Received-SPF: pass (google.com: domain of linux-kernel+bounces-118360-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240301203023.2197451-1-jsavitz@redhat.com> <87jzmduiva.fsf@kernel.org>
In-Reply-To: <87jzmduiva.fsf@kernel.org>
From: Joel Savitz <jsavitz@redhat.com>
Date: Tue, 26 Mar 2024 00:45:16 -0400
Message-ID: <CAL1p7m5BoxFDeK0MryQCmTDCeBLN3rMLRGx3cHa6teS02wsgZw@mail.gmail.com>
Subject: Re: [PATCH] powerpc: align memory_limit to 16MB in early_parse_mem
To: "Aneesh Kumar K.V" <aneesh.kumar@kernel.org>
Cc: linux-kernel@vger.kernel.org, Michael Ellerman <mpe@ellerman.id.au>, 
	Nicholas Piggin <npiggin@gmail.com>, Christophe Leroy <christophe.leroy@csgroup.eu>, 
	"Naveen N. Rao" <naveen.n.rao@linux.ibm.com>, Benjamin Gray <bgray@linux.ibm.com>, 
	Paul Mackerras <paulus@ozlabs.org>, linuxppc-dev@lists.ozlabs.org, 
	Gonzalo Siero <gsierohu@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Mar 8, 2024 at 5:18=E2=80=AFAM Aneesh Kumar K.V <aneesh.kumar@kerne=
l.org> wrote:
>
> Joel Savitz <jsavitz@redhat.com> writes:
>
> > On 64-bit powerpc, usage of a non-16MB-aligned value for the mem=3D ker=
nel
> > cmdline parameter results in a system hang at boot.
> >
> > For example, using 'mem=3D4198400K' will always reproduce this issue.
> >
> > This patch fixes the problem by aligning any argument to mem=3D to 16MB
> > corresponding with the large page size on powerpc.
> >
> > Fixes: 2babf5c2ec2f ("[PATCH] powerpc: Unify mem=3D handling")
> > Co-developed-by: Gonzalo Siero <gsierohu@redhat.com>
> > Signed-off-by: Gonzalo Siero <gsierohu@redhat.com>
> > Signed-off-by: Joel Savitz <jsavitz@redhat.com>
> > ---
> >  arch/powerpc/kernel/prom.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
> > index 0b5878c3125b..8cd3e2445d8a 100644
> > --- a/arch/powerpc/kernel/prom.c
> > +++ b/arch/powerpc/kernel/prom.c
> > @@ -82,8 +82,12 @@ static int __init early_parse_mem(char *p)
> >  {
> >       if (!p)
> >               return 1;
> > -
> > +#ifdef CONFIG_PPC64
> > +     /* Align to 16 MB =3D=3D size of ppc64 large page */
> > +     memory_limit =3D ALIGN(memparse(p, &p), 0x1000000);
> > +#else
> >       memory_limit =3D PAGE_ALIGN(memparse(p, &p));
> > +#endif
> >       DBG("memory limit =3D 0x%llx\n", memory_limit);
> >
> >       return 0;
> > --
> > 2.43.0
>
> Can you try this change?
>
> commit 5555bc55e1aa71f545cff31e1eccdb4a2e39df84
> Author: Aneesh Kumar K.V (IBM) <aneesh.kumar@kernel.org>
> Date:   Fri Mar 8 14:45:26 2024 +0530
>
>     powerpc/mm: Align memory_limit value specified using mem=3D kernel pa=
rameter
>
>     The value specified for the memory limit is used to set a restriction=
 on
>     memory usage. It is important to ensure that this restriction is with=
in
>     the linear map kernel address space range. The hash page table
>     translation uses a 16MB page size to map the kernel linear map addres=
s
>     space. htab_bolt_mapping() function aligns down the size of the range
>     while mapping kernel linear address space. Since the memblock limit i=
s
>     enforced very early during boot, before we can detect the type of mem=
ory
>     translation (radix vs hash), we align the memory limit value specifie=
d
>     as a kernel parameter to 16MB. This alignment value will work for bot=
h
>     hash and radix translations.
>
>     Signed-off-by: Aneesh Kumar K.V (IBM) <aneesh.kumar@kernel.org>
>
> diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
> index 0b5878c3125b..9bd965d35352 100644
> --- a/arch/powerpc/kernel/prom.c
> +++ b/arch/powerpc/kernel/prom.c
> @@ -824,8 +824,11 @@ void __init early_init_devtree(void *params)
>                 reserve_crashkernel();
>         early_reserve_mem();
>
> -       /* Ensure that total memory size is page-aligned. */
> -       limit =3D ALIGN(memory_limit ?: memblock_phys_mem_size(), PAGE_SI=
ZE);
> +       if (memory_limit > memblock_phys_mem_size())
> +               memory_limit =3D 0;
> +
> +       /* Align down to 16 MB which is large page size with hash page tr=
anslation */
> +       limit =3D ALIGN_DOWN(memory_limit ?: memblock_phys_mem_size(), SZ=
_16M);
>         memblock_enforce_memory_limit(limit);
>
>  #if defined(CONFIG_PPC_BOOK3S_64) && defined(CONFIG_PPC_4K_PAGES)
> diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_i=
nit.c
> index e67effdba85c..d6410549e141 100644
> --- a/arch/powerpc/kernel/prom_init.c
> +++ b/arch/powerpc/kernel/prom_init.c
> @@ -817,8 +817,8 @@ static void __init early_cmdline_parse(void)
>                 opt +=3D 4;
>                 prom_memory_limit =3D prom_memparse(opt, (const char **)&=
opt);
>  #ifdef CONFIG_PPC64
> -               /* Align to 16 MB =3D=3D size of ppc64 large page */
> -               prom_memory_limit =3D ALIGN(prom_memory_limit, 0x1000000)=
;
> +               /* Align down to 16 MB which is large page size with hash=
 page translation */
> +               prom_memory_limit =3D ALIGN_DOWN(prom_memory_limit, SZ_16=
M);
>  #endif
>         }
>
>

Sorry for the delayed reply. I just tested this patch and it fixes the
bug for me.