digitalmars.D.bugs - weird
- David Medlock (9/9) Dec 23 2004 I have gotten the following error several times, and I cannot seem to
- Thomas Kuehne (7/15) Dec 24 2004 1) Cut the code down until any further reductions removes the error mess...
- David Medlock (27/27) Dec 24 2004 Thomas Kuehne wrote:
-
Simon Buchan
(17/17)
Dec 24 2004
On Fri, 24 Dec 2004 09:20:40 -0500, David Medlock
... - Thomas Kuehne (12/14) Dec 26 2004 -----BEGIN PGP SIGNED MESSAGE-----
I have gotten the following error several times, and I cannot seem to isolate the problem at all: solver.obj Error 42: Symbol Undefined __arguments_Aaik --- errorlevel 1 SMAKE fatal error: command "dmd" returned with error code 1 I am guessing 'Aaik' is a type signature, but I cannot decode it... I would like to hunt this down, but I honestly don't know where to look. Any help is appreciated.
Dec 23 2004
"David Medlock" <amedlock nospam.org> schrieb im Newsbeitrag news:cqg81u$1pos$1 digitaldaemon.com...I have gotten the following error several times, and I cannot seem to isolate the problem at all: solver.obj Error 42: Symbol Undefined __arguments_Aaik --- errorlevel 1 SMAKE fatal error: command "dmd" returned with error code 1 I am guessing 'Aaik' is a type signature, but I cannot decode it... I would like to hunt this down, but I honestly don't know where to look.1) Cut the code down until any further reductions removes the error message. 2) Try a search for "aik" in the source code. 3) Enable debug info and compile(without linking), search for "aik" in the object file. 4) Try to write a dummy "std/typeinfo/ti_*" class. Thomas
Dec 24 2004
Thomas Kuehne wrote: <snip> *some helpful stuff* Thanks Thomas. This is pretty strange. Its difficult to debug this because it is in a stack template class I wrote(attached). Believe it or not it is the writefln statement in the following routine: // Note: negative indices are from the end of the stack // ( this[-1] = last, this[0]=first ) T opIndex( int n ) in { assert( mCount>0 ); } body { alias std.string.toString str ; if ( n<0 ) n = mCount + n; if( (n<0) || (n>=this.mCount)) { char[] msg = "Attempting to index item %s %s items" ~ str(n) ~ " with " ~ str( mCount ) ~ " items." ; writefln("Attempting to index item %s with %s items", n, mCount ); throw new Exception("Stack Underflow: " ~ msg ); } else return data[n]; } If I comment out the writefln statement, everything links fine. It even happens when i qualify using std.stdio.writefln. Its a bit strange because (unless I am mistaken) the A in Aaik is a class reference type and I am passing no class references...
Dec 24 2004
On Fri, 24 Dec 2004 09:20:40 -0500, David Medlock <amedlock nospam.org> wrote: That code links fine with my system (dmd 0.109 XP). This may have something to do with your "SMAKE" whatever that is. -- "Unhappy Microsoft customers have a funny way of becoming Linux, Salesforce.com and Oracle customers." - www.microsoft-watch.com: "The Year in Review: Microsoft Opens Up" -- "I plan on at least one critical patch every month, and I haven't been disappointed." - Adam Hansen, manager of security at Sonnenschein Nath & Rosenthal LLP (Quote from http://www.eweek.com/article2/0,1759,1736104,00.asp) -- "It's been a challenge to "reteach or retrain" Web users to pay for content, said Pizey" -Wired website: "The Incredible Shrinking Comic"
Dec 24 2004
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Simon Buchan schrieb:That code links fine with my system (dmd 0.109 XP). This may have something to do with your "SMAKE" whatever that is.I can't reproduce it either (dmd & gdc Linux). Thomas PS: If it's a compiler bug, I am sure you can reduce this test case's size further. -----BEGIN PGP SIGNATURE----- iD8DBQFBzzPl3w+/yD4P9tIRAkRfAKC+8KoP1TUcohRNmyUFnznFH5rXCACgtqk/ ooFWosOevI8YlArVcbT8oMU= =yzYZ -----END PGP SIGNATURE-----
Dec 26 2004