I know this one is not new but...
> Michael Nelson <nelson@seahunt.imat.com> wrote:
> > What _is_ your officially approved compiler these days? I was under the
> > impression it was gcc-2.7.2.3. What do YOU use to compile the kernel?
>
> I still use gcc-2.7.2.1. And yes, it also has the bug, but I just never
> saw it because I only compiled for SMP.
>
> However, unlike 2.0.x and Alan, I do NOT have a rule that says that "only
> compiler version xxx" is supported. I'm happy having people compile v2.1
> with newer compilers - it should work. So I don't have a blacklist,
> although I also haven't tried to keep up with the "bug of the month" wrt
> the newer compilers (the snapshots have had some problems).
>
> My current version of this particular fix will actually do the right thing
> for gcc-2.8.x, where it just does the empty initializer. So as of 2.1.115,
> gcc-2.8 will create better code than 2.7.x for this case on UP. Go wild,
> but know that I haven't tested that particular compiler..
>
> Linus
Maybe it is time to upgrade to gcc-2.7.3, so that this patch could
be applied in the main stream :
-linux-old/arch/i386/Makefile
+linux/arch/i386/Makefile
CFLAGS_PIPE := -pipe
-CFLAGS_NSR := -fno-strength-reduce
+CFLAGS_NSR :=
CFLAGS := $(CFLAGS) $(CFLAGS_PIPE) $(CFLAGS_NSR)
And then the average "old" PCs can have a speedup, better than
going from 2.0 to 2.1 , even with standard distributions.
Moreover, performance measures will be more meaningfull,
and the kernel should not have compiler specific...
gcc-2.7.1 is no more really the "bug of the month", but one day
this option should go away, isn't it ?
Etienne.
-- -- The world belong to its organizer.- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html