[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20260128022924.4152852-1-kuba@kernel.org>
Date: Tue, 27 Jan 2026 18:29:24 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: rkannoth@...vell.com
Cc: Jakub Kicinski <kuba@...nel.org>,
sbhatta@...vell.com,
edumazet@...gle.com,
andrew+netdev@...n.ch,
netdev@...r.kernel.org,
sgoutham@...vell.com,
sumang@...vell.com,
pabeni@...hat.com,
davem@...emloft.net,
linux-kernel@...r.kernel.org
Subject: Re: [net-next,v5,06/13] octeontx2-af: npc: cn20k: Use common APIs
This is an AI-generated review of your patch. The human sending this
email has considered the AI review valid, or at least plausible.
For local reproduction steps see https://netdev-ai.bots.linux.dev/ai-local.html
---
octeontx2-af: npc: cn20k: Use common APIs
This patch updates the common NPC APIs to support both cn20k and previous
silicon variants by adding cn20k-specific register access and dispatching
to cn20k-specific routines for MCAM entry operations.
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> index 246f3568a795..59988a54506a 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
[ ... ]
> @@ -2063,15 +2154,13 @@ static void rvu_npc_setup_interfaces(struct rvu *rvu, int blkaddr)
> struct npc_mcam *mcam = &rvu->hw->mcam;
> struct rvu_hwinfo *hw = rvu->hw;
> u64 nibble_ena, rx_kex, tx_kex;
> - u64 *keyx_cfg;
> + u64 *keyx_cfg, reg;
> u8 intf;
>
> - if (is_cn20k(rvu->pdev)) {
> + if (is_cn20k(rvu->pdev))
> keyx_cfg = mkex_extr->keyx_cfg;
> - goto skip_miss_cntr;
> - }
> -
> - keyx_cfg = mkex->keyx_cfg;
> + else
> + keyx_cfg = mkex->keyx_cfg;
>
> /* Reserve last counter for MCAM RX miss action which is set to
> * drop packet. This way we will know how many pkts didn't match
> @@ -2080,7 +2169,6 @@ static void rvu_npc_setup_interfaces(struct rvu *rvu, int blkaddr)
> mcam->counters.max--;
> mcam->rx_miss_act_cntr = mcam->counters.max;
^^^^^^^^^^
Was the removal of "goto skip_miss_cntr" intentional for cn20k? Previously,
cn20k would skip the counter reservation here. Now cn20k reserves a counter
via mcam->counters.max-- but never programs NPC_AF_INTFX_MISS_STAT_ACT (due
to the "continue" later in the RX interface loop). This appears to waste
one counter on cn20k.
> -skip_miss_cntr:
> rx_kex = keyx_cfg[NIX_INTF_RX];
> tx_kex = keyx_cfg[NIX_INTF_TX];
>
Powered by blists - more mailing lists