Thanks, upgrading to 6.4.2 seems to have done the trick. I've obviously
put it off too long!
Rich
___
Glasgow-haskell-users mailing list
Glasgow-haskell-users@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
I'm trying to use heap profiling with +RTS -hc -i1 options and running
my program for about 30 seconds. However, I only get around 7 samples
with seemingly bogus timetags (i.e. 0.00, 3.69, 3.73, 3.10, 4.05,
4.12). What's going on?
I'm running GHC 6.4.2 on Windows (MSYS/MinGW).
Thanks,
Rich
Yep, it was buried in the matrix inversion. Thanks!
On 7/20/06, Neil Mitchell <[EMAIL PROTECTED]> wrote:
Hi,
> I have some code that just says
> <>
This behaviour is also caused by black holes, for example:
main = let x = x + 5 in print x
gives this same error. Note that the value of x depen
Hi,
I have some code that just says
<>
This behaviour is also caused by black holes, for example:
main = let x = x + 5 in print x
gives this same error. Note that the value of x depends on the value
of x. Hence you get a circular problem, which would non-terminate, but
the compiler can spot
Hi,
I have some code that just says
<>
when it's run. I found a FAQ that says something about finalizers, but
I'm not using those (at least not consciously). Here is my main:
import qualified Data.ByteString.Lazy.Char8 as L
main = do ls <- liftM L.lines $ L.readFile "output.txt"
mapM_
Chris Kuklewicz wrote:
I was working on pulling library calling code from JRegex into
Text.Regex.Lazy
and I am wondering if this code for compiling regex.h regular
expressions is a
potential memory leak:
regcomp pattern flags = do
regex_fptr <- mallocForeignPtrBytes (#const sizeof(regex_t))
Joel Reymont wrote:
I get a lot of errors like this:
utils/Panic.lhs:27:0:
Failed to load interface for `Config':
Bad interface file: stage1/main/Config.hi
mismatched interface file versions: expected 6050, found
390518464
Do you guys always clean after a pull or is t
Joel Reymont wrote:
What I usually do is build GHC from darcs and make-install it. I then
do a darcs pull and use the previously built compiler to bootstrap. I
guess this is not optimal and I should just use the last stable GHC to
recompile the darcs tree.
Yes, that isn't a "supported" wor
Hello Rich,
Tuesday, July 18, 2006, 11:47:58 PM, you wrote:
> I have a server application that I am building using GHC 6.4 (yes, an
> update to 6.4.2 is on the horizon, but
> not in the immediate future - unless it fixes this problem :) ) under
6.4.2 (or better a current stable build) fixes a lo
