Re: [PATCH net-next] net: ethtool: Don't check if RSS context exists in case of context 0

From
Edward Cree <ecree.xilinx@gmail.com>
Date
2025-02-27 19:54:20
Reply to
Jakub Kicinski, 2025-02-27
On 27/02/2025 16:56, Jakub Kicinski wrote:
On Thu, 27 Feb 2025 16:24:52 +0000 Edward Cree wrote:
quoted
quoted
I never uttered the thought that lead me to opposing. 
ctx 0 is a poor man's pass / accept. If someone needs a pass we should
add an explicit "action pass".  
To me 'pass' is just a shorthand for whatever specific behaviour
 happens to match the default.  I.e. if you can already express
 that behaviour directly, then 'pass' is a strictly nonorthogonal
 addition and therefore bad interface design.
I presume sfc matches only one rule. For devices which terminate on
first hit and rules are ordered by ntuple ID, the following rules*:

 src-ip 192.168.0.2                 context 0
 src-ip 192.0.0.0   m 0.255.255.255 action -1

implement allowing only 192.168.0.2 out of the 192.0.0.0/8 subnet. 
Ah, so the point is that "pass" wasn't expressible before RSS contexts
 came along, and may therefore be surprising and new.  I get you now.
The device may not even support RSS contexts.
I think in that case the driver will reject the first filter because it
 doesn't recognise flow_type == FLOW_RSS | WHATEVER_FLOW.  Unless the
 driver writer has deliberately added support for RSS filters despite
 only having context 0, which seems like a signal they know it means
 'pass' and explicitly wanted that.
Thread (16 messages) 16 messages, 4 authors, last reply 2025-03-04
Keyboard shortcuts
hback out one level
jnext message in thread / next series-diff position
kprevious message in thread / previous series-diff position
ldrill in
Escclose help / fold thread tree
?toggle this help