[PATCH] riscv: Implement smp_cond_load8/16() with Zawrs

[Guo Ren]

On Mon, Dec 16, 2024 at 11:42 PM Andrew Jones <ajones at ventanamicro.com> wrote:
>
> On Sun, Dec 15, 2024 at 10:22:53PM -0500, guoren at kernel.org wrote:
> > From: Guo Ren <guoren at linux.alibaba.com>
> >
> > RISC-V code uses the queued spinlock implementation, which calls
> > the macros smp_cond_load_acquire for one byte. So, complement the
> > implementation of byte and halfword versions.
> >
> > Signed-off-by: Guo Ren <guoren at linux.alibaba.com>
> > Signed-off-by: Guo Ren <guoren at kernel.org>
> > ---
> >  arch/riscv/include/asm/cmpxchg.h | 38 +++++++++++++++++++++++++++++---
> >  1 file changed, 35 insertions(+), 3 deletions(-)
> >
> > diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
> > index 4cadc56220fe..2bd42a11ff8f 100644
> > --- a/arch/riscv/include/asm/cmpxchg.h
> > +++ b/arch/riscv/include/asm/cmpxchg.h
> > @@ -365,16 +365,48 @@ static __always_inline void __cmpwait(volatile void *ptr,
> >  {
> >       unsigned long tmp;
> >
> > +     u32 *__ptr32b;
> > +     ulong __s, __val, __mask;
> > +
> >       asm goto(ALTERNATIVE("j %l[no_zawrs]", "nop",
> >                            0, RISCV_ISA_EXT_ZAWRS, 1)
> >                : : : : no_zawrs);
> >
> >       switch (size) {
> >       case 1:
> > -             fallthrough;
> > +             __ptr32b = (u32 *)((ulong)(ptr) & ~0x3);
> > +             __s = ((ulong)(ptr) & 0x3) * BITS_PER_BYTE;
> > +             __val = val << __s;
> > +             __mask = 0xf << __s;
>
> This mask should be 0xff and the mask below should be 0xffff.
Thx for catching it; it's hard to test it out. I will correct it in
the next version.

>
> > +
> > +             asm volatile(
> > +             "       lr.w    %0, %1\n"
> > +             "       and     %0, %0, %3\n"
> > +             "       xor     %0, %0, %2\n"
> > +             "       bnez    %0, 1f\n"
> > +                     ZAWRS_WRS_NTO "\n"
> > +             "1:"
> > +             : "=&r" (tmp), "+A" (*(__ptr32b))
> > +             : "r" (__val), "r" (__mask)
> > +             : "memory");
> > +             break;
> >       case 2:
> > -             /* RISC-V doesn't have lr instructions on byte and half-word. */
> > -             goto no_zawrs;
> > +             __ptr32b = (u32 *)((ulong)(ptr) & ~0x3);
> > +             __s = ((ulong)(ptr) & 0x2) * BITS_PER_BYTE;
> > +             __val = val << __s;
> > +             __mask = 0xff << __s;
> > +
> > +             asm volatile(
> > +             "       lr.w    %0, %1\n"
> > +             "       and     %0, %0, %3\n"
> > +             "       xor     %0, %0, %2\n"
> > +             "       bnez    %0, 1f\n"
> > +                     ZAWRS_WRS_NTO "\n"
> > +             "1:"
> > +             : "=&r" (tmp), "+A" (*(__ptr32b))
> > +             : "r" (__val), "r" (__mask)
> > +             : "memory");
> > +             break;
> >       case 4:
> >               asm volatile(
> >               "       lr.w    %0, %1\n"
> > --
> > 2.40.1
> >
>
> Thanks,
> drew



-- 
Best Regards
 Guo Ren