From: Allen Hubbe <Allen.Hubbe@emc.com>
To: linux-ntb@googlegroups.com
Cc: linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
Jon Mason <jdmason@kudzu.us>, Dave Jiang <dave.jiang@intel.com>,
Allen Hubbe <Allen.Hubbe@emc.com>
Subject: [PATCH v2 15/17] NTB: Default to CPU memcpy for performance
Date: Thu, 21 May 2015 04:20:59 -0400 [thread overview]
Message-ID: <f81e011a326e20b8ef0cb2c863f3ace4b0e23318.1432196075.git.Allen.Hubbe@emc.com> (raw)
In-Reply-To: <cover.1432192681.git.Allen.Hubbe@emc.com>
In-Reply-To: <cover.1432196075.git.Allen.Hubbe@emc.com>
From: Dave Jiang <dave.jiang@intel.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@intel.com>
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.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 95252ea..9c266e1 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 {
@@ -1571,10 +1575,15 @@ 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
next prev parent reply other threads:[~2015-05-21 13:31 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-21 8:20 [PATCH v2 00/17] NTB: Add NTB hardware abstraction layer Allen Hubbe
2015-05-21 9:45 ` Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 01/17] NTB: Move files in preparation for NTB abstraction Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 02/17] NTB: Add NTB hardware abstraction layer Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 03/17] NTB: Enable link for Intel root port mode in probe Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 04/17] NTB: Check the device ID to set errata flags Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 05/17] NTB: Add parameters for Intel SNB B2B addresses Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 06/17] NTB: Add ping pong test client Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 07/17] NTB: Add tool " Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 08/17] NTB: Rate limit ntb_qp_link_work Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 09/17] NTB: Differentiate transport link down messages Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 10/17] NTB: Do not advance transport RX on link down Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 11/17] NTB: Reset transport QP link stats on down Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 12/17] NTB: Use NUMA memory and DMA chan in transport Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 13/17] NTB: Use NUMA memory in Intel driver Allen Hubbe
2015-05-21 8:20 ` [PATCH v2 14/17] NTB: Improve performance with write combining Allen Hubbe
2015-05-21 8:20 ` Allen Hubbe [this message]
2015-05-21 8:21 ` [PATCH v2 16/17] NTB: Fix small code format issues in transport Allen Hubbe
2015-05-21 8:21 ` [PATCH v2 17/17] NTB: Rename intel hw to proper platform names Allen Hubbe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=f81e011a326e20b8ef0cb2c863f3ace4b0e23318.1432196075.git.Allen.Hubbe@emc.com \
--to=allen.hubbe@emc.com \
--cc=dave.jiang@intel.com \
--cc=jdmason@kudzu.us \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-ntb@googlegroups.com \
--cc=linux-pci@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®