Messages in this thread | | | Date | Mon, 04 Feb 2013 08:50:33 -0500 | From | Rik van Riel <> | Subject | Re: [PATCH -v4 5/5] x86,smp: limit spinlock delay on virtual machines |
| |
On 01/26/2013 07:47 AM, Borislav Petkov wrote: > On Sat, Jan 26, 2013 at 01:00:43PM +0100, Ingo Molnar wrote: >>> +void __init init_spinlock_delay(void) >>> +{ >>> + if (x86_hyper) >>> + max_spinlock_delay = MAX_SPINLOCK_DELAY_GUEST; >> >> I realize that you took existing code and extended it, but that >> chunk of code looks pretty disgusting visually now - at minimum >> it should be vertically aligned as most other kernel code does. >> >> The comment should also tell that the unit of these values is >> 'spinlock-op loops' or so.
Will do.
> Also, with currently making PARAVIRT_GUEST optional, x86_hyper is maybe > a bad choice of a variable to test. > > Maybe instead to this: > > if (IS_ENABLED(CONFIG_PARAVIRT_GUEST)) > ...
We need to know whether we are actually running on top of a hypervisor, not whether we have the code compiled in to do so.
After all, the majority of distribution kernels will have CONFIG_PARAVIRT_GUEST set, but the majority of those kernels will be running bare metal...
-- All rights reversed
| |