Hi As TDX support has been merged into linux-next, we thought it's the time to do a new round of regression testing on normal VM and it might be useful to share the results on the list. This time, we tested 2 configurations to increase the test coverage: 1) Disable TDX in BIOS and kernel Kconfig 2) Enable TDX in BIOS and initialize TDX successfully on host. The test results of these 2 configurations are the same. During the testing, we encountered 2 PMU bugs[0][1], which are not related to TDX patches. No other new issues were hit. Details -------------------------- Test Environment CPU: Granite Rapids Kernel: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git QEMU: https://github.com/intel-staging/qemu-tdx/releases/tag/tdx-qemu-upstream-v8 OVMF: https://github.com/tianocore/edk2/tree/edk2-stable202411 Tested features or cases: - VM basic boot, stress boot - Boot multiple VMs, boot various distros(including Windows) - Boot VM with huge resource/complex cpu topology - Memory hotplug/unplug, memory with NX hugepage on - Memory workload in high/low memory VMs with NX hugepage on - Device passthrough(NIC) - Live migration - Nested - PMU/vPMU - SGX - 5 level paging - Intel key features [0] https://bugzilla.kernel.org/show_bug.cgi?id=220019 [1] https://lore.kernel.org/all/a0e3eccd-314a-4073-a570-0fe7b27c25c8@xxxxxxxxxxxxxxx/#t Best Regards Fan
