On 03/04/2019 11:06, Thomas Gleixner wrote: > On Wed, 3 Apr 2019, Martin Schwidefsky wrote: > >> On Mon, 1 Apr 2019 12:51:50 +0100 >> Vincenzo Frascino <vincenzo.frascino@xxxxxxx> wrote: >> >>> clock_getres in the vDSO library has to preserve the same behaviour >>> of posix_get_hrtimer_res(). >>> >>> In particular, posix_get_hrtimer_res() does: >>> sec = 0; >>> ns = hrtimer_resolution; >>> and hrtimer_resolution depends on the enablement of the high >>> resolution timers that can happen either at compile or at run time. >>> >>> Fix the s390 vdso implementation of clock_getres keeping a copy of >>> hrtimer_resolution in vdso data and using that directly. >>> >>> Cc: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> >>> Cc: Heiko Carstens <heiko.carstens@xxxxxxxxxx> >>> Signed-off-by: Vincenzo Frascino <vincenzo.frascino@xxxxxxx> >>> --- >>> arch/s390/include/asm/vdso.h | 1 + >>> arch/s390/kernel/asm-offsets.c | 2 +- >>> arch/s390/kernel/time.c | 1 + >>> arch/s390/kernel/vdso32/clock_getres.S | 17 ++++++++++++----- >>> arch/s390/kernel/vdso64/clock_getres.S | 15 ++++++++++----- >>> 5 files changed, 25 insertions(+), 11 deletions(-) >> >> I tried this patch and in principle this works. In that regard >> Acked-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> >> >> But I wonder if the loop to check the update counter is really >> necessary. The hrtimer_resolution value can only changes once with >> the first call to hrtimer_switch_to_hres(). With the TOD clock >> as the only clock available on s390 we always have the ability >> to do hrtimer. It then all depends on the highres=[on|off] kernel >> parameter what value we get with clock_getres(). > > Yes, it's not changing after boot anymore. > > Thanks, > > tglx > Ok, I will remove the loop from both the implementations and post it with v2. -- Regards, Vincenzo
