On Sun, 7 Jul 2019 at 20:39, Andrew Lunn [off-list ref] wrote:
On Sun, Jul 07, 2019 at 08:29:17PM +0300, Vladimir Oltean wrote:
quoted
From: Vinicius Costa Gomes <vinicius.gomes@intel.com>
This allows taprio to offload the schedule enforcement to capable
network cards, resulting in more precise windows and less CPU usage.
The important detail here is the difference between the gate_mask in
taprio and gate_mask for the network driver. For the driver, each bit
in gate_mask references a transmission queue: bit 0 for queue 0, bit 1
for queue 1, and so on. This is done so the driver doesn't need to
know about traffic classes.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Voon Weifeng <redacted>
Hi Vladimir
Your SOB is also needed here.
Andrew
Hi Andrew,
I thought I'd added it, but it looks like I edited the patch and then
overwrote it, instead of operating on the git history.
Next time!
-Vladimir