Quote from: nybble41 on July 28, 2010, 04:58:42 PM

Actually, I re-examined my changes based on your results and I think I may have messed up the rate calculation. Since it calculates four hashes per loop iteration I multiplied the increment to nHashCounter by four, but in retrospect it's already accounting for that by incrementing the nonce four times as quickly. Ergo, the rate was displayed as 4x higher than actual. For reference, I've generated 300 BTC (six blocks) since I started using the program on the 11th of this month, which works out to about one block out of every 400 (0.25%). Is that about what you're getting? It looks like your version should be a bit faster than mine, which is only to be expected--this was my first attempt at using SSE, or for that matter any kind of SIMD optimization.

I should've known it wouldn't be quite so simple.

Oh yeah, I had to review the khash/s code a few times to get it right. At one point it displayed 12000 but I somehow didn't believe it

I settled for this: in each thread, for each hash that I calculate, increase a counter. Every couple iterations save the khash/s *for this thread* in an array. Then every 30 seconds one thread sums up all the khash/s values and prints the total.

Quote from: nybble41 on July 28, 2010, 04:58:42 PM

Quote from: wereHamster on July 28, 2010, 03:39:51 PM

I don't think pure SSE can be exactly four times as fast as a well optimized C code, mostly because SSE lacks rotate instructions. Did you implement SHA completely in SSE or are you mixing SSE and C?

I implemented the rotates as ((x >> y) | (x << (32-y))), using SSE opcodes for the shifts. It's implemented completely in C, using GCC's vector extensions--no direct assembly code. I did use intrinsics for the shift operations, since the shift operators aren't implemented for vectors. The rest looks much like the original version.

Yep, that's more or less what I did. Except I used intrinsics instead of the gcc vector extension, I think that should be more portable. It was pretty easy, I took an existing implementation as a base and then only had to change some macros. Comparing my SSE version with the base yields a speedup of (only) 2.5x. It's not 4x, mainly because the lack of rotate operations in SSE. Packing and unpacking also cause a small decrease in speed.

I also heard that some compilers can generate suboptimal, sometimes even outright wrong, code from intrinsics. I was advised to stay away from them and use pure assembler instead.