Hello! I would like to ask whether there are any ongoing efforts towards accelerating the QEMU User mode Translation by conducting path exploration; binary decoding (For example supported by an Intel tool like XED2 or academic work like Jakstab) (ahead-of-time, before translation starts) and more importantly, I would be very interested to hear your opinion about a TranslationBlock (TCG, possibly LLVM [1]) disk cache that shall be created during or after the dynamic translation process.
My assumption is that a combination of static exploration and disassembly along with hard cached TranslationBlocks (Like an LLVM bit code file) could lead towards a significant speed-up after a single execution round and would give the project direction towards a static binary translation component. I am very much looking forward to hearing some of your opinions on this! Thanks, zulla [1] http://infoscience.epfl.ch/record/149975/files/x86-llvm-translator-chipounov_2.pdf
