On Thu, Feb 06, 2014 at 07:08:26PM +0100, Peter Zijlstra wrote:
On Thu, Feb 06, 2014 at 06:37:27PM +0100, Torsten Duwe wrote:
quoted
I must admit that I haven't tested the patch on non-pseries ppc64 nor on
ppc32. Only ppc64 has the ldarx and I tried to atomically replace the
holder along with the locks. That might prove unneccessary.
But what is the holder for? Can't we do away with that field?
Scott, Peter: good questions.
The conditional is wrong because I confused pSeries with ppc64 CPUs with
64-bit kernels. I got deluded by the LOCK_TOKEN definition above. Is that
correctly ifdef'd, with PPC64? The holder field should be ifdef'd
CONFIG_PPC_SPLPAR, independent of ppc64.
It is an advisory performance hint, and doesn't need to be updated atomically
with the lock; this and the above are 2 reasons to drop the asm string
operand size voodoo as well.
Thanks,
Torsten