Details
- Reviewers
jhb - Commits
- rS313785: Use symbolic constants for OSC support / control negotiations.
Diff Detail
- Repository
- rS FreeBSD src repository - subversion
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
| sys/dev/pci/pcireg.h | ||
|---|---|---|
| 1046 ↗ | (On Diff #25193) | So this is not PCI specific. It's meaning is defined for all _OSC invocations regardless of GUID. If we want constants for this, I think they need to be in some ACPI header along with the bitmasks (the query flag before invoking _OSC and the various status flags in this word after it returns). |
| 1058 ↗ | (On Diff #25193) | This is about ownership of the single PCI-e capability (the PCI cap, not the extended ones above 0x100). We should probably check the firmware spec to find out if we need to set this before we start dinking around with any of those registers. |
| sys/dev/pci/pcireg.h | ||
|---|---|---|
| 1046 ↗ | (On Diff #25193) | Agreed. Didn't want to complicate things too much by adding this to ACPI. |
| 1058 ↗ | (On Diff #25193) | It was unclear to me what this bit means, given that the AER bit is separate, but AER is also mentioned here. Since we're not doing any of the other things, my forthcoming AER patches don't set it. But that's an excellent AI to chase down. |