linux-kernel - [PATCH 15/16] ntb: default to cpu memcpy for performance

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <0979e7faf4814d1059640cd448a5ad8496604a98.1432135632.git.Allen.Hubbe@emc.com>
Date:	Wed, 20 May 2015 11:41:16 -0400
From:	Allen Hubbe <Allen.Hubbe@....com>
To:	linux-ntb@...glegroups.com
Cc:	linux-kernel@...r.kernel.org, linux-pci@...r.kernel.org,
	Jon Mason <jdmason@...zu.us>,
	Dave Jiang <dave.jiang@...el.com>,
	Allen Hubbe <Allen.Hubbe@....com>
Subject: [PATCH 15/16] ntb: default to cpu memcpy for performance

From: Dave Jiang <dave.jiang@...el.com>

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 <dave.jiang@...el.com>
Signed-off-by: Allen Hubbe <Allen.Hubbe@....com>
---
 drivers/ntb/ntb_transport.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/ntb/ntb_transport.c b/drivers/ntb/ntb_transport.c
index cc0b7ff..237f76f 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 {
@@ -1570,10 +1574,15 @@ 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);
-- 
2.4.0.rc0.43.gcf8a8c6

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/