On Mon, 11 Dec 2000, Chris Faulhaber wrote:
> No, but you may want to review the mailing list archives since GCC
> optimization problems have been rehash many times (see also the notes
> in /etc/defaults/make.conf WRT optimization).
I hadn't seen those comments. I didn't think gcc could be *that
Hi,
On a PIII 600E, compiling a week old 4.2S kernel with the following
options:
-O2 -fomit-frame-pointer -mpentium -march=pentium -pipe
Cause the kernel to produce TCP CRC errors when talking to certain IP
addresses. It generally seemed to get the 2nd octet of the CRC out by one
i.e. 0xb9f7 b
On Mon, Dec 11, 2000 at 03:34:19PM +, Mark Powell wrote:
> On Mon, 11 Dec 2000, Thomas T. Veldhouse wrote:
>
> > I generally don't compile the kernel with any optimizations except
> > "-O -pipe". I never have a problem - and I don't have any performance
> > differences that have been noticab
On Mon, 11 Dec 2000, Thomas T. Veldhouse wrote:
> I generally don't compile the kernel with any optimizations except
> "-O -pipe". I never have a problem - and I don't have any performance
> differences that have been noticable.
Well okay. But this isn't supposed to happen. Is it?
Cheers.
Ma
lt;[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, December 11, 2000 9:24 AM
Subject: Different gcc optimisations cause IP packet CRC problems
> Hi,
> On a PIII 600E, compiling a week old 4.2S kernel with the following
> options:
>
> -O2 -fomit-frame-pointer -mpentium
