[PATCH] powerpc/vdso64: Add support for CLOCK_{REALTIME/MONOTONIC}_COARSE

[Santosh Sivaraj]

* Michael Ellerman <mpe at ellerman.id.au> wrote (on 2017-07-20 23:18:26 +1000):

> Santosh Sivaraj <santosh at fossix.org> writes:
> 
> > Current vDSO64 implementation does not have support for coarse
> > clocks (CLOCK_MONOTONIC_COARSE, CLOCK_REALTIME_COARSE), for which it falls
> > back to system call. Below is a benchmark of the difference in execution
> > time with and without vDSO support.
> 
> Hi Santosh,
> 
> Great patch! Always good to see asm replaced with C.
> 
> > diff --git a/arch/powerpc/kernel/vdso64/gettime.c b/arch/powerpc/kernel/vdso64/gettime.c
> > new file mode 100644
> > index 0000000..01f411f
> > --- /dev/null
> > +++ b/arch/powerpc/kernel/vdso64/gettime.c
> > @@ -0,0 +1,162 @@
> ...
> > +static notrace int gettime_syscall_fallback(clockid_t clk_id,
> > +					     struct timespec *tp)
> > +{
> > +	register clockid_t id asm("r3") = clk_id;
> > +	register struct timespec *t asm("r4") = tp;
> > +	register int nr asm("r0") = __NR_clock_gettime;
> > +	register int ret asm("r3");
> 
> I guess this works. I've always been a bit nervous about register
> variables TBH.
Yes, this works. It falls back to syscall in the case of CLOCK_BOOTTIME.

> 
> > +	asm volatile("sc"
> > +		     : "=r" (ret)
> > +		     : "r"(nr), "r"(id), "r"(t)
> > +		     : "memory");
> 
> Not sure we need the memory clobber?
> 
> It can clobber more registers than that though.
> 
> See: Documentation/powerpc/syscall64-abi.txt
> 
> > diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S
> > index 3820213..1258009 100644
> > --- a/arch/powerpc/kernel/vdso64/gettimeofday.S
> > +++ b/arch/powerpc/kernel/vdso64/gettimeofday.S
> > @@ -51,85 +53,21 @@ V_FUNCTION_BEGIN(__kernel_gettimeofday)
> ...
> > +	stwu	r1,-112(r1)
> > +  .cfi_register lr,r6
> > +	std	r6,24(r1)
> > +	bl	V_LOCAL_FUNC(kernel_clock_gettime)
> >  	crclr	cr0*4+so
> 
> Clearing CR0[SO] says that the syscall always succeeded.
> 
> What happens if you call this with a completely bogus clock id?
>

In case of a bogus clock id, the default case sets 'ret' to -1, which forces it
to fallback to the syscall.
> I think the solution is probably to do the syscall fallback in asm, and
> everything else in C.

Ok. That's what Sergey also sugested. I will send a v2.

Thanks,
Santosh
> 
> cheers

--