Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.
The FDP team is no longer accepting new bugs in Bugzilla. Please report your issues under FDP project in Jira. Thanks.

Bug 1783763

Summary: Unable to run the testpmd program using dpdk-18.11.2 package
Product: Red Hat Enterprise Linux Fast Datapath Reporter: Sebastian Scheinkman <sscheink>
Component: DPDKAssignee: Jens Freimann <jfreiman>
DPDK sub component: other QA Contact: liting <tli>
Status: CLOSED DUPLICATE Docs Contact:
Severity: unspecified    
Priority: unspecified CC: amorenoz, atragler, augol, ctrautma, dmarchan, jfreiman, jhsiao, ktraynor, mmirecki
Version: RHEL 8.0   
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2020-01-13 12:54:51 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1771572    

Description Sebastian Scheinkman 2019-12-15 12:26:39 UTC
Description of problem:

Unable to run the testpmd program using dpdk-18.11.2

testpmd -l 70,72,74,76,78 -w 0000:19:00.5 -- -i --portmask=0x1 --nb-cores=2 --forward-mode=macswap --port-topology=loop
EAL: Detected 80 lcore(s)
EAL: Detected 2 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Probing VFIO support...
EAL: Cannot obtain physical addresses: No such file or directory. Only vfio will function.
error allocating rte services array
EAL: FATAL: rte_service_init() failed
EAL: rte_service_init() failed
PANIC in main():
Cannot init EAL
5: [testpmd(_start+0x2e) [0x56436aae6b5e]]
4: [/lib64/libc.so.6(__libc_start_main+0xf3) [0x7f7bde26c873]]
3: [testpmd(+0x4ddf1) [0x56436aae5df1]]
2: [/lib64/librte_eal.so.9(__rte_panic+0xc1) [0x7f7bdf4721e6]]
1: [/lib64/librte_eal.so.9(rte_dump_stack+0x32) [0x7f7bdf47ee32]]
Aborted (core dumped)


Version-Release number of selected component (if applicable):


How reproducible:
100%

Steps to Reproduce:
1. install latest dpdk from red hat Fast Datapath repo
2. run the testpmd program

Actual results:
core dump

Expected results:
be able to run the application

Additional info:

Comment 1 Adrián Moreno 2019-12-16 09:23:45 UTC
Hi Sebastian,
It seems your application is defaulting to IOVA_PA mode. Can you please check:
- You have enough permissions to access physical addresses (i.e: running as root)?
- The issue is not reproducible with "--iova-mode va" in the EAL parameters
- There is no other dpdk application (including ovs) running in the system.
- Does it always fail in the same point? i.e: just after "error allocating rte services array"

Thanks.

Comment 2 Marcin Mirecki 2019-12-17 08:30:50 UTC
Let me add more information on the issue.
- we are running on an openshift environment with mellanox sriov nic's.
- dpdk is running inside a container, build from ubi8, with the dpdk (downstream) package installed on it.
- inside the container we are working on a VF from the sriov nic, passed to the container by means of the Sriov CNI
- we are using the mlx5_core driver for the nic.

We also tried this with a custom build upstream version of dpdk, and it works fine.

Comment 3 David Marchand 2020-01-07 12:32:58 UTC
It should be a fairly similar issue to bz1785933.
I put a scratch build of the 18.11 dpdk package in it, can you have a try?
Thanks.

Comment 4 David Marchand 2020-01-13 10:23:33 UTC
Sebastian, Marcin,

Can you confirm this is the same issue than bz1785933?
Thanks.

Comment 5 Sebastian Scheinkman 2020-01-13 12:20:39 UTC
Hi David,

Yes, this is the same issue.
I was able to run the container without the privileged=true using the scratch build

Comment 6 Jens Freimann 2020-01-13 12:54:51 UTC

*** This bug has been marked as a duplicate of bug 1785933 ***

Comment 7 Marcin Mirecki 2023-01-23 09:09:30 UTC
Closed, clearing needinfo