[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#378599: [m68k] gcc-4.1 -O2 optimization problem



On Mon, Jul 17, 2006 at 09:08:42PM +0200, Falk Hueffner wrote:
> Stephen R Marenka <stephen@marenka.net> writes:
> 
> > We now have about sixteen packages that end up with a segfault or an
> > illegal instruction. I haven't tried recompiling all these to verify that 
> > this is the same problem, but I will if it will help.
> >
> > The problem is I have no idea how to isolate this bug.
> 
> Select the most minimal package where the problem is reproducible. Try
> to compile single .c files with -O0 to identify the miscompiled file.
> Split the miscompiled file by function to idenify the miscompiled
> function. Make a stand-alone testcase involving only this function.

The only way I know to reproduce the problem is to build a whole library
and run the test case. 

libgc is the quickest to compile of the problem packages I know about --
it takes about three hours for a full compile.

So you're saying the only way is to compile, link, and test each of the
55 source files? Yikes, that's going to take a while.

-- 
Stephen R. Marenka     If life's not fun, you're not doing it right!
<stephen@marenka.net>

Attachment: signature.asc
Description: Digital signature


Reply to: