[Snappi] Validate PFC frames should not be counted as RX_OK, TX_OK, RX_DRP, TX_DRP#9775
Conversation
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
| logger = logging.getLogger(__name__) | ||
|
|
||
| SNAPPI_POLL_DELAY_SEC = 2 | ||
| CONTINUOUS_MODE = -5 |
There was a problem hiding this comment.
What is continuous mode? What is the meaning of 5?
There was a problem hiding this comment.
It's an arbitrary number that isn't used anywhere else - continuous mode means the traffic flows constantly without any fixed duration or fixed packets at some percentage rate
| """ | ||
| Verify correct frame count for tx frames on the DUT | ||
| (OK and DROPS) when the traffic is expected to be paused on the DUT. | ||
| DUT is polled after it stops receiving PFC pause frames from TGEN. |
There was a problem hiding this comment.
When you poll DUT, has DUT drained all the packets in the buffer?
| .format(prios[prio])) | ||
| else: | ||
| # PFC pause frames should not be transmitted when test traffic is not paused | ||
| pytest_assert(pfc_pause_tx_frames <= PFC_FRAME_COUNT_TOL, |
There was a problem hiding this comment.
have you ever seen pfc_pause_tx_frames >0 in this case?
There was a problem hiding this comment.
No - this is just there as a back stopper. If you want me to lower the threshold to 2 or 3, that is also possible. It is currently set to 5.
There was a problem hiding this comment.
Reset to 0
Can you please remove PFC_FRAME_COUNT_TOL?
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
| .format(prios[prio])) | ||
| else: | ||
| # PFC pause frames should not be transmitted when test traffic is not paused | ||
| pytest_assert(pfc_pause_tx_frames <= PFC_FRAME_COUNT_TOL, |
There was a problem hiding this comment.
Reset to 0
Can you please remove PFC_FRAME_COUNT_TOL?
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
…nter polling (#9847) Summary: Currently, all verification for PFC pause frames are done by polling the ixia device at the end of the test. We need to add functionality to poll the appropriate counters on the DUT while the DUT continues to receive PFC pause frames from the neighboring device. This PR is dependent on #9775. Fixes (issue) #9843
|
The pre-commit check detected issues in the files touched by this pull request. Detailed pre-commit check results: To run the pre-commit checks locally, you can follow below steps:
|
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
@developfast PR conflicts with 202012 branch |
|
@developfast PR conflicts with 202305 branch |
|
@developfast PR conflicts with 202205 branch |
…nter polling (sonic-net#9847) Summary: Currently, all verification for PFC pause frames are done by polling the ixia device at the end of the test. We need to add functionality to poll the appropriate counters on the DUT while the DUT continues to receive PFC pause frames from the neighboring device. This PR is dependent on sonic-net#9775. Fixes (issue) sonic-net#9843
…nter polling (#9847) Summary: Currently, all verification for PFC pause frames are done by polling the ixia device at the end of the test. We need to add functionality to poll the appropriate counters on the DUT while the DUT continues to receive PFC pause frames from the neighboring device. This PR is dependent on #9775. Fixes (issue) #9843
Description of PR
Summary: While receiving or sending out PFC frames, the PFC counter should be the only counter increasing on the device. The other counters are not expected to increase. The regular TX and RX counters are also polled to ensure they do not increment by the PFC count.
Fixes (issue) #6638
Type of change
Back port request
Approach
What is the motivation for this PR?
PFC validation and behavior on the DUT
How did you do it?
How did you verify/test it?
Tested on lab device
Any platform specific information?
Supported testbed topology if it's a new test case?
Documentation