Bug 1898074
| Summary: | [nodedev-dumpxml] improve css cap property naming | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux Advanced Virtualization | Reporter: | smitterl |
| Component: | libvirt | Assignee: | Thomas Huth <thuth> |
| Status: | CLOSED UPSTREAM | QA Contact: | smitterl |
| Severity: | low | Docs Contact: | |
| Priority: | low | ||
| Version: | 8.4 | CC: | berrange, cohuck, dzheng, jsuchane, mprivozn, thuth, virt-maint |
| Target Milestone: | rc | Keywords: | Triaged, Upstream |
| Target Release: | 8.5 | ||
| Hardware: | s390x | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-02-02 12:08:52 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: | |||
Daniel, do you think we could still the output of nodedev-dumpxml here without breaking the compatibility promise of libvirt? (In reply to Thomas Huth from comment #1) > Daniel, do you think we could still the output of nodedev-dumpxml here > without breaking the compatibility promise of libvirt? Afraid not, the best we can do is improve documentation. If I've got Cornelia's blog right (https://virtualpenguins.blogspot.com/2017/03/channel-io-whats-in-channel-subsystem.html), the third value here is not the "schid" itself anyway, but the subchannel number. All three values, cssid, ssid and subchannel number can then be used to create a subchannel identification word ("schid"). Anyway, I'll try to come up with a patch for libvirt that improves the documentation here a little bit. (In reply to Thomas Huth from comment #3) > If I've got Cornelia's blog right > (https://virtualpenguins.blogspot.com/2017/03/channel-io-whats-in-channel- > subsystem.html), the third value here is not the "schid" itself anyway, but > the subchannel number. All three values, cssid, ssid and subchannel number > can then be used to create a subchannel identification word ("schid"). Yes, that's right (also see the PoP pg 14-1 for how the schid is constructed, even though it does not mention the cssid). Merged upstream as: b943099c87 docs: Clarify the documentation of the <css> elements v7.0.0-185-gb943099c87 Thanks everybody. I checked on upstream https://libvirt.org/formatnode.html |
Description of problem: As css ids and ccw ids have the same format it can become confusing. Proposal: Use "schid" instead of "devno" for subchannel id. Version-Release number of selected component (if applicable): libvirt-6.0.0-30.module+el8.4.0+8705+34397d87.s390x How reproducible: 100% Steps to Reproduce: 1. Confirm that the host has ccw devices, virsh nodedev-list should list for example ccw_0_0_570d and it's corresponding channel css_0_0_25ac. 2. virsh nodedev-dumpxml css_0_0_25ac. Actual results: <device> ... <capability type='css'> <cssid>0x0</cssid> <ssid>0x0</ssid> <devno>0x25ac</devno> </capability> </device> Expected results: <device> ... <capability type='css'> <cssid>0x0</cssid> <ssid>0x0</ssid> <schid>0x25ac</schid> </capability> </device> Additional info: By [2] "devno" is for the actual device behind the subchannel while "schid" is for example used in kernel s390 drivers [1]. [1] https://www.kernel.org/doc/html/v5.8/driver-api/s390-drivers.html [2] https://www.ibm.com/support/knowledgecenter/en/linuxonibm/com.ibm.linux.z.lkdd/lkdd_r_sys_view_subchan.html