From: Dave Jiang Date: Tue, 19 May 2015 20:52:04 +0000 (-0400) Subject: NTB: Default to CPU memcpy for performance X-Git-Tag: v4.2-rc1~5^2~5 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=a41ef053f700618f5f55a1dd658908a71163400b;p=profile%2Fcommon%2Fplatform%2Fkernel%2Flinux-artik7.git NTB: Default to CPU memcpy for performance Disable DMA usage by default, since the CPU provides much better performance with write combining. Provide a module parameter to enable DMA usage when offloading the memcpy is preferred. Signed-off-by: Dave Jiang Signed-off-by: Allen Hubbe Signed-off-by: Jon Mason --- diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c index 7a765d3..e07b056 100644 --- a/drivers/ntb/ntb_transport.c +++ b/drivers/ntb/ntb_transport.c @@ -88,6 +88,10 @@ static unsigned int copy_bytes = 1024; module_param(copy_bytes, uint, 0644); MODULE_PARM_DESC(copy_bytes, "Threshold under which NTB will use the CPU to copy instead of DMA"); +static bool use_dma; +module_param(use_dma, bool, 0644); +MODULE_PARM_DESC(use_dma, "Use DMA engine to perform large data copy"); + static struct dentry *nt_debugfs_dir; struct ntb_queue_entry { @@ -1589,10 +1593,15 @@ ntb_transport_create_queue(void *data, struct device *client_dev, dma_cap_zero(dma_mask); dma_cap_set(DMA_MEMCPY, dma_mask); - qp->dma_chan = dma_request_channel(dma_mask, ntb_dma_filter_fn, - (void *)(unsigned long)node); - if (!qp->dma_chan) - dev_info(&pdev->dev, "Unable to allocate DMA channel, using CPU instead\n"); + if (use_dma) { + qp->dma_chan = dma_request_channel(dma_mask, ntb_dma_filter_fn, + (void *)(unsigned long)node); + if (!qp->dma_chan) + dev_info(&pdev->dev, "Unable to allocate DMA channel\n"); + } else { + qp->dma_chan = NULL; + } + dev_dbg(&pdev->dev, "Using %s memcpy\n", qp->dma_chan ? "DMA" : "CPU"); for (i = 0; i < NTB_QP_DEF_NUM_ENTRIES; i++) { entry = kzalloc_node(sizeof(*entry), GFP_ATOMIC, node);