Build error with custom afi
44 views
Skip to first unread message
Varun Gandhi
Feb 11, 2021, 1:26:01 PM2/11/21
to fir...@googlegroups.com
Hi,
When trying to build a custom afi, I’m getting the following error:
2021-02-09 12:47:17,982 [flush ] [DEBUG] [localhost] local: aws s3 cp 21_02_09-062239.Developer_CL.tar s3://firesim-734394535448/dcp/21_02_09-062239.Developer_CL.tar-192.168.0.193-EFYZKRGMPC.tar
2021-02-09 12:47:18,127 [flush ] [DEBUG] Fatal error: local() encountered an error (return code 1) while executing 'aws s3 cp 21_02_09-062239.Developer_CL.tar s3://firesim-734394535448/dcp/21_02_09-062239.Developer_CL.tar-192.168.0.193-EFYZKRGMPC.tar'
2021-02-09 12:47:18,127 [flush ] [DEBUG] Aborting.
2021-02-09 12:47:18,151 [<module> ] [ERROR] Fatal error.
Traceback (most recent call last):
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 334, in <module>
main(args)
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 269, in main
globals()[args.task](buildconfig)
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 149, in buildafi
execute(aws_build, globalbuildconf, hosts=globalbuildconf.get_build_instance_ips())
File "/usr/lib64/python2.7/site-packages/fabric/tasks.py", line 420, in execute
error(err)
File "/usr/lib64/python2.7/site-packages/fabric/utils.py", line 358, in error
return func(message)
File "/usr/lib64/python2.7/site-packages/fabric/utils.py", line 62, in abort
raise e
SystemExit: 1
Also, in the console log generated, I see the following message “build complete”…so it’s a bit contradictory.
[192.168.0.193] out: # if { [file exists $CL_DIR/build/checkpoints/to_aws/${timestamp}.Developer_CL.tar] } { ─
[192.168.0.193] out: # puts "Deleting old tar file with same name.";
[192.168.0.193] out: # file delete -force $CL_DIR/build/checkpoints/to_aws/${timestamp}.Developer_CL.tar
[192.168.0.193] out: # }
[192.168.0.193] out: # cd $CL_DIR/build/checkpoints
[192.168.0.193] out: # tar::create to_aws/${timestamp}.Developer_CL.tar [glob to_aws/${timestamp}*]
[192.168.0.193] out: # puts "AWS FPGA: ([clock format [clock seconds] -format %T]) - Finished creating final tar file in to_aws directo
ry.";
[192.168.0.193] out: AWS FPGA: (12:45:30) - Finished creating final tar file in to_aws directory.
[192.168.0.193] out: # if {[string compare $notify_via_sns "1"] == 0} {
[192.168.0.193] out: # puts "AWS FPGA: ([clock format [clock seconds] -format %T]) - Calling notification script to send e-mail to $e
nv(EMAIL)";
[192.168.0.193] out: # exec $env(AWS_FPGA_REPO_DIR)/shared/bin/scripts/notify_via_sns.py
[192.168.0.193] out: # }
[192.168.0.193] out: # puts "AWS FPGA: ([clock format [clock seconds] -format %T]) - Build complete.";
[192.168.0.193] out: AWS FPGA: (12:45:30) - Build complete.
[192.168.0.193] out: INFO: [Common 17-206] Exiting Vivado at Tue Feb 9 12:45:30 2021...
[192.168.0.193] out:
Fatal error: One or more hosts failed while executing task 'aws_build'
Fatal error.
Traceback (most recent call last):
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 334, in <module>
main(args)
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 269, in main
globals()[args.task](buildconfig)
File "/home/centos/chipyard/sims/firesim/deploy/firesim", line 149, in buildafi
execute(aws_build, globalbuildconf, hosts=globalbuildconf.get_build_instance_ips())
File "/usr/lib64/python2.7/site-packages/fabric/tasks.py", line 420, in execute
error(err)
File "/usr/lib64/python2.7/site-packages/fabric/utils.py", line 358, in error
return func(message)
File "/usr/lib64/python2.7/site-packages/fabric/utils.py", line 62, in abort
raise e
SystemExit: 1
Best,
Varun
David Biancolin
Feb 11, 2021, 2:38:40 PM2/11/21
to FireSim
Hey Varun,
This is a little unintuitive, but there are two phases of building the AFI. The first part builds most of the design (including the FireSim part), the second stage involes uploading this partition to s3 so that it can be linked with an amazon provided partition, and the final bitstream can be generated. The first stage completely successfully.
What seems to have failed was upooading the checkpoint to s3:
Can you try to upload a file manually using the command above (with a file that exists). And can you confirm that the bucket exists in your aws console?
This is a little unintuitive, but there are two phases of building the AFI. The first part builds most of the design (including the FireSim part), the second stage involes uploading this partition to s3 so that it can be linked with an amazon provided partition, and the final bitstream can be generated. The first stage completely successfully.
What seems to have failed was upooading the checkpoint to s3:
2021-02-09 12:47:17,982 [flush ] [DEBUG] [localhost] local: aws s3 cp 21_02_09-062239.Developer_CL.tar s3://firesim-734394535448/dcp/21_02_09-062239.Developer_CL.tar-192.168.0.193-EFYZKRGMPC.tar
2021-02-09 12:47:18,127 [flush ] [DEBUG] Fatal error: local() encountered an error (return code 1) while executing 'aws s3 cp 21_02_09-062239.Developer_CL.tar s3://firesim-734394535448/dcp/21_02_09-062239.Developer_CL.tar-192.168.0.193-EFYZKRGMPC.tar' Reply all
Reply to author
Forward
0 new messages