mirror of
https://github.com/yuzu-emu/unicorn.git
synced 2025-01-23 06:31:09 +00:00
ae3e22a689
For some workloads such as arm bootup, tb_phys_hash is performance-critical. The is due to the high frequency of accesses to the hash table, originated by (frequent) TLB flushes that wipe out the cpu-private tb_jmp_cache's. More info: https://lists.nongnu.org/archive/html/qemu-devel/2016-03/msg05098.html To dig further into this I modified an arm image booting debian jessie to immediately shut down after boot. Analysis revealed that quite a bit of time is unnecessarily spent in tb_phys_hash: the cause is poor hashing that results in very uneven loading of chains in the hash table's buckets; the longest observed chain had ~550 elements. The appended addresses this with two changes: 1) Use xxhash as the hash table's hash function. xxhash is a fast, high-quality hashing function. 2) Feed the hashing function with not just tb_phys, but also pc and flags. This improves performance over using just tb_phys for hashing, since that resulted in some hash buckets having many TB's, while others getting very few; with these changes, the longest observed chain on a single hash bucket is brought down from ~550 to ~40. Tests show that the other element checked for in tb_find_physical, cs_base, is always a match when tb_phys+pc+flags are a match, so hashing cs_base is wasteful. It could be that this is an ARM-only thing, though. UPDATE: On Tue, Apr 05, 2016 at 08:41:43 -0700, Richard Henderson wrote: > The cs_base field is only used by i386 (in 16-bit modes), and sparc (for a TB > consisting of only a delay slot). > It may well still turn out to be reasonable to ignore cs_base for hashing. BTW, after this change the hash table should not be called "tb_hash_phys" anymore; this is addressed later in this series. This change gives consistent bootup time improvements. I tested two host machines: - Intel Xeon E5-2690: 11.6% less time - Intel i7-4790K: 19.2% less time Increasing the number of hash buckets yields further improvements. However, using a larger, fixed number of buckets can degrade performance for other workloads that do not translate as many blocks (600K+ for debian-jessie arm bootup). This is dealt with later in this series. Backports commit 42bd32287f3a18d823f2258b813824a39ed7c6d9 from qemu |
||
|---|---|---|
| bindings | ||
| docs | ||
| include | ||
| msvc | ||
| qemu | ||
| samples | ||
| tests | ||
| .appveyor.yml | ||
| .gitignore | ||
| .travis.yml | ||
| AUTHORS.TXT | ||
| Brewfile | ||
| ChangeLog | ||
| config.mk | ||
| COPYING | ||
| COPYING.LGPL2 | ||
| COPYING_GLIB | ||
| CREDITS.TXT | ||
| install-cmocka-linux.sh | ||
| list.c | ||
| make.sh | ||
| Makefile | ||
| msvc.bat | ||
| pkgconfig.mk | ||
| README.md | ||
| uc.c | ||
| windows_export.bat | ||
Unicorn Engine
Unicorn is a lightweight, multi-platform, multi-architecture CPU emulator framework based on QEMU.
Unicorn offers some unparalleled features:
- Multi-architecture: ARM, ARM64 (ARMv8), M68K, MIPS, SPARC, and X86 (16, 32, 64-bit)
- Clean/simple/lightweight/intuitive architecture-neutral API
- Implemented in pure C language, with bindings for Crystal, Clojure, Visual Basic, Perl, Rust, Ruby, Python, Java, .NET, Go, Delphi/Free Pascal and Haskell.
- Native support for Windows & *nix (with Mac OSX, Linux, *BSD & Solaris confirmed)
- High performance via Just-In-Time compilation
- Support for fine-grained instrumentation at various levels
- Thread-safety by design
- Distributed under free software license GPLv2
Further information is available at http://www.unicorn-engine.org
License
This project is released under the GPL license.
Compilation & Docs
See docs/COMPILE.md file for how to compile and install Unicorn.
More documentation is available in docs/README.md.
Contact
Contact us via mailing list, email or twitter for any questions.
Contribute
If you want to contribute, please pick up something from our Github issues.
We also maintain a list of more challenged problems in a TODO list.
CREDITS.TXT records important contributors of our project.