Re: [PATCH net-next v3 7/7] bnxt_en: Fix ring accounting and validation when rings are constrained

[Gregory Price <gourry@gourry.net>]

On Mon, Oct 12, 2026 at 03:08:27PM -0700, Michael Chan wrote:
> When __bnxt_reserve_rings() reserves fewer TX rings than requested,
> and an XDP program is attached, the driver used to blindly subtract
> bp->tx_nr_rings_xdp from bp->tx_nr_rings, potentially causing an integer
> underflow.
> 

This is outside my normal space, was reviewing this for testing and had
an agent spot what it thinks is a bug.  It seemed reasonable to me but
if you think this is wrong, please ignore.

> @@ -11602,25 +11673,23 @@ int bnxt_reserve_rings(struct bnxt *bp, bool irq_re_init)
>  		if (!rc)
>  			rc = bnxt_init_int_mode(bp);
>  		bnxt_ulp_irq_restart(bp, rc);
> -	} else if (irq_change && !rc) {
> -		if (bnxt_change_msix(bp, irqs_required) != irqs_required)
> -			rc = -ENOSPC;
> +	} else if (irq_re_init && BNXT_NEW_RM(bp) && !rc) {

should this be:

       } else if (irq_re_init && BNXT_NEW_RM(bp) && bp->irq_tbl && !rc) {

Note from the LLM was:

  Unlike irq_change, this no longer implies bp->irq_tbl != NULL.  After a
  FW/AER reset irq_tbl is gone but total_irqs is stale, so if the rings
  shrink we call bnxt_change_msix() with no table and fail with -ENOSPC.

  Should this also check bp->irq_tbl?

~Gregory