On Thu, Nov 23, 2017 at 09:55:11AM +0100, Peter Zijlstra wrote:
> On Tue, Nov 21, 2017 at 09:18:53AM -0500, Mathieu Desnoyers wrote:
> > +static inline __attribute__((always_inline))
> > +int rseq_cmpeqv_storev(intptr_t *v, intptr_t expect, intptr_t newv,
> > + int cpu)
> > +{
> > + __asm__ __volatile__ goto (
> > + RSEQ_ASM_DEFINE_TABLE(3, __rseq_table, 0x0, 0x0, 1f, 2f-1f, 4f)
> > + RSEQ_ASM_STORE_RSEQ_CS(1, 3b, rseq_cs)
>
> > + RSEQ_ASM_CMP_CPU_ID(cpu_id, current_cpu_id, 4f)
> > + "cmpq %[v], %[expect]\n\t"
> > + "jnz 5f\n\t"
Also, I'm confused between the abort and cmpfail cases.
In would expect the cpu_id compare to also result in cmpfail, that is, I
would only expect the kernel to result in abort.
> > + /* final store */
> > + "movq %[newv], %[v]\n\t"
> > + "2:\n\t"
> > + RSEQ_ASM_DEFINE_ABORT(4, __rseq_failure, RSEQ_SIG, "", abort)
> > + RSEQ_ASM_DEFINE_CMPFAIL(5, __rseq_failure, "", cmpfail)
> > + : /* gcc asm goto does not allow outputs */
> > + : [cpu_id]"r"(cpu),
> > + [current_cpu_id]"m"(__rseq_abi.cpu_id),
> > + [rseq_cs]"m"(__rseq_abi.rseq_cs),
> > + [v]"m"(*v),
> > + [expect]"r"(expect),
> > + [newv]"r"(newv)
> > + : "memory", "cc", "rax"
> > + : abort, cmpfail
> > + );
> > + return 0;
> > +abort:
> > + return -1;
Which then would suggest this be -EINTR or something like that.
> > +cmpfail:
> > + return 1;
> > +}
