[PATCH] arm64: cache: Lower ARCH_DMA_MINALIGN to 64 (L1_CACHE_BYTES)

[Jeffrey Hugo]

On Wed, Jul 7, 2021 at 8:41 AM Jeffrey Hugo <jeffrey.l.hugo at gmail.com> wrote:
>
> On Wed, Jul 7, 2021 at 3:30 AM Arnd Bergmann <arnd at arndb.de> wrote:
> >
> > On Tue, Jul 6, 2021 at 6:20 PM Will Deacon <will at kernel.org> wrote:
> > >
> > > I think the million dollar question is whether the 128-byte cache-lines
> > > live in a cache above the PoC or not. If it's just a system level cache
> > > through which all DMA is "coherent", then it doesn't matter.
> >
> > On Wed, Jul 7, 2021 at 10:24 AM Yassine Oudjana
> > <y.oudjana at protonmail.com> wrote:
> > >
> > > On Wednesday, July 7th, 2021 at 12:33 AM, Arnd Bergmann <arnd at arndb.de> wrote:
> > > > On Tue, Jul 6, 2021 at 7:15 PM Yassine Oudjana y.oudjana at protonmail.com wrote:
> > > > >
> > > > > $ numactl -C 0 line -M 1M
> > > > > 128
> > > > > $ numactl -C 3 line -M 1M
> > > > > 128
> > > >
> > > >     Can you rerun the the 'line' test with '-M 128K' to see if that confirms the 64
> > > >     byte L1 line size that the 'cache' test reported?
> > >
> > > $ numactl -C 0 line -M 128K
> > > 64
> > > $ numactl -C 3 line -M 128K
> > > 64
> >
> > Ok, so this seems to confirm that the L1 uses 64 byte lines, while the L2 (or
> > possibly L3) uses 128 byte lines.
> >
> > On Wed, Jul 7, 2021 at 12:27 AM Bjorn Andersson
> > <bjorn.andersson at linaro.org> wrote:
> > >
> > > I can confirm that MSM8996, and a few derivatives, has 128 byte cache lines.
> >
> > Ok, thanks. Assuming this is an outer cache and the L1 indeed has a smaller line
> > size, can you also confirm that this 128 byte lines are north of the point of
> > coherency?
>
> Finding this old documentation has been painful  :)
>
> L0 I 64 byte cacheline
> L1 I 64
> L1 D 64
> L2 unified 128 (shared between the CPUs of a duplex)
>
> I believe L2 is within the POC, but I'm trying to dig up the old
> documentation to confirm.

Was able to track down a friendly hardware designer.  The POC lies
between L2 and L3.  Hope this helps.

> > In other words, does the CTR_EL0.DminLine field also show 128 bytes
> > (in which case
> > it seems we already lost)? And if not, does a CPU store to the second half of a
> > 128 byte L2 line cause DMA data in the first half to be clobbered?
>
> Per the documentation I'm seeing, CTR_EL0.DminLine should show 128
> bytes.  I don't have hardware handy to confirm.