On Thu, Oct 30, 2014 at 09:55:35PM -0400, Ted Unangst wrote:
On Thu, Oct 30, 2014 at 22:10, Alexander Bluhm wrote:
+
+ /* Avoid user land starvation. */
+ yield();
I think this is the responsibility of the taskq thread, not the
individual task.
I am not sure about this. Without an
On 4 Nov 2014, at 09:50, Alexander Bluhm alexander.bl...@gmx.net wrote:
On Thu, Oct 30, 2014 at 09:55:35PM -0400, Ted Unangst wrote:
On Thu, Oct 30, 2014 at 22:10, Alexander Bluhm wrote:
+
+ /* Avoid user land starvation. */
+ yield();
I think this is the responsibility of the
On 31 Oct 2014, at 22:37, Alexander Bluhm alexander.bl...@gmx.net wrote:
On Fri, Oct 31, 2014 at 02:50:00PM +1000, David Gwynne wrote:
so without splicing, the payloads from multiple tcp packets (at least all of
the ones in a single softnet run?) get bundled up into a buffer that
On Fri, Oct 31, 2014 at 02:50:00PM +1000, David Gwynne wrote:
so without splicing, the payloads from multiple tcp packets (at least all of
the ones in a single softnet run?) get bundled up into a buffer that userland
reads and then writes out again in a single go. right?
you're suggesting
Hi,
Some performance measurements showed that socket splicing for TCP
can be made faster. The main slowdown was that tcp_output() got
called for every incomming packet. When copying through user-land
this cannot happen as the scheduler gets involved.
So my idea is to do the socket splicing for
On Thu, Oct 30, 2014 at 22:10, Alexander Bluhm wrote:
+
+ /* Avoid user land starvation. */
+ yield();
I think this is the responsibility of the taskq thread, not the
individual task.
Regarding the second pool diff, that looks very reasonable. Quite a
savings for a feature many
On 31 Oct 2014, at 07:10, Alexander Bluhm alexander.bl...@gmx.net wrote:
Hi,
Some performance measurements showed that socket splicing for TCP
can be made faster. The main slowdown was that tcp_output() got
called for every incomming packet. When copying through user-land
this cannot