From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from out-183.mta1.migadu.com (out-183.mta1.migadu.com [95.215.58.183]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id DFF4535A388 for ; Tue, 24 Mar 2026 16:09:28 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=95.215.58.183 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774368570; cv=none; b=fZqND6ah8WVUABq3zPJEeFQ9UBg3dleBsT3heJ0YVSrxJ0rC1TTXiS4UiYfvFZTmcE0FExqo2n40hJzi4/mimzBI8foi5i5LD2cnt+H7UvFnpgV+1iyng6ThpdPn6pD+a9+1Vt08LuZDumZf0X4U2iEmiQ/lqn38tMzujRpRtmg= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774368570; c=relaxed/simple; bh=Sz9kUo6deDneVXClMdOf2aaDC4MhNRkUAlmaVpOCVXs=; h=Message-ID:Date:MIME-Version:Subject:To:Cc:References:From: In-Reply-To:Content-Type; b=eVxb9VnSVUGV0wjl8xg5UN2n5Xs+SXzNwgazLvZD8g52w9LdiG7Plklv3xdCCCzghmy4TpgUVlTq3ox8wqFtjxZ18i5PYISvomtbettK3YPnWdn55qIMRtK0GVM9RKV+uGELSnAFQS9w81YhaG0FLpNkDPg3M8koPex7JbqZfac= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=qSiWh7yY; arc=none smtp.client-ip=95.215.58.183 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="qSiWh7yY" Message-ID: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1774368557; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=q9HnuHlSuBejVcQaUC7IGTPI3zmgypr9vBTmv2lZWRQ=; b=qSiWh7yYiVFnUisxmjTZTQQ0ii6l53riqvrgCbjqQKxS3pV5P2amhE0GVBoPnWQxqBpebR /uozWB1BNPZ4OuxnnhyCR+mGwiyuf7/q+eHoy8TWn3nazEwyD6FKdDReqeQZhBrUihMrR4 8rhB0NCI9wY5hq0ndYx+9Td05Jkjd5U= Date: Tue, 24 Mar 2026 12:09:04 -0400 Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Subject: Re: [PATCH net 2/2] net: xilinx: axienet: Fix BQL accounting for multi-BD TX packets To: Suraj Gupta , andrew+netdev@lunn.ch, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, michal.simek@amd.com, radhey.shyam.pandey@amd.com, horms@kernel.org Cc: netdev@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, harini.katakam@amd.com References: <20260324145353.1899248-1-suraj.gupta2@amd.com> <20260324145353.1899248-3-suraj.gupta2@amd.com> Content-Language: en-US X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: Sean Anderson In-Reply-To: <20260324145353.1899248-3-suraj.gupta2@amd.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Migadu-Flow: FLOW_OUT On 3/24/26 10:53, Suraj Gupta wrote: > When a TX packet spans multiple buffer descriptors (scatter-gather), > the per-BD byte count is accumulated into a local variable that resets > on each NAPI poll. If the BDs for a single packet complete across > different polls, the earlier bytes are lost and never credited to BQL. > This causes BQL to think bytes are permanently in-flight, eventually > stalling the TX queue. > > Fix this by replacing the local accumulator with a persistent counter > (tx_compl_bytes) that survives across polls and is reset only after > updating BQL and stats. Do we need this? Can't we just do something like diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c index 415e9bc252527..1ea8a6592bce1 100644 --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c @@ -768,6 +768,7 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 *sizep, int budge if (cur_p->skb) { struct axienet_cb *cb = (void *)cur_p->skb->cb; + *sizep += skb->len; dma_unmap_sgtable(lp->dev, &cb->sgt, DMA_TO_DEVICE, 0); sg_free_table_chained(&cb->sgt, XAE_INLINE_SG_CNT); napi_consume_skb(cur_p->skb, budget); @@ -783,8 +784,6 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 *sizep, int budge wmb(); cur_p->cntrl = 0; cur_p->status = 0; - - *sizep += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK; } smp_store_release(&lp->tx_bd_ci, (ci + i) & (lp->tx_bd_num - 1)); > Fixes: c900e49d58eb ("net: xilinx: axienet: Implement BQL") > Signed-off-by: Suraj Gupta > --- > drivers/net/ethernet/xilinx/xilinx_axienet.h | 3 +++ > .../net/ethernet/xilinx/xilinx_axienet_main.c | 20 +++++++++---------- > 2 files changed, 13 insertions(+), 10 deletions(-) > > diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet.h b/drivers/net/ethernet/xilinx/xilinx_axienet.h > index 602389843342..a4444c939451 100644 > --- a/drivers/net/ethernet/xilinx/xilinx_axienet.h > +++ b/drivers/net/ethernet/xilinx/xilinx_axienet.h > @@ -509,6 +509,8 @@ struct skbuf_dma_descriptor { > * complete. Only updated at runtime by TX NAPI poll. > * @tx_bd_tail: Stores the index of the next Tx buffer descriptor in the ring > * to be populated. > + * @tx_compl_bytes: Accumulates TX completion length until a full packet is > + * reported to the stack. > * @tx_packets: TX packet count for statistics > * @tx_bytes: TX byte count for statistics > * @tx_stat_sync: Synchronization object for TX stats > @@ -592,6 +594,7 @@ struct axienet_local { > u32 tx_bd_num; > u32 tx_bd_ci; > u32 tx_bd_tail; > + u32 tx_compl_bytes; > u64_stats_t tx_packets; > u64_stats_t tx_bytes; > struct u64_stats_sync tx_stat_sync; > diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c > index b06e4c37ff61..95bf61986cb7 100644 > --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c > +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c > @@ -692,6 +692,8 @@ static void axienet_dma_stop(struct axienet_local *lp) > axienet_lock_mii(lp); > __axienet_device_reset(lp); > axienet_unlock_mii(lp); > + > + lp->tx_compl_bytes = 0; > } > > /** > @@ -770,8 +772,6 @@ static int axienet_device_reset(struct net_device *ndev) > * @first_bd: Index of first descriptor to clean up > * @nr_bds: Max number of descriptors to clean up > * @force: Whether to clean descriptors even if not complete > - * @sizep: Pointer to a u32 filled with the total sum of all bytes > - * in all cleaned-up descriptors. Ignored if NULL. > * @budget: NAPI budget (use 0 when not called from NAPI poll) > * > * Would either be called after a successful transmit operation, or after > @@ -780,7 +780,7 @@ static int axienet_device_reset(struct net_device *ndev) > * Return: The number of packets handled. > */ > static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd, > - int nr_bds, bool force, u32 *sizep, int budget) > + int nr_bds, bool force, int budget) > { > struct axidma_bd *cur_p; > unsigned int status; > @@ -819,8 +819,8 @@ static int axienet_free_tx_chain(struct axienet_local *lp, u32 first_bd, > cur_p->cntrl = 0; > cur_p->status = 0; > > - if (sizep) > - *sizep += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK; > + if (!force) > + lp->tx_compl_bytes += status & XAXIDMA_BD_STS_ACTUAL_LEN_MASK; > } > > if (!force) { > @@ -999,18 +999,18 @@ static int axienet_tx_poll(struct napi_struct *napi, int budget) > { > struct axienet_local *lp = container_of(napi, struct axienet_local, napi_tx); > struct net_device *ndev = lp->ndev; > - u32 size = 0; > int packets; > > packets = axienet_free_tx_chain(lp, lp->tx_bd_ci, lp->tx_bd_num, false, > - &size, budget); > + budget); > > if (packets) { > - netdev_completed_queue(ndev, packets, size); > + netdev_completed_queue(ndev, packets, lp->tx_compl_bytes); > u64_stats_update_begin(&lp->tx_stat_sync); > u64_stats_add(&lp->tx_packets, packets); > - u64_stats_add(&lp->tx_bytes, size); > + u64_stats_add(&lp->tx_bytes, lp->tx_compl_bytes); > u64_stats_update_end(&lp->tx_stat_sync); > + lp->tx_compl_bytes = 0; > > /* Matches barrier in axienet_start_xmit */ > smp_mb(); > @@ -1115,7 +1115,7 @@ axienet_start_xmit(struct sk_buff *skb, struct net_device *ndev) > netdev_err(ndev, "TX DMA mapping error\n"); > ndev->stats.tx_dropped++; > axienet_free_tx_chain(lp, orig_tail_ptr, ii + 1, > - true, NULL, 0); > + true, 0); > dev_kfree_skb_any(skb); > return NETDEV_TX_OK; > }