Messages in this thread | | | From | Andy Lutomirski <> | Date | Tue, 21 Jul 2015 11:54:30 -0700 | Subject | Re: Kernel broken on processors without performance counters |
| |
On Tue, Jul 21, 2015 at 11:50 AM, Jason Baron <jasonbaron0@gmail.com> wrote: > > > On 07/21/2015 02:15 PM, Borislav Petkov wrote: >> On Tue, Jul 21, 2015 at 06:12:15PM +0200, Peter Zijlstra wrote: >>> Yes, if you start out false, you must be unlikely. If you start out >>> true, you must be likely. >>> >>> We could maybe try and untangle that if there really is a good use case, >>> but this is the current state. >>> >>> The whole reason this happened is because 'false' is like: >>> >>> >>> ... >>> <nop> >>> 1: >>> ... >>> >>> >>> >>> label: >>> <unlikely code> >>> jmp 1b >>> >>> >>> Where the code if out-of-line by default. The enable will rewrite the >>> <nop> with a jmp label. >> Btw, native_sched_clock() is kinda botched because of that, see below. >> >> I'd want that RDTSC to come first with a NOP preceding it which can >> become a JMP in case some idiotic CPU can't do RDTSC and needs to use >> jiffies. Instead, we *unconditionally* jump to RDTSC which is WTF?! We >> can just as well do a normal unlikely() without the static_key: > > hmmm...so this is a case where need to the default the branch > to the out-of-line branch at boot. That is, we can't just enable > the out-of-line branch at boot time, b/c it might be too late at > that point? IE native_sched_clock() gets called very early? >
Could this be done at link time, or perhaps when compressing the kernel image, instead of at boot time?
--Andy
| |