On Mon, 1 Jun 2020, Suravee Suthikulpanit wrote: > > Moving init_iommu_perf_ctr just after iommu_flush_all_caches resolves > > the issue. This is the earliest point in amd_iommu_init_pci where the > > call succeeds on my laptop. > > According to your description, it should just need to be anywhere after the > pci_enable_device() is called for the IOMMU device, isn't it? So, on your > system, what if we just move the init_iommu_perf_ctr() here:
No, this doesn't work, as I already said in the paragraph you are responding to. See my last sentence in the quoted part. So the implication is init_device_table_dma together with subsequent cache flush is also setting up something that is necessary for counters to be writable. Alexander
