TODO.nvmm revision 1.4
11.1SmaxvKnown issues in NVMM, low priority in most cases. 21.1Smaxv 31.1Smaxv====== KERNEL NVMM DRIVER ====== 41.1Smaxv 51.2Smaxv * 32bit-PAE guests can misbehave on Intel, because we need to manually 61.2Smaxv install the PDPTEs, and currently we don't do it. In practice they don't 71.2Smaxv misbehave because the emulator never has to interfere with CR3. 81.1Smaxv 91.4Smaxv * AMD: we don't support VCPU_CONF_TPR, would be nice to. 101.1Smaxv 111.4Smaxv * AMD: need to do comprehensive CPUID filtering. 121.1Smaxv 131.4Smaxv * Intel: we have comprehensive CPUID filtering, but should we limit the highest 141.4Smaxv leaf? 151.1Smaxv 161.1Smaxv====== LIBNVMM ====== 171.1Smaxv 181.1Smaxv * There are still a few twisted corner cases we don't handle in the instruction 191.1Smaxv emulator. For example if the guest makes an MMIO access relative to RSP, we 201.1Smaxv must base the GVA on %SS and not %DS. This is tiring, and in practice, no 211.1Smaxv guest is dumb enough to perform such accesses. 221.1Smaxv 231.4Smaxv * Maybe the __areas should have a rwlock? I don't think Qemu unmaps memory 241.4Smaxv while VCPUs are running, but still. 25