On Tue, Sep 02, 2014 at 11:40:16AM +0200, Peter Zijlstra wrote: > On Mon, Sep 01, 2014 at 09:04:38PM +0200, Peter Zijlstra wrote: > > On Mon, Sep 01, 2014 at 09:46:42AM -0700, Guenter Roeck wrote: > > > Yes, the problem is still seen in next-140829. See qemu test results at > > > http://server.roeck-us.net:8010/builders. > > > > > > I provided sparc64 images at http://server.roeck-us.net/qemu/sparc64/. > > > The README file includes instructions on how to run the image in qemu > > > and how to build a kernel. > > > > > > I can do the same for mips and mips64 if needed; let me know. > > > > Thanks, that's enough to get started. > > This makes sparc64 go again, silly oversight indeed. > > --- > Subject: sparc64: Fix atomics > > The patch folding the atomic ops had a silly fail in the _return primitives. > > Fixes: 4f3316c2b5fe ("locking,arch,sparc: Fold atomic_ops") > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx> Confirmed working. Tested-by: Guenter Roeck <linux@xxxxxxxxxxxx> > --- > diff --git a/arch/sparc/lib/atomic_64.S b/arch/sparc/lib/atomic_64.S > index 96d70b4..05dac43 100644 > --- a/arch/sparc/lib/atomic_64.S > +++ b/arch/sparc/lib/atomic_64.S > @@ -37,7 +37,7 @@ ENTRY(atomic_##op##_return) /* %o0 = increment, %o1 = atomic_ptr */ \ > cas [%o1], %g1, %g7; \ > cmp %g1, %g7; \ > bne,pn %icc, BACKOFF_LABEL(2f, 1b); \ > - add %g1, %o0, %g1; \ > + op %g1, %o0, %g1; \ > retl; \ > sra %g1, 0, %o0; \ > 2: BACKOFF_SPIN(%o2, %o3, 1b); \ > @@ -76,7 +76,7 @@ ENTRY(atomic64_##op##_return) /* %o0 = increment, %o1 = atomic_ptr */ \ > bne,pn %xcc, BACKOFF_LABEL(2f, 1b); \ > nop; \ > retl; \ > - add %g1, %o0, %o0; \ > + op %g1, %o0, %o0; \ > 2: BACKOFF_SPIN(%o2, %o3, 1b); \ > ENDPROC(atomic64_##op##_return); > > > -- To unsubscribe from this list: send the line "unsubscribe linux-next" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
