Hi, Palmer,
On Sat, Apr 30, 2022 at 9:22 PM Palmer Dabbelt <palmer@xxxxxxxxxxx> wrote:
>
> On Sat, 30 Apr 2022 03:41:59 PDT (-0700), Arnd Bergmann wrote:
> > On Sat, Apr 30, 2022 at 12:00 PM Huacai Chen <chenhuacai@xxxxxxxxx> wrote:
> >>
> >> On Sat, Apr 30, 2022 at 5:50 PM Arnd Bergmann <arnd@xxxxxxxx> wrote:
> >> >
> >> > On Sat, Apr 30, 2022 at 11:05 AM Huacai Chen <chenhuacai@xxxxxxxxxxx> wrote:
> >> >
> >> > > +unsigned long __xchg_small(volatile void *ptr, unsigned long val, unsigned int size)
> >> > > +{
> >> > > + u32 old32, mask, temp;
> >> > > + volatile u32 *ptr32;
> >> > > + unsigned int shift;
> >> > > +
> >> > > + /* Check that ptr is naturally aligned */
> >> >
> >> > As discussed, please remove this function and all the references to it.
> >>
> >> It seems that "generic ticket spinlock" hasn't been merged in 5.18?
> >
> > No, but we can merge it together with the loongarch architecture for 5.19.
> >
> > I suggested you coordinate with Guo Ren and Palmer Dabbelt about how
> > to best merge it. The latest version was pasted two weeks ago [1], and
> > it sounds like there are only minor issues to work out and that I can merge
> > v4 into the asm-generic tree before merging the loongarch code in the
> > same place.
> >
> > Arnd
> >
> > [1] https://lore.kernel.org/lkml/20220414220214.24556-1-palmer@xxxxxxxxxxxx/
>
> I can just send another version, IIRC it was just that discussion about
> the memory barrier and there's already prototype code so it shouldn't be
> too bad. I was hoping to do it sooner, sorry.
I've seen your v4 patches, then I will adjust LoongArch code to use
generic ticket spinlocks.
Huacai
