[v6,0/8] AMD IOMMU: further improvements

Message

Jan Beulich Sept. 19, 2019, 1:08 p.m. UTC

The main goal continues to be to reduce the huge memory overhead
that we've noticed. On the way there a number of other things were
once again noticed. All patches have now been tested on a Fam15
and a Fam17 system.

01: don't blindly allocate interrupt remapping tables
02: make phantom functions share interrupt remapping tables
03: x86/PCI: read maximum MSI vector count early
04: replace INTREMAP_ENTRIES
05: restrict interrupt remapping table sizes
06: tidy struct ivrs_mappings
07: allocate one device table per PCI segment
08: pre-fill all DTEs right after table allocation

Jan