From: Ian Campbell <hidden> Date: 2016-08-03 14:13:11
This file is included from DTS files under arch/arm64 too (via
broadcom/bcm2837-rpi-3-b.dts and broadcom/bcm2837.dtsi). There is a desire
not to have skeleton.dtsi for ARM64. See commit 3ebee5a2e141 ("arm64: dts:
kill skeleton.dtsi") for rationale for its removal.
As well as the addition of #*-cells also requires adding the device_type to
the rpi memory node explicitly.
Note that this change results in the removal of an empty /aliases node from
bcm2835-rpi-a.dtb and bcm2835-rpi-a-plus.dtb. I have no hardware to check
if this is a problem or not.
It also results in some reordering of the nodes in the DTBs (the /aliases
and /memory nodes come later). This isn't supposed to matter but, again,
I've no hardware to check if it is true in this particular case.
Signed-off-by: Ian Campbell <redacted>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <redacted>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Frank Rowand <redacted>
Cc: Eric Anholt <redacted>
Cc: Stephen Warren <redacted>
Cc: Lee Jones <lee@kernel.org>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: devicetree at vger.kernel.org
Cc: linux-kernel at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Cc: linux-rpi-kernel at lists.infradead.org
Cc: arm at kernel.org
---
v2: New patch to avoid needing to add skeleton.dtsi to arch/arm64
---
arch/arm/boot/dts/bcm2835-rpi.dtsi | 1 +
arch/arm/boot/dts/bcm283x.dtsi | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
@@ -2,7 +2,6 @@#include<dt-bindings/clock/bcm2835.h>#include<dt-bindings/clock/bcm2835-aux.h>#include<dt-bindings/gpio/gpio.h>-#include"skeleton.dtsi"/* This include file covers the common peripherals and configuration between*bcm2835andbcm2836implementations,leavingtheCPUconfigurationto
From: Ian Campbell <hidden> Date: 2016-08-03 14:13:16
The ../../../arm... style cross-references added by commit 9d56c22a7861
("ARM: bcm2835: Add devicetree for the Raspberry Pi 3.") do not work in the
context of the split device-tree repository[0] (where the directory
structure differs). As with commit 8ee57b8182c4 ("ARM64: dts: vexpress: Use
a symlink to vexpress-v2m-rs1.dtsi from arch=arm") use symlinks instead.
[0] https://git.kernel.org/cgit/linux/kernel/git/devicetree/devicetree-rebasing.git/
Signed-off-by: Ian Campbell <redacted>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <redacted>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Frank Rowand <redacted>
Cc: Eric Anholt <redacted>
Cc: Stephen Warren <redacted>
Cc: Lee Jones <lee@kernel.org>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: devicetree at vger.kernel.org
Cc: linux-kernel at vger.kernel.org
Cc: linux-arm-kernel at lists.infradead.org
Cc: linux-rpi-kernel at lists.infradead.org
Cc: arm at kernel.org
---
v2: No more need for skeleton.dtsi since this was cleaned up in the previous
patch
---
arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi | 1 +
arch/arm64/boot/dts/broadcom/bcm2837-rpi-3-b.dts | 4 ++--
arch/arm64/boot/dts/broadcom/bcm2837.dtsi | 2 +-
arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi | 1 +
arch/arm64/boot/dts/broadcom/bcm283x.dtsi | 1 +
5 files changed, 6 insertions(+), 3 deletions(-)
create mode 120000 arch/arm64/boot/dts/broadcom/bcm2835-rpi.dtsi
create mode 120000 arch/arm64/boot/dts/broadcom/bcm283x-rpi-smsc9514.dtsi
create mode 120000 arch/arm64/boot/dts/broadcom/bcm283x.dtsi
From: Mark Rutland <mark.rutland@arm.com> Date: 2016-08-03 14:55:45
On Wed, Aug 03, 2016 at 03:12:45PM +0100, Ian Campbell wrote:
The ../../../arm... style cross-references added by commit 9d56c22a7861
("ARM: bcm2835: Add devicetree for the Raspberry Pi 3.") do not work in the
context of the split device-tree repository[0] (where the directory
structure differs). As with commit 8ee57b8182c4 ("ARM64: dts: vexpress: Use
a symlink to vexpress-v2m-rs1.dtsi from arch=arm") use symlinks instead.
[0] https://git.kernel.org/cgit/linux/kernel/git/devicetree/devicetree-rebasing.git/
Signed-off-by: Ian Campbell <redacted>
As this matches what we did in 8ee57b8182c4, and no longer involves
raising skeletons back from the dead:
Acked-by: Mark Rutland <mark.rutland@arm.com>
Mark.
