[committed] Fix genemit segfault

Kaz Kojima kkojima@rr.iij4u.or.jp
Tue Nov 14 04:27:00 GMT 2006


Michael Matz <matz@suse.de> wrote:
> genemit segfault for me and valgrind points to the source of the problem, 
> this fixes it.  Committed as obvious (in case it isn't: the used[] array 
> will be subscribed by the actual operand number where scratch and dup 
> operands also count, but the 'operands' variable only holds the number 
> without scratches or dups).

I introduced that breakage in r118746.  Sorry about the breakage,
and thanks for fixing it.

Regards,
	kaz



More information about the Gcc-patches mailing list