[mercury-users] SIREG again

[Fergus Henderson]

On 26-Feb-2003, Michael Day <mikeday at yeslogic.com> wrote:
> 
> unable to find a register to spill in class `SIREG'
> this is the insn:
> (insn 110 107 113 (parallel[
>             (set (reg:SI 2 ecx [74])
>                 (const_int 0 [0x0]))
> 		.....
> 
> I no longer get this kind of error when building the compiler since I
> upgraded to Redhat 8 with gcc 3.2, but I still get it from time to time in
> my own code when using C foreign procs.
> 
> I'll hack around it; but is this something that will be fixed in gcc 3.4?

Probably yes.  The GCC bug with using memcpy(), memcmp(), memset(), bzero()
or structure assignment in code which declares "esi" as a global register
variable has been fixed in the GCC CVS repository, and the fix will be
included in gcc 3.4 (and possibly also in gcc 3.3).  So if that is the
one that is causing this problem, then yes it will be fixed.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
The University of Melbourne         |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------