From: Mark Rutland <mark.rutland@arm.com> Date: 2016-08-03 15:00:49
On Wed, Aug 03, 2016 at 03:12:44PM +0100, Ian Campbell wrote:
This file is included from DTS files under arch/arm64 too (via
broadcom/bcm2837-rpi-3-b.dts and broadcom/bcm2837.dtsi). There is a desire
not to have skeleton.dtsi for ARM64. See commit 3ebee5a2e141 ("arm64: dts:
kill skeleton.dtsi") for rationale for its removal.
As well as the addition of #*-cells also requires adding the device_type to
the rpi memory node explicitly.
Note that this change results in the removal of an empty /aliases node from
bcm2835-rpi-a.dtb and bcm2835-rpi-a-plus.dtb. I have no hardware to check
if this is a problem or not.
It also results in some reordering of the nodes in the DTBs (the /aliases
and /memory nodes come later). This isn't supposed to matter but, again,
I've no hardware to check if it is true in this particular case.
Signed-off-by: Ian Campbell <redacted>
I also don't have the relevant hardware to test with, but this looks
generally like the right thing. So FWIW:
Acked-by: Mark Rutland <mark.rutland@arm.com>
Mark.
@@ -2,7 +2,6 @@#include<dt-bindings/clock/bcm2835.h>#include<dt-bindings/clock/bcm2835-aux.h>#include<dt-bindings/gpio/gpio.h>-#include"skeleton.dtsi"/* This include file covers the common peripherals and configuration between*bcm2835andbcm2836implementations,leavingtheCPUconfigurationto
From: Ian Campbell <hidden> Date: 2016-08-09 10:48:49
On Wed, 2016-08-03 at 15:48 +0100, Mark Rutland wrote:
quoted
[...]Signed-off-by: Ian Campbell [off-list ref]
I also don't have the relevant hardware to test with, but this looks
generally like the right thing. So FWIW:
Acked-by: Mark Rutland <mark.rutland@arm.com>
Thanks (and Stefan too).
I think these things generally go via the arm-soc tree? Arnd & Olaf,
would it be possible to get this fix in for rc2 please (or in any event
for 4.8). Although it's an external tree I believe build breakage in
the split-out DT git repo is worth addressing.
Ian.
From: Ian Campbell <hidden> Date: 2016-08-18 07:22:09
On Tue, 2016-08-09 at 11:48 +0100, Ian Campbell wrote:
On Wed, 2016-08-03 at 15:48 +0100, Mark Rutland wrote:
quoted
?
quoted
[...]Signed-off-by: Ian Campbell [off-list ref]
?
I also don't have the relevant hardware to test with, but this
looks
quoted
generally like the right thing. So FWIW:
?
Acked-by: Mark Rutland <mark.rutland@arm.com>
Thanks (and Stefan too).
I think these things generally go via the arm-soc tree? Arnd & Olaf,
would it be possible to get this fix in for rc2 please (or in any event
for 4.8). Although it's an external tree I believe build breakage in
the split-out DT git repo is worth addressing.
This change doesn't appear to be in either v4.8-rc2, Linus' tree nor in
the arm-soc tree that I can see. So, ping?
Or am I barking up the wrong tree pointing this patch towards arm-soc
(via arm at k.o)?
Ian
From: Mark Rutland <mark.rutland@arm.com> Date: 2016-08-18 10:23:59
On Thu, Aug 18, 2016 at 08:21:56AM +0100, Ian Campbell wrote:
On Tue, 2016-08-09 at 11:48 +0100, Ian Campbell wrote:
quoted
On Wed, 2016-08-03 at 15:48 +0100, Mark Rutland wrote:
quoted
?
quoted
[...]Signed-off-by: Ian Campbell [off-list ref]
?
I also don't have the relevant hardware to test with, but this
looks
quoted
generally like the right thing. So FWIW:
?
Acked-by: Mark Rutland <mark.rutland@arm.com>
Thanks (and Stefan too).
I think these things generally go via the arm-soc tree? Arnd & Olaf,
would it be possible to get this fix in for rc2 please (or in any event
for 4.8). Although it's an external tree I believe build breakage in
the split-out DT git repo is worth addressing.
This change doesn't appear to be in either v4.8-rc2, Linus' tree nor in
the arm-soc tree that I can see. So, ping?
Or am I barking up the wrong tree pointing this patch towards arm-soc
(via arm at k.o)?
That's the right place.
The best thing to do would be to resend the patch, with all tags
accumulated, with arm-soc, Arnd, and Olof in the To line (rather than
Cc'd), so it's clear they need to action it.
Thanks,
Mark.
On Thursday, August 18, 2016 11:23:49 AM CEST Mark Rutland wrote:
On Thu, Aug 18, 2016 at 08:21:56AM +0100, Ian Campbell wrote:
quoted
On Tue, 2016-08-09 at 11:48 +0100, Ian Campbell wrote:
quoted
On Wed, 2016-08-03 at 15:48 +0100, Mark Rutland wrote:
quoted
quoted
[...]Signed-off-by: Ian Campbell [off-list ref]
I also don't have the relevant hardware to test with, but this
looks
quoted
generally like the right thing. So FWIW:
Acked-by: Mark Rutland <mark.rutland@arm.com>
Thanks (and Stefan too).
