Page MenuHomeFreeBSD

hyperv vmbus_pcib: Use pci_domain_*_bus for PCI_RES_BUS resources
ClosedPublic

Authored by jhb on Jan 12 2024, 10:32 PM.
Tags
None
Referenced Files
Unknown Object (File)
Tue, Dec 9, 9:37 PM
Unknown Object (File)
Mon, Dec 8, 3:32 AM
Unknown Object (File)
Sun, Dec 7, 11:15 PM
Unknown Object (File)
Sat, Dec 6, 10:58 PM
Unknown Object (File)
Sat, Dec 6, 12:27 AM
Unknown Object (File)
Sat, Nov 29, 12:33 AM
Unknown Object (File)
Fri, Nov 28, 8:49 AM
Unknown Object (File)
Tue, Nov 25, 10:21 AM
Subscribers
None

Diff Detail

Repository
rG FreeBSD src repository
Lint
Lint Skipped
Unit
Tests Skipped
Build Status
Buildable 55405
Build 52294: arc lint + arc unit

Event Timeline

jhb requested review of this revision.Jan 12 2024, 10:32 PM
jhb created this revision.
sys/dev/hyperv/pcib/vmbus_pcib.c
2011

Please check the indentation here.

sys/dev/hyperv/pcib/vmbus_pcib.c
2011

The raw version looks fine. I think the previous version was not fully consistent. Here is the patched method table:

	/* Bus interface */
	DEVMETHOD(bus_read_ivar,		vmbus_pcib_read_ivar),
	DEVMETHOD(bus_write_ivar,		vmbus_pcib_write_ivar),
	DEVMETHOD(bus_alloc_resource,		vmbus_pcib_alloc_resource),
	DEVMETHOD(bus_adjust_resource,		vmbus_pcib_adjust_resource),
	DEVMETHOD(bus_release_resource,		vmbus_pcib_release_resource),
	DEVMETHOD(bus_activate_resource,   	vmbus_pcib_activate_resource),
	DEVMETHOD(bus_deactivate_resource, 	vmbus_pcib_deactivate_resource),
	DEVMETHOD(bus_setup_intr,	   bus_generic_setup_intr),
	DEVMETHOD(bus_teardown_intr,	   bus_generic_teardown_intr),
	DEVMETHOD(bus_get_cpus,			vmbus_pcib_get_cpus),
This revision was not accepted when it landed; it landed in state Needs Review.Jan 23 2024, 5:46 PM
This revision was automatically updated to reflect the committed changes.

Code looks fine to me, but has the new functions been tested on Azure ? As in local Hyper-V we mostly dont use pci passthrough.

Code looks fine to me, but has the new functions been tested on Azure ? As in local Hyper-V we mostly dont use pci passthrough.

I do not have a means to test it, no. It is part of a larger API sweep/cleanup series.