Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Haibo Li <haibo.li@mediatek.com>
To:     <angelogioacchino.delregno@collabora.com>
CC:     <a.anurag@samsung.com>, <alexander.sverdlin@nokia.com>,
        <amergnat@baylibre.com>, <ardb@kernel.org>,
        <catalin.marinas@arm.com>, <haibo.li@mediatek.com>,
        <linus.walleij@linaro.org>, <linux-arm-kernel@lists.infradead.org>,
        <linux-kernel@vger.kernel.org>,
        <linux-mediatek@lists.infradead.org>, <linux@armlinux.org.uk>,
        <matthias.bgg@gmail.com>, <rmk+kernel@armlinux.org.uk>,
        <xiaoming.yu@mediatek.com>
Subject: Re: [PATCH v2] ARM:unwind:fix unwind abort for uleb128 case
Date:   Fri, 14 Apr 2023 13:57:04 +0800
Message-ID: <20230414055704.123841-1-haibo.li@mediatek.com>
In-Reply-To: <22bb4f8f-8f4b-6efb-74ab-b33eabc1fbb9@collabora.com>
References: <22bb4f8f-8f4b-6efb-74ab-b33eabc1fbb9@collabora.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain
Precedence: bulk

> Il 13/04/23 09:34, Haibo Li ha scritto:=0D
> > When unwind instruction is 0xb2,the subsequent instructions are=0D
> > uleb128 bytes.=0D
> > For now,it uses only the first uleb128 byte in code.=0D
> >=0D
> > For vsp increments of 0x204~0x400,use one uleb128 byte like below:=0D
> > 0xc06a00e4 <unwind_test_work>: 0x80b27fac=0D
> >    Compact model index: 0=0D
> >    0xb2 0x7f vsp =3D vsp + 1024=0D
> >    0xac      pop {r4, r5, r6, r7, r8, r14}=0D
> >=0D
> > For vsp increments larger than 0x400,use two uleb128 bytes like below:=
=0D
> > 0xc06a00e4 <unwind_test_work>: @0xc0cc9e0c=0D
> >    Compact model index: 1=0D
> >    0xb2 0x81 0x01 vsp =3D vsp + 1032=0D
> >    0xac      pop {r4, r5, r6, r7, r8, r14}=0D
> > The unwind works well since the decoded uleb128 byte is also 0x81.=0D
> >=0D
> > For vsp increments larger than 0x600,use two uleb128 bytes like below:=
=0D
> > 0xc06a00e4 <unwind_test_work>: @0xc0cc9e0c=0D
> >    Compact model index: 1=0D
> >    0xb2 0x81 0x02 vsp =3D vsp + 1544=0D
> >    0xac      pop {r4, r5, r6, r7, r8, r14}=0D
> > In this case,the decoded uleb128 result is 0x101(vsp=3D0x204+(0x101<<2)=
).=0D
> > While the uleb128 used in code is 0x81(vsp=3D0x204+(0x81<<2)).=0D
> > The unwind aborts at this frame since it gets incorrect vsp.=0D
> >=0D
> > To fix this,add uleb128 decode to cover all the above case.=0D
> >=0D
> > Signed-off-by: Haibo Li <haibo.li@mediatek.com>=0D
> > ---=0D
> > v2:=0D
> > - As Linus Walleij and Alexandre Mergnat suggested,add comments for=0D
> > unwind_decode_uleb128=0D
> > - As Alexandre Mergnat suggested,change variables declaration in=0D
> > Alphabetical order=0D
> > ---=0D
> >   arch/arm/kernel/unwind.c | 25 ++++++++++++++++++++++++-=0D
> >   1 file changed, 24 insertions(+), 1 deletion(-)=0D
> >=0D
> > diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c index=
=0D
> > 53be7ea6181b..f37e55fcf81d 100644=0D
> > --- a/arch/arm/kernel/unwind.c=0D
> > +++ b/arch/arm/kernel/unwind.c=0D
> > @@ -308,6 +308,29 @@ static int=0D
> unwind_exec_pop_subset_r0_to_r3(struct unwind_ctrl_block *ctrl,=0D
> >       return URC_OK;=0D
> >   }=0D
> >=0D
> > +static unsigned long unwind_decode_uleb128(struct unwind_ctrl_block=0D
> > +*ctrl) {=0D
> > +     unsigned long bytes =3D 0;=0D
> > +     unsigned long insn;=0D
> > +     unsigned long result =3D 0;=0D
> > +=0D
> > +     /* unwind_get_byte() will advance ctrl one instruction at a time,=
=0D
> > +      * we loop until we get an instruction byte where bit 7 is not se=
t.=0D
> > +      * Note:It decodes max 4 bytes to output 28bits data.=0D
> > +      * 28bits data(0xfffffff) covers vsp increments of 1073742336.=0D
> > +      * It is sufficent for unwinding stack.=0D
> > +      */=0D
> =0D
> /*=0D
>   * unwind_get_byte() will advance `ctrl` one instruction at a time, so=0D
>   * loop until we get an instruction byte where bit 7 is not set.=0D
>   *=0D
>   * Note: This decodes a maximum of 4 bytes to output 28 bits data where=
=0D
>   * max is 0xfffffff: that will cover a vsp increment of 1073742336, henc=
e=0D
>   * it is sufficient for unwinding the stack.=0D
>   */=0D
Looks much better.Thanks.=0D
> =0D
> > +     do {=0D
> > +             insn =3D unwind_get_byte(ctrl);=0D
> > +             result |=3D (insn & 0x7f) << (bytes * 7);=0D
> > +             bytes++;=0D
> =0D
> also, I would do ...=0D
> =0D
>         } while (!!(insn & 0x80) && bytes !=3D sizeof(result));=0D
> =0D
> ...compressing the code and not creating any human readability concern.=0D
> =0D
> after which, you can get my=0D
> =0D
> Reviewed-by: AngeloGioacchino Del Regno=0D
> <angelogioacchino.delregno@collabora.com>=0D
get it.I will make a new patch.=0D