I think these things generally go via the arm-soc tree? Arnd & Olaf,
would it be possible to get this fix in for rc2 please (or in any event
for 4.8). Although it's an external tree I believe build breakage in
the split-out DT git repo is worth addressing.
This change doesn't appear to be in either v4.8-rc2, Linus' tree nor in
the arm-soc tree that I can see. So, ping?
Or am I barking up the wrong tree pointing this patch towards arm-soc
(via arm at k.o)?
That's the right place.
The best thing to do would be to resend the patch, with all tags
accumulated, with arm-soc, Arnd, and Olof in the To line (rather than
Cc'd), so it's clear they need to action it.
I had skipped the new message as well while sorting through 5000
messages after my vacation, but I've now put it into my TODO folder
and will get to it eventually.
We sometimes miss stuff that is meant for arm-soc when it comes
from people that don't normally send us patches. If you want to
be sure to catch the attention, stick a 'GIT PULL' into the subject
or send a ping on IRC.
The easiest way for us is when things get funnelled through the
platform maintainers that handle all the other patches, but I
can see how that may be annoying for a patch like this.
Arnd
From: Ian Campbell <hidden> Date: 2016-09-09 14:46:06
On Tue, 2016-08-23 at 12:09 +0200, Arnd Bergmann wrote:
I had skipped the new message as well while sorting through 5000
messages after my vacation, but I've now put it into my TODO folder
and will get to it eventually.
Is it still on your TODO or should I do...
We sometimes miss stuff that is meant for arm-soc when it comes
from people that don't normally send us patches. If you want to
be sure to catch the attention, stick a 'GIT PULL' into the subject
or send a ping on IRC.
... one of these?
It'd be great to get this fixed in v4.8 so that the corresponding
device-tree.git tag will build.
Thanks,
Ian.
On Friday, September 9, 2016 3:45:56 PM CEST Ian Campbell wrote:
On Tue, 2016-08-23 at 12:09 +0200, Arnd Bergmann wrote:
quoted
I had skipped the new message as well while sorting through 5000
messages after my vacation, but I've now put it into my TODO folder
and will get to it eventually.
Is it still on your TODO or should I do...
quoted
We sometimes miss stuff that is meant for arm-soc when it comes
from people that don't normally send us patches. If you want to
be sure to catch the attention, stick a 'GIT PULL' into the subject
or send a ping on IRC.
... one of these?
It'd be great to get this fixed in v4.8 so that the corresponding
device-tree.git tag will build.
Oh, it ended up in my v4.9-todo folder, not 4.8, I misunderstood
the urgency, sorry.
I've applied it to the fixes branch now.
Arnd
From: Ian Campbell <hidden> Date: 2016-09-09 16:21:12
On Fri, 2016-09-09 at 17:47 +0200, Arnd Bergmann wrote:
On Friday, September 9, 2016 3:45:56 PM CEST Ian Campbell wrote:
quoted
On Tue, 2016-08-23 at 12:09 +0200, Arnd Bergmann wrote:
quoted
I had skipped the new message as well while sorting through 5000
messages after my vacation, but I've now put it into my TODO
folder
and will get to it eventually.
Is it still on your TODO or should I do...
quoted
We sometimes miss stuff that is meant for arm-soc when it comes
from people that don't normally send us patches. If you want to
be sure to catch the attention, stick a 'GIT PULL' into the
subject
or send a ping on IRC.
... one of these?
It'd be great to get this fixed in v4.8 so that the corresponding
device-tree.git tag will build.
Oh, it ended??up in my v4.9-todo folder, not 4.8, I misunderstood
the urgency, sorry.
From: Stefan Wahren <hidden> Date: 2016-08-03 16:57:53
Hi Ian,
Ian Campbell [off-list ref] hat am 3. August 2016 um 16:12 geschrieben:
This file is included from DTS files under arch/arm64 too (via
broadcom/bcm2837-rpi-3-b.dts and broadcom/bcm2837.dtsi). There is a desire
not to have skeleton.dtsi for ARM64. See commit 3ebee5a2e141 ("arm64: dts:
kill skeleton.dtsi") for rationale for its removal.
As well as the addition of #*-cells also requires adding the device_type to
the rpi memory node explicitly.
Note that this change results in the removal of an empty /aliases node from
bcm2835-rpi-a.dtb and bcm2835-rpi-a-plus.dtb. I have no hardware to check
if this is a problem or not.
It also results in some reordering of the nodes in the DTBs (the /aliases
and /memory nodes come later). This isn't supposed to matter but, again,
I've no hardware to check if it is true in this particular case.
i tested this patch with a Raspberry Pi Model B successfully.
Tested-by: Stefan Wahren <redacted>