[<prev] [next>] [day] [month] [year] [list]
Message-Id: <1453801470-3331-1-git-send-email-decui@microsoft.com>
Date: Tue, 26 Jan 2016 01:44:30 -0800
From: Dexuan Cui <decui@...rosoft.com>
To: gregkh@...uxfoundation.org, davem@...emloft.net,
stephen@...workplumber.org, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org,
driverdev-devel@...uxdriverproject.org, olaf@...fle.de,
apw@...onical.com, jasowang@...hat.com, kys@...rosoft.com,
vkuznets@...hat.com
Cc: pebolle@...cali.nl, stefanha@...hat.com, dan.carpenter@...cle.com
Subject: [PATCH V6 3/8] Drivers: hv: vmbus: vmbus_sendpacket_ctl: hvsock: avoid unnecessary signaling
When the hvsock channel's outbound ringbuffer is full (i.e.,
hv_ringbuffer_write() returns -EAGAIN), we should avoid the unnecessary
signaling the host.
Signed-off-by: Dexuan Cui <decui@...rosoft.com>
---
drivers/hv/channel.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c
index 1161d68..3f04533 100644
--- a/drivers/hv/channel.c
+++ b/drivers/hv/channel.c
@@ -659,6 +659,9 @@ int vmbus_sendpacket_ctl(struct vmbus_channel *channel, void *buffer,
* If we cannot write to the ring-buffer; signal the host
* even if we may not have written anything. This is a rare
* enough condition that it should not matter.
+ * NOTE: in this case, the hvsock channel is an exception, because
+ * it looks the host side's hvsock implementation has a throttling
+ * mechanism which can hurt the performance otherwise.
*/
if (channel->signal_policy)
@@ -666,7 +669,8 @@ int vmbus_sendpacket_ctl(struct vmbus_channel *channel, void *buffer,
else
kick_q = true;
- if (((ret == 0) && kick_q && signal) || (ret))
+ if (((ret == 0) && kick_q && signal) ||
+ (ret && !is_hvsock_channel(channel)))
vmbus_setevent(channel);
return ret;
--
2.1.0
Powered by blists - more mailing lists