From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: MIME-Version: 1.0 In-Reply-To: References: <1488313886-17155-1-git-send-email-jon.mason@broadcom.com> <1488313886-17155-2-git-send-email-jon.mason@broadcom.com> From: Jon Mason Date: Thu, 2 Mar 2017 14:00:45 -0500 Message-ID: Subject: Re: [PATCH 1/2] ARM: dts: bcm5301x: Add TWD WD Support to DT Content-Type: multipart/alternative; boundary=001a11443ef6e95fdd0549c40b37 To: =?UTF-8?B?UmFmYcWCIE1pxYJlY2tp?= Cc: Hauke Mehrtens , Rob Herring , Mark Rutland , Florian Fainelli , "open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS" , linux-arm-kernel , open list , BCM Kernel Feedback , Jon Mason List-ID: --001a11443ef6e95fdd0549c40b37 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Thu, Mar 2, 2017 at 1:54 PM, Rafa=C5=82 Mi=C5=82ecki = wrote: > On 02/28/2017 09:31 PM, Jon Mason wrote: > >> From: Jon Mason >> >> Add support for the ARM TWD Watchdog to the bcm5301x device tree. The >> ARM TWD timer allocated the register space for the WDT, so this patch >> necessitated shrinking that. Also, the GIC masks were added for these. >> >> Signed-off-by: Jon Mason >> --- >> arch/arm/boot/dts/bcm5301x.dtsi | 15 ++++++++++++--- >> 1 file changed, 12 insertions(+), 3 deletions(-) >> >> diff --git a/arch/arm/boot/dts/bcm5301x.dtsi >> b/arch/arm/boot/dts/bcm5301x.dtsi >> index 4fbb089..3fbc450 100644 >> --- a/arch/arm/boot/dts/bcm5301x.dtsi >> +++ b/arch/arm/boot/dts/bcm5301x.dtsi >> @@ -70,10 +70,19 @@ >> clocks =3D <&periph_clk>; >> }; >> >> - local-timer@20600 { >> + timer@20600 { >> compatible =3D "arm,cortex-a9-twd-timer"; >> - reg =3D <0x20600 0x100>; >> - interrupts =3D ; >> + reg =3D <0x20600 0x20>; >> + interrupts =3D > + IRQ_TYPE_LEVEL_HIGH)>; >> + clocks =3D <&periph_clk>; >> + }; >> > > If you follow my recent e-mail thread: > BCM5301X: GIC: PPI11 is secure or misconfigured (same for PPI13) > you'll see IRQ_TYPE_LEVEL_HIGH type isn't correct. It should be > IRQ_TYPE_EDGE_RISING. > > I believe patch switching to IRQ_TYPE_EDGE_RISING should be sent with Cc > stable > for kernels 4.8+. > > The same change is needed for "arm,cortex-a9-global-timer". > > Would you find time to revise this patch? > I'll do 2 patches. One to revise this one and one to address the issue you've discovered. Will that be okay for you? --001a11443ef6e95fdd0549c40b37 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


On Thu, Mar 2, 2017 at 1:54 PM, Rafa=C5=82 Mi=C5=82ecki <zajec5@gmail.c= om> wrote:
On 02/28/2017 09:31 PM, Jon Mason wrote:
From: Jon Mason <jonmason@broadcom.com>

Add support for the ARM TWD Watchdog to the bcm5301x device tree.=C2=A0 The=
ARM TWD timer allocated the register space for the WDT, so this patch
necessitated shrinking that.=C2=A0 Also, the GIC masks were added for these= .

Signed-off-by: Jon Mason <jonmason@broadcom.com>
---
=C2=A0arch/arm/boot/dts/bcm5301x.dtsi | 15 ++++++++++++---
=C2=A01 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/arch/arm/boot/dts/bcm5301x.dtsi b/arch/arm/boot/dts/bcm53= 01x.dtsi
index 4fbb089..3fbc450 100644
--- a/arch/arm/boot/dts/bcm5301x.dtsi
+++ b/arch/arm/boot/dts/bcm5301x.dtsi
@@ -70,10 +70,19 @@
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 clocks =3D <&periph_clk>;
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 };

-=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0local-timer@20600 {=
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0timer@20600 {
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 compatible =3D "arm,cortex-a9-twd-timer";
-=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0reg =3D <0x20600 0x100>;
-=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0interrupts =3D <GIC_PPI 13 IRQ_TYPE_LEVEL_HIGH>;
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0reg =3D <0x20600 0x20>;
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0interrupts =3D <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(2) |
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0IRQ_TYPE_LEVEL_HIGH)>;
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0clocks =3D <&periph_clk>;
+=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0};

If you follow my recent e-mail thread:
BCM5301X: GIC: PPI11 is secure or misconfigured (same for PPI13)
you'll see IRQ_TYPE_LEVEL_HIGH type isn't correct. It should be
IRQ_TYPE_EDGE_RISING.

I believe patch switching to IRQ_TYPE_EDGE_RISING should be sent with Cc st= able
for kernels 4.8+.

The same change is needed for "arm,cortex-a9-global-timer".

Would you find time to revise this patch?

I'll do 2 patches.=C2=A0 One to revise this one and one to address t= he issue you've discovered.=C2=A0 Will that be okay for you?
= =C2=A0

--001a11443ef6e95fdd0549c40b37--