Page MenuHomeFreeBSD

Force PCI bus interrupt rerouting
ClosedPublic

Authored by bz on Feb 29 2016, 4:42 PM.
Tags
None
Referenced Files
Unknown Object (File)
Nov 17 2024, 4:53 AM
Unknown Object (File)
Nov 17 2024, 4:52 AM
Unknown Object (File)
Nov 4 2024, 4:37 AM
Unknown Object (File)
Sep 29 2024, 1:45 AM
Unknown Object (File)
Sep 7 2024, 1:45 AM
Unknown Object (File)
Jun 21 2024, 8:28 PM
Unknown Object (File)
Jun 15 2024, 11:44 PM
Unknown Object (File)
May 21 2024, 9:40 AM
Subscribers

Details

Summary

Seems to be needed for at least gem5. Someone should explain this and see if it's correct to add it or not.

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

bz retitled this revision from to Force PCI bus interrupt rerouting.
bz updated this object.
bz edited the test plan for this revision. (Show Details)
bz added reviewers: andrew, jhb.
bz set the repository for this revision to rS FreeBSD src repository - subversion.
andrew edited edge metadata.
andrew added inline comments.
sys/arm64/include/param.h
45 ↗(On Diff #13887)

Should be #define<tab>__PCI...

This revision is now accepted and ready to land.Mar 1 2016, 12:02 PM
sys/arm64/include/param.h
45 ↗(On Diff #13887)

ACK, will check before committing.

In theory all platforms should do this if their PCI bridge drivers know how to manage routing. Note also that this only matters for legacy INTx interrupts, not MSI.

In D5494#117542, @jhb wrote:

In theory all platforms should do this if their PCI bridge drivers know how to manage routing. Note also that this only matters for legacy INTx interrupts, not MSI.

I think we missed this because we've always used MSI/MSI-X interrupts on arm64.

This revision was automatically updated to reflect the committed changes.