Re: [Chicken-hackers] [PATCH] [LONG] Fix (hopefully) for #1068

[Christian Kellermann]

* Peter Bex <address@hidden> [131229 20:12]:
> Since I'm having some problems with my laptop I'm on a slow machine
> right now, so I can't do extensive testing or benchmarking.  It would
> be great if someone looking at this patch could perform a quick
> Salmonella run to see if no other eggs are broken, and run a few
> benchmarks comparing it against current master or 4.8.0.5 (you could
> use the one from https://github.com/mario-goulart/chicken-benchmarks).
> Please remember that you should compare optimized builds.  That means,
> without DEBUGBUILD, because this has been changed to check more things
> now.

I have run the chicken-benchmarks on it and it is still in the same ballpark on 
my machine.

; ./run.scm  
Repeating each program 10 times
Using /home/ckeen/chickens/master/bin/csc 

(1/43)  0.......................................0s
(2/43)  binarytrees.............................0.98s
(3/43)  boyer...................................1.47s
(4/43)  browse..................................1.28s
(5/43)  conform.................................1.19s
(6/43)  cpstak..................................5.04s
(7/43)  ctak....................................2.38s
(8/43)  dderiv..................................11.4s
(9/43)  deriv...................................10.81s
(10/43) destructive.............................0.94s
(11/43) div-iter................................0.34s
(12/43) div-rec.................................1.31s
(13/43) dynamic.................................1.83s
(14/43) earley..................................0.37s
(15/43) fft.....................................0.36s
(16/43) fib.....................................1.75s
(17/43) fibc....................................15.49s
(18/43) fprint..................................1.52s
(19/43) fread...................................11.48s
(20/43) graphs..................................9.26s
(21/43) hanoi...................................5.84s
(22/43) kanren..................................32.83s
(23/43) lattice.................................133.53s
(24/43) maze....................................1.42s
(25/43) mazefun.................................62.01s
(26/43) nbody...................................47.39s
(27/43) nboyer..................................323.02s
(28/43) nestedloop..............................29.52s
(29/43) nfa.....................................42.13s
(30/43) nqueens.................................0.7s
(31/43) nucleic2................................41.03s
(32/43) psyntax.................................12.27s
(33/43) puzzle..................................0.39s
(34/43) sboyer..................................812.42s
(35/43) scheme..................................0.16s
(36/43) slatex..................................67.08s
(37/43) sort1...................................42.38s
(38/43) tak.....................................3.96s
(39/43) takl....................................2.18s
(40/43) takr....................................5.86s
(41/43) traverse................................2.52s
(42/43) travinit................................0.54s
(43/43) triangl.................................6.81s

Total run time: 34m57s

I have pushed this patch to master to have a full salmonella run with it.

Thanks for all the bug hunting Peter!

Kind regards,

Christian

-- 
In the world, there is nothing more submissive and weak than
water. Yet for attacking that which is hard and strong, nothing can
surpass it. --- Lao Tzu