Re: [PATCH net] net: xilinx: axienet: Fix IRQ coalescing packet count overflow

[Sean Anderson]

On 9/5/24 10:55, Sean Anderson wrote:
> On 9/4/24 13:19, Pandey, Radhey Shyam wrote:
>>> -----Original Message-----
>>> From: Sean Anderson <sean.anderson@xxxxxxxxx>
>>> Sent: Tuesday, September 3, 2024 11:31 PM
>>> To: Pandey, Radhey Shyam <radhey.shyam.pandey@xxxxxxx>; David S .
>>> Miller <davem@xxxxxxxxxxxxx>; Eric Dumazet <edumazet@xxxxxxxxxx>;
>>> Jakub Kicinski <kuba@xxxxxxxxxx>; Paolo Abeni <pabeni@xxxxxxxxxx>;
>>> netdev@xxxxxxxxxxxxxxx
>>> Cc: Simek, Michal <michal.simek@xxxxxxx>; linux-arm-
>>> kernel@xxxxxxxxxxxxxxxxxxx; Ariane Keller <ariane.keller@xxxxxxxxxxxxxx>;
>>> linux-kernel@xxxxxxxxxxxxxxx; Daniel Borkmann <daniel@xxxxxxxxxxxxx>;
>>> Andy Chiu <andy.chiu@xxxxxxxxxx>; Sean Anderson
>>> <sean.anderson@xxxxxxxxx>
>>> Subject: [PATCH net] net: xilinx: axienet: Fix IRQ coalescing packet count
>>> overflow
>>> 
>>> If coalesce_count is greater than 255 it will not fit in the register and
>>> will overflow. Clamp it to 255 for more-predictable results.
>>> 
>>> Signed-off-by: Sean Anderson <sean.anderson@xxxxxxxxx>
>>> Fixes: 8a3b7a252dca ("drivers/net/ethernet/xilinx: added Xilinx AXI Ethernet
>>> driver")
>>> ---
>>> 
>>>  drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 6 ++++--
>>>  1 file changed, 4 insertions(+), 2 deletions(-)
>>> 
>>> diff --git a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>>> b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>>> index 9aeb7b9f3ae4..5f27fc1c4375 100644
>>> --- a/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>>> +++ b/drivers/net/ethernet/xilinx/xilinx_axienet_main.c
>>> @@ -252,7 +252,8 @@ static u32 axienet_usec_to_timer(struct axienet_local
>>> *lp, u32 coalesce_usec)
>>>  static void axienet_dma_start(struct axienet_local *lp)
>>>  {
>>>  	/* Start updating the Rx channel control register */
>>> -	lp->rx_dma_cr = (lp->coalesce_count_rx <<
>>> XAXIDMA_COALESCE_SHIFT) |
>>> +	lp->rx_dma_cr = (min(lp->coalesce_count_rx, 255) <<
>>> +			 XAXIDMA_COALESCE_SHIFT) |
>>>  			XAXIDMA_IRQ_IOC_MASK |
>> 
>> Instead of every time clamping coalesce_count_rx on read I think better 
>> to do it place where it set in axienet_ethtools_set_coalesce()? It would
>> also represent the coalesce count state that is reported by get_coalesce()
>> and same is being used in programming.
> 
> The parameter which will be trickier is the timer, which is also clamped
> but depends on the (DMA) clock speed. So theoretically it may be
> different if the clock gets changed at runtime between when we set
> coalesce and when we apply it. But do we even support dynamic rate
> changes for that clock?
> 
> In either case, I think this will be easier to do as part of [1], since I
> am already rearranging the calculation in that patch.

Implemented as https://lore.kernel.org/netdev/20240909235208.1331065-6-sean.anderson@xxxxxxxxx/

--Sean

> --Sean
> 
> [1] https://lore.kernel.org/netdev/20240903192524.4158713-2-sean.anderson@xxxxxxxxx/
> 
>>> XAXIDMA_IRQ_ERROR_MASK;
>>>  	/* Only set interrupt delay timer if not generating an interrupt on
>>>  	 * the first RX packet. Otherwise leave at 0 to disable delay interrupt.
>>> @@ -264,7 +265,8 @@ static void axienet_dma_start(struct axienet_local
>>> *lp)
>>>  	axienet_dma_out32(lp, XAXIDMA_RX_CR_OFFSET, lp->rx_dma_cr);
>>> 
>>>  	/* Start updating the Tx channel control register */
>>> -	lp->tx_dma_cr = (lp->coalesce_count_tx <<
>>> XAXIDMA_COALESCE_SHIFT) |
>>> +	lp->tx_dma_cr = (min(lp->coalesce_count_tx, 255) <<
>>> +			 XAXIDMA_COALESCE_SHIFT) |
>>>  			XAXIDMA_IRQ_IOC_MASK |
>>> XAXIDMA_IRQ_ERROR_MASK;
>>>  	/* Only set interrupt delay timer if not generating an interrupt on
>>>  	 * the first TX packet. Otherwise leave at 0 to disable delay interrupt.
>>> --
>>> 2.35.1.1320.gc452695387.dirty
>>