Fedor Indutny indutny

Organizations

@bem @Voxer @libuv @js-js @iojs
indutny commented on issue iojs/io.js#1522
@indutny

I have a fix for lowering 1.1gb, btw. Will push a PR tomorrow.

indutny commented on issue iojs/io.js#1522
@indutny

@ChALkeR I have modified the test case to work with localhost server (partly because I'm on 3G internet, partly because I'm not so patient). And it…

indutny commented on pull request iojs/io.js#1506
@indutny

@chrisdickinson wut? was it fixed by re-introducing SealHandleScope?

indutny deleted branch feature/update-http-parser-2.5.0 at indutny/io.js
indutny closed pull request iojs/io.js#1517
@indutny
deps: update http_parser to 2.5.0
indutny commented on pull request iojs/io.js#1517
@indutny

Landed in b3a7da1, thank you!

indutny pushed to v1.x at iojs/io.js
@indutny
  • @indutny b3a7da1
    deps: update http_parser to 2.5.0
indutny commented on pull request iojs/io.js#1517
@indutny

Benchmark results: https://gist.github.com/indutny/f4aef68005a0242a937e . Unfortunately, did it in reverse order so the - is actually a + and means…

@indutny
QUIC implementation?
indutny commented on issue indutny/node-spdy#118
@indutny

Now I'm ready to answer. No, if it'll be implemented - it'll be implemented in another module.

indutny commented on pull request iojs/io.js#1457
@indutny

@mscdex Thank you!

indutny commented on pull request iojs/io.js#1517
@indutny

@santigimeno cat over the keyboard?

indutny commented on pull request iojs/io.js#1517
@indutny

Running benchmarks on the Digital Ocean server... (Makes me wonder if Digital Ocean could sponsor a server for benchmarks for us... :) )

indutny opened pull request iojs/io.js#1517
@indutny
deps: update http_parser to 2.5.0
1 commit with 914 additions and 302 deletions
@indutny
@indutny
@indutny
indutny commented on pull request iojs/io.js#1457
@indutny

@mscdex could you please try comparing with latest http-parser master? I think it should be much faster than the current one.

indutny commented on pull request indutny/bn.js#27
@indutny

@alexeykudinkin thank you very much for helping me figure out! What particular algorithm are you implementing here? Could you please mention it in …

indutny commented on pull request indutny/bn.js#27
@indutny

Technically, twdf_ = exp(i * PI * j / s) here. Except that after j hit l it is twdf_ = exp(i * PI * (j - 1) / s)

indutny commented on pull request indutny/bn.js#27
@indutny

Why do you need this?

indutny commented on pull request indutny/bn.js#27
@indutny

Hm... perhaps I am missing something but p + j + s is bound by 3 * N here, while both arrays have N size?

indutny commented on pull request indutny/bn.js#27
@indutny

// wst[i] = wst[i] * nwst[i] Shouldn't it be rnwst in used terminology?

indutny commented on pull request indutny/bn.js#27
@indutny

So overall this loop reads like, this? for (var s = 1; s < N; s <<= 1) { var l = s << 1; var twdf = exp(i * PI / s); for (var p = 0; p < N; p += l)…

indutny commented on pull request indutny/bn.js#27
@indutny

NOTE: order of this two statements is actually reversed

indutny commented on pull request indutny/bn.js#27
@indutny

// twdf_ *= twdf