Re: [PATCH 00/15] Device Evidence and Trust for PCI Security Protocol (TDISP)

Ankit Agrawal <[email protected]> Sun, 2 Aug 2026 18:12:09 +0000
Newsgroups dev.linux.lists.driver-core,dev.linux.lists.linux-coco,org.kernel.vger.linux-pci
Message-ID <SAWPR12MB9992676CE5F9BC9CAA15591536B0D62@SAWPR12MB999267.namprd12.prod.outlook.com>
> The direction appears to be set, imho the immediate TODO list is:=0A=
> - address "not-netlink-y enough" concerns=0A=
> - teach pci_iomap() to map mixed private+shared BARs (ones which contain =
MSIX for the starter)=0A=
> - always call pci_tsm_ops::enable_dma/disable_dma=0A=
> - ditch the module trust parameter?=0A=
> - add a knob to enable/disable DMA and keep TDI in RUN=0A=
=0A=
Agreed, couple of items I'll add.=0A=
=0A=
- Donald's suggestion for subsystem name like spdm, tsm or tsm-device (uapi=
-breaking)=0A=
- Fix bugs such as missing pci_tsm_disable_dma() in pci_dma_cleanup()=