Email based retests for the Loongarch lab

[Patrick Robb]

Hi Min Zhou,

Thanks for the response and for getting some retest support online! I agree
the next reasonable steps are to continue to align out processes, in part
so that the lab behavior is predictable according to what is documented on
dpdk.org.

On Sat, Jun 7, 2025 at 8:50 PM zhoumin <zhoumin at loongson.cn> wrote:

> Yes, we don't support the rebase argument. Meanwhile, for the sake of
> simplicity, the rechecking behavior at Loongson lab is a little different
> from the default behavior. we recheck the patches on the latest HEAD of the
> branch selected by pw_maintainers_cli.py script. This behavior is same to
> that of the first testing for the commit at Loongson lab. It's ok to be
> consistent with the defualt behavior. Maybe we can support both it and
> rebase argument after you have synced some infrastructures.
>
Good point, we need to make sure the support for the rebase={branch}
argument is integrated into the get_reruns.py script in dpdk-ci. The UNH
team made that addition and it is available for patch review:
https://patchwork.dpdk.org/project/ci/patch/20250611205849.72165-1-probb@iol.unh.edu/

Maybe you can apply it, give it a run and add a tested by tag to the patch
if it is working for you? In order to make sure it was working I submitted
a placeholder loongson recheck request with the rebase argument (requesting
the main branch) included on the DPDK mailing list. Right now if I run the
script with the following args it produces this json:

python3 tools/get_reruns.py -ts 2025-06-10 --contexts loongarch-compilation

output json:

{
    "retests": {
        "34851": {
            "contexts": [
                "loongarch-compilation"
            ],
            "arguments": {
                "rebase": "main"
            }
        }
    },
    "last_comment_timestamp": "2025-06-11T20:29:39.327426"
}
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mails.dpdk.org/archives/ci/attachments/20250611/05715d43/attachment.htm>