From: Thomas Petazzoni <thomas.petazz...@free-electrons.com>
Date: Wed, 21 Dec 2016 11:28:49 +0100
> Since commit 71ce391dfb784 ("net: mvpp2: enable proper per-CPU TX
> buffers unmapping"), we are not correctly DMA unmapping TX buffers for
> fragments.
>
> Indeed, the mvpp2_txq_inc_put() function only stores in the
> txq_cpu->tx_buffs[] array the physical address of the buffer to be
> DMA-unmapped when skb != NULL. In addition, when DMA-unmapping, we use
> skb_headlen(skb) to get the size to be unmapped. Both of this works fine
> for TX descriptors that are associated directly to a SKB, but not the
> ones that are used for fragments, with a NULL pointer as skb:
>
> - We have a NULL physical address when calling DMA unmap
> - skb_headlen(skb) crashes because skb is NULL
>
> This causes random crashes when fragments are used.
>
> To solve this problem, we need to:
>
> - Store the physical address of the buffer to be unmapped
> unconditionally, regardless of whether it is tied to a SKB or not.
>
> - Store the length of the buffer to be unmapped, which requires a new
> field.
>
> Instead of adding a third array to store the length of the buffer to be
> unmapped, and as suggested by David Miller, this commit refactors the
> tx_buffs[] and tx_skb[] arrays of 'struct mvpp2_txq_pcpu' into a
> separate structure 'mvpp2_txq_pcpu_buf', to which a 'size' field is
> added. Therefore, instead of having three arrays to allocate/free, we
> have a single one, which also improve data locality, reducing the
> impact on the CPU cache.
>
> Fixes: 71ce391dfb784 ("net: mvpp2: enable proper per-CPU TX buffers
> unmapping")
> Reported-by: Raphael G <raphael.g...@corp.ovh.com>
> Cc: Raphael G <raphael.g...@corp.ovh.com>
> Cc: sta...@vger.kernel.org
> Signed-off-by: Thomas Petazzoni <thomas.petazz...@free-electrons.com>
> ---
> Changes since v1:
> - As requested by David Miller, move the per TX buffer fields into a
> separate structure, 'struct mvpp2_txq_pcpu_buf', so that instead of
> allocating three arrays, we allocate a single array.
Indeed, this looks a lot better.
Applied, thanks.
