Starting with tag:
[TAG Initial conversion from CVS complete
John Goerzen <jgoerzen@complete.org>**20051128162047] [subvert recent optimisations in GHC
Simon Marlow <simonmar@microsoft.com>**20060207140859
 
 Some of these tests were using 'take' in their attempts to duplicate
 some work and hence run longer, but since GHC just started deforesting
 'take' this stopped working.  Now we use {-# NOILINE #-} as a
 hopefully more robust way to subvert optimisations.
] 
[add stdout for x86_64
Simon Marlow <simonmar@microsoft.com>**20060329075554] 
[Increase maximum heap size allowed for imaginary/wheel-sieve2
Ian Lynagh <igloo@earth.li>**20060905195843] 
[Add some SMP and threading benchmarks I have lying around
Simon Marlow <simonmar@microsoft.com>**20060327130230] 
[make some of these compile with GHC 6.6
Simon Marlow <simonmar@microsoft.com>**20061027080007] 
[knock up a more up to date README
Simon Marlow <simonmar@microsoft.com>**20061016105641] 
[TAG 6.6 release
Ian Lynagh <igloo@earth.li>**20061011124740] 
[Put the bernouilli argument in PROG_ARGS rather than SRC_RUNTEST_OPTS
Ian Lynagh <igloo@earth.li>**20061211132653
 The current way didn't work when running nofib with ghci.
] 
[Make primetest main be IO () so ghci doesn't print the result
Ian Lynagh <igloo@earth.li>**20061211134513] 
[Increase heap for wheel-sieve2 (was too small for mode=slow)
Ian Lynagh <igloo@earth.li>**20061212211118] 
[Look at the stdout/stderr files appropriate for mode
Ian Lynagh <igloo@earth.li>**20061220220611] 
[Tell runstdtest about binary outputs
Simon Marlow <simonmar@microsoft.com>**20070123123911] 
[Initial contribution: buggy tests
josep**20070130191656
 
 This patch also deletes unmodified, non-buggy tests
 
] 
[Updated Readme
josep**20070131181751] 
[fix rfib
josep**20070619111747] 
