Re: [PATCH 5/7] igb: eliminate duplicate barriers on weakly-ordered archs

[Alexander Duyck]

On Tue, Mar 13, 2018 at 8:20 PM, Sinan Kaya  wrote:
> Code includes wmb() followed by writel(). writel() already has a barrier
> on some architectures like arm64.
>
> This ends up CPU observing two barriers back to back before executing the
> register write.
>
> Since code already has an explicit barrier call, changing writel() to
> writel_relaxed().
>
> Signed-off-by: Sinan Kaya 
> ---
>  drivers/net/ethernet/intel/igb/igb_main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c 
> b/drivers/net/ethernet/intel/igb/igb_main.c
> index b88fae7..ba8ccb5 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -8072,7 +8072,7 @@ void igb_alloc_rx_buffers(struct igb_ring *rx_ring, u16 
> cleaned_count)
>  * such as IA-64).
>  */
> wmb();
> -   writel(i, rx_ring->tail);
> +   writel_relaxed(i, rx_ring->tail);
> }
>  }
>

This one missed the writel at the end of igb_tx_map().

Re: [PATCH 5/7] igb: eliminate duplicate barriers on weakly-ordered archs

[Alexander Duyck]

On Tue, Mar 13, 2018 at 8:20 PM, Sinan Kaya  wrote:
> Code includes wmb() followed by writel(). writel() already has a barrier
> on some architectures like arm64.
>
> This ends up CPU observing two barriers back to back before executing the
> register write.
>
> Since code already has an explicit barrier call, changing writel() to
> writel_relaxed().
>
> Signed-off-by: Sinan Kaya 
> ---
>  drivers/net/ethernet/intel/igb/igb_main.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/intel/igb/igb_main.c 
> b/drivers/net/ethernet/intel/igb/igb_main.c
> index b88fae7..ba8ccb5 100644
> --- a/drivers/net/ethernet/intel/igb/igb_main.c
> +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> @@ -8072,7 +8072,7 @@ void igb_alloc_rx_buffers(struct igb_ring *rx_ring, u16 
> cleaned_count)
>  * such as IA-64).
>  */
> wmb();
> -   writel(i, rx_ring->tail);
> +   writel_relaxed(i, rx_ring->tail);
> }
>  }
>

This one missed the writel at the end of igb_tx_map().