Re: Breakage in arch/i386/kernel/time.c
Richard Gooch (rgooch@ras.ucalgary.ca)
Fri, 3 Dec 1999 11:37:01 -0700
Benjamin C. R. LaHaise writes:
> On Fri, 3 Dec 1999, Richard Gooch wrote:
>
> > Sergey Kubushin writes:
>
> > > Don't use ancient kgcc(Kernel GNU C Compiler)-2.7.x.x, use mainstream
> > > gcc-2.95.2 instead :))
> >
> > That's not a helpful response. Firstly, AFAIK, gcc 2.95.x is still
> > suspect (it's not had the rigorous testing that gcc 2.7.3.x has
> > had). Secondly, gcc 2.7.3.x is the recommended compiler in
> > Documentation/Changes.
>
> The breakage comes about due to the changes to the spinlock code.
> time.c does some really braindead things by forcing the use of
> specific registers for the variables passed to the asm. One fix
> would be to remove the register specifications, another is to make
> the function non-inline and another is to deprecate the use of gcc
> 2.7.2. Personally, given the bugs in 2.7.2 wrt register allocation,
> I'm in favour of getting people to move forward to gcc 2.95 asap.
But do_gettimeofday() doesn't have any explicit asm code. Sure, it
does locking, but those lock functions shouldn't affect the C code
it's called from. If there's leakage from the lock functions, that
should be fixed.
Regards,
Richard....
Permanent: rgooch@atnf.csiro.au
Current: rgooch@ras.ucalgary.ca
-
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.tux.org/lkml/