HackerTrans
TopNewTrendsCommentsPastAskShowJobs

corsix

no profile record

Submissions

(Ab)using gf2p8affineqb to turn indices into bits

corsix.org
2 points·by corsix·il y a 2 ans·0 comments

Polyfill-Glibc

github.com
5 points·by corsix·il y a 2 ans·1 comments

comments

corsix
·l’année dernière·discuss
In this case, "atomic 128-bit store" is the special instruction, with the twist that half of those 128 bits contain a pointer.
corsix
·l’année dernière·discuss
You can start with this idea, and then make it _very_ performant by using bit counting instructions. See https://www.corsix.org/content/higher-quality-random-floats for an exposition.
corsix
·il y a 2 ans·discuss
Indeed, LuaJIT support for Windows/Arm64 was added in https://github.com/LuaJIT/LuaJIT/issues/593, and there’s experimental out-of-tree support for Windows/Arm64EC in https://github.com/LuaJIT/LuaJIT/issues/1096
corsix
·il y a 2 ans·discuss
For an implementation of logical immediate encoding without the loop, see https://github.com/LuaJIT/LuaJIT/blob/04dca7911ea255f37be799...
corsix
·il y a 2 ans·discuss
AArch64 NEON has the URSQRTE instruction, which gets closer to the OP's question than you might think; view a 32-bit value as a fixed-precision integer with 32 fractional bits (so the representable range is evenly spaced 0 through 1-ε, where ε=2^-32), then URSQRTE computes the approximate inverse square root, halves it, then clamps it to the range 0 through 1-ε. Fixed-precision integers aren't quite integers, and approximate inverse square root isn't quite square root, but it might get you somewhere close.

The related FRSQRTE instruction is much more conventional, operating on 32-bit floats, again giving approximate inverse square root.
corsix
·il y a 2 ans·discuss
Unfortunately things aren't so simple, as when doing JIT compilation, LuaJIT _will_ try to shorten the lifetimes of local variables. Using the latest available version of LuaJIT (https://github.com/LuaJIT/LuaJIT/commit/0d313b243194a0b8d239...), the following reliably fails for me:

  local ffi = require"ffi"
  local function collect_lots()
    for i = 1, 20 do collectgarbage() end
  end
  local function f(s)
    local blob = ffi.new"int[2]"
    local interior = blob + 1
    interior[0] = 13 -- should become the return value
    s:gsub(".", collect_lots)
    return interior[0] -- kept alive by blob?
  end
  for i = 1, 60 do
    local str = ("x"):rep(i - 59)
    assert(f(str) == 13) -- can fail!!
  end
corsix
·il y a 3 ans·discuss
Some compute might be on the AMX units (dedicated matrix multiplication coprocessor, closely attached to the CPU, distinct from both ANE and GPU). They gained bf16 support in M2.
corsix
·il y a 3 ans·discuss
close() is documented as a cancellation point, and is the kind of syscall that might crop up in a destructor.
corsix
·il y a 3 ans·discuss
Using your cdf framing, while there is a point about [0, 1) versus (0, 1] intervals, the bigger point of the article is about whether said cdf holds for any IEEE-754 double-precision p, or whether it only holds for p of the form i*2^-53 (for integer i).
corsix
·il y a 3 ans·discuss
Oh, cute. It looks like they ever-so-slightly overweight the probability of values whose mantissa is entirely zeroes though. For example, the probability of hitting exactly 0.5 should be 2^-54 + 2^-55, whereas zig looks to give it 2^-54 + 2^-54.
corsix
·il y a 3 ans·discuss
https://www.corsix.org/content/whirlwind-tour-aarch64-vector... is my take on NEON, albeit not quite the same form factor as the OP.
corsix
·il y a 3 ans·discuss
I think https://github.com/LuaJIT/LuaJIT/commit/6a2163a6b45d6d251599... improved things a bit, notably making automatic tarballs work again.
corsix
·il y a 3 ans·discuss
I’m pleased to see FUTEX2_SIZE_U64, but saddened that it isn’t actually implemented. It has always seemed like a very useful primitive to have.
corsix
·il y a 3 ans·discuss
Also interesting is that https://luajit.org/status.html now states “LuaJIT is actively developed and maintained” (whereas for the last ~5 years, “actively” isn’t a word I’d have used), and makes reference to a TBA development branch.
corsix
·il y a 3 ans·discuss
Per https://www.stateof.ai/compute, one of the players in the market has ten thousand GPUs in a private cloud. Out-computing just that one player is hard enough, let alone out-computing the whole market.
corsix
·il y a 3 ans·discuss
The trie structure described in the article can be (ab)used to export an infinite number of symbols from a library: https://www.corsix.org/content/exporting-an-infinite-number-...
corsix
·il y a 3 ans·discuss
From a hardware perspective, vector instructions operate on small 1D vectors, whereas tensor instructions operate on small 2D matrices. I say “instructions”, but it’s really only matrix multiply or matrix multiply and accumulate - most other instructions are fine staying as 1D.
corsix
·il y a 3 ans·discuss
Assuming that you're after "round to nearest with ties toward even", then the quoted numpy code gets very close to `vcvtps2ph`, and one minor tweak gets it to bitwise identical: replace `ret += (ret == 0x7c00u)` with `ret |= 0x200`. Alternatively, the quoted Maratyszcza code gets to the same place if you replace `& 0x7c00u` with `& 0x7dffu`.
corsix
·il y a 3 ans·discuss
The first niche that came to mind was x86 code running under Rosetta 2; despite ARM having an equivalent to F16C, Rosetta 2 doesn’t translate AVX, and F16C doesn’t have a non-AVX encoding.