vhost_net: disable zerocopy by default
Subsystems (5)
PATCH 1/1 vhost_net: disable zerocopy by default
Vhost_net was known to suffer from HOL[1] issues which is not easy to fix. Several downstream disable the feature by default. What's more, the datapath was split and datacopy path got the support of batching and XDP support recently which makes it faster than zerocopy part for small packets transmission.
It looks to me that disable zerocopy by default is more appropriate. It cold be enabled by default again in the future if we fix the above issues.
[1] https://patchwork.kernel.org/patch/3787671/
--- drivers/vhost/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vhost/net.c +1 -1
@@ -36,7 +36,7 @@#include "vhost.h" -static int experimental_zcopytx =1 ; +static int experimental_zcopytx =0 ;module_param (experimental_zcopytx, int,0444 );MODULE_PARM_DESC (experimental_zcopytx,"Enable Zero Copy TX;" " 1 -Enable; 0 - Disable" );
Signature · 1 line
2.18.1
Quoted 6 lines · Jason Wang
Vhost_netwas known to suffer from HOL[1] issues which is not easy to fix. Several downstream disable the feature by default. What's more, the datapath was split and datacopy path got the support of batching and XDP support recently which makes it faster than zerocopy part for small packets transmission.
Quoted 7 lines
It looks to me that disable zerocopy by default is more appropriate. It cold be enabled by default again in the future if we fix the above issues.
[1] https://patchwork.kernel.org/patch/3787671/
Signed-off-by Jason Wang <[email protected]>
Quoted 19 lines
--- drivers/vhost/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/vhost/net.c +1
-1@@ -36,7 +36,7 @@#include "vhost.h" -static int experimental_zcopytx =1 ; +static int experimental_zcopytx =0 ;module_param (experimental_zcopytx, int,0444 );MODULE_PARM_DESC (experimental_zcopytx,"Enable Zero Copy TX;" " 1 -Enable; 0 - Disable" );Signature · 1 line
2.18.1
Quoted 6 lines · Jason Wang
Vhost_netwas known to suffer from HOL[1] issues which is not easy to fix. Several downstream disable the feature by default. What's more, the datapath was split and datacopy path got the support of batching and XDP support recently which makes it faster than zerocopy part for small packets transmission.
Quoted 7 lines
It looks to me that disable zerocopy by default is more appropriate. It cold be enabled by default again in the future if we fix the above issues.
[1] https://patchwork.kernel.org/patch/3787671/
Signed-off-by Jason Wang <[email protected]>
Quoted 19 lines
--- drivers/vhost/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/vhost/net.c +1
-1@@ -36,7 +36,7 @@#include "vhost.h" -static int experimental_zcopytx =1 ; +static int experimental_zcopytx =0 ;module_param (experimental_zcopytx, int,0444 );MODULE_PARM_DESC (experimental_zcopytx,"Enable Zero Copy TX;" " 1 -Enable; 0 - Disable" );Signature · 1 line
2.18.1
From: Jason Wang <[email protected]>
Date: Mon, 17 Jun 2019 05:20:54 -0400
Quoted 6 lines
Vhost_netwas known to suffer from HOL[1] issues which is not easy to fix. Several downstream disable the feature by default. What's more, the datapath was split and datacopy path got the support of batching and XDP support recently which makes it faster than zerocopy part for small packets transmission.
Quoted 7 lines
It looks to me that disable zerocopy by default is more appropriate. It cold be enabled by default again in the future if we fix the above issues.
[1] https://patchwork.kernel.org/patch/3787671/
Signed-off-by Jason Wang <[email protected]>
Applied, thanks Jason.
From: Jason Wang <[email protected]>
Date: Mon, 17 Jun 2019 05:20:54 -0400
Quoted 6 lines
Vhost_netwas known to suffer from HOL[1] issues which is not easy to fix. Several downstream disable the feature by default. What's more, the datapath was split and datacopy path got the support of batching and XDP support recently which makes it faster than zerocopy part for small packets transmission.
Quoted 7 lines
It looks to me that disable zerocopy by default is more appropriate. It cold be enabled by default again in the future if we fix the above issues.
[1] https://patchwork.kernel.org/patch/3787671/
Signed-off-by Jason Wang <[email protected]>
Applied, thanks Jason.