From: Andrew Lunn Subject: Re: [PATCH 6/8] async_tx: adding mult and sum_product flags Date: Tue, 12 May 2015 18:05:35 +0200 Message-ID: <20150512160535.GK19927@lunn.ch> References: <1431445063-20226-1-git-send-email-maxime.ripard@free-electrons.com> <1431445063-20226-7-git-send-email-maxime.ripard@free-electrons.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Vinod Koul , Dan Williams , Gregory Clement , Jason Cooper , Sebastian Hesselbarth , dmaengine@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, linux-crypto@vger.kernel.org, Lior Amsalem , Thomas Petazzoni , Herbert Xu , "David S. Miller" To: Maxime Ripard Return-path: Content-Disposition: inline In-Reply-To: <1431445063-20226-7-git-send-email-maxime.ripard@free-electrons.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Tue, May 12, 2015 at 05:37:41PM +0200, Maxime Ripard wrote: > From: Lior Amsalem > > Some engines (like Marvell mv_xor) do not support mult and sum_product > operations as part of the pq support. > > This patch adds new flags: DMA_PREP_PQ_MULT & DMA_PREP_PQ_SUM_PRODUCT these > flags helps the driver identify such operations. > > Signed-off-by: Lior Amsalem > Reviewed-by: Ofer Heifetz > Reviewed-by: Nadav Haklai > Signed-off-by: Maxime Ripard > --- > crypto/async_tx/async_raid6_recov.c | 4 ++-- > include/linux/dmaengine.h | 4 ++++ > 2 files changed, 6 insertions(+), 2 deletions(-) > > diff --git a/crypto/async_tx/async_raid6_recov.c b/crypto/async_tx/async_raid6_recov.c > index 934a84981495..2db5486fd873 100644 > --- a/crypto/async_tx/async_raid6_recov.c > +++ b/crypto/async_tx/async_raid6_recov.c > @@ -47,7 +47,7 @@ async_sum_product(struct page *dest, struct page **srcs, unsigned char *coef, > struct device *dev = dma->dev; > dma_addr_t pq[2]; > struct dma_async_tx_descriptor *tx; > - enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P; > + enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P | DMA_PREP_PQ_SUM_PRODUCT; > > if (submit->flags & ASYNC_TX_FENCE) > dma_flags |= DMA_PREP_FENCE; > @@ -111,7 +111,7 @@ async_mult(struct page *dest, struct page *src, u8 coef, size_t len, > dma_addr_t dma_dest[2]; > struct device *dev = dma->dev; > struct dma_async_tx_descriptor *tx; > - enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P; > + enum dma_ctrl_flags dma_flags = DMA_PREP_PQ_DISABLE_P | DMA_PREP_PQ_MULT; > > if (submit->flags & ASYNC_TX_FENCE) > dma_flags |= DMA_PREP_FENCE; > diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h > index ad419757241f..f19ecebb4d3f 100644 > --- a/include/linux/dmaengine.h > +++ b/include/linux/dmaengine.h > @@ -174,6 +174,8 @@ struct dma_interleaved_template { > * operation it continues the calculation with new sources > * @DMA_PREP_FENCE - tell the driver that subsequent operations depend > * on the result of this operation > + * @DMA_PREP_PQ_MULT - tell the driver that this is a mult request > + * @DMA_PREP_PQ_SUM_PRODUCT - tell the driver that this is a sum product request > */ > enum dma_ctrl_flags { > DMA_PREP_INTERRUPT = (1 << 0), > @@ -182,6 +184,8 @@ enum dma_ctrl_flags { > DMA_PREP_PQ_DISABLE_Q = (1 << 3), > DMA_PREP_CONTINUE = (1 << 4), > DMA_PREP_FENCE = (1 << 5), > + DMA_PREP_PQ_MULT = (1 << 10), > + DMA_PREP_PQ_SUM_PRODUCT = (1 << 11), Any reason for skipping 6 to 9? Andrew