Hello Paul,

  I think that %SYMBOL-VALUE is quite trivial, and the stepper can ignore it without a significant loss of the debugging features. Step on it is not likely to provide any useful info. 

 Best Regards,

2011/5/12 Paul Khuong <pvk@pvk.ca>
On 2011-05-12, at 6:42 AM, Nikodemus Siivola wrote:

> Update of /cvsroot/sbcl/sbcl/src/compiler
> In directory vz-cvs-3.sog:/tmp/cvs-serv3523/src/compiler
> Modified Files:
>       ir1tran.lisp
> Log Message:
> more conservative global variable conversion
> Based on patch by Roman Marynchak.
> Fixes lp#722734.
> * Modify IR1-CONVERT-VAR to emit SYMBOL-VALUE wrapper
>   for all global variables except those which are
> * Modify IR1-STEP-FORM-P to return false for SYMBOL-VALUE
>   with constant argument, now that virtually all global
>   variables are accessed with a function call (which gets
>   converted via a VOP, so the final machine code remains
>   the same.)

I'm not sure how I feel about this. I've never used the stepper, but it seems to me that its users can expect a function call to trigger a step. Would it be so bad to have %SYMBOL-VALUE or some such that doesn't trigger a step?

Paul Khuong
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
Sbcl-devel mailing list