Re: [PATCH 4.4 03/16] devres: Align data[] to ARCH_KMALLOC_MINALIGN

[Geert Uytterhoeven <geert@linux-m68k.org>]

Hi Greg,

On Wed, May 13, 2020 at 11:35 AM Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
> On Thu, Apr 23, 2020 at 09:40:01PM +0100, Lee Jones wrote:
> > From: Alexey Brodkin <alexey.brodkin@synopsys.com>
> >
> > [ Upstream commit a66d972465d15b1d89281258805eb8b47d66bd36 ]
> >
> > Initially we bumped into problem with 32-bit aligned atomic64_t
> > on ARC, see [1]. And then during quite lengthly discussion Peter Z.
> > mentioned ARCH_KMALLOC_MINALIGN which IMHO makes perfect sense.
> > If allocation is done by plain kmalloc() obtained buffer will be
> > ARCH_KMALLOC_MINALIGN aligned and then why buffer obtained via
> > devm_kmalloc() should have any other alignment?
> >
> > This way we at least get the same behavior for both types of
> > allocation.
> >
> > [1] http://lists.infradead.org/pipermail/linux-snps-arc/2018-July/004009.html
> > [2] http://lists.infradead.org/pipermail/linux-snps-arc/2018-July/004036.html
> >
> > Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
> > Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > Cc: Geert Uytterhoeven <geert@linux-m68k.org>
> > Cc: David Laight <David.Laight@ACULAB.COM>
> > Cc: Peter Zijlstra <peterz@infradead.org>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: Vineet Gupta <vgupta@synopsys.com>
> > Cc: Will Deacon <will.deacon@arm.com>
> > Cc: Greg KH <greg@kroah.com>
> > Cc: <stable@vger.kernel.org> # 4.8+
> > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > Signed-off-by: Lee Jones <lee.jones@linaro.org>
> > ---
> >  drivers/base/devres.c | 10 ++++++++--
> >  1 file changed, 8 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/base/devres.c b/drivers/base/devres.c
> > index 8fc654f0807bf..9763325a9c944 100644
> > --- a/drivers/base/devres.c
> > +++ b/drivers/base/devres.c
> > @@ -24,8 +24,14 @@ struct devres_node {
> >
> >  struct devres {
> >       struct devres_node              node;
> > -     /* -- 3 pointers */
> > -     unsigned long long              data[]; /* guarantee ull alignment */
> > +     /*
> > +      * Some archs want to perform DMA into kmalloc caches
> > +      * and need a guaranteed alignment larger than
> > +      * the alignment of a 64-bit integer.
> > +      * Thus we use ARCH_KMALLOC_MINALIGN here and get exactly the same
> > +      * buffer alignment as if it was allocated by plain kmalloc().
> > +      */
> > +     u8 __aligned(ARCH_KMALLOC_MINALIGN) data[];
> >  };
> >
> >  struct devres_group {
> > --
> > 2.25.1
> >
>
> I don't want to apply this to older kernels as it could cause extra
> memory usage for no good reason.  I have no idea why a non ARC system
> would want it :(

I think the reference to ARC is a red herring.
The real issue is that buffers used for DMA may not have the required
alignment, which is not limited to ARC systems.

Note that I'm also not super happy with the extra memory usage.
But devm_*() conveniences come with their penalties...

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds