Warn at XDP parallel start if the NIC firmware LLDP agent is active - #223
Open
jason-sirius wants to merge 1 commit into
Open
Warn at XDP parallel start if the NIC firmware LLDP agent is active#223jason-sirius wants to merge 1 commit into
jason-sirius wants to merge 1 commit into
Conversation
On Intel i40e/X710 NICs the firmware LLDP agent applies a default FCoE-aware DCB configuration that splits the queue pairs into two traffic classes (e.g. TC0 = queues 0-7, TC1 = queues 8-15). Hardware RSS only distributes within a traffic class, so the upper half of the receive queues -- and the libtrace processing threads reading them -- silently receive no packets. Every software-visible setting (channel count, RSS indirection table, sysfs queues) looks correct throughout, which makes this painful to diagnose: we lost several days to it and only found it via the driver's debugfs VSI dump (numtc=2). Add linux_get_nic_fw_lldp_enabled(), which looks up the "disable-fw-lldp" private flag by name via ETHTOOL_GDRVINFO / ETHTOOL_GSTRINGS / ETHTOOL_GPFLAGS (driver-agnostic; returns -1 when the driver has no such flag), and warn from linux_xdp_pstart_input() with the one-line remedy (ethtool --set-priv-flags <if> disable-fw-lldp on, a persistent NVM write).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
On Intel i40e/X710 NICs the firmware LLDP agent applies a default FCoE-aware DCB configuration that splits the queue pairs into two traffic classes (e.g. TC0 = queues 0-7, TC1 = queues 8-15). Hardware RSS only distributes within a traffic class, so the upper half of the receive queues, and the
libtraceprocessing threads reading them, silently receive no packets. Every software-visible setting (channel count, RSS indirection table, sysfs queues) looks correct throughout, which makes this painful to diagnose: we lost several days to it and only found it via the driver's debugfs VSI dump (numtc=2).Add
linux_get_nic_fw_lldp_enabled(), which looks up the "disable-fw-lldp" private flag by name viaETHTOOL_GDRVINFO/ETHTOOL_GSTRINGS /ETHTOOL_GPFLAGS(driver-agnostic; returns -1 when the driver has no such flag), and warn fromlinux_xdp_pstart_input()with the one-line remedy (ethtool --set-priv-flags <if> disable-fw-lldpon, a persistent NVM write).