Re: [PATCH net-next v16 4/5] eea: create/destroy rx,tx queues for netdevice open and stop

[Jakub Kicinski <kuba@kernel.org>]

On Mon, 24 Nov 2025 09:42:50 +0800 Xuan Zhuo wrote:
> +static void enet_bind_new_q_and_cfg(struct eea_net *enet,
> +				    struct eea_net_init_ctx *ctx)
> +{
> +	struct eea_net_rx *rx;
> +	struct eea_net_tx *tx;
> +	int i;
> +
> +	enet->cfg = ctx->cfg;
> +
> +	enet->rx = ctx->rx;
> +	enet->tx = ctx->tx;
> +
> +	for (i = 0; i < ctx->cfg.rx_ring_num; i++) {
> +		rx = ctx->rx[i];
> +		tx = &ctx->tx[i];
> +
> +		rx->enet = enet;
> +		tx->enet = enet;

I think you need to move it up sooner? Or re-implement the cleanup
paths to handle partial initialization correctly.. Right now if
eea_alloc_rx_hdr() fails, eea_alloc_rx()->eea_free_rx()->eea_free_rx_hdr()
will dereference rx->enet before it's set

> +static void eea_free_meta(struct eea_net_tx *tx)
> +{
> +	struct eea_sq_free_stats stats;
> +	struct eea_tx_meta *meta;
> +	int i;
> +
> +	while ((meta = eea_tx_meta_get(tx)))
> +		meta->skb = NULL;
> +
> +	for (i = 0; i < tx->enet->cfg.tx_ring_num; i++) {

Similarly is it okay to use enet->cfg here? Could be that we're freeing
cfg before it's bound to a device.