Bitcoin x86 for Windows
Figured I’d make a new topic since anyone on x86 Windows probably won’t even bother to read the x64 thread, however, since you’re here, I suggest you read page 5 of that thread after reading this.
OK, so essentially I’ve compiled 2 builds of Bitcoin with the new SHA caching optimisation, one build has full optimisation for the SSE instruction sets and will require a modern CPU, and the other is compiled without any SSE optimisation at all and should therefore run on pretty much any CPU capable of running XP or higher. The SSE version is a bit faster than the non-SSE version and both are inferior to the x64 builds, if you have a 64-bit OS, don’t bother with these.
Beware however that the libeay32.dll that I included may have SSE and therefore, if you can’t get either to run on your machine, replace that DLL with the one bundled with the stock Bitcoin
No need to be hostile. I do have a build environment set up, but I am completely unable to compile your changes without your source code. This just sends up a big red flag for me as you’ve obviously modified the source for some of your builds and are unwilling to provide those changes for peer review. I think this goes against the spirit of open source. Modified binary only releases don’t help progress the software and create an unnecessary dependency on an individual to provide those binaries.
Quote from: BlackEye on July 26, 2010, 12:38:27 PM
No need to be hostile. I do have a build environment set up, but I am completely unable to compile your changes without your source code. This just sends up a big red flag for me as you’ve obviously modified the source for some of your builds and are unwilling to provide those changes for peer review. I think this goes against the spirit of open source. Modified binary only releases don’t help progress the software and create an unnecessary dependency on an individual to provide those binaries.
Crypto++ 5.6.0: http://www.cryptopp.com/ Cached SHA256: http://pastebin.com/rJAYZJ32 (although I’m pretty sure this is publically submitted elsewhere, I was linked to it on IRC)
Or those with servers stuck on 32-bit OSes. :-D Quad core Xeon@1.6GHz Stock: 1100kh/s Full Opt: 2600kh/s
THANKS!
More than doubled my khash/sec from 800 to 1900.
AMD Phenon II X2 550
Quote from: sgtstein on July 26, 2010, 02:28:25 PM
Or those with servers stuck on 32-bit OSes. :-D Quad core Xeon@1.6GHz Stock: 1100kh/s Full Opt: 2600kh/s
THANKS!
BitCoins are always appreciated, address in my sig
Quote from: Olipro on July 26, 2010, 01:04:41 PM
Crypto++ 5.6.0: http://www.cryptopp.com/
Cached SHA256: http://pastebin.com/rJAYZJ32 (although I’m pretty sure this is publically submitted elsewhere, I was linked to it on IRC) I added the cached SHA256 state idea to the SVN, rev 113. The speedup is about 70%. I credited it to tcatm based on your post in the x64 thread.
I can compile the Crypto++ 5.6.0 ASM SHA code with MinGW but as soon as it runs it crashes. It says its for MASM (Microsoft’s assembler) and the sample command line they give looks like Visual C++. Does it only work with the MSVC and Intel compilers?
Quote from: Olipro on July 26, 2010, 03:42:01 PM
Quote from: sgtstein on July 26, 2010, 02:28:25 PM
Or those with servers stuck on 32-bit OSes. :-D Quad core Xeon@1.6GHz Stock: 1100kh/s Full Opt: 2600kh/s
THANKS!
BitCoins are always appreciated, address in my sig
I just sent you a wad of coin Wink
Quote from: satoshi on July 27, 2010, 01:29:42 AM
I can compile the Crypto++ 5.6.0 ASM SHA code with MinGW but as soon as it runs it crashes. It says its for MASM (Microsoft’s assembler) and the sample command line they give looks like Visual C++. Does it only work with the MSVC and Intel compilers?
I recently also made an attempt to use Crypto++ 5.6.0 (as an external library) instead of the old integrated code, with the same result - it crashed on the first invocation of CryptoPP::SHA256::Transform. Only I built everything with VC++ 2008. I haven’t investigated in depth, but someone mentioned Crypto++‘s routine required aligned input - maybe that’s the reason, or we may have other bug we don’t figure.
You need to change the assembly instructions that require aligned input to unaligned - topic 453 or make the blocks that are being hashed aligned. I haven’t tried yet, but this assembly code combined with the state caching modification should make this blazing fast.
Quote from: BlackEye on July 25, 2010, 10:12:23 PM
I was able to integrate the SHA256 functionality from Crypto++ 5.6.0 into Bitcoin. This is the fastest SHA256 yet using the SSE2 assembly code. Since Bitcoin was sending unaligned data to the block hash function, I had to change the MOVDQA instruction to MOVDQU.
I think using the SHA256 functionality from Crypto++ 5.6.0 is the way forward right now. I added a subset of the Crypto++ 5.6.0 library to the SVN. I stripped it down to just SHA and 11 general dependency files. There shouldn’t be any other crypto in there other than SHA.
I aligned the data fields and it worked. The ASM SHA-256 is about 48% faster. The combined speedup is about 2.5x faster than version 0.3.3.
I guess it’s using SSE2. It automatically sets its build configuration at compile time based on the compiler environment.
It looks like it has some SSE2 detection at runtime, but it’s hard to tell if it actually uses it to fall back if it’s not available. I want the release builds to have SSE2. SSE2 has been around since the first Pentium 4. A Pentium 3 or older would be so slow, you’d be wasting your electricity trying to generate on it anyway.
This is SVN rev 114.
OK, thanks. I’d also like to know if it runs fine as long as you don’t turn on Generate. You’d think as long as it doesn’t actually execute any SSE2 instructions, it would still load. At least Pentium 3’s could run it without generating.