omap-sham: use prefetch mechanism to increase transfer speed
cover omap-sham: use prefetch mechanism to increase transfer speed
Samu Onkalo (1):
crypto: omap-sham: Adjust DMA parameters
drivers/crypto/omap-sham.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-)
PATCH 1/1 crypto: omap-sham: Adjust DMA parameters
From: Samu Onkalo <[email protected]>
DMA is set to use burst mode also for source channel. It should descrease memory bandwidth needs.
DMA synchronization is set to use prefetch mechanism. SHAM block is behind L4 bus and it doesn't have fifo. SHAM block is stalling as long as the new data is available. It takes time to fetch data from memory and transfer it via L4 bus. With prefetch enabled, data is waiting in DMA fifo and SHAM block receives new data block faster.
This increases SHA processing speed up to 30 percent depending on the bus / memory load.
--- drivers/crypto/omap-sham.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-)
drivers/crypto/omap-sham.c +5 -1
@@ -311,7 +311,8 @@ static int omap_sham_xmit_dma(struct omap_sham_dev *dd, dma_addr_t dma_addr, len32 =DIV_ROUND_UP (length,sizeof (u32));omap_set_dma_transfer_params (dd->dma_lch,OMAP_DMA_DATA_TYPE_S32 , len32, -1 ,OMAP_DMA_SYNC_PACKET , dd->dma,OMAP_DMA_DST_SYNC ); +1 ,OMAP_DMA_SYNC_PACKET , dd->dma, +OMAP_DMA_DST_SYNC_PREFETCH );omap_set_dma_src_params (dd->dma_lch,0 ,OMAP_DMA_AMODE_POST_INC , dma_addr,0 ,0 );
@@ -1072,6 +1073,9 @@ static int omap_sham_dma_init(struct omap_sham_dev *dd)omap_set_dma_dest_burst_mode (dd->dma_lch,OMAP_DMA_DATA_BURST_16 ); +omap_set_dma_src_burst_mode (dd->dma_lch, +OMAP_DMA_DATA_BURST_4 ); +return 0 ; }
Signature · 1 line
1.7.0.4
Herbert Xu
maintainerRe: [PATCH 1/1] crypto: omap-sham: Adjust DMA parameters
Quoted 13 lines · Dmitry Kasatkin
From: Samu Onkalo <[email protected]>
DMA is set to use burst mode also for source channel. It should descrease memory bandwidth needs.
DMA synchronization is set to use prefetch mechanism. SHAM block is behind L4 bus and it doesn't have fifo. SHAM block is stalling as long as the new data is available. It takes time to fetch data from memory and transfer it via L4 bus. With prefetch enabled, data is waiting in DMA fifo and SHAM block receives new data block faster.
This increases SHA processing speed up to 30 percent depending on the bus / memory load.
Quoted 2 lines
Signed-off-by Samu Onkalo <[email protected]>Signed-off-by Dmitry Kasatkin <[email protected]>
Patch applied. Thanks!
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt