ffi.new vs ffi.C.malloc speed
- From: Ronan Collobert <ronan@xxxxxxxxxxxxx>
- To: luajit@xxxxxxxxxxxxx
- Date: Wed, 10 Oct 2012 16:43:03 +0200
Hi,
Assuming I want to deal with large floating point vectors, it looks to me I
cannot use ffi.new. Reading luajit mailing list, I understand ffi.new arrays
are constrained to the 2GB luajit memory management limit. So, a simple...
> = ffi.new('float[?]', 1000000000)
bad argument #1 to '?' (size of C type is unknown or too large)
...which tries to allocate slightly less than 4GB, fails.
Still reading luajit mailing list, I understand than I can instead use
ffi.C.malloc. In which case I need to do three calls: (1) ffi.C.malloc, (2)
ffi.cast such that my vector is usable and (3) ffi.gc such that it is properly
garbage collected. This works great, but I noticed this process is much slower
(~ x10) than the ffi.new way (looks like the GC takes quite a lot of time in
that case). See benchmem.lua in attachment:
time luajit benchmem.lua vla
real 0m1.505s
user 0m1.500s
sys 0m0.000s
time luajit benchmem.lua malloc
real 0m13.928s
user 0m13.140s
sys 0m0.770s
Is there any better solution? What should I expect in terms of speed
improvement in the future (say, with the new GC in luajit 2.1)?
Thanks,
Ronan.
Other related posts:
