will try to decrease the same refcount again.
This is wrong. To fix this issue, we only call vhost_net_ubuf_put()
when vq->heads[nvq->desc].len == VHOST_DMA_IN_PROGRESS.
Fixes:
|
||
---|---|---|
.. | ||
iotlb.c | ||
Kconfig | ||
Makefile | ||
net.c | ||
scsi.c | ||
test.c | ||
test.h | ||
vdpa.c | ||
vhost.c | ||
vhost.h | ||
vringh.c | ||
vsock.c |