--- Day changed Mon Aug 24 2015 02:44 -!- jtimon [~quassel@172.56.40.38] has joined #secp256k1 02:58 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has joined #secp256k1 03:46 -!- jtimon [~quassel@172.56.40.38] has quit [Ping timeout: 245 seconds] 03:53 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has quit [Ping timeout: 246 seconds] 03:58 -!- midnightmagic [~midnightm@unaffiliated/midnightmagic] has joined #secp256k1 04:57 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has quit [Ping timeout: 240 seconds] 04:57 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has joined #secp256k1 05:00 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has quit [Client Quit] 05:08 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has joined #secp256k1 07:21 -!- GAit1 [~GAit@2-228-102-98.ip191.fastwebnet.it] has joined #secp256k1 07:22 -!- GAit [~GAit@2-228-102-98.ip191.fastwebnet.it] has quit [Ping timeout: 245 seconds] 07:24 -!- GAit1 [~GAit@2-228-102-98.ip191.fastwebnet.it] has quit [Client Quit] 09:04 <@andytoshi> so yesterday i found another 10% or so perf gain in the GCD code, by noticing when the numbers are reduced to single words then switching entirely out of bignum code 09:04 < sipa> ha, nice 09:05 <@andytoshi> i could not improve this by doing this once the numbers reduced to dwords (uint128_t's) interestingly, and i spent a ton of time getting that to work 09:06 <@andytoshi> (cuz after doing the extgcd with single words, you wind up with a linear transformation that you have to apply to the bignums, since those are the actual result of the modular inversion. applying a uint128_t linear transformation to bignums while tracking carry is nasty) 09:09 <@andytoshi> today i think i'll be distracted by elements work, but i'd like to experiment when i get a chance with replacing the uint128_t/uint64_t calculation (which currently just uses the gcc types) with something from the dword/word papers i've read 09:09 <@andytoshi> not too optimistic about that; the gcc people know what they're doing 10:31 -!- jtimon [~quassel@172.56.41.255] has joined #secp256k1 12:16 -!- jtimon [~quassel@172.56.41.255] has quit [Ping timeout: 252 seconds] 13:26 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has joined #secp256k1 13:32 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has quit [Ping timeout: 260 seconds] 14:56 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has joined #secp256k1 15:12 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has quit [Remote host closed the connection] 15:16 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has joined #secp256k1 16:32 -!- jtimon [~quassel@184-23-239-227.dedicated.static.sonic.net] has quit [Ping timeout: 252 seconds]