On 11/22/23 17:09, David Marchand wrote: > The DPDK unit test only runs if vfio or igb_uio kernel modules are loaded: > on systems with only mlx5, this test is always skipped. > > Besides, the test tries to grab the first device listed by dpdk-devbind.py, > regardless of the PCI device status regarding kmod binding. > > Remove dependency on this DPDK script and use a minimal script that > reads PCI sysfs. > > This script is not perfect, as one can imagine PCI devices bound to > vfio-pci for virtual machines. > Plus, this script only tries to take over vfio-pci devices. mlx5 devices > can't be taken over blindly as it could mean losing connectivity to the > machine if the netdev was in use for this system. > > For those two reasons, add a new environment variable DPDK_PCI_ADDR for > testers to select the PCI device of their liking. > For consistency and grep, the temporary file PCI_ADDR is renamed > to DPDK_PCI_ADDR. > > Reviewed-by: Maxime Coquelin <maxime.coque...@redhat.com> > Acked-by: Eelco Chaudron <echau...@redhat.com> > Signed-off-by: David Marchand <david.march...@redhat.com> > --- > Changes since v4: > - separated from the original series, > - rebased, > - dropped mlx5 devices from the discovery script, > - documented DPDK_PCI_ADDR env variable, > > Changes since v3: > - fixed nit from Maxime, > > Changes since v2: > - sorted logs alphabetically, > > --- > Documentation/topics/testing.rst | 11 ++++++--- > tests/automake.mk | 1 + > tests/system-dpdk-find-device.py | 39 ++++++++++++++++++++++++++++++++ > tests/system-dpdk-macros.at | 10 ++------ > tests/system-dpdk.at | 14 ++++++------ > 5 files changed, 57 insertions(+), 18 deletions(-) > create mode 100755 tests/system-dpdk-find-device.py > > diff --git a/Documentation/topics/testing.rst > b/Documentation/topics/testing.rst > index 5f6940b84d..fb9b3e77b1 100644 > --- a/Documentation/topics/testing.rst > +++ b/Documentation/topics/testing.rst > @@ -343,15 +343,20 @@ To see a list of all the available tests, run:: > > These tests support a `DPDK supported NIC`_. The tests operate on a wider > set of > environments, for instance, when a virtual port is used. > -They do require proper DPDK variables (``DPDK_DIR`` and ``DPDK_BUILD``). > Moreover you need to have root privileges to load the required modules and > to bind > -the NIC to the DPDK-compatible driver. > +a PCI device to the DPDK-compatible driver. > > .. _DPDK supported NIC: https://core.dpdk.org/supported/#nics > > +The phy test will skip if no suitable PCI device is found. > +It is possible to select which PCI device is used for this test by setting > the > +DPDK_PCI_ADDR environment variable, which is especially useful when testing > +with a mlx5 device:: > + > + # DPDK_PCI_ADDR=0000:82:00.0 make check-dpdk > + > All tests are skipped if no hugepages are configured. User must look into > the DPDK > manual to figure out how to `Configure hugepages`_. > -The phy test will skip if no compatible physical device is available. > > .. _Configure hugepages: > https://doc.dpdk.org/guides-22.11/linux_gsg/sys_reqs.html > > diff --git a/tests/automake.mk b/tests/automake.mk > index f8a925012d..69117514cc 100644 > --- a/tests/automake.mk > +++ b/tests/automake.mk > @@ -192,6 +192,7 @@ SYSTEM_OFFLOADS_TESTSUITE_AT = \ > > SYSTEM_DPDK_TESTSUITE_AT = \ > tests/system-common-macros.at \ > + tests/system-dpdk-find-device.py \
I didn't review this version in full, but I think, this should go to CHECK_PYFILES instead. The *_AT macros are for .at files. Maybe Kevin can make this change before applying. Assuming the flake8-check will pass. > tests/system-dpdk-macros.at \ > tests/system-dpdk-testsuite.at \ > tests/system-dpdk.at > diff --git a/tests/system-dpdk-find-device.py > b/tests/system-dpdk-find-device.py > new file mode 100755 > index 0000000000..b0997380bc > --- /dev/null > +++ b/tests/system-dpdk-find-device.py > @@ -0,0 +1,39 @@ > +#!/usr/bin/env python3 > +# Copyright (c) 2021 Red Hat, Inc. 2024, I guess? :) Best regards, Ilya Maximets. _______________________________________________ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev
