On Wed, May 06, 2026 at 07:34:20PM -0700, Jakub Kicinski wrote: > On Mon, 04 May 2026 17:27:49 -0700 Bobby Eshleman wrote: > > + if (bind_dev != netdev) > > + netdev_lock(bind_dev); > > + dma_dev = netdev_queue_get_dma_dev(bind_dev, 0, NETDEV_QUEUE_TYPE_TX); > > + if (bind_dev != netdev) > > + netdev_unlock(bind_dev); > > + binding = net_devmem_bind_dmabuf(bind_dev, > > + bind_dev != netdev ? netdev : NULL, > > + dma_dev, DMA_TO_DEVICE, dmabuf_fd, > > + priv, info->extack); > > Not sure if it matters but are we intentionally releasing the bind_dev > lock before calling net_devmem_bind_dmabuf() ? Previously more code here > was covered by the physical netdev's lock.
True, lock needs to be held at least until after binding. I have this fixed in the next rev. Best, Bobby
