On 8 Feb 2023, at 17:50, Simon Horman wrote:

> On Tue, Feb 07, 2023 at 03:07:24PM +0100, Eelco Chaudron wrote:
>> With some datapaths, read TC, it takes a bit longer to update the
>> OpenFlow statistics. Rather than adding an additional delay, try
>> to read the counters multiple times until we get the desired value.
>>
>> Signed-off-by: Eelco Chaudron <[email protected]>
>> Acked-by: Roi Dayan <[email protected]>
>
> Hi Eelco,
>
> I'm not sure how much this buys us.
> I notice that in the following patch the tests are disabled
> (maybe I am reading that wrong) as they are not reliable.
> And, FWIIW, in my environment, I couldn't get these tests to pass
> at all (17 iterations).
>
> I'd be happy to set up a Fedora, or other environment, to verify
> this patch if you think it's worthwhile.

I think it’s fine for now. For me, it failed every now and then on 50 runs. I 
decided to keep them out in the final patch to avoid re-run errors. As I’ve 
seen that, if it fails in the first run, the changes are high it will fail in 
the re-run. But I did not find any clue yet why.

I have some revalidator patches pending upstreanm, and my plan is to revisit 
these two errors once those changes are in.

> Lastly, just to clarify my response to other patches.
> The tests were for the SW TC datapath.

Thanks for doing this!

Cheers,

Eelco

_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to