On 06/08/2011 19:38, Ian Lynagh wrote:
On Fri, Aug 05, 2011 at 03:09:11PM +0100, Simon Marlow wrote:
I get almost identical results on my Linux box here. Ok, so to
summarise the T4801 results:
peak MB: Linux; 61
OS X : 72
allocations: Linux: 72GB
OS X: 81GB
those differences are large enough to be a concern, I think (>10%).
There are no differences in the RTS that would cause extra
allocation on OS X. It seems to be just the compiler tests that
need more memory - so something in GHC maybe?
Here's T4801 run with -hT:
http://lambda.haskell.org/~igloo/linux.png
http://lambda.haskell.org/~igloo/mac.png
It looks like they're the same until the code gen stage, where we use
more memory to generate MachO code than ELF code. Could be due to some
missing strictness, as it looks like there is quite a bit of extra
THUNK.
Ah yes, so the NCG on OS X does some extra things that I'm not really
familiar with. There might be some low hanging fruit in there that we
don't normally notice because we don't profile on OS X.
Cheers,
Simon
FWIW, here are the compiler stats:
Linux:
[("bytes allocated", "723833288")
,("num_GCs", "1297")
,("average_bytes_used", "15074188")
,("max_bytes_used", "22636088")
,("num_byte_usage_samples", "1297")
,("peak_megabytes_allocated", "59")
,("init_cpu_seconds", "0.00")
,("init_wall_seconds", "0.00")
,("mutator_cpu_seconds", "0.91")
,("mutator_wall_seconds", "1.05")
,("GC_cpu_seconds", "75.07")
,("GC_wall_seconds", "75.07")
]
Mac:
[("bytes allocated", "812618832")
,("num_GCs", "1466")
,("average_bytes_used", "16302266")
,("max_bytes_used", "27078816")
,("num_byte_usage_samples", "1466")
,("peak_megabytes_allocated", "69")
,("init_cpu_seconds", "0.00")
,("init_wall_seconds", "0.00")
,("mutator_cpu_seconds", "0.79")
,("mutator_wall_seconds", "1.06")
,("GC_cpu_seconds", "73.02")
,("GC_wall_seconds", "74.44")
]
Thanks
Ian
_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc
_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc
