Re: [PATCH v1 1/3] powerpc/code-patching: Test patch_instructions() during boot

[Benjamin Gray <bgray@linux.ibm.com>]

 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
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Date: Mon, 18 Mar 2024 08:55:02 +1100
MIME-Version: 1.0
User-Agent: Evolution 3.50.4 (3.50.4-1.fc39) 
X-Trend-IP-HD: ip=[9.192.253.14]helo={ozlabs.au.ibm.com}sender=(bgray@linux.ibm.com)recipient=<christophe.leroy@csgroup.eu;mpe@ellerman.id.au;linuxppc-dev@lists.ozlabs.org>

On Mon, 2024-03-18 at 08:38 +1100, Benjamin Gray wrote:
> On Fri, 2024-03-15 at 07:14 +0000, Christophe Leroy wrote:
> >=20
> >=20
> > Le 15/03/2024 =C3=A0 03:57, Benjamin Gray a =C3=A9crit=C2=A0:
> > > patch_instructions() introduces new behaviour with a couple of
> > > variations. Test each case of
> > >=20
> > > =C2=A0=C2=A0 * a repeated 32-bit instruction,
> > > =C2=A0=C2=A0 * a repeated 64-bit instruction (ppc64), and
> > > =C2=A0=C2=A0 * a copied sequence of instructions
> > >=20
> > > for both on a single page and when it crosses a page boundary.
> > >=20
> > > Signed-off-by: Benjamin Gray <bgray@linux.ibm.com>
> > > ---
> > > =C2=A0 arch/powerpc/lib/test-code-patching.c | 92
> > > +++++++++++++++++++++++++++
> > > =C2=A0 1 file changed, 92 insertions(+)
> > >=20
> > > diff --git a/arch/powerpc/lib/test-code-patching.c
> > > b/arch/powerpc/lib/test-code-patching.c
> > > index c44823292f73..35a3756272df 100644
> > > --- a/arch/powerpc/lib/test-code-patching.c
> > > +++ b/arch/powerpc/lib/test-code-patching.c
> > > @@ -347,6 +347,97 @@ static void __init
> > > test_prefixed_patching(void)
> > > =C2=A0=C2=A0	check(!memcmp(iptr, expected, sizeof(expected)));
> > > =C2=A0 }
> > > =C2=A0=20
> > > +static void __init test_multi_instruction_patching(void)
> > > +{
> > > +	u32 code[256];
> >=20
> > Build failure:
> >=20
> > =C2=A0=C2=A0 CC=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 arch/powerpc/lib/test-cod=
e-patching.o
> > arch/powerpc/lib/test-code-patching.c: In function=20
> > 'test_multi_instruction_patching':
> > arch/powerpc/lib/test-code-patching.c:439:1: error: the frame size
> > of
> > 1040 bytes is larger than 1024 bytes [-Werror=3Dframe-larger-than=3D]
> > =C2=A0=C2=A0 439 | }
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | ^
> > cc1: all warnings being treated as errors
> > make[4]: *** [scripts/Makefile.build:243:=20
> > arch/powerpc/lib/test-code-patching.o] Error 1
> >=20
> >=20
> > I have to avoid big arrays on the stack.
>=20
> All good, I can do that.
>=20
> I do run my patches through a couple of 32-bit configs, but I didn't
> see this error. Is this a standard config I should be testing with?
>=20

Specifically I build pmac32_defconfig and ppc44x_defconfig