script - jjin-choi/study_note GitHub Wiki

Q: What happens when dvlr start runs? Many uses wonder why it takes so long time for simulate the testcases. A DVA job proceeds in the following order.

• sync the files that you submitted with -l option
• execute the script by getting -pre option.
• compile for all of design units. If you entered multiple design units with de option, simulation doesn’t start until all of compiles are completed. Also, even if one of compile has been failed, simulation step would be skipped.
• If all of design unit ended with pass, then simulation gets started. It takes roughly 10 minutes to reflect results in dvlr show/report results.

Q: How do I get started DVLR when I have no permission? When you encounter these kinds of errors, it means there is no DVLR system workspace for your account. It could be happened for three reasons. First, you might make a typo in your ID so please check the ID and version. If the command is correct, the email address on Attribute file or the information of toss database might be incorrect. In this case, please make sure whether the email address registered in toss database is same as the one of attribute file. You could refer to the guideline on how to check and how to update the toss information on sWiki. Or this error could occur if you have multiple VWP IDs. When creating system workspaces by new-label command, the workspaces are mapped to certain ID for each user. You can check the result of the configuration on the web site. We’re looking for the ways how can be provided the configuration by command line interface. If your issue is not covered by any of them I mentioned, then please contact AT part directly. To have the issue resolved as quickly as possible, please provide the following information.

Q: I get an error message like the below when using dvlr submit with -l option. Some errors may occur during dvlr submit, too. Please make sure that your current workspace has that files.
Note that this error means that your system workspace has the same revision of files with your current workspace, so no files to submit. If you want to find out the difference between your local workspace and system workspace, please use dvlr diff command.

Q: Why can’t I start another DVLR job? Users are not allowed to run multiple jobs in one system workspace. Please wait for running jobs such as sanity check, compile, elaboration to be finished and try again. Note that vManager simulation are not under consideration of running jobs. Normally, you could resolve this error by only 2 steps. You can get job status and DVA JOB ID that is still running by dvlr show. If you want to discard the job and execute it again, please use dvlr flush.

Q: What’s the difference between dvlr report and dvlr show? dvlr show command literally show all the historical results of testcases. Therefore, testcases once failed, then the result of dvlr show could remain failed if the name, or owner, or attribute status has been changed. On the other hand, dvlr report command provides you the latest result of testcases reflecting the most up-to-date version of attribute file that has been updated by dvlr attr.

Q. Why are my jobs still failed even if I submit/start DVLR job? Each time a rerun is performed, a log file is generated in a different directory. So please check log file has been written on new directory by dvlr report command, and that means the testcase has been simulated. You could check the status of testcase on VM_RUN_STATUS file on the same directory with log file. If the testcase has been passed but dvlr report said it’s still failed or running, it is highly possible that DVLR database doesn’t match with vManager. It may take 10 minute to update status on DVLR database.

Q. Why are my jobs still failed even if I submit/start DVLR job? Other issues could be happened because of vManager, LSF, and DVLR daemon. We’re constantly working to ensure a robust automation system, and make DVLR even more effective and efficient for you. We’re on development to detect different kinds of errors and to handle these errors. And it will be provided through CLI command as soon as possible