[linux-6.6.y] Zhaoxin: Fix DMA RW sequence disorder issue #371
Conversation
Identify kh-40000 platforms by specific PCI device's version number. Signed-off-by: leoliu-oc <[email protected]>
Add 'kh40000_direct_dma_ops' to replace 'direct_dma_ops' for KH-40000 platform. For coherent DMA access, memory can be allocated only from the memory node of the node where the device resides. For streaming DMA access, add a PCI read operation at the end of DMA access. Signed-off-by: leoliu-oc <[email protected]>
Add 'kh40000_iommu_dma_ops' to replace 'intel_dma_ops' for KH-40000 platform. For coherent DMA access, memory can be allocated only from the memory node of the node where the device resides. For streaming DMA access, add a PCI read operation at the end of DMA access. Signed-off-by: leoliu-oc <[email protected]>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
Hi @leoliu-oc. Thanks for your PR. I'm waiting for a deepin-community member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
3 participants
The Zhaoxin KH-40000 platform can't keep the PCIE transaction order for DMA writes, whose target addresses are located in different NUMA nodes, from the same device. Patch this issue by flushing the target DMA write with a subsequent PCIE configuration space read operation.