Couple of questions about details of the RT patch
I was looking over the RT patch and I had a couple of questions:
Is there a reason that the PREEMPT_LAZY code is not ifdef-ed out when PREEMPT_LAZY is disabled. I can't see a reason to leave it in, it's just dead code at that point. I'd be happy to do a patch to fix this.
Why does preempt_disable_rt() become a barrier when PREEMPT_RT is disabled? Why isn't it just nothing?
Thanks,
-corey
Quoted 8 lines · Corey Minyard
I was looking over the RT patch and I had a couple of questions:
Is there a reason that the
PREEMPT_LAZYcode is not ifdef-ed out whenPREEMPT_LAZYis disabled. I can't see a reason to leave it in, it's just dead code at that point. I'd be happy to do a patch to fix this.Why does
preempt_disable_rt()become a barrier whenPREEMPT_RTis disabled? Why isn't it just nothing?
Because I'm a lazy bastard and I fundamentaly hate #ifdefs :)
Thanks,
tglx
PATCHES for 3.14
Guys and Mainteiners, throw 3.12 and other non LTS version, please enjoy a 3.14 kernel. How will release, we will be "stable" patches